On Thu, Oct 5, 2017 at 2:22 PM, Mark Brown <[email protected]> wrote:
> On Thu, Oct 05, 2017 at 02:11:17PM +0200, Arnd Bergmann wrote:
>
>> This adds a hard Kconfig dependency on HWSPINLOCK, which is the
>> simplest solution since we don't care about building for the case
>> without HWSPINLOCK.
>
> Given that there are stubs for HWSPINLOCK it'd be good to also have an
> || COMPILE_TEST in there, otherwise what's the point in the stubs?
As far as I'm concerned, we can remove those stubs, it's easy enough
to compile-test with hwspinlock enabled, and the existing in-kernel
users all require hwspinlock anyway.
I've sent a second version now, please pick whichever you prefer.
Arnd
