Re: [RFC/PATCH] drivers/of/platform: Add powerpc 4xx embedded busses to default list

[Rob Herring <robh-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>]

On Fri, Sep 1, 2017 at 6:47 PM, Benjamin Herrenschmidt
<benh-XVmvHMARGAS8U2dJNN8I7kB+6BGkLq7r@public.gmane.org> wrote:
> On Fri, 2017-09-01 at 10:24 -0500, Rob Herring wrote:
>> On Thu, Aug 31, 2017 at 10:51 PM, Benjamin Herrenschmidt
>> <benh-XVmvHMARGAS8U2dJNN8I7kB+6BGkLq7r@public.gmane.org> wrote:
>> > This allow to (slowly) migrate those embedded platforms
>> > to of_platform_default_populate()
>> >
>> > Signed-off-by: Benjamin Herrenschmidt <benh-XVmvHMARGAS8U2dJNN8I7kB+6BGkLq7r@public.gmane.org>
>> > ---
>> >
>> > I'm here to collect acks (or comments :-) I'd like this to go via
>> > the powerpc tree along with the patches converting some of the
>> > platforms. I'll be adding more bus types if/when I start tackling
>> > other powerpc embedded families but for now I'm dealing with 4xx.
>>
>> Glad to see it.
>
> So my end game is to remove the #ifndef CONFIG_PPC around
> of_platform_default_populate_init(void). However, for that
> to work, I need to add a way to disable that on some platforms.
>
> However, it might take time, especially when it comes to dealing
> with the old Macs or some more obscure embedded platforms.
>
> So in the meantime, I'm thinking adding some kind of runtime way
> of disabling this default populate. Would you be ok with that ?
>
> Something like:
>
>         bool arch_wants_of_platform_defaults(void)
>
> With a weak implementation returning true.

That seems fine. The intention was that you can call
of_platform_default_populate multiple times with arch specific match
lists and the default. There could be ordering issues both with
initcall levels and device creation ordering.

Rob
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html