Re: [PATCH v1 10/10] core: platform: use fwnode_driver_match_device()

["Rafael J. Wysocki" <rjw@rjwysocki.net>]

On Friday, July 10, 2015 02:53:53 PM Tomeu Vizoso wrote:
> On 2 July 2015 at 01:46, Rafael J. Wysocki <rjw@rjwysocki.net> wrote:
> > On Tuesday, June 30, 2015 04:55:08 PM Tomeu Vizoso wrote:
> >> Instead of calling both of_driver_match_device() and
> >> acpi_driver_match_device(), call fwnode_driver_match_device() which
> >> should be able to sort out what firmware describes the device in
> >> question.
> >>
> >> Signed-off-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
> >
> > This one should be part of the other set IMO.
> 
> Yeah, I wasn't sure about it myself, but in any case the other series
> won't depend on this patch any more because I have moved the match
> delay into platform.c. I plan though to resend this with Rob's
> comments addressed in the next revision, so they can be taken if they
> are seen worthy in themselves.
> 
> > Or please just combine the two sets so it is more clear what the new helpers
> > are intended for.
> 
> Sorry about that, I did the split because Mark suggested so in the
> previous review round. I understand that depending on each person's
> workflow it may be more convenient in one way or the other, so I'm not
> sure I can find a way that fits everybody's preferences.

The ordering/form in which it is merged need not reflect the ordering/form
in which it is reviewed and it is much easier to follow changes having full
context available.

So you can split it later if need be, but for now let's keep them all together
for the completeness of the context.

Thanks,
Rafael