[PATCH 00/04] ARM: shmobile: r8a73a4 SoC support V3

[horms@verge.net.au (Simon Horman)]

On Wed, Mar 27, 2013 at 05:19:55PM +0900, Magnus Damm wrote:
> On Wed, Mar 27, 2013 at 12:46 PM, Simon Horman <horms@verge.net.au> wrote:
> > On Tue, Mar 26, 2013 at 11:17:43PM +0900, Magnus Damm wrote:
> >> On Tue, Mar 26, 2013 at 8:48 PM, Arnd Bergmann <arnd@arndb.de> wrote:
> >> > On Tuesday 26 March 2013, Magnus Damm wrote:
> >> >> ARM: shmobile: r8a73a4 SoC support V3
> >> >> [PATCH 01/04] ARM: shmobile: Initial r8a73a4 SoC support V3
> >> >> [PATCH 02/04] ARM: shmobile: r8a73a4 SCIF support V3
> >> >> [PATCH 03/04] ARM: shmobile: r8a73a4 IRQC support V2
> >> >> [PATCH 04/04] ARM: shmobile: r8a73a4 PFC support
> >> >>
> >> >> This series is V3 of initial support for the r8a73a4 SoC. The SoC
> >> >> is known as r8a73a4 which is built around Cortex-A15 and makes use
> >> >> of SCIF for serial, IRQC interrupt controller for external IRQs and
> >> >> PFC for combined support for pinmux and GPIO.
> >> >>
> >> >
> >> > Acked-by: Arnd Bergmann <arnd@arndb.de>
> >>
> >> Thanks for your help!
> >
> > Indeed.
> >
> > Magnus, please let me know if/when you would like me to queue-up these
> > patches.
> 
> Thanks. Since you bring up the topic, I'd like you to queue up this
> patch series more or less right away if possible. =)

Sure, I think it is possible.

> About how to handle the dependencies, I am not sure if you intend to
> keep the SoC bits separate from the PFC and the board code. It is
> possible to split it out by series but if that is needed or not
> depends on how the ARM SoC guys want to be fed.

I would prefer if they were separated out so that I can apply
them to separate branches, in keeping with arm-soc requirements.

The branches can depend on each other which is how any dependencies
can be handled.

At this point I am thinking of merging the soc branch, with this
series added, into the pinmux branch and then applying the r8a73a4 pfc
patches to the result.

> Regardless, this r8a73a4 SoC V3 series depends on the following patch
> to compile:
> [PATCH/RFC] ARM: shmobile: Disallow PINCTRL without GPIOLIB
> 
> Laurent seems OK with the patch above, how about you? I am quite OK
> with the above patch as a short term fix. But I believe Linus W and
> Laurent at some point need to decide if it is valid to have PINCTL
> selected without GPIO.

I think I will take it into my tree as this series relies on it.