[PATCH] ARM: shmobile: Use shmobile_init_late() on r8a7740

[Magnus Damm]

Hi Simon,

On Fri, May 16, 2014 at 1:45 PM, Simon Horman <horms at verge.net.au> wrote:
> On Fri, May 16, 2014 at 01:39:11PM +0900, Magnus Damm wrote:
>> Hi Simon,
>>
>> On Fri, May 16, 2014 at 1:08 PM, Simon Horman <horms at verge.net.au> wrote:
>> > On Tue, May 13, 2014 at 05:10:20PM +0900, Simon Horman wrote:
>> >> On Mon, May 12, 2014 at 09:42:15AM +0200, Geert Uytterhoeven wrote:
>> >> > On Mon, May 12, 2014 at 1:10 AM, Magnus Damm <magnus.damm at gmail.com> wrote:
>> >> > > From: Magnus Damm <damm+renesas at opensource.se>
>> >> > >
>> >> > > Hook up ->init_late for r8a7740 to initialize Suspend-to-RAM
>> >> > > and CPUIdle in case of C-code less board support for r8a7740.
>> >> > >
>> >> > > Signed-off-by: Magnus Damm <damm+renesas at opensource.se>
>> >> >
>> >> > Acked-by: Geert Uytterhoeven <geert at linux-m68k.org>
>> >>
>> >> Thanks, I will queue this up.
>> >
>> > Hi Magnus, Hi Geert,
>> >
>> > As I am currently doing some work for the r8a7779
>> > I am wondering if a change like this one is appropriate for that SoC.
>> > If so I'll cook one up.
>>
>> Yes, we want all SoCs to be converted over. Please take care that some
>> DTS change may be needed to specify maximum frequency, so this may
>> require a DTB update for the user. Which makes it less nice user
>> experience for legacy code - if we care about that at this point. But
>> legacy code also often tends to use early devices and other stuff, so
>> this probably affects DT SoC code and DT reference board code mostly.
>
> Does CPU frequency affect this patch (shmobile_init_late) as
> well as the ones to use shmobile_init_delay?

Oops, wrong patch from my side. =) Please ignore my comment.

But we still want the SoCs to be converted over.

/ magnus