[PATCH] net: sh_eth: Add support of device tree probe

[Nobuhiro Iwamatsu]

Hi,

On Thu, Feb 7, 2013 at 10:42 AM, Kuninori Morimoto
<kuninori.morimoto.gx at renesas.com> wrote:
>
> Hi Iwamatsu-san
>
>> >> +Required properties:
>> >> +- compatible:                   "renesas,sh-eth";
>> >> +- interrupt-parent:             The phandle for the interrupt controller that
>> >> +                                services interrupts for this device.
>> >> +- reg:                          Offset and length of the register set for the
>> >> +                                device.
>> >> +- interrupts:                   Interrupt mapping for the sh_eth interrupt
>> >> +                                sources (vector id).
>> >> +- phy-mode:              String, operation mode of the PHY interface.
>> >> +- sh-eth,edmac-endian:          String, endian of sh_eth dmac.
>> >
>> > I believe endian is only little/big.
>> > It can be "bool" settings, not "string" ?
>> >
>>
>> If we use bool , we may use this as "sh-eth,edmac-endian-big" or
>> sh-eth,edmac-endian-little.
>> Or we define big-endian, and when using big-endian, we specify this.
>> Either way is okay.
>
> Sorry, if it means
>
> "we define big-endian as default settings,
> and user can select little-endian by using sh-eth,edmac-little-endian"
>
> then, it is nice.
> Then, it can be "Option" properties.

OK, I will update.

>
>> >> +- sh-eth,endian:                String, endian of sh_eth dmac.
>> >
>> > Is this really used it in this driver ?
>> >
>>
>> Yes. In many cases, this uses little-endian, but user may specify
>> big-endian, I think.
>
> I meant that I could find "sh-eth,edmac-endian" operation,
> but, couldn't find "sh-eth,endian" operation on sh_eth_parse_dt()
> and on Example either.
>

OK,I will remove this sentence.

Thanks,
  Nobuhiro
-- 
Nobuhiro Iwamatsu