Re: FEC_MPC52xx_MDIO oddness.

[Gabriel C <nix.or.die@googlemail.com>]

Domen Puncer wrote:
> On 31/10/07 17:30 +0100, Gabriel C wrote:
>> Domen Puncer wrote:
>>> On 31/10/07 10:22 -0400, Dave Jones wrote:
>>>> The Kconfig for this reads..
>>>>
>>>> +config FEC_MPC52xx_MDIO
>>>> +       bool "MPC52xx FEC MDIO bus driver"
>>>> +       depends on FEC_MPC52xx
>>>> +       default y
>>>> +       ---help---
>>>> +         The MPC5200's FEC can connect to the Ethernet either with
>>>> +         an external MII PHY chip or 10 Mbps 7-wire interface
>>>> +         (Motorola? industry standard).
>>>> +         If your board uses an external PHY connected to FEC, enable this.
>>>> +         If not sure, enable.
>>>> +         If compiled as module, it will be called 'fec_mpc52xx_phy.ko'.
>>>>
>>>> Because this is bool, it'll never be compiled as a module.
>>>> Given that fec_mpc52xx_phy.c has a MODULE_LICENSE()
>>>> Should this be a tristate perhaps ?
>>> It's ok, it will be compiled as module if fec driver is compiled as
>>> module, and built-in if fec is built-in.
>>>
>>> from Makefile:
>>> obj-$(CONFIG_FEC_MPC52xx) += fec_mpc52xx.o
>>> ifeq ($(CONFIG_FEC_MPC52xx_MDIO),y)
>>>         obj-$(CONFIG_FEC_MPC52xx) += fec_mpc52xx_phy.o
>>> endif
>>>
>>> Yes, a bit weird, but it's best I could come up with.
>>> (fec_mdio=m and fec=y is an invalid case, that I wanted to avoid).
>> Could you please fix this warnings as well ?
>>
>>
>> http://groups.google.com/group/fa.linux.kernel/browse_thread/thread/13c9d93a7afa60e2
> 
> I saw the bug reports, but unfortunately don't have a smart idea for
> the solution (moving the driver into a separate Kconfig might do it),
> so I'm waiting for Sam's input on this.

Added Sam and Adrian to CC , maybe Adrian knows some trick for this select stuff as well.