[Arm-netbook] EOMA-68 I/O Board start-up details?

[luke.leighton]

On Mon, Apr 1, 2013 at 8:08 PM, luke.leighton <luke.leighton at gmail.com> wrote:
> On Mon, Apr 1, 2013 at 7:40 PM, Scott Sullivan <scott at ss.org> wrote:
>> So Luke,
>>
>> At the moment details on the EOMA-68 start-up procedures are a little thin.
>> http://elinux.org/Embedded_Open_Modular_Architecture/EOMA-68#Start-up_procedure
>>
>> What information is available for those wishing to design their own I/O
>> boards ahead of or in parallel with the CPU card releases?
>
>  the only major bit missing is the I2C EEPROM address.

 0x70.


>> Format of Device Information? Binary or Plain Text?
>
>  haven't got that far yet.

 ok answer should have been: "A list of Linux Device Trees".

 the list will be length-separated groups.   i.e. length first, then
data.  length, data.

 also we'll need a revision number, a descriptive string (length then
text), then the number of device-tree entries.

 what i'm leaning towards is a PURPOSE-DEFINED device tree (with
associated driver) not a NUMERIC device tree for the GPIO.

 so, there would be a driver named "Reset Button" for example, and its
associated device tree would say "this manufacturer dedicated GPIO pin
N to the purpose of fulfilling the 'reset button' capability".

 or there would be a driver named "USB Hub Power-up" for example, and
its associated device tree would say "this manufacturer of this I/O
board dedicated GPIO pin N to the purpose of powering up the I/O
board's USB Hub".

 or there would be a driver named "External Interrupt from the
On-Board Embedded Controller".

 or, maybe, just have an enum (and descriptive string) in the same
header file in the same "GPIO Driver" source code, which goes into a
struct that has GPIO number, enum, I/O direction and initial state.

 l.