[PATCHv4 1/2] drivers: input: keypad: Add device tree support

[Poddar, Sourav]

Hi Roland,

On Wed, Apr 25, 2012 at 5:08 PM, Roland Stigge <stigge at antcom.de> wrote:
> Hi Russell and Sourav,
>
> On 04/25/2012 12:52 PM, Russell King - ARM Linux wrote:
>> I think you and Roland Stigge need to put your heads together and come
>> up with a common DT binding for matrix keypad devices, rather than
>> inventing N different platform specific bindings.
>>
>>> +Required Board Specific Properties, in addition to those specified by
>>> +the shared matrix-keyboard bindings:
>>> +- keypad,num-rows: Number of row lines connected to the keypad
>>> +  controller.
>>> +
>>> +- keypad,num-columns: Number of column lines connected to the
>>> +  keypad controller.
>
> Right - the LPC32xx key scanner always has a square matrix design (#rows
> == #cols), but I'm fine with re-using properties "keypad,num-rows" and
> "keypad,num-columns" instead of "nxp,matrix-size".
>
> Besides, I still have "nxp,debounce-delay-ms" and "nxp,scan-delay-ms" -
> do I need to change this to some default properties I don't know of (yet)?
>
I think these are the custom properties which should be kept in your bindings.
I dont see anything of these sort in
Documentation/devicetree/bindings/input/matrix-keymap.txt .
> And: I re-used the DT subnodes strategy for the keycodes ("linux,code")
> of the samsung-keypad driver, see
> Documentation/devicetree/bindings/input/samsung-keypad.txt - can I
> continue using it?

Even I was using the bindings as used by samsung-keypad driver. But according
to the comments and logically too, we should use the common matrix keyboard
bindings already descibed in
Documentation/devicetree/bindings/input/matrix-keymap.txt.
So that we can get rid of multiple bindings for different platform.

~Sourav
> Thanks in advance,
>
> Roland