[U-Boot] [PATCH v2 1/2] gpio: Add GPIO driver framework for Marvell SoCs

[Albert ARIBAUD]

Hi Simon,

On 04/08/2011 02:04, Simon Guinot wrote:
> Hi Ajay,
>
> On Wed, Aug 03, 2011 at 10:10:00AM +0530, Ajay Bhargav wrote:
>> ----- "Simon Guinot"<simon at sequanux.org>  wrote:
>>
>>> AFAIK, Orion and Kirkwood SoCs don't provide bitwise set/clear for
>>> GPIO output/direction registers. Instead, a register must be read
>>> first to leave other bits unchanged (see __set_direction in
>>> kw_gpio.c).
>>>
>>> Is it possible to handle Armada SoCs GPIOs in a same way ? maybe
>>> using
>>> the pin registers (gpxx in the Armada struct gpio_reg array) ?
>>>
>>> If not, this code is not Marvell generic but rather specific for
>>> Armada
>>> SoCs and then maybe armada_gpio is a better name...
>>>
>>> Regards,
>>>
>>> Simon
>>
>> Hi Simon,
>>
>> Yes its possible to implement code that way, Armada SoC does have GPIO
>> registers for set/clear. what about register naming?? I think they are
>> different for Kirkwood and Orion.
>
> I think that the register names could be OK. But here is a most
> important problem: On Orion/Kirkwood SoCs, a single GPIO output register
> is available (no set/clear variants as for Armada). I missed that point
> at my first look. It is quite problematic because only two registers are
> shared between the different Marvell SoCs: level and direction. In fact,
> this registers are probably relevant on every machines providing GPIOs...
>
> Maybe that having two common registers is not enough to add
> Orion/Kirkwood support to the mvgpio driver ?
 >
>> One more thing which can be done to make this code generic is to have
>> some macros which can be defined by individual arch for specific registers
>> which are going to be in use e.g.
>>
>> #define GPIO_PIN_LEVEL_REG
>> #define GPIO_DIR_REG
>> #define GPIO_PIN_SET_REG
>> #define GPIO_PIN_CLR_REG
>
> Yes, but how to handle both a single GPI0 output register and some GPIO
> {set,clear} output registers (in a nice way) ?

Two distinct gipo drivers for the two marvell variants?

Or a single driver with a single API but implemented differently 
according to the presence / absence of a variant flag, e.g. 
CONFIG_MVGIPO_HAS_SET_AND_CLR_REGS?

> Regards,
>
> Simon

Amicalement,
-- 
Albert.