ucb1400_gpio driver Kconfig entry

ucb1400_gpio driver Kconfig entry

[Ferenc Wagner]

Hi Marek,

drivers/gpio/Kconfig says:

config GPIO_UCB1400
        bool "Philips UCB1400 GPIO"
        depends on UCB1400_CORE
        help
          This enables support for the Philips UCB1400 GPIO pins.
          The UCB1400 is an AC97 audio codec.

          To compile this driver as a module, choose M here: the
          module will be called ucb1400_gpio.

But you use bool above, so choosing M isn't actually an option.
Maybe you want to resolve this inconsistency.
-- 
Regards,
Feri.

Re: ucb1400_gpio driver Kconfig entry

[Marek Vasut]

On Tuesday 01 February 2011 14:05:13 Ferenc Wagner wrote:
> Hi Marek,
> 
> drivers/gpio/Kconfig says:
> 
> config GPIO_UCB1400
>         bool "Philips UCB1400 GPIO"
>         depends on UCB1400_CORE
>         help
>           This enables support for the Philips UCB1400 GPIO pins.
>           The UCB1400 is an AC97 audio codec.
> 
>           To compile this driver as a module, choose M here: the
>           module will be called ucb1400_gpio.
> 
> But you use bool above, so choosing M isn't actually an option.
> Maybe you want to resolve this inconsistency.

I don't really have any device with ucb1400 available now. I will have some 
probably sometimes soon though. On the other hand, if you know it's a correct 
fix, you can send a patch.

Also, CC linux-input please.

Cheers

Re: ucb1400_gpio driver Kconfig entry

[Ferenc Wagner]

Marek Vasut <marek.vasut@gmail.com> writes:

> On Tuesday 01 February 2011 14:05:13 Ferenc Wagner wrote:
> 
>> drivers/gpio/Kconfig says:
>> 
>> config GPIO_UCB1400
>>         bool "Philips UCB1400 GPIO"
>>         depends on UCB1400_CORE
>>         help
>>           This enables support for the Philips UCB1400 GPIO pins.
>>           The UCB1400 is an AC97 audio codec.
>> 
>>           To compile this driver as a module, choose M here: the
>>           module will be called ucb1400_gpio.
>> 
>> But you use bool above, so choosing M isn't actually an option.
>> Maybe you want to resolve this inconsistency.
>
> I don't really have any device with ucb1400 available now. I will have
> some probably sometimes soon though. On the other hand, if you know
> it's a correct fix, you can send a patch.

This note hasn't got much to do with the driver code itself.  Either you
want to enable compiling the driver as a module, then use tristate
instead of bool; or it isn't an option and then remove the corresponding
paragraph from the documentation.  The latter is improbable, and a
quick look at the code seems to confirm that the driver was actually
intended to be used as a module.  So I suggest replacing bool with
tristate.  But I don't know anything at all about this driver or device.
-- 
Regards,
Feri.

Re: ucb1400_gpio driver Kconfig entry

[Marek Vasut]

On Tuesday 01 February 2011 18:26:24 Ferenc Wagner wrote:
> Marek Vasut <marek.vasut@gmail.com> writes:
> > On Tuesday 01 February 2011 14:05:13 Ferenc Wagner wrote:
> >> drivers/gpio/Kconfig says:
> >> 
> >> config GPIO_UCB1400
> >> 
> >>         bool "Philips UCB1400 GPIO"
> >>         depends on UCB1400_CORE
> >>         help
> >>         
> >>           This enables support for the Philips UCB1400 GPIO pins.
> >>           The UCB1400 is an AC97 audio codec.
> >>           
> >>           To compile this driver as a module, choose M here: the
> >>           module will be called ucb1400_gpio.
> >> 
> >> But you use bool above, so choosing M isn't actually an option.
> >> Maybe you want to resolve this inconsistency.
> > 
> > I don't really have any device with ucb1400 available now. I will have
> > some probably sometimes soon though. On the other hand, if you know
> > it's a correct fix, you can send a patch.
> 
> This note hasn't got much to do with the driver code itself.  Either you
> want to enable compiling the driver as a module, then use tristate
> instead of bool; or it isn't an option and then remove the corresponding
> paragraph from the documentation.  The latter is improbable, and a
> quick look at the code seems to confirm that the driver was actually
> intended to be used as a module.  So I suggest replacing bool with
> tristate.  But I don't know anything at all about this driver or device.

I know what this thing is -- it's a MFD chip with audio/touchscreen/gpio on it. 
I'll get to a device with such thing soon, so don't worry. I'll keep this in 
ToDo. I think it should be possible to compile this as a module btw.

Cheers