Re: awful kconfig help texts.

[Lars-Peter Clausen <lars@metafoo.de>]

On 08/01/2012 12:04 PM, Borislav Petkov wrote:
> On Wed, Aug 01, 2012 at 11:38:16AM +0200, Lars-Peter Clausen wrote:
>> You don't see any drivers, because the subsystem is still young and no
>> such arch independent drivers have been added yet, but they will get
>> added in the future. The arch independent companion or PWM expander
>> chips usually interface via I2S or SPI and I would consider it quite
>> likely that you'll also find them on some embedded X86 boards. If we
>> add a arch restriction to the config item now we'd quite likely have
>> to remove it again in the next release.
> 
> Yes please.
> 
> Kconfig is overcrowded as it is now and adding yet another option which
> is irrelevant for some arches (for now, as you say) simply causes
> confusion to people with absolutely no gain.
> 
> Simply take a look at all arch/<archname>/Kconfig files and look at all
> the "select ..." statements right at the beginning of the respective
> Kconfig file.

Yes and these select statements make sense, what you suggest though doesn't,
at least from my point of view. What you want is that you don't get
presented the option to select the PWM system if there no PWM driver
available based on your other config options. But whether a PWM driver is
available or not is not a issue of which arch you are building for.

You could do that by using a construct like below, but well...

config HAS_PWM
    bool

config PWM
    bool "PWM
    depends on HAS_PWM

config PWM_CAN_BUILD_DRIVER_X
    defbool I2C
    select HAS_PWM

config PWM_DRIVER_X
    tristate "PWM chip X support"
    depends on PWM && PWM_CAN_BUILD_DRIVER_X

config PWM_CAN_BUILD_DRIVER_Y
    defbool ARCH_Y
    select HAS_PWM

config PWM_DRIVER_Y
    tristate "PWM chip Y support"
    depends on PWM && PWM_CAN_BUILD_DRIVER_Y