Re: [PATCH v3] leds: provide helper to register "leds-gpio" devices

[Uwe Kleine-König]

Hi Andrew,

On Mon, May 09, 2011 at 03:02:54PM -0700, Andrew Morton wrote:
> On Mon, 11 Apr 2011 22:35:57 +0200
> Uwe Kleine-K__nig  <u.kleine-koenig@xxxxxxxxxxxxxx> wrote:
> 
> > This function makes a deep copy of the platform data to allow it to live
> > in init memory.
> > The definition cannot go into leds-gpio.c because it needs to be builtin
> > to be usable by platforms.
> > 
> 
> Well...  why?  The changelog tells us what the function does but
> provides no information explaining why you think it's needed, nor how
> it is expected to be used, etc.
> 
> Please send a complete and useful changelog!
OK, will try to do better for v4.

> > --- a/drivers/leds/Kconfig
> > +++ b/drivers/leds/Kconfig
> > @@ -14,6 +14,11 @@ config LEDS_CLASS
> >  	  This option enables the led sysfs class in /sys/class/leds.  You'll
> >  	  need this to do anything useful with LEDs.  If unsure, say N.
> >  
> > +config LED_REGISTER_GPIO
> > +	bool
> > +	help
> > +	  This option provides the function gpio_led_register_device.
> > +
> 
> Every other LEDS Kconfig symbol uses "LEDS", not "LED".  I'll make that
> change.
I used LED only to match led-register.c. There led- seemed reasonable to
me because it's only the drivers that use leds-, but there is
led-core.o, led-class.o and led-triggers.o.
Hmm, I don't care much.
 
> >  if NEW_LEDS
> >  
> >  comment "LED drivers"
> > [...]
> > --- a/include/linux/leds.h
> > +++ b/include/linux/leds.h
> > @@ -207,5 +207,17 @@ struct gpio_led_platform_data {
> >  					unsigned long *delay_off);
> >  };
> >  
> > +/**
> > + * gpio_led_register_device - register a gpio-led device
> > + * @pdata: the platform data used for the new device
> > + *
> > + * Use this function instead of platform_device_add()ing a static struct
> > + * platform_device.
> > + *
> > + * Note: as pdata and pdata->leds is copied these usually can and should be
> > + * __initdata.
> > + */
> > +struct platform_device *gpio_led_register_device(
> > +		int id, const struct gpio_led_platform_data *pdata);
> 
> It's unusual to document functions in the .h file.  There's a bit of
> precedent for that in leds.h, but there is more precedent in
> drivers/leds/*.c
> 
> Personally I think that putting the documentation in .h is rather sucky
> - it happens so rarely that one just doesn't think of looking in there.
> 
> The description isn't terribly useful if the reader doesn't know what
> "platform_device_add()ing a static struct platform_device" is for.  Can
> we describe this in some manner whcih doesn't refer to something else
> which is probably undocumented?
> 
> The comment doesn't document return values.
OK, will fix in v4.

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/