Re: [PATCH] led: core: fix misleading comment after workqueue removal from drivers

[Jacek Anaszewski <j.anaszewski@samsung.com>]

Hi Heiner,

On 01/10/2016 06:16 PM, Heiner Kallweit wrote:
> Now that workqueues have been removed from individual drivers and
> were replaced with a core-internal workqueue we shouldn't
> encourage people to add new workqueues to drivers.
>
> Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
> ---
>   include/linux/leds.h | 5 ++++-
>   1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/include/linux/leds.h b/include/linux/leds.h
> index bc1476f..4429887 100644
> --- a/include/linux/leds.h
> +++ b/include/linux/leds.h
> @@ -50,7 +50,10 @@ struct led_classdev {
>   #define LED_DEV_CAP_FLASH	(1 << 23)
>
>   	/* Set LED brightness level */
> -	/* Must not sleep, use a workqueue if needed */
> +	/* Must not sleep. If no non-blocking version can be provided
> +	 * set brightness_set_blocking only. The LED core will use an
> +	 * internal work queue to create a non-blocking version.
> +	 */

This comment isn't easily comprehensible at first glance.

How about:

"Must not sleep. Use brightness_set_blocking for drivers
that can sleep while setting brightness."

>   	void		(*brightness_set)(struct led_classdev *led_cdev,
>   					  enum led_brightness brightness);
>   	/*
>


-- 
Best Regards,
Jacek Anaszewski