On Sat, Feb 12, 2011 at 21:15, Bill Gatliff wrote:
> --- a/drivers/pwm/Kconfig
> +++ b/drivers/pwm/Kconfig
> +config GPIO_PWM
> + Â Â Â tristate "GPIO+hrtimer PWM device emulation"
> + Â Â Â depends on GENERIC_PWM
> + Â Â Â help
> + Â Â Â Â This option enables code that emulates a single-channel
> + Â Â Â Â PWM device using a high-resolution timer and a GPIO
> + Â Â Â Â pin. ÂThe PWM framework allows you to create as many
> + Â Â Â Â of these devices as desired, subject to CPU overhead
> + Â Â Â Â and GPIO pin availability. ÂIf unsure, say N.
document the module name in the help ?
> --- /dev/null
> +++ b/drivers/pwm/gpio-pwm.c
> +#define DRIVER_NAME "gpio-pwm"
use KBUILD_MODNAME instead ?
> +static void gpio_pwm_work (struct work_struct *work)
no space before the "("
> +static struct pwm_device_ops gpio_pwm_device_ops = {
> +    .config     = gpio_pwm_config,
> + Â Â Â .config_nosleep = gpio_pwm_config_nosleep,
> +    .request    Â= gpio_pwm_request,
> +};
is this struct not constified ? same for some of the other structs in
this file ...
> + Â Â Â gp = kzalloc(sizeof(*gp), GFP_KERNEL);
> + Â Â Â if (IS_ERR_OR_NULL(gp))
i dont think the kmalloc funcs return a pointer error. it's either
NULL or it's a valid pointer. this comes up a few times ...
> +EXPORT_SYMBOL(gpio_pwm_create);
> +EXPORT_SYMBOL(gpio_pwm_destroy);
are people expected to call these directly ? is that useful if they
cant call any of the config funcs ?
-mike
--
To unsubscribe from this list: send the line "unsubscribe linux-embedded" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
