On Thu, Nov 8, 2012 at 9:34 AM, Péter Ujfalusi <peter.ujfalusi@xxxxxx> wrote:
> On 11/07/2012 07:12 PM, Grazvydas Ignotas wrote:
>>> +static int twl4030_pwmled_config(struct pwm_chip *chip, struct pwm_device *pwm,
>>> + int duty_ns, int period_ns)
>>> +{
>>> + int duty_cycle = (duty_ns * TWL4030_LED_MAX) / period_ns;
>>> + u8 on_time;
>>> + u8 pwm_config[2];
>>> + int base, ret;
>>> +
>>> + if (duty_cycle >= TWL4030_LED_MAX)
>>> + on_time = TWL4030_LED_MAX;
>>> + else if (!duty_cycle)
>>> + on_time = TWL4030_LED_MAX - 1;
>>> + else
>>> + on_time = TWL4030_LED_MAX - duty_cycle;
>>> +
>>> + base = pwm->hwpwm * 2 + TWL4030_PWMA_REG;
>>> +
>>> + pwm_config[0] = on_time;
>>> + pwm_config[1] = TWL4030_LED_MAX;
>>> +
>>> + ret = twl_i2c_write(TWL4030_MODULE_LED, pwm_config, base, 2);
>>
>> Shouldn't this use TWL4030_MODULE_PWMA and TWL4030_MODULE_PWMB as
>> first argument? I can guess it works your way too, but
>> TWL4030_MODULE_PWMx would match the TRM better.
>
> I don't have strong opinion regarding to this. You mean changing from:
>
> base = pwm->hwpwm * 2 + TWL4030_PWMA_REG;
> ret = twl_i2c_write(TWL4030_MODULE_LED, pwm_config, base, 2);
>
> to:
>
> if (pwm->hwpwm)
> module = TWL4030_MODULE_PWMB;
> else
> module = TWL4030_MODULE_PWMA;
> ret = twl_i2c_write(module, pwm_config, 0, 2);
>
> But I want to note that I'm currently trying to clean up the mess around
> twl-core. In my view we have quite a bit of redundancy in there. The PWM A/B
> is for driving the LED A/B outputs. We should have only these modules for
> PWM/LED in twl-core:
> TWL_MODULE_PWM - offset for PWM0ON register in twl4030 and PWM1ON for twl6030
> TWL_MODULE_LED - offset for LEDEN register in twl4030 and LED_PWM_CTRL1
> for twl6030
>
> From here the driver can figure out what to do IMHO.
>
> There's no need to have separate TWL 'modules' for:
> TWL4030_BASEADD_PWM0
> TWL4030_BASEADD_PWM1
> TWL4030_BASEADD_PWMA
> TWL4030_BASEADD_PWMB
Well all these seem to come from TRM, no hard feelings here too but if
you are going to remove them, probably worth adding a comment.
--
Gražvydas
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
