Hi Denis, On 3/16/20 9:24 PM, Denis Osterland-Heim wrote: > Hi Jacek, > > Am Montag, den 16.03.2020, 19:36 +0100 schrieb Jacek Anaszewski: >> Hi Denis, >> >> On 3/16/20 1:53 PM, Denis Osterland-Heim wrote: > ... >>> >>> @@ -92,13 +96,27 @@ static int led_pwm_add(struct device *dev, struct led_pwm_priv *priv, >>> >>> pwm_init_state(led_data->pwm, &led_data->pwmstate); >>> >>> + if (led->default_state == LEDS_PWM_DEFSTATE_ON) >>> + led_data->cdev.brightness = led->max_brightness; >>> + else if (led->default_state == LEDS_PWM_DEFSTATE_KEEP) { >>> + uint64_t brightness; >>> + >>> + pwm_get_state(led_data->pwm, &led_data->pwmstate); >> >> This seems to not be reading the device state, i.e. what you tried >> to address by direct call to pwm->chip->ops->get_state() before. >> >> Am I missing something? >> > > well, not you, but I missed cfc4c189bc70b1acc17e6f1abf1dc1c0ae890bd8. > Since this commit pwm_get_state() is sufficient. I assume you tested it? With that, for the whole set: Acked-by: Jacek Anaszewski <jacek.anaszewski@xxxxxxxxx> -- Best regards, Jacek Anaszewski