On Wed, Mar 19, 2014 at 02:03:21PM +0100, Alexandre Belloni wrote: > Set inversed polarity when .pwm_active_low is set in the platform_data. With > device tree, this is taken care of by of_pwm_xlate_with_flags(), called from > of_pwm_get(). > > Signed-off-by: Alexandre Belloni <alexandre.belloni@xxxxxxxxxxxxxxxxxx> > --- > drivers/video/backlight/pwm_bl.c | 8 ++++++++ > include/linux/pwm_backlight.h | 1 + > 2 files changed, 9 insertions(+) > > diff --git a/drivers/video/backlight/pwm_bl.c b/drivers/video/backlight/pwm_bl.c > index b75201ff46f6..ffdd3b2b2742 100644 > --- a/drivers/video/backlight/pwm_bl.c > +++ b/drivers/video/backlight/pwm_bl.c > @@ -309,6 +309,14 @@ static int pwm_backlight_probe(struct platform_device *pdev) > if (data->pwm_period_ns > 0) > pwm_set_period(pb->pwm, data->pwm_period_ns); > > + /* > + * The DT case is taking care of polarity in of_pwm_get(). For the > + * non-DT case, set the polarity from platform data. > + */ > + if (data->pwm_active_low) > + if (pwm_set_polarity(pb->pwm, PWM_POLARITY_INVERSED)) > + dev_err(&pdev->dev, "impossible to invert polarity\n"); Could this perhaps be solved by adding a field to the PWM lookup table that's used for non-DT cases? The idea being that we can use the same interfaces for DT and non-DT cases, so we should try to move both towards the same representation. Thierry
Attachment:
pgpt4mdPri2Oa.pgp
Description: PGP signature
