[PATCH] ARM: dts: imx53: add support for Ka-Ro TX53 modules

[Sascha Hauer]

On Fri, Dec 13, 2013 at 01:42:40PM +0100, Lothar Waßmann wrote:
> Hi,
> 
> Shawn Guo wrote:
> > On Fri, Dec 13, 2013 at 10:54:17AM +0100, Lothar Waßmann wrote:
> > > > > +		compatible = "pwm-backlight";
> > > > > +		pwms = <&pwm2 0 500000>;
> > > > > +		power-supply = <&reg_3v3>;
> > > > > +		brightness-levels = <
> > > > > +			100 99 98 97 96 95 94 93 92 91
> > > > > +			 90 89 88 87 86 85 84 83 82 81
> > > > > +			 80 79 78 77 76 75 74 73 72 71
> > > > > +			 70 69 68 67 66 65 64 63 62 61
> > > > > +			 60 59 58 57 56 55 54 53 52 51
> > > > > +			 50 49 48 47 46 45 44 43 42 41
> > > > > +			 40 39 38 37 36 35 34 33 32 31
> > > > > +			 30 29 28 27 26 25 24 23 22 21
> > > > > +			 20 19 18 17 16 15 14 13 12 11
> > > > > +			 10  9  8  7  6  5  4  3  2  1
> > > > > +			  0
> > > > 
> > > > Why is it so unique to start with 100 and end with 0?  Does it even
> > > > work?  Here is what I read from bindings doc.
> > > > 
> > > Yes, it works. The backlight control of the displays we typically use
> > > is active low. This is a poor man's way to get an inverted PWM output!
> > > (And one in which the number written to the sysfs file corresponds to
> > > the actual duty cycle of the PWM output)
> > 
> > I think Linux pwm driver should be improved for this case rather than
> > manipulate device tree in a way that violate the defined bindings.
> >
> I admit I would be happier with a 'polarity' flag like other PWM
> drivers provide it. I tried to add polarity support in the generic PWM
> layer, so that all HW drivers that do not support inversion in HW can
> make use of it, but this gives problems with client drivers that don't
> know about the inversion and disable the PWM when the duty cycle is set
> to 0.
> Maybe I'll find some time to find a proper solution for this issue.

Actually the i.MX supports inversion in hardware. It's bit 18 in the
PWMCR register.

As a sidenote IMO client drivers shouldn't make assumptions about the
pwm output when a pwm is disabled anyway. If clients want to have a
constant output they should set the duty cycle to 0 or 100%.

Sascha


-- 
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |