Hi,

On Mon, Jul 24, 2017 at 09:20:17PM +0200, Pavel Machek wrote:

On Fri 2017-07-14 12:01:49, Sebastian Reichel wrote:

quoted

Add DT binding document for PWM controlled vibrator devices.

Signed-off-by: Sebastian Reichel <redacted>

Acked-by: Pavel Machek <redacted>

quoted

index 000000000000..09145d18491d

--- /dev/null
+++ b/Documentation/devicetree/bindings/input/pwm-vibrator.txt

@@ -0,0 +1,66 @@
+* PWM vibrator device tree bindings
+
+Registers a PWM device as vibrator. It is expected, that the vibrator's
+strength increases based on the duty cycle of the enable PWM channel
+(100% duty cycle meaning strongest vibration, 0% meaning no vibration).
+
+The binding supports an optional direction PWM channel, that can be
+driven at fixed duty cycle. If available this is can be used to increase
+the vibration effect of some devices.

Actually what "direction" does would be nice to explain, because I
don't know. Does it make the motor turn the other way around?

Yes, at least that's how I understand it. IIUIC this will increase
the imbalance effect and thus the vibration.

quoted

+Required properties:
+- compatible: should contain "pwm-vibrator"

should->Should.

quoted

+- pwm-names: Should contain "enable" and optionally "direction"
+- pwms: Should contain a PWM handle for each entry in pwm-names
+
+Optional properties:
+- vcc-supply: Phandle for the regulator supplying power
+- direction-duty-cycle-ns: Duty cycle of the direction PWM channel in
+                           nanoseconds, defaults to 50% of the channel's
+			   period.

Is nanoseconds right unit here? It drives a motor...

Yes, for the driving the motor the values will be very big
and we could use milliseconds. But the PWM specifier [0] uses
nanoseconds for the period and I think period and duty cycle
should be described in the same scale.

[0] Documentation/devicetree/bindings/pwm/pwm.txt

-- Sebastian