[RFC,3/4] dt-bindings: iio: add support for GPIO triggers

Document Industrial I/O GPIO trigger support.

Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
---
 .../bindings/iio/trigger/iio-trig-gpio.txt         | 26 ++++++++++++++++++++++
 1 file changed, 26 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/iio/trigger/iio-trig-gpio.txt

On 17/02/17 16:03, Fabrice Gasnier wrote:
> Document Industrial I/O GPIO trigger support.
> 
> Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
Whilst I think this needs redoing for the interrupt trigger driver,
please be aware that bindings must be OS agnostic i.e. you can't
name things after linux specific subsystem.  We get a lot
of stick for the iio-hwmon bindings specifically for doing this!

Which makes this 'interesting' to describe.

Once we have it updated feedback from Rob and Mark will definitely be needed on
this one.

Jonathan
> ---
>  .../bindings/iio/trigger/iio-trig-gpio.txt         | 26 ++++++++++++++++++++++
>  1 file changed, 26 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/iio/trigger/iio-trig-gpio.txt
> 
> diff --git a/Documentation/devicetree/bindings/iio/trigger/iio-trig-gpio.txt b/Documentation/devicetree/bindings/iio/trigger/iio-trig-gpio.txt
> new file mode 100644
> index 0000000..a8dbf5c
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/iio/trigger/iio-trig-gpio.txt
> @@ -0,0 +1,26 @@
> +Industrial I/O - GPIO based trigger
> +
> +GPIOs may be used as IIO trigger provider, when trigger is connected to
> +GPIO lines.
> +
> +Required properties:
> +- compatible: Should be "iio-gpio-trigger"
> +- gpios: Should specify one GPIO line used as trigger source.
> +  See Documentation/devicetree/bindings/gpio/gpio.txt
> +- #io-trigger-cells: Should be 0, as simple trigger provider.
> +  See Documentation/devicetree/bindings/iio/iio-bindings.txt
> +
> +Optional properties:
> +- label: String to specifiy trigger name.
> +- gpio-trigger-rising-edge: A rising edge on GPIO will trigger.
> +- gpio-trigger-falling-edge: A falling edge on GPIO will trigger.
> +  Note: Both rising and falling edge may be used. In case none of
> +  rising or falling edge is selected, rising edge is selected by
> +  default.
> +
> +Example:
> +	gpiotrig0: iio-gpio-trigger0 {
> +		#io-trigger-cells = <0>;
> +		compatible = "iio-gpio-trigger";
> +		gpios = <&gpioa 11 0>;
> +	}
> 

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

On Fri, Feb 17, 2017 at 5:03 PM, Fabrice Gasnier <fabrice.gasnier@st.com> wrote:

> Document Industrial I/O GPIO trigger support.
>
> Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
(...)
> +- gpio-trigger-rising-edge: A rising edge on GPIO will trigger.
> +- gpio-trigger-falling-edge: A falling edge on GPIO will trigger.
> +  Note: Both rising and falling edge may be used. In case none of
> +  rising or falling edge is selected, rising edge is selected by
> +  default.

We have GPIO_ACTIVE*_HIGH and GPIO_ACTIVE_LOW in the
gpios second cell already.

Can't we simply use that and specify that for this usecase, reading
that cell as GPIO_ACTIVE_HIGH == trigger on rising edge and
GPIO_ACTIVE_LOW == trigger on falling edge?

Yours,
Linus Walleij
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

On 02/23/2017 04:03 PM, Linus Walleij wrote:
> On Fri, Feb 17, 2017 at 5:03 PM, Fabrice Gasnier <fabrice.gasnier@st.com> wrote:
>
>> Document Industrial I/O GPIO trigger support.
>>
>> Signed-off-by: Fabrice Gasnier <fabrice.gasnier@st.com>
> (...)
>> +- gpio-trigger-rising-edge: A rising edge on GPIO will trigger.
>> +- gpio-trigger-falling-edge: A falling edge on GPIO will trigger.
>> +  Note: Both rising and falling edge may be used. In case none of
>> +  rising or falling edge is selected, rising edge is selected by
>> +  default.
>
> We have GPIO_ACTIVE*_HIGH and GPIO_ACTIVE_LOW in the
> gpios second cell already.
>
> Can't we simply use that and specify that for this usecase, reading
> that cell as GPIO_ACTIVE_HIGH == trigger on rising edge and
> GPIO_ACTIVE_LOW == trigger on falling edge?

Hi Linus,

Thanks for reviewing,

I think I'll simply drop patch 3 & 4, as suggested by Lars and Jonathan,
on patch 4, to use interrupt binding directly. Then standard interrupt
binding can be used instead.
Only thing is to add OF support for iio_interrupt_trigger.

BR,
Fabrice
>
> Yours,
> Linus Walleij
>
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html