Re: [PATCH 3/3] input: dt-bindings: Add binding for bma150 sensor

[Jonathan Bakker]

On 2019-01-25 5:28 p.m., Dmitry Torokhov wrote:
> On Fri, Jan 25, 2019 at 07:44:00PM +0100, PaweÅ Chmiel wrote:
>> From: Jonathan Bakker <xc-racer2@xxxxxxx>
>>
>> Add device tree bindings for Bosch BMA150 Accelerometer Sensor
>>
>> Signed-off-by: Jonathan Bakker <xc-racer2@xxxxxxx>
>> Signed-off-by: PaweÅ Chmiel <pawel.mikolaj.chmiel@xxxxxxxxx>
>> ---
>>  .../bindings/input/bosch,bma150.txt           | 20 +++++++++++++++++++
>>  1 file changed, 20 insertions(+)
>>  create mode 100644 Documentation/devicetree/bindings/input/bosch,bma150.txt
>>
>> diff --git a/Documentation/devicetree/bindings/input/bosch,bma150.txt b/Documentation/devicetree/bindings/input/bosch,bma150.txt
>> new file mode 100644
>> index 000000000000..290c60e38c70
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/input/bosch,bma150.txt
>> @@ -0,0 +1,20 @@
>> +* Bosch BMA150 Accelerometer Sensor
>> +
>> +Also works for the SMB380 and BMA023 accelerometers
>> +
>> +Required properties:
>> +- compatible : Should be "bosch,bma150"
>> +- reg : The I2C address of the sensor
>> +
>> +Optional properties:
>> +- interrupt-parent : should be the phandle for the interrupt controller
>> +- interrupts : Interrupt mapping for IRQ.  If not present device will be polled
>> +
>> +Example:
>> +
>> +bma150@38 {
>> +	compatible = "bosch,bma150";
>> +	reg = <0x38>;
>> +	interrupt-parent = <&gph0>;
>> +	interrupts = <1 IRQ_TYPE_LEVEL_HIGH>;
> 
> Hmm, here you say that IRQ_TYPE_LEVEL_HIGH, so it is level interrupts,
> but the driver overrides to rising edge unconditionally. Since you are
> the first to add DT support please make separate patch to driver to drop
> the ORQ trigger from request_theraded_irq() leaving only IRQF_ONESHOT.
> 
This was simply an oversight on my part, my device was using the the polled method as opposed to an interrupt.  I'll correct this in v2.
> Also please create patch removing platform data support as noone is
> using it upstream.
> 
Will do.
> What about the rest of config parameters from bma150_cfg? They should be
> handled as device properties too.
> 
Ok, I can add them as well.  I didn't bother as the comment in the source code says that the default values are the ones recommended by Bosch.
> Thanks.
> 
Thanks,
Jonathan