Re: [RFC PATCH 1/3] dt-bindings: iio: adc: Add TI ADS1120 binding

[David Lechner <dlechner@baylibre.com>]

On 10/30/25 12:12 PM, Jonathan Cameron wrote:
> On Thu, 30 Oct 2025 22:04:09 +0530
> Ajith Anandhan <ajithanandhan0406@gmail.com> wrote:
> 
>> Add device tree binding documentation for the Texas Instruments
>> ADS1120.
>>
>> The binding defines required properties like compatible, reg, and
>> SPI configuration parameters.
>>
>> Link: https://www.ti.com/lit/gpn/ads1120
> Datasheet: https://www.ti.com/lit/gpn/ads1120
> 
> Is a somewhat official tag for these. Though better to put it in the dt-binding
> doc itself as well or instead of here.
> 
>> Signed-off-by: Ajith Anandhan <ajithanandhan0406@gmail.com>
>> ---
>>  .../bindings/iio/adc/ti,ads1120.yaml          | 50 +++++++++++++++++++
>>  1 file changed, 50 insertions(+)
>>  create mode 100644 Documentation/devicetree/bindings/iio/adc/ti,ads1120.yaml
>>
>> diff --git a/Documentation/devicetree/bindings/iio/adc/ti,ads1120.yaml b/Documentation/devicetree/bindings/iio/adc/ti,ads1120.yaml
>> new file mode 100644
>> index 000000000..09285c981
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/iio/adc/ti,ads1120.yaml
>> @@ -0,0 +1,50 @@
>> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
>> +%YAML 1.2
>> +---
>> +$id: http://devicetree.org/schemas/iio/adc/ti,ads1120.yaml#
>> +$schema: http://devicetree.org/meta-schemas/core.yaml#
>> +
>> +title: Texas Instruments ADS1120 4-channel, 16-bit, 2kSPS ADC
>> +
>> +maintainers:
>> +  - Ajith Anandhan <ajithanandhan0406@gmail.com>
>> +
>> +properties:
>> +  compatible:
>> +    const: ti,ads1120
>> +
>> +  reg:
>> +    maxItems: 1
>> +
>> +  spi-max-frequency:
>> +    maximum: 4000000
>> +
>> +  spi-cpha: true
>> +
>> +  "#io-channel-cells":
>> +    const: 1
> 
> Power supplies should be here and required (even if real boards
> rely on stub regulators). 
> 
> Looks like there is an optional reference as well - so include that
> but not as required (use internal ref if not supplied).

Actually, there are two. REF{P,N}1 is an alternative function of
the AIN{0,3} pins.

It is also possible that the analog power supply can be used as
a reference source instead of the internal one.

This came up recently and we glossed over it. However, I think
it would make sense to have a flag property that means "the
AVSS supply is of sufficient quality that it is better than 
the internal reference supply", e.g. ti,avdss-is-ref. And
drivers can use this info to decide if they want to select it
as the reference voltage or not.

> 
> There is a data ready pin as well so I'd expect an interrupt.

There is actually two DRDY pins. One is shared with DOUT, so we
should have two interrupts and interrupt-names so we know which
pin is actually wired up.

> 
> All these should be in the binding from the start as we want it
> to be as complete as possible.  The driver doesn't have to use everything
> the binding supplies.
> 
Another trivial one is an optional clocks property for the external
clock. It doesn't need clock-names since there is only one.

Additional bindings needed when this is used with temperature sensors
are not so trivial though, so we don't need to add those until someone
actually needs them.