Message-ID: <20230225171723.15e822ec@jic23-huawei>
Date:   Sat, 25 Feb 2023 17:17:23 +0000
From:   Jonathan Cameron <jic23@...nel.org>
To:     Krzysztof Kozlowski <krzk@...nel.org>
Cc:     marius.cristea@...rochip.com, lars@...afoo.de, robh+dt@...nel.org,
        linux-iio@...r.kernel.org, devicetree@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v1 1/2] dt-bindings: iio: adc: adding dt-bindings for
 PAC193X

Hi Marius

Welcome to IIO.

I'll comment on top of Krzystof's review to avoid duplication.


> > 
> > Signed-off-by: Marius Cristea <marius.cristea@...rochip.com>
> > ---
> >  .../bindings/iio/adc/microchip,pac193x.yaml   | 122 ++++++++++++++++++

Wild cards in names bite us far too often.  Name this and the driver after one of
the supported parts.

> >  1 file changed, 122 insertions(+)
> >  create mode 100644 Documentation/devicetree/bindings/iio/adc/microchip,pac193x.yaml
> > 
> > diff --git a/Documentation/devicetree/bindings/iio/adc/microchip,pac193x.yaml b/Documentation/devicetree/bindings/iio/adc/microchip,pac193x.yaml
> > new file mode 100644
> > index 000000000000..e4ea560991e9
> > --- /dev/null
> > +++ b/Documentation/devicetree/bindings/iio/adc/microchip,pac193x.yaml
> > @@ -0,0 +1,122 @@
> > +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
> > +%YAML 1.2
> > +---
> > +$id: http://devicetree.org/schemas/iio/adc/microchip,pac193x.yaml#
> > +$schema: http://devicetree.org/meta-schemas/core.yaml#
> > +
> > +title: Microchip PAC193X Power Monitors with Accumulator
> > +
> > +maintainers:
> > +  - Marius Cristea <marius.cristea@...rochip.com>
> > +
> > +properties:
> > +  compatible:
> > +    enum:
> > +      - microchip,pac1934
> > +      - microchip,pac1933
> > +      - microchip,pac1932
> > +      - microchip,pac1931  
> 
> Maybe order these numerically (ascending)?
> 
> > +
> > +  reg:
> > +    maxItems: 1
> > +
> > +  "#address-cells":
> > +    const: 1
> > +
> > +  "#size-cells":
> > +    const: 0
> > +
> > +  microchip,samp-rate:
> > +    $ref: /schemas/types.yaml#/definitions/uint32
> > +    description: Sampling rate for all device's channels.  
> 
> What are the units? rate is usually in hz, which should be expressed in
> unit suffix (property name)]

It's unusual for sampling rate to be a property of the hardware and hence
suitable for DT binding. Normally we make this a userspace control instead.
If there is a reason for doing it from DT, that wants to be mentioned here.

> 
> > +    enum: [8, 64, 256, 1024]
> > +    default: 1024
> > +
> > +required:  
> 
> required goes after all properties, so after patternProperties
> 
> > +  - compatible
> > +  - reg
> > +  - microchip,samp-rate
> > +  - "#address-cells"
> > +  - "#size-cells"
> > +
> > +patternProperties:
> > +  "^channel([1-4])@[1-4]+$":  
> 
> Why first reg-pattern ([1-4])? For what do you need it?
> 
> > +    $ref: "adc.yaml"  
> 
> Drop quotes
> 
> > +    type: object
> > +    description: Represents the external channels which are connected to the ADC.
> > +
> > +    properties:
> > +      reg:
> > +        description: The channel number.
> > +          It can have up to 4 channels, numbered from 1 to 4.
> > +        items:
> > +          - minimum: 1
> > +            maximum: 4
> > +
> > +      microchip,uohms-shunt-res:  
> 
> Not a correct unit suffix. Use standard properties:
> git grep shunt -- Documentation/devicetree/

Check for existing relevant properties. Here we have
shunt-resistor-micro-ohms in common use in both IIO and hwmon.


> 
> 
> > +        $ref: /schemas/types.yaml#/definitions/uint32
> > +        description: |
> > +          Value in micro Ohms of the shunt resistor connected between
> > +          the SENSE+ and SENSE- inputs, across which the current is measured. Value
> > +          is needed to compute the scaling of the measured current.
> > +
> > +      microchip,rail-name:
> > +        $ref: /schemas/types.yaml#/definitions/string
> > +        description: Name of the monitored power rail.  
> 
> Why do you need it? Why this is a property of DT? Aren't you now
> duplicating label?


Agreed. Sounds like label to me as well.

> 
> > +
> > +      microchip,bi-directional:
> > +        description: Whether the channel is bi-directional.  
> 
> Describe here what is a "bi-directional" channel for ADC... Maybe it is
> obvious, maybe not. For me it is not and none of other devices have it.
> 
From datasheet this seems to be a novel description of whether the ADC
is measuring just positive or allows for negative voltage as well (i.e.
the current through the shut resistor is going on the other direction).
Can we relate this to a biopolar / unipolar decision on a simple voltage ADC?
For that we have standard DT bindings in iio/adc/adc.yaml


> > +        type: boolean
> > +
> > +    required:
> > +      - reg
> > +      - microchip,uohms-shunt-res
> > +      - microchip,rail-name
> > +
> > +    additionalProperties: false
> > +
> > +additionalProperties: false
> > +
> > +examples:
> > +  - |
> > +    i2c {
> > +        #address-cells = <1>;
> > +        #size-cells = <0>;
> > +
> > +        pac193x: pac193x@10 {  
> 
> Node names should be generic.
> https://devicetree-specification.readthedocs.io/en/latest/chapter2-devicetree-basics.html#generic-names-recommendation
> 
> > +            compatible = "microchip,pac1934";
> > +            reg = <0x10>;
> > +
> > +            #address-cells = <1>;
> > +            #size-cells = <0>;
> > +            microchip,samp-rate = <64>;
> > +
> > +            channel1@1 {  
> 
> 1@1 is for sure not generic...

Standard naming for channel nodes in adc.yaml
Some bindings predate that, but for new ones please match it.

Thanks,

Jonathan



> 
> 
> 
> Best regards,
> Krzysztof
> 

Powered by blists - more mailing lists