Re: [PATCH v6 4/6] dt-bindings: iio: Add adis16550 bindings

[Jonathan Cameron <jic23@kernel.org>]

On Wed, 5 Feb 2025 19:52:37 +0000
Conor Dooley <conor@kernel.org> wrote:

> On Wed, Feb 05, 2025 at 04:11:51PM +0000, Nuno Sá wrote:
> > On Tue, 2025-02-04 at 19:25 +0000, Conor Dooley wrote:  
> > > On Tue, Feb 04, 2025 at 04:36:08PM +0200, Robert Budai wrote:  
> > > > Document the ADIS16550 device devicetree bindings.
> > > > 
> > > > Co-developed-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
> > > > Signed-off-by: Antoniu Miclaus <antoniu.miclaus@analog.com>
> > > > Signed-off-by: Ramona Gradinariu <ramona.gradinariu@analog.com>
> > > > Signed-off-by: Robert Budai <robert.budai@analog.com>
> > > > ---
> > > > 
> > > > v6:
> > > > - applied blank line suggestions
> > > > - added clock-frequency dependency change suggestions
> > > > - yamllint corrections
> > > > 
> > > >  .../bindings/iio/imu/adi,adis16550.yaml       | 83 +++++++++++++++++++
> > > >  MAINTAINERS                                   |  9 ++
> > > >  2 files changed, 92 insertions(+)
> > > >  create mode 100644
> > > > Documentation/devicetree/bindings/iio/imu/adi,adis16550.yaml
> > > > 
> > > > diff --git a/Documentation/devicetree/bindings/iio/imu/adi,adis16550.yaml
> > > > b/Documentation/devicetree/bindings/iio/imu/adi,adis16550.yaml
> > > > new file mode 100644
> > > > index 000000000000..8750bb937979
> > > > --- /dev/null
> > > > +++ b/Documentation/devicetree/bindings/iio/imu/adi,adis16550.yaml
> > > > @@ -0,0 +1,83 @@
> > > > +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
> > > > +%YAML 1.2
> > > > +---
> > > > +$id: http://devicetree.org/schemas/iio/imu/adi,adis16550.yaml#
> > > > +$schema: http://devicetree.org/meta-schemas/core.yaml#
> > > > +
> > > > +title: Analog Devices ADIS16550 and similar IMUs
> > > > +
> > > > +maintainers:
> > > > +  - Nuno Sa <nuno.sa@analog.com>
> > > > +  - Ramona Gradinariu <ramona.gradinariu@analog.com>
> > > > +  - Antoniu Miclaus <antoniu.miclaus@analog.com>
> > > > +
> > > > +properties:
> > > > +  compatible:
> > > > +    enum:
> > > > +      - adi,adis16550
> > > > +
> > > > +  reg:
> > > > +    maxItems: 1
> > > > +
> > > > +  spi-cpha: true
> > > > +  spi-cpol: true
> > > > +
> > > > +  spi-max-frequency:
> > > > +    maximum: 15000000
> > > > +
> > > > +  vdd-supply: true
> > > > +
> > > > +  interrupts:
> > > > +    maxItems: 1
> > > > +
> > > > +  reset-gpios:
> > > > +    description:
> > > > +      Must be the device tree identifier of the RESET pin. If specified,
> > > > +      it will be asserted during driver probe. As the line is active low,
> > > > +      it should be marked GPIO_ACTIVE_LOW.
> > > > +    maxItems: 1
> > > > +
> > > > +  clocks:
> > > > +    description: If not provided, then the internal clock is used.
> > > > +    maxItems: 1
> > > > +
> > > > +  clock-frequency:
> > > > +    description: Clock frequency in Hz when an external clock is used.
> > > > +    oneOf:
> > > > +      - minimum: 1
> > > > +        maximum: 128
> > > > +      - minimum: 3000
> > > > +        maximum: 4500  
> > > 
> > > I don't get why this is a property, to be honest. When you've got an
> > > external clock, why isn't the frequency obtained from the clock provider
> > > node?
> > >   
> > 
> > The main purpose of this property is actually to show/document the constrains of
> > the external clock. We can very well just error out in the driver (and we do
> > that) and not have this property. I mentioned this property to Robert some
> > revisions ago and I also pointed out that I wasn't really sure if it should be
> > used or not (I guess this is more for fixed clock providers...). IIRC, I did
> > asked for some advice/comments but we got none so I assume Robert just decided
> > to use it and see what you guys had to say about it.  
> 
> NGL, this is one of the kinda of things where if you're relying on
> dt-bindings to avoid cocking up your board design, things have already
> gotten pretty badly wrong! That said, "clock-frequency" is a
> property for cpus, fixed-frequency clock providers and i2c buses, you'd
> need a vendor prefix and a unit suffix here IMO. Also, I don't really
> think that it actually does anything at all, given it does not constrain the
> clock you're linking to with the clocks property. This may as well just be
> a comment in the description of the clocks property, for all that it does.

I'd just drop it. 

Jonathan