Re: [PATCH] dt-bindings: iio: st-sensors: Add IIS2MDC magnetometer

[Conor Dooley <conor@kernel.org>]

[-- Attachment #1: Type: text/plain, Size: 2616 bytes --]

On Wed, Feb 21, 2024 at 08:45:18PM +0100, Marco Felsch wrote:
> Hi Conor,
> 
> On 24-02-21, Conor Dooley wrote:
> > On Wed, Feb 21, 2024 at 08:16:44PM +0100, Marco Felsch wrote:
> > > On 24-02-21, Conor Dooley wrote:
> > > > On Wed, Feb 21, 2024 at 06:58:10PM +0100, Marco Felsch wrote:
> > > > > Add the iis2mdc magnetometer support which is equivalent to the lis2mdl.
> > > > > 
> > > > > Signed-off-by: Marco Felsch <m.felsch@pengutronix.de>
> > > > > ---
> > > > >  Documentation/devicetree/bindings/iio/st,st-sensors.yaml | 1 +
> > > > >  1 file changed, 1 insertion(+)
> > > > > 
> > > > > diff --git a/Documentation/devicetree/bindings/iio/st,st-sensors.yaml b/Documentation/devicetree/bindings/iio/st,st-sensors.yaml
> > > > > index fff7e3d83a02..ee593c8bbb65 100644
> > > > > --- a/Documentation/devicetree/bindings/iio/st,st-sensors.yaml
> > > > > +++ b/Documentation/devicetree/bindings/iio/st,st-sensors.yaml
> > > > > @@ -64,6 +64,7 @@ properties:
> > > > >            - st,lsm9ds0-gyro
> > > > >        - description: STMicroelectronics Magnetometers
> > > > >          enum:
> > > > > +          - st,iis2mdc
> > > > 
> > > > Without a fallback compatible to the equivilent device, how does a
> > > > driver bind to this device?
> > > 
> > > I skimed the datasheets and the driver already handles this binding
> > > exactly the same as the st,lis2mdl, so my assumption is they do match.
> > > 
> > > Why do I you think we need a fallback compatible here?
> > 
> > I didn't look at the driver, there was no mention of the driver already
> > having (undocumented) support for it. Since there was no driver change
> > alongside this patch, I thought you'd need a fallback compatible to
> > allow the driver to match against a compatible it recognises.
> 
> I explicitly did not mention the driver in the commit message else I
> would have got a response like "dt-bindings have no dependency to
> drivers" ;)

Putting it under the --- line is always an option. Where there are
existing users but the compatible is just undocumented, this it's
helpful to do.

> > Besides, having fallback compatibles is the norm when one device has the
> > same programming model as another.
> 
> Not for this binding according the driver.

If they don't have the same programming model, then describing them as
"equivalent" wouldn't be correct. That said, they seem to use the same
sensor settings when alls said and done (see st_magn_sensors_settings),
so I think they are actually compatible even if the driver has separate
match data for each.

Cheers,
Conor.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]