Re: [PATCH v1 2/8] dt-bindings: iio: dac: maxim,ds4424: add maxim,rfs-ohms property

[Jonathan Cameron <jic23@kernel.org>]

On Fri, 23 Jan 2026 11:03:38 +0100
Oleksij Rempel <o.rempel@pengutronix.de> wrote:

> On Fri, Jan 23, 2026 at 09:28:19AM +0000, Jonathan Cameron wrote:
> > On Mon, 19 Jan 2026 18:27:52 +0000
> > Conor Dooley <conor@kernel.org> wrote:
> >   
> > > On Mon, Jan 19, 2026 at 07:24:18PM +0100, Oleksij Rempel wrote:  
> > > > Introduce the 'maxim,rfs-ohms' property. The full-scale output current
> > > > of these DACs is determined by external resistors (Rfs) connected to
> > > > the FS pins. The driver requires the physical resistance values to
> > > > calculate the correct current scale (Amps per step) for the IIO
> > > > subsystem.
> > > > 
> > > > Keep it optional to avoid forcing updates of existing DTs; without it
> > > > the driver cannot derive a correct IIO scale.    
> > > 
> > > I don't really follow the logic here, if the driver doesn't work
> > > properly without it, shouldn't it be a required property even if that
> > > means existing devicetrees get new warnings? Warnings are preferable to
> > > the drivers malfunctioning on those devices, after all!  
> > 
> > Agreed. The driver can paper over holes or I guess we could provide a default
> > if the datasheet has some reference value or similar?    
> 
> No, this DAC controls the current. The current range is application specific and
> defined by the load resistors. Is it better to not provide scale if this
> information not available, guessing it will make things worse and
> potentially damage some HW.
> 
Then does this need to be a fix on the driver side?
Or are you just suggesting not proving the scale if we don't have enough info?
That's fine if so.

> > DT binding should require it.  
> 
> ACK