Re: [PATCH RESEND v3 03/17] dt-bindings: iio: adc: ad7768-1: add trigger-sources property

[Rob Herring <robh@kernel.org>]

On Thu, Feb 27, 2025 at 06:30:04PM -0300, Jonathan Santos wrote:
> On 02/13, Conor Dooley wrote:
> > On Wed, Feb 12, 2025 at 03:16:16PM -0300, Jonathan Santos wrote:
> > > In addition to GPIO synchronization, The AD7768-1 also supports
> > > synchronization over SPI, which use is recommended when the GPIO
> > > cannot provide a pulse synchronous with the base MCLK signal. It
> > > consists of looping back the SYNC_OUT to the SYNC_IN pin and send
> > > a command via SPI to trigger the synchronization.
> > > 
> > > Add a new trigger-sources property to enable synchronization over SPI
> > > and future multiple devices support. This property references the
> > > main device (or trigger provider) responsible for generating the
> > > SYNC_OUT pulse to drive the SYNC_IN of device.
> > > 
> > > While at it, add description to the interrupts property.
> > > 
> > > Signed-off-by: Jonathan Santos <Jonathan.Santos@analog.com>
> > > ---
> > > v3 Changes:
> > > * Fixed dt-bindings errors.
> > > * Trigger-source is set as an alternative to sync-in-gpios, so we
> > >   don't break the previous ABI.
> > > * increased maxItems from trigger-sources to 2.
> > > 
> > > v2 Changes:
> > > * Patch added as replacement for adi,sync-in-spi patch.
> > > * addressed the request for a description to interrupts property.
> > > ---
> > >  .../bindings/iio/adc/adi,ad7768-1.yaml        | 28 +++++++++++++++++--
> > >  1 file changed, 25 insertions(+), 3 deletions(-)
> > > 
> > > diff --git a/Documentation/devicetree/bindings/iio/adc/adi,ad7768-1.yaml b/Documentation/devicetree/bindings/iio/adc/adi,ad7768-1.yaml
> > > index 3ce59d4d065f..4bcc9e20fab9 100644
> > > --- a/Documentation/devicetree/bindings/iio/adc/adi,ad7768-1.yaml
> > > +++ b/Documentation/devicetree/bindings/iio/adc/adi,ad7768-1.yaml
> > > @@ -26,7 +26,19 @@ properties:
> > >    clock-names:
> > >      const: mclk
> > >  
> > > +  trigger-sources:
> > > +    description:
> > > +      Specifies the device responsible for driving the synchronization pin,
> > > +      as an alternative to adi,sync-in-gpios. If the own device node is
> > > +      referenced, The synchronization over SPI is enabled and the SYNC_OUT
> > > +      output will drive the SYNC_IN pin.
> > 
> > Maybe a silly question, but why is self-reference needed here?
> > sync-in-gpios is a required property at present, so why can't you
> > operate under the assumption that neither the trigger-sources when
> > neither are present? Is it because only one of the sources could be
> > external and one internal, or there could be either one or two internal
> > sources? Self-referencing properties always feel like a bit of a
> > mistake.
> > 
> We have one internal synchronization source and two external ones.
> The internal is when /SYNC_OUT drives /SYNC_IN (in this case with self
> referencing). The external sources include using a GPIO to drive the 
> /SYNC_IN (sync-in-gpio case) or using an external /SYNC_OUT, the last one
> is described in the datasheet for multi-devices setup.
> 
> The trigger-sources property is inteded to indicate wheter the /SYNC_OUT
> signal originates from within the device or from another source. currently
> we are only handling the internal case. We are not sure yet how to handle
> the external sources.
> 
> > > +    maxItems: 2

From the above, I don't follow how you have 2 entries. In any case, you 
need to define what each entry is in the schema.

I'm also wondering if adi,sync-in-gpios should be instead represented in 
trigger-sources. This would mean we need a 'gpio-trigger' binding. That 
would make the handling of trigger sources more uniform. But maybe not 
worth the complexity.

Rob