Re: [PATCH 1/4] dt-bindings: iio: adc: adi,ad7124: fix clocks properties

["Rob Herring (Arm)" <robh@kernel.org>]

On Thu, 24 Jul 2025 18:25:22 -0500, David Lechner wrote:
> Use correct clocks properties for the AD7124 family of ADCs.
> 
> These ADCs have an internal clock along with an optional external clock
> that can be connected to the CLK pin. This pin can be wired up 3 ways:
> 1. Not connected - the internal clock is used.
> 2. Connected to an external clock (input) - the external clock is used.
> 3. Connected to the CLK pin on another ADC (output) - the internal clock
>    is used on one and the other is configured for an external clock.
> 
> The new bindings describe these 3 cases by picking one of the following:
> 1. Omit both clocks and #clock-cells properties.
> 2. Include only the clocks property with a phandle to the external clock.
> 3. Include only the #clock-cells property on the ADC providing the output.
> 
> The clock-names property is now deprecated and should not be used. The
> MCLK signal that it refers to is an internal counter in the ADC and
> therefore does not make sense as a devicetree property as it can't be
> connected to anything external to the ADC. Since there is only one
> possible external clock, the clock-names property is not needed anyway.
> Based on the implementation of the Linux driver, it looks like the
> "mclk" clock was basically being used as a control to select the power
> mode of the ADC, which is not something that should be done in the
> devicetree.
> 
> Signed-off-by: David Lechner <dlechner@baylibre.com>
> ---
>  .../devicetree/bindings/iio/adc/adi,ad7124.yaml     | 21 ++++++++++++++++-----
>  1 file changed, 16 insertions(+), 5 deletions(-)
> 

Reviewed-by: Rob Herring (Arm) <robh@kernel.org>