On Wed, Aug 28, 2019 at 8:03 AM Michał Mirosław <mirq-linux@xxxxxxxxxxxx> wrote: > > On Tue, Aug 27, 2019 at 05:37:16PM -0500, Rob Herring wrote: > > On Sat, Aug 24, 2019 at 10:26:55PM +0200, Michał Mirosław wrote: > > > Add single-pin LRCLK source options for Atmel SSC module. > > > > > > Signed-off-by: Michał Mirosław <mirq-linux@xxxxxxxxxxxx> > > > > > > --- > > > v2: split from implementation patch > > > > > > --- > > > Documentation/devicetree/bindings/misc/atmel-ssc.txt | 5 +++++ > > > 1 file changed, 5 insertions(+) > > > > > > diff --git a/Documentation/devicetree/bindings/misc/atmel-ssc.txt b/Documentation/devicetree/bindings/misc/atmel-ssc.txt > > > index f9fb412642fe..c98e96dbec3a 100644 > > > --- a/Documentation/devicetree/bindings/misc/atmel-ssc.txt > > > +++ b/Documentation/devicetree/bindings/misc/atmel-ssc.txt > > > @@ -24,6 +24,11 @@ Optional properties: > > > this parameter to choose where the clock from. > > > - By default the clock is from TK pin, if the clock from RK pin, this > > > property is needed. > > > + - atmel,lrclk-from-tf-pin: bool property. > > > + - atmel,lrclk-from-rf-pin: bool property. > > > + - SSC in slave mode gets LRCLK from RF for receive and TF for transmit > > > + data direction. This property makes both use single TF (or RF) pin > > > + as LRCLK. At most one can be present. > > > > A single property taking 1 of possible 2 values would prevent the error > > of more than 1 property present. > > It still would need a validation check in the code, though: you > could put wrong value then. It seems more consistent with the > existing parameters to have two bool properties. It was validation using schema that I was thinking about. Expressing a possible set of values for a property is easier than inter-property constraints. But if you really prefer as-is: Reviewed-by: Rob Herring <robh@xxxxxxxxxx>
