On Monday 02 April 2018 09:45 PM, David Lechner wrote: > On 04/02/2018 06:12 AM, Sekhar Nori wrote: >> On Friday 16 March 2018 10:50 PM, David Lechner wrote: >>> On 03/15/2018 09:52 PM, David Lechner wrote: >>>> This adds clock provider nodes for da850 and wires them up to all of >>>> the >>>> devices. >>>> >>>> Signed-off-by: David Lechner <david@xxxxxxxxxxxxxx> >>>> --- >>> >>> ... >>> >>> This is the mcasp0: mcasp@100000 node... >>> >>>> @@ -560,6 +720,7 @@ >>>> dmas = <&edma0 1 1>, >>>> <&edma0 0 1>; >>>> dma-names = "tx", "rx"; >>>> + clocks = <&psc1 7>; >>> >>> After some testing, it looks like it needs to be: >>> >>> + power-domains = <&psc1 7>; >>> >>> instead of >>> >>> + clocks = <&psc1 7>; >> >> We should probably have both clocks and power-domains properties for all >> PSC clocks. This way, the driver can change without a corresponding DT >> update dependency. >> >> Thanks, >> Sekhar >> > > That's fine with me. I just didn't know how people felt about using > properties > that are not documented. Good point. How to ease documentation for generic properties like these was discussed in the past, but there is no guidance in Documentation/devicetree/bindings that I can see. So, in the interest of reduced controversy, its probably better to do what you already have and only populate properties already documented in the bindings. Thanks, Sekhar -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
