Quoting Mike Looijmans (2018-03-19 08:04:15)
> On 18-03-18 13:52, Rob Herring wrote:
> > On Thu, Mar 15, 2018 at 12:50:08PM +0100, Mike Looijmans wrote:
> >> +Required properties:
> >> + - compatible: One of "silabs,si514a", "silabs,si514b" "silabs,si514c" according
> >> + to the speed grade of the chip.
> >> + - reg: I2C device address.
> >> + - #clock-cells: From common clock bindings: Shall be 0.
> >> +
> >> +Optional properties:
> >> + - clock-output-names: From common clock bindings. Recommended to be "si544".
> >
> > What's the point if there is only 1 clock output?
>
> I have a board with 6 of these chips, and the clock registration fails if they
> don't have unique names. Providing a clock-output-names property was the
> easiest way around it.
>
> Should I just omit the line or change the wording to reflect the above or is
> there a better solution (maybe I can use the DT node name as clock name in the
> driver, will look into that)
I would leave clock-output-names for now. One day we can get off of
strings in CCF. Soon perhaps.
>
> >
> >> +
> >> +Example:
> >> + si544: clock-generator@55 {
> >
> > clock-controller@55 is the standard node name.
>
> I'll change and post patch v3
>
I made some comments on v1. Please fold those into v3.
--
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
