Hi Peter,

On Tue, 30 Sep 2014 19:12:22 +0200
Peter Korsgaard [off-list ref] wrote:

quoted

quoted

quoted

quoted

quoted

"Boris" == Boris Brezillon [off-list ref] writes:

 > Document DT bindings of Atmel's TRNG (True Random Number Generator) IP.
 > Signed-off-by: Boris Brezillon [off-list ref]
 > ---
 >  Documentation/devicetree/bindings/hwrng/atmel-trng.txt | 16 ++++++++++++++++
 >  1 file changed, 16 insertions(+)
 >  create mode 100644 Documentation/devicetree/bindings/hwrng/atmel-trng.txt

 > diff --git a/Documentation/devicetree/bindings/hwrng/atmel-trng.txt b/Documentation/devicetree/bindings/hwrng/atmel-trng.txt
 > new file mode 100644
 > index 0000000..4ac5aaa
 > --- /dev/null
 > +++ b/Documentation/devicetree/bindings/hwrng/atmel-trng.txt
 > @@ -0,0 +1,16 @@
 > +Atmel TRNG (True Random Number Generator) block
 > +
 > +Required properties:
 > +- compatible : Should be "atmel,at91sam9g45-trng"
 > +- reg : Offset and length of the register set of this block
 > +- interrupts : the interrupt number for the TRNG block

The interrupt isn't strictly speaking required as it isn't used by the
driver, but as 9G45 has the signal wired up I guess it makes sense.

This irq might be used in a near future (in order to wait for data as
requested by hwrng code when the wait argument is true).
Thus I prefer to keep it as required.

Best Regards,

Boris

-- 
Boris Brezillon, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com