On Thu, Jun 08, 2017 at 02:12:02PM +0200, Thomas Petazzoni wrote:
> Hello,
>
> On Wed, 7 Jun 2017 17:33:17 -0500, Rob Herring wrote:
>
> > > +Example:
> > > +
> > > +icu: interrupt-controller@1e0000 {
> > > + compatible = "marvell,cp110-icu";
> > > + reg = <0x1e0000 0x10>;
> > > + #interrupt-cells = <3>;
> > > + interrupt-controller;
> > > + interrupt-parent = <&gic>;
> >
> > If you have a parent, then you should have some interrupts. I guess that
> > would be your ranges property? I suppose that is fine.
>
> The ranges of interrupts available is defined by the gicp node. Indeed,
> as explained in the cover letter:
>
> - We have one GICP in the SoC, providing a number of GIC SPI interrupts
>
> - We have one ICU per CP in the SoC. So for example in the Armada 8K,
> we have two CPs, and therefore two ICUs.
>
> So the range of available GIC SPI interrupts it not associated to each
> ICU, it's a global range of GIC SPI interrupts: each can freely be
> allocated by any of the ICUs in the system.
Okay, I guess I have no issues with this one.
Acked-by: Rob Herring <robh@xxxxxxxxxx>
--
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
