[PATCH v4 01/15] arm64: dts: renesas: r8a7795: Add IPMMU device nodes
Simon Horman
horms at verge.net.au
Fri Nov 10 03:08:08 PST 2017
On Tue, Nov 07, 2017 at 10:11:56AM +0100, Geert Uytterhoeven wrote:
> Hi Simon,
>
> On Wed, Nov 1, 2017 at 11:34 AM, Simon Horman
> <horms+renesas at verge.net.au> wrote:
> > From: Magnus Damm <damm+renesas at opensource.se>
> >
> > Add r8a7795 IPMMU nodes and keep all disabled by default.
> >
> > This includes all IPMMU devices for r8a7795 ES2.0. Those
> > not present in r8a7795 ES1.x are removed from the DT for those
> > SoCs using delete-node. A follow-up patch will add IPMMU devices
> > to ES1.x which are not also present in ES2.0.
> >
> > Signed-off-by: Magnus Damm <damm+renesas at opensource.se>
> > Reviewed-by: Laurent Pinchart <laurent.pinchart at ideasonboard.com>
> > Signed-off-by: Simon Horman <horms+renesas at verge.net.au>
>
> Thanks for your patch!
>
> > --- a/arch/arm64/boot/dts/renesas/r8a7795-es1.dtsi
> > +++ b/arch/arm64/boot/dts/renesas/r8a7795-es1.dtsi
> > @@ -86,6 +91,22 @@
> > };
> > };
> >
> > +&ipmmu_vi0 {
> > + renesas,ipmmu-main = <&ipmmu_mm 11>;
> > +};
> > +
> > +&ipmmu_vp0 {
> > + renesas,ipmmu-main = <&ipmmu_mm 12>;
> > +};
> > +
> > +&ipmmu_vc0 {
> > + renesas,ipmmu-main = <&ipmmu_mm 12>;
>
> Should be 9.
Thanks, it looks like I made a copy-paste error.
>
> > +};
>
> Missing override:
>
> &ipmmu_vc1 {
> renesas,ipmmu-main = <&ipmmu_mm 10>;
> }
Thanks, added.
>
> > --- a/arch/arm64/boot/dts/renesas/r8a7795.dtsi
> > +++ b/arch/arm64/boot/dts/renesas/r8a7795.dtsi
> > @@ -421,6 +421,133 @@
> > resets = <&cpg 407>;
> > };
> >
> > + ipmmu_vi0: mmu at febd0000 {
> > + compatible = "renesas,ipmmu-r8a7795";
> > + reg = <0 0xfebd0000 0 0x1000>; /* IPMMU-VI0 */
>
> Given the label names, do we need comments like "IPMMU-VI0"?
Thanks, I will drop them.
>
> > + ipmmu_vp0: mmu at fe990000 {
> > + compatible = "renesas,ipmmu-r8a7795";
> > + reg = <0 0xfe990000 0 0x1000>; /* IPMMU-VP0 */
> > + renesas,ipmmu-main = <&ipmmu_mm 16>;
> > + #iommu-cells = <1>;
>
> According to Table 16.2, some IPMMU instances (e.g VP0) are part of a power
> domain. While I doubt the kernel code handles that correctly, it should be
> described in DT nevertheless.
Thanks, will do.
More information about the linux-arm-kernel
mailing list