Re: [GIT PULL 2/2] second round of Rockchip dts32 changes for 4.14

[Heiko Stuebner <heiko@sntech.de>]

Am Donnerstag, 24. August 2017, 15:50:51 CEST schrieb Arnd Bergmann:
> On Thu, Aug 24, 2017 at 10:42 AM, Heiko Stuebner <heiko@sntech.de> wrote:
> > The following changes since commit 0f4dc7e154d054303c4dd6b5c3e3207f9a7f8714:
> >
> >   ARM: dts: rockchip: fix property-ordering in rv1108 mmc nodes (2017-07-22 22:41:35 +0200)
> >
> > are available in the git repository at:
> >
> >   git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip.git tags/v4.14-rockchip-dts32-2
> >
> > for you to fetch changes up to 274ff50d3ad951bca0adbfb9ffd628d35ba6db6f:
> >
> >   ARM: dts: rockchip: enable usb for rv1108-evb (2017-08-23 09:24:51 +0200)
> >
> > ----------------------------------------------------------------
> > A lot of attention for the rv1108 soc targetted at media-processing
> > (usb, operating points, spi, pwm, adc, watchdog, i2c and devices for
> > its evb).
> >
> > RK3228/3229 gets iommu and spi nodes. Similar to the rk3288 which
> > also gets some more iommu nodes as well as getting converted to 64
> > bit addresses due to wanting to address more than 4GB of memory
> > via LPAE.
> 
> Pulled into next/dt with a small fixup, see below. Please check.
> 
> Thanks!
> 
>       Arnd
> 
> 
> diff --cc arch/arm/boot/dts/rk3288.dtsi
> index 858e1fed762a,c0c04e99e159..356ed1e62452
> --- a/arch/arm/boot/dts/rk3288.dtsi
> +++ b/arch/arm/boot/dts/rk3288.dtsi
> @@@ -1126,9 -1145,27 +1145,27 @@@
>                 };
>         };
> 
> +       vpu_mmu: iommu@ff9a0800 {
> +               compatible = "rockchip,iommu";
> +               reg = <0x0 0xff9a0800 0x0 0x100>;
> +               interrupts = <GIC_SPI 11 IRQ_TYPE_LEVEL_HIGH>;
> +               interrupt-names = "vpu_mmu";
> +               #iommu-cells = <0>;
> +               status = "disabled";
> +       };
> +
> +       hevc_mmu: iommu@ff9c0440 {
> +               compatible = "rockchip,iommu";
> +               reg = <0x0 0xff9c0440 0x0 0x40>, <0x0 0xff9c0480 0x0 0x40>;
> +               interrupts = <GIC_SPI 111 IRQ_TYPE_LEVEL_HIGH>;
> +               interrupt-names = "hevc_mmu";
> +               #iommu-cells = <0>;
> +               status = "disabled";
> +       };
> +
>  -      gpu: mali@ffa30000 {
>  -              compatible = "rockchip,rk3288-mali", "arm,mali-t760",
> "arm,mali-midgard";
>  +      gpu: gpu@ffa30000 {
>  +              compatible = "rockchip,rk3288-mali", "arm,mali-t760";
> -               reg = <0xffa30000 0x10000>;
> +               reg = <0x0 0xffa30000 0x0 0x10000>;
>                 interrupts = <GIC_SPI 6 IRQ_TYPE_LEVEL_HIGH>,
>                              <GIC_SPI 7 IRQ_TYPE_LEVEL_HIGH>,
>                              <GIC_SPI 8 IRQ_TYPE_LEVEL_HIGH>;

yep looks good. Seems like the mali compatible-fix that was going into
an 4.13-rc interfered here.


Thanks
Heiko