Re: [PATCH v5 4/7] riscv: dts: renesas: Add initial devicetree for Renesas RZ/Five SoC

[Conor Dooley]

On Sun, Oct 30, 2022 at 08:02:10AM +0800, Guo Ren wrote:
> On Sun, Oct 30, 2022 at 3:11 AM Lad, Prabhakar
> <prabhakar.csengg@xxxxxxxxx> wrote:
> >
> > Hi Guo,
> >
> > Thank you for the review.
> >
> > On Sat, Oct 29, 2022 at 5:25 AM Guo Ren <guoren@xxxxxxxxxx> wrote:
> > >
> > > On Sat, Oct 29, 2022 at 12:59 AM Prabhakar <prabhakar.csengg@xxxxxxxxx> wrote:
> > > >
> > > > From: Lad Prabhakar <prabhakar.mahadev-lad.rj@xxxxxxxxxxxxxx>
> > > >
> > > > Add initial device tree for Renesas RZ/Five RISC-V CPU Core (AX45MP
> > > > Single).
> > > >
> > > > RZ/Five SoC is almost identical to RZ/G2UL Type-1 SoC (ARM64) hence we
> > > > will be reusing r9a07g043.dtsi [0] as a base DTSI for both the SoC's.
> > > > r9a07g043f.dtsi includes RZ/Five SoC specific blocks.
> > > >
> > > > Below are the RZ/Five SoC specific blocks added in the initial DTSI which
> > > > can be used to boot via initramfs on RZ/Five SMARC EVK:
> > > > - AX45MP CPU
> > > > - PLIC
> > > >
> > > > [0] arch/arm64/boot/dts/renesas/r9a07g043.dtsi
> > > >
> > > > Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@xxxxxxxxxxxxxx>
> > > > ---
> > > > v4 -> v5
> > > > * Fixed riscv,ndev value (should be 511)
> > > > * Reworked completely (sort of new patch)
> > > >
> > > > v3 -> v4
> > > > * No change
> > > >
> > > > v2 -> v3
> > > > * Fixed clock entry for CPU core
> > > > * Fixed timebase frequency to 12MHz
> > > > * Fixed sorting of the nodes
> > > > * Included RB tags
> > > >
> > > > v1 -> v2
> > > > * Dropped including makefile change
> > > > * Updated ndev count
> > > > ---
> > > >  arch/riscv/boot/dts/renesas/r9a07g043f.dtsi | 57 +++++++++++++++++++++
> > > >  1 file changed, 57 insertions(+)
> > > >  create mode 100644 arch/riscv/boot/dts/renesas/r9a07g043f.dtsi
> > > >
> > > > diff --git a/arch/riscv/boot/dts/renesas/r9a07g043f.dtsi b/arch/riscv/boot/dts/renesas/r9a07g043f.dtsi
> > > > new file mode 100644
> > > > index 000000000000..50134be548f5
> > > > --- /dev/null
> > > > +++ b/arch/riscv/boot/dts/renesas/r9a07g043f.dtsi
> > > > @@ -0,0 +1,57 @@
> > > > +// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
> > > > +/*
> > > > + * Device Tree Source for the RZ/Five SoC
> > > > + *
> > > > + * Copyright (C) 2022 Renesas Electronics Corp.
> > > > + */
> > > > +
> > > > +#include <dt-bindings/interrupt-controller/irq.h>
> > > > +
> > > > +#define SOC_PERIPHERAL_IRQ(nr) (nr + 32)
> > > > +
> > > > +#include <arm64/renesas/r9a07g043.dtsi>
> > > The initial patch shouldn't be broken. Combine them together with the
> > > minimal components and add others late. Don't separate the DTS files.
> > >
> > r9a07g043.dtsi [0] already exists in the kernel. r9a07g043.dtsi is
> > shared with the RZ/G2UL SoC (ARM64) and the RZ/Five SoC. There are two
> > more patches [1] which are required and are currently queued up in the
> > Renesas tree for v6.2 (Ive mentioned the dependencies in the cover
> > letter).
> 
> You could just move the below part to the second dtsi patch. Then
> compile won't be broken.
> 
>             clocks = <&cpg CPG_MOD R9A07G043_NCEPLIC_ACLK>;
>             power-domains = <&cpg>;
>             resets = <&cpg R9A07G043_NCEPLIC_ARESETN>;

The makefile for this directory is not added until the next patch right?
The compile shouldn't be broken here since it therefore cannot be
compiled?

Slightly confused,
Conor.

> 
> >
> > [0] arch/arm64/boot/dts/renesas/r9a07g043.dtsi
> > [1] https://patchwork.kernel.org/project/linux-renesas-soc/cover/20221025220629.79321-1-prabhakar.mahadev-lad.rj@xxxxxxxxxxxxxx/
> >
> > > > +
> > > > +/ {
> > > > +       cpus {
> > > > +               #address-cells = <1>;
> > > > +               #size-cells = <0>;
> > > > +               timebase-frequency = <12000000>;
> > > > +
> > > > +               cpu0: cpu@0 {
> > > > +                       compatible = "andestech,ax45mp", "riscv";
> > > > +                       device_type = "cpu";
> > > > +                       reg = <0x0>;
> > > > +                       status = "okay";
> > > > +                       riscv,isa = "rv64imafdc";
> > > > +                       mmu-type = "riscv,sv39";
> > > > +                       i-cache-size = <0x8000>;
> > > > +                       i-cache-line-size = <0x40>;
> > > > +                       d-cache-size = <0x8000>;
> > > > +                       d-cache-line-size = <0x40>;
> > > > +                       clocks = <&cpg CPG_CORE R9A07G043_CLK_I>;
> > > > +
> > > > +                       cpu0_intc: interrupt-controller {
> > > > +                               #interrupt-cells = <1>;
> > > > +                               compatible = "riscv,cpu-intc";
> > > > +                               interrupt-controller;
> > > > +                       };
> > > > +               };
> > > > +       };
> > > > +};
> > > > +
> > > > +&soc {
> > > > +       interrupt-parent = <&plic>;
> > > > +
> > > > +       plic: interrupt-controller@12c00000 {
> > > > +               compatible = "renesas,r9a07g043-plic", "andestech,nceplic100";
> > > > +               #interrupt-cells = <2>;
> > > > +               #address-cells = <0>;
> > > > +               riscv,ndev = <511>;
> > > > +               interrupt-controller;
> > > > +               reg = <0x0 0x12c00000 0 0x400000>;
> > > > +               clocks = <&cpg CPG_MOD R9A07G043_NCEPLIC_ACLK>;
> > > > +               power-domains = <&cpg>;
> > > > +               resets = <&cpg R9A07G043_NCEPLIC_ARESETN>;
> > > Ditto, Where is cpg? in r9a07g043.dtsi?
> > >
> > Yes CPG node is in r9a07g043.dtsi.
> >
> > Cheers,
> > Prabhakar
> 
> 
> 
> --
> Best Regards
>  Guo Ren