[PATCH v2] arm64: renesas: r8a779g0: add MFIS node

[PATCH v2] arm64: renesas: r8a779g0: add MFIS node

[Wolfram Sang]

Describe the MFIS core which is used for various tasks including
inter-processor communication. Interrupt numbers look irregular but they
all work as expected on a Renesas R-Car V4H SparrowHawk board.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
---

Change since v1:
* fixed sorting (Sashiko)

 arch/arm64/boot/dts/renesas/r8a779g0.dtsi | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/arch/arm64/boot/dts/renesas/r8a779g0.dtsi b/arch/arm64/boot/dts/renesas/r8a779g0.dtsi
index 82a7278836e5..08d7210629e5 100644
--- a/arch/arm64/boot/dts/renesas/r8a779g0.dtsi
+++ b/arch/arm64/boot/dts/renesas/r8a779g0.dtsi
@@ -589,6 +589,23 @@ tmu4: timer@ffc00000 {
 			status = "disabled";
 		};
 
+		mfis: system-controller@e6260000 {
+			compatible = "renesas,r8a779g0-mfis";
+			reg = <0 0xe6260000 0 0xf000>;
+			reg-names = "common";
+			interrupts = <GIC_SPI 123 IRQ_TYPE_LEVEL_HIGH>, <GIC_SPI 125 IRQ_TYPE_LEVEL_HIGH>,
+				     <GIC_SPI 127 IRQ_TYPE_LEVEL_HIGH>, <GIC_SPI 129 IRQ_TYPE_LEVEL_HIGH>,
+				     <GIC_SPI 916 IRQ_TYPE_LEVEL_HIGH>, <GIC_SPI 918 IRQ_TYPE_LEVEL_HIGH>,
+				     <GIC_SPI 920 IRQ_TYPE_LEVEL_HIGH>, <GIC_SPI 922 IRQ_TYPE_LEVEL_HIGH>,
+				     <GIC_SPI 131 IRQ_TYPE_LEVEL_HIGH>, <GIC_SPI 133 IRQ_TYPE_LEVEL_HIGH>,
+				     <GIC_SPI 135 IRQ_TYPE_LEVEL_HIGH>, <GIC_SPI 137 IRQ_TYPE_LEVEL_HIGH>;
+			interrupt-names = "ch0e",  "ch1e",  "ch2e",  "ch3e",  "ch4e",  "ch5e",  "ch6e",  "ch7e",  "ch8e",  "ch9e",
+					  "ch10e", "ch11e";
+			#hwlock-cells = <1>;
+			#mbox-cells = <2>;
+			status = "disabled";
+		};
+
 		tsn0: ethernet@e6460000 {
 			compatible = "renesas,r8a779g0-ethertsn", "renesas,rcar-gen4-ethertsn";
 			reg = <0 0xe6460000 0 0x7000>,
-- 
2.51.0

Re: [PATCH v2] arm64: renesas: r8a779g0: add MFIS node

[Geert Uytterhoeven]

Hi Wolfram,

On Sat, 30 May 2026 at 10:03, Wolfram Sang
<wsa+renesas@sang-engineering.com> wrote:
> Describe the MFIS core which is used for various tasks including
> inter-processor communication. Interrupt numbers look irregular but they
> all work as expected on a Renesas R-Car V4H SparrowHawk board.
>
> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>

Thanks for your patch!

> --- a/arch/arm64/boot/dts/renesas/r8a779g0.dtsi
> +++ b/arch/arm64/boot/dts/renesas/r8a779g0.dtsi
> @@ -589,6 +589,23 @@ tmu4: timer@ffc00000 {
>                         status = "disabled";
>                 };
>
> +               mfis: system-controller@e6260000 {
> +                       compatible = "renesas,r8a779g0-mfis";
> +                       reg = <0 0xe6260000 0 0xf000>;

Why 0xf000 and not 0x10000?

> +                       reg-names = "common";
> +                       interrupts = <GIC_SPI 123 IRQ_TYPE_LEVEL_HIGH>, <GIC_SPI 125 IRQ_TYPE_LEVEL_HIGH>,
> +                                    <GIC_SPI 127 IRQ_TYPE_LEVEL_HIGH>, <GIC_SPI 129 IRQ_TYPE_LEVEL_HIGH>,
> +                                    <GIC_SPI 916 IRQ_TYPE_LEVEL_HIGH>, <GIC_SPI 918 IRQ_TYPE_LEVEL_HIGH>,
> +                                    <GIC_SPI 920 IRQ_TYPE_LEVEL_HIGH>, <GIC_SPI 922 IRQ_TYPE_LEVEL_HIGH>,
> +                                    <GIC_SPI 131 IRQ_TYPE_LEVEL_HIGH>, <GIC_SPI 133 IRQ_TYPE_LEVEL_HIGH>,
> +                                    <GIC_SPI 135 IRQ_TYPE_LEVEL_HIGH>, <GIC_SPI 137 IRQ_TYPE_LEVEL_HIGH>;
> +                       interrupt-names = "ch0e",  "ch1e",  "ch2e",  "ch3e",  "ch4e",  "ch5e",  "ch6e",  "ch7e",  "ch8e",  "ch9e",

This is way too longer for a single line, and the double spacing
doesn't really help (yeah, it does align ch1e and ch11e...).

> +                                         "ch10e", "ch11e";
> +                       #hwlock-cells = <1>;
> +                       #mbox-cells = <2>;
> +                       status = "disabled";
> +               };
> +
>                 tsn0: ethernet@e6460000 {
>                         compatible = "renesas,r8a779g0-ethertsn", "renesas,rcar-gen4-ethertsn";
>                         reg = <0 0xe6460000 0 0x7000>,

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH v2] arm64: renesas: r8a779g0: add MFIS node

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 764 bytes --]

Hi Geert,

thanks for the review!

> > +               mfis: system-controller@e6260000 {
> > +                       compatible = "renesas,r8a779g0-mfis";
> > +                       reg = <0 0xe6260000 0 0xf000>;
> 
> Why 0xf000 and not 0x10000?

The sheet listing the memory map has a "MFI - Region 14" entry for
0xe626e000, but no region 15 for 0xe626f000. I thought I play safe.
Can change if you want...

> > +                       interrupt-names = "ch0e",  "ch1e",  "ch2e",  "ch3e",  "ch4e",  "ch5e",  "ch6e",  "ch7e",  "ch8e",  "ch9e",
> 
> This is way too longer for a single line, and the double spacing
> doesn't really help (yeah, it does align ch1e and ch11e...).

You are right, will fix this.

Happy hacking,

   Wolfram


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [PATCH v2] arm64: renesas: r8a779g0: add MFIS node

[Geert Uytterhoeven]

Hi Wolfram,

On Tue, 2 Jun 2026 at 13:03, Wolfram Sang
<wsa+renesas@sang-engineering.com> wrote:
> > > +               mfis: system-controller@e6260000 {
> > > +                       compatible = "renesas,r8a779g0-mfis";
> > > +                       reg = <0 0xe6260000 0 0xf000>;
> >
> > Why 0xf000 and not 0x10000?
>
> The sheet listing the memory map has a "MFI - Region 14" entry for
> 0xe626e000, but no region 15 for 0xe626f000. I thought I play safe.

Indeed...

However, according to attachment R-CarV4H_UM_018_AXI-bus_D_r1p30.xlsx,
MFI14 applies to MFIS H'E626_E000 - H'E626_FFFF, thus covers 8 KiB,
i.e. double of the other regions.

> Can change if you want...

Yes please.
Thanks!

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH v2] arm64: renesas: r8a779g0: add MFIS node

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 414 bytes --]

Hi Geert,

> However, according to attachment R-CarV4H_UM_018_AXI-bus_D_r1p30.xlsx,
> MFI14 applies to MFIS H'E626_E000 - H'E626_FFFF, thus covers 8 KiB,
> i.e. double of the other regions.

Thank you for pointing me to that document! I did look for the size of
the memory area but could not find it.

> > Can change if you want...
> 
> Yes please.

Will change it.

Happy hacking,

   Wolfram


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]