Re: [PATCH 7/7] arm64: dts: allwinner: add a523 msgbox and remoteproc nodes

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [Low] dts: dt-binding schema violation in remoteproc compatible property
- [Low] dts: simple-bus node sorting violation for mailbox@3003000
--

commit 38cf09c8de53fb563929c96a81218691b32c1fef
Author: Tim Michals <tcmichals@gmail.com>

arm64: dts: allwinner: add a523 msgbox and remoteproc nodes

This commit adds the hardware mailbox and XuanTie RISC-V remoteproc nodes for
Allwinner A523/A527 SoCs. It also connects the remoteproc node to the
necessary clocks, resets, and mailboxes for lifecycle management and IPC.

> diff --git a/arch/arm64/boot/dts/allwinner/sun55i-a523.dtsi b/arch/arm64/boot/dts/allwinner/sun55i-a523.dtsi
> index 85ef492ffeae7..7c5c39c568583 100644
> --- a/arch/arm64/boot/dts/allwinner/sun55i-a523.dtsi
> +++ b/arch/arm64/boot/dts/allwinner/sun55i-a523.dtsi
> @@ -1144,5 +1144,44 @@ npu: npu@7122000 {
>  			resets = <&mcu_ccu RST_BUS_MCU_NPU>;
>  			power-domains = <&ppu PD_NPU>;
>  		};
> +
> +		msgbox: mailbox@3003000 {

[Severity: Low]
Does this placement violate the device tree simple-bus unit address sorting
conventions?  Typically, nodes within a simple-bus are sorted sequentially by
unit address in ascending order.  Placing mailbox@3003000 after npu@7122000
breaks this ordering.

Should this node be moved earlier in the file to sit between
dma-controller@3002000 and efuse@3006000 to maintain correct sorting?

> +			compatible = "allwinner,sun55i-a523-msgbox";
> +			reg = <0x03003000 0x1000>,
> +			      <0x07094000 0x1000>,
> +			      <0x07120000 0x1000>,
> +			      <0x07136000 0x1000>;
> +			reg-names = "arm", "dsp", "cpus", "rv";

[ ... ]

> +		rproc: remoteproc@7130000 {
> +			compatible = "allwinner,sun55i-a523-rproc",
> +				     "allwinner,sun55i-a527-rproc";

[Severity: Low]
Will this multi-string compatible property pass device tree validation?

The dt-binding for allwinner,sun55i-rproc.yaml defines the compatible
property strictly as an enum containing allwinner,sun55i-a523-rproc,
allwinner,sun55i-a527-rproc, and allwinner,sun55i-t527-rproc.

Since the schema uses enum instead of an items list, it requires the value to
be exactly one of the listed strings.  Could this be changed to specify a
single compatible string to prevent dtbs_check warnings?

> +			reg = <0x07130000 0x1000>,
> +			      <0x07280000 0x40000>,

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260922034711.190253-1-tcmichals@gmail.com?part=7