Re: [PATCH v2 4/7] dt-bindings: remoteproc: add Allwinner sun55i-rproc schema

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] dt-bindings: missing constraint for required 'r_sram'/'r_sram1' in reg-names
--

commit cc7b4893d8220b754d0e23e60a9d997c33b1c0c3
Author: Tim Michals <tcmichals@gmail.com>
Subject: dt-bindings: remoteproc: add Allwinner sun55i-rproc schema

This commit adds a Device Tree binding schema for the Allwinner XuanTie
E906/E907 RISC-V remoteproc on sun55i SoCs. It defines resources such as
memory-mapped registers, clocks, resets, and mailboxes required to manage
the co-processor.

> diff --git a/Documentation/devicetree/bindings/remoteproc/allwinner,sun55i-rproc.yaml b/Documentation/devicetree/bindings/remoteproc/allwinner,sun55i-rproc.yaml
> new file mode 100644
> index 0000000000000..53e3f4faaa768
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/remoteproc/allwinner,sun55i-rproc.yaml

[ ... ]

> +  reg:
> +    minItems: 1
> +    maxItems: 4
> +    description:
> +      Memory-mapped register regions. The following named regions are
> +      supported (all optional except at least one of r_sram or r_sram1) -
> +      "cfg" for RISC-V core control and boot-vector registers,
> +      "r_sram" for dedicated MCU SRAM Space 0,
> +      "r_sram1" for switchable MCU SRAM Space 1,
> +      "remap" for the hardware remap control register.
> +
> +  reg-names:
> +    minItems: 1
> +    maxItems: 4
> +    items:
> +      enum:
> +        - cfg
> +        - r_sram
> +        - r_sram1
> +        - remap

[Severity: Medium]
The description for reg explicitly states that at least one of r_sram or
r_sram1 is required, but the schema for reg-names allows any single item
from the enum (such as just cfg) to pass validation.

If a device tree node is created with only cfg or remap in reg-names, it
will silently pass offline device tree validation despite lacking the
required SRAM regions.

Should an additional constraint be added to enforce the presence of at least
one SRAM region so that incomplete device trees are caught during validation?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260927002021.797069-1-tcmichals@gmail.com?part=4