Re: [PATCH 1/2] dt-bindings: spi: renesas,sh-msiof: Add support for R-Car X5H

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] dt-bindings: weakened schema validation for older generations' dma counts
--

commit 339efeecd4721be8f1b5b945500191b6aa6436e7
Author: Thanh Quan <thanh.quan.xn@renesas.com>

dt-bindings: spi: renesas,sh-msiof: Add support for R-Car X5H

Document support for Clock-Synchronized Serial Interfaces with FIFO
(MSIOF) in the Renesas R-Car X5H (R8A78000) SoC, which is the first
member of the R-Car Gen5 family.

> diff --git a/Documentation/devicetree/bindings/spi/renesas,sh-msiof.yaml b/Documentation/devicetree/bindings/spi/renesas,sh-msiof.yaml
> index e0c7047ae8adb..41e423a42654d 100644
> --- a/Documentation/devicetree/bindings/spi/renesas,sh-msiof.yaml
> +++ b/Documentation/devicetree/bindings/spi/renesas,sh-msiof.yaml

[ ... ]

> @@ -100,11 +105,11 @@ properties:
>  
>    dmas:
>      minItems: 2
> -    maxItems: 4
> +    maxItems: 6

[Severity: Medium]
Does this unintentionally weaken schema validation for older generations?

By relaxing the global maximum to 6, a device tree for older hardware (which
only supports up to 4 DMAs) could now silently pass validation with an invalid
number of DMA channels.

Should this preserve maxItems: 4 globally and only override it to 6 inside an
allOf: if-then conditional block for the new Gen5 compatible?

>  
>    dma-names:
>      minItems: 2
> -    maxItems: 4
> +    maxItems: 6

[Severity: Medium]
Similarly, does relaxing this limit globally allow older generations to define
more dma-names than supported without generating a validation warning?

>      items:
>        enum: [ tx, rx ]
>

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/cover.1790065451.git.geert+renesas@glider.be?part=1