Re: [PATCH 1/3] dt-bindings: iio: adc: add bindings for stm32 mdf filter

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [Low] dt-bindings: invalid decimal-only regex for sitf and filter unit addresses
- [Low] dt-bindings: faulty regex in mdf-dai and channel patternProperties
--

commit 0433e9502551ab7438d761bd69f15064435b1ccc
Author: Olivier Moysan <olivier.moysan@foss.st.com>

dt-bindings: iio: adc: add bindings for stm32 mdf filter

Add bindings that describes STM32 MDF settings to support
digital filtering for Pulse Density Modulation (PDM) microphones
and analog sigma delta modulators.

> diff --git a/Documentation/devicetree/bindings/iio/adc/st,stm32-mdf-adc.yaml b/Documentation/devicetree/bindings/iio/adc/st,stm32-mdf-adc.yaml
> new file mode 100644
> index 0000000000000..f2fbc3e150e8c
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/iio/adc/st,stm32-mdf-adc.yaml
[ ... ]
> +patternProperties:
> +  "^sitf@[0-9]+$":

[Severity: Low]
Is this regex overly restrictive? Device tree unit addresses commonly contain
hexadecimal characters, so an address like sitf@1a0 would fail to match this
pattern.

Since the top-level schema has additionalProperties: false, valid nodes might
be rejected. Would "^sitf@[0-9a-f]+$" be more appropriate?

> +    type: object
> +    description: Serial interface child node
[ ... ]
> +  "^filter@[0-9]+$":

[Severity: Low]
Similarly, does this restrict filter unit addresses strictly to decimal digits?

A node with a hexadecimal offset, such as filter@1e4, would not match this
pattern and would fail schema validation.

> +    type: object
> +    description: Digital filter path child node
[ ... ]
> +    patternProperties:
> +      "^channel@([0-7])$":

[Severity: Low]
Does this capture group differ from the allOf constraint key? The allOf section
later in the file uses "^channel@[0-7]$".

Mismatches in literal keys can sometimes interfere with schema aggregation
tools that rely on exact string matching to properly merge subschemas.

> +        type: object
> +        $ref: adc.yaml
[ ... ]
> +        then:
> +          patternProperties:
> +            "^mdf-dai+$":

[Severity: Low]
Could this regex pattern be a typo? The '+' quantifier applies to the character
'i' rather than matching multiple child nodes or strictly matching the string.

If it is meant to match exactly one string, would "^mdf-dai$" be intended?

> +              type: object
> +              description: child node

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261001145702.2628429-1-olivier.moysan@foss.st.com?part=1