Re: [PATCH v3 1/4] dt-bindings: rtc: Add NXP PCF8525 RTC

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] The `interrupt-names` property schema incorrectly rejects valid hardware configurations where only INTB is connected.
--

commit 620f353657ea4242bf73ac8431d118e755e9de62
Author: Shiv Prakash Gupta <shivprakash.gupta@nxp.com>
Subject: dt-bindings: rtc: Add NXP PCF8525 RTC

This commit adds a Devicetree binding for the NXP PCF8525 I2C RTC. It 
describes properties for the device, including the optional INTA and INTB 
interrupts, crystal load capacitance, and temperature-compensation model.

> diff --git a/Documentation/devicetree/bindings/rtc/nxp,pcf8525.yaml b/Documentation/devicetree/bindings/rtc/nxp,pcf8525.yaml
> new file mode 100644
> index 0000000000000..bee71ec3da808
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/rtc/nxp,pcf8525.yaml
> @@ -0,0 +1,86 @@

[ ... ]

> +  interrupts:
> +    minItems: 1
> +    maxItems: 2
> +    description:
> +      INTA, followed by INTB when both interrupt outputs are connected.
> +
> +  interrupt-names:
> +    minItems: 1
> +    items:
> +      - const: inta
> +      - const: intb

[Severity: Medium]
Does this interrupt-names array definition inadvertently prevent
configurations where only intb is connected? 

Using an items array with - const: inta at the first position and
- const: intb at the second strictly requires the first item to be inta
during dt_binding_check schema validation.

If a device tree author specifies a single interrupt and uses
interrupt-names = "intb" because only the INTB pin is routed to the SoC
(for example, to be used solely for the watchdog), this schema will reject it.

Would it be better to use oneOf to explicitly permit the valid combinations,
or define the items using an enum so that intb is permitted at the first
index?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260910060912.2443551-1-shivprakash.gupta@nxp.com?part=1