Re: [PATCH v6 15/17] ARM: dts: microchip: sama7d65: add cells for temperature calibration

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:

New issues:
- [Low] Adding `nvmem-cells` and `nvmem-cell-names` properties to the ADC device tree node without updating the corresponding binding schema causes validation failures.

Pre-existing issues:
- [High] This problem wasn't introduced by this patch, but the ADC driver completely ignores -EPROBE_DEFER from nvmem_cell_get(), leading to missing temperature channels on boot.
--

--- Patch [15]: [PATCH v6 15/17] ARM: dts: microchip: sama7d65: add cells for temperature calibration ---
commit a0118b79f35c5e22daa318dd83acca0c7e480c2e
Author: Varshini Rajendran <varshini.rajendran@microchip.com>

ARM: dts: microchip: sama7d65: add cells for temperature calibration
    
Add nvmem-cells for temperature calibration data to the ADC node.
The calibration data is read from the OTPC.

> diff --git a/arch/arm/boot/dts/microchip/sama7d65.dtsi b/arch/arm/boot/dts/microchip/sama7d65.dtsi
> index aac2e22d96cb3..a8b18ad0ef7f7 100644
> --- a/arch/arm/boot/dts/microchip/sama7d65.dtsi
> +++ b/arch/arm/boot/dts/microchip/sama7d65.dtsi
> @@ -318,6 +318,8 @@ adc: adc@e1000000 {
>  			dmas = <&dma0 AT91_XDMAC_DT_PERID(0)>;
>  			dma-names = "rx";
>  			#io-channel-cells = <1>;
> +			nvmem-cells = <&temperature_calib>;
> +			nvmem-cell-names = "temperature_calib";

[Severity: Low]
Will adding these properties without updating the binding schema cause
dtbs_check to fail? 

Documentation/devicetree/bindings/iio/adc/atmel,sama5d2-adc.yaml enforces
additionalProperties: false, which might reject the newly added nvmem-cells
and nvmem-cell-names properties.

[Severity: High]
This is a pre-existing issue, but does the ADC driver incorrectly swallow
probe deferrals when trying to fetch this nvmem cell?

If the OTPC provider has not finished probing, nvmem_cell_get() in
drivers/iio/adc/at91-sama5d2_adc.c:at91_adc_temp_sensor_init() will return
-EPROBE_DEFER.

However, the caller at91_adc_probe() appears to ignore the return value:

    ret = at91_adc_temp_sensor_init(st, &pdev->dev);
    /* Don't register temperature channel if initialization failed. */
    if (ret)
        num_channels = st->soc_info.platform->max_channels - 1;
    else
        num_channels = st->soc_info.platform->max_channels;

Could this non-deterministically drop the temperature channel on boot
instead of properly deferring the probe?

>  			atmel,min-sample-rate-hz = <200000>;
>  			atmel,max-sample-rate-hz = <20000000>;
>  			atmel,trigger-edge-type = <IRQ_TYPE_EDGE_RISING>;

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260806074024.531259-1-varshini.rajendran@microchip.com?part=15