Re: [PATCH 1/2] arm64: dts: qcom: kodiak: Sort pinctrl subnodes by pins

["Luca Weiss" <luca.weiss@fairphone.com>]

On Fri Jun 12, 2026 at 2:59 PM CEST, Vladimir Zapolskiy wrote:
> As documented in the "Devicetree Sources (DTS) Coding Style" document,
> pinctrl subnodes should be sorted by the pins property. Do this once for
> kodiak.dtsi so that future additions can be added at the right places.
>
> No functional change intended, verified with dtx_diff.
>
> Signed-off-by: Luca Weiss <luca.weiss@fairphone.com>
> ---
>   arch/arm64/boot/dts/qcom/kodiak.dtsi | 1382 +++++++++++++++++-----------------
>   1 file changed, 691 insertions(+), 691 deletions(-)
>
> diff --git a/arch/arm64/boot/dts/qcom/kodiak.dtsi b/arch/arm64/boot/dts/qcom/kodiak.dtsi
> index fa540d8c2615..62daef726d32 100644
> --- a/arch/arm64/boot/dts/qcom/kodiak.dtsi
> +++ b/arch/arm64/boot/dts/qcom/kodiak.dtsi

<snip>

> +			qup_uart12_cts: qup-uart12-cts-state {
> +				pins = "gpio48";
> +				function = "qup14";
> +			};
> +
> +			qup_uart12_rts: qup-uart12-rts-state {
> +				pins = "gpio49";
> +				function = "qup14";
> +			};
> +
> +			qup_uart12_tx: qup-uart12-tx-state {
> +				pins = "gpio50";
> +				function = "qup14";
> +			};
>
> I understand and support the intention to keep this change non-functional,
> but this pad "gpio50" is for qup16 also, right?

According to my QCM6490 data sheet, GPIO_50 has these functions:
* UART for qup14 (OK)
* SPI for qup14 (OK)
* SPI for qup16 (no pinctrl)

>
> Similarly pads "gpio54"/"gpio55" for qup14 function, "gpio62"/"gpio63"
> for qup16 function, I find all of these are missing on the original list.

GPIO_54:
* UART qup15 (OK)
* SPI qup15 (OK)
* SPI qup14 (no pinctrl)

GPIO_55:
* UART qup15 (OK)
* SPI qup15 (OK)
* SPI qup14 (no pinctrl)

GPIO_62:
* UART qup17 (OK)
* SPI qup17 (OK)
* SPI qup16 (no pinctrl)

GPIO_63:
* UART qup16 (?)
* SPI qup16 (lane 3) (?)
* SPI qup16 (lane 5) (?)

But the GPIO_63 looks weird, is the data sheet wrong?! Where would
UART_RX of QUP1 SE7 go? Maybe it should be UART qup17 and SPI qup17 and
then SPI qup16 ??

Can somebody at Qualcomm please check 80-20659-1 Rev. AM and maybe make
the apppriate people there aware?

So yes Vladimir, you're correct. Some pinctrl definitions for those SPI
QUPs are not defined. And the datasheet seems wrong as well.

> Reviewed-by: Vladimir Zapolskiy <vladimir.zapolskiy@linaro.org>

Thanks for checking!

Regards
Luca