Re: [PATCH v6 3/3] riscv64: dts: sophgo: add initial Milk-V Duo S board support

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Contradictory eMMC voltage configuration artificially restricts eMMC speed capabilities
--

commit 9d652e59c7623b46a5dd7a0a5954061467e5aea7
Author: Joshua Milas <josh.milas@gmail.com>

riscv64: dts: sophgo: add initial Milk-V Duo S board support

This adds initial riscv support for the Milk-V Duo S board
making it possible to boot Linux to the command line.

Link: https://milkv.io/duo-s [1]

> diff --git a/arch/riscv/boot/dts/sophgo/sg2000-milkv-duo-s.dts b/arch/riscv/boot/dts/sophgo/sg2000-milkv-duo-s.dts
> new file mode 100644
> index 0000000000000..f56bfddeacc40
> --- /dev/null
> +++ b/arch/riscv/boot/dts/sophgo/sg2000-milkv-duo-s.dts

[ ... ]

> +&pinctrl {
> +	emmc_cfg: emmc-cfg {
> +		emmc-rstn-pins {
> +			pinmux = <PINMUX(PIN_EMMC_RSTN, 0)>;
> +			bias-pull-up;
> +			drive-strength-microamp = <10800>;
> +			power-source = <1800>;
> +		};

[ ... ]

> +&emmc {
> +	bus-width = <4>;
> +	no-1-8-v;

[Severity: Medium]
Is this contradictory eMMC voltage configuration intentional?

The pinctrl node emmc_cfg explicitly configures the eMMC pins to
operate at 1.8V (power-source = <1800>), but the &emmc node here
specifies the no-1-8-v property.

Since the hardware pads are physically configured for 1.8V, the interface
actually does support 1.8V signaling. Specifying no-1-8-v here artificially
restricts the eMMC to legacy, slower modes instead of negotiating
faster modes like HS200.

> +	cap-mmc-hw-reset;
> +	no-sd;
> +	no-sdio;
> +	non-removable;
> +	pinctrl-0 = <&emmc_cfg>;
> +	pinctrl-names = "default";
> +	status = "okay";
> +};

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260615120324.1527881-1-josh.milas@gmail.com?part=3