Re: [PATCH v2 2/2] arm64: dts: rockchip: Add rk3576 evb2 board

[Chaoyi Chen <chaoyi.chen@rock-chips.com>]

Hi Heiko,

On 1/7/2026 4:21 PM, Heiko Stübner wrote:
> Am Mittwoch, 7. Januar 2026, 08:56:04 Mitteleuropäische Normalzeit schrieb Alexey Charkov:
>> Hi Chaoyi,
>>
>> On Wed, Jan 7, 2026 at 11:04 AM Chaoyi Chen <kernel@airkyi.com> wrote:
>>>
>>> From: Chaoyi Chen <chaoyi.chen@rock-chips.com>
>>>
>>> General features for rk3576 evb2 board:
>>>     - Rockchip RK3576
>>>     - LPDDR4/4X
>>>     - eMMC5.1
>>>     - RK806-2x2pcs + DiscretePower
>>>     - 1x HDMI2.1 TX / HDMI2.0 RX
>>>     - 1x full size DP1.4 TX (Only 2 Lanes)
>>>     - 2x 10/100/1000M Ethernet
>>>     - 5x SATA3.0 7Pin Slot
>>>     - 2x USB3.2 Gen1 Host
>>>     - 3x USB2.0 Host
>>>     - WIFI/BT
>>>     - ...
>>>
>>> Tested with eMMC/SDMMC/HDMI/USB/Ethernet/WIFI/BT module.
>>>
>>> Signed-off-by: Chaoyi Chen <chaoyi.chen@rock-chips.com>
> 
> [...]
> 
>>> +       vbus5v0_typec: regulator-vbus5v0-typec {
>>> +               compatible = "regulator-fixed";
>>> +               regulator-name = "vbus5v0_typec";
>>
>> This might better be renamed, given that last time you mentioned this
>> board doesn't have a Type-C connector. Perhaps regulator-vbus5v0-otg?
> 
> Alternatively a comment above it.
> 
> I.e. regulator-naming should always follow the naming used in the
> schematics, so that it gets easier to reference between schematics
> and devicetree.
> 

Thanks for the explanation. I will fix this in v3.

> 
>>> +               regulator-min-microvolt = <5000000>;
>>> +               regulator-max-microvolt = <5000000>;
>>> +               enable-active-high;
>>> +               gpio = <&gpio0 RK_PD1 GPIO_ACTIVE_HIGH>;
>>> +               vin-supply = <&vcc5v0_device>;
>>> +               pinctrl-names = "default";
>>> +               pinctrl-0 = <&usb_otg0_pwren>;
>>> +       };
>>> +
>>> +       vcc12v_dcin: regulator-vcc12v-dcin {
>>> +               compatible = "regulator-fixed";
>>> +               regulator-name = "vcc12v_dcin";
>>> +               regulator-always-on;
>>> +               regulator-boot-on;
>>> +               regulator-min-microvolt = <12000000>;
>>> +               regulator-max-microvolt = <12000000>;
>>> +       };
>>> +
>>> +       vcc1v2_ufs_vccq_s0: regulator-vcc1v2-ufs-vccq-s0 {
>>> +               compatible = "regulator-fixed";
>>> +               regulator-name = "vcc1v2_ufs_vccq_s0";
>>> +               regulator-boot-on;
>>> +               regulator-always-on;
>>> +               regulator-min-microvolt = <1200000>;
>>> +               regulator-max-microvolt = <1200000>;
>>> +               vin-supply = <&vcc_sys>;
>>> +       };
>>> +
>>> +       vcc1v8_ufs_vccq2_s0: regulator-vcc1v8-ufs-vccq2-s0 {
>>> +               compatible = "regulator-fixed";
>>> +               regulator-name = "vcc1v8_ufs_vccq2_s0";
>>> +               regulator-boot-on;
>>> +               regulator-always-on;
>>> +               regulator-min-microvolt = <1800000>;
>>> +               regulator-max-microvolt = <1800000>;
>>> +               vin-supply = <&vcc_1v8_s3>;
>>> +       };
>>> +
>>> +       vcc3v3_hubreset: vcc3v3-hubreset {
>>> +               compatible = "regulator-fixed";
>>> +               regulator-name = "vcc3v3_hubreset";
>>> +               regulator-boot-on;
>>> +               regulator-always-on;
>>
>> If this regulator supplies a soldered-on discrete hub and is required
>> to power it up, won't it be better to describe the hub in the device
>> tree (see binding at [1]), make the regulator its supply, and perhaps
>> drop the "regulator-boot-on/regulator-always-on" annotation here,
>> letting the regulator core deal with its enabling instead?
>>
>> [1] https://github.com/torvalds/linux/blob/master/Documentation/devicetree/bindings/usb/usb-device.yaml
> 
> Yep, it would be nicer to it this way.
> A live example can be found in the Rock 5 ITX [2]
> 
> [2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/arm64/boot/dts/rockchip/rk3588-rock-5-itx.dts#n1266

Thank you for the great example. BTW the hub used here is CH344. It
looks like we need to add a new binding :)

> 
> 
> Heiko
> 
> 
>> [snip]
>>
>> Other than these, LGTM - thanks for addressing my comments from v1!
>> Feel free to include my:
>>
>> Reviewed-by: Alexey Charkov <alchark@gmail.com>
>>
>> Best regards,
>> Alexey
>>
> 
> 
> 
> 
> 
> 

-- 
Best, 
Chaoyi