Hi Potin,
On Thu, 2025-03-20 at 23:21 +0800, Potin Lai wrote:
> Add device tree nodes for the NCT7363 fan controllers on the second-
> source
> Power Distribution Board (PDB).
>
> Signed-off-by: Potin Lai <potin.lai.pt@xxxxxxxxx>
> ---
> .../dts/aspeed/aspeed-bmc-facebook-catalina.dts | 76
> ++++++++++++++++++++++
> 1 file changed, 76 insertions(+)
>
> diff --git a/arch/arm/boot/dts/aspeed/aspeed-bmc-facebook-
> catalina.dts b/arch/arm/boot/dts/aspeed/aspeed-bmc-facebook-
> catalina.dts
> index fd22add90449..9cbb296dcd9e 100644
> --- a/arch/arm/boot/dts/aspeed/aspeed-bmc-facebook-catalina.dts
> +++ b/arch/arm/boot/dts/aspeed/aspeed-bmc-facebook-catalina.dts
> @@ -467,6 +467,82 @@ i2c1mux0ch2: i2c@2 {
> #size-cells = <0>;
> reg = <0x2>;
>
> + hwmon0: hwmon@1 {
There might be instances of calling nodes and labels 'hwmon' in other
(BMC) devicetrees, but I don't think we should continue that practice.
I think in this case it should be 'fan-controller'.
hwmon is the name of a kernel subsystem, not a class of hardware. The
devicetrees should describe the hardware (and not concepts from the
kernel).
> + compatible = "nuvoton,nct7363";
> + reg = <0x01>;
> + #pwm-cells = <2>;
> +
> + fan-9 {
> + pwms = <&hwmon0 0 40000>;
> + tach-ch = /bits/ 8 <0x09>;
> + };
> + fan-11 {
> + pwms = <&hwmon0 0 40000>;
> + tach-ch = /bits/ 8 <0x0b>;
> + };
> + fan-10 {
> + pwms = <&hwmon0 4 40000>;
> + tach-ch = /bits/ 8 <0x0a>;
> + };
> + fan-13 {
> + pwms = <&hwmon0 4 40000>;
> + tach-ch = /bits/ 8 <0x0d>;
> + };
> + fan-15 {
> + pwms = <&hwmon0 6 40000>;
> + tach-ch = /bits/ 8 <0x0f>;
> + };
> + fan-1 {
> + pwms = <&hwmon0 6 40000>;
> + tach-ch = /bits/ 8 <0x01>;
> + };
> + fan-0 {
> + pwms = <&hwmon0 10 40000>;
> + tach-ch = /bits/ 8 <0x00>;
> + };
> + fan-3 {
> + pwms = <&hwmon0 10 40000>;
> + tach-ch = /bits/ 8 <0x03>;
> + };
> + };
> + hwmon1: hwmon@2 {
As above.
Andrew
