Re: [PATCH 1/2] arm64: dts: rk3328: add gpu opp table

[Johan Jonker]

On 10/16/21 5:45 PM, Trevor Woerner wrote:
> Add an operating-points table and cooling entry to the GPU on the
> RK3328 SoC to improve its performance. According to its datasheet[1]
> the maximum frequency of the Mali-450 MP2 GPU found on the RK3328 SoC
> is 500MHz.
> 
> On my rock64 device, under x11, glmark2-es2 performance increased from
> around 60 to just over 100. Same device running glmark2-es2 under
> wayland/weston improved from just over 100 to just over 200.
> 
> [1] https://rockchip.fr/RK3328%20datasheet%20V1.2.pdf
> 
> Signed-off-by: Trevor Woerner <twoerner@xxxxxxxxx>
> ---
>  arch/arm64/boot/dts/rockchip/rk3328.dtsi | 26 +++++++++++++++++++++++-
>  1 file changed, 25 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/arm64/boot/dts/rockchip/rk3328.dtsi b/arch/arm64/boot/dts/rockchip/rk3328.dtsi
> index 8c821acb21ff..5e1dcf71e414 100644
> --- a/arch/arm64/boot/dts/rockchip/rk3328.dtsi
> +++ b/arch/arm64/boot/dts/rockchip/rk3328.dtsi
> @@ -532,7 +532,8 @@ map0 {
>  					cooling-device = <&cpu0 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>  							 <&cpu1 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
>  							 <&cpu2 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
> -							 <&cpu3 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
> +							 <&cpu3 THERMAL_NO_LIMIT THERMAL_NO_LIMIT>,
> +							 <&gpu THERMAL_NO_LIMIT THERMAL_NO_LIMIT>;
>  					contribution = <4096>;
>  				};
>  			};
> @@ -617,6 +618,29 @@ gpu: gpu@ff300000 {
>  		clocks = <&cru ACLK_GPU>, <&cru ACLK_GPU>;
>  		clock-names = "bus", "core";
>  		resets = <&cru SRST_GPU_A>;
> +		operating-points-v2 = <&gpu_opp_table>;
> +		#cooling-cells = <2>;
> +	};
> +

> +	gpu_opp_table: gpu-opp-table {

After the conversion to YAML of the Operating Performance Points(OPP)
binding the operating-points-v2 property expects the nodename to have
the '^opp-table(-[a-z0-9]+)?$' format.

make ARCH=arm64 dtbs_check
DT_SCHEMA_FILES=Documentation/devicetree/bindings/opp/opp-v2.yaml

> +		compatible = "operating-points-v2";
> +
> +		opp-200000000 {
> +			opp-hz = /bits/ 64 <200000000>;
> +			opp-microvolt = <1100000>;
> +		};
> +		opp-300000000 {
> +			opp-hz = /bits/ 64 <300000000>;
> +			opp-microvolt = <1100000>;
> +		};
> +		opp-400000000 {
> +			opp-hz = /bits/ 64 <400000000>;
> +			opp-microvolt = <1100000>;
> +		};
> +		opp-500000000 {
> +			opp-hz = /bits/ 64 <500000000>;
> +			opp-microvolt = <1100000>;
> +		};
>  	};

opp-microvolt has the same value for every node vs. table below?

See also previous discussion:

https://lore.kernel.org/linux-rockchip/3c95c29b-6c07-5945-ac22-d683997e1ca0@xxxxxxx/

Is that now fixed/checked?

Copy from manufacturer tree:

	gpu_opp_table: gpu-opp-table {
		compatible = "operating-points-v2";

		rockchip,leakage-voltage-sel = <
			1   10    0
			11  254   1
		>;
		nvmem-cells = <&logic_leakage>;
		nvmem-cell-names = "gpu_leakage";

		opp-200000000 {
			opp-hz = /bits/ 64 <200000000>;
			opp-microvolt = <950000>;
			opp-microvolt-L0 = <950000>;
			opp-microvolt-L1 = <950000>;
		};
		opp-300000000 {
			opp-hz = /bits/ 64 <300000000>;
			opp-microvolt = <975000>;
			opp-microvolt-L0 = <975000>;
			opp-microvolt-L1 = <950000>;
		};
		opp-400000000 {
			opp-hz = /bits/ 64 <400000000>;
			opp-microvolt = <1050000>;
			opp-microvolt-L0 = <1050000>;
			opp-microvolt-L1 = <1025000>;
		};
		opp-500000000 {
			opp-hz = /bits/ 64 <500000000>;
			opp-microvolt = <1150000>;
			opp-microvolt-L0 = <1150000>;
			opp-microvolt-L1 = <1100000>;
		};
	};
>  
>  	h265e_mmu: iommu@ff330200 {
>