On 24/11/2020 13:17, Christian Hewitt wrote:
> GXM (S912) is a big-little design with CPUs 0-3 clocked at 1.5GHz
> and CPUs 4-7 at 1.0GHz. Adding capacity-dmips-mhz attributes allows
> the scheduler to factor the different clock speeds into capacity
> calculations and prefer the higher-clocked cluster to improve
> overall performance.
> 
> This was inspired by the similar change for G12B [0] boards. The
> diference here is that all cores are A53's so the same dmips-mhz
> value is used.
> 
> VIM2:~ # cat /sys/devices/system/cpu/cpu*/cpufreq/scaling_cur_freq
> 1512000
> 1512000
> 1512000
> 1512000
> 1000000
> 1000000
> 1000000
> 1000000
> 
> before:
> 
> VIM2:~ # cat /sys/devices/system/cpu/cpu*/cpu_capacity
> 1024
> 1024
> 1024
> 1024
> 1024
> 1024
> 1024
> 1024
> 
> after:
> 
> VIM2:~ # cat /sys/devices/system/cpu/cpu*/cpu_capacity
> 1024
> 1024
> 1024
> 1024
> 677
> 677
> 677
> 677
> 
> The after value matches my table-napkin calculation:
> 
> (1000000 / 1512000 = 0.661) * 1024 = 677
> 
> [0] 
> https://github.com/torvalds/linux/commit/6eeaf4d2452ec8b1ece58776812140734fc2e088
> 
> Signed-off-by: Christian Hewitt <christianshew...@gmail.com>
> ---
>  arch/arm64/boot/dts/amlogic/meson-gxm.dtsi | 20 ++++++++++++++++++++
>  1 file changed, 20 insertions(+)
> 
> diff --git a/arch/arm64/boot/dts/amlogic/meson-gxm.dtsi 
> b/arch/arm64/boot/dts/amlogic/meson-gxm.dtsi
> index fe4145112295..411cc312fc62 100644
> --- a/arch/arm64/boot/dts/amlogic/meson-gxm.dtsi
> +++ b/arch/arm64/boot/dts/amlogic/meson-gxm.dtsi
> @@ -42,11 +42,28 @@
>                       };
>               };
>  
> +             cpu0: cpu@0 {
> +                     capacity-dmips-mhz = <1024>;
> +             };
> +
> +             cpu1: cpu@1 {
> +                     capacity-dmips-mhz = <1024>;
> +             };
> +
> +             cpu2: cpu@2 {
> +                     capacity-dmips-mhz = <1024>;
> +             };
> +
> +             cpu3: cpu@3 {
> +                     capacity-dmips-mhz = <1024>;
> +             };
> +
>               cpu4: cpu@100 {
>                       device_type = "cpu";
>                       compatible = "arm,cortex-a53";
>                       reg = <0x0 0x100>;
>                       enable-method = "psci";
> +                     capacity-dmips-mhz = <1024>;
>                       next-level-cache = <&l2>;
>                       clocks = <&scpi_dvfs 1>;
>                       #cooling-cells = <2>;
> @@ -57,6 +74,7 @@
>                       compatible = "arm,cortex-a53";
>                       reg = <0x0 0x101>;
>                       enable-method = "psci";
> +                     capacity-dmips-mhz = <1024>;
>                       next-level-cache = <&l2>;
>                       clocks = <&scpi_dvfs 1>;
>                       #cooling-cells = <2>;
> @@ -67,6 +85,7 @@
>                       compatible = "arm,cortex-a53";
>                       reg = <0x0 0x102>;
>                       enable-method = "psci";
> +                     capacity-dmips-mhz = <1024>;
>                       next-level-cache = <&l2>;
>                       clocks = <&scpi_dvfs 1>;
>                       #cooling-cells = <2>;
> @@ -77,6 +96,7 @@
>                       compatible = "arm,cortex-a53";
>                       reg = <0x0 0x103>;
>                       enable-method = "psci";
> +                     capacity-dmips-mhz = <1024>;
>                       next-level-cache = <&l2>;
>                       clocks = <&scpi_dvfs 1>;
>                       #cooling-cells = <2>;
> 

Re-reading the bindings, the value is correct and necessary to have the 
cpu_capacity calculated
to help the scheduler.

Thanks!

Reviewed-by: Neil Armstrong <narmstr...@baylibre.com>

Reply via email to