Received-SPF: pass (google.com: domain of linux-kernel+bounces-38373-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Precedence: bulk
MIME-Version: 1.0
References: <20240125-rk-dts-additions-v1-0-5879275db36f@gmail.com>
 <20240125-rk-dts-additions-v1-4-5879275db36f@gmail.com> <731aac66-f698-4a1e-b9ee-46a7f24ecae5@linaro.org>
In-Reply-To: <731aac66-f698-4a1e-b9ee-46a7f24ecae5@linaro.org>
From: Alexey Charkov <alchark@gmail.com>
Date: Thu, 25 Jan 2024 14:17:21 +0400
Message-ID: <CABjd4YzWJie91kcbHom_Zso=QQR9gPmAVvJb1vbqa0Qwu5egKg@mail.gmail.com>
Subject: Re: [PATCH 4/4] arm64: dts: rockchip: Add OPP data for CPU cores on RK3588
To: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Rob Herring <robh+dt@kernel.org>, 
	Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>, Conor Dooley <conor+dt@kernel.org>, 
	Heiko Stuebner <heiko@sntech.de>, Dragan Simic <dsimic@manjaro.org>, devicetree@vger.kernel.org, 
	linux-arm-kernel@lists.infradead.org, linux-rockchip@lists.infradead.org, 
	linux-kernel@vger.kernel.org, Viresh Kumar <viresh.kumar@linaro.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Thu, Jan 25, 2024 at 1:30=E2=80=AFPM Daniel Lezcano
<daniel.lezcano@linaro.org> wrote:
>
>
> Hi Alexey,
>
> Adding Viresh
>
> On 24/01/2024 21:30, Alexey Charkov wrote:
> > By default the CPUs on RK3588 start up in a conservative performance
> > mode. Add frequency and voltage mappings to the device tree to enable
> > dynamic scaling via cpufreq
> >
> > Signed-off-by: Alexey Charkov <alchark@gmail.com>
> > ---
> >   arch/arm64/boot/dts/rockchip/rk3588s.dtsi | 209 +++++++++++++++++++++=
+++++++++
> >   1 file changed, 209 insertions(+)
> >
> > diff --git a/arch/arm64/boot/dts/rockchip/rk3588s.dtsi b/arch/arm64/boo=
t/dts/rockchip/rk3588s.dtsi
> > index 131b9eb21398..e605be531a0f 100644
> > --- a/arch/arm64/boot/dts/rockchip/rk3588s.dtsi
> > +++ b/arch/arm64/boot/dts/rockchip/rk3588s.dtsi
> > @@ -97,6 +97,7 @@ cpu_l0: cpu@0 {
> >                       clocks =3D <&scmi_clk SCMI_CLK_CPUL>;
> >                       assigned-clocks =3D <&scmi_clk SCMI_CLK_CPUL>;
> >                       assigned-clock-rates =3D <816000000>;
> > +                     operating-points-v2 =3D <&cluster0_opp_table>;
> >                       cpu-idle-states =3D <&CPU_SLEEP>;
> >                       i-cache-size =3D <32768>;
> >                       i-cache-line-size =3D <64>;
> > @@ -116,6 +117,7 @@ cpu_l1: cpu@100 {
> >                       enable-method =3D "psci";
> >                       capacity-dmips-mhz =3D <530>;
> >                       clocks =3D <&scmi_clk SCMI_CLK_CPUL>;
> > +                     operating-points-v2 =3D <&cluster0_opp_table>;
> >                       cpu-idle-states =3D <&CPU_SLEEP>;
> >                       i-cache-size =3D <32768>;
> >                       i-cache-line-size =3D <64>;
> > @@ -135,6 +137,7 @@ cpu_l2: cpu@200 {
> >                       enable-method =3D "psci";
> >                       capacity-dmips-mhz =3D <530>;
> >                       clocks =3D <&scmi_clk SCMI_CLK_CPUL>;
> > +                     operating-points-v2 =3D <&cluster0_opp_table>;
> >                       cpu-idle-states =3D <&CPU_SLEEP>;
> >                       i-cache-size =3D <32768>;
> >                       i-cache-line-size =3D <64>;
> > @@ -154,6 +157,7 @@ cpu_l3: cpu@300 {
> >                       enable-method =3D "psci";
> >                       capacity-dmips-mhz =3D <530>;
> >                       clocks =3D <&scmi_clk SCMI_CLK_CPUL>;
> > +                     operating-points-v2 =3D <&cluster0_opp_table>;
> >                       cpu-idle-states =3D <&CPU_SLEEP>;
> >                       i-cache-size =3D <32768>;
> >                       i-cache-line-size =3D <64>;
> > @@ -175,6 +179,7 @@ cpu_b0: cpu@400 {
> >                       clocks =3D <&scmi_clk SCMI_CLK_CPUB01>;
> >                       assigned-clocks =3D <&scmi_clk SCMI_CLK_CPUB01>;
> >                       assigned-clock-rates =3D <816000000>;
> > +                     operating-points-v2 =3D <&cluster1_opp_table>;
> >                       cpu-idle-states =3D <&CPU_SLEEP>;
> >                       i-cache-size =3D <65536>;
> >                       i-cache-line-size =3D <64>;
> > @@ -194,6 +199,7 @@ cpu_b1: cpu@500 {
> >                       enable-method =3D "psci";
> >                       capacity-dmips-mhz =3D <1024>;
> >                       clocks =3D <&scmi_clk SCMI_CLK_CPUB01>;
> > +                     operating-points-v2 =3D <&cluster1_opp_table>;
> >                       cpu-idle-states =3D <&CPU_SLEEP>;
> >                       i-cache-size =3D <65536>;
> >                       i-cache-line-size =3D <64>;
> > @@ -215,6 +221,7 @@ cpu_b2: cpu@600 {
> >                       clocks =3D <&scmi_clk SCMI_CLK_CPUB23>;
> >                       assigned-clocks =3D <&scmi_clk SCMI_CLK_CPUB23>;
> >                       assigned-clock-rates =3D <816000000>;
> > +                     operating-points-v2 =3D <&cluster2_opp_table>;
> >                       cpu-idle-states =3D <&CPU_SLEEP>;
> >                       i-cache-size =3D <65536>;
> >                       i-cache-line-size =3D <64>;
> > @@ -234,6 +241,7 @@ cpu_b3: cpu@700 {
> >                       enable-method =3D "psci";
> >                       capacity-dmips-mhz =3D <1024>;
> >                       clocks =3D <&scmi_clk SCMI_CLK_CPUB23>;
> > +                     operating-points-v2 =3D <&cluster2_opp_table>;
> >                       cpu-idle-states =3D <&CPU_SLEEP>;
> >                       i-cache-size =3D <65536>;
> >                       i-cache-line-size =3D <64>;
> > @@ -348,6 +356,207 @@ l3_cache: l3-cache {
> >               };
> >       };
> >
> > +     cluster0_opp_table: opp-table-cluster0 {
> > +             compatible =3D "operating-points-v2";
> > +             opp-shared;
> > +
> > +             opp-408000000 {
> > +                     opp-hz =3D /bits/ 64 <408000000>;
> > +                     opp-microvolt =3D <675000 675000 950000>;
> > +                     clock-latency-ns =3D <40000>;
> > +             };
> > +             opp-600000000 {
> > +                     opp-hz =3D /bits/ 64 <600000000>;
> > +                     opp-microvolt =3D <675000 675000 950000>;
> > +                     clock-latency-ns =3D <40000>;
> > +             };
> > +             opp-816000000 {
> > +                     opp-hz =3D /bits/ 64 <816000000>;
> > +                     opp-microvolt =3D <675000 675000 950000>;
> > +                     clock-latency-ns =3D <40000>;
> > +             };
> > +             opp-1008000000 {
> > +                     opp-hz =3D /bits/ 64 <1008000000>;
> > +                     opp-microvolt =3D <675000 675000 950000>;
> > +                     clock-latency-ns =3D <40000>;
> > +             };
>
> It is not useful to introduce OPP with the same voltage. There is no
> gain in terms of energy efficiency as the compute capacity is linearly
> tied with power consumption (P=3DCxFxV=C2=B2) in this case.
>
> For example, opp-408 consumes 2 bogoWatts and opp-816 consumes 4
> bogoWatts (because of the same voltage).
>
> For a workload, opp-408 takes 10 sec and opp-816 takes 5 sec because it
> is twice faster.
>
> The energy consumption is:
>
> opp-408 =3D 10 x 2 =3D 20 BogoJoules
> opp-816 =3D 5 x 4 =3D 20 BogoJoules

I see, thank you. Will drop all "lower frequency - same voltage"
instances and resubmit in the next iteration.

Best regards,
Alexey