| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <20190514161310.GF1824@centauri.ideon.se>
Date: Tue, 14 May 2019 18:13:10 +0200
From: Niklas Cassel <niklas.cassel@...aro.org>
To: Amit Kucheria <amit.kucheria@...aro.org>
Cc: linux-kernel@...r.kernel.org, linux-arm-msm@...r.kernel.org,
bjorn.andersson@...aro.org, andy.gross@...aro.org,
David Brown <david.brown@...aro.org>,
Li Yang <leoyang.li@....com>, Shawn Guo <shawnguo@...nel.org>,
"Raju P.L.S.S.S.N" <rplsssn@...eaurora.org>,
devicetree@...r.kernel.org, mkshah@...eaurora.org,
ulf.hansson@...aro.org
Subject: Re: [PATCHv1 8/8] arm64: dts: qcom: sdm845: Add PSCI cpuidle low
power states
On Fri, May 10, 2019 at 04:59:46PM +0530, Amit Kucheria wrote:
> From: "Raju P.L.S.S.S.N" <rplsssn@...eaurora.org>
>
> Add device bindings for cpuidle states for cpu devices.
>
> [amit: rename the idle-states to more generic names and fixups]
>
> Cc: <devicetree@...r.kernel.org>
> Cc: <mkshah@...eaurora.org>
> Signed-off-by: Raju P.L.S.S.S.N <rplsssn@...eaurora.org>
> Reviewed-by: Evan Green <evgreen@...omium.org>
> Signed-off-by: Amit Kucheria <amit.kucheria@...aro.org>
> ---
> arch/arm64/boot/dts/qcom/sdm845.dtsi | 62 ++++++++++++++++++++++++++++
> 1 file changed, 62 insertions(+)
>
> diff --git a/arch/arm64/boot/dts/qcom/sdm845.dtsi b/arch/arm64/boot/dts/qcom/sdm845.dtsi
> index 5308f1671824..2c8c54e4bd77 100644
> --- a/arch/arm64/boot/dts/qcom/sdm845.dtsi
> +++ b/arch/arm64/boot/dts/qcom/sdm845.dtsi
> @@ -119,6 +119,7 @@
> compatible = "qcom,kryo385";
> reg = <0x0 0x0>;
> enable-method = "psci";
> + cpu-idle-states = <&LITTLE_CPU_PD &LITTLE_CPU_RPD &CLUSTER_PD>;
> qcom,freq-domain = <&cpufreq_hw 0>;
> #cooling-cells = <2>;
> next-level-cache = <&L2_0>;
> @@ -136,6 +137,7 @@
> compatible = "qcom,kryo385";
> reg = <0x0 0x100>;
> enable-method = "psci";
> + cpu-idle-states = <&LITTLE_CPU_PD &LITTLE_CPU_RPD &CLUSTER_PD>;
> qcom,freq-domain = <&cpufreq_hw 0>;
> #cooling-cells = <2>;
> next-level-cache = <&L2_100>;
> @@ -150,6 +152,7 @@
> compatible = "qcom,kryo385";
> reg = <0x0 0x200>;
> enable-method = "psci";
> + cpu-idle-states = <&LITTLE_CPU_PD &LITTLE_CPU_RPD &CLUSTER_PD>;
> qcom,freq-domain = <&cpufreq_hw 0>;
> #cooling-cells = <2>;
> next-level-cache = <&L2_200>;
> @@ -164,6 +167,7 @@
> compatible = "qcom,kryo385";
> reg = <0x0 0x300>;
> enable-method = "psci";
> + cpu-idle-states = <&LITTLE_CPU_PD &LITTLE_CPU_RPD &CLUSTER_PD>;
> qcom,freq-domain = <&cpufreq_hw 0>;
> #cooling-cells = <2>;
> next-level-cache = <&L2_300>;
> @@ -178,6 +182,7 @@
> compatible = "qcom,kryo385";
> reg = <0x0 0x400>;
> enable-method = "psci";
> + cpu-idle-states = <&BIG_CPU_PD &BIG_CPU_RPD &CLUSTER_PD>;
> qcom,freq-domain = <&cpufreq_hw 1>;
> #cooling-cells = <2>;
> next-level-cache = <&L2_400>;
> @@ -192,6 +197,7 @@
> compatible = "qcom,kryo385";
> reg = <0x0 0x500>;
> enable-method = "psci";
> + cpu-idle-states = <&BIG_CPU_PD &BIG_CPU_RPD &CLUSTER_PD>;
> qcom,freq-domain = <&cpufreq_hw 1>;
> #cooling-cells = <2>;
> next-level-cache = <&L2_500>;
> @@ -206,6 +212,7 @@
> compatible = "qcom,kryo385";
> reg = <0x0 0x600>;
> enable-method = "psci";
> + cpu-idle-states = <&BIG_CPU_PD &BIG_CPU_RPD &CLUSTER_PD>;
> qcom,freq-domain = <&cpufreq_hw 1>;
> #cooling-cells = <2>;
> next-level-cache = <&L2_600>;
> @@ -220,6 +227,7 @@
> compatible = "qcom,kryo385";
> reg = <0x0 0x700>;
> enable-method = "psci";
> + cpu-idle-states = <&BIG_CPU_PD &BIG_CPU_RPD &CLUSTER_PD>;
> qcom,freq-domain = <&cpufreq_hw 1>;
> #cooling-cells = <2>;
> next-level-cache = <&L2_700>;
> @@ -228,6 +236,60 @@
> next-level-cache = <&L3_0>;
> };
> };
> +
> + idle-states {
> + entry-method = "psci";
> +
> + LITTLE_CPU_PD: little-power-down {
> + compatible = "arm,idle-state";
> + idle-state-name = "little-power-down";
> + arm,psci-suspend-param = <0x40000003>;
> + entry-latency-us = <350>;
> + exit-latency-us = <461>;
> + min-residency-us = <1890>;
> + local-timer-stop;
> + };
> +
> + LITTLE_CPU_RPD: little-rail-power-down {
> + compatible = "arm,idle-state";
> + idle-state-name = "little-rail-power-down";
> + arm,psci-suspend-param = <0x40000004>;
> + entry-latency-us = <360>;
> + exit-latency-us = <531>;
> + min-residency-us = <3934>;
> + local-timer-stop;
> + };
> +
> + BIG_CPU_PD: big-power-down {
> + compatible = "arm,idle-state";
> + idle-state-name = "big-power-down";
> + arm,psci-suspend-param = <0x40000003>;
> + entry-latency-us = <264>;
> + exit-latency-us = <621>;
> + min-residency-us = <952>;
> + local-timer-stop;
> + };
> +
> + BIG_CPU_RPD: big-rail-power-down {
> + compatible = "arm,idle-state";
> + idle-state-name = "big-rail-power-down";
> + arm,psci-suspend-param = <0x40000004>;
> + entry-latency-us = <702>;
> + exit-latency-us = <1061>;
> + min-residency-us = <4488>;
> + local-timer-stop;
> + };
> +
> + CLUSTER_PD: cluster-power-down {
> + compatible = "arm,idle-state";
> + idle-state-name = "cluster-power-down";
> + arm,psci-suspend-param = <0x400000F4>;
> + entry-latency-us = <3263>;
> + exit-latency-us = <6562>;
> + min-residency-us = <9987>;
> + local-timer-stop;
I'm surprised that this power state is not defined in downstream node qcom,pm-cluster@0
https://source.codeaurora.org/quic/la/kernel/msm-4.9/tree/arch/arm64/boot/dts/qcom/sdm845-pm.dtsi?h=msm-4.9
Also note that Ulf and Lina's cluster idling patch series:
https://patchwork.kernel.org/project/linux-arm-msm/list/?series=117055
hasn't been merged yet.
Is it really safe to add a cluster power state without this series?
If the firmware fails to enter this cluster power state, won't the
cpuidle governor continue to try to enter this power state?
Thus basically disabling cpuidle, since the governor will never try
to enter the per cpu power states?
It would be interesting with statistics of how many times we've tried
to enter this power state, together with how many times entering this
power state has failed. If this percentage is very high, then we probably
need the cluster idling patch series before enabling this power state.
> + };
> + };
> };
>
> pmu {
> --
> 2.17.1
>
Powered by blists - more mailing lists