2024-02-17 13:00:27

by Konrad Dybcio

[permalink] [raw]
Subject: [PATCH] arm64: dts: qcom: sc8280xp: Add missing LMH interrupts

Hook up the interrupts that signal the Limits Management Hardware has
started some sort of throttling action.

In testing, you may notice the A78C cluster throttle IRQ fire count stays
at zero. After an hour of painful experiments on an X13s, I was able to
get that cluster to heat up near 90 degC, after which the IRQ has indeed
fired. So it stands to reason that the heat output difference between the
A78C and X1C clusters is so massive that LMH rarely decides to throttle
the "little" one based on its power metrics.

Fixes: 152d1faf1e2f ("arm64: dts: qcom: add SC8280XP platform")
Signed-off-by: Konrad Dybcio <[email protected]>
---
arch/arm64/boot/dts/qcom/sc8280xp.dtsi | 5 +++++
1 file changed, 5 insertions(+)

diff --git a/arch/arm64/boot/dts/qcom/sc8280xp.dtsi b/arch/arm64/boot/dts/qcom/sc8280xp.dtsi
index a5b194813079..daaf054efca3 100644
--- a/arch/arm64/boot/dts/qcom/sc8280xp.dtsi
+++ b/arch/arm64/boot/dts/qcom/sc8280xp.dtsi
@@ -4966,6 +4966,11 @@ cpufreq_hw: cpufreq@18591000 {
<0 0x18592000 0 0x1000>;
reg-names = "freq-domain0", "freq-domain1";

+ interrupts = <GIC_SPI 31 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 32 IRQ_TYPE_LEVEL_HIGH>;
+ interrupt-names = "dcvsh-irq-0",
+ "dcvsh-irq-1";
+
clocks = <&rpmhcc RPMH_CXO_CLK>, <&gcc GCC_GPLL0>;
clock-names = "xo", "alternate";


---
base-commit: d37e1e4c52bc60578969f391fb81f947c3e83118
change-id: 20240217-topic-8280_lmh-1e690efcf7a8

Best regards,
--
Konrad Dybcio <[email protected]>



2024-03-19 02:50:40

by Bjorn Andersson

[permalink] [raw]
Subject: Re: [PATCH] arm64: dts: qcom: sc8280xp: Add missing LMH interrupts


On Sat, 17 Feb 2024 14:00:07 +0100, Konrad Dybcio wrote:
> Hook up the interrupts that signal the Limits Management Hardware has
> started some sort of throttling action.
>
> In testing, you may notice the A78C cluster throttle IRQ fire count stays
> at zero. After an hour of painful experiments on an X13s, I was able to
> get that cluster to heat up near 90 degC, after which the IRQ has indeed
> fired. So it stands to reason that the heat output difference between the
> A78C and X1C clusters is so massive that LMH rarely decides to throttle
> the "little" one based on its power metrics.
>
> [...]

Applied, thanks!

[1/1] arm64: dts: qcom: sc8280xp: Add missing LMH interrupts
commit: 9fa6a0bad7077cac60ba98d801e13a74581ec46a

Best regards,
--
Bjorn Andersson <[email protected]>