Devicetree
 help / color / mirror / Atom feed
* [PATCH v2] arm64: dts: st: add PSCI idle state to STM32MP23 CPU domains
@ 2026-11-19 17:42 Patrick Delaunay
  2026-11-19 17:51 ` sashiko-bot
  0 siblings, 1 reply; 2+ messages in thread
From: Patrick Delaunay @ 2026-11-19 17:42 UTC (permalink / raw)
  To: Rob Herring, Krzysztof Kozlowski, Conor Dooley, Maxime Coquelin,
	Alexandre Torgue, Amelie Delaunay
  Cc: Linux STM32, linux-arm-kernel, Patrick Delaunay, devicetree,
	linux-kernel

Define a PSCI CPU suspend state and reference it from both STM32MP23
CPU power domains.

In OSI mode, PSCI power domains are initialized as off. The idle-state
allows the CPU power-domain governor to manage them, avoiding probe
stalls for devices that depend on the cluster power domain, including
SDMMC1.

Fixes: e9b03ef21386 ("arm64: dts: st: introduce stm32mp23 SoCs family")
Link: https://lore.kernel.org/all/20261030102907.1.I0ef2a7ac525ef18fa683368382d93e98cc36ee25@changeid/
Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com>
---
Hi,

It is v2 of previous patch [1], with only update of commit message.

I add a link on previous patch [2] which changes the timer node with
compatible arm,armv8-timer, see Sashiko issue [3] for details.

[1] [PATCH] arm64: dts: st: add PSCI idle state to STM32MP23 CPU domains
https://patchwork.kernel.org/project/linux-arm-kernel/patch/20261104133023.1.If2b6b99285c1e60ca11b6e7fccf7d9b130ca20dd@changeid/
https://lore.kernel.org/all/20261104133023.1.If2b6b99285c1e60ca11b6e7fccf7d9b130ca20dd@changeid/

[2] [PATCH] arm64: dts: st: mark STM32MP23 timer as stopping in suspend
https://patchwork.kernel.org/project/linux-arm-kernel/patch/20261030102907.1.I0ef2a7ac525ef18fa683368382d93e98cc36ee25@changeid/
https://lore.kernel.org/all/20261030102907.1.I0ef2a7ac525ef18fa683368382d93e98cc36ee25@changeid/

[3] [High] arm64: dts: st: stm32mp231 cpu-sleep local-timer-stop conflicts with always-on
https://lore.kernel.org/all/sashiko-outbox-197300@kernel.org/

Patrick


Changes in v2:
- Add "Fixes", requested by Amelie Delaunay
- Add "Link" to arm generic timer patch

 arch/arm64/boot/dts/st/stm32mp231.dtsi | 14 ++++++++++++++
 arch/arm64/boot/dts/st/stm32mp233.dtsi |  1 +
 2 files changed, 15 insertions(+)

diff --git a/arch/arm64/boot/dts/st/stm32mp231.dtsi b/arch/arm64/boot/dts/st/stm32mp231.dtsi
index fefbf6f4f153..27231e140fbb 100644
--- a/arch/arm64/boot/dts/st/stm32mp231.dtsi
+++ b/arch/arm64/boot/dts/st/stm32mp231.dtsi
@@ -24,6 +24,19 @@ cpu0: cpu@0 {
 			power-domains = <&cpu0_pd>;
 			power-domain-names = "psci";
 		};
+
+		idle-states {
+			entry-method = "psci";
+
+			cpu_sleep: cpu-sleep {
+				compatible = "arm,idle-state";
+				arm,psci-suspend-param = <0x00000002>;
+				local-timer-stop;
+				entry-latency-us = <150>;
+				exit-latency-us = <200>;
+				min-residency-us = <1000>;
+			};
+		};
 	};
 
 	arm-pmu {
@@ -123,6 +136,7 @@ psci {
 
 		cpu0_pd: power-domain-cpu0 {
 			#power-domain-cells = <0>;
+			domain-idle-states = <&cpu_sleep>;
 			power-domains = <&cluster_pd>;
 		};
 
diff --git a/arch/arm64/boot/dts/st/stm32mp233.dtsi b/arch/arm64/boot/dts/st/stm32mp233.dtsi
index 78f4059fca5d..cebefd99906f 100644
--- a/arch/arm64/boot/dts/st/stm32mp233.dtsi
+++ b/arch/arm64/boot/dts/st/stm32mp233.dtsi
@@ -26,6 +26,7 @@ arm-pmu {
 	psci {
 		cpu1_pd: power-domain-cpu1 {
 			#power-domain-cells = <0>;
+			domain-idle-states = <&cpu_sleep>;
 			power-domains = <&cluster_pd>;
 		};
 	};
---
base-commit: aebb7274476403040d582c61478df8f4a7bf9cd3
branch: master

-- 
2.43.0


^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [PATCH v2] arm64: dts: st: add PSCI idle state to STM32MP23 CPU domains
  2026-11-19 17:42 [PATCH v2] arm64: dts: st: add PSCI idle state to STM32MP23 CPU domains Patrick Delaunay
@ 2026-11-19 17:51 ` sashiko-bot
  0 siblings, 0 replies; 2+ messages in thread
From: sashiko-bot @ 2026-11-19 17:51 UTC (permalink / raw)
  To: Patrick Delaunay; +Cc: devicetree, robh, conor+dt

Thank you for your contribution! Sashiko AI review found 1 potential
issue to consider:
- [High] arm64: dts: st: stm32mp231 cpu-sleep local-timer-stop conflicts
  with always-on
--

commit 7e079de649dbd4d8c4228cc0cad8b0197a5e36f1
Author: Patrick Delaunay <patrick.delaunay@foss.st.com>

arm64: dts: st: add PSCI idle state to STM32MP23 CPU domains

This commit adds a PSCI cpu-sleep idle state and references it from the
STM32MP23 CPU power domains so the power-domain governor can manage them in
OSI mode.

Link: https://lore.kernel.org/all/20261030102907.1.I0ef2a7ac525ef18fa683368382d93e98cc36ee25@changeid/

> diff --git a/arch/arm64/boot/dts/st/stm32mp231.dtsi b/arch/arm64/boot/dts/st/stm32mp231.dtsi
> index fefbf6f4f1537..27231e140fbb5 100644
> --- a/arch/arm64/boot/dts/st/stm32mp231.dtsi
> +++ b/arch/arm64/boot/dts/st/stm32mp231.dtsi
> @@ -24,6 +24,19 @@ cpu0: cpu@0 {
>  			power-domains = <&cpu0_pd>;
>  			power-domain-names = "psci";
>  		};
> +
> +		idle-states {
> +			entry-method = "psci";
> +
> +			cpu_sleep: cpu-sleep {
> +				compatible = "arm,idle-state";
> +				arm,psci-suspend-param = <0x00000002>;
> +				local-timer-stop;
> +				entry-latency-us = <150>;
> +				exit-latency-us = <200>;
> +				min-residency-us = <1000>;
> +			};
> +		};
>  	};

[Severity: High]
Does setting local-timer-stop on cpu_sleep while the arm,armv8-timer node in
stm32mp231.dtsi still has always-on cause the CPU to enter PSCI suspend with
its local timer stopped and no broadcast wakeup timer armed?

arch/arm64/boot/dts/st/stm32mp231.dtsi {
    ...
	timer {
		compatible = "arm,armv8-timer";
		interrupt-parent = <&intc>;
		interrupts = <GIC_PPI 13 (GIC_CPU_MASK_SIMPLE(1) | IRQ_TYPE_LEVEL_LOW)>,
			     <GIC_PPI 14 (GIC_CPU_MASK_SIMPLE(1) | IRQ_TYPE_LEVEL_LOW)>,
			     <GIC_PPI 11 (GIC_CPU_MASK_SIMPLE(1) | IRQ_TYPE_LEVEL_LOW)>,
			     <GIC_PPI 10 (GIC_CPU_MASK_SIMPLE(1) | IRQ_TYPE_LEVEL_LOW)>;
		always-on;
	};
    ...
}

Because always-on is present on the timer node, arch_timer_of_init() sets
arch_timer_c3stop to false and __arch_timer_setup() omits
CLOCK_EVT_FEAT_C3STOP on the local clockevent device.  When
cpuidle_enter_state() enters cpu_sleep, tick_broadcast_enter() ->
tick_broadcast_oneshot_control() sees CLOCK_EVT_FEAT_C3STOP clear and
returns 0 immediately without arming a broadcast timer or returning -EBUSY
to fall back to WFI, leaving the CPU asleep until an external non-timer
interrupt arrives.

The commit message adds a Link: to the separate patch "arm64: dts: st: mark
STM32MP23 timer as stopping in suspend", but that prerequisite is not applied
in this tree (and is not pulled in when backporting via Fixes:).

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261119184209.v2.1.If2b6b99285c1e60ca11b6e7fccf7d9b130ca20dd@changeid?part=1

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2026-11-19 17:51 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-19 17:42 [PATCH v2] arm64: dts: st: add PSCI idle state to STM32MP23 CPU domains Patrick Delaunay
2026-11-19 17:51 ` sashiko-bot

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox