[PATCH v3 3/4] arm64: dts: mediatek: add Genio 720-EVK board

zoan37 agentzoan at gmail.com
Sat Oct 10 17:09:22 PDT 2026


Hi Louis-Alexis,

A note from using this series' mt8189.dtsi as the base for an MT8189
Chromebook (Lenovo IdeaPad Slim 3 Chromebook, board "quigon" of the
skywalker family) on next-20261008.

The dtsi has only the cpu_off_l/cpu_off_b idle states (PSCI
CPU_SUSPEND 0x00010000). This firmware has no PSCI SYSTEM_SUSPEND, so
suspend is s2idle, and with those states the SoC never really sleeps:
the EC doesn't see the sleep signal and the RTC alarm doesn't wake it.
ChromeOS's mt8189.dtsi adds a system-level state that only s2idle uses
(infinite min-residency); when the last CPU enters it, TF-A and the SPM
put the SoC to sleep:

	cpu_s2idle: cpu-s2idle {
		compatible = "arm,idle-state";
		arm,psci-suspend-param = <0x020180ff>;
		local-timer-stop;
		entry-latency-us = <10000>;
		exit-latency-us = <10000>;
		min-residency-us = <0xffffffff>;
	};

with every CPU listing it after its cpu_off state
(cpu-idle-states = <&cpu_off_l &cpu_s2idle>, <&cpu_off_b &cpu_s2idle>
on the big cores). With that, suspend is real SPM sleep on this board:
the EC sees S0->S3->S0, RTC, lid and keyboard wake work, it passes
repeated suspend/resume cycles (20/20, then 10/10 twice), and a
10-minute sleep used about 20 mWh of battery.

I can only speak for the Chromebook firmware, so I don't know whether
the Genio EVKs' TF-A implements this power level, but it may be worth
adding to the SoC dtsi if it does.

The board files are at
https://github.com/zoan37/lenovo-slim3-chromebook-omarchy
(port/quigon/mainline/mt8189-quigon-suspend.dtsi).

This was found and tested with help from an AI assistant (Claude).

Thanks,
zoan37



More information about the Linux-mediatek mailing list