[PATCH] Input: pm8941-pwrkey - report wakeup events
From: Michael Auer
Date: Sat Nov 21 2026 - 15:41:24 EST
The power key (and resin, with "wakeup-source") is a wakeup source, but
the driver never reports a wakeup event. Its wakeup_count in
/sys/class/wakeup stays 0 even after it woke the system, so user space
cannot tell a key press from other wakeups. PowerDevil, for example,
uses the wakeup counts to decide whether to turn the display on after
resume.
Report a wakeup event from the interrupt handler, as other power key
drivers do.
Signed-off-by: Michael Auer <mick@xxxxxxxx>
---
Tested on a OnePlus 6 (sdm845, PM8998) with an sdm845-mainline 7.1-rc1
based kernel: a power key wakeup from suspend moves wakeup_count from 0
to 1. On input/next it was only build tested (W=1, arm64 and x86_64
allmodconfig).
The event is reported for every interrupt, press and release. After
resume the press can be read as a release (the handler then simulates
the press), so reporting only presses could miss the wakeup.
Tool use: the analysis, the code and the commit messages were
written with an LLM coding assistant (Claude Code). I reviewed all
of it and tested it on my phone as described above.
drivers/input/misc/pm8941-pwrkey.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/input/misc/pm8941-pwrkey.c b/drivers/input/misc/pm8941-pwrkey.c
index 34b4afbb1984..3827e751297a 100644
--- a/drivers/input/misc/pm8941-pwrkey.c
+++ b/drivers/input/misc/pm8941-pwrkey.c
@@ -161,6 +161,9 @@ static irqreturn_t pm8941_pwrkey_irq(int irq, void *_data)
}
}
+ /* Let user space see that the key woke the system */
+ pm_wakeup_event(pwrkey->dev, 0);
+
err = regmap_read(pwrkey->regmap, pwrkey->baseaddr + PON_RT_STS, &sts);
if (err)
return IRQ_HANDLED;
base-commit: ab13b5fc808e043a13f861a7c1c2ba8173c22b84
--
2.43.0