* [PATCH v4] ACPI: button: Report wakeup key only for power button wakeups
2026-09-17 14:31 [PATCH v3] " Baorui.Liu
@ 2026-09-22 2:42 ` Baorui.Liu
0 siblings, 0 replies; 2+ messages in thread
From: Baorui.Liu @ 2026-09-22 2:42 UTC (permalink / raw)
To: rafael
Cc: lenb, robert.moore, linux-acpi, linux-kernel, acpica-devel,
Baorui.Liu
The ACPI button driver reports KEY_WAKEUP from the Power Button input
device to let userspace know that the system was resumed by a power
button wakeup.
However, reporting KEY_WAKEUP from generic system resume paths can make
userspace observe a Power Button wakeup even when the system was resumed
by a different wake source.
This is reproducible on an AMD Android 15 Xen guest. With a kernel
without this fix, a non-power-button S3 resume:
echo mem > /sys/power/state
xl trigger android s3resume
makes the Power Button input device report KEY_WAKEUP. The same test on
a kernel with this fix no longer reports KEY_WAKEUP from the Power Button
input device.
Track whether a power button event/notify is observed while the ACPI
button device is suspended, and report KEY_WAKEUP on resume only in that
case.
When the fixed power-button status bit is set during early S3 resume,
clear it to avoid a deferred KEY_POWER delivery to userspace, and mark
wakeup_pending so button resume can still report KEY_WAKEUP. Also clear
wakeup_pending before setting suspended to avoid a race with a power
button SCI during suspend entry.
Signed-off-by: Baorui.Liu <baorliu@amd.com>
---
drivers/acpi/button.c | 46 ++++++++++++++++++++++++++++++++++++++++++-
drivers/acpi/sleep.c | 27 +++++++++++--------------
include/acpi/button.h | 5 +++++
3 files changed, 62 insertions(+), 16 deletions(-)
diff --git a/drivers/acpi/button.c b/drivers/acpi/button.c
index cdbb1023a8ee..c07de417df5d 100644
--- a/drivers/acpi/button.c
+++ b/drivers/acpi/button.c
@@ -191,6 +191,7 @@ struct acpi_button {
bool last_state;
ktime_t last_time;
bool suspended;
+ bool wakeup_pending;
bool lid_state_initialized;
bool gpe_enabled;
};
@@ -397,6 +398,34 @@ static void acpi_lid_forget(struct acpi_device *adev)
}
/* Driver Interface */
+/**
+ * acpi_button_power_wakeup_pending - Mark fixed power button wakeup pending.
+ *
+ * Used when the fixed power-button status bit is observed and cleared during
+ * early S3 resume, so KEY_WAKEUP can still be reported from button resume
+ * without delivering a deferred KEY_POWER to userspace.
+ */
+void acpi_button_power_wakeup_pending(void)
+{
+ struct acpi_device *adev;
+ struct device *phys_dev;
+ struct acpi_button *button;
+
+ adev = acpi_dev_get_first_match_dev(ACPI_BUTTON_HID_POWERF, NULL, -1);
+ if (!adev)
+ return;
+
+ phys_dev = acpi_get_first_physical_node(adev);
+ if (phys_dev) {
+ button = dev_get_drvdata(phys_dev);
+ if (button && button->type == ACPI_BUTTON_TYPE_POWER)
+ button->wakeup_pending = true;
+ }
+
+ acpi_dev_put(adev);
+}
+EXPORT_SYMBOL_GPL(acpi_button_power_wakeup_pending);
+
int acpi_lid_open(void)
{
guard(mutex)(&acpi_lid_lock);
@@ -476,6 +505,9 @@ static void acpi_button_notify(acpi_handle handle, u32 event, void *data)
acpi_pm_wakeup_event(button->dev);
+ if (button->type == ACPI_BUTTON_TYPE_POWER && button->suspended)
+ button->wakeup_pending = true;
+
if (button->suspended || event == ACPI_BUTTON_NOTIFY_WAKE)
return;
@@ -498,6 +530,11 @@ static void acpi_button_notify_run(void *data)
static u32 acpi_button_event(void *data)
{
+ struct acpi_button *button = data;
+
+ if (button->type == ACPI_BUTTON_TYPE_POWER && button->suspended)
+ button->wakeup_pending = true;
+
acpi_os_execute(OSL_NOTIFY_HANDLER, acpi_button_notify_run, data);
return ACPI_INTERRUPT_HANDLED;
}
@@ -507,6 +544,12 @@ static int acpi_button_suspend(struct device *dev)
{
struct acpi_button *button = dev_get_drvdata(dev);
+ /*
+ * Clear wakeup_pending before marking suspended. Otherwise a power
+ * button SCI between the two stores could set wakeup_pending and then
+ * be overwritten here, losing a real wakeup.
+ */
+ button->wakeup_pending = false;
button->suspended = true;
return 0;
}
@@ -523,12 +566,13 @@ static int acpi_button_resume(struct device *dev)
acpi_lid_initialize_state(button);
}
- if (button->type == ACPI_BUTTON_TYPE_POWER) {
+ if (button->type == ACPI_BUTTON_TYPE_POWER && button->wakeup_pending) {
input = button->input;
input_report_key(input, KEY_WAKEUP, 1);
input_sync(input);
input_report_key(input, KEY_WAKEUP, 0);
input_sync(input);
+ button->wakeup_pending = false;
}
return 0;
}
diff --git a/drivers/acpi/sleep.c b/drivers/acpi/sleep.c
index 132a9df98471..30e268987e37 100644
--- a/drivers/acpi/sleep.c
+++ b/drivers/acpi/sleep.c
@@ -22,6 +22,7 @@
#include <linux/syscore_ops.h>
#include <asm/io.h>
#include <trace/events/power.h>
+#include <acpi/button.h>
#include "internal.h"
#include "sleep.h"
@@ -504,12 +505,11 @@ static void acpi_pm_finish(void)
acpi_resume_power_resources();
- /* If we were woken with the fixed power button, provide a small
- * hint to userspace in the form of a wakeup event on the fixed power
- * button device (if it can be found).
- *
- * We delay the event generation til now, as the PM layer requires
- * timekeeping to be running before we generate events. */
+ /*
+ * If woken by the fixed power button, provide a wakeup event on that
+ * device. KEY_WAKEUP input reporting is handled by the button driver
+ * via wakeup_pending; do not synthesize KEY_POWER here.
+ */
if (!pwr_btn_event_pending)
return;
@@ -626,14 +626,11 @@ static int acpi_suspend_enter(suspend_state_t pm_state)
/* Reprogram control registers */
acpi_leave_sleep_state_prep(acpi_state);
- /* ACPI 3.0 specs (P62) says that it's the responsibility
- * of the OSPM to clear the status bit [ implying that the
- * POWER_BUTTON event should not reach userspace ]
- *
- * However, we do generate a small hint for userspace in the form of
- * a wakeup event. We flag this condition for now and generate the
- * event later, as we're currently too early in resume to be able to
- * generate wakeup events.
+ /*
+ * ACPI 3.0 (P62): OSPM should clear the fixed power-button status bit
+ * so the event does not reach userspace as KEY_POWER. Remember that a
+ * power-button wake occurred so button resume can report KEY_WAKEUP
+ * and acpi_pm_finish() can emit a PM wakeup event.
*/
if (ACPI_SUCCESS(status) && (acpi_state == ACPI_STATE_S3)) {
acpi_event_status pwr_btn_status = ACPI_EVENT_FLAG_DISABLED;
@@ -642,8 +639,8 @@ static int acpi_suspend_enter(suspend_state_t pm_state)
if (pwr_btn_status & ACPI_EVENT_FLAG_STATUS_SET) {
acpi_clear_event(ACPI_EVENT_POWER_BUTTON);
- /* Flag for later */
pwr_btn_event_pending = true;
+ acpi_button_power_wakeup_pending();
}
}
diff --git a/include/acpi/button.h b/include/acpi/button.h
index af2fce5d2ee3..213c6cfd0264 100644
--- a/include/acpi/button.h
+++ b/include/acpi/button.h
@@ -8,11 +8,16 @@
#if IS_ENABLED(CONFIG_ACPI_BUTTON)
extern int acpi_lid_open(void);
+extern void acpi_button_power_wakeup_pending(void);
#else
static inline int acpi_lid_open(void)
{
return 1;
}
+
+static inline void acpi_button_power_wakeup_pending(void)
+{
+}
#endif /* IS_ENABLED(CONFIG_ACPI_BUTTON) */
#endif /* ACPI_BUTTON_H */
--
2.34.1
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH v4] ACPI: button: Report wakeup key only for power button wakeups
@ 2026-09-29 2:52 kernel test robot
0 siblings, 0 replies; 2+ messages in thread
From: kernel test robot @ 2026-09-29 2:52 UTC (permalink / raw)
To: oe-kbuild; +Cc: lkp
::::::
:::::: Manual check reason: "linux-review patch is more than 7 days old, verify it wasn't already superseded"
::::::
BCC: lkp@intel.com
CC: oe-kbuild-all@lists.linux.dev
In-Reply-To: <20260922024232.816-1-baorliu@amd.com>
References: <20260922024232.816-1-baorliu@amd.com>
TO: "Baorui.Liu" <baorliu@amd.com>
TO: rafael@kernel.org
CC: lenb@kernel.org
CC: linux-acpi@vger.kernel.org
CC: linux-kernel@vger.kernel.org
CC: acpica-devel@lists.linuxfoundation.org
CC: "Baorui.Liu" <baorliu@amd.com>
Hi Baorui.Liu,
kernel test robot noticed the following build errors:
[auto build test ERROR on rafael-pm/linux-next]
[also build test ERROR on linus/master v7.3-rc5 next-20260928]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/Baorui-Liu/ACPI-button-Report-wakeup-key-only-for-power-button-wakeups/20260922-104232
base: https://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git linux-next
patch link: https://lore.kernel.org/r/20260922024232.816-1-baorliu%40amd.com
patch subject: [PATCH v4] ACPI: button: Report wakeup key only for power button wakeups
:::::: branch date: 7 days ago
:::::: commit date: 7 days ago
config: x86_64-randconfig-2100-20260928 (https://download.01.org/0day-ci/archive/20260929/202609290454.LJrhMCb3-lkp@intel.com/config)
compiler: gcc-14 (Debian 14.2.0-19) 14.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20260929/202609290454.LJrhMCb3-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/r/202609290454.LJrhMCb3-lkp@intel.com/
All errors (new ones prefixed by >>):
ld: vmlinux.o: in function `acpi_suspend_enter':
>> drivers/acpi/sleep.c:643:(.text+0x147f324): undefined reference to `acpi_button_power_wakeup_pending'
vim +643 drivers/acpi/sleep.c
e9b3aba887f47f drivers/acpi/sleep/main.c Rafael J. Wysocki 2007-07-17 589
^1da177e4c3f41 drivers/acpi/sleep/main.c Linus Torvalds 2005-04-16 590 /**
2c6e33c366bff2 drivers/acpi/sleep/main.c Len Brown 2008-04-23 591 * acpi_suspend_enter - Actually enter a sleep state.
e9b3aba887f47f drivers/acpi/sleep/main.c Rafael J. Wysocki 2007-07-17 592 * @pm_state: ignored
^1da177e4c3f41 drivers/acpi/sleep/main.c Linus Torvalds 2005-04-16 593 *
50ad147aa09c82 drivers/acpi/sleep/main.c Rafael J. Wysocki 2007-07-24 594 * Flush caches and go to sleep. For STR we have to call arch-specific
50ad147aa09c82 drivers/acpi/sleep/main.c Rafael J. Wysocki 2007-07-24 595 * assembly, which in turn call acpi_enter_sleep_state().
^1da177e4c3f41 drivers/acpi/sleep/main.c Linus Torvalds 2005-04-16 596 * It's unfortunate, but it works. Please fix if you're feeling frisky.
^1da177e4c3f41 drivers/acpi/sleep/main.c Linus Torvalds 2005-04-16 597 */
2c6e33c366bff2 drivers/acpi/sleep/main.c Len Brown 2008-04-23 598 static int acpi_suspend_enter(suspend_state_t pm_state)
^1da177e4c3f41 drivers/acpi/sleep/main.c Linus Torvalds 2005-04-16 599 {
^1da177e4c3f41 drivers/acpi/sleep/main.c Linus Torvalds 2005-04-16 600 acpi_status status = AE_OK;
e9b3aba887f47f drivers/acpi/sleep/main.c Rafael J. Wysocki 2007-07-17 601 u32 acpi_state = acpi_target_sleep_state;
979f11b060c0b3 drivers/acpi/sleep.c Rafael J. Wysocki 2011-02-08 602 int error;
^1da177e4c3f41 drivers/acpi/sleep/main.c Linus Torvalds 2005-04-16 603
bb3632c6101b2f drivers/acpi/sleep.c Todd E Brandt 2014-06-06 604 trace_suspend_resume(TPS("acpi_suspend"), acpi_state, true);
e9b3aba887f47f drivers/acpi/sleep/main.c Rafael J. Wysocki 2007-07-17 605 switch (acpi_state) {
e9b3aba887f47f drivers/acpi/sleep/main.c Rafael J. Wysocki 2007-07-17 606 case ACPI_STATE_S1:
^1da177e4c3f41 drivers/acpi/sleep/main.c Linus Torvalds 2005-04-16 607 barrier();
3f6f49c7854c92 drivers/acpi/sleep.c Len Brown 2012-07-26 608 status = acpi_enter_sleep_state(acpi_state);
^1da177e4c3f41 drivers/acpi/sleep/main.c Linus Torvalds 2005-04-16 609 break;
^1da177e4c3f41 drivers/acpi/sleep/main.c Linus Torvalds 2005-04-16 610
e9b3aba887f47f drivers/acpi/sleep/main.c Rafael J. Wysocki 2007-07-17 611 case ACPI_STATE_S3:
d6a77ead21b69c drivers/acpi/sleep.c Konrad Rzeszutek Wilk 2013-05-14 612 if (!acpi_suspend_lowlevel)
d6a77ead21b69c drivers/acpi/sleep.c Konrad Rzeszutek Wilk 2013-05-14 613 return -ENOSYS;
f1a2003e22f6b5 drivers/acpi/sleep.c Rafael J. Wysocki 2011-02-08 614 error = acpi_suspend_lowlevel();
979f11b060c0b3 drivers/acpi/sleep.c Rafael J. Wysocki 2011-02-08 615 if (error)
979f11b060c0b3 drivers/acpi/sleep.c Rafael J. Wysocki 2011-02-08 616 return error;
f5ee87df7a4dab drivers/acpi/sleep.c Hanjun Guo 2021-06-02 617 pr_info("Low-level resume complete\n");
ef25ba04760159 drivers/acpi/sleep.c Rafael J. Wysocki 2015-10-07 618 pm_set_resume_via_firmware();
^1da177e4c3f41 drivers/acpi/sleep/main.c Linus Torvalds 2005-04-16 619 break;
^1da177e4c3f41 drivers/acpi/sleep/main.c Linus Torvalds 2005-04-16 620 }
bb3632c6101b2f drivers/acpi/sleep.c Todd E Brandt 2014-06-06 621 trace_suspend_resume(TPS("acpi_suspend"), acpi_state, false);
872d83d00f6702 drivers/acpi/sleep/main.c Arnaud Patard 2006-04-27 622
b6dacf63e9fb2e drivers/acpi/sleep.c Matthew Garrett 2010-05-11 623 /* This violates the spec but is required for bug compatibility. */
50ffba1bd3120b drivers/acpi/sleep.c Bob Moore 2009-02-23 624 acpi_write_bit_register(ACPI_BITREG_SCI_ENABLE, 1);
65df78473ffbf3 drivers/acpi/sleep/main.c Rafael J. Wysocki 2008-11-26 625
3f6f49c7854c92 drivers/acpi/sleep.c Len Brown 2012-07-26 626 /* Reprogram control registers */
3f6f49c7854c92 drivers/acpi/sleep.c Len Brown 2012-07-26 627 acpi_leave_sleep_state_prep(acpi_state);
c95d47a868f35c drivers/acpi/sleep/main.c Rafael J. Wysocki 2008-01-08 628
81320b20cf47de drivers/acpi/sleep.c Baorui.Liu 2026-09-22 629 /*
81320b20cf47de drivers/acpi/sleep.c Baorui.Liu 2026-09-22 630 * ACPI 3.0 (P62): OSPM should clear the fixed power-button status bit
81320b20cf47de drivers/acpi/sleep.c Baorui.Liu 2026-09-22 631 * so the event does not reach userspace as KEY_POWER. Remember that a
81320b20cf47de drivers/acpi/sleep.c Baorui.Liu 2026-09-22 632 * power-button wake occurred so button resume can report KEY_WAKEUP
81320b20cf47de drivers/acpi/sleep.c Baorui.Liu 2026-09-22 633 * and acpi_pm_finish() can emit a PM wakeup event.
872d83d00f6702 drivers/acpi/sleep/main.c Arnaud Patard 2006-04-27 634 */
c10d7a13846bff drivers/acpi/sleep.c Daniel Drake 2012-05-10 635 if (ACPI_SUCCESS(status) && (acpi_state == ACPI_STATE_S3)) {
51468a9d3a8024 drivers/acpi/sleep.c Al Stone 2013-11-20 636 acpi_event_status pwr_btn_status = ACPI_EVENT_FLAG_DISABLED;
c10d7a13846bff drivers/acpi/sleep.c Daniel Drake 2012-05-10 637
c10d7a13846bff drivers/acpi/sleep.c Daniel Drake 2012-05-10 638 acpi_get_event_status(ACPI_EVENT_POWER_BUTTON, &pwr_btn_status);
c10d7a13846bff drivers/acpi/sleep.c Daniel Drake 2012-05-10 639
18864cc4892d20 drivers/acpi/sleep.c Lv Zheng 2016-08-04 640 if (pwr_btn_status & ACPI_EVENT_FLAG_STATUS_SET) {
872d83d00f6702 drivers/acpi/sleep/main.c Arnaud Patard 2006-04-27 641 acpi_clear_event(ACPI_EVENT_POWER_BUTTON);
c10d7a13846bff drivers/acpi/sleep.c Daniel Drake 2012-05-10 642 pwr_btn_event_pending = true;
81320b20cf47de drivers/acpi/sleep.c Baorui.Liu 2026-09-22 @643 acpi_button_power_wakeup_pending();
c10d7a13846bff drivers/acpi/sleep.c Daniel Drake 2012-05-10 644 }
c10d7a13846bff drivers/acpi/sleep.c Daniel Drake 2012-05-10 645 }
872d83d00f6702 drivers/acpi/sleep/main.c Arnaud Patard 2006-04-27 646
a3627f67b1d5dd drivers/acpi/sleep/main.c Shaohua Li 2007-06-20 647 /*
22db06337f590d drivers/acpi/sleep.c Rafael J. Wysocki 2023-06-14 648 * Disable all GPE and clear their status bits before interrupts are
22db06337f590d drivers/acpi/sleep.c Rafael J. Wysocki 2023-06-14 649 * enabled. Some GPEs (like wakeup GPEs) have no handlers and this can
a508939e15356d drivers/acpi/sleep.c Chu Guangqing 2025-11-25 650 * prevent them from producing spurious interrupts.
22db06337f590d drivers/acpi/sleep.c Rafael J. Wysocki 2023-06-14 651 *
22db06337f590d drivers/acpi/sleep.c Rafael J. Wysocki 2023-06-14 652 * acpi_leave_sleep_state() will reenable specific GPEs later.
22db06337f590d drivers/acpi/sleep.c Rafael J. Wysocki 2023-06-14 653 *
22db06337f590d drivers/acpi/sleep.c Rafael J. Wysocki 2023-06-14 654 * Because this code runs on one CPU with disabled interrupts (all of
22db06337f590d drivers/acpi/sleep.c Rafael J. Wysocki 2023-06-14 655 * the other CPUs are offline at this time), it need not acquire any
22db06337f590d drivers/acpi/sleep.c Rafael J. Wysocki 2023-06-14 656 * sleeping locks which may trigger an implicit preemption point even
22db06337f590d drivers/acpi/sleep.c Rafael J. Wysocki 2023-06-14 657 * if there is no contention, so avoid doing that by using a low-level
22db06337f590d drivers/acpi/sleep.c Rafael J. Wysocki 2023-06-14 658 * library routine here.
a3627f67b1d5dd drivers/acpi/sleep/main.c Shaohua Li 2007-06-20 659 */
22db06337f590d drivers/acpi/sleep.c Rafael J. Wysocki 2023-06-14 660 acpi_hw_disable_all_gpes();
d5a64513c6a171 drivers/acpi/sleep.c Rafael J. Wysocki 2010-04-09 661 /* Allow EC transactions to happen. */
c2b46d679b30c5 drivers/acpi/sleep.c Lv Zheng 2016-08-03 662 acpi_ec_unblock_transactions();
a3627f67b1d5dd drivers/acpi/sleep/main.c Shaohua Li 2007-06-20 663
2a6b69765ad794 drivers/acpi/sleep.c Matthew Garrett 2010-05-28 664 suspend_nvs_restore();
2a6b69765ad794 drivers/acpi/sleep.c Matthew Garrett 2010-05-28 665
^1da177e4c3f41 drivers/acpi/sleep/main.c Linus Torvalds 2005-04-16 666 return ACPI_SUCCESS(status) ? 0 : -EFAULT;
^1da177e4c3f41 drivers/acpi/sleep/main.c Linus Torvalds 2005-04-16 667 }
^1da177e4c3f41 drivers/acpi/sleep/main.c Linus Torvalds 2005-04-16 668
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-09-29 2:52 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-09-29 2:52 [PATCH v4] ACPI: button: Report wakeup key only for power button wakeups kernel test robot
-- strict thread matches above, loose matches on Subject: below --
2026-09-17 14:31 [PATCH v3] " Baorui.Liu
2026-09-22 2:42 ` [PATCH v4] " Baorui.Liu
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.