[PATCH] gpiolib: acpi: Ignore AC adapter wakeup on ASUS FA507

[PATCH] gpiolib: acpi: Ignore AC adapter wakeup on ASUS FA507

[Bartu Alev]

The ASUS TUF Gaming A15 FA507 wakes from s2idle whenever the AC
adapter is plugged in or unplugged.

In the CPMGPIO0 SSDT, GPIO pin 23 (0x0017) is declared in the
\_SB.GPIO._AEI resource template as:

    GpioInt (Edge, ActiveBoth, ExclusiveAndWake, PullNone, 0x0000,
             "\\_SB.GPIO", 0x00, ResourceConsumer, ,)
             { 0x0017 }

and the corresponding \_SB.GPIO._EVT handler issues a device wake
notification for the AC adapter on pin 23 events:

    Case (0x17)
    {
        Notify (\_SB.ACAD, 0x02) // Device Wake
        Sleep (0x05)
        Notify (\_SB.ACAD, 0x80) // Status Change
    }

Both AC plug and unplug transitions therefore trigger a spurious
wakeup from s2idle. Add an ignore_wake quirk for this pin.

Signed-off-by: Bartu Alev <bartualev@gmail.com>
---
Hi,

I tested this quirk on my FA507NV and it completely resolves the
s2idle wake issue.

FA507 ACPI disassembly:
https://gitlab.com/voidvore/reverse-engineering/-/blob/fa507/fa507/FA507NV_FA507NV.318/disassembly/ssdt26-cpmgpio0.dsl#L263-269

FA506 (same pin 23 handler):
https://gitlab.com/asus-linux/reverse-engineering/-/blob/master/FA506NCR_FA506NCR.304/disassembly/ssdt8.dsl#L263-268

GA403UI (same pin 23 handler):
https://gitlab.com/asus-linux/reverse-engineering/-/blob/master/uncategorized/GA403UI/ssdt18.dsl#L267-273

I have one question: this issue is probably present across the whole
TUF series. If I acpidump more TUF models in the future and they all map
to the same GPIO pin, should the quirk list the board names one by
one (FA507, FA506, FA707, ...) or match on the product family ('ASUS TUF
Gaming A15', 'ASUS TUF Gaming A16' or 'ASUS TUF Gaming')?

The DMI outputs on my FA507NV laptop are as follows:

  board_name:     FA507NV
  board_vendor:   ASUSTeK COMPUTER INC.
  product_family: ASUS TUF Gaming A15
  product_name:   ASUS TUF Gaming A15 FA507NV_FA507NV
  
Thanks,
Bartu

 drivers/gpio/gpiolib-acpi-quirks.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/gpio/gpiolib-acpi-quirks.c b/drivers/gpio/gpiolib-acpi-quirks.c
index a0116f004975..6d56ea0d0581 100644
--- a/drivers/gpio/gpiolib-acpi-quirks.c
+++ b/drivers/gpio/gpiolib-acpi-quirks.c
@@ -392,6 +392,15 @@ static const struct dmi_system_id gpiolib_acpi_quirks[] __initconst = {
 			.ignore_wake = "VEN_0488:00@355",
 		},
 	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "FA507"),
+		},
+		.driver_data = &(struct acpi_gpiolib_dmi_quirk) {
+			.ignore_wake = "AMDI0030:00@23",
+		},
+	},
 	{} /* Terminating entry */
 };
 
-- 
2.55.0

Re: [PATCH] gpiolib: acpi: Ignore AC adapter wakeup on ASUS FA507

[Mika Westerberg]

Hi,

On Sun, Sep 27, 2026 at 04:42:12AM +0300, Bartu Alev wrote:
> The ASUS TUF Gaming A15 FA507 wakes from s2idle whenever the AC
> adapter is plugged in or unplugged.
> 
> In the CPMGPIO0 SSDT, GPIO pin 23 (0x0017) is declared in the
> \_SB.GPIO._AEI resource template as:
> 
>     GpioInt (Edge, ActiveBoth, ExclusiveAndWake, PullNone, 0x0000,
>              "\\_SB.GPIO", 0x00, ResourceConsumer, ,)
>              { 0x0017 }
> 
> and the corresponding \_SB.GPIO._EVT handler issues a device wake
> notification for the AC adapter on pin 23 events:
> 
>     Case (0x17)
>     {
>         Notify (\_SB.ACAD, 0x02) // Device Wake
>         Sleep (0x05)
>         Notify (\_SB.ACAD, 0x80) // Status Change
>     }
> 
> Both AC plug and unplug transitions therefore trigger a spurious
> wakeup from s2idle. Add an ignore_wake quirk for this pin.

I would think this is by design like that.

What is the issue? You unplug the device from AC with lid closed and it
wakes up? Userspace should put it back to sleep in these cases. In Windows
and ChromeOS there is something called "dark resume" that deals with this
but I'm not sure if generic distros have that yet.

Re: [PATCH] gpiolib: acpi: Ignore AC adapter wakeup on ASUS FA507

[Andy Shevchenko]

On Mon, Sep 28, 2026 at 11:41:56AM +0200, Mika Westerberg wrote:
> On Sun, Sep 27, 2026 at 04:42:12AM +0300, Bartu Alev wrote:
> > The ASUS TUF Gaming A15 FA507 wakes from s2idle whenever the AC
> > adapter is plugged in or unplugged.
> > 
> > In the CPMGPIO0 SSDT, GPIO pin 23 (0x0017) is declared in the
> > \_SB.GPIO._AEI resource template as:
> > 
> >     GpioInt (Edge, ActiveBoth, ExclusiveAndWake, PullNone, 0x0000,
> >              "\\_SB.GPIO", 0x00, ResourceConsumer, ,)
> >              { 0x0017 }
> > 
> > and the corresponding \_SB.GPIO._EVT handler issues a device wake
> > notification for the AC adapter on pin 23 events:
> > 
> >     Case (0x17)
> >     {
> >         Notify (\_SB.ACAD, 0x02) // Device Wake
> >         Sleep (0x05)
> >         Notify (\_SB.ACAD, 0x80) // Status Change
> >     }
> > 
> > Both AC plug and unplug transitions therefore trigger a spurious
> > wakeup from s2idle. Add an ignore_wake quirk for this pin.
> 
> I would think this is by design like that.

Yep, we don't have quirks for AC adapter events, touch pad and USB are
the majority of misbehaving cases.

With that being said, I defer this patch (you can discuss more and if
the conclusion is to include, it may be done in the next release).

> What is the issue? You unplug the device from AC with lid closed and it
> wakes up? Userspace should put it back to sleep in these cases. In Windows
> and ChromeOS there is something called "dark resume" that deals with this
> but I'm not sure if generic distros have that yet.

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH] gpiolib: acpi: Ignore AC adapter wakeup on ASUS FA507

[Mario Limonciello]

On 9/28/26 04:41, Mika Westerberg wrote:
> Hi,
> 
> On Sun, Sep 27, 2026 at 04:42:12AM +0300, Bartu Alev wrote:
>> The ASUS TUF Gaming A15 FA507 wakes from s2idle whenever the AC
>> adapter is plugged in or unplugged.
>>
>> In the CPMGPIO0 SSDT, GPIO pin 23 (0x0017) is declared in the
>> \_SB.GPIO._AEI resource template as:
>>
>>      GpioInt (Edge, ActiveBoth, ExclusiveAndWake, PullNone, 0x0000,
>>               "\\_SB.GPIO", 0x00, ResourceConsumer, ,)
>>               { 0x0017 }
>>
>> and the corresponding \_SB.GPIO._EVT handler issues a device wake
>> notification for the AC adapter on pin 23 events:
>>
>>      Case (0x17)
>>      {
>>          Notify (\_SB.ACAD, 0x02) // Device Wake
>>          Sleep (0x05)
>>          Notify (\_SB.ACAD, 0x80) // Status Change
>>      }
>>
>> Both AC plug and unplug transitions therefore trigger a spurious
>> wakeup from s2idle. Add an ignore_wake quirk for this pin.
> 
> I would think this is by design like that.
> 
> What is the issue? You unplug the device from AC with lid closed and it
> wakes up? Userspace should put it back to sleep in these cases. In Windows
> and ChromeOS there is something called "dark resume" that deals with this
> but I'm not sure if generic distros have that yet.

I tend to agree with Mika.  If there is dedicated GPIO being triggered 
on AC adapter events with a Notify(0x02) it was OEM intended to wake the 
system.

FWIW I did have a design proposal to systemd for generic distros to add 
some behavior around dark resume.

https://github.com/systemd/systemd/issues/27077

I also had started a PR but didn't really garner interest:

https://github.com/systemd/systemd/pull/37142

As you have a system this can strongly benefit, feel free to take the 
torch on trying to develop a userland solution.

Re: [PATCH] gpiolib: acpi: Ignore AC adapter wakeup on ASUS FA507

[Andy Shevchenko]

On Sun, Sep 27, 2026 at 04:42:12AM +0300, Bartu Alev wrote:
> The ASUS TUF Gaming A15 FA507 wakes from s2idle whenever the AC
> adapter is plugged in or unplugged.
> 
> In the CPMGPIO0 SSDT, GPIO pin 23 (0x0017) is declared in the
> \_SB.GPIO._AEI resource template as:
> 
>     GpioInt (Edge, ActiveBoth, ExclusiveAndWake, PullNone, 0x0000,
>              "\\_SB.GPIO", 0x00, ResourceConsumer, ,)
>              { 0x0017 }
> 
> and the corresponding \_SB.GPIO._EVT handler issues a device wake
> notification for the AC adapter on pin 23 events:
> 
>     Case (0x17)
>     {
>         Notify (\_SB.ACAD, 0x02) // Device Wake
>         Sleep (0x05)
>         Notify (\_SB.ACAD, 0x80) // Status Change
>     }
> 
> Both AC plug and unplug transitions therefore trigger a spurious
> wakeup from s2idle. Add an ignore_wake quirk for this pin.

...

> I have one question: this issue is probably present across the whole
> TUF series. If I acpidump more TUF models in the future and they all map
> to the same GPIO pin, should the quirk list the board names one by
> one (FA507, FA506, FA707, ...) or match on the product family ('ASUS TUF
> Gaming A15', 'ASUS TUF Gaming A16' or 'ASUS TUF Gaming')?
> 
> The DMI outputs on my FA507NV laptop are as follows:
> 
>   board_name:     FA507NV
>   board_vendor:   ASUSTeK COMPUTER INC.
>   product_family: ASUS TUF Gaming A15
>   product_name:   ASUS TUF Gaming A15 FA507NV_FA507NV

I can't guarantee that, so the safest approach is to go with the one-by-one
quirk(s). If you get a confirmation from the vendor (ASUSTeK) in writing
supporting your idea, the patch will be welcome, of course!

-- 
With Best Regards,
Andy Shevchenko