All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] gpiolib: acpi: Add ignore_wake quirk for ASUS ZenBook UX425UA_UM425UA
@ 2026-08-21  7:06 Kivan
  2026-08-21 20:49 ` Mario Limonciello
  0 siblings, 1 reply; 4+ messages in thread
From: Kivan @ 2026-08-21  7:06 UTC (permalink / raw)
  To: Mika Westerberg, Andy Shevchenko
  Cc: Linus Walleij, Bartosz Golaszewski, Mario Limonciello,
	Hans de Goede, linux-gpio, linux-acpi, linux-kernel, Kivan

The ASUS ZenBook UX425UA_UM425UA (with AMD Ryzen 7 5700U SoC) fails
to stay suspended in s2idle (Modern Standby), prematurely waking
after 1-3 seconds with 0% hardware sleep achieved.

In the laptop's ACPI SSDT tables (SSDT15), GPIO Pin 24 (hex 0x18) is
defined in _AEI as ExclusiveAndWake targeting the PCIe bridge device
\_SB.PCI0.GPP3:

    GpioInt (Edge, ActiveLow, ExclusiveAndWake, PullNone, 0x0000,
             "\\_SB.GPIO", ...) { 0x0018 }

When the SoC powers down PCIe root ports into low-power states during
s2idle, an electrical edge transition triggers a spurious wake interrupt
on GPIO Pin 24. This signals the AMD GPIO controller (AMDI0030:00 / IRQ 7
via pinctrl_amd), aborting sleep across the system.

Booting with kernel parameter "gpiolib_acpi.ignore_wake=AMDI0030:00@24"
completely resolves the issue: amd_gpio reports ignoring the wakeup on
pin 24, and interrupt counts in /proc/interrupts remain at 0 during sleep
cycles, allowing uninterrupted s2idle until intentionally woken by the
user.

Add a DMI quirk entry to ignore wake events on AMDI0030:00@24 for this
system.

Signed-off-by: Kivan <vaknins33@gmail.com>
---
 drivers/gpio/gpiolib-acpi-quirks.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/gpio/gpiolib-acpi-quirks.c b/drivers/gpio/gpiolib-acpi-quirks.c
index a0116f0..dd4db58 100644
--- a/drivers/gpio/gpiolib-acpi-quirks.c
+++ b/drivers/gpio/gpiolib-acpi-quirks.c
@@ -392,6 +392,19 @@ static const struct dmi_system_id gpiolib_acpi_quirks[] __initconst = {
 			.ignore_wake = "VEN_0488:00@355",
 		},
 	},
+	{
+		/*
+		 * Spurious wakeups from GPP3 PCIe bridge interrupt
+		 * Found in BIOS UX425UA.301
+		 */
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "ZenBook UX425UA_UM425UA"),
+		},
+		.driver_data = &(struct acpi_gpiolib_dmi_quirk) {
+			.ignore_wake = "AMDI0030:00@24",
+		},
+	},
 	{} /* Terminating entry */
 };
 
-- 
2.55.0


^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH] gpiolib: acpi: Add ignore_wake quirk for ASUS ZenBook UX425UA_UM425UA
  2026-08-21  7:06 [PATCH] gpiolib: acpi: Add ignore_wake quirk for ASUS ZenBook UX425UA_UM425UA Kivan
@ 2026-08-21 20:49 ` Mario Limonciello
  2026-08-21 21:46   ` Aviv
  0 siblings, 1 reply; 4+ messages in thread
From: Mario Limonciello @ 2026-08-21 20:49 UTC (permalink / raw)
  To: Kivan, Mika Westerberg, Andy Shevchenko
  Cc: Linus Walleij, Bartosz Golaszewski, Hans de Goede, linux-gpio,
	linux-acpi, linux-kernel

What is connected to this GPP3?

Can you please share your acpidump somewhere?

On 8/21/26 02:06, Kivan wrote:
> The ASUS ZenBook UX425UA_UM425UA (with AMD Ryzen 7 5700U SoC) fails
> to stay suspended in s2idle (Modern Standby), prematurely waking
> after 1-3 seconds with 0% hardware sleep achieved.
> 
> In the laptop's ACPI SSDT tables (SSDT15), GPIO Pin 24 (hex 0x18) is
> defined in _AEI as ExclusiveAndWake targeting the PCIe bridge device
> \_SB.PCI0.GPP3:
> 
>      GpioInt (Edge, ActiveLow, ExclusiveAndWake, PullNone, 0x0000,
>               "\\_SB.GPIO", ...) { 0x0018 }
> 
> When the SoC powers down PCIe root ports into low-power states during
> s2idle, an electrical edge transition triggers a spurious wake interrupt
> on GPIO Pin 24. This signals the AMD GPIO controller (AMDI0030:00 / IRQ 7
> via pinctrl_amd), aborting sleep across the system.
> 
> Booting with kernel parameter "gpiolib_acpi.ignore_wake=AMDI0030:00@24"
> completely resolves the issue: amd_gpio reports ignoring the wakeup on
> pin 24, and interrupt counts in /proc/interrupts remain at 0 during sleep
> cycles, allowing uninterrupted s2idle until intentionally woken by the
> user.
> 
> Add a DMI quirk entry to ignore wake events on AMDI0030:00@24 for this
> system.
> 
> Signed-off-by: Kivan <vaknins33@gmail.com>
> ---
>   drivers/gpio/gpiolib-acpi-quirks.c | 13 +++++++++++++
>   1 file changed, 13 insertions(+)
> 
> diff --git a/drivers/gpio/gpiolib-acpi-quirks.c b/drivers/gpio/gpiolib-acpi-quirks.c
> index a0116f0..dd4db58 100644
> --- a/drivers/gpio/gpiolib-acpi-quirks.c
> +++ b/drivers/gpio/gpiolib-acpi-quirks.c
> @@ -392,6 +392,19 @@ static const struct dmi_system_id gpiolib_acpi_quirks[] __initconst = {
>   			.ignore_wake = "VEN_0488:00@355",
>   		},
>   	},
> +	{
> +		/*
> +		 * Spurious wakeups from GPP3 PCIe bridge interrupt
> +		 * Found in BIOS UX425UA.301
> +		 */
> +		.matches = {
> +			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
> +			DMI_MATCH(DMI_PRODUCT_NAME, "ZenBook UX425UA_UM425UA"),
> +		},
> +		.driver_data = &(struct acpi_gpiolib_dmi_quirk) {
> +			.ignore_wake = "AMDI0030:00@24",
> +		},
> +	},
>   	{} /* Terminating entry */
>   };
>   


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] gpiolib: acpi: Add ignore_wake quirk for ASUS ZenBook UX425UA_UM425UA
  2026-08-21 20:49 ` Mario Limonciello
@ 2026-08-21 21:46   ` Aviv
  2026-08-21 22:00     ` Mario Limonciello
  0 siblings, 1 reply; 4+ messages in thread
From: Aviv @ 2026-08-21 21:46 UTC (permalink / raw)
  To: Mario Limonciello
  Cc: Mika Westerberg, Andy Shevchenko, Linus Walleij,
	Bartosz Golaszewski, Hans de Goede, linux-gpio, linux-acpi,
	linux-kernel

Hi Mario,

Nothing is physically connected to GPP3. The bridge (00:02.1 /
\_SB.PCI0.GPP3) is unpopulated on this board (the ZenBook UX425UA has
no Ethernet or secondary PCIe devices there).

The DSDT seems to have leftover definitions for RTL8/RUSB on GPP3 from
a shared BIOS base, and routes GPIO pin 24 to it in _EVT, causing
phantom wake events during s2idle.

Here is the acpidump:
https://gist.github.com/vaknin/142f5cbad15958e9025cd81960114f6e

Thanks,
Kivan


On Fri, Aug 21, 2026 at 11:49 PM Mario Limonciello
<mario.limonciello@amd.com> wrote:
>
> What is connected to this GPP3?
>
> Can you please share your acpidump somewhere?
>
> On 8/21/26 02:06, Kivan wrote:
> > The ASUS ZenBook UX425UA_UM425UA (with AMD Ryzen 7 5700U SoC) fails
> > to stay suspended in s2idle (Modern Standby), prematurely waking
> > after 1-3 seconds with 0% hardware sleep achieved.
> >
> > In the laptop's ACPI SSDT tables (SSDT15), GPIO Pin 24 (hex 0x18) is
> > defined in _AEI as ExclusiveAndWake targeting the PCIe bridge device
> > \_SB.PCI0.GPP3:
> >
> >      GpioInt (Edge, ActiveLow, ExclusiveAndWake, PullNone, 0x0000,
> >               "\\_SB.GPIO", ...) { 0x0018 }
> >
> > When the SoC powers down PCIe root ports into low-power states during
> > s2idle, an electrical edge transition triggers a spurious wake interrupt
> > on GPIO Pin 24. This signals the AMD GPIO controller (AMDI0030:00 / IRQ 7
> > via pinctrl_amd), aborting sleep across the system.
> >
> > Booting with kernel parameter "gpiolib_acpi.ignore_wake=AMDI0030:00@24"
> > completely resolves the issue: amd_gpio reports ignoring the wakeup on
> > pin 24, and interrupt counts in /proc/interrupts remain at 0 during sleep
> > cycles, allowing uninterrupted s2idle until intentionally woken by the
> > user.
> >
> > Add a DMI quirk entry to ignore wake events on AMDI0030:00@24 for this
> > system.
> >
> > Signed-off-by: Kivan <vaknins33@gmail.com>
> > ---
> >   drivers/gpio/gpiolib-acpi-quirks.c | 13 +++++++++++++
> >   1 file changed, 13 insertions(+)
> >
> > diff --git a/drivers/gpio/gpiolib-acpi-quirks.c b/drivers/gpio/gpiolib-acpi-quirks.c
> > index a0116f0..dd4db58 100644
> > --- a/drivers/gpio/gpiolib-acpi-quirks.c
> > +++ b/drivers/gpio/gpiolib-acpi-quirks.c
> > @@ -392,6 +392,19 @@ static const struct dmi_system_id gpiolib_acpi_quirks[] __initconst = {
> >                       .ignore_wake = "VEN_0488:00@355",
> >               },
> >       },
> > +     {
> > +             /*
> > +              * Spurious wakeups from GPP3 PCIe bridge interrupt
> > +              * Found in BIOS UX425UA.301
> > +              */
> > +             .matches = {
> > +                     DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
> > +                     DMI_MATCH(DMI_PRODUCT_NAME, "ZenBook UX425UA_UM425UA"),
> > +             },
> > +             .driver_data = &(struct acpi_gpiolib_dmi_quirk) {
> > +                     .ignore_wake = "AMDI0030:00@24",
> > +             },
> > +     },
> >       {} /* Terminating entry */
> >   };
> >
>

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] gpiolib: acpi: Add ignore_wake quirk for ASUS ZenBook UX425UA_UM425UA
  2026-08-21 21:46   ` Aviv
@ 2026-08-21 22:00     ` Mario Limonciello
  0 siblings, 0 replies; 4+ messages in thread
From: Mario Limonciello @ 2026-08-21 22:00 UTC (permalink / raw)
  To: Aviv
  Cc: Mika Westerberg, Andy Shevchenko, Linus Walleij,
	Bartosz Golaszewski, Hans de Goede, linux-gpio, linux-acpi,
	linux-kernel



On 8/21/26 16:46, Aviv wrote:
> Hi Mario,
> 
> Nothing is physically connected to GPP3. The bridge (00:02.1 /
> \_SB.PCI0.GPP3) is unpopulated on this board (the ZenBook UX425UA has
> no Ethernet or secondary PCIe devices there).
> 
> The DSDT seems to have leftover definitions for RTL8/RUSB on GPP3 from
> a shared BIOS base, and routes GPIO pin 24 to it in _EVT, causing
> phantom wake events during s2idle.
> 
> Here is the acpidump:
> https://gist.github.com/vaknin/142f5cbad15958e9025cd81960114f6e
> 
> Thanks,
> Kivan

Thanks.  The dump agrees with your assessment.

Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>>
> 
> On Fri, Aug 21, 2026 at 11:49 PM Mario Limonciello
> <mario.limonciello@amd.com> wrote:
>>
>> What is connected to this GPP3?
>>
>> Can you please share your acpidump somewhere?
>>
>> On 8/21/26 02:06, Kivan wrote:
>>> The ASUS ZenBook UX425UA_UM425UA (with AMD Ryzen 7 5700U SoC) fails
>>> to stay suspended in s2idle (Modern Standby), prematurely waking
>>> after 1-3 seconds with 0% hardware sleep achieved.
>>>
>>> In the laptop's ACPI SSDT tables (SSDT15), GPIO Pin 24 (hex 0x18) is
>>> defined in _AEI as ExclusiveAndWake targeting the PCIe bridge device
>>> \_SB.PCI0.GPP3:
>>>
>>>       GpioInt (Edge, ActiveLow, ExclusiveAndWake, PullNone, 0x0000,
>>>                "\\_SB.GPIO", ...) { 0x0018 }
>>>
>>> When the SoC powers down PCIe root ports into low-power states during
>>> s2idle, an electrical edge transition triggers a spurious wake interrupt
>>> on GPIO Pin 24. This signals the AMD GPIO controller (AMDI0030:00 / IRQ 7
>>> via pinctrl_amd), aborting sleep across the system.
>>>
>>> Booting with kernel parameter "gpiolib_acpi.ignore_wake=AMDI0030:00@24"
>>> completely resolves the issue: amd_gpio reports ignoring the wakeup on
>>> pin 24, and interrupt counts in /proc/interrupts remain at 0 during sleep
>>> cycles, allowing uninterrupted s2idle until intentionally woken by the
>>> user.
>>>
>>> Add a DMI quirk entry to ignore wake events on AMDI0030:00@24 for this
>>> system.
>>>
>>> Signed-off-by: Kivan <vaknins33@gmail.com>
>>> ---
>>>    drivers/gpio/gpiolib-acpi-quirks.c | 13 +++++++++++++
>>>    1 file changed, 13 insertions(+)
>>>
>>> diff --git a/drivers/gpio/gpiolib-acpi-quirks.c b/drivers/gpio/gpiolib-acpi-quirks.c
>>> index a0116f0..dd4db58 100644
>>> --- a/drivers/gpio/gpiolib-acpi-quirks.c
>>> +++ b/drivers/gpio/gpiolib-acpi-quirks.c
>>> @@ -392,6 +392,19 @@ static const struct dmi_system_id gpiolib_acpi_quirks[] __initconst = {
>>>                        .ignore_wake = "VEN_0488:00@355",
>>>                },
>>>        },
>>> +     {
>>> +             /*
>>> +              * Spurious wakeups from GPP3 PCIe bridge interrupt
>>> +              * Found in BIOS UX425UA.301
>>> +              */
>>> +             .matches = {
>>> +                     DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
>>> +                     DMI_MATCH(DMI_PRODUCT_NAME, "ZenBook UX425UA_UM425UA"),
>>> +             },
>>> +             .driver_data = &(struct acpi_gpiolib_dmi_quirk) {
>>> +                     .ignore_wake = "AMDI0030:00@24",
>>> +             },
>>> +     },
>>>        {} /* Terminating entry */
>>>    };
>>>
>>


^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2026-08-21 22:00 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-08-21  7:06 [PATCH] gpiolib: acpi: Add ignore_wake quirk for ASUS ZenBook UX425UA_UM425UA Kivan
2026-08-21 20:49 ` Mario Limonciello
2026-08-21 21:46   ` Aviv
2026-08-21 22:00     ` Mario Limonciello

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.