Re: [PATCH] gpiolib: acpi: Add quirk for ASUS ROG Strix G614 series

[Andy Shevchenko]

On Tue, Jun 16, 2026 at 12:58:31PM +0200, Marco Scardovi wrote:
> On Tue, Jun 16, 2026 at 12:24:51 +0200, Mika Westerberg wrote:
> > On Tue, Jun 16, 2026 at 12:58:04PM +0300, Andy Shevchenko wrote:
> > > Mika, do you see any problems with disabled (deferred?) interrupt for
> > > this type of event? I think it might break the suspend-resume (on
> > > touchpad event). Also do you remember if we ever had touchpad event to
> > > be ActiveBoth?
> > 
> > For starters, why this is ACPI event in the first place? The AML is not
> > supposed to use the touchpad.
> > 
> > Touchpad should be using just regular GpioInt(). Does it still work after
> > this patch? If yes then I think this is definitely okay.
> 
> Yes, the touchpad continues to work perfectly after this patch (even the
> suspend/resume you asked about).
> 
> The touchpad itself is initialized and driven by the i2c-hid driver using a 
> standard GpioInt() resource defined in the touchpad device's _CRS method. 
> The ACPI event handler (pin 21) registered under the GPIO controller's _AEI 
> is distinct from this.
> 
> The DMI quirk (.no_edge_events_on_boot = true) only prevents the initial 
> boot-time trigger of this ACPI event handler when the line is asserted low 
> during initialization. It does not prevent the touchpad driver from 
> requesting and receiving its interrupts.
> 
> Furthermore, unlike using the "gpiolib_acpi.ignore_interrupt=AMDI0030:00@21" 
> workaround (which would completely disable the ACPI handler), this patch still 
> keeps the ACPI event handler registered. If any edge events occur on this pin 
> post-boot (for example, if the BIOS needs to toggle touchpad states or track 
> device status), the handler will still run normally, avoiding any potential 
> breakage of runtime ACPI functionality.

Okay, perhaps this all needs to be elaborated and summarized in the commit message.

> Sidenote
> Personally I'd wait for Mario for further info: after all it's him the
> maintainer for AMD side and surely knows more than me.

Sure, I am with you on this.

-- 
With Best Regards,
Andy Shevchenko