Re: Failed IRQ assignment for INT0002 on Braswell

From: Hans de Goede
Date: Wed Nov 22 2017 - 05:48:57 EST


Hi,

On 21-11-17 23:35, Oleksandr Natalenko wrote:
Hi, Hans.

v4.13 kernel introduced new fancy warning in the dmesg:

===
kernel: acpi INT0002:00: Device [GPED] is in always present list
kernel: genirq: Flags mismatch irq 9. 00010084 (INT0002) vs. 00002080 (acpi)
kernel: INT0002 Virtual GPIO INT0002:00: Error requesting IRQ 9: -16
kernel: INT0002 Virtual GPIO: probe of INT0002:00 failed with error -16
===

Looking at git log, I've found that this driver was indeed introduced in v4.13
by the following commit:

===
commit 63dada87f7ef7d4a536765c816fbbe7c4b9f3c85
Author: Hans de Goede <hdegoede@xxxxxxxxxx>
Date: Mon Jun 12 22:55:46 2017 +0200
platform/x86: Add driver for ACPI INT0002 Virtual GPIO device
===

I've checked DSDT disassembly for this device, and here is relevant snippet
[1]. The only thing I currently understand there is that IRQ 9 is hard coded
(line 64), but I'm not sure where those flags come from.

The hardware in J3710 CPU on ASRock J3710-ITX motherboard.

Next, I've found that you also faced this warning and posted some RFC patch
[2], but unfortunately I do not see what that discussion ended up with.

It is not that I really need GPIO on this board, but it would be nice to get
rid of this warning. Could you please point me to a possible fix or a way for
further investigation?

This should be fixed by:

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/kernel/irq?id=382bd4de61827dbaaf5fb4fb7b1f4be4a86505e7

Which is in 4.13, but the trigger-type does not seem to be the problem in
your case, the problem likely is the ONESHOT flag:

#define IRQF_ONESHOT 0x00002000

Which appears to be set in the flags for the acpi irq handler:

> kernel: genirq: Flags mismatch irq 9. 00010084 (INT0002) vs. 00002080 (acpi)

But that irq is requested here:

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/acpi/osl.c#n570

:

if (request_irq(irq, acpi_irq, IRQF_SHARED, "acpi", acpi_irq)) {
printk(KERN_ERR PREFIX "SCI (IRQ%d) allocation failed\n", irq);
...

And IRQF_ONESHOT is not passed, so I do not understand where the 00002000 in the
acpi irq handler flags is coming from ...

Regards,

Hans