Bug 219836 - amd_gpio AMDI0030:00: Failed to request GPIO for pin 0x0000
Summary: amd_gpio AMDI0030:00: Failed to request GPIO for pin 0x0000
Status: RESOLVED INVALID
Alias: None
Product: Drivers
Classification: Unclassified
Component: I2C (show other bugs)
Hardware: All Linux
: P3 normal
Assignee: Drivers/I2C virtual user
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2025-03-03 23:58 UTC by Atiq
Modified: 2025-03-27 12:35 UTC (History)
4 users (show)

See Also:
Kernel Version:
Subsystem:
Regression: No
Bisected commit-id:


Attachments
ACPI Dump (1.58 MB, text/plain)
2025-03-04 00:30 UTC, Atiq
Details
kernel full log hostname anonymized (506.25 KB, text/plain)
2025-03-04 00:43 UTC, Atiq
Details

Description Atiq 2025-03-03 23:58:14 UTC
Entire line from log:

```plain
amd_gpio AMDI0030:00: Failed to request GPIO for pin 0x0000, err -16
```

On AMD Ryzen AI 9 HX 370 w/ Radeon 890M, seeing this oops on kernel-6.13.5

Distro: fedora 41
Device: Asus P16 H7606WV
Desktop: gnome
Xorg or wayland: wayland
Kernel: 6.13.5
Comment 1 Atiq 2025-03-04 00:00:27 UTC
* seeing this error (not oops) on kernel logs
Comment 2 Atiq 2025-03-04 00:17:20 UTC
let me know if you need acpi dump, dmidecode or any other info.
Comment 3 Mario Limonciello (AMD) 2025-03-04 00:18:30 UTC
Yeah you have provided very little here. We need a full kernel log, an acpidump and information about when this regressed.
Comment 4 Atiq 2025-03-04 00:30:10 UTC
Created attachment 307736 [details]
ACPI Dump

via `sudo acpidump > acpi_dump.txt`
Comment 5 Atiq 2025-03-04 00:43:48 UTC
Created attachment 307737 [details]
kernel full log hostname anonymized

via `journalctl --no-hostname --boot --no-pager > kernel_boot_log.txt`
Comment 6 Atiq 2025-03-04 00:45:23 UTC
Hi Mario,
Thank you for the reply. I attached full kernel log and an acpidump. I see this GPIO error during every boot. I remember I saw this error in older kernels as well, before upgrading to newer kernels.

Let me know if you need any further info.

Sincerely,
Atiq
Comment 7 Mario Limonciello (AMD) 2025-03-04 01:28:33 UTC
This is a BIOS bug.  There are two devices that are declaring to use GPIO 0, but with different configuration.

Here is the first one:

>         Device (BTNS)
>         {
>             Name (_HID, "ACPI0011" /* Generic Buttons Device */)  // _HID:
>             Hardware ID
>             Name (_DDN, "Generic buttons device")  // _DDN: DOS Device Name
>             Method (_CRS, 0, NotSerialized)  // _CRS: Current Resource
>             Settings
>             {
>                 Name (RBUF, ResourceTemplate ()
>                 {
>                     GpioInt (Edge, ActiveBoth, Exclusive, PullUp, 0x0000,
>                         "\\_SB.GPIO", 0x00, ResourceConsumer, ,
>                         )
>                         {   // Pin list
>                             0x0000
>                         }
>                 })
>                 Return (RBUF) /* \_SB_.BTNS._CRS.RBUF */
>             }

And here is the second one:

>     Scope (\_SB.GPIO)
>     {
>         Method (_AEI, 0, NotSerialized)  // _AEI: ACPI Event Interrupts
>         {
>             Name (BUF0, ResourceTemplate ()
>             {
>                 GpioInt (Edge, ActiveHigh, ExclusiveAndWake, PullDefault,
>                 0x1388,
>                     "\\_SB.GPIO", 0x00, ResourceConsumer, ,
>                     )
>                     {   // Pin list
>                         0x0000
>                     }

The kernel is behaving properly.  Please check for a BIOS update and report it to ASUS if it's happening on the latest BIOS.
Comment 8 Atiq 2025-03-12 22:40:00 UTC
Escalated this to Asus Support.

Case# N2503015169-0001
Comment 9 Atiq 2025-03-27 00:10:06 UTC
Asus Support said this,

> We regret to inform you that we are currently limited in the support we are
> able to provide for Linux operating systems. For the best possible software
> support, please contact the software manufacturer for further assistance.

I replied on their comment. Let's see if this moves forward!

Note You need to log in before you can comment on or make changes to this bug.