[Bug 221449] New: Wireless gamepad stopped working through Bluetooth

[Bug 221449] New: Wireless gamepad stopped working through Bluetooth

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=221449

            Bug ID: 221449
           Summary: Wireless gamepad stopped working through Bluetooth
           Product: Drivers
           Version: 2.5
          Hardware: Intel
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P3
         Component: Bluetooth
          Assignee: linux-bluetooth@vger.kernel.org
          Reporter: arvin.ignaci@gmail.com
        Regression: No

I'm running Fedora Silverblue 44 (Universal Blue image) with kernel version
6.19.14. Some time in the last month or so, a Bluetooth gamepad I had stopped
working. It started with inputs being mapped incorrectly and certain inputs not
being recognized. Now, I don't get any input at all. The same gamepad used to
work just fine over Bluetooth and still works over USB.

The gamepad is a GuliKit KK3 set to the "Android" pairing mode; It shows up as
an "Xbox Wireless Controller" in Linux. My Bluetooth controller appears to be
an Intel AX200 modem (according to lsusb). When I connect the gamepad via
Bluetooth and try pressing some buttons or moving the sticks around, the dmesg
output is filled with these messages:

> [ 2041.273972] hid_report_raw_event: 14 callbacks suppressed
> [ 2041.273976] microsoft 0005:045E:02FD.000C: Event data for report 1 was too
> short (16 vs 15)
> [ 2041.277912] microsoft 0005:045E:02FD.000C: Event data for report 1 was too
> short (16 vs 15)
> [ 2041.280044] microsoft 0005:045E:02FD.000C: Event data for report 1 was too
> short (16 vs 15)
> [ 2041.288927] microsoft 0005:045E:02FD.000C: Event data for report 1 was too
> short (16 vs 15)
> [ 2041.300915] microsoft 0005:045E:02FD.000C: Event data for report 1 was too
> short (16 vs 15)
> [ 2041.309919] microsoft 0005:045E:02FD.000C: Event data for report 1 was too
> short (16 vs 15)

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.

[Bug 221449] Wireless gamepad stopped working through Bluetooth

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=221449

--- Comment #1 from Arvin Ignaci (arvin.ignaci@gmail.com) ---
To clarify, the gamepad stopped working after a system update, which is why I
suspected a kernel regression and posted this issue here.

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.

[Bug 221449] Wireless gamepad stopped working through Bluetooth

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=221449

The Linux kernel's regression tracker (Thorsten Leemhuis) (regressions@leemhuis.info) changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |regressions@leemhuis.info

--- Comment #2 from The Linux kernel's regression tracker (Thorsten Leemhuis) (regressions@leemhuis.info) ---
6.19.y is EOL. Does 7.0 work? Testing 7.1-rc2 would be good, too. And can you
please verify is it's really the kernel that broke things?

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.

[Bug 221449] Wireless gamepad stopped working through Bluetooth

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=221449

--- Comment #3 from Arvin Ignaci (arvin.ignaci@gmail.com) ---
Wow, didn't realize the kernel moved that fast. I'm on the latest version of
Fedora, which I thought was a pretty up-to-date distro, and 6.19 seems to be
the only version available in the repos. Looks like I'll have to figure out how
to install from source on Silverblue...

What's a good way to isolate the kernel as the issue? Install and reboot with
an older and newer version and test the behavior under each, or is there a
better way?

(Sorry if these are really basic questions; I haven't really done any kernel
development before.)

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.

[Bug 221449] Wireless gamepad stopped working through Bluetooth

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=221449

--- Comment #4 from The Linux kernel's regression tracker (Thorsten Leemhuis) (regressions@leemhuis.info) ---
Fedora is preparing switching to 7.0. For the latest kernels, try
https://fedoraproject.org/wiki/Kernel_Vanilla_Repositories

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.

[Bug 221449] Wireless gamepad stopped working through Bluetooth

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=221449

Paul Menzel (pmenzel+bugzilla.kernel.org@molgen.mpg.de) changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |pmenzel+bugzilla.kernel.org
                   |                            |@molgen.mpg.de

--- Comment #5 from Paul Menzel (pmenzel+bugzilla.kernel.org@molgen.mpg.de) ---
Next to trying the Linux kernel, could you please run `sudo journalctl` and
look at the date in the beginning and see if that goes back far enough? If the
gamepad worked over Bluetooth back then, you could look at the first occurrence
of the error message, and see if these are an indicator.

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.

[Bug 221449] Wireless gamepad stopped working through Bluetooth

[bugzilla-daemon]

https://bugzilla.kernel.org/show_bug.cgi?id=221449

--- Comment #6 from Paul Menzel (pmenzel+bugzilla.kernel.org@molgen.mpg.de) ---
> What's a good way to isolate the kernel as the issue? Install and reboot with
> an older and newer version and test the behavior under each, or is there a
> better way?

That’s the easiest way. Sometimes it’s possible to do all this in a virtual
machine and the Bluetooth controller is passed into the virtual machine. But
setting this up is unfortunately, at least last I checked, non-trivial.

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.