public inbox for linux-usb@vger.kernel.org
 help / color / mirror / Atom feed
From: Shuah Khan <skhan@linuxfoundation.org>
To: Marco Schuschnig <m.schuster91@googlemail.com>,
	valentina.manea.m@gmail.com, shuah@kernel.org
Cc: linux-usb@vger.kernel.org, Shuah Khan <skhan@linuxfoundation.org>
Subject: Re: usbip: Kernel oops when trying to bind devices from Windows client
Date: Tue, 6 Jan 2026 15:07:29 -0700	[thread overview]
Message-ID: <7220f3ee-141a-4fb3-b515-c2961ebd77ea@linuxfoundation.org> (raw)
In-Reply-To: <CAEhB=Bt13wr1+cc1rfzwPxnEEuawMgTew=RJic2FsVb=Gccm+A@mail.gmail.com>

On 1/4/26 16:13, Marco Schuschnig wrote:
> Hello all,
> 
> I am trying to get usbip working with a Raspberry Pi Zero 2 W
> (Raspberry Pi OS / Debian Trixie, kernel 6.2.47) as host and a Windows
> 10 machine as client (using usbip-win2 0.9.7.3). The device to be
> shared is a "145f:0276 Trust GXT 165 Gaming Mouse".

Are you using vhci_hcd equivalent on Windows and usbip_host on
Linux? This won't work because usbip_host and vhci_hcd have to
be both Linux systems.

> 
> Upon attaching from the Windows machine, the Raspberry Pi
> kernel-oopses pretty much immediately. Sometimes, the Pi locks up hard
> until it is either power-cycled or the watchdog kicks in, sometimes it
> keeps running. In the latter case, when attempting to detach on the
> Windows machine, another kernel oops follows like in the attached
> stacktrace (sec. 158).
> 
> The issue is fully reproducible, the attached stacktrace was produced
> with usbip_debug_flag=0xFFFFFFFF.
> 
> An issue with the hardware can be ruled out, as when using a Raspberry
> Pi 5 as client (same kernel version) the usbip connection works
> flawlessly. 

Yes this is a supported configuration.

Incidentally, when using the Raspberry Pi 5 as host
> together with the Windows 10 machine as client, usbip works just fine.

I am surprised this works. What's the kernel version Raspberry Pi 5?
> I have tested with the same results an USB keyboard and an Android
> smartphone to rule out a misbehaving USB device.
> 
> It might be possible that the culprit is something Windows does to USB
> devices or a bug in the usbip-win2 package, but even if Windows sends
> corrupt USB packets, these should IMHO not lead to a kernel-oops.

The reason is Windows usbip client and Linux host aren't compatible.
We don't test this case and support this case.

thanks,
-- Shuah

      parent reply	other threads:[~2026-01-06 22:07 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-01-04 23:13 usbip: Kernel oops when trying to bind devices from Windows client Marco Schuschnig
2026-01-05  8:08 ` Greg KH
2026-01-05  8:38   ` Marco Schuschnig
2026-01-05  8:52     ` Greg KH
2026-01-06 22:07 ` Shuah Khan [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=7220f3ee-141a-4fb3-b515-c2961ebd77ea@linuxfoundation.org \
    --to=skhan@linuxfoundation.org \
    --cc=linux-usb@vger.kernel.org \
    --cc=m.schuster91@googlemail.com \
    --cc=shuah@kernel.org \
    --cc=valentina.manea.m@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox