Re: [PATCH] USB: serial: simple: narrow OWON match to avoid breaking VDS1022

From: Huang Wei

Date: Wed Nov 18 2026 - 20:54:52 EST


On Tue, Nov 17, 2026 at 03:05:09PM +0100, Johan Hovold wrote:
> As there apparently are now users of the in-kernel driver and libusb
> has a mechanism for unbinding kernel drivers when userspace wants to
> handle a device itself, my preference is option 1.

Understood - thanks for the direction. I'll withdraw the v2 (Greg:
please drop it) and leave the owon match as is.

> I've tried to skim these threads and IIUC the only issue here is that
> some java-wrapper around libusb doesn't provide the unbind mechanism?

That matches the reporter's setup. He uses the software packaged at
https://github.com/florentbr/Owon-VDS1022, which installs the
official Java application (with a Python API as well), and its udev
rule only sets device permissions:

SUBSYSTEMS=="usb", ATTRS{idVendor}=="5345", ATTRS{idProduct}=="1234", MODE="0666"

There is no unbind on either the Java or the Python side, and the
reporter confirmed the application could access the device only
after blacklisting the usb-serial module.

> And there should be nothing preventing a userspace application from
> using the tty interface even if a device does not speak SCPI, right?

At the byte level I agree - the generic usbserial tty is a
transparent channel. In practice the VDS1022 tooling is written
against libusb only (the SCPI subset is served through the
application's socket, not through a serial port), so existing users
still depend on an udev unbind rule or module blacklisting.

I'll update bug 220193 with the outcome and the workaround so the
reporter can close it.

Thanks,
Huang Wei