linux-kernel - Re: [PATCH 0/1] Input: xpad - add disable

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <CAB7A79w8dBiOAbpgZzNxo3whk7UUx=bxUWcMvuT+sLpuPKJAOQ@mail.gmail.com>
Date: Sun, 25 May 2025 03:41:14 -0700
From: Apoorv Parle <apparle@...il.com>
To: "A. Bosch" <progandy@...gandy.de>
Cc: Dmitry Torokhov <dmitry.torokhov@...il.com>, linux-input@...r.kernel.org, 
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 0/1] Input: xpad - add disable_xboxone module parameter

Any thoughts on considering this patch?

On Wed, May 21, 2025 at 2:45 PM Apoorv Parle <apparle@...il.com> wrote:
>
> From what I've read on udev rules so far, I'm not sure if there's a
> clean deterministic way of unbinding one driver and binding it to
> another.
> As I understand, I'll have to write a script to run for either the
> "add" action or the "bind" action. The "add" script runs asynchronous
> to kernel's binding, so it could happen before or after xpad binding
> is done; and even if I unbind it, kernel can re-bind it again while
> scanning the USB bus, which again creates a race with the script. If I
> use the "bind" action instead of "add" action, the script can recurse
> depending on timing of kernel's attempt to rebind. I tried to look for
> similar examples but couldn't find any -- just a few very high cpu
> usage bugs due to similar recursion.
> Other similar scenarios (like graphics drivers) that I tried to find,
> just blacklist whole kernel modules or used some module param.
>
> Though I'll admit, my knowledge of udev rules is somewhat limited.
>
>
> On Mon, May 19, 2025 at 11:23 PM A. Bosch <progandy@...gandy.de> wrote:
> >
> > Hi,
> >
> > could you not create udev rules to perform these actions?
> >
> > - Andreas
> >
> >
> > Am 20. Mai 2025 06:04:36 MESZ schrieb Apoorv Parle <apparle@...il.com>:
> >>
> >> Ah, yes, that's possible. But it's too hard to generally use -
> >> requires manual command line intervention each time the xbox dongle is
> >> plugged-in and/or computer reboots.There's no easy way to robustly
> >> automate this especially for layperson linux gamers.
> >> A stripped-down and hand-compiled version of xpad is at least fully
> >> automated, but only until the linux kernel is updated by the distro.
> >>
> >> On Mon, May 19, 2025 at 2:36 PM Dmitry Torokhov
> >> <dmitry.torokhov@...il.com> wrote:
> >>>
> >>>
> >>>  I understand. However you can unbind and bind individual devices to
> >>>  individual drivers via sysfs by writing into
> >>>  /sys/bus/usb/drivers/{xpad|xone}/{un}bind.
> >>>
> >>>  Thanks.
> >>>
> >>>  --
> >>>  Dmitry
> >>
> >>