Re: [PATCH] HID: universal-pidff: stop the device when force-feedback init fails

Thanks for looking at it.  I agree with the rule, but I don't think those t=
wo
exits can take the stop, and I would rather explain why than send a v2 that
changes them.

After hid_hw_start() there is exactly one error return in
universal_pidff_probe(), the init_function() one this patch changes.  The t=
wo
exits on lines 92 and 101 return 0, which means the probe succeeds, and tha=
t
is the whole difference between them.  Because they succeed, hdev->driver
stays set, and since universal_pidff has no .remove callback,
hid_device_remove() falls back to the default remove, which is hid_hw_stop(=
).
The device those two exits leave running is therefore already stopped for u=
s
when it is unbound or unplugged, and nothing leaks there.  The leak exists
only on the path where the probe fails, because __hid_device_probe() clears
hdev->driver on failure and hid_device_remove() then skips the unwind
entirely.

If we call hid_hw_stop() on those two exits as well, it ends up running twi=
ce
on the same device, and usbhid_stop() does not survive that.  It clears the
three URB pointers on the way out - that is what the "don't mess up next
start" comment is about - but the buffers it releases immediately afterward=
s
through hid_free_buffers() are left dangling: inbuf, outbuf and ctrlbuf go
back through usb_free_coherent() and usbhid->cr through kfree(), and none o=
f
those four pointers is cleared.  There is no HID_STARTED check at the top o=
f
usbhid_stop() either, so the second call frees all four again.  Every unplu=
g
of a device that took one of those exits would be a double free, which is a
good deal worse than the leak this patch is fixing.

The other problem is that stopping there takes down the very device those
exits exist for.  hid_hw_stop() calls hid_disconnect(), which unregisters t=
he
input device through hidinput_disconnect() and the character device through
hidraw_disconnect().  The additional "device" you mention, the one that is
there just for the extra buttons and axes, is by definition the one without=
 a
PID usage page in its descriptor, so it is exactly the one that takes the e=
xit
on line 92.  Stopping it would leave this driver bound to a device with no
input node and no hidraw node at all, and returning an error there instead =
of
0 does not help either, because hid-generic will not adopt the interface:
hid_generic_match() walks the bus and gives up as soon as it finds another
driver that matches the device.  Either way the buttons and axes disappear,
which I do not think is what you are after.

2026=EB=85=84 8=EC=9B=94 3=EC=9D=BC (=EC=9B=94) =EC=98=A4=EC=A0=84 5:41, To=
masz Paku=C5=82a <[email protected]>=EB=8B=98=EC=9D=B4 =EC=
=9E=91=EC=84=B1:
>
> On Wed, 2026-07-29 at 23:24 +0900, Baul Lee wrote:
> > universal_pidff_probe() starts the device with hid_hw_start() and then,=
 if
> > force-feedback initialisation fails, returns the error through a label =
that
> > only does "return error".  The device is left started.
> >
> > The HID core does not unwind on the driver's behalf.  __hid_device_prob=
e()
> > releases the devres group, closes the report and clears hdev->driver:
> >
> >       if (ret) {
> >               devres_release_group(&hdev->dev, hdev->devres_group_id);
> >               hid_close_report(hdev);
> >               hdev->driver =3D NULL;
> >       }
> >
> > The hidraw character device that hid_hw_start() registered through
> > hid_connect() is allocated with kzalloc() and added with cdev_device_ad=
d(),
> > so it is not devres-managed and survives that.  With hdev->driver NULL,
> > hid_device_remove() skips hid_hw_stop() as well, because it only unwind=
s
> > while a driver is still attached.  The registration therefore outlives =
the
> > device on both paths.
> >
> > Opening the surviving /dev/hidrawX writes into freed memory.  KASAN rep=
orts
> > a use-after-free write from hidraw_open() -> hid_hw_open() -> the
> > transport's open callback, which takes a spinlock inside the freed obje=
ct.
> > A descriptor that carries a PID usage page and no input reports is enou=
gh:
> > hidraw claims the device so hid_hw_start() succeeds, while hid->inputs
> > stays empty so force-feedback init fails.  The other failure returns in
> > hid_pidff_init_with_quirks() - no output reports, an allocation failure=
,
> > pidff_init_fields(), pidff_check_autocenter(), an unusable effect count=
,
> > input_ff_create() - all reach the same label.
> >
> > Stop the device on that path.  hid-dr.c and hid-emsff.c, which start th=
e
> > device with the same HID_CONNECT_DEFAULT & ~HID_CONNECT_FF mask, alread=
y do
> > this.  The two earlier gotos must keep returning without hid_hw_stop(),
> > since neither has a started device, so give the path that fails after t=
he
> > start its own label.
> >
> > Discovered by XBOW, triaged by Baul Lee <[email protected]>
> >
> > Fixes: f06bf8d94fff ("HID: Add hid-universal-pidff driver and supported=
 device ids")
> > Cc: [email protected]
> > Signed-off-by: Baul Lee <[email protected]>
> > ---
> >  drivers/hid/hid-universal-pidff.c | 4 +++-
> >  1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/hid/hid-universal-pidff.c b/drivers/hid/hid-univer=
sal-pidff.c
> > index 549dac555d40..60180467a0bf 100644
> > --- a/drivers/hid/hid-universal-pidff.c
> > +++ b/drivers/hid/hid-universal-pidff.c
> > @@ -104,12 +104,14 @@ static int universal_pidff_probe(struct hid_devic=
e *hdev,
> >       error =3D init_function(hdev, id->driver_data);
> >       if (error) {
> >               hid_warn(hdev, "Error initialising force feedback\n");
> > -             goto err;
> > +             goto err_stop;
> >       }
> >
> >       hid_info(hdev, "Universal pidff driver loaded successfully!");
> >
> >       return 0;
> > +err_stop:
> > +     hid_hw_stop(hdev);
> >  err:
> >       return error;
> >  }
>
> Actually, now that I look at it, this isn't the only place where the
> stop should happen. Any error after successful hid_hw_start() should
> trigger hid_hw_stop().
>
> So for this to truly matter, you should replace
>         return 0;
> with
>         goto err_stop;
> on lines 92 and 101. Just add error value clearing so error =3D 0 because
> even though we're exiting there, it's not an error. Some devices just
> show up as multiple separate HID devices.
>
> Tomasz
>