[PATCH v1] xhci: dbgtty: fix device unregister

[PATCH v1] xhci: dbgtty: fix device unregister

[Łukasz Bartosik]

From: Łukasz Bartosik <ukaszb@chromium.org>

When DbC is disconnected then xhci_dbc_tty_unregister_device()
is called. However if there is any user space process blocked
on write to DbC terminal device then it will never be signalled
and thus stay blocked indifinitely.

This fix adds a tty_hangup() call in xhci_dbc_tty_unregister_device().
The tty_hangup() wakes up any blocked writers and causes subsequent
write attempts to DbC terminal device to fail.

Cc: stable@vger.kernel.org
Fixes: dfba2174dc42 ("usb: xhci: Add DbC support in xHCI driver")
Signed-off-by: Łukasz Bartosik <ukaszb@chromium.org>
---
 drivers/usb/host/xhci-dbgtty.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/usb/host/xhci-dbgtty.c b/drivers/usb/host/xhci-dbgtty.c
index d894081d8d15..6ea31af576c7 100644
--- a/drivers/usb/host/xhci-dbgtty.c
+++ b/drivers/usb/host/xhci-dbgtty.c
@@ -535,6 +535,13 @@ static void xhci_dbc_tty_unregister_device(struct xhci_dbc *dbc)
 
 	if (!port->registered)
 		return;
+	/*
+	 * Hang up the TTY. This wakes up any blocked
+	 * writers and causes subsequent writes to fail.
+	 */
+	if (port->port.tty)
+		tty_hangup(port->port.tty);
+
 	tty_unregister_device(dbc_tty_driver, port->minor);
 	xhci_dbc_tty_exit_port(port);
 	port->registered = false;
-- 
2.52.0.rc1.455.g30608eb744-goog

Re: [PATCH v1] xhci: dbgtty: fix device unregister

[Mathias Nyman]

Hi Łukasz

On 11/14/25 17:01, Łukasz Bartosik wrote:
> From: Łukasz Bartosik <ukaszb@chromium.org>
> 
> When DbC is disconnected then xhci_dbc_tty_unregister_device()
> is called. However if there is any user space process blocked
> on write to DbC terminal device then it will never be signalled
> and thus stay blocked indifinitely.
> 
> This fix adds a tty_hangup() call in xhci_dbc_tty_unregister_device().
> The tty_hangup() wakes up any blocked writers and causes subsequent
> write attempts to DbC terminal device to fail.

Nice catch

> 
> Cc: stable@vger.kernel.org
> Fixes: dfba2174dc42 ("usb: xhci: Add DbC support in xHCI driver")
> Signed-off-by: Łukasz Bartosik <ukaszb@chromium.org>
> ---
>   drivers/usb/host/xhci-dbgtty.c | 7 +++++++
>   1 file changed, 7 insertions(+)
> 
> diff --git a/drivers/usb/host/xhci-dbgtty.c b/drivers/usb/host/xhci-dbgtty.c
> index d894081d8d15..6ea31af576c7 100644
> --- a/drivers/usb/host/xhci-dbgtty.c
> +++ b/drivers/usb/host/xhci-dbgtty.c
> @@ -535,6 +535,13 @@ static void xhci_dbc_tty_unregister_device(struct xhci_dbc *dbc)
>   
>   	if (!port->registered)
>   		return;
> +	/*
> +	 * Hang up the TTY. This wakes up any blocked
> +	 * writers and causes subsequent writes to fail.
> +	 */
> +	if (port->port.tty)
> +		tty_hangup(port->port.tty);

I'm not a tty expert but would the tty_port_tty_vhangup(&port->port) make
sense here?

No need to check for port->port.tty, and it does all the needed locking and
tty reference counting.

It is also synchronous which should probably be ok as this is either called
from a delayed workqueue, during suspend, or remove()

Thanks
Mathias

Re: [PATCH v1] xhci: dbgtty: fix device unregister

[Łukasz Bartosik]

On Mon, Nov 17, 2025 at 4:23 PM Mathias Nyman
<mathias.nyman@linux.intel.com> wrote:
>
> Hi Łukasz
>
> On 11/14/25 17:01, Łukasz Bartosik wrote:
> > From: Łukasz Bartosik <ukaszb@chromium.org>
> >
> > When DbC is disconnected then xhci_dbc_tty_unregister_device()
> > is called. However if there is any user space process blocked
> > on write to DbC terminal device then it will never be signalled
> > and thus stay blocked indifinitely.
> >
> > This fix adds a tty_hangup() call in xhci_dbc_tty_unregister_device().
> > The tty_hangup() wakes up any blocked writers and causes subsequent
> > write attempts to DbC terminal device to fail.
>
> Nice catch
>
> >
> > Cc: stable@vger.kernel.org
> > Fixes: dfba2174dc42 ("usb: xhci: Add DbC support in xHCI driver")
> > Signed-off-by: Łukasz Bartosik <ukaszb@chromium.org>
> > ---
> >   drivers/usb/host/xhci-dbgtty.c | 7 +++++++
> >   1 file changed, 7 insertions(+)
> >
> > diff --git a/drivers/usb/host/xhci-dbgtty.c b/drivers/usb/host/xhci-dbgtty.c
> > index d894081d8d15..6ea31af576c7 100644
> > --- a/drivers/usb/host/xhci-dbgtty.c
> > +++ b/drivers/usb/host/xhci-dbgtty.c
> > @@ -535,6 +535,13 @@ static void xhci_dbc_tty_unregister_device(struct xhci_dbc *dbc)
> >
> >       if (!port->registered)
> >               return;
> > +     /*
> > +      * Hang up the TTY. This wakes up any blocked
> > +      * writers and causes subsequent writes to fail.
> > +      */
> > +     if (port->port.tty)
> > +             tty_hangup(port->port.tty);
>
> I'm not a tty expert but would the tty_port_tty_vhangup(&port->port) make
> sense here?
>
> No need to check for port->port.tty, and it does all the needed locking and
> tty reference counting.
>
> It is also synchronous which should probably be ok as this is either called
> from a delayed workqueue, during suspend, or remove()
>

Hi Mathias,

Thanks. Good point to call tty_vhangup instead of tty_hangup.

Let me test it and if it looks good I will send an updated patch.

Thanks,
Łukasz

> Thanks
> Mathias