2023-05-05 09:25:55

by Prashanth K

Subject: [PATCH] usb: gadget: u_serial: Add null pointer check in gserial_suspend

Consider a case where gserial_disconnect has already cleared
gser->ioport. And if gserial_suspend gets called afterwards,
it will lead to accessing of gser->ioport and thus causing
null pointer dereference.

Avoid this by adding a null pointer check. Added a static
spinlock to prevent gser->ioport from becoming null after
the newly added null pointer check.

Fixes: aba3a8d01d62 ("usb: gadget: u_serial: add suspend resume callbacks")
Signed-off-by: Prashanth K <[email protected]>
---
drivers/usb/gadget/function/u_serial.c | 13 +++++++++++--
1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/gadget/function/u_serial.c b/drivers/usb/gadget/function/u_serial.c
index a0ca47f..e5d522d 100644
--- a/drivers/usb/gadget/function/u_serial.c
+++ b/drivers/usb/gadget/function/u_serial.c
@@ -1420,10 +1420,19 @@ EXPORT_SYMBOL_GPL(gserial_disconnect);

void gserial_suspend(struct gserial *gser)
{
- struct gs_port *port = gser->ioport;
+ struct gs_port *port;
unsigned long flags;

- spin_lock_irqsave(&port->port_lock, flags);
+ spin_lock_irqsave(&serial_port_lock, flags);
+ port = gser->ioport;
+
+ if (!port) {
+ spin_unlock_irqrestore(&serial_port_lock, flags);
+ return;
+ }
+
+ spin_lock(&port->port_lock);
+ spin_unlock(&serial_port_lock);
port->suspended = true;
spin_unlock_irqrestore(&port->port_lock, flags);
}
--
2.7.4

2023-05-06 15:55:46

by Alan Stern

Subject: Re: [PATCH] usb: gadget: u_serial: Add null pointer check in gserial_suspend

On Fri, May 05, 2023 at 02:48:37PM +0530, Prashanth K wrote:
> Consider a case where gserial_disconnect has already cleared
> gser->ioport. And if gserial_suspend gets called afterwards,
> it will lead to accessing of gser->ioport and thus causing
> null pointer dereference.
>
> Avoid this by adding a null pointer check. Added a static
> spinlock to prevent gser->ioport from becoming null after
> the newly added null pointer check.
>
> Fixes: aba3a8d01d62 ("usb: gadget: u_serial: add suspend resume callbacks")
> Signed-off-by: Prashanth K <[email protected]>
> ---
> drivers/usb/gadget/function/u_serial.c | 13 +++++++++++--
> 1 file changed, 11 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/usb/gadget/function/u_serial.c b/drivers/usb/gadget/function/u_serial.c
> index a0ca47f..e5d522d 100644
> --- a/drivers/usb/gadget/function/u_serial.c
> +++ b/drivers/usb/gadget/function/u_serial.c
> @@ -1420,10 +1420,19 @@ EXPORT_SYMBOL_GPL(gserial_disconnect);
>
> void gserial_suspend(struct gserial *gser)
> {
> - struct gs_port *port = gser->ioport;
> + struct gs_port *port;
> unsigned long flags;
>
> - spin_lock_irqsave(&port->port_lock, flags);
> + spin_lock_irqsave(&serial_port_lock, flags);
> + port = gser->ioport;
> +
> + if (!port) {
> + spin_unlock_irqrestore(&serial_port_lock, flags);
> + return;
> + }
> +
> + spin_lock(&port->port_lock);
> + spin_unlock(&serial_port_lock);
> port->suspended = true;
> spin_unlock_irqrestore(&port->port_lock, flags);
> }

This looks fine to me, but I'm not a serial-gadget maintainer.
In fact, it looks like we don't have a serial-gadget maintainer.

Alan Stern