[PATCH] fdti_sio: Fix atomicity violation in get_serial_info()

Qiu-ji Chen posted 1 patch 2 weeks, 3 days ago
drivers/usb/serial/ftdi_sio.c | 2 ++
1 file changed, 2 insertions(+)
[PATCH] fdti_sio: Fix atomicity violation in get_serial_info()
Posted by Qiu-ji Chen 2 weeks, 3 days ago
Our static checker found a bug where set_serial_info() uses a mutex, but 
get_serial_info() does not. Fortunately, the impact of this is relatively 
minor. It doesn't cause a crash or any other serious issues. However, if a 
race condition occurs between set_serial_info() and get_serial_info(), 
there is a chance that the data returned by get_serial_info() will be 
meaningless.

Similar issues have been submitted or confirmed: 
https://lore.kernel.org/all/20241002125845.785464707@linuxfoundation.org/ 
https://lore.kernel.org/all/20241106095819.15194-1-chenqiuji666@gmail.com/

Signed-off-by: Qiu-ji Chen <chenqiuji666@gmail.com>
Fixes: 3ae36bed3a93 ("fdti_sio: switch to ->[sg]et_serial()")
---
 drivers/usb/serial/ftdi_sio.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c
index c6f17d732b95..e07c5e3eb18c 100644
--- a/drivers/usb/serial/ftdi_sio.c
+++ b/drivers/usb/serial/ftdi_sio.c
@@ -1443,9 +1443,11 @@ static void get_serial_info(struct tty_struct *tty, struct serial_struct *ss)
 	struct usb_serial_port *port = tty->driver_data;
 	struct ftdi_private *priv = usb_get_serial_port_data(port);
 
+	mutex_lock(&priv->cfg_lock);
 	ss->flags = priv->flags;
 	ss->baud_base = priv->baud_base;
 	ss->custom_divisor = priv->custom_divisor;
+	mutex_unlock(&priv->cfg_lock);
 }
 
 static int set_serial_info(struct tty_struct *tty, struct serial_struct *ss)
-- 
2.34.1
Re: [PATCH] fdti_sio: Fix atomicity violation in get_serial_info()
Posted by Johan Hovold 2 weeks, 3 days ago
On Wed, Nov 06, 2024 at 06:18:09PM +0800, Qiu-ji Chen wrote:
> Our static checker found a bug where set_serial_info() uses a mutex, but 
> get_serial_info() does not. Fortunately, the impact of this is relatively 
> minor. It doesn't cause a crash or any other serious issues. However, if a 
> race condition occurs between set_serial_info() and get_serial_info(), 
> there is a chance that the data returned by get_serial_info() will be 
> meaningless.

Perhaps "inconsistent" is a better word here.

> Similar issues have been submitted or confirmed: 
> https://lore.kernel.org/all/20241002125845.785464707@linuxfoundation.org/ 
> https://lore.kernel.org/all/20241106095819.15194-1-chenqiuji666@gmail.com/

This should go under the --- line so that it's not included in the git
logs.

> Signed-off-by: Qiu-ji Chen <chenqiuji666@gmail.com>
> Fixes: 3ae36bed3a93 ("fdti_sio: switch to ->[sg]et_serial()")

And please use the expected patch prefix:

	USB: serial: ftdi_sio: ...

Johan