Re: [PATCH tty-next 21/22] serial: core: Remove superfluous ldisc flush from uart_close()

[Peter Hurley]

On 06/16/2014 09:17 AM, Peter Hurley wrote:
> The tty_ldisc_flush() after port hardware shutdown is unnecessary;
> the ldisc flush was just performed before the hardware shutdown
> in tty_port_close_start() and the ldisc will be released when
> uart_close() returns (because the last port close implies the
> last tty close).
>
> Signed-off-by: Peter Hurley <peter@xxxxxxxxxxxxxxxxxx>
> ---
>   drivers/tty/serial/serial_core.c | 3 ---
>   1 file changed, 3 deletions(-)
>
> diff --git a/drivers/tty/serial/serial_core.c b/drivers/tty/serial/serial_core.c
> index 212ee07..15212d7 100644
> --- a/drivers/tty/serial/serial_core.c
> +++ b/drivers/tty/serial/serial_core.c
> @@ -1351,9 +1351,6 @@ static void uart_close(struct tty_struct *tty, struct file *filp)
>
>   	mutex_lock(&port->mutex);
>   	uart_shutdown(tty, state);
> -
> -	tty_ldisc_flush(tty);
> -

This isn't exactly correct. There is a small window of time between the
ldisc flush being performed in tty_port_close_start() and the
subsequent stop_rx() in uart_close(). This might allow for data to be
received, and a racing tty reopen to receive that stale data. Unlikely, but
possible.

>   	tty_port_tty_set(port, NULL);
>   	tty->closing = 0;
>   	spin_lock_irqsave(&port->lock, flags);

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/