Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1332336 > unrolled thread

[PATCH v1] tty: serial: 8250: Fix possible race in serial8250_em485_destroy()

Back to article view | Back to linux.kernel

Contents

  [PATCH v1] tty: serial: 8250: Fix possible race in serial8250_em485_destroy() "Matwey V. Kornilov" <matwey@sai.msu.ru> - 2016-02-11 20:40 +0100
    Re: [PATCH v1] tty: serial: 8250: Fix possible race in serial8250_em485_destroy() "Matwey V. Kornilov" <matwey@sai.msu.ru> - 2016-02-11 21:10 +0100

#1332336 — [PATCH v1] tty: serial: 8250: Fix possible race in serial8250_em485_destroy()

Fix possbile race in serial8250_em485_destroy() when timer handlers can
dereference p->em485 which is alread destroyed but not yet NULLed.

Signed-off-by: Matwey V. Kornilov <matwey@sai.msu.ru>
---
I've found that Greg applied initial patchset, so this erratum goes as separate patch.

 drivers/tty/serial/8250/8250_port.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/tty/serial/8250/8250_port.c b/drivers/tty/serial/8250/8250_port.c
index c908b77..d962de2 100644
--- a/drivers/tty/serial/8250/8250_port.c
+++ b/drivers/tty/serial/8250/8250_port.c
@@ -617,14 +617,16 @@ EXPORT_SYMBOL_GPL(serial8250_em485_init);
  */
 void serial8250_em485_destroy(struct uart_8250_port *p)
 {
-	if (p->em485 == NULL)
+	struct uart_8250_em485 *em485 = p->em485;
+
+	if (!em485)
 		return;
 
-	del_timer(&p->em485->start_tx_timer);
-	del_timer(&p->em485->stop_tx_timer);
+	del_timer(&em485->start_tx_timer);
+	del_timer(&em485->stop_tx_timer);
 
-	kfree(p->em485);
 	p->em485 = NULL;
+	kfree(em485);
 }
 EXPORT_SYMBOL_GPL(serial8250_em485_destroy);
 
-- 
2.7.0

[toc] | [next] | [standalone]

#1332345

I am sorry, please ignore it. There is no issue actually. The timer
handlers and rs485_config callbacks are protected by the same
spinlock, so they are never run in parallel.

2016-02-11 22:32 GMT+03:00 Matwey V. Kornilov <matwey@sai.msu.ru>:
> Fix possbile race in serial8250_em485_destroy() when timer handlers can
> dereference p->em485 which is alread destroyed but not yet NULLed.
>
> Signed-off-by: Matwey V. Kornilov <matwey@sai.msu.ru>
> ---
> I've found that Greg applied initial patchset, so this erratum goes as separate patch.
>
>  drivers/tty/serial/8250/8250_port.c | 10 ++++++----
>  1 file changed, 6 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/tty/serial/8250/8250_port.c b/drivers/tty/serial/8250/8250_port.c
> index c908b77..d962de2 100644
> --- a/drivers/tty/serial/8250/8250_port.c
> +++ b/drivers/tty/serial/8250/8250_port.c
> @@ -617,14 +617,16 @@ EXPORT_SYMBOL_GPL(serial8250_em485_init);
>   */
>  void serial8250_em485_destroy(struct uart_8250_port *p)
>  {
> -       if (p->em485 == NULL)
> +       struct uart_8250_em485 *em485 = p->em485;
> +
> +       if (!em485)
>                 return;
>
> -       del_timer(&p->em485->start_tx_timer);
> -       del_timer(&p->em485->stop_tx_timer);
> +       del_timer(&em485->start_tx_timer);
> +       del_timer(&em485->stop_tx_timer);
>
> -       kfree(p->em485);
>         p->em485 = NULL;
> +       kfree(em485);
>  }
>  EXPORT_SYMBOL_GPL(serial8250_em485_destroy);
>
> --
> 2.7.0
>



-- 
With best regards,
Matwey V. Kornilov.
Sternberg Astronomical Institute, Lomonosov Moscow State University, Russia
119991, Moscow, Universitetsky pr-k 13, +7 (495) 9392382

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web