Re: [PATCH] tty race on con_close and con_flush_chars

From: Tachino Nobuhiro (tachino@open.nm.fujitsu.co.jp)
Date: Wed Oct 31 2001 - 00:08:44 EST


Hello,

At 29 Oct 2001 22:13:55 -0500,
Robert Love wrote:
>
> Someone pointed out (via private email) that a race can still exist
> between checking that vt is non-zero and acquiring the semaphore,
> especially since we can sleep doing that.
>
> I agree; the following patch should alleviate that race.

Your patch only fixes con_flush_chars(), but the same races exists in
do_con_write() and maybe con_unthrottle(). I think the patch below is
better because the oops would never happen with this patch.

diff -r -u -N linux.org/drivers/char/console.c linux/drivers/char/console.c
--- linux.org/drivers/char/console.c Tue Oct 30 18:55:17 2001
+++ linux/drivers/char/console.c Tue Oct 30 19:07:08 2001
@@ -2424,7 +2424,6 @@
                 return;
         if (tty->count != 1) return;
         vcs_make_devfs (MINOR (tty->device) - tty->driver.minor_start, 1);
- tty->driver_data = 0;
 }
 
 static void vc_init(unsigned int currcons, unsigned int rows, unsigned int cols, int do_clear)

>
> diff -u linux-2.4.13-ac5/drivers/char/console.c
> linux/drivers/char/console.c
> --- linux-2.4.13-ac5/drivers/char/console.c Mon Oct 29 17:27:19 2001
> +++ linux/drivers/char/console.c Mon Oct 29 22:11:27 2001
> @@ -2387,8 +2387,14 @@
> return;
>
> pm_access(pm_con);
> +
> + /*
> + * If we raced with con_close(), `vt' may be null.
> + * Hence this bandaid. - akpm
> + */
> acquire_console_sem();
> - set_cursor(vt->vc_num);
> + if (vt)
> + set_cursor(vt->vc_num);
> release_console_sem();
> }
>
>
> Robert Love
>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Wed Oct 31 2001 - 21:00:43 EST