Messages in this thread Patch in this message | | | Date | Sat, 25 Jul 2009 13:07:03 +0100 | From | Alan Cox <> | Subject | Re: [Regression] kdesu broken |
| |
> > See the thread starting here: ("possible regression with pty.c commit") > > http://lkml.org/lkml/2009/7/11/125 > > I am also facing a similar problem. > > http://bugzilla.kernel.org/show_bug.cgi?id=13815
Probably something like this fixes it. I'll be working on that Monday/Tuesday along with various other bugs that need a review.
--
pty: ensure writes hit the reader before close
From: Alan Cox <alan@linux.intel.com>
Logically we move the buffering from one side to the other ---
drivers/char/pty.c | 11 +++++++++++ 1 files changed, 11 insertions(+), 0 deletions(-)
diff --git a/drivers/char/pty.c b/drivers/char/pty.c index 6e6942c..7555890 100644 --- a/drivers/char/pty.c +++ b/drivers/char/pty.c @@ -36,6 +36,15 @@ static struct tty_driver *ptm_driver; static struct tty_driver *pts_driver; #endif +static int pty_empty(struct tty_struct *tty) +{ + if (tty->buf.memory_used == 0) + return 1; + if (test_bit(TTY_HUPPED, &tty->flags)) + return 1; + return 0; +} + static void pty_close(struct tty_struct *tty, struct file *filp) { BUG_ON(!tty); @@ -47,9 +56,11 @@ static void pty_close(struct tty_struct *tty, struct file *filp) } wake_up_interruptible(&tty->read_wait); wake_up_interruptible(&tty->write_wait); + tty->packet = 0; if (!tty->link) return; + wait_event_interruptible(tty->write_wait, pty_empty(tty->link)); tty->link->packet = 0; set_bit(TTY_OTHER_CLOSED, &tty->link->flags); wake_up_interruptible(&tty->link->read_wait);
| |