linux-kernel - PROBLEM: ppp over serial gets stuck in drivers/char/tty

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Date:	Thu, 12 Mar 2009 11:07:35 +0100 (CET)
From:	yann.poupet@...e.fr
To:	linux-kernel@...r.kernel.org
Subject: PROBLEM: ppp over serial gets stuck in drivers/char/tty_ioctl.c

Hello,

I'm working on ARM with kernel 2.6.18.2, and maybe I found a problem in
drivers/char/tty_ioctl.c, function tty_wait_until_sent().

void tty_wait_until_sent(struct tty_struct * tty, long timeout)
{
    DECLARE_WAITQUEUE(wait, current);

#ifdef TTY_DEBUG_WAIT_UNTIL_SENT
    char buf[64];

    printk(KERN_DEBUG "%s wait until sent...\n", tty_name(tty, buf));
#endif
    if (!tty->driver->chars_in_buffer)
        return;
    add_wait_queue(&tty->write_wait, &wait);

[snip...]

While trying to get PPP to work over serial line, I found ppp getting stuck in
this function, because if tty buffer is empty when getting here, it might wait
forever IIRC (I found the problem a few month ago and can't remember the
details of my investigations).
This is why I added a test to see whether the buffer is not already empty:

[snip...]
    if (!tty->driver->chars_in_buffer)
        return;
    /* if buffer is empty, return */
    if (!tty->driver->chars_in_buffer(tty))
        return;
    add_wait_queue(&tty->write_wait, &wait);
[snip...]

After this modification, ppp over serial line worked fine.

Is it on purpose one tests whether tty->driver->chars_in_buffer is != 0 ?
I felt like this should be if (!tty->driver->chars_in_buffer(tty)), but in
doubts I left both tests.

I saw tty_wait_until_sent() has been modified between 2.6.25 and 2.6.26,
so probably this problem is gone in newer kernels.
However, if a kernel version up to 2.6.25 is still maintained, or if someone
gets the same problem for kernel up to 2.6.25, this might be usefull.

Since this is only related to older kernel, maybe I shouldn't post to this
ML, but I haven't found where I should send my message.

Regards,

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