Using serial devices under linux

大林 | 25 Jul 2012 03:47

n_tty_read panic

Hello.

Here's a kernel panic caused by n_tty_read, linux kernel version is 2.6.38.8.

There are many discussions on the subject, this bug solved yet? Where can I download the patch?

thanks.

Jul 24 01:30:25 AnShion <8> klogd: [432687.542510] ------------[ cut here ]------------
Jul 24 01:30:25 AnShion <10> klogd: [432687.598836] kernel BUG at drivers/tty/n_tty.c:1725!
Jul 24 01:30:25 AnShion <8> klogd: [432687.658218] invalid opcode: 0000 [#384] SMP 
Jul 24 01:30:25 AnShion <8> klogd: [432687.704085] last sysfs file: /sys/devices/pci0000:00/0000:00:1e.0/0000:05:05.0/resource
Jul 24 01:30:25 AnShion <12> klogd: [432687.704087] Modules linked in: igb e1000e
Jul 24 01:30:25 AnShion <12> klogd: [432687.704093] 
Jul 24 01:30:25 AnShion <12> klogd: [432687.704095] Pid: 6856, comm: login Tainted: G      D     2.6.38.8 #412 To be
filled by O.E.M. To be filled by O.E.M./P8B-X series
Jul 24 01:30:25 AnShion <12> klogd: [432687.704098] EIP: 0060:[<c041b415>] EFLAGS: 00010246 CPU: 5
Jul 24 01:30:25 AnShion <12> klogd: [432687.704101] EIP is at n_tty_read+0x735/0x750
Jul 24 01:30:25 AnShion <12> klogd: [432687.704103] EAX: 00000000 EBX: eda6609c ECX: eda66128 EDX: 00000000
Jul 24 01:30:25 AnShion <12> klogd: [432687.704104] ESI: 00000000 EDI: eda66000 EBP: e4f59f3c ESP: e4f59ed0
Jul 24 01:30:25 AnShion <12> klogd: [432687.704105]  DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068
Jul 24 01:30:25 AnShion <8> klogd: [432687.704106] Process login (pid: 6856, ti=e4f58000 task=dda1b200 task.ti=e4f58000)
Jul 24 01:30:25 AnShion <8> klogd: [432687.704107] Stack:
Jul 24 01:30:25 AnShion <12> klogd: [432687.704108]  00000001 dda1b200 dda1b200 b7571000 ee6e3a80
eda66124 eda663a8 eda6609c
Jul 24 01:30:25 AnShion <12> klogd: [432687.704110]  00000000 00000000 fffffffb 7fffffff 00000000
e4f59f14 eda66128 eda66190
Jul 24 01:30:25 AnShion <12> klogd: [432687.704112]  eda66400 00000000 dda1b200 c01419c0 00100100
00200200 00000000 b7571000
Jul 24 01:30:25 AnShion <8> klogd: [432687.704114] Call Trace:

(Continue reading)

headers

Alan Cox | 25 Jul 2012 13:44

Re: n_tty_read panic

On Wed, 25 Jul 2012 09:47:09 +0800
"大林" <yourdarling1999 <at> sina.com> wrote:

> Hello.
> 
> Here's a kernel panic caused by n_tty_read, linux kernel version is 2.6.38.8.
> 
> 
> There are many discussions on the subject, this bug solved yet? Where can I download the patch?

Thats a different trace to the others I've seen. Several races have been
fixed, there are certainly a couple of others left to sort eventually. I
don't believe anyone is currently working on them (and indeed until the
rest of the locking and lock splitting work is done it's difficult to see
how any useful progress could be made here). Patches are however welcome.

Alan
--
To unsubscribe from this list: send the line "unsubscribe linux-serial" in
the body of a message to majordomo <at> vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

headers

Stanislav Kozina | 30 Jul 2012 12:41

Re: n_tty_read panic

Seems that tty->read_buf == NULL at the very beginning of n_tty_read():

1723 do_it_again:
1724
1725         BUG_ON(!tty->read_buf);

tty->read_buf access needs tty->read_lock to be held, correct? So 
perhaps the patch would be:

diff --git a/drivers/tty/n_tty.c b/drivers/tty/n_tty.c
index ee1c268..9decb1c 100644
--- a/drivers/tty/n_tty.c
+++ b/drivers/tty/n_tty.c
 <at>  <at>  -1728,7 +1728,11  <at>  <at>  static ssize_t n_tty_read(struct tty_struct *tty, 
struct file *file,

  do_it_again:

+#ifdef CONFIG_BUG
+       spin_lock_irqsave(&tty->read_lock, flags);
         BUG_ON(!tty->read_buf);
+       spin_unlock_irqrestore(&tty->read_lock, flags);
+#endif

         c = job_control(tty, file);
         if (c < 0)

Regards,
-Stanislav Kozina

(Continue reading)