On Mon, Oct 01, 2012 at 11:48:58PM +0200, Jiri Slaby wrote:

On 10/01/2012 08:30 PM, Greg KH wrote:

quoted

Stanislav Kozina (2):
      tty: Fix possible race in n_tty_read()

The (pretty hopeless) commit log says:
    Fix possible panic caused by unlocked access to tty->read_cnt in
    while-loop condition in n_tty_read().

Just curious, what kind of panic that can cause, can you be more
concrete on that? I suppose two readers, one eats everything, the other
is a killer? And should we backport to -stable?

Stanislav, you are the one who could reproduce this, any answers here?

thanks,

greg k-h