On Fri, 11 Jul 2014 11:07:15 +0200
Markus Armbruster <arm...@redhat.com> wrote:
> Cc'ing folks involved in the flawed commit.
>
> Liliang <liangx.z...@intel.com> writes:
>
> > From: Li Liang <liangx.z...@intel.com>
> >
> > This bug was introduced in the commit
> > 9005b2a7589540a3733b3abdcfbccfe7746cd1a1,
> > it will cause deadlock when create a vm with the parameter "-monitor pty"
> > and
> > then try to read from /dev/pts/x.
> >
> > Signed-off-by: Li Liang <liangx.z...@intel.com>
>
> Long lines in commit message, and the subject isn't as informative as it
> could be. Suggest something like:
>
> qemu-char: Fix deadlock in pty character device
>
> To reproduce, run with "-monitor pty", then try to read from the
> slave /dev/pts/FOO created for it.
>
> Broken in commit 9005b2a.
Paolo posted a pull request today with a fix for this issue included:
[PULL 2/5] qemu-char: fix deadlock with "-monitor pty"
Thanks Paolo!
>
> > ---
> > qemu-char.c | 4 +++-
> > 1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/qemu-char.c b/qemu-char.c
> > index 55e372c..55cdded 100644
> > --- a/qemu-char.c
> > +++ b/qemu-char.c
> > @@ -1103,9 +1103,11 @@ static gboolean pty_chr_timer(gpointer opaque)
> > s->timer_tag = 0;
> > if (!s->connected) {
> > /* Next poll ... */
> > + qemu_mutex_unlock(&chr->chr_write_lock);
> > pty_chr_update_read_handler_locked(chr);
> > + } else {
> > + qemu_mutex_unlock(&chr->chr_write_lock);
> > }
> > - qemu_mutex_unlock(&chr->chr_write_lock);
> > return FALSE;
> > }
>
