Hi Lizardo,
* Anderson Lizardo <anderson.lizardo@xxxxxxxxxxxxx> [2011-04-04 16:50:13 -0400]:
> Hi Gustavo,
>
> (patch too big for me to review properly...)
>
> On Fri, Apr 1, 2011 at 6:33 PM, Gustavo F. Padovan
> <padovan@xxxxxxxxxxxxxx> wrote:
> > static void l2cap_retrans_timeout(unsigned long arg)
> > {
> > - struct sock *sk = (void *) arg;
> > + struct l2cap_chan *chan = (void *) arg;
> > + struct sock *sk = chan->sk;
> >
> > BT_DBG("sk %p", sk);
> >
> > @@ -978,9 +983,9 @@ static void l2cap_retrans_timeout(unsigned long arg)
> > l2cap_pi(sk)->retry_count = 1;
> > __mod_monitor_timer();
> >
> > - l2cap_pi(sk)->conn_state |= L2CAP_CONN_WAIT_F;
> > + chan->conn_state |= L2CAP_CONN_WAIT_F;
>
> I suppose the above requires a lock ?
It has one, the socket lock. It's completely safe use it here.
>
> In general it seems that by creating the additional "l2cap_chan"
> layer, all lock usages need to be reviewed (IMHO).
Yes, that will be done in an next step. I'm going to remove any sock reference
from l2cap_core.c, so a new locking system will be needed.
--
Gustavo F. Padovan
http://profusion.mobi
--
To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
