Yes, it is protected. I missed that.
On Tue, Oct 8, 2013 at 3:49 PM, Ben Pfaff <b...@nicira.com> wrote: > I think that it is protected by the lock. The line above the first > context line takes the lock, no? > > On Tue, Oct 08, 2013 at 02:42:54PM -0700, Alex Wang wrote: > > Looks good to me, just one point, > > > > I think the following code should be protected by the lock. > > > > @@ -503,6 +533,14 @@ netdev_dummy_send(struct netdev *netdev, const void > > > *buffer, size_t size) > > > dev->stats.tx_packets++; > > > dev->stats.tx_bytes += size; > > > > > > + if (dev->tx_pcap) { > > > + struct ofpbuf packet; > > > + > > > + ofpbuf_use_const(&packet, buffer, size); > > > + pcap_write(dev->tx_pcap, &packet); > > > + fflush(dev->tx_pcap); > > > + } > > > + > > > for (i = 0; i < dev->n_streams; i++) { > > > struct dummy_stream *s = &dev->streams[i] > > > >
_______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev