Thanks, applied.
On Tue, Jan 19, 2016 at 02:27:19PM -0800, Jarno Rajahalme wrote:
> With minor questions below,
>
> Acked-by: Jarno Rajahalme <[email protected]>
...
> > if (property) {
> > ofpbuf_use_const(property, msg->data, len);
> > - msg->header = msg->data;
> > - msg->msg = ((uint8_t *) msg->data
> > - + (type < min_exp
> > - ? sizeof(struct ofp_prop_header)
> > - : sizeof(struct ofp_prop_experimenter)));
> > + property->header = property->data;
> > + property->msg = ((uint8_t *) property->data
> > + + (type < min_exp
> > + ? sizeof(struct ofp_prop_header)
> > + : sizeof(struct ofp_prop_experimenter)));
>
> Is this a bug fix to a previous patch or just a convention change?
Ouch, that is a bug fix.
I squashed that into the previous patch.
> > @@ -4492,7 +4489,7 @@ pull_table_feature_property(struct ofpbuf *msg,
> > struct ofpbuf *payload,
> >
> > error = ofpprop_pull(msg, payload, typep);
> > if (payload && !error) {
> > - ofpbuf_pull(payload, (uint8_t *)msg->msg - (uint8_t *)msg->header);
> > + ofpbuf_pull(payload, (char *)payload->msg - (char *)payload->data);
>
> Why not ‘payload->header’?
I guess they're the same, so that it doesn't really matter, but I
changed it to 'payload->header' now.
_______________________________________________
dev mailing list
[email protected]
http://openvswitch.org/mailman/listinfo/dev
