> > Were you going to try it this way? > > /* First byte of dgid signals multicast/broadcast when 0xff */ > if ((wc->wc_flags & IB_WC_GRH) && > ((struct ib_grh *)skb->data)->dgid.raw[0] == 0xff) { > if (memcmp(((struct ib_grh *)skb->data)->dgid.raw, > dev->broadcast + 4, sizeof(union ib_gid)) == 0) > skb->pkt_type = PACKET_BROADCAST; > else > skb->pkt_type = PACKET_MULTICAST; > } > else > skb->pkt_type = PACKET_HOST; > > I think doing the memcmp only in the multicast path should be > reasonable overhead wise. >
Shouldn't struct ib_grh be packed to make this really work? The code looks a little messy to me anyway... How about using a local var which is a ptr to packed struct ib_grh? The compiler will probably eliminate it anyway. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html