IB/rxe: optimize mcast recv process

In mcast recv process, the function skb_clone is used. In fact,
the refcount can be increased to replace cloning a new skb since
the original skb will not be modified before it is freed.

This can make the performance better and save the memory.

CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

drivers/infiniband/sw/rxe/rxe_recv.c

@@ -276,7 +276,6 @@ static void rxe_rcv_mcast_pkt(struct rxe_dev *rxe, struct sk_buff *skb)
 {
 	struct rxe_pkt_info *pkt = SKB_TO_PKT(skb);
 	struct rxe_mc_grp *mcg;
-	struct sk_buff *skb_copy;
 	struct rxe_mc_elem *mce;
 	struct rxe_qp *qp;
 	union ib_gid dgid;
@@ -309,18 +308,14 @@ static void rxe_rcv_mcast_pkt(struct rxe_dev *rxe, struct sk_buff *skb)
 			continue;
 
 		/* if *not* the last qp in the list
-		 * make a copy of the skb to post to the next qp
+		 * increase the users of the skb then post to the next qp
 		 */
-		skb_copy = (mce->qp_list.next != &mcg->qp_list) ?
-				skb_clone(skb, GFP_ATOMIC) : NULL;
+		if (mce->qp_list.next != &mcg->qp_list)
+			refcount_inc(&skb->users);
 
 		pkt->qp = qp;
 		rxe_add_ref(qp);
 		rxe_rcv_pkt(rxe, pkt, skb);
-
-		skb = skb_copy;
-		if (!skb)
-			break;
 	}
 
 	spin_unlock_bh(&mcg->mcg_lock);
@@ -328,8 +323,7 @@ static void rxe_rcv_mcast_pkt(struct rxe_dev *rxe, struct sk_buff *skb)
 	rxe_drop_ref(mcg);	/* drop ref from rxe_pool_get_key. */
 
 err1:
-	if (skb)
-		kfree_skb(skb);
+	kfree_skb(skb);
 }
 
 static int rxe_match_dgid(struct rxe_dev *rxe, struct sk_buff *skb)