ibmveth: Switch order of ibmveth_helper calls.

[ Upstream commit 5ce9ad8 ]

ibmveth_rx_csum_helper() must be called after ibmveth_rx_mss_helper()
as ibmveth_rx_csum_helper() may alter ip and tcp checksum values.

Fixes: 66aa067 ("ibmveth: Support to enable LSO/CSO for Trunk VEA.")
Signed-off-by: David Wilder <dwilder@us.ibm.com>
Reviewed-by: Thomas Falcon <tlfalcon@linux.ibm.com>
Reviewed-by: Cristobal Forno <cris.forno@ibm.com>
Reviewed-by: Pradeep Satyanarayana <pradeeps@linux.vnet.ibm.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/net/ethernet/ibm/ibmveth.c

@@ -1385,16 +1385,16 @@ static int ibmveth_poll(struct napi_struct *napi, int budget)
 			skb_put(skb, length);
 			skb->protocol = eth_type_trans(skb, netdev);
 
-			if (csum_good) {
-				skb->ip_summed = CHECKSUM_UNNECESSARY;
-				ibmveth_rx_csum_helper(skb, adapter);
-			}
-
 			if (length > netdev->mtu + ETH_HLEN) {
 				ibmveth_rx_mss_helper(skb, mss, lrg_pkt);
 				adapter->rx_large_packets++;
 			}
 
+			if (csum_good) {
+				skb->ip_summed = CHECKSUM_UNNECESSARY;
+				ibmveth_rx_csum_helper(skb, adapter);
+			}
+
 			napi_gro_receive(napi, skb);	/* send it up */
 
 			netdev->stats.rx_packets++;