net/mlx5e: IPoIB, handle RX packet correctly

IPoIB packet contains the pseudo header area, we need to pull it prior
to reset_mac_header in order to let the GRO work well.

In more details:
GRO checks the mac address of the new coming packet, it does that by
comparing the hard_header_len size of the current packet to the previous
one in that session, the comparison is over hard_header_len size.
Now, the driver prepares that area in the skb by allocating area from the
reserved part and resetting the correct mac header to it.

Fixes: 9d6bd75 ("net/mlx5e: IPoIB, RX handler")
Signed-off-by: Erez Shitrit <erezsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

drivers/net/ethernet/mellanox/mlx5/core/en_rx.c

@@ -1041,13 +1041,16 @@ void mlx5e_free_xdpsq_descs(struct mlx5e_xdpsq *sq)
 #define MLX5_IB_GRH_BYTES       40
 #define MLX5_IPOIB_ENCAP_LEN    4
 #define MLX5_GID_SIZE           16
+#define MLX5_IPOIB_PSEUDO_LEN   20
+#define MLX5_IPOIB_HARD_LEN     (MLX5_IPOIB_PSEUDO_LEN + MLX5_IPOIB_ENCAP_LEN)
 
 static inline void mlx5i_complete_rx_cqe(struct mlx5e_rq *rq,
 					 struct mlx5_cqe64 *cqe,
 					 u32 cqe_bcnt,
 					 struct sk_buff *skb)
 {
 	struct net_device *netdev = rq->netdev;
+	char *pseudo_header;
 	u8 *dgid;
 	u8 g;
 
@@ -1076,8 +1079,11 @@ static inline void mlx5i_complete_rx_cqe(struct mlx5e_rq *rq,
 	if (likely(netdev->features & NETIF_F_RXHASH))
 		mlx5e_skb_set_hash(cqe, skb);
 
+	/* 20 bytes of ipoib header and 4 for encap existing */
+	pseudo_header = skb_push(skb, MLX5_IPOIB_PSEUDO_LEN);
+	memset(pseudo_header, 0, MLX5_IPOIB_PSEUDO_LEN);
 	skb_reset_mac_header(skb);
-	skb_pull(skb, MLX5_IPOIB_ENCAP_LEN);
+	skb_pull(skb, MLX5_IPOIB_HARD_LEN);
 
 	skb->dev = netdev;