i40e/i40evf: carefully fill tx ring

We need to make sure that we stay away from the cache line
where the DD bit (done) may be getting written back for
the transmit ring since the hardware may be writing the
whole cache line for a partial update.

Change-ID: Id0b6dfc01f654def6a2a021af185803be1915d7e
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

drivers/net/ethernet/intel/i40e/i40e_txrx.c

@@ -2102,7 +2102,7 @@ static int i40e_xmit_descriptor_count(struct sk_buff *skb,
 
 	/* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD,
 	 *       + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD,
-	 *       + 2 desc gap to keep tail from touching head,
+	 *       + 4 desc gap to avoid the cache line where head is,
 	 *       + 1 desc for context descriptor,
 	 * otherwise try next time
 	 */
@@ -2113,7 +2113,7 @@ static int i40e_xmit_descriptor_count(struct sk_buff *skb,
 	count += skb_shinfo(skb)->nr_frags;
 #endif
 	count += TXD_USE_COUNT(skb_headlen(skb));
-	if (i40e_maybe_stop_tx(tx_ring, count + 3)) {
+	if (i40e_maybe_stop_tx(tx_ring, count + 4 + 1)) {
 		tx_ring->tx_stats.tx_busy++;
 		return 0;
 	}

drivers/net/ethernet/intel/i40evf/i40e_txrx.c

@@ -1482,7 +1482,7 @@ static int i40e_xmit_descriptor_count(struct sk_buff *skb,
 
 	/* need: 1 descriptor per page * PAGE_SIZE/I40E_MAX_DATA_PER_TXD,
 	 *       + 1 desc for skb_head_len/I40E_MAX_DATA_PER_TXD,
-	 *       + 2 desc gap to keep tail from touching head,
+	 *       + 4 desc gap to avoid the cache line where head is,
 	 *       + 1 desc for context descriptor,
 	 * otherwise try next time
 	 */
@@ -1493,7 +1493,7 @@ static int i40e_xmit_descriptor_count(struct sk_buff *skb,
 	count += skb_shinfo(skb)->nr_frags;
 #endif
 	count += TXD_USE_COUNT(skb_headlen(skb));
-	if (i40e_maybe_stop_tx(tx_ring, count + 3)) {
+	if (i40e_maybe_stop_tx(tx_ring, count + 4 + 1)) {
 		tx_ring->tx_stats.tx_busy++;
 		return 0;
 	}