mlx4: add a new xmit_more counter

ethtool -S reports a new counter, tracking number of time doorbell
was not triggered, because skb->xmit_more was set.

$ ethtool -S eth0 | egrep "tx_packet|xmit_more"
     tx_packets: 2413288400
     xmit_more: 666121277

I merged the tso_packet false sharing avoidance in this patch as well.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

drivers/net/ethernet/mellanox/mlx4/en_ethtool.c

@@ -112,6 +112,7 @@ static const char main_strings[][ETH_GSTRING_LEN] = {
 
 	/* port statistics */
 	"tso_packets",
+	"xmit_more",
 	"queue_stopped", "wake_queue", "tx_timeout", "rx_alloc_failed",
 	"rx_csum_good", "rx_csum_none", "tx_chksum_offload",
 

drivers/net/ethernet/mellanox/mlx4/en_port.c

@@ -150,14 +150,19 @@ int mlx4_en_DUMP_ETH_STATS(struct mlx4_en_dev *mdev, u8 port, u8 reset)
 	priv->port_stats.tx_chksum_offload = 0;
 	priv->port_stats.queue_stopped = 0;
 	priv->port_stats.wake_queue = 0;
+	priv->port_stats.tso_packets = 0;
+	priv->port_stats.xmit_more = 0;
 
 	for (i = 0; i < priv->tx_ring_num; i++) {
-		stats->tx_packets += priv->tx_ring[i]->packets;
-		stats->tx_bytes += priv->tx_ring[i]->bytes;
-		priv->port_stats.tx_chksum_offload += priv->tx_ring[i]->tx_csum;
-		priv->port_stats.queue_stopped +=
-			priv->tx_ring[i]->queue_stopped;
-		priv->port_stats.wake_queue += priv->tx_ring[i]->wake_queue;
+		const struct mlx4_en_tx_ring *ring = priv->tx_ring[i];
+
+		stats->tx_packets += ring->packets;
+		stats->tx_bytes += ring->bytes;
+		priv->port_stats.tx_chksum_offload += ring->tx_csum;
+		priv->port_stats.queue_stopped     += ring->queue_stopped;
+		priv->port_stats.wake_queue        += ring->wake_queue;
+		priv->port_stats.tso_packets       += ring->tso_packets;
+		priv->port_stats.xmit_more         += ring->xmit_more;
 	}
 
 	stats->rx_errors = be64_to_cpu(mlx4_en_stats->PCS) +

drivers/net/ethernet/mellanox/mlx4/en_tx.c

@@ -840,7 +840,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
 		 * note that we already verified that it is linear */
 		memcpy(tx_desc->lso.header, skb->data, lso_header_size);
 
-		priv->port_stats.tso_packets++;
+		ring->tso_packets++;
 		i = ((skb->len - lso_header_size) / skb_shinfo(skb)->gso_size) +
 			!!((skb->len - lso_header_size) % skb_shinfo(skb)->gso_size);
 		tx_info->nr_bytes = skb->len + (i - 1) * lso_header_size;
@@ -910,6 +910,8 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
 			wmb();
 			iowrite32be(ring->doorbell_qpn,
 				    ring->bf.uar->map + MLX4_SEND_DOORBELL);
+		} else {
+			ring->xmit_more++;
 		}
 	}
 

drivers/net/ethernet/mellanox/mlx4/mlx4_en.h

@@ -279,6 +279,8 @@ struct mlx4_en_tx_ring {
 	unsigned long tx_csum;
 	unsigned long queue_stopped;
 	unsigned long wake_queue;
+	unsigned long tso_packets;
+	unsigned long xmit_more;
 	struct mlx4_bf bf;
 	bool bf_enabled;
 	bool bf_alloced;
@@ -426,6 +428,7 @@ struct mlx4_en_pkt_stats {
 
 struct mlx4_en_port_stats {
 	unsigned long tso_packets;
+	unsigned long xmit_more;
 	unsigned long queue_stopped;
 	unsigned long wake_queue;
 	unsigned long tx_timeout;
@@ -433,7 +436,7 @@ struct mlx4_en_port_stats {
 	unsigned long rx_chksum_good;
 	unsigned long rx_chksum_none;
 	unsigned long tx_chksum_offload;
-#define NUM_PORT_STATS		8
+#define NUM_PORT_STATS		9
 };
 
 struct mlx4_en_perf_stats {