ibmvnic: Use ndo_get_stats64 to fix inaccurate SAR reporting

VNIC testing on multi-core Power systems showed SAR stats drift
and packet rate inconsistencies under load.

Implements ndo_get_stats64 to provide safe aggregation of queue-level
atomic64 counters into rtnl_link_stats64 for use by tools like 'ip -s',
'ifconfig', and 'sar'. Switch to ndo_get_stats64 to align SAR reporting
with the standard kernel interface for retrieving netdev stats.

This removes redundant per-adapter stat updates, reduces overhead,
eliminates cacheline bouncing from hot path updates, and improves
the accuracy of reported packet rates.

Signed-off-by: Mingming Cao <[email protected]>
Reviewed-by: Brian King <[email protected]>
Reviewed-by: Dave Marquardt <[email protected]>
Reviewed-by: Simon Horman <[email protected]>

----
Changes since v3:
link to v3: https://www.spinics.net/lists/netdev/msg1107999.html
-- keep per queue counters as u64 (this patch) and drop off patch 1 in v3

Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

Author: Mingming Cao

drivers/net/ethernet/ibm/ibmvnic.c

@@ -2312,8 +2312,6 @@ static void ibmvnic_tx_scrq_clean_buffer(struct ibmvnic_adapter *adapter,
 					  tx_pool->num_buffers - 1 :
 					  tx_pool->consumer_index - 1;
 		tx_buff = &tx_pool->tx_buff[index];
-		adapter->netdev->stats.tx_packets--;
-		adapter->netdev->stats.tx_bytes -= tx_buff->skb->len;
 		adapter->tx_stats_buffers[queue_num].batched_packets--;
 		adapter->tx_stats_buffers[queue_num].bytes -=
 						tx_buff->skb->len;
@@ -2647,9 +2645,6 @@ static netdev_tx_t ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev)
 	}
 out:
 	rcu_read_unlock();
-	netdev->stats.tx_dropped += tx_dropped;
-	netdev->stats.tx_bytes += tx_bytes;
-	netdev->stats.tx_packets += tx_bpackets + tx_dpackets;
 	adapter->tx_send_failed += tx_send_failed;
 	adapter->tx_map_failed += tx_map_failed;
 	adapter->tx_stats_buffers[queue_num].batched_packets += tx_bpackets;
@@ -3452,6 +3447,25 @@ static int ibmvnic_reset(struct ibmvnic_adapter *adapter,
 	return -ret;
 }
 
+static void ibmvnic_get_stats64(struct net_device *netdev,
+				struct rtnl_link_stats64 *stats)
+{
+	struct ibmvnic_adapter *adapter = netdev_priv(netdev);
+	int i;
+
+	for (i = 0; i < adapter->req_rx_queues; i++) {
+		stats->rx_packets += adapter->rx_stats_buffers[i].packets;
+		stats->rx_bytes   += adapter->rx_stats_buffers[i].bytes;
+	}
+
+	for (i = 0; i < adapter->req_tx_queues; i++) {
+		stats->tx_packets += adapter->tx_stats_buffers[i].batched_packets;
+		stats->tx_packets += adapter->tx_stats_buffers[i].direct_packets;
+		stats->tx_bytes   += adapter->tx_stats_buffers[i].bytes;
+		stats->tx_dropped += adapter->tx_stats_buffers[i].dropped_packets;
+	}
+}
+
 static void ibmvnic_tx_timeout(struct net_device *dev, unsigned int txqueue)
 {
 	struct ibmvnic_adapter *adapter = netdev_priv(dev);
@@ -3567,8 +3581,6 @@ static int ibmvnic_poll(struct napi_struct *napi, int budget)
 
 		length = skb->len;
 		napi_gro_receive(napi, skb); /* send it up */
-		netdev->stats.rx_packets++;
-		netdev->stats.rx_bytes += length;
 		adapter->rx_stats_buffers[scrq_num].packets++;
 		adapter->rx_stats_buffers[scrq_num].bytes += length;
 		frames_processed++;
@@ -3678,6 +3690,7 @@ static const struct net_device_ops ibmvnic_netdev_ops = {
 	.ndo_set_rx_mode	= ibmvnic_set_multi,
 	.ndo_set_mac_address	= ibmvnic_set_mac,
 	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_get_stats64	= ibmvnic_get_stats64,
 	.ndo_tx_timeout		= ibmvnic_tx_timeout,
 	.ndo_change_mtu		= ibmvnic_change_mtu,
 	.ndo_features_check     = ibmvnic_features_check,