net: Enable Tx queue selection based on Rx queues

This patch adds support to pick Tx queue based on the Rx queue(s) map
configuration set by the admin through the sysfs attribute
for each Tx queue. If the user configuration for receive queue(s) map
does not apply, then the Tx queue selection falls back to CPU(s) map
based selection and finally to hashing.

Signed-off-by: Amritha Nambiar <amritha.nambiar@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

include/net/sock.h

@@ -1730,6 +1730,16 @@ static inline void sk_rx_queue_clear(struct sock *sk)
 #endif
 }
 
+#ifdef CONFIG_XPS
+static inline int sk_rx_queue_get(const struct sock *sk)
+{
+	if (sk && sk->sk_rx_queue_mapping != NO_QUEUE_MAPPING)
+		return sk->sk_rx_queue_mapping;
+
+	return -1;
+}
+#endif
+
 static inline void sk_set_socket(struct sock *sk, struct socket *sock)
 {
 	sk_tx_queue_clear(sk);

net/core/dev.c

@@ -3459,35 +3459,63 @@ sch_handle_egress(struct sk_buff *skb, int *ret, struct net_device *dev)
 }
 #endif /* CONFIG_NET_EGRESS */
 
-static inline int get_xps_queue(struct net_device *dev, struct sk_buff *skb)
+#ifdef CONFIG_XPS
+static int __get_xps_queue_idx(struct net_device *dev, struct sk_buff *skb,
+			       struct xps_dev_maps *dev_maps, unsigned int tci)
+{
+	struct xps_map *map;
+	int queue_index = -1;
+
+	if (dev->num_tc) {
+		tci *= dev->num_tc;
+		tci += netdev_get_prio_tc_map(dev, skb->priority);
+	}
+
+	map = rcu_dereference(dev_maps->attr_map[tci]);
+	if (map) {
+		if (map->len == 1)
+			queue_index = map->queues[0];
+		else
+			queue_index = map->queues[reciprocal_scale(
+						skb_get_hash(skb), map->len)];
+		if (unlikely(queue_index >= dev->real_num_tx_queues))
+			queue_index = -1;
+	}
+	return queue_index;
+}
+#endif
+
+static int get_xps_queue(struct net_device *dev, struct sk_buff *skb)
 {
 #ifdef CONFIG_XPS
 	struct xps_dev_maps *dev_maps;
-	struct xps_map *map;
+	struct sock *sk = skb->sk;
 	int queue_index = -1;
 
 	if (!static_key_false(&xps_needed))
 		return -1;
 
 	rcu_read_lock();
-	dev_maps = rcu_dereference(dev->xps_cpus_map);
+	if (!static_key_false(&xps_rxqs_needed))
+		goto get_cpus_map;
+
+	dev_maps = rcu_dereference(dev->xps_rxqs_map);
 	if (dev_maps) {
-		unsigned int tci = skb->sender_cpu - 1;
+		int tci = sk_rx_queue_get(sk);
 
-		if (dev->num_tc) {
-			tci *= dev->num_tc;
-			tci += netdev_get_prio_tc_map(dev, skb->priority);
-		}
+		if (tci >= 0 && tci < dev->num_rx_queues)
+			queue_index = __get_xps_queue_idx(dev, skb, dev_maps,
+							  tci);
+	}
 
-		map = rcu_dereference(dev_maps->attr_map[tci]);
-		if (map) {
-			if (map->len == 1)
-				queue_index = map->queues[0];
-			else
-				queue_index = map->queues[reciprocal_scale(skb_get_hash(skb),
-									   map->len)];
-			if (unlikely(queue_index >= dev->real_num_tx_queues))
-				queue_index = -1;
+get_cpus_map:
+	if (queue_index < 0) {
+		dev_maps = rcu_dereference(dev->xps_cpus_map);
+		if (dev_maps) {
+			unsigned int tci = skb->sender_cpu - 1;
+
+			queue_index = __get_xps_queue_idx(dev, skb, dev_maps,
+							  tci);
 		}
 	}
 	rcu_read_unlock();