Merge branch 'udp_tunnel-gro-optimizations'

Paolo Abeni says:

====================
udp_tunnel: GRO optimizations

The UDP tunnel GRO stage is source of measurable overhead for workload
based on UDP-encapsulated traffic: each incoming packets requires a full
UDP socket lookup and an indirect call.

In the most common setups a single UDP tunnel device is used. In such
case we can optimize both the lookup and the indirect call.

Patch 1 tracks per netns the active UDP tunnels and replaces the socket
lookup with a single destination port comparison when possible.

Patch 2 tracks the different types of UDP tunnels and replaces the
indirect call with a static one when there is a single UDP tunnel type
active.

I measure ~10% performance improvement in TCP over UDP tunnel stream
tests on top of this series.

v4: https://lore.kernel.org/[e-mail]
v3: https://lore.kernel.org/[e-mail]
v2: https://lore.kernel.org/[e-mail]
v1: https://lore.kernel.org/[e-mail]
====================

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

include/linux/udp.h

···
 
 	/* Cache friendly copy of sk->sk_peek_off >= 0 */
 	bool		peeking_with_offset;
+
+	/*
+	 * Accounting for the tunnel GRO fastpath.
+	 * Unprotected by compilers guard, as it uses space available in
+	 * the last UDP socket cacheline.
+	 */
+	struct hlist_node	tunnel_list;
 };
 
 #define udp_test_bit(nr, sk)			\
···
 #endif
 
 #define IS_UDPLITE(__sk) (__sk->sk_protocol == IPPROTO_UDPLITE)
+
+static inline struct sock *udp_tunnel_sk(const struct net *net, bool is_ipv6)
+{
+#if IS_ENABLED(CONFIG_NET_UDP_TUNNEL)
+	return rcu_dereference(net->ipv4.udp_tunnel_gro[is_ipv6].sk);
+#else
+	return NULL;
+#endif
+}
 
 #endif	/* _LINUX_UDP_H */

include/net/netns/ipv4.h

···
 };
 #endif
 
+struct udp_tunnel_gro {
+	struct sock __rcu *sk;
+	struct hlist_head list;
+};
+
 struct netns_ipv4 {
 	/* Cacheline organization can be found documented in
 	 * Documentation/networking/net_cachelines/netns_ipv4_sysctl.rst.
···
 
 	struct inet_timewait_death_row tcp_death_row;
 	struct udp_table *udp_table;
+
+#if IS_ENABLED(CONFIG_NET_UDP_TUNNEL)
+	/* Not in a pernet subsys because need to be available at GRO stage */
+	struct udp_tunnel_gro udp_tunnel_gro[2];
+#endif
 
 #ifdef CONFIG_SYSCTL
 	struct ctl_table_header	*forw_hdr;
···
 	struct hlist_head	*inet_addr_lst;
 	struct delayed_work	addr_chk_work;
 };
+
 #endif

include/net/udp.h

···
 	struct udp_sock *up = udp_sk(sk);
 
 	skb_queue_head_init(&up->reader_queue);
+	INIT_HLIST_NODE(&up->tunnel_list);
 	up->forward_threshold = sk->sk_rcvbuf >> 2;
 	set_bit(SOCK_CUSTOM_SOCKOPT, &sk->sk_socket->flags);
 }

include/net/udp_tunnel.h

···
 }
 #endif
 
+#if IS_ENABLED(CONFIG_NET_UDP_TUNNEL)
+void udp_tunnel_update_gro_lookup(struct net *net, struct sock *sk, bool add);
+void udp_tunnel_update_gro_rcv(struct sock *sk, bool add);
+#else
+static inline void udp_tunnel_update_gro_lookup(struct net *net,
+						struct sock *sk, bool add) {}
+static inline void udp_tunnel_update_gro_rcv(struct sock *sk, bool add) {}
+#endif
+
+static inline void udp_tunnel_cleanup_gro(struct sock *sk)
+{
+	udp_tunnel_update_gro_rcv(sk, false);
+	udp_tunnel_update_gro_lookup(sock_net(sk), sk, false);
+}
+
 static inline void udp_tunnel_encap_enable(struct sock *sk)
 {
 	if (udp_test_and_set_bit(ENCAP_ENABLED, sk))
···
 	if (READ_ONCE(sk->sk_family) == PF_INET6)
 		ipv6_stub->udpv6_encap_enable();
 #endif
+	udp_tunnel_update_gro_rcv(sk, true);
 	udp_encap_enable();
 }
 

net/ipv4/udp.c

···
 			if (encap_destroy)
 				encap_destroy(sk);
 		}
-		if (udp_test_bit(ENCAP_ENABLED, sk))
+		if (udp_test_bit(ENCAP_ENABLED, sk)) {
 			static_branch_dec(&udp_encap_needed_key);
+			udp_tunnel_cleanup_gro(sk);
+		}
 	}
 }
 
···
 
 static int __net_init udp_pernet_init(struct net *net)
 {
+#if IS_ENABLED(CONFIG_NET_UDP_TUNNEL)
+	int i;
+
+	/* No tunnel is configured */
+	for (i = 0; i < ARRAY_SIZE(net->ipv4.udp_tunnel_gro); ++i) {
+		INIT_HLIST_HEAD(&net->ipv4.udp_tunnel_gro[i].list);
+		RCU_INIT_POINTER(net->ipv4.udp_tunnel_gro[i].sk, NULL);
+	}
+#endif
 	udp_sysctl_init(net);
 	udp_set_table(net);
 

net/ipv4/udp_offload.c

···
 #include <net/udp.h>
 #include <net/protocol.h>
 #include <net/inet_common.h>
+#include <net/udp_tunnel.h>
+
+#if IS_ENABLED(CONFIG_NET_UDP_TUNNEL)
+
+/*
+ * Dummy GRO tunnel callback, exists mainly to avoid dangling/NULL
+ * values for the udp tunnel static call.
+ */
+static struct sk_buff *dummy_gro_rcv(struct sock *sk,
+				     struct list_head *head,
+				     struct sk_buff *skb)
+{
+	NAPI_GRO_CB(skb)->flush = 1;
+	return NULL;
+}
+
+typedef struct sk_buff *(*udp_tunnel_gro_rcv_t)(struct sock *sk,
+						struct list_head *head,
+						struct sk_buff *skb);
+
+struct udp_tunnel_type_entry {
+	udp_tunnel_gro_rcv_t gro_receive;
+	refcount_t count;
+};
+
+#define UDP_MAX_TUNNEL_TYPES (IS_ENABLED(CONFIG_GENEVE) + \
+			      IS_ENABLED(CONFIG_VXLAN) * 2 + \
+			      IS_ENABLED(CONFIG_NET_FOU) * 2 + \
+			      IS_ENABLED(CONFIG_XFRM) * 2)
+
+DEFINE_STATIC_CALL(udp_tunnel_gro_rcv, dummy_gro_rcv);
+static DEFINE_STATIC_KEY_FALSE(udp_tunnel_static_call);
+static struct mutex udp_tunnel_gro_type_lock;
+static struct udp_tunnel_type_entry udp_tunnel_gro_types[UDP_MAX_TUNNEL_TYPES];
+static unsigned int udp_tunnel_gro_type_nr;
+static DEFINE_SPINLOCK(udp_tunnel_gro_lock);
+
+void udp_tunnel_update_gro_lookup(struct net *net, struct sock *sk, bool add)
+{
+	bool is_ipv6 = sk->sk_family == AF_INET6;
+	struct udp_sock *tup, *up = udp_sk(sk);
+	struct udp_tunnel_gro *udp_tunnel_gro;
+
+	spin_lock(&udp_tunnel_gro_lock);
+	udp_tunnel_gro = &net->ipv4.udp_tunnel_gro[is_ipv6];
+	if (add)
+		hlist_add_head(&up->tunnel_list, &udp_tunnel_gro->list);
+	else if (up->tunnel_list.pprev)
+		hlist_del_init(&up->tunnel_list);
+
+	if (udp_tunnel_gro->list.first &&
+	    !udp_tunnel_gro->list.first->next) {
+		tup = hlist_entry(udp_tunnel_gro->list.first, struct udp_sock,
+				  tunnel_list);
+
+		rcu_assign_pointer(udp_tunnel_gro->sk, (struct sock *)tup);
+	} else {
+		RCU_INIT_POINTER(udp_tunnel_gro->sk, NULL);
+	}
+
+	spin_unlock(&udp_tunnel_gro_lock);
+}
+EXPORT_SYMBOL_GPL(udp_tunnel_update_gro_lookup);
+
+void udp_tunnel_update_gro_rcv(struct sock *sk, bool add)
+{
+	struct udp_tunnel_type_entry *cur = NULL;
+	struct udp_sock *up = udp_sk(sk);
+	int i, old_gro_type_nr;
+
+	if (!UDP_MAX_TUNNEL_TYPES || !up->gro_receive)
+		return;
+
+	mutex_lock(&udp_tunnel_gro_type_lock);
+
+	/* Check if the static call is permanently disabled. */
+	if (udp_tunnel_gro_type_nr > UDP_MAX_TUNNEL_TYPES)
+		goto out;
+
+	for (i = 0; i < udp_tunnel_gro_type_nr; i++)
+		if (udp_tunnel_gro_types[i].gro_receive == up->gro_receive)
+			cur = &udp_tunnel_gro_types[i];
+
+	old_gro_type_nr = udp_tunnel_gro_type_nr;
+	if (add) {
+		/*
+		 * Update the matching entry, if found, or add a new one
+		 * if needed
+		 */
+		if (cur) {
+			refcount_inc(&cur->count);
+			goto out;
+		}
+
+		if (unlikely(udp_tunnel_gro_type_nr == UDP_MAX_TUNNEL_TYPES)) {
+			pr_err_once("Too many UDP tunnel types, please increase UDP_MAX_TUNNEL_TYPES\n");
+			/* Ensure static call will never be enabled */
+			udp_tunnel_gro_type_nr = UDP_MAX_TUNNEL_TYPES + 1;
+		} else {
+			cur = &udp_tunnel_gro_types[udp_tunnel_gro_type_nr++];
+			refcount_set(&cur->count, 1);
+			cur->gro_receive = up->gro_receive;
+		}
+	} else {
+		/*
+		 * The stack cleanups only successfully added tunnel, the
+		 * lookup on removal should never fail.
+		 */
+		if (WARN_ON_ONCE(!cur))
+			goto out;
+
+		if (!refcount_dec_and_test(&cur->count))
+			goto out;
+
+		/* Avoid gaps, so that the enable tunnel has always id 0 */
+		*cur = udp_tunnel_gro_types[--udp_tunnel_gro_type_nr];
+	}
+
+	if (udp_tunnel_gro_type_nr == 1) {
+		static_call_update(udp_tunnel_gro_rcv,
+				   udp_tunnel_gro_types[0].gro_receive);
+		static_branch_enable(&udp_tunnel_static_call);
+	} else if (old_gro_type_nr == 1) {
+		static_branch_disable(&udp_tunnel_static_call);
+		static_call_update(udp_tunnel_gro_rcv, dummy_gro_rcv);
+	}
+
+out:
+	mutex_unlock(&udp_tunnel_gro_type_lock);
+}
+EXPORT_SYMBOL_GPL(udp_tunnel_update_gro_rcv);
+
+static void udp_tunnel_gro_init(void)
+{
+	mutex_init(&udp_tunnel_gro_type_lock);
+}
+
+static struct sk_buff *udp_tunnel_gro_rcv(struct sock *sk,
+					  struct list_head *head,
+					  struct sk_buff *skb)
+{
+	if (static_branch_likely(&udp_tunnel_static_call)) {
+		if (unlikely(gro_recursion_inc_test(skb))) {
+			NAPI_GRO_CB(skb)->flush |= 1;
+			return NULL;
+		}
+		return static_call(udp_tunnel_gro_rcv)(sk, head, skb);
+	}
+	return call_gro_receive_sk(udp_sk(sk)->gro_receive, sk, head, skb);
+}
+
+#else
+
+static void udp_tunnel_gro_init(void) {}
+
+static struct sk_buff *udp_tunnel_gro_rcv(struct sock *sk,
+					  struct list_head *head,
+					  struct sk_buff *skb)
+{
+	return call_gro_receive_sk(udp_sk(sk)->gro_receive, sk, head, skb);
+}
+
+#endif
 
 static struct sk_buff *__skb_udp_tunnel_segment(struct sk_buff *skb,
 	netdev_features_t features,
···
 
 	skb_gro_pull(skb, sizeof(struct udphdr)); /* pull encapsulating udp header */
 	skb_gro_postpull_rcsum(skb, uh, sizeof(struct udphdr));
-	pp = call_gro_receive_sk(udp_sk(sk)->gro_receive, sk, head, skb);
+	pp = udp_tunnel_gro_rcv(sk, head, skb);
 
 out:
 	skb_gro_flush_final(skb, pp, flush);
···
 {
 	const struct iphdr *iph = skb_gro_network_header(skb);
 	struct net *net = dev_net_rcu(skb->dev);
+	struct sock *sk;
 	int iif, sdif;
+
+	sk = udp_tunnel_sk(net, false);
+	if (sk && dport == htons(sk->sk_num))
+		return sk;
 
 	inet_get_iif_sdif(skb, &iif, &sdif);
 
···
 			.gro_complete =	udp4_gro_complete,
 		},
 	};
+
+	udp_tunnel_gro_init();
 	return inet_add_offload(&net_hotdata.udpv4_offload, IPPROTO_UDP);
 }

net/ipv4/udp_tunnel_core.c

···
 }
 EXPORT_SYMBOL(udp_sock_create4);
 
+static bool sk_saddr_any(struct sock *sk)
+{
+#if IS_ENABLED(CONFIG_IPV6)
+	return ipv6_addr_any(&sk->sk_v6_rcv_saddr);
+#else
+	return !sk->sk_rcv_saddr;
+#endif
+}
+
 void setup_udp_tunnel_sock(struct net *net, struct socket *sock,
 			   struct udp_tunnel_sock_cfg *cfg)
 {
···
 	udp_sk(sk)->gro_complete = cfg->gro_complete;
 
 	udp_tunnel_encap_enable(sk);
+
+	if (!sk->sk_dport && !sk->sk_bound_dev_if && sk_saddr_any(sk) &&
+	    sk->sk_kern_sock)
+		udp_tunnel_update_gro_lookup(net, sk, true);
 }
 EXPORT_SYMBOL_GPL(setup_udp_tunnel_sock);
 

net/ipv6/udp.c

···
 #include <net/tcp_states.h>
 #include <net/ip6_checksum.h>
 #include <net/ip6_tunnel.h>
+#include <net/udp_tunnel.h>
 #include <net/xfrm.h>
 #include <net/inet_hashtables.h>
 #include <net/inet6_hashtables.h>
···
 		if (udp_test_bit(ENCAP_ENABLED, sk)) {
 			static_branch_dec(&udpv6_encap_needed_key);
 			udp_encap_disable();
+			udp_tunnel_cleanup_gro(sk);
 		}
 	}
 }

net/ipv6/udp_offload.c

···
 {
 	const struct ipv6hdr *iph = skb_gro_network_header(skb);
 	struct net *net = dev_net_rcu(skb->dev);
+	struct sock *sk;
 	int iif, sdif;
+
+	sk = udp_tunnel_sk(net, true);
+	if (sk && dport == htons(sk->sk_num))
+		return sk;
 
 	inet6_get_iif_sdif(skb, &iif, &sdif);