tcp: Add stats for socket migration.

This commit adds two stats for the socket migration feature to evaluate the
effectiveness: LINUX_MIB_TCPMIGRATEREQ(SUCCESS|FAILURE).

If the migration fails because of the own_req race in receiving ACK and
sending SYN+ACK paths, we do not increment the failure stat. Then another
CPU is responsible for the req.

Link: https://lore.kernel.org/bpf/CAK6E8=cgFKuGecTzSCSQ8z3YJ_163C0uwO9yRvfDSE7vOe9mJA@mail.gmail.com/
Suggested-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.co.jp>
Acked-by: Yuchung Cheng <ycheng@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

3 files changed, 18 insertions, 2 deletions

diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 0eea878edc30..754013fa393b 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -703,6 +703,8 @@ static struct request_sock *inet_reqsk_clone(struct request_sock *req,
 
 	nreq = kmem_cache_alloc(req->rsk_ops->slab, GFP_ATOMIC | __GFP_NOWARN);
 	if (!nreq) {
+		__NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMIGRATEREQFAILURE);
+
 		/* paired with refcount_inc_not_zero() in reuseport_migrate_sock() */
 		sock_put(sk);
 		return NULL;
@@ -876,9 +878,10 @@ static void reqsk_timer_handler(struct timer_list *t)
 		if (!inet_ehash_insert(req_to_sk(nreq), req_to_sk(oreq), NULL)) {
 			/* delete timer */
 			inet_csk_reqsk_queue_drop(sk_listener, nreq);
-			goto drop;
+			goto no_ownership;
 		}
 
+		__NET_INC_STATS(net, LINUX_MIB_TCPMIGRATEREQSUCCESS);
 		reqsk_migrate_reset(oreq);
 		reqsk_queue_removed(&inet_csk(oreq->rsk_listener)->icsk_accept_queue, oreq);
 		reqsk_put(oreq);
@@ -887,17 +890,19 @@ static void reqsk_timer_handler(struct timer_list *t)
 		return;
 	}
 
-drop:
 	/* Even if we can clone the req, we may need not retransmit any more
 	 * SYN+ACKs (nreq->num_timeout > max_syn_ack_retries, etc), or another
 	 * CPU may win the "own_req" race so that inet_ehash_insert() fails.
 	 */
 	if (nreq) {
+		__NET_INC_STATS(net, LINUX_MIB_TCPMIGRATEREQFAILURE);
+no_ownership:
 		reqsk_migrate_reset(nreq);
 		reqsk_queue_removed(queue, nreq);
 		__reqsk_free(nreq);
 	}
 
+drop:
 	inet_csk_reqsk_queue_drop_and_put(oreq->rsk_listener, oreq);
 }
 
@@ -1135,11 +1140,13 @@ struct sock *inet_csk_complete_hashdance(struct sock *sk, struct sock *child,
 
 			refcount_set(&nreq->rsk_refcnt, 1);
 			if (inet_csk_reqsk_queue_add(sk, nreq, child)) {
+				__NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMIGRATEREQSUCCESS);
 				reqsk_migrate_reset(req);
 				reqsk_put(req);
 				return child;
 			}
 
+			__NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMIGRATEREQFAILURE);
 			reqsk_migrate_reset(nreq);
 			__reqsk_free(nreq);
 		} else if (inet_csk_reqsk_queue_add(sk, req, child)) {
@@ -1188,8 +1195,12 @@ void inet_csk_listen_stop(struct sock *sk)
 				refcount_set(&nreq->rsk_refcnt, 1);
 
 				if (inet_csk_reqsk_queue_add(nsk, nreq, child)) {
+					__NET_INC_STATS(sock_net(nsk),
+							LINUX_MIB_TCPMIGRATEREQSUCCESS);
 					reqsk_migrate_reset(req);
 				} else {
+					__NET_INC_STATS(sock_net(nsk),
+							LINUX_MIB_TCPMIGRATEREQFAILURE);
 					reqsk_migrate_reset(nreq);
 					__reqsk_free(nreq);
 				}
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index 6d46297a99f8..b0d3a09dc84e 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -295,6 +295,8 @@ static const struct snmp_mib snmp4_net_list[] = {
 	SNMP_MIB_ITEM("TcpDuplicateDataRehash", LINUX_MIB_TCPDUPLICATEDATAREHASH),
 	SNMP_MIB_ITEM("TCPDSACKRecvSegs", LINUX_MIB_TCPDSACKRECVSEGS),
 	SNMP_MIB_ITEM("TCPDSACKIgnoredDubious", LINUX_MIB_TCPDSACKIGNOREDDUBIOUS),
+	SNMP_MIB_ITEM("TCPMigrateReqSuccess", LINUX_MIB_TCPMIGRATEREQSUCCESS),
+	SNMP_MIB_ITEM("TCPMigrateReqFailure", LINUX_MIB_TCPMIGRATEREQFAILURE),
 	SNMP_MIB_SENTINEL
 };
 
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index f258a4c0da71..0a4f3f16140a 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -786,6 +786,9 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb,
 	return inet_csk_complete_hashdance(sk, child, req, own_req);
 
 listen_overflow:
+	if (sk != req->rsk_listener)
+		__NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMIGRATEREQFAILURE);
+
 	if (!sock_net(sk)->ipv4.sysctl_tcp_abort_on_overflow) {
 		inet_rsk(req)->acked = 1;
 		return NULL;