net: optimize INET input path further

Followup of commit b178bb3dfc30 (net: reorder struct sock fields)

Optimize INET input path a bit further, by :

1) moving sk_refcnt close to sk_lock.

This reduces number of dirtied cache lines by one on 64bit arches (and
64 bytes cache line size).

2) moving inet_daddr & inet_rcv_saddr at the beginning of sk

(same cache line than hash / family / bound_dev_if / nulls_node)

This reduces number of accessed cache lines in lookups by one, and dont
increase size of inet and timewait socks.
inet and tw sockets now share same place-holder for these fields.

Before patch :

offsetof(struct sock, sk_refcnt) = 0x10
offsetof(struct sock, sk_lock) = 0x40
offsetof(struct sock, sk_receive_queue) = 0x60
offsetof(struct inet_sock, inet_daddr) = 0x270
offsetof(struct inet_sock, inet_rcv_saddr) = 0x274

After patch :

offsetof(struct sock, sk_refcnt) = 0x44
offsetof(struct sock, sk_lock) = 0x48
offsetof(struct sock, sk_receive_queue) = 0x68
offsetof(struct inet_sock, inet_daddr) = 0x0
offsetof(struct inet_sock, inet_rcv_saddr) = 0x4

compute_score() (udp or tcp) now use a single cache line per ignored
item, instead of two.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

3 files changed, 11 insertions, 11 deletions

diff --git a/net/core/sock.c b/net/core/sock.c
index fb60801..bcdb6ff 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -992,17 +992,18 @@ static inline void sock_lock_init(struct sock *sk)
 /*
  * Copy all fields from osk to nsk but nsk->sk_refcnt must not change yet,
  * even temporarly, because of RCU lookups. sk_node should also be left as is.
+ * We must not copy fields between sk_dontcopy_begin and sk_dontcopy_end
  */
 static void sock_copy(struct sock *nsk, const struct sock *osk)
 {
 #ifdef CONFIG_SECURITY_NETWORK
 	void *sptr = nsk->sk_security;
 #endif
-	BUILD_BUG_ON(offsetof(struct sock, sk_copy_start) !=
-		     sizeof(osk->sk_node) + sizeof(osk->sk_refcnt) +
-		     sizeof(osk->sk_tx_queue_mapping));
-	memcpy(&nsk->sk_copy_start, &osk->sk_copy_start,
-	       osk->sk_prot->obj_size - offsetof(struct sock, sk_copy_start));
+	memcpy(nsk, osk, offsetof(struct sock, sk_dontcopy_begin));
+
+	memcpy(&nsk->sk_dontcopy_end, &osk->sk_dontcopy_end,
+	       osk->sk_prot->obj_size - offsetof(struct sock, sk_dontcopy_end));
+
 #ifdef CONFIG_SECURITY_NETWORK
 	nsk->sk_security = sptr;
 	security_sk_clone(osk, nsk);
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 06f5f8f..25e3181 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -55,7 +55,6 @@ EXPORT_SYMBOL(inet_get_local_port_range);
 int inet_csk_bind_conflict(const struct sock *sk,
 			   const struct inet_bind_bucket *tb)
 {
-	const __be32 sk_rcv_saddr = inet_rcv_saddr(sk);
 	struct sock *sk2;
 	struct hlist_node *node;
 	int reuse = sk->sk_reuse;
@@ -75,9 +74,9 @@ int inet_csk_bind_conflict(const struct sock *sk,
 		     sk->sk_bound_dev_if == sk2->sk_bound_dev_if)) {
 			if (!reuse || !sk2->sk_reuse ||
 			    sk2->sk_state == TCP_LISTEN) {
-				const __be32 sk2_rcv_saddr = inet_rcv_saddr(sk2);
-				if (!sk2_rcv_saddr || !sk_rcv_saddr ||
-				    sk2_rcv_saddr == sk_rcv_saddr)
+				const __be32 sk2_rcv_saddr = sk_rcv_saddr(sk2);
+				if (!sk2_rcv_saddr || !sk_rcv_saddr(sk) ||
+				    sk2_rcv_saddr == sk_rcv_saddr(sk))
 					break;
 			}
 		}
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index b541a4e..7aad127 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -54,8 +54,8 @@ int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2)
 {
 	const struct in6_addr *sk_rcv_saddr6 = &inet6_sk(sk)->rcv_saddr;
 	const struct in6_addr *sk2_rcv_saddr6 = inet6_rcv_saddr(sk2);
-	__be32 sk1_rcv_saddr = inet_sk(sk)->inet_rcv_saddr;
-	__be32 sk2_rcv_saddr = inet_rcv_saddr(sk2);
+	__be32 sk1_rcv_saddr = sk_rcv_saddr(sk);
+	__be32 sk2_rcv_saddr = sk_rcv_saddr(sk2);
 	int sk_ipv6only = ipv6_only_sock(sk);
 	int sk2_ipv6only = inet_v6_ipv6only(sk2);
 	int addr_type = ipv6_addr_type(sk_rcv_saddr6);