txhash: Make rethinking txhash behavior configurable via sysctl

Add a per ns sysctl that controls the txhash rethink behavior:
net.core.txrehash. When enabled, the same behavior is retained,
when disabled, rethink is not performed. Sysctl is enabled by default.

Signed-off-by: Akhmat Karakotov <hmukos@yandex-team.ru>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

5 files changed, 39 insertions, 15 deletions

diff --git a/include/net/netns/core.h b/include/net/netns/core.h
index 552bc25b1933..388244e315e7 100644
--- a/include/net/netns/core.h
+++ b/include/net/netns/core.h
@@ -10,6 +10,7 @@ struct netns_core {
 	struct ctl_table_header	*sysctl_hdr;
 
 	int	sysctl_somaxconn;
+	u8	sysctl_txrehash;
 
 #ifdef CONFIG_PROC_FS
 	struct prot_inuse __percpu *prot_inuse;
diff --git a/include/net/sock.h b/include/net/sock.h
index ff9b508d9c5f..0540e1b2aeb0 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -587,6 +587,18 @@ static inline bool sk_user_data_is_nocopy(const struct sock *sk)
 			   __tmp | SK_USER_DATA_NOCOPY);		\
 })
 
+static inline
+struct net *sock_net(const struct sock *sk)
+{
+	return read_pnet(&sk->sk_net);
+}
+
+static inline
+void sock_net_set(struct sock *sk, struct net *net)
+{
+	write_pnet(&sk->sk_net, net);
+}
+
 /*
  * SK_CAN_REUSE and SK_NO_REUSE on a socket mean that the socket is OK
  * or not whether his port will be reused by someone else. SK_FORCE_REUSE
@@ -2054,10 +2066,18 @@ static inline void sk_set_txhash(struct sock *sk)
 
 static inline bool sk_rethink_txhash(struct sock *sk)
 {
-	if (sk->sk_txhash) {
+	u8 rehash;
+
+	if (!sk->sk_txhash)
+		return false;
+
+	rehash = READ_ONCE(sock_net(sk)->core.sysctl_txrehash);
+
+	if (rehash) {
 		sk_set_txhash(sk);
 		return true;
 	}
+
 	return false;
 }
 
@@ -2704,18 +2724,6 @@ static inline void sk_eat_skb(struct sock *sk, struct sk_buff *skb)
 	__kfree_skb(skb);
 }
 
-static inline
-struct net *sock_net(const struct sock *sk)
-{
-	return read_pnet(&sk->sk_net);
-}
-
-static inline
-void sock_net_set(struct sock *sk, struct net *net)
-{
-	write_pnet(&sk->sk_net, net);
-}
-
 static inline bool
 skb_sk_is_prefetched(struct sk_buff *skb)
 {
diff --git a/include/uapi/linux/socket.h b/include/uapi/linux/socket.h
index eb0a9a5b6e71..0accd6102ece 100644
--- a/include/uapi/linux/socket.h
+++ b/include/uapi/linux/socket.h
@@ -31,4 +31,7 @@ struct __kernel_sockaddr_storage {
 
 #define SOCK_BUF_LOCK_MASK (SOCK_SNDBUF_LOCK | SOCK_RCVBUF_LOCK)
 
+#define SOCK_TXREHASH_DISABLED	0
+#define SOCK_TXREHASH_ENABLED	1
+
 #endif /* _UAPI_LINUX_SOCKET_H */
diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
index c53d9aab38ab..8711350085d6 100644
--- a/net/core/net_namespace.c
+++ b/net/core/net_namespace.c
@@ -364,6 +364,8 @@ out_undo:
 static int __net_init net_defaults_init_net(struct net *net)
 {
 	net->core.sysctl_somaxconn = SOMAXCONN;
+	net->core.sysctl_txrehash = SOCK_TXREHASH_ENABLED;
+
 	return 0;
 }
 
diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c
index 7b4d485aac7a..dbeb8ecbcd98 100644
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -593,6 +593,15 @@ static struct ctl_table netns_core_table[] = {
 		.extra1		= SYSCTL_ZERO,
 		.proc_handler	= proc_dointvec_minmax
 	},
+	{
+		.procname	= "txrehash",
+		.data		= &init_net.core.sysctl_txrehash,
+		.maxlen		= sizeof(u8),
+		.mode		= 0644,
+		.extra1		= SYSCTL_ZERO,
+		.extra2		= SYSCTL_ONE,
+		.proc_handler	= proc_dou8vec_minmax,
+	},
 	{ }
 };
 
@@ -611,7 +620,7 @@ __setup("fb_tunnels=", fb_tunnels_only_for_init_net_sysctl_setup);
 
 static __net_init int sysctl_core_net_init(struct net *net)
 {
-	struct ctl_table *tbl;
+	struct ctl_table *tbl, *tmp;
 
 	tbl = netns_core_table;
 	if (!net_eq(net, &init_net)) {
@@ -619,7 +628,8 @@ static __net_init int sysctl_core_net_init(struct net *net)
 		if (tbl == NULL)
 			goto err_dup;
 
-		tbl[0].data = &net->core.sysctl_somaxconn;
+		for (tmp = tbl; tmp->procname; tmp++)
+			tmp->data += (char *)net - (char *)&init_net;
 
 		/* Don't export any sysctls to unprivileged users */
 		if (net->user_ns != &init_user_ns) {