net: sock_copy() fixes

Commit e912b114
(net: sk_prot_alloc() should not blindly overwrite memory)
took care of not zeroing whole new socket at allocation time.

sock_copy() is another spot where we should be very careful.
We should not set refcnt to a non null value, until
we are sure other fields are correctly setup, or
a lockless reader could catch this socket by mistake,
while not fully (re)initialized.

This patch puts sk_node & sk_refcnt to the very beginning
of struct sock to ease sock_copy() & sk_prot_alloc() job.

We add appropriate smp_wmb() before sk_refcnt initializations
to match our RCU requirements (changes to sock keys should
be committed to memory before sk_refcnt setting)
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

include/net/sock.h

@@ -104,15 +104,15 @@ struct net;
 
 /**
  *	struct sock_common - minimal network layer representation of sockets
+ *	@skc_node: main hash linkage for various protocol lookup tables
+ *	@skc_nulls_node: main hash linkage for UDP/UDP-Lite protocol
+ *	@skc_refcnt: reference count
+ *	@skc_hash: hash value used with various protocol lookup tables
  *	@skc_family: network address family
  *	@skc_state: Connection state
  *	@skc_reuse: %SO_REUSEADDR setting
  *	@skc_bound_dev_if: bound device index if != 0
- *	@skc_node: main hash linkage for various protocol lookup tables
- *	@skc_nulls_node: main hash linkage for UDP/UDP-Lite protocol
  *	@skc_bind_node: bind hash linkage for various protocol lookup tables
- *	@skc_refcnt: reference count
- *	@skc_hash: hash value used with various protocol lookup tables
  *	@skc_prot: protocol handlers inside a network family
  *	@skc_net: reference to the network namespace of this socket
  *
@@ -120,17 +120,21 @@ struct net;
  *	for struct sock and struct inet_timewait_sock.
  */
 struct sock_common {
-	unsigned short		skc_family;
-	volatile unsigned char	skc_state;
-	unsigned char		skc_reuse;
-	int			skc_bound_dev_if;
+	/*
+	 * first fields are not copied in sock_copy()
+	 */
 	union {
 		struct hlist_node	skc_node;
 		struct hlist_nulls_node skc_nulls_node;
 	};
-	struct hlist_node	skc_bind_node;
 	atomic_t		skc_refcnt;
+
 	unsigned int		skc_hash;
+	unsigned short		skc_family;
+	volatile unsigned char	skc_state;
+	unsigned char		skc_reuse;
+	int			skc_bound_dev_if;
+	struct hlist_node	skc_bind_node;
 	struct proto		*skc_prot;
 #ifdef CONFIG_NET_NS
 	struct net	 	*skc_net;
@@ -208,15 +212,17 @@ struct sock {
 	 * don't add nothing before this first member (__sk_common) --acme
 	 */
 	struct sock_common	__sk_common;
+#define sk_node			__sk_common.skc_node
+#define sk_nulls_node		__sk_common.skc_nulls_node
+#define sk_refcnt		__sk_common.skc_refcnt
+
+#define sk_copy_start		__sk_common.skc_hash
+#define sk_hash			__sk_common.skc_hash
 #define sk_family		__sk_common.skc_family
 #define sk_state		__sk_common.skc_state
 #define sk_reuse		__sk_common.skc_reuse
 #define sk_bound_dev_if		__sk_common.skc_bound_dev_if
-#define sk_node			__sk_common.skc_node
-#define sk_nulls_node		__sk_common.skc_nulls_node
 #define sk_bind_node		__sk_common.skc_bind_node
-#define sk_refcnt		__sk_common.skc_refcnt
-#define sk_hash			__sk_common.skc_hash
 #define sk_prot			__sk_common.skc_prot
 #define sk_net			__sk_common.skc_net
 	kmemcheck_bitfield_begin(flags);

net/core/sock.c

@@ -919,13 +919,19 @@ static inline void sock_lock_init(struct sock *sk)
 			af_family_keys + sk->sk_family);
 }
 
+/*
+ * 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.
+ */
 static void sock_copy(struct sock *nsk, const struct sock *osk)
 {
 #ifdef CONFIG_SECURITY_NETWORK
 	void *sptr = nsk->sk_security;
 #endif
-
-	memcpy(nsk, osk, osk->sk_prot->obj_size);
+	BUILD_BUG_ON(offsetof(struct sock, sk_copy_start) !=
+		     sizeof(osk->sk_node) + sizeof(osk->sk_refcnt));
+	memcpy(&nsk->sk_copy_start, &osk->sk_copy_start,
+	       osk->sk_prot->obj_size - offsetof(struct sock, sk_copy_start));
 #ifdef CONFIG_SECURITY_NETWORK
 	nsk->sk_security = sptr;
 	security_sk_clone(osk, nsk);
@@ -1140,6 +1146,11 @@ struct sock *sk_clone(const struct sock *sk, const gfp_t priority)
 
 		newsk->sk_err	   = 0;
 		newsk->sk_priority = 0;
+		/*
+		 * Before updating sk_refcnt, we must commit prior changes to memory
+		 * (Documentation/RCU/rculist_nulls.txt for details)
+		 */
+		smp_wmb();
 		atomic_set(&newsk->sk_refcnt, 2);
 
 		/*
@@ -1855,6 +1866,11 @@ void sock_init_data(struct socket *sock, struct sock *sk)
 
 	sk->sk_stamp = ktime_set(-1L, 0);
 
+	/*
+	 * Before updating sk_refcnt, we must commit prior changes to memory
+	 * (Documentation/RCU/rculist_nulls.txt for details)
+	 */
+	smp_wmb();
 	atomic_set(&sk->sk_refcnt, 1);
 	atomic_set(&sk->sk_wmem_alloc, 1);
 	atomic_set(&sk->sk_drops, 0);