Merge commit 'linus/master' into merge-linus

Conflicts:

	arch/x86/kvm/i8254.c

1 files changed, 86 insertions, 101 deletions

diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 8e42fbbd576..2095abc3cab 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -8,7 +8,7 @@
  * Authors:	Ross Biro
  *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
  *		Arnt Gulbrandsen, <agulbra@nvg.unit.no>
- *		Alan Cox, <Alan.Cox@linux.org>
+ *		Alan Cox, <alan@lxorguk.ukuu.org.uk>
  *		Hirokazu Takahashi, <taka@valinux.co.jp>
  *
  * Fixes:
@@ -108,9 +108,6 @@
  *	Snmp MIB for the UDP layer
  */
 
-DEFINE_SNMP_STAT(struct udp_mib, udp_stats_in6) __read_mostly;
-EXPORT_SYMBOL(udp_stats_in6);
-
 struct hlist_head udp_hash[UDP_HTABLE_SIZE];
 DEFINE_RWLOCK(udp_hash_lock);
 
@@ -125,14 +122,23 @@ EXPORT_SYMBOL(sysctl_udp_wmem_min);
 atomic_t udp_memory_allocated;
 EXPORT_SYMBOL(udp_memory_allocated);
 
-static inline int __udp_lib_lport_inuse(struct net *net, __u16 num,
-					const struct hlist_head udptable[])
+static int udp_lib_lport_inuse(struct net *net, __u16 num,
+			       const struct hlist_head udptable[],
+			       struct sock *sk,
+			       int (*saddr_comp)(const struct sock *sk1,
+						 const struct sock *sk2))
 {
-	struct sock *sk;
+	struct sock *sk2;
 	struct hlist_node *node;
 
-	sk_for_each(sk, node, &udptable[udp_hashfn(net, num)])
-		if (net_eq(sock_net(sk), net) && sk->sk_hash == num)
+	sk_for_each(sk2, node, &udptable[udp_hashfn(net, num)])
+		if (net_eq(sock_net(sk2), net)			&&
+		    sk2 != sk					&&
+		    sk2->sk_hash == num				&&
+		    (!sk2->sk_reuse || !sk->sk_reuse)		&&
+		    (!sk2->sk_bound_dev_if || !sk->sk_bound_dev_if
+			|| sk2->sk_bound_dev_if == sk->sk_bound_dev_if) &&
+		    (*saddr_comp)(sk, sk2))
 			return 1;
 	return 0;
 }
@@ -149,83 +155,37 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum,
 					 const struct sock *sk2 )    )
 {
 	struct hlist_head *udptable = sk->sk_prot->h.udp_hash;
-	struct hlist_node *node;
-	struct hlist_head *head;
-	struct sock *sk2;
 	int    error = 1;
 	struct net *net = sock_net(sk);
 
 	write_lock_bh(&udp_hash_lock);
 
 	if (!snum) {
-		int i, low, high, remaining;
-		unsigned rover, best, best_size_so_far;
+		int low, high, remaining;
+		unsigned rand;
+		unsigned short first;
 
 		inet_get_local_port_range(&low, &high);
 		remaining = (high - low) + 1;
 
-		best_size_so_far = UINT_MAX;
-		best = rover = net_random() % remaining + low;
-
-		/* 1st pass: look for empty (or shortest) hash chain */
-		for (i = 0; i < UDP_HTABLE_SIZE; i++) {
-			int size = 0;
-
-			head = &udptable[udp_hashfn(net, rover)];
-			if (hlist_empty(head))
-				goto gotit;
-
-			sk_for_each(sk2, node, head) {
-				if (++size >= best_size_so_far)
-					goto next;
-			}
-			best_size_so_far = size;
-			best = rover;
-		next:
-			/* fold back if end of range */
-			if (++rover > high)
-				rover = low + ((rover - low)
-					       & (UDP_HTABLE_SIZE - 1));
-
-
-		}
-
-		/* 2nd pass: find hole in shortest hash chain */
-		rover = best;
-		for (i = 0; i < (1 << 16) / UDP_HTABLE_SIZE; i++) {
-			if (! __udp_lib_lport_inuse(net, rover, udptable))
-				goto gotit;
-			rover += UDP_HTABLE_SIZE;
-			if (rover > high)
-				rover = low + ((rover - low)
-					       & (UDP_HTABLE_SIZE - 1));
+		rand = net_random();
+		snum = first = rand % remaining + low;
+		rand |= 1;
+		while (udp_lib_lport_inuse(net, snum, udptable, sk,
+					   saddr_comp)) {
+			do {
+				snum = snum + rand;
+			} while (snum < low || snum > high);
+			if (snum == first)
+				goto fail;
 		}
-
-
-		/* All ports in use! */
+	} else if (udp_lib_lport_inuse(net, snum, udptable, sk, saddr_comp))
 		goto fail;
 
-gotit:
-		snum = rover;
-	} else {
-		head = &udptable[udp_hashfn(net, snum)];
-
-		sk_for_each(sk2, node, head)
-			if (sk2->sk_hash == snum                             &&
-			    sk2 != sk                                        &&
-			    net_eq(sock_net(sk2), net)			     &&
-			    (!sk2->sk_reuse        || !sk->sk_reuse)         &&
-			    (!sk2->sk_bound_dev_if || !sk->sk_bound_dev_if
-			     || sk2->sk_bound_dev_if == sk->sk_bound_dev_if) &&
-			    (*saddr_comp)(sk, sk2)                             )
-				goto fail;
-	}
-
 	inet_sk(sk)->num = snum;
 	sk->sk_hash = snum;
 	if (sk_unhashed(sk)) {
-		head = &udptable[udp_hashfn(net, snum)];
-		sk_add_node(sk, head);
+		sk_add_node(sk, &udptable[udp_hashfn(net, snum)]);
 		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	}
 	error = 0;
@@ -302,6 +262,28 @@ static struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr,
 	return result;
 }
 
+static inline struct sock *__udp4_lib_lookup_skb(struct sk_buff *skb,
+						 __be16 sport, __be16 dport,
+						 struct hlist_head udptable[])
+{
+	struct sock *sk;
+	const struct iphdr *iph = ip_hdr(skb);
+
+	if (unlikely(sk = skb_steal_sock(skb)))
+		return sk;
+	else
+		return __udp4_lib_lookup(dev_net(skb->dst->dev), iph->saddr, sport,
+					 iph->daddr, dport, inet_iif(skb),
+					 udptable);
+}
+
+struct sock *udp4_lib_lookup(struct net *net, __be32 saddr, __be16 sport,
+			     __be32 daddr, __be16 dport, int dif)
+{
+	return __udp4_lib_lookup(net, saddr, sport, daddr, dport, dif, udp_hash);
+}
+EXPORT_SYMBOL_GPL(udp4_lib_lookup);
+
 static inline struct sock *udp_v4_mcast_next(struct sock *sk,
 					     __be16 loc_port, __be32 loc_addr,
 					     __be16 rmt_port, __be32 rmt_addr,
@@ -951,6 +933,27 @@ int udp_disconnect(struct sock *sk, int flags)
 	return 0;
 }
 
+static int __udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
+{
+	int is_udplite = IS_UDPLITE(sk);
+	int rc;
+
+	if ((rc = sock_queue_rcv_skb(sk, skb)) < 0) {
+		/* Note that an ENOMEM error is charged twice */
+		if (rc == -ENOMEM)
+			UDP_INC_STATS_BH(sock_net(sk), UDP_MIB_RCVBUFERRORS,
+					 is_udplite);
+		goto drop;
+	}
+
+	return 0;
+
+drop:
+	UDP_INC_STATS_BH(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
+	kfree_skb(skb);
+	return -1;
+}
+
 /* returns:
  *  -1: error
  *   0: success
@@ -989,9 +992,7 @@ int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
 		    up->encap_rcv != NULL) {
 			int ret;
 
-			bh_unlock_sock(sk);
 			ret = (*up->encap_rcv)(sk, skb);
-			bh_lock_sock(sk);
 			if (ret <= 0) {
 				UDP_INC_STATS_BH(sock_net(sk),
 						 UDP_MIB_INDATAGRAMS,
@@ -1044,17 +1045,16 @@ int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
 			goto drop;
 	}
 
-	if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) {
-		/* Note that an ENOMEM error is charged twice */
-		if (rc == -ENOMEM) {
-			UDP_INC_STATS_BH(sock_net(sk),
-					UDP_MIB_RCVBUFERRORS, is_udplite);
-			atomic_inc(&sk->sk_drops);
-		}
-		goto drop;
-	}
+	rc = 0;
 
-	return 0;
+	bh_lock_sock(sk);
+	if (!sock_owned_by_user(sk))
+		rc = __udp_queue_rcv_skb(sk, skb);
+	else
+		sk_add_backlog(sk, skb);
+	bh_unlock_sock(sk);
+
+	return rc;
 
 drop:
 	UDP_INC_STATS_BH(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
@@ -1092,15 +1092,7 @@ static int __udp4_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
 				skb1 = skb_clone(skb, GFP_ATOMIC);
 
 			if (skb1) {
-				int ret = 0;
-
-				bh_lock_sock(sk);
-				if (!sock_owned_by_user(sk))
-					ret = udp_queue_rcv_skb(sk, skb1);
-				else
-					sk_add_backlog(sk, skb1);
-				bh_unlock_sock(sk);
-
+				int ret = udp_queue_rcv_skb(sk, skb1);
 				if (ret > 0)
 					/* we should probably re-process instead
 					 * of dropping packets here. */
@@ -1191,17 +1183,10 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
 		return __udp4_lib_mcast_deliver(net, skb, uh,
 				saddr, daddr, udptable);
 
-	sk = __udp4_lib_lookup(net, saddr, uh->source, daddr,
-			uh->dest, inet_iif(skb), udptable);
+	sk = __udp4_lib_lookup_skb(skb, uh->source, uh->dest, udptable);
 
 	if (sk != NULL) {
-		int ret = 0;
-		bh_lock_sock(sk);
-		if (!sock_owned_by_user(sk))
-			ret = udp_queue_rcv_skb(sk, skb);
-		else
-			sk_add_backlog(sk, skb);
-		bh_unlock_sock(sk);
+		int ret = udp_queue_rcv_skb(sk, skb);
 		sock_put(sk);
 
 		/* a return value > 0 means to resubmit the input, but
@@ -1494,7 +1479,7 @@ struct proto udp_prot = {
 	.sendmsg	   = udp_sendmsg,
 	.recvmsg	   = udp_recvmsg,
 	.sendpage	   = udp_sendpage,
-	.backlog_rcv	   = udp_queue_rcv_skb,
+	.backlog_rcv	   = __udp_queue_rcv_skb,
 	.hash		   = udp_lib_hash,
 	.unhash		   = udp_lib_unhash,
 	.get_port	   = udp_v4_get_port,