Merge with /home/shaggy/git/linus-clean/

Signed-off-by: Dave Kleikamp <shaggy@austin.ibm.com>

8 files changed, 76 insertions, 85 deletions

diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index c886b28..e278cb9 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -593,10 +593,13 @@ static void fib_hash_move(struct hlist_head *new_info_hash,
 			  struct hlist_head *new_laddrhash,
 			  unsigned int new_size)
 {
+	struct hlist_head *old_info_hash, *old_laddrhash;
 	unsigned int old_size = fib_hash_size;
-	unsigned int i;
+	unsigned int i, bytes;
 
 	write_lock(&fib_info_lock);
+	old_info_hash = fib_info_hash;
+	old_laddrhash = fib_info_laddrhash;
 	fib_hash_size = new_size;
 
 	for (i = 0; i < old_size; i++) {
@@ -636,6 +639,10 @@ static void fib_hash_move(struct hlist_head *new_info_hash,
 	fib_info_laddrhash = new_laddrhash;
 
 	write_unlock(&fib_info_lock);
+
+	bytes = old_size * sizeof(struct hlist_head *);
+	fib_hash_free(old_info_hash, bytes);
+	fib_hash_free(old_laddrhash, bytes);
 }
 
 struct fib_info *
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 279f57a..3d78464 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -936,8 +936,7 @@ int icmp_rcv(struct sk_buff *skb)
 	case CHECKSUM_HW:
 		if (!(u16)csum_fold(skb->csum))
 			break;
-		NETDEBUG(if (net_ratelimit())
-				printk(KERN_DEBUG "icmp v4 hw csum failure\n"));
+		LIMIT_NETDEBUG(printk(KERN_DEBUG "icmp v4 hw csum failure\n"));
 	case CHECKSUM_NONE:
 		if ((u16)csum_fold(skb_checksum(skb, 0, skb->len, 0)))
 			goto error;
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 7f68e27..eb377ae 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -377,7 +377,7 @@ static struct ipq *ip_frag_create(unsigned hash, struct iphdr *iph, u32 user)
 	return ip_frag_intern(hash, qp);
 
 out_nomem:
-	NETDEBUG(if (net_ratelimit()) printk(KERN_ERR "ip_frag_create: no memory left !\n"));
+	LIMIT_NETDEBUG(printk(KERN_ERR "ip_frag_create: no memory left !\n"));
 	return NULL;
 }
 
@@ -625,10 +625,8 @@ static struct sk_buff *ip_frag_reasm(struct ipq *qp, struct net_device *dev)
 	return head;
 
 out_nomem:
- 	NETDEBUG(if (net_ratelimit())
-	         printk(KERN_ERR 
-			"IP: queue_glue: no memory for gluing queue %p\n",
-			qp));
+ 	LIMIT_NETDEBUG(printk(KERN_ERR "IP: queue_glue: no memory for gluing "
+			      "queue %p\n", qp));
 	goto out_fail;
 out_oversize:
 	if (net_ratelimit())
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index fc7c481..ff4bd06 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -848,6 +848,9 @@ mc_msf_out:
  
 		case IP_IPSEC_POLICY:
 		case IP_XFRM_POLICY:
+			err = -EPERM;
+			if (!capable(CAP_NET_ADMIN))
+				break;
 			err = xfrm_user_policy(sk, optname, optval, optlen);
 			break;
 
diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c
index bc59d0d..91d5ea1 100644
--- a/net/ipv4/netfilter/ip_nat_standalone.c
+++ b/net/ipv4/netfilter/ip_nat_standalone.c
@@ -102,6 +102,10 @@ ip_nat_fn(unsigned int hooknum,
 		return NF_ACCEPT;
 	}
 
+	/* Don't try to NAT if this packet is not conntracked */
+	if (ct == &ip_conntrack_untracked)
+		return NF_ACCEPT;
+
 	switch (ctinfo) {
 	case IP_CT_RELATED:
 	case IP_CT_RELATED+IP_CT_IS_REPLY:
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 62f62bb..5d91213 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1494,12 +1494,11 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
 			 * to destinations, already remembered
 			 * to the moment of synflood.
 			 */
-			NETDEBUG(if (net_ratelimit()) \
-					printk(KERN_DEBUG "TCP: drop open "
-							  "request from %u.%u."
-							  "%u.%u/%u\n", \
-					       NIPQUAD(saddr),
-					       ntohs(skb->h.th->source)));
+			LIMIT_NETDEBUG(printk(KERN_DEBUG "TCP: drop open "
+					      "request from %u.%u."
+					      "%u.%u/%u\n",
+					      NIPQUAD(saddr),
+					      ntohs(skb->h.th->source)));
 			dst_release(dst);
 			goto drop_and_free;
 		}
@@ -1627,8 +1626,7 @@ static int tcp_v4_checksum_init(struct sk_buff *skb)
 				  skb->nh.iph->daddr, skb->csum))
 			return 0;
 
-		NETDEBUG(if (net_ratelimit())
-				printk(KERN_DEBUG "hw tcp v4 csum failed\n"));
+		LIMIT_NETDEBUG(printk(KERN_DEBUG "hw tcp v4 csum failed\n"));
 		skb->ip_summed = CHECKSUM_NONE;
 	}
 	if (skb->len <= 76) {
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index e3f8ea1..7d076f0d 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -403,11 +403,9 @@ static void tcp_queue_skb(struct sock *sk, struct sk_buff *skb)
 		sk->sk_send_head = skb;
 }
 
-static void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb)
+static void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb, unsigned int mss_now)
 {
-	struct tcp_sock *tp = tcp_sk(sk);
-
-	if (skb->len <= tp->mss_cache ||
+	if (skb->len <= mss_now ||
 	    !(sk->sk_route_caps & NETIF_F_TSO)) {
 		/* Avoid the costly divide in the normal
 		 * non-TSO case.
@@ -417,10 +415,10 @@ static void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb)
 	} else {
 		unsigned int factor;
 
-		factor = skb->len + (tp->mss_cache - 1);
-		factor /= tp->mss_cache;
+		factor = skb->len + (mss_now - 1);
+		factor /= mss_now;
 		skb_shinfo(skb)->tso_segs = factor;
-		skb_shinfo(skb)->tso_size = tp->mss_cache;
+		skb_shinfo(skb)->tso_size = mss_now;
 	}
 }
 
@@ -429,7 +427,7 @@ static void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb)
  * packet to the list.  This won't be called frequently, I hope. 
  * Remember, these are still headerless SKBs at this point.
  */
-static int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len)
+static int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss_now)
 {
 	struct tcp_sock *tp = tcp_sk(sk);
 	struct sk_buff *buff;
@@ -492,8 +490,8 @@ static int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len)
 	}
 
 	/* Fix up tso_factor for both original and new SKB.  */
-	tcp_set_skb_tso_segs(sk, skb);
-	tcp_set_skb_tso_segs(sk, buff);
+	tcp_set_skb_tso_segs(sk, skb, mss_now);
+	tcp_set_skb_tso_segs(sk, buff, mss_now);
 
 	if (TCP_SKB_CB(skb)->sacked & TCPCB_LOST) {
 		tp->lost_out += tcp_skb_pcount(skb);
@@ -569,7 +567,7 @@ int tcp_trim_head(struct sock *sk, struct sk_buff *skb, u32 len)
 	 * factor and mss.
 	 */
 	if (tcp_skb_pcount(skb) > 1)
-		tcp_set_skb_tso_segs(sk, skb);
+		tcp_set_skb_tso_segs(sk, skb, tcp_current_mss(sk, 1));
 
 	return 0;
 }
@@ -734,12 +732,14 @@ static inline unsigned int tcp_cwnd_test(struct tcp_sock *tp, struct sk_buff *sk
 /* This must be invoked the first time we consider transmitting
  * SKB onto the wire.
  */
-static inline int tcp_init_tso_segs(struct sock *sk, struct sk_buff *skb)
+static inline int tcp_init_tso_segs(struct sock *sk, struct sk_buff *skb, unsigned int mss_now)
 {
 	int tso_segs = tcp_skb_pcount(skb);
 
-	if (!tso_segs) {
-		tcp_set_skb_tso_segs(sk, skb);
+	if (!tso_segs ||
+	    (tso_segs > 1 &&
+	     skb_shinfo(skb)->tso_size != mss_now)) {
+		tcp_set_skb_tso_segs(sk, skb, mss_now);
 		tso_segs = tcp_skb_pcount(skb);
 	}
 	return tso_segs;
@@ -817,7 +817,7 @@ static unsigned int tcp_snd_test(struct sock *sk, struct sk_buff *skb,
 	struct tcp_sock *tp = tcp_sk(sk);
 	unsigned int cwnd_quota;
 
-	tcp_init_tso_segs(sk, skb);
+	tcp_init_tso_segs(sk, skb, cur_mss);
 
 	if (!tcp_nagle_test(tp, skb, cur_mss, nonagle))
 		return 0;
@@ -854,7 +854,7 @@ int tcp_may_send_now(struct sock *sk, struct tcp_sock *tp)
  * know that all the data is in scatter-gather pages, and that the
  * packet has never been sent out before (and thus is not cloned).
  */
-static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len)
+static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len, unsigned int mss_now)
 {
 	struct sk_buff *buff;
 	int nlen = skb->len - len;
@@ -887,8 +887,8 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len)
 	skb_split(skb, buff, len);
 
 	/* Fix up tso_factor for both original and new SKB.  */
-	tcp_set_skb_tso_segs(sk, skb);
-	tcp_set_skb_tso_segs(sk, buff);
+	tcp_set_skb_tso_segs(sk, skb, mss_now);
+	tcp_set_skb_tso_segs(sk, buff, mss_now);
 
 	/* Link BUFF into the send queue. */
 	skb_header_release(buff);
@@ -972,19 +972,18 @@ static int tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle)
 	if (unlikely(sk->sk_state == TCP_CLOSE))
 		return 0;
 
-	skb = sk->sk_send_head;
-	if (unlikely(!skb))
-		return 0;
-
-	tso_segs = tcp_init_tso_segs(sk, skb);
-	cwnd_quota = tcp_cwnd_test(tp, skb);
-	if (unlikely(!cwnd_quota))
-		goto out;
-
 	sent_pkts = 0;
-	while (likely(tcp_snd_wnd_test(tp, skb, mss_now))) {
+	while ((skb = sk->sk_send_head)) {
+		tso_segs = tcp_init_tso_segs(sk, skb, mss_now);
 		BUG_ON(!tso_segs);
 
+		cwnd_quota = tcp_cwnd_test(tp, skb);
+		if (!cwnd_quota)
+			break;
+
+		if (unlikely(!tcp_snd_wnd_test(tp, skb, mss_now)))
+			break;
+
 		if (tso_segs == 1) {
 			if (unlikely(!tcp_nagle_test(tp, skb, mss_now,
 						     (tcp_skb_is_last(sk, skb) ?
@@ -1006,11 +1005,11 @@ static int tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle)
 					limit = skb->len - trim;
 			}
 			if (skb->len > limit) {
-				if (tso_fragment(sk, skb, limit))
+				if (tso_fragment(sk, skb, limit, mss_now))
 					break;
 			}
 		} else if (unlikely(skb->len > mss_now)) {
-			if (unlikely(tcp_fragment(sk, skb,  mss_now)))
+			if (unlikely(tcp_fragment(sk, skb,  mss_now, mss_now)))
 				break;
 		}
 
@@ -1026,27 +1025,12 @@ static int tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle)
 
 		tcp_minshall_update(tp, mss_now, skb);
 		sent_pkts++;
-
-		/* Do not optimize this to use tso_segs. If we chopped up
-		 * the packet above, tso_segs will no longer be valid.
-		 */
-		cwnd_quota -= tcp_skb_pcount(skb);
-
-		BUG_ON(cwnd_quota < 0);
-		if (!cwnd_quota)
-			break;
-
-		skb = sk->sk_send_head;
-		if (!skb)
-			break;
-		tso_segs = tcp_init_tso_segs(sk, skb);
 	}
 
 	if (likely(sent_pkts)) {
 		tcp_cwnd_validate(sk, tp);
 		return 0;
 	}
-out:
 	return !tp->packets_out && sk->sk_send_head;
 }
 
@@ -1076,7 +1060,7 @@ void tcp_push_one(struct sock *sk, unsigned int mss_now)
 
 	BUG_ON(!skb || skb->len < mss_now);
 
-	tso_segs = tcp_init_tso_segs(sk, skb);
+	tso_segs = tcp_init_tso_segs(sk, skb, mss_now);
 	cwnd_quota = tcp_snd_test(sk, skb, mss_now, TCP_NAGLE_PUSH);
 
 	if (likely(cwnd_quota)) {
@@ -1093,11 +1077,11 @@ void tcp_push_one(struct sock *sk, unsigned int mss_now)
 					limit = skb->len - trim;
 			}
 			if (skb->len > limit) {
-				if (unlikely(tso_fragment(sk, skb, limit)))
+				if (unlikely(tso_fragment(sk, skb, limit, mss_now)))
 					return;
 			}
 		} else if (unlikely(skb->len > mss_now)) {
-			if (unlikely(tcp_fragment(sk, skb, mss_now)))
+			if (unlikely(tcp_fragment(sk, skb, mss_now, mss_now)))
 				return;
 		}
 
@@ -1388,7 +1372,7 @@ int tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb)
 		int old_factor = tcp_skb_pcount(skb);
 		int new_factor;
 
-		if (tcp_fragment(sk, skb, cur_mss))
+		if (tcp_fragment(sk, skb, cur_mss, cur_mss))
 			return -ENOMEM; /* We'll try again later. */
 
 		/* New SKB created, account for it. */
@@ -1991,7 +1975,7 @@ int tcp_write_wakeup(struct sock *sk)
 			    skb->len > mss) {
 				seg_size = min(seg_size, mss);
 				TCP_SKB_CB(skb)->flags |= TCPCB_FLAG_PSH;
-				if (tcp_fragment(sk, skb, seg_size))
+				if (tcp_fragment(sk, skb, seg_size, mss))
 					return -1;
 				/* SWS override triggered forced fragmentation.
 				 * Disable TSO, the connection is too sick. */
@@ -2000,7 +1984,7 @@ int tcp_write_wakeup(struct sock *sk)
 					sk->sk_route_caps &= ~NETIF_F_TSO;
 				}
 			} else if (!tcp_skb_pcount(skb))
-				tcp_set_skb_tso_segs(sk, skb);
+				tcp_set_skb_tso_segs(sk, skb, mss);
 
 			TCP_SKB_CB(skb)->flags |= TCPCB_FLAG_PSH;
 			TCP_SKB_CB(skb)->when = tcp_time_stamp;
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 7c24e64..dc4d073 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -628,7 +628,7 @@ back_from_confirm:
 		/* ... which is an evident application bug. --ANK */
 		release_sock(sk);
 
-		NETDEBUG(if (net_ratelimit()) printk(KERN_DEBUG "udp cork app bug 2\n"));
+		LIMIT_NETDEBUG(printk(KERN_DEBUG "udp cork app bug 2\n"));
 		err = -EINVAL;
 		goto out;
 	}
@@ -693,7 +693,7 @@ static int udp_sendpage(struct sock *sk, struct page *page, int offset,
 	if (unlikely(!up->pending)) {
 		release_sock(sk);
 
-		NETDEBUG(if (net_ratelimit()) printk(KERN_DEBUG "udp cork app bug 3\n"));
+		LIMIT_NETDEBUG(printk(KERN_DEBUG "udp cork app bug 3\n"));
 		return -EINVAL;
 	}
 
@@ -1102,7 +1102,7 @@ static int udp_checksum_init(struct sk_buff *skb, struct udphdr *uh,
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 		if (!udp_check(uh, ulen, saddr, daddr, skb->csum))
 			return 0;
-		NETDEBUG(if (net_ratelimit()) printk(KERN_DEBUG "udp v4 hw csum failure.\n"));
+		LIMIT_NETDEBUG(printk(KERN_DEBUG "udp v4 hw csum failure.\n"));
 		skb->ip_summed = CHECKSUM_NONE;
 	}
 	if (skb->ip_summed != CHECKSUM_UNNECESSARY)
@@ -1181,14 +1181,13 @@ int udp_rcv(struct sk_buff *skb)
 	return(0);
 
 short_packet:
-	NETDEBUG(if (net_ratelimit())
-		printk(KERN_DEBUG "UDP: short packet: From %u.%u.%u.%u:%u %d/%d to %u.%u.%u.%u:%u\n",
-			NIPQUAD(saddr),
-			ntohs(uh->source),
-			ulen,
-			len,
-			NIPQUAD(daddr),
-			ntohs(uh->dest)));
+	LIMIT_NETDEBUG(printk(KERN_DEBUG "UDP: short packet: From %u.%u.%u.%u:%u %d/%d to %u.%u.%u.%u:%u\n",
+			      NIPQUAD(saddr),
+			      ntohs(uh->source),
+			      ulen,
+			      len,
+			      NIPQUAD(daddr),
+			      ntohs(uh->dest)));
 no_header:
 	UDP_INC_STATS_BH(UDP_MIB_INERRORS);
 	kfree_skb(skb);
@@ -1199,13 +1198,12 @@ csum_error:
 	 * RFC1122: OK.  Discards the bad packet silently (as far as 
 	 * the network is concerned, anyway) as per 4.1.3.4 (MUST). 
 	 */
-	NETDEBUG(if (net_ratelimit())
-		 printk(KERN_DEBUG "UDP: bad checksum. From %d.%d.%d.%d:%d to %d.%d.%d.%d:%d ulen %d\n",
-			NIPQUAD(saddr),
-			ntohs(uh->source),
-			NIPQUAD(daddr),
-			ntohs(uh->dest),
-			ulen));
+	LIMIT_NETDEBUG(printk(KERN_DEBUG "UDP: bad checksum. From %d.%d.%d.%d:%d to %d.%d.%d.%d:%d ulen %d\n",
+			      NIPQUAD(saddr),
+			      ntohs(uh->source),
+			      NIPQUAD(daddr),
+			      ntohs(uh->dest),
+			      ulen));
 drop:
 	UDP_INC_STATS_BH(UDP_MIB_INERRORS);
 	kfree_skb(skb);