From 7489aec8eed4f2f1eb3b4d35763bd3ea30b32ef5 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 31 May 2010 16:41:35 +0200
Subject: netfilter: xtables: stackptr should be percpu

commit f3c5c1bfd4 (netfilter: xtables: make ip_tables reentrant)
introduced a performance regression, because stackptr array is shared by
all cpus, adding cache line ping pongs. (16 cpus share a 64 bytes cache
line)

Fix this using alloc_percpu()

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-By: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 net/ipv6/netfilter/ip6_tables.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index 6f517bd83692..9d2d68f0e605 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -363,7 +363,7 @@ ip6t_do_table(struct sk_buff *skb,
 	cpu        = smp_processor_id();
 	table_base = private->entries[cpu];
 	jumpstack  = (struct ip6t_entry **)private->jumpstack[cpu];
-	stackptr   = &private->stackptr[cpu];
+	stackptr   = per_cpu_ptr(private->stackptr, cpu);
 	origptr    = *stackptr;
 
 	e = get_entry(table_base, private->hook_entry[hook]);
-- 
cgit v1.2.3


From b1faf5666438090a4dc4fceac8502edc7788b7e3 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 31 May 2010 23:44:05 -0700
Subject: net: sock_queue_err_skb() dont mess with sk_forward_alloc

Correct sk_forward_alloc handling for error_queue would need to use a
backlog of frames that softirq handler could not deliver because socket
is owned by user thread. Or extend backlog processing to be able to
process normal and error packets.

Another possibility is to not use mem charge for error queue, this is
what I implemented in this patch.

Note: this reverts commit 29030374
(net: fix sk_forward_alloc corruptions), since we dont need to lock
socket anymore.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/udp.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 3048f906c042..87be58673b55 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -466,11 +466,9 @@ void __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	if (sk->sk_state != TCP_ESTABLISHED && !np->recverr)
 		goto out;
 
-	if (np->recverr) {
-		bh_lock_sock(sk);
+	if (np->recverr)
 		ipv6_icmp_error(sk, skb, err, uh->dest, ntohl(info), (u8 *)(uh+1));
-		bh_unlock_sock(sk);
-	}
+
 	sk->sk_err = err;
 	sk->sk_error_report(sk);
 out:
-- 
cgit v1.2.3


From 5d55354f147bcf82b7a330bf9617b82a692b7d49 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Mon, 31 May 2010 17:23:22 +0000
Subject: net/ipv6/mcast.c: Remove unnecessary kmalloc casts

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/mcast.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index 59f1881968c7..7b5fb43c227c 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -1995,8 +1995,7 @@ static int sf_setstate(struct ifmcaddr6 *pmc)
 				    &psf->sf_addr))
 					break;
 			if (!dpsf) {
-				dpsf = (struct ip6_sf_list *)
-					kmalloc(sizeof(*dpsf), GFP_ATOMIC);
+				dpsf = kmalloc(sizeof(*dpsf), GFP_ATOMIC);
 				if (!dpsf)
 					continue;
 				*dpsf = *psf;
-- 
cgit v1.2.3


From aac4dddc358acfd9d98b20024a42c34dfab31c39 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Tue, 1 Jun 2010 00:26:58 -0700
Subject: ipv6: get rid of ipip6_prl_lock

As noticed by Julia Lawall, ipip6_tunnel_add_prl() incorrectly calls
kzallloc(..., GFP_KERNEL) while a spinlock is held. She provided
a patch to use GFP_ATOMIC instead.

One possibility would be to convert this spinlock to a mutex, or
preallocate the thing before taking the lock.

After RCU conversion, it appears we dont need this lock, since
caller already holds RTNL

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/sit.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index e51e650ea80b..702c532ec21e 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -249,8 +249,6 @@ failed:
 	return NULL;
 }
 
-static DEFINE_SPINLOCK(ipip6_prl_lock);
-
 #define for_each_prl_rcu(start)			\
 	for (prl = rcu_dereference(start);	\
 	     prl;				\
@@ -340,7 +338,7 @@ ipip6_tunnel_add_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a, int chg)
 	if (a->addr == htonl(INADDR_ANY))
 		return -EINVAL;
 
-	spin_lock(&ipip6_prl_lock);
+	ASSERT_RTNL();
 
 	for (p = t->prl; p; p = p->next) {
 		if (p->addr == a->addr) {
@@ -370,7 +368,6 @@ ipip6_tunnel_add_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a, int chg)
 	t->prl_count++;
 	rcu_assign_pointer(t->prl, p);
 out:
-	spin_unlock(&ipip6_prl_lock);
 	return err;
 }
 
@@ -397,7 +394,7 @@ ipip6_tunnel_del_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a)
 	struct ip_tunnel_prl_entry *x, **p;
 	int err = 0;
 
-	spin_lock(&ipip6_prl_lock);
+	ASSERT_RTNL();
 
 	if (a && a->addr != htonl(INADDR_ANY)) {
 		for (p = &t->prl; *p; p = &(*p)->next) {
@@ -419,7 +416,6 @@ ipip6_tunnel_del_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a)
 		}
 	}
 out:
-	spin_unlock(&ipip6_prl_lock);
 	return err;
 }
 
-- 
cgit v1.2.3


From c2d9ba9bce8d7323ca96f239e1f505c14d6244fb Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Tue, 1 Jun 2010 06:51:19 +0000
Subject: net: CONFIG_NET_NS reduction

Use read_pnet() and write_pnet() to reduce number of ifdef CONFIG_NET_NS

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/addrlabel.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/addrlabel.c b/net/ipv6/addrlabel.c
index 8c4348cb1950..f0e774cea386 100644
--- a/net/ipv6/addrlabel.c
+++ b/net/ipv6/addrlabel.c
@@ -53,11 +53,7 @@ static struct ip6addrlbl_table
 static inline
 struct net *ip6addrlbl_net(const struct ip6addrlbl_entry *lbl)
 {
-#ifdef CONFIG_NET_NS
-	return lbl->lbl_net;
-#else
-	return &init_net;
-#endif
+	return read_pnet(&lbl->lbl_net);
 }
 
 /*
-- 
cgit v1.2.3


From 20c59de2e6b6bc74bbf714dcd4e720afe8d516cf Mon Sep 17 00:00:00 2001
From: Arnaud Ebalard <arno@natisbad.org>
Date: Tue, 1 Jun 2010 21:35:01 +0000
Subject: ipv6: Refactor update of IPv6 flowi destination address for srcrt
 (RH) option

There are more than a dozen occurrences of following code in the
IPv6 stack:

    if (opt && opt->srcrt) {
            struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt;
            ipv6_addr_copy(&final, &fl.fl6_dst);
            ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
            final_p = &final;
    }

Replace those with a helper. Note that the helper overrides final_p
in all cases. This is ok as final_p was previously initialized to
NULL when declared.

Signed-off-by: Arnaud Ebalard <arno@natisbad.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/af_inet6.c              |  9 ++-------
 net/ipv6/datagram.c              | 18 ++++--------------
 net/ipv6/exthdrs.c               | 24 ++++++++++++++++++++++++
 net/ipv6/inet6_connection_sock.c |  9 ++-------
 net/ipv6/raw.c                   | 10 ++--------
 net/ipv6/syncookies.c            |  9 ++-------
 net/ipv6/tcp_ipv6.c              | 27 ++++++---------------------
 net/ipv6/udp.c                   | 11 +++--------
 8 files changed, 45 insertions(+), 72 deletions(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index e733942dafe1..94b1b9c954bf 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -651,7 +651,7 @@ int inet6_sk_rebuild_header(struct sock *sk)
 
 	if (dst == NULL) {
 		struct inet_sock *inet = inet_sk(sk);
-		struct in6_addr *final_p = NULL, final;
+		struct in6_addr *final_p, final;
 		struct flowi fl;
 
 		memset(&fl, 0, sizeof(fl));
@@ -665,12 +665,7 @@ int inet6_sk_rebuild_header(struct sock *sk)
 		fl.fl_ip_sport = inet->inet_sport;
 		security_sk_classify_flow(sk, &fl);
 
-		if (np->opt && np->opt->srcrt) {
-			struct rt0_hdr *rt0 = (struct rt0_hdr *) np->opt->srcrt;
-			ipv6_addr_copy(&final, &fl.fl6_dst);
-			ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
-			final_p = &final;
-		}
+		final_p = fl6_update_dst(&fl, np->opt, &final);
 
 		err = ip6_dst_lookup(sk, &dst, &fl);
 		if (err) {
diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c
index 712684687c9a..7d929a22cbc2 100644
--- a/net/ipv6/datagram.c
+++ b/net/ipv6/datagram.c
@@ -38,10 +38,11 @@ int ip6_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 	struct sockaddr_in6	*usin = (struct sockaddr_in6 *) uaddr;
 	struct inet_sock      	*inet = inet_sk(sk);
 	struct ipv6_pinfo      	*np = inet6_sk(sk);
-	struct in6_addr		*daddr, *final_p = NULL, final;
+	struct in6_addr		*daddr, *final_p, final;
 	struct dst_entry	*dst;
 	struct flowi		fl;
 	struct ip6_flowlabel	*flowlabel = NULL;
+	struct ipv6_txoptions   *opt;
 	int			addr_type;
 	int			err;
 
@@ -155,19 +156,8 @@ ipv4_connected:
 
 	security_sk_classify_flow(sk, &fl);
 
-	if (flowlabel) {
-		if (flowlabel->opt && flowlabel->opt->srcrt) {
-			struct rt0_hdr *rt0 = (struct rt0_hdr *) flowlabel->opt->srcrt;
-			ipv6_addr_copy(&final, &fl.fl6_dst);
-			ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
-			final_p = &final;
-		}
-	} else if (np->opt && np->opt->srcrt) {
-		struct rt0_hdr *rt0 = (struct rt0_hdr *)np->opt->srcrt;
-		ipv6_addr_copy(&final, &fl.fl6_dst);
-		ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
-		final_p = &final;
-	}
+	opt = flowlabel ? flowlabel->opt : np->opt;
+	final_p = fl6_update_dst(&fl, opt, &final);
 
 	err = ip6_dst_lookup(sk, &dst, &fl);
 	if (err)
diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index 8a659f92d17a..853a633a94d4 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -874,3 +874,27 @@ struct ipv6_txoptions *ipv6_fixup_options(struct ipv6_txoptions *opt_space,
 	return opt;
 }
 
+/**
+ * fl6_update_dst - update flowi destination address with info given
+ *                  by srcrt option, if any.
+ *
+ * @fl: flowi for which fl6_dst is to be updated
+ * @opt: struct ipv6_txoptions in which to look for srcrt opt
+ * @orig: copy of original fl6_dst address if modified
+ *
+ * Returns NULL if no txoptions or no srcrt, otherwise returns orig
+ * and initial value of fl->fl6_dst set in orig
+ */
+struct in6_addr *fl6_update_dst(struct flowi *fl,
+				const struct ipv6_txoptions *opt,
+				struct in6_addr *orig)
+{
+	if (!opt || !opt->srcrt)
+		return NULL;
+
+	ipv6_addr_copy(orig, &fl->fl6_dst);
+	ipv6_addr_copy(&fl->fl6_dst, ((struct rt0_hdr *)opt->srcrt)->addr);
+	return orig;
+}
+
+EXPORT_SYMBOL_GPL(fl6_update_dst);
diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c
index 0c5e3c3b7fd5..8a1628023bd1 100644
--- a/net/ipv6/inet6_connection_sock.c
+++ b/net/ipv6/inet6_connection_sock.c
@@ -185,7 +185,7 @@ int inet6_csk_xmit(struct sk_buff *skb)
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct flowi fl;
 	struct dst_entry *dst;
-	struct in6_addr *final_p = NULL, final;
+	struct in6_addr *final_p, final;
 
 	memset(&fl, 0, sizeof(fl));
 	fl.proto = sk->sk_protocol;
@@ -199,12 +199,7 @@ int inet6_csk_xmit(struct sk_buff *skb)
 	fl.fl_ip_dport = inet->inet_dport;
 	security_sk_classify_flow(sk, &fl);
 
-	if (np->opt && np->opt->srcrt) {
-		struct rt0_hdr *rt0 = (struct rt0_hdr *)np->opt->srcrt;
-		ipv6_addr_copy(&final, &fl.fl6_dst);
-		ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
-		final_p = &final;
-	}
+	final_p = fl6_update_dst(&fl, np->opt, &final);
 
 	dst = __inet6_csk_dst_check(sk, np->dst_cookie);
 
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 4a4dcbe4f8b2..864eb8e03b1b 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -725,7 +725,7 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk,
 {
 	struct ipv6_txoptions opt_space;
 	struct sockaddr_in6 * sin6 = (struct sockaddr_in6 *) msg->msg_name;
-	struct in6_addr *daddr, *final_p = NULL, final;
+	struct in6_addr *daddr, *final_p, final;
 	struct inet_sock *inet = inet_sk(sk);
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct raw6_sock *rp = raw6_sk(sk);
@@ -847,13 +847,7 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk,
 	if (ipv6_addr_any(&fl.fl6_src) && !ipv6_addr_any(&np->saddr))
 		ipv6_addr_copy(&fl.fl6_src, &np->saddr);
 
-	/* merge ip6_build_xmit from ip6_output */
-	if (opt && opt->srcrt) {
-		struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt;
-		ipv6_addr_copy(&final, &fl.fl6_dst);
-		ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
-		final_p = &final;
-	}
+	final_p = fl6_update_dst(&fl, opt, &final);
 
 	if (!fl.oif && ipv6_addr_is_multicast(&fl.fl6_dst))
 		fl.oif = np->mcast_oif;
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index 34d1f0690d7e..12383705dbad 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -240,17 +240,12 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	 * me if there is a preferred way.
 	 */
 	{
-		struct in6_addr *final_p = NULL, final;
+		struct in6_addr *final_p, final;
 		struct flowi fl;
 		memset(&fl, 0, sizeof(fl));
 		fl.proto = IPPROTO_TCP;
 		ipv6_addr_copy(&fl.fl6_dst, &ireq6->rmt_addr);
-		if (np->opt && np->opt->srcrt) {
-			struct rt0_hdr *rt0 = (struct rt0_hdr *) np->opt->srcrt;
-			ipv6_addr_copy(&final, &fl.fl6_dst);
-			ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
-			final_p = &final;
-		}
+		final_p = fl6_update_dst(&fl, np->opt, &final);
 		ipv6_addr_copy(&fl.fl6_src, &ireq6->loc_addr);
 		fl.oif = sk->sk_bound_dev_if;
 		fl.mark = sk->sk_mark;
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 2b7c3a100e2c..e487080d02db 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -129,7 +129,7 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 	struct inet_connection_sock *icsk = inet_csk(sk);
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct tcp_sock *tp = tcp_sk(sk);
-	struct in6_addr *saddr = NULL, *final_p = NULL, final;
+	struct in6_addr *saddr = NULL, *final_p, final;
 	struct flowi fl;
 	struct dst_entry *dst;
 	int addr_type;
@@ -250,12 +250,7 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 	fl.fl_ip_dport = usin->sin6_port;
 	fl.fl_ip_sport = inet->inet_sport;
 
-	if (np->opt && np->opt->srcrt) {
-		struct rt0_hdr *rt0 = (struct rt0_hdr *)np->opt->srcrt;
-		ipv6_addr_copy(&final, &fl.fl6_dst);
-		ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
-		final_p = &final;
-	}
+	final_p = fl6_update_dst(&fl, np->opt, &final);
 
 	security_sk_classify_flow(sk, &fl);
 
@@ -477,7 +472,7 @@ static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req,
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct sk_buff * skb;
 	struct ipv6_txoptions *opt = NULL;
-	struct in6_addr * final_p = NULL, final;
+	struct in6_addr * final_p, final;
 	struct flowi fl;
 	struct dst_entry *dst;
 	int err = -1;
@@ -494,12 +489,7 @@ static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req,
 	security_req_classify_flow(req, &fl);
 
 	opt = np->opt;
-	if (opt && opt->srcrt) {
-		struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt;
-		ipv6_addr_copy(&final, &fl.fl6_dst);
-		ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
-		final_p = &final;
-	}
+	final_p = fl6_update_dst(&fl, opt, &final);
 
 	err = ip6_dst_lookup(sk, &dst, &fl);
 	if (err)
@@ -1392,18 +1382,13 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb,
 		goto out_overflow;
 
 	if (dst == NULL) {
-		struct in6_addr *final_p = NULL, final;
+		struct in6_addr *final_p, final;
 		struct flowi fl;
 
 		memset(&fl, 0, sizeof(fl));
 		fl.proto = IPPROTO_TCP;
 		ipv6_addr_copy(&fl.fl6_dst, &treq->rmt_addr);
-		if (opt && opt->srcrt) {
-			struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt;
-			ipv6_addr_copy(&final, &fl.fl6_dst);
-			ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
-			final_p = &final;
-		}
+		final_p = fl6_update_dst(&fl, opt, &final);
 		ipv6_addr_copy(&fl.fl6_src, &treq->loc_addr);
 		fl.oif = sk->sk_bound_dev_if;
 		fl.mark = sk->sk_mark;
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 3048f906c042..4aea57dec75f 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -929,7 +929,7 @@ int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk,
 	struct inet_sock *inet = inet_sk(sk);
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *) msg->msg_name;
-	struct in6_addr *daddr, *final_p = NULL, final;
+	struct in6_addr *daddr, *final_p, final;
 	struct ipv6_txoptions *opt = NULL;
 	struct ip6_flowlabel *flowlabel = NULL;
 	struct flowi fl;
@@ -1099,14 +1099,9 @@ do_udp_sendmsg:
 		ipv6_addr_copy(&fl.fl6_src, &np->saddr);
 	fl.fl_ip_sport = inet->inet_sport;
 
-	/* merge ip6_build_xmit from ip6_output */
-	if (opt && opt->srcrt) {
-		struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt;
-		ipv6_addr_copy(&final, &fl.fl6_dst);
-		ipv6_addr_copy(&fl.fl6_dst, rt0->addr);
-		final_p = &final;
+	final_p = fl6_update_dst(&fl, opt, &final);
+	if (final_p)
 		connected = 0;
-	}
 
 	if (!fl.oif && ipv6_addr_is_multicast(&fl.fl6_dst)) {
 		fl.oif = np->mcast_oif;
-- 
cgit v1.2.3


From af9b4738574b46025de7ccbe75c7b24fd8914379 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 3 Jun 2010 00:43:44 +0000
Subject: syncookies: avoid unneeded tcp header flag double check

caller: if (!th->rst && !th->syn && th->ack)
callee: if (!th->ack)

make the caller only check for !syn (common for 3whs), and move
the !rst / ack test to the callee.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/syncookies.c | 2 +-
 net/ipv6/tcp_ipv6.c   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index 12383705dbad..9fcb3ec7c838 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -174,7 +174,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 	struct dst_entry *dst;
 	__u8 rcv_wscale;
 
-	if (!sysctl_tcp_syncookies || !th->ack)
+	if (!sysctl_tcp_syncookies || !th->ack || th->rst)
 		goto out;
 
 	if (tcp_synq_no_recent_overflow(sk) ||
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index e487080d02db..5887141ad641 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1157,7 +1157,7 @@ static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb)
 	}
 
 #ifdef CONFIG_SYN_COOKIES
-	if (!th->rst && !th->syn && th->ack)
+	if (!th->syn)
 		sk = cookie_v6_check(sk, skb);
 #endif
 	return sk;
-- 
cgit v1.2.3


From 5918e2fb9035b35164002c3a268feaf70b7c04d5 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 3 Jun 2010 00:43:57 +0000
Subject: syncookies: update mss tables

- ipv6 msstab: account for ipv6 header size
- ipv4 msstab: add mss for Jumbograms.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/syncookies.c | 39 +++++++++++++++------------------------
 1 file changed, 15 insertions(+), 24 deletions(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index 9fcb3ec7c838..70d330f8c990 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -27,28 +27,17 @@ extern __u32 syncookie_secret[2][16-4+SHA_DIGEST_WORDS];
 #define COOKIEBITS 24	/* Upper bits store count */
 #define COOKIEMASK (((__u32)1 << COOKIEBITS) - 1)
 
-/*
- * This table has to be sorted and terminated with (__u16)-1.
- * XXX generate a better table.
- * Unresolved Issues: HIPPI with a 64k MSS is not well supported.
- *
- * Taken directly from ipv4 implementation.
- * Should this list be modified for ipv6 use or is it close enough?
- * rfc 2460 8.3 suggests mss values 20 bytes less than ipv4 counterpart
- */
+/* Table must be sorted. */
 static __u16 const msstab[] = {
-	64 - 1,
-	256 - 1,
-	512 - 1,
-	536 - 1,
-	1024 - 1,
-	1440 - 1,
-	1460 - 1,
-	4312 - 1,
-	(__u16)-1
+	64,
+	512,
+	536,
+	1280 - 60,
+	1480 - 60,
+	1500 - 60,
+	4460 - 60,
+	9000 - 60,
 };
-/* The number doesn't include the -1 terminator */
-#define NUM_MSS (ARRAY_SIZE(msstab) - 1)
 
 /*
  * This (misnamed) value is the age of syncookie which is permitted.
@@ -134,9 +123,11 @@ __u32 cookie_v6_init_sequence(struct sock *sk, struct sk_buff *skb, __u16 *mssp)
 
 	tcp_synq_overflow(sk);
 
-	for (mssind = 0; mss > msstab[mssind + 1]; mssind++)
-		;
-	*mssp = msstab[mssind] + 1;
+	for (mssind = ARRAY_SIZE(msstab) - 1; mssind ; mssind--)
+		if (mss >= msstab[mssind])
+			break;
+
+	*mssp = msstab[mssind];
 
 	NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_SYNCOOKIESSENT);
 
@@ -154,7 +145,7 @@ static inline int cookie_check(struct sk_buff *skb, __u32 cookie)
 					    th->source, th->dest, seq,
 					    jiffies / (HZ * 60), COUNTER_TRIES);
 
-	return mssind < NUM_MSS ? msstab[mssind] + 1 : 0;
+	return mssind < ARRAY_SIZE(msstab) ? msstab[mssind] : 0;
 }
 
 struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
-- 
cgit v1.2.3


From 72e09ad107e78d69ff4d3b97a69f0aad2b77280f Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Sat, 5 Jun 2010 03:03:30 -0700
Subject: ipv6: avoid high order allocations

With mtu=9000, mld_newpack() use order-2 GFP_ATOMIC allocations, that
are very unreliable, on machines where PAGE_SIZE=4K

Limit allocated skbs to be at most one page. (order-0 allocations)

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/mcast.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index 59f1881968c7..ab1622d7d409 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -1356,7 +1356,10 @@ static struct sk_buff *mld_newpack(struct net_device *dev, int size)
 		     IPV6_TLV_PADN, 0 };
 
 	/* we assume size > sizeof(ra) here */
-	skb = sock_alloc_send_skb(sk, size + LL_ALLOCATED_SPACE(dev), 1, &err);
+	size += LL_ALLOCATED_SPACE(dev);
+	/* limit our allocations to order-0 page */
+	size = min_t(int, size, SKB_MAX_ORDER(0, 0));
+	skb = sock_alloc_send_skb(sk, size, 1, &err);
 
 	if (!skb)
 		return NULL;
-- 
cgit v1.2.3


From 8ffb335e8d696affc04f963bf73ce2196f80edb9 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Sun, 6 Jun 2010 15:34:40 -0700
Subject: ip6mr: fix a typo in ip6mr_for_each_table()

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/ip6mr.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c
index 073071f2b75b..89c0b077c7aa 100644
--- a/net/ipv6/ip6mr.c
+++ b/net/ipv6/ip6mr.c
@@ -120,7 +120,7 @@ static void mroute_clean_tables(struct mr6_table *mrt);
 static void ipmr_expire_process(unsigned long arg);
 
 #ifdef CONFIG_IPV6_MROUTE_MULTIPLE_TABLES
-#define ip6mr_for_each_table(mrt, met) \
+#define ip6mr_for_each_table(mrt, net) \
 	list_for_each_entry_rcu(mrt, &net->ipv6.mr6_tables, list)
 
 static struct mr6_table *ip6mr_get_table(struct net *net, u32 id)
-- 
cgit v1.2.3


From 1789a640f55658d9a54c1868cc3405e4d85dbd8e Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Thu, 3 Jun 2010 22:23:57 +0000
Subject: raw: avoid two atomics in xmit

Avoid two atomic ops per raw_send_hdrinc() call

Avoid two atomic ops per raw6_send_hdrinc() call

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/raw.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 864eb8e03b1b..968b96490729 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -602,13 +602,14 @@ out:
 }
 
 static int rawv6_send_hdrinc(struct sock *sk, void *from, int length,
-			struct flowi *fl, struct rt6_info *rt,
+			struct flowi *fl, struct dst_entry **dstp,
 			unsigned int flags)
 {
 	struct ipv6_pinfo *np = inet6_sk(sk);
 	struct ipv6hdr *iph;
 	struct sk_buff *skb;
 	int err;
+	struct rt6_info *rt = (struct rt6_info *)*dstp;
 
 	if (length > rt->u.dst.dev->mtu) {
 		ipv6_local_error(sk, EMSGSIZE, fl, rt->u.dst.dev->mtu);
@@ -626,7 +627,8 @@ static int rawv6_send_hdrinc(struct sock *sk, void *from, int length,
 
 	skb->priority = sk->sk_priority;
 	skb->mark = sk->sk_mark;
-	skb_dst_set(skb, dst_clone(&rt->u.dst));
+	skb_dst_set(skb, &rt->u.dst);
+	*dstp = NULL;
 
 	skb_put(skb, length);
 	skb_reset_network_header(skb);
@@ -886,9 +888,9 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk,
 		goto do_confirm;
 
 back_from_confirm:
-	if (inet->hdrincl) {
-		err = rawv6_send_hdrinc(sk, msg->msg_iov, len, &fl, (struct rt6_info*)dst, msg->msg_flags);
-	} else {
+	if (inet->hdrincl)
+		err = rawv6_send_hdrinc(sk, msg->msg_iov, len, &fl, &dst, msg->msg_flags);
+	else {
 		lock_sock(sk);
 		err = ip6_append_data(sk, ip_generic_getfrag, msg->msg_iov,
 			len, 0, hlimit, tclass, opt, &fl, (struct rt6_info*)dst,
-- 
cgit v1.2.3


From 035320d54758e21227987e3aae0d46e7a04f4ddc Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Sun, 6 Jun 2010 23:48:40 +0000
Subject: ipmr: dont corrupt lists

ipmr_rules_exit() and ip6mr_rules_exit() free a list of items, but
forget to properly remove these items from list. List head is not
changed and still points to freed memory.

This can trigger a fault later when icmpv6_sk_exit() is called.

Fix is to either reinit list, or use list_del() to properly remove items
from list before freeing them.

bugzilla report : https://bugzilla.kernel.org/show_bug.cgi?id=16120

Introduced by commit d1db275dd3f6e4 (ipv6: ip6mr: support multiple
tables) and commit f0ad0860d01e (ipv4: ipmr: support multiple tables)

Reported-by: Alex Zhavnerchik <alex.vizor@gmail.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/ip6mr.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c
index 89c0b077c7aa..66078dad7fe8 100644
--- a/net/ipv6/ip6mr.c
+++ b/net/ipv6/ip6mr.c
@@ -254,8 +254,10 @@ static void __net_exit ip6mr_rules_exit(struct net *net)
 {
 	struct mr6_table *mrt, *next;
 
-	list_for_each_entry_safe(mrt, next, &net->ipv6.mr6_tables, list)
+	list_for_each_entry_safe(mrt, next, &net->ipv6.mr6_tables, list) {
+		list_del(&mrt->list);
 		ip6mr_free_table(mrt);
+	}
 	fib_rules_unregister(net->ipv6.mr6_rules_ops);
 }
 #else
-- 
cgit v1.2.3


From bb69ae049fcc986fcd742eb90ca0d44a7a49c9f1 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 7 Jun 2010 11:42:13 +0000
Subject: anycast: Some RCU conversions

- dev_get_by_flags() changed to dev_get_by_flags_rcu()

- ipv6_sock_ac_join() dont touch dev & idev refcounts
- ipv6_sock_ac_drop() dont touch dev & idev refcounts
- ipv6_sock_ac_close() dont touch dev & idev refcounts
- ipv6_dev_ac_dec() dount touch idev refcount
- ipv6_chk_acast_addr() dont touch idev refcount

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/anycast.c | 90 +++++++++++++++++++++++++-----------------------------
 1 file changed, 42 insertions(+), 48 deletions(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c
index b5b07054508a..f058fbd808c8 100644
--- a/net/ipv6/anycast.c
+++ b/net/ipv6/anycast.c
@@ -77,41 +77,40 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr)
 	pac->acl_next = NULL;
 	ipv6_addr_copy(&pac->acl_addr, addr);
 
+	rcu_read_lock();
 	if (ifindex == 0) {
 		struct rt6_info *rt;
 
 		rt = rt6_lookup(net, addr, NULL, 0, 0);
 		if (rt) {
 			dev = rt->rt6i_dev;
-			dev_hold(dev);
 			dst_release(&rt->u.dst);
 		} else if (ishost) {
 			err = -EADDRNOTAVAIL;
-			goto out_free_pac;
+			goto error;
 		} else {
 			/* router, no matching interface: just pick one */
-
-			dev = dev_get_by_flags(net, IFF_UP, IFF_UP|IFF_LOOPBACK);
+			dev = dev_get_by_flags_rcu(net, IFF_UP,
+						   IFF_UP | IFF_LOOPBACK);
 		}
 	} else
-		dev = dev_get_by_index(net, ifindex);
+		dev = dev_get_by_index_rcu(net, ifindex);
 
 	if (dev == NULL) {
 		err = -ENODEV;
-		goto out_free_pac;
+		goto error;
 	}
 
-	idev = in6_dev_get(dev);
+	idev = __in6_dev_get(dev);
 	if (!idev) {
 		if (ifindex)
 			err = -ENODEV;
 		else
 			err = -EADDRNOTAVAIL;
-		goto out_dev_put;
+		goto error;
 	}
 	/* reset ishost, now that we have a specific device */
 	ishost = !idev->cnf.forwarding;
-	in6_dev_put(idev);
 
 	pac->acl_ifindex = dev->ifindex;
 
@@ -124,26 +123,22 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr)
 		if (ishost)
 			err = -EADDRNOTAVAIL;
 		if (err)
-			goto out_dev_put;
+			goto error;
 	}
 
 	err = ipv6_dev_ac_inc(dev, addr);
-	if (err)
-		goto out_dev_put;
-
-	write_lock_bh(&ipv6_sk_ac_lock);
-	pac->acl_next = np->ipv6_ac_list;
-	np->ipv6_ac_list = pac;
-	write_unlock_bh(&ipv6_sk_ac_lock);
-
-	dev_put(dev);
-
-	return 0;
+	if (!err) {
+		write_lock_bh(&ipv6_sk_ac_lock);
+		pac->acl_next = np->ipv6_ac_list;
+		np->ipv6_ac_list = pac;
+		write_unlock_bh(&ipv6_sk_ac_lock);
+		pac = NULL;
+	}
 
-out_dev_put:
-	dev_put(dev);
-out_free_pac:
-	sock_kfree_s(sk, pac, sizeof(*pac));
+error:
+	rcu_read_unlock();
+	if (pac)
+		sock_kfree_s(sk, pac, sizeof(*pac));
 	return err;
 }
 
@@ -176,11 +171,12 @@ int ipv6_sock_ac_drop(struct sock *sk, int ifindex, struct in6_addr *addr)
 
 	write_unlock_bh(&ipv6_sk_ac_lock);
 
-	dev = dev_get_by_index(net, pac->acl_ifindex);
-	if (dev) {
+	rcu_read_lock();
+	dev = dev_get_by_index_rcu(net, pac->acl_ifindex);
+	if (dev)
 		ipv6_dev_ac_dec(dev, &pac->acl_addr);
-		dev_put(dev);
-	}
+	rcu_read_unlock();
+
 	sock_kfree_s(sk, pac, sizeof(*pac));
 	return 0;
 }
@@ -199,13 +195,12 @@ void ipv6_sock_ac_close(struct sock *sk)
 	write_unlock_bh(&ipv6_sk_ac_lock);
 
 	prev_index = 0;
+	rcu_read_lock();
 	while (pac) {
 		struct ipv6_ac_socklist *next = pac->acl_next;
 
 		if (pac->acl_ifindex != prev_index) {
-			if (dev)
-				dev_put(dev);
-			dev = dev_get_by_index(net, pac->acl_ifindex);
+			dev = dev_get_by_index_rcu(net, pac->acl_ifindex);
 			prev_index = pac->acl_ifindex;
 		}
 		if (dev)
@@ -213,8 +208,7 @@ void ipv6_sock_ac_close(struct sock *sk)
 		sock_kfree_s(sk, pac, sizeof(*pac));
 		pac = next;
 	}
-	if (dev)
-		dev_put(dev);
+	rcu_read_unlock();
 }
 
 #if 0
@@ -363,33 +357,32 @@ int __ipv6_dev_ac_dec(struct inet6_dev *idev, struct in6_addr *addr)
 	return 0;
 }
 
+/* called with rcu_read_lock() */
 static int ipv6_dev_ac_dec(struct net_device *dev, struct in6_addr *addr)
 {
-	int ret;
-	struct inet6_dev *idev = in6_dev_get(dev);
+	struct inet6_dev *idev = __in6_dev_get(dev);
+
 	if (idev == NULL)
 		return -ENODEV;
-	ret = __ipv6_dev_ac_dec(idev, addr);
-	in6_dev_put(idev);
-	return ret;
+	return __ipv6_dev_ac_dec(idev, addr);
 }
 
 /*
  *	check if the interface has this anycast address
+ *	called with rcu_read_lock()
  */
 static int ipv6_chk_acast_dev(struct net_device *dev, struct in6_addr *addr)
 {
 	struct inet6_dev *idev;
 	struct ifacaddr6 *aca;
 
-	idev = in6_dev_get(dev);
+	idev = __in6_dev_get(dev);
 	if (idev) {
 		read_lock_bh(&idev->lock);
 		for (aca = idev->ac_list; aca; aca = aca->aca_next)
 			if (ipv6_addr_equal(&aca->aca_addr, addr))
 				break;
 		read_unlock_bh(&idev->lock);
-		in6_dev_put(idev);
 		return aca != NULL;
 	}
 	return 0;
@@ -403,14 +396,15 @@ int ipv6_chk_acast_addr(struct net *net, struct net_device *dev,
 {
 	int found = 0;
 
-	if (dev)
-		return ipv6_chk_acast_dev(dev, addr);
 	rcu_read_lock();
-	for_each_netdev_rcu(net, dev)
-		if (ipv6_chk_acast_dev(dev, addr)) {
-			found = 1;
-			break;
-		}
+	if (dev)
+		found = ipv6_chk_acast_dev(dev, addr);
+	else
+		for_each_netdev_rcu(net, dev)
+			if (ipv6_chk_acast_dev(dev, addr)) {
+				found = 1;
+				break;
+			}
 	rcu_read_unlock();
 	return found;
 }
-- 
cgit v1.2.3


From 96b52e61be1ad4d4f8de39b9deaf253da804ea3b Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 7 Jun 2010 21:05:02 +0000
Subject: ipv6: mcast: RCU conversions

- ipv6_sock_mc_join() : doesnt touch dev refcount

- ipv6_sock_mc_drop() : doesnt touch dev/idev refcounts

- ip6_mc_find_dev() becomes ip6_mc_find_dev_rcu() (called from rcu),
                    and doesnt touch dev/idev refcounts

- ipv6_sock_mc_close() : doesnt touch dev/idev refcounts

- ip6_mc_source() uses ip6_mc_find_dev_rcu()

- ip6_mc_msfilter() uses ip6_mc_find_dev_rcu()

- ip6_mc_msfget() uses ip6_mc_find_dev_rcu()

- ipv6_dev_mc_dec(), ipv6_chk_mcast_addr(),
  igmp6_event_query(), igmp6_event_report(),
  mld_sendpack(), igmp6_send() dont touch idev refcount

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/mcast.c | 183 ++++++++++++++++++++++++++-----------------------------
 1 file changed, 87 insertions(+), 96 deletions(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index 8752e8084806..3e36d1538b6e 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -152,18 +152,19 @@ int ipv6_sock_mc_join(struct sock *sk, int ifindex, const struct in6_addr *addr)
 	mc_lst->next = NULL;
 	ipv6_addr_copy(&mc_lst->addr, addr);
 
+	rcu_read_lock();
 	if (ifindex == 0) {
 		struct rt6_info *rt;
 		rt = rt6_lookup(net, addr, NULL, 0, 0);
 		if (rt) {
 			dev = rt->rt6i_dev;
-			dev_hold(dev);
 			dst_release(&rt->u.dst);
 		}
 	} else
-		dev = dev_get_by_index(net, ifindex);
+		dev = dev_get_by_index_rcu(net, ifindex);
 
 	if (dev == NULL) {
+		rcu_read_unlock();
 		sock_kfree_s(sk, mc_lst, sizeof(*mc_lst));
 		return -ENODEV;
 	}
@@ -180,8 +181,8 @@ int ipv6_sock_mc_join(struct sock *sk, int ifindex, const struct in6_addr *addr)
 	err = ipv6_dev_mc_inc(dev, addr);
 
 	if (err) {
+		rcu_read_unlock();
 		sock_kfree_s(sk, mc_lst, sizeof(*mc_lst));
-		dev_put(dev);
 		return err;
 	}
 
@@ -190,7 +191,7 @@ int ipv6_sock_mc_join(struct sock *sk, int ifindex, const struct in6_addr *addr)
 	np->ipv6_mc_list = mc_lst;
 	write_unlock_bh(&ipv6_sk_mc_lock);
 
-	dev_put(dev);
+	rcu_read_unlock();
 
 	return 0;
 }
@@ -213,18 +214,17 @@ int ipv6_sock_mc_drop(struct sock *sk, int ifindex, const struct in6_addr *addr)
 			*lnk = mc_lst->next;
 			write_unlock_bh(&ipv6_sk_mc_lock);
 
-			dev = dev_get_by_index(net, mc_lst->ifindex);
+			rcu_read_lock();
+			dev = dev_get_by_index_rcu(net, mc_lst->ifindex);
 			if (dev != NULL) {
-				struct inet6_dev *idev = in6_dev_get(dev);
+				struct inet6_dev *idev = __in6_dev_get(dev);
 
 				(void) ip6_mc_leave_src(sk, mc_lst, idev);
-				if (idev) {
+				if (idev)
 					__ipv6_dev_mc_dec(idev, &mc_lst->addr);
-					in6_dev_put(idev);
-				}
-				dev_put(dev);
 			} else
 				(void) ip6_mc_leave_src(sk, mc_lst, NULL);
+			rcu_read_unlock();
 			sock_kfree_s(sk, mc_lst, sizeof(*mc_lst));
 			return 0;
 		}
@@ -234,43 +234,36 @@ int ipv6_sock_mc_drop(struct sock *sk, int ifindex, const struct in6_addr *addr)
 	return -EADDRNOTAVAIL;
 }
 
-static struct inet6_dev *ip6_mc_find_dev(struct net *net,
-					 struct in6_addr *group,
-					 int ifindex)
+/* called with rcu_read_lock() */
+static struct inet6_dev *ip6_mc_find_dev_rcu(struct net *net,
+					     struct in6_addr *group,
+					     int ifindex)
 {
 	struct net_device *dev = NULL;
 	struct inet6_dev *idev = NULL;
 
 	if (ifindex == 0) {
-		struct rt6_info *rt;
+		struct rt6_info *rt = rt6_lookup(net, group, NULL, 0, 0);
 
-		rt = rt6_lookup(net, group, NULL, 0, 0);
 		if (rt) {
 			dev = rt->rt6i_dev;
 			dev_hold(dev);
 			dst_release(&rt->u.dst);
 		}
 	} else
-		dev = dev_get_by_index(net, ifindex);
+		dev = dev_get_by_index_rcu(net, ifindex);
 
 	if (!dev)
-		goto nodev;
-	idev = in6_dev_get(dev);
+		return NULL;
+	idev = __in6_dev_get(dev);
 	if (!idev)
-		goto release;
+		return NULL;;
 	read_lock_bh(&idev->lock);
-	if (idev->dead)
-		goto unlock_release;
-
+	if (idev->dead) {
+		read_unlock_bh(&idev->lock);
+		return NULL;
+	}
 	return idev;
-
-unlock_release:
-	read_unlock_bh(&idev->lock);
-	in6_dev_put(idev);
-release:
-	dev_put(dev);
-nodev:
-	return NULL;
 }
 
 void ipv6_sock_mc_close(struct sock *sk)
@@ -286,19 +279,17 @@ void ipv6_sock_mc_close(struct sock *sk)
 		np->ipv6_mc_list = mc_lst->next;
 		write_unlock_bh(&ipv6_sk_mc_lock);
 
-		dev = dev_get_by_index(net, mc_lst->ifindex);
+		rcu_read_lock();
+		dev = dev_get_by_index_rcu(net, mc_lst->ifindex);
 		if (dev) {
-			struct inet6_dev *idev = in6_dev_get(dev);
+			struct inet6_dev *idev = __in6_dev_get(dev);
 
 			(void) ip6_mc_leave_src(sk, mc_lst, idev);
-			if (idev) {
+			if (idev)
 				__ipv6_dev_mc_dec(idev, &mc_lst->addr);
-				in6_dev_put(idev);
-			}
-			dev_put(dev);
 		} else
 			(void) ip6_mc_leave_src(sk, mc_lst, NULL);
-
+		rcu_read_unlock();
 		sock_kfree_s(sk, mc_lst, sizeof(*mc_lst));
 
 		write_lock_bh(&ipv6_sk_mc_lock);
@@ -327,14 +318,17 @@ int ip6_mc_source(int add, int omode, struct sock *sk,
 	if (!ipv6_addr_is_multicast(group))
 		return -EINVAL;
 
-	idev = ip6_mc_find_dev(net, group, pgsr->gsr_interface);
-	if (!idev)
+	rcu_read_lock();
+	idev = ip6_mc_find_dev_rcu(net, group, pgsr->gsr_interface);
+	if (!idev) {
+		rcu_read_unlock();
 		return -ENODEV;
+	}
 	dev = idev->dev;
 
 	err = -EADDRNOTAVAIL;
 
-	read_lock_bh(&ipv6_sk_mc_lock);
+	read_lock(&ipv6_sk_mc_lock);
 	for (pmc=inet6->ipv6_mc_list; pmc; pmc=pmc->next) {
 		if (pgsr->gsr_interface && pmc->ifindex != pgsr->gsr_interface)
 			continue;
@@ -358,7 +352,7 @@ int ip6_mc_source(int add, int omode, struct sock *sk,
 		pmc->sfmode = omode;
 	}
 
-	write_lock_bh(&pmc->sflock);
+	write_lock(&pmc->sflock);
 	pmclocked = 1;
 
 	psl = pmc->sflist;
@@ -433,11 +427,10 @@ int ip6_mc_source(int add, int omode, struct sock *sk,
 	ip6_mc_add_src(idev, group, omode, 1, source, 1);
 done:
 	if (pmclocked)
-		write_unlock_bh(&pmc->sflock);
-	read_unlock_bh(&ipv6_sk_mc_lock);
+		write_unlock(&pmc->sflock);
+	read_unlock(&ipv6_sk_mc_lock);
 	read_unlock_bh(&idev->lock);
-	in6_dev_put(idev);
-	dev_put(dev);
+	rcu_read_unlock();
 	if (leavegroup)
 		return ipv6_sock_mc_drop(sk, pgsr->gsr_interface, group);
 	return err;
@@ -463,14 +456,17 @@ int ip6_mc_msfilter(struct sock *sk, struct group_filter *gsf)
 	    gsf->gf_fmode != MCAST_EXCLUDE)
 		return -EINVAL;
 
-	idev = ip6_mc_find_dev(net, group, gsf->gf_interface);
+	rcu_read_lock();
+	idev = ip6_mc_find_dev_rcu(net, group, gsf->gf_interface);
 
-	if (!idev)
+	if (!idev) {
+		rcu_read_unlock();
 		return -ENODEV;
+	}
 	dev = idev->dev;
 
 	err = 0;
-	read_lock_bh(&ipv6_sk_mc_lock);
+	read_lock(&ipv6_sk_mc_lock);
 
 	if (gsf->gf_fmode == MCAST_INCLUDE && gsf->gf_numsrc == 0) {
 		leavegroup = 1;
@@ -512,7 +508,7 @@ int ip6_mc_msfilter(struct sock *sk, struct group_filter *gsf)
 		(void) ip6_mc_add_src(idev, group, gsf->gf_fmode, 0, NULL, 0);
 	}
 
-	write_lock_bh(&pmc->sflock);
+	write_lock(&pmc->sflock);
 	psl = pmc->sflist;
 	if (psl) {
 		(void) ip6_mc_del_src(idev, group, pmc->sfmode,
@@ -522,13 +518,12 @@ int ip6_mc_msfilter(struct sock *sk, struct group_filter *gsf)
 		(void) ip6_mc_del_src(idev, group, pmc->sfmode, 0, NULL, 0);
 	pmc->sflist = newpsl;
 	pmc->sfmode = gsf->gf_fmode;
-	write_unlock_bh(&pmc->sflock);
+	write_unlock(&pmc->sflock);
 	err = 0;
 done:
-	read_unlock_bh(&ipv6_sk_mc_lock);
+	read_unlock(&ipv6_sk_mc_lock);
 	read_unlock_bh(&idev->lock);
-	in6_dev_put(idev);
-	dev_put(dev);
+	rcu_read_unlock();
 	if (leavegroup)
 		err = ipv6_sock_mc_drop(sk, gsf->gf_interface, group);
 	return err;
@@ -551,11 +546,13 @@ int ip6_mc_msfget(struct sock *sk, struct group_filter *gsf,
 	if (!ipv6_addr_is_multicast(group))
 		return -EINVAL;
 
-	idev = ip6_mc_find_dev(net, group, gsf->gf_interface);
+	rcu_read_lock();
+	idev = ip6_mc_find_dev_rcu(net, group, gsf->gf_interface);
 
-	if (!idev)
+	if (!idev) {
+		rcu_read_unlock();
 		return -ENODEV;
-
+	}
 	dev = idev->dev;
 
 	err = -EADDRNOTAVAIL;
@@ -577,8 +574,7 @@ int ip6_mc_msfget(struct sock *sk, struct group_filter *gsf,
 	psl = pmc->sflist;
 	count = psl ? psl->sl_count : 0;
 	read_unlock_bh(&idev->lock);
-	in6_dev_put(idev);
-	dev_put(dev);
+	rcu_read_unlock();
 
 	copycount = count < gsf->gf_numsrc ? count : gsf->gf_numsrc;
 	gsf->gf_numsrc = count;
@@ -604,8 +600,7 @@ int ip6_mc_msfget(struct sock *sk, struct group_filter *gsf,
 	return 0;
 done:
 	read_unlock_bh(&idev->lock);
-	in6_dev_put(idev);
-	dev_put(dev);
+	rcu_read_unlock();
 	return err;
 }
 
@@ -822,6 +817,7 @@ int ipv6_dev_mc_inc(struct net_device *dev, const struct in6_addr *addr)
 	struct ifmcaddr6 *mc;
 	struct inet6_dev *idev;
 
+	/* we need to take a reference on idev */
 	idev = in6_dev_get(dev);
 
 	if (idev == NULL)
@@ -860,7 +856,7 @@ int ipv6_dev_mc_inc(struct net_device *dev, const struct in6_addr *addr)
 	setup_timer(&mc->mca_timer, igmp6_timer_handler, (unsigned long)mc);
 
 	ipv6_addr_copy(&mc->mca_addr, addr);
-	mc->idev = idev;
+	mc->idev = idev; /* (reference taken) */
 	mc->mca_users = 1;
 	/* mca_stamp should be updated upon changes */
 	mc->mca_cstamp = mc->mca_tstamp = jiffies;
@@ -915,16 +911,18 @@ int __ipv6_dev_mc_dec(struct inet6_dev *idev, const struct in6_addr *addr)
 
 int ipv6_dev_mc_dec(struct net_device *dev, const struct in6_addr *addr)
 {
-	struct inet6_dev *idev = in6_dev_get(dev);
+	struct inet6_dev *idev;
 	int err;
 
-	if (!idev)
-		return -ENODEV;
-
-	err = __ipv6_dev_mc_dec(idev, addr);
+	rcu_read_lock();
 
-	in6_dev_put(idev);
+	idev = __in6_dev_get(dev);
+	if (!idev)
+		err = -ENODEV;
+	else
+		err = __ipv6_dev_mc_dec(idev, addr);
 
+	rcu_read_unlock();
 	return err;
 }
 
@@ -965,7 +963,8 @@ int ipv6_chk_mcast_addr(struct net_device *dev, const struct in6_addr *group,
 	struct ifmcaddr6 *mc;
 	int rv = 0;
 
-	idev = in6_dev_get(dev);
+	rcu_read_lock();
+	idev = __in6_dev_get(dev);
 	if (idev) {
 		read_lock_bh(&idev->lock);
 		for (mc = idev->mc_list; mc; mc=mc->next) {
@@ -992,8 +991,8 @@ int ipv6_chk_mcast_addr(struct net_device *dev, const struct in6_addr *group,
 				rv = 1; /* don't filter unspecified source */
 		}
 		read_unlock_bh(&idev->lock);
-		in6_dev_put(idev);
 	}
+	rcu_read_unlock();
 	return rv;
 }
 
@@ -1104,6 +1103,7 @@ static int mld_marksources(struct ifmcaddr6 *pmc, int nsrcs,
 	return 1;
 }
 
+/* called with rcu_read_lock() */
 int igmp6_event_query(struct sk_buff *skb)
 {
 	struct mld2_query *mlh2 = NULL;
@@ -1127,7 +1127,7 @@ int igmp6_event_query(struct sk_buff *skb)
 	if (!(ipv6_addr_type(&ipv6_hdr(skb)->saddr) & IPV6_ADDR_LINKLOCAL))
 		return -EINVAL;
 
-	idev = in6_dev_get(skb->dev);
+	idev = __in6_dev_get(skb->dev);
 
 	if (idev == NULL)
 		return 0;
@@ -1137,10 +1137,8 @@ int igmp6_event_query(struct sk_buff *skb)
 	group_type = ipv6_addr_type(group);
 
 	if (group_type != IPV6_ADDR_ANY &&
-	    !(group_type&IPV6_ADDR_MULTICAST)) {
-		in6_dev_put(idev);
+	    !(group_type&IPV6_ADDR_MULTICAST))
 		return -EINVAL;
-	}
 
 	if (len == 24) {
 		int switchback;
@@ -1161,10 +1159,9 @@ int igmp6_event_query(struct sk_buff *skb)
 	} else if (len >= 28) {
 		int srcs_offset = sizeof(struct mld2_query) -
 				  sizeof(struct icmp6hdr);
-		if (!pskb_may_pull(skb, srcs_offset)) {
-			in6_dev_put(idev);
+		if (!pskb_may_pull(skb, srcs_offset))
 			return -EINVAL;
-		}
+
 		mlh2 = (struct mld2_query *)skb_transport_header(skb);
 		max_delay = (MLDV2_MRC(ntohs(mlh2->mld2q_mrc))*HZ)/1000;
 		if (!max_delay)
@@ -1173,28 +1170,23 @@ int igmp6_event_query(struct sk_buff *skb)
 		if (mlh2->mld2q_qrv)
 			idev->mc_qrv = mlh2->mld2q_qrv;
 		if (group_type == IPV6_ADDR_ANY) { /* general query */
-			if (mlh2->mld2q_nsrcs) {
-				in6_dev_put(idev);
+			if (mlh2->mld2q_nsrcs)
 				return -EINVAL; /* no sources allowed */
-			}
+
 			mld_gq_start_timer(idev);
-			in6_dev_put(idev);
 			return 0;
 		}
 		/* mark sources to include, if group & source-specific */
 		if (mlh2->mld2q_nsrcs != 0) {
 			if (!pskb_may_pull(skb, srcs_offset +
-			    ntohs(mlh2->mld2q_nsrcs) * sizeof(struct in6_addr))) {
-				in6_dev_put(idev);
+			    ntohs(mlh2->mld2q_nsrcs) * sizeof(struct in6_addr)))
 				return -EINVAL;
-			}
+
 			mlh2 = (struct mld2_query *)skb_transport_header(skb);
 			mark = 1;
 		}
-	} else {
-		in6_dev_put(idev);
+	} else
 		return -EINVAL;
-	}
 
 	read_lock_bh(&idev->lock);
 	if (group_type == IPV6_ADDR_ANY) {
@@ -1227,12 +1219,11 @@ int igmp6_event_query(struct sk_buff *skb)
 		}
 	}
 	read_unlock_bh(&idev->lock);
-	in6_dev_put(idev);
 
 	return 0;
 }
 
-
+/* called with rcu_read_lock() */
 int igmp6_event_report(struct sk_buff *skb)
 {
 	struct ifmcaddr6 *ma;
@@ -1260,7 +1251,7 @@ int igmp6_event_report(struct sk_buff *skb)
 	    !(addr_type&IPV6_ADDR_LINKLOCAL))
 		return -EINVAL;
 
-	idev = in6_dev_get(skb->dev);
+	idev = __in6_dev_get(skb->dev);
 	if (idev == NULL)
 		return -ENODEV;
 
@@ -1280,7 +1271,6 @@ int igmp6_event_report(struct sk_buff *skb)
 		}
 	}
 	read_unlock_bh(&idev->lock);
-	in6_dev_put(idev);
 	return 0;
 }
 
@@ -1396,12 +1386,14 @@ static void mld_sendpack(struct sk_buff *skb)
 	struct mld2_report *pmr =
 			      (struct mld2_report *)skb_transport_header(skb);
 	int payload_len, mldlen;
-	struct inet6_dev *idev = in6_dev_get(skb->dev);
+	struct inet6_dev *idev;
 	struct net *net = dev_net(skb->dev);
 	int err;
 	struct flowi fl;
 	struct dst_entry *dst;
 
+	rcu_read_lock();
+	idev = __in6_dev_get(skb->dev);
 	IP6_UPD_PO_STATS(net, idev, IPSTATS_MIB_OUT, skb->len);
 
 	payload_len = (skb->tail - skb->network_header) - sizeof(*pip6);
@@ -1441,8 +1433,7 @@ out:
 	} else
 		IP6_INC_STATS_BH(net, idev, IPSTATS_MIB_OUTDISCARDS);
 
-	if (likely(idev != NULL))
-		in6_dev_put(idev);
+	rcu_read_unlock();
 	return;
 
 err_out:
@@ -1779,7 +1770,8 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type)
 					 IPPROTO_ICMPV6,
 					 csum_partial(hdr, len, 0));
 
-	idev = in6_dev_get(skb->dev);
+	rcu_read_lock();
+	idev = __in6_dev_get(skb->dev);
 
 	dst = icmp6_dst_alloc(skb->dev, NULL, &ipv6_hdr(skb)->daddr);
 	if (!dst) {
@@ -1806,8 +1798,7 @@ out:
 	} else
 		IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTDISCARDS);
 
-	if (likely(idev != NULL))
-		in6_dev_put(idev);
+	rcu_read_unlock();
 	return;
 
 err_out:
-- 
cgit v1.2.3


From 00d9d6a185de89edc0649ca4ead58f0283dfcbac Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 7 Jun 2010 22:24:44 +0000
Subject: ipv6: fix ICMP6_MIB_OUTERRORS

In commit 1f8438a85366 (icmp: Account for ICMP out errors), I did a typo
on IPV6 side, using ICMP6_MIB_OUTMSGS instead of ICMP6_MIB_OUTERRORS

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/icmp.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index ce7992982557..03e62f94ff8e 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -483,7 +483,7 @@ route_done:
 			      np->tclass, NULL, &fl, (struct rt6_info*)dst,
 			      MSG_DONTWAIT, np->dontfrag);
 	if (err) {
-		ICMP6_INC_STATS_BH(net, idev, ICMP6_MIB_OUTMSGS);
+		ICMP6_INC_STATS_BH(net, idev, ICMP6_MIB_OUTERRORS);
 		ip6_flush_pending_frames(sk);
 		goto out_put;
 	}
@@ -565,7 +565,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb)
 				np->dontfrag);
 
 	if (err) {
-		ICMP6_INC_STATS_BH(net, idev, ICMP6_MIB_OUTMSGS);
+		ICMP6_INC_STATS_BH(net, idev, ICMP6_MIB_OUTERRORS);
 		ip6_flush_pending_frames(sk);
 		goto out_put;
 	}
-- 
cgit v1.2.3


From d8d1f30b95a635dbd610dcc5eb641aca8f4768cf Mon Sep 17 00:00:00 2001
From: Changli Gao <xiaosuo@gmail.com>
Date: Thu, 10 Jun 2010 23:31:35 -0700
Subject: net-next: remove useless union keyword

remove useless union keyword in rtable, rt6_info and dn_route.

Since there is only one member in a union, the union keyword isn't useful.

Signed-off-by: Changli Gao <xiaosuo@gmail.com>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/addrconf.c   |  10 +-
 net/ipv6/anycast.c    |   6 +-
 net/ipv6/fib6_rules.c |  10 +-
 net/ipv6/ip6_fib.c    |  30 ++---
 net/ipv6/ip6_output.c |  38 +++----
 net/ipv6/ip6_tunnel.c |   8 +-
 net/ipv6/mcast.c      |   4 +-
 net/ipv6/ndisc.c      |   8 +-
 net/ipv6/raw.c        |  12 +-
 net/ipv6/route.c      | 300 +++++++++++++++++++++++++-------------------------
 net/ipv6/sit.c        |   8 +-
 11 files changed, 214 insertions(+), 220 deletions(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index e1a698df5706..b97bb1f30808 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -557,7 +557,7 @@ void inet6_ifa_finish_destroy(struct inet6_ifaddr *ifp)
 		pr_warning("Freeing alive inet6 address %p\n", ifp);
 		return;
 	}
-	dst_release(&ifp->rt->u.dst);
+	dst_release(&ifp->rt->dst);
 
 	call_rcu(&ifp->rcu, inet6_ifa_finish_destroy_rcu);
 }
@@ -823,7 +823,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp)
 				rt->rt6i_flags |= RTF_EXPIRES;
 			}
 		}
-		dst_release(&rt->u.dst);
+		dst_release(&rt->dst);
 	}
 
 out:
@@ -1863,7 +1863,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len)
 					      dev, expires, flags);
 		}
 		if (rt)
-			dst_release(&rt->u.dst);
+			dst_release(&rt->dst);
 	}
 
 	/* Try to figure out our local address for this prefix */
@@ -4093,11 +4093,11 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
 		if (ifp->idev->cnf.forwarding)
 			addrconf_leave_anycast(ifp);
 		addrconf_leave_solict(ifp->idev, &ifp->addr);
-		dst_hold(&ifp->rt->u.dst);
+		dst_hold(&ifp->rt->dst);
 
 		if (ifp->state == INET6_IFADDR_STATE_DEAD &&
 		    ip6_del_rt(ifp->rt))
-			dst_free(&ifp->rt->u.dst);
+			dst_free(&ifp->rt->dst);
 		break;
 	}
 }
diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c
index f058fbd808c8..0e5e943446f0 100644
--- a/net/ipv6/anycast.c
+++ b/net/ipv6/anycast.c
@@ -84,7 +84,7 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr)
 		rt = rt6_lookup(net, addr, NULL, 0, 0);
 		if (rt) {
 			dev = rt->rt6i_dev;
-			dst_release(&rt->u.dst);
+			dst_release(&rt->dst);
 		} else if (ishost) {
 			err = -EADDRNOTAVAIL;
 			goto error;
@@ -244,7 +244,7 @@ static void aca_put(struct ifacaddr6 *ac)
 {
 	if (atomic_dec_and_test(&ac->aca_refcnt)) {
 		in6_dev_put(ac->aca_idev);
-		dst_release(&ac->aca_rt->u.dst);
+		dst_release(&ac->aca_rt->dst);
 		kfree(ac);
 	}
 }
@@ -350,7 +350,7 @@ int __ipv6_dev_ac_dec(struct inet6_dev *idev, struct in6_addr *addr)
 	write_unlock_bh(&idev->lock);
 	addrconf_leave_solict(idev, &aca->aca_addr);
 
-	dst_hold(&aca->aca_rt->u.dst);
+	dst_hold(&aca->aca_rt->dst);
 	ip6_del_rt(aca->aca_rt);
 
 	aca_put(aca);
diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c
index 8e44f8f9c188..b1108ede18e1 100644
--- a/net/ipv6/fib6_rules.c
+++ b/net/ipv6/fib6_rules.c
@@ -43,8 +43,8 @@ struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl,
 	if (arg.result)
 		return arg.result;
 
-	dst_hold(&net->ipv6.ip6_null_entry->u.dst);
-	return &net->ipv6.ip6_null_entry->u.dst;
+	dst_hold(&net->ipv6.ip6_null_entry->dst);
+	return &net->ipv6.ip6_null_entry->dst;
 }
 
 static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
@@ -86,7 +86,7 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
 			struct in6_addr saddr;
 
 			if (ipv6_dev_get_saddr(net,
-					       ip6_dst_idev(&rt->u.dst)->dev,
+					       ip6_dst_idev(&rt->dst)->dev,
 					       &flp->fl6_dst,
 					       rt6_flags2srcprefs(flags),
 					       &saddr))
@@ -99,12 +99,12 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
 		goto out;
 	}
 again:
-	dst_release(&rt->u.dst);
+	dst_release(&rt->dst);
 	rt = NULL;
 	goto out;
 
 discard_pkt:
-	dst_hold(&rt->u.dst);
+	dst_hold(&rt->dst);
 out:
 	arg->result = rt;
 	return rt == NULL ? -EAGAIN : 0;
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 92a122b7795d..b6a585909d35 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -165,7 +165,7 @@ static __inline__ void node_free(struct fib6_node * fn)
 static __inline__ void rt6_release(struct rt6_info *rt)
 {
 	if (atomic_dec_and_test(&rt->rt6i_ref))
-		dst_free(&rt->u.dst);
+		dst_free(&rt->dst);
 }
 
 static void fib6_link_table(struct net *net, struct fib6_table *tb)
@@ -278,7 +278,7 @@ static int fib6_dump_node(struct fib6_walker_t *w)
 	int res;
 	struct rt6_info *rt;
 
-	for (rt = w->leaf; rt; rt = rt->u.dst.rt6_next) {
+	for (rt = w->leaf; rt; rt = rt->dst.rt6_next) {
 		res = rt6_dump_route(rt, w->args);
 		if (res < 0) {
 			/* Frame is full, suspend walking */
@@ -619,7 +619,7 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt,
 
 	ins = &fn->leaf;
 
-	for (iter = fn->leaf; iter; iter=iter->u.dst.rt6_next) {
+	for (iter = fn->leaf; iter; iter=iter->dst.rt6_next) {
 		/*
 		 *	Search for duplicates
 		 */
@@ -647,7 +647,7 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt,
 		if (iter->rt6i_metric > rt->rt6i_metric)
 			break;
 
-		ins = &iter->u.dst.rt6_next;
+		ins = &iter->dst.rt6_next;
 	}
 
 	/* Reset round-robin state, if necessary */
@@ -658,7 +658,7 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt,
 	 *	insert node
 	 */
 
-	rt->u.dst.rt6_next = iter;
+	rt->dst.rt6_next = iter;
 	*ins = rt;
 	rt->rt6i_node = fn;
 	atomic_inc(&rt->rt6i_ref);
@@ -799,7 +799,7 @@ out:
 			atomic_inc(&pn->leaf->rt6i_ref);
 		}
 #endif
-		dst_free(&rt->u.dst);
+		dst_free(&rt->dst);
 	}
 	return err;
 
@@ -810,7 +810,7 @@ out:
 st_failure:
 	if (fn && !(fn->fn_flags & (RTN_RTINFO|RTN_ROOT)))
 		fib6_repair_tree(info->nl_net, fn);
-	dst_free(&rt->u.dst);
+	dst_free(&rt->dst);
 	return err;
 #endif
 }
@@ -1108,7 +1108,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
 	RT6_TRACE("fib6_del_route\n");
 
 	/* Unlink it */
-	*rtp = rt->u.dst.rt6_next;
+	*rtp = rt->dst.rt6_next;
 	rt->rt6i_node = NULL;
 	net->ipv6.rt6_stats->fib_rt_entries--;
 	net->ipv6.rt6_stats->fib_discarded_routes++;
@@ -1122,14 +1122,14 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp,
 	FOR_WALKERS(w) {
 		if (w->state == FWS_C && w->leaf == rt) {
 			RT6_TRACE("walker %p adjusted by delroute\n", w);
-			w->leaf = rt->u.dst.rt6_next;
+			w->leaf = rt->dst.rt6_next;
 			if (w->leaf == NULL)
 				w->state = FWS_U;
 		}
 	}
 	read_unlock(&fib6_walker_lock);
 
-	rt->u.dst.rt6_next = NULL;
+	rt->dst.rt6_next = NULL;
 
 	/* If it was last route, expunge its radix tree node */
 	if (fn->leaf == NULL) {
@@ -1168,7 +1168,7 @@ int fib6_del(struct rt6_info *rt, struct nl_info *info)
 	struct rt6_info **rtp;
 
 #if RT6_DEBUG >= 2
-	if (rt->u.dst.obsolete>0) {
+	if (rt->dst.obsolete>0) {
 		WARN_ON(fn != NULL);
 		return -ENOENT;
 	}
@@ -1195,7 +1195,7 @@ int fib6_del(struct rt6_info *rt, struct nl_info *info)
 	 *	Walk the leaf entries looking for ourself
 	 */
 
-	for (rtp = &fn->leaf; *rtp; rtp = &(*rtp)->u.dst.rt6_next) {
+	for (rtp = &fn->leaf; *rtp; rtp = &(*rtp)->dst.rt6_next) {
 		if (*rtp == rt) {
 			fib6_del_route(fn, rtp, info);
 			return 0;
@@ -1334,7 +1334,7 @@ static int fib6_clean_node(struct fib6_walker_t *w)
 		.nl_net = c->net,
 	};
 
-	for (rt = w->leaf; rt; rt = rt->u.dst.rt6_next) {
+	for (rt = w->leaf; rt; rt = rt->dst.rt6_next) {
 		res = c->func(rt, c->arg);
 		if (res < 0) {
 			w->leaf = rt;
@@ -1448,8 +1448,8 @@ static int fib6_age(struct rt6_info *rt, void *arg)
 		}
 		gc_args.more++;
 	} else if (rt->rt6i_flags & RTF_CACHE) {
-		if (atomic_read(&rt->u.dst.__refcnt) == 0 &&
-		    time_after_eq(now, rt->u.dst.lastuse + gc_args.timeout)) {
+		if (atomic_read(&rt->dst.__refcnt) == 0 &&
+		    time_after_eq(now, rt->dst.lastuse + gc_args.timeout)) {
 			RT6_TRACE("aging clone %p\n", rt);
 			return -1;
 		} else if ((rt->rt6i_flags & RTF_GATEWAY) &&
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 89425af0684c..d40b330c0ee6 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -698,7 +698,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *))
 		ipv6_hdr(skb)->payload_len = htons(first_len -
 						   sizeof(struct ipv6hdr));
 
-		dst_hold(&rt->u.dst);
+		dst_hold(&rt->dst);
 
 		for (;;) {
 			/* Prepare header of the next frame,
@@ -726,7 +726,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *))
 
 			err = output(skb);
 			if(!err)
-				IP6_INC_STATS(net, ip6_dst_idev(&rt->u.dst),
+				IP6_INC_STATS(net, ip6_dst_idev(&rt->dst),
 					      IPSTATS_MIB_FRAGCREATES);
 
 			if (err || !frag)
@@ -740,9 +740,9 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *))
 		kfree(tmp_hdr);
 
 		if (err == 0) {
-			IP6_INC_STATS(net, ip6_dst_idev(&rt->u.dst),
+			IP6_INC_STATS(net, ip6_dst_idev(&rt->dst),
 				      IPSTATS_MIB_FRAGOKS);
-			dst_release(&rt->u.dst);
+			dst_release(&rt->dst);
 			return 0;
 		}
 
@@ -752,9 +752,9 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *))
 			frag = skb;
 		}
 
-		IP6_INC_STATS(net, ip6_dst_idev(&rt->u.dst),
+		IP6_INC_STATS(net, ip6_dst_idev(&rt->dst),
 			      IPSTATS_MIB_FRAGFAILS);
-		dst_release(&rt->u.dst);
+		dst_release(&rt->dst);
 		return err;
 	}
 
@@ -785,7 +785,7 @@ slow_path:
 		 *	Allocate buffer.
 		 */
 
-		if ((frag = alloc_skb(len+hlen+sizeof(struct frag_hdr)+LL_ALLOCATED_SPACE(rt->u.dst.dev), GFP_ATOMIC)) == NULL) {
+		if ((frag = alloc_skb(len+hlen+sizeof(struct frag_hdr)+LL_ALLOCATED_SPACE(rt->dst.dev), GFP_ATOMIC)) == NULL) {
 			NETDEBUG(KERN_INFO "IPv6: frag: no memory for new fragment!\n");
 			IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)),
 				      IPSTATS_MIB_FRAGFAILS);
@@ -798,7 +798,7 @@ slow_path:
 		 */
 
 		ip6_copy_metadata(frag, skb);
-		skb_reserve(frag, LL_RESERVED_SPACE(rt->u.dst.dev));
+		skb_reserve(frag, LL_RESERVED_SPACE(rt->dst.dev));
 		skb_put(frag, len + hlen + sizeof(struct frag_hdr));
 		skb_reset_network_header(frag);
 		fh = (struct frag_hdr *)(skb_network_header(frag) + hlen);
@@ -1156,24 +1156,24 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to,
 
 			/* need source address above miyazawa*/
 		}
-		dst_hold(&rt->u.dst);
-		inet->cork.dst = &rt->u.dst;
+		dst_hold(&rt->dst);
+		inet->cork.dst = &rt->dst;
 		inet->cork.fl = *fl;
 		np->cork.hop_limit = hlimit;
 		np->cork.tclass = tclass;
 		mtu = np->pmtudisc == IPV6_PMTUDISC_PROBE ?
-		      rt->u.dst.dev->mtu : dst_mtu(rt->u.dst.path);
+		      rt->dst.dev->mtu : dst_mtu(rt->dst.path);
 		if (np->frag_size < mtu) {
 			if (np->frag_size)
 				mtu = np->frag_size;
 		}
 		inet->cork.fragsize = mtu;
-		if (dst_allfrag(rt->u.dst.path))
+		if (dst_allfrag(rt->dst.path))
 			inet->cork.flags |= IPCORK_ALLFRAG;
 		inet->cork.length = 0;
 		sk->sk_sndmsg_page = NULL;
 		sk->sk_sndmsg_off = 0;
-		exthdrlen = rt->u.dst.header_len + (opt ? opt->opt_flen : 0) -
+		exthdrlen = rt->dst.header_len + (opt ? opt->opt_flen : 0) -
 			    rt->rt6i_nfheader_len;
 		length += exthdrlen;
 		transhdrlen += exthdrlen;
@@ -1186,7 +1186,7 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to,
 		mtu = inet->cork.fragsize;
 	}
 
-	hh_len = LL_RESERVED_SPACE(rt->u.dst.dev);
+	hh_len = LL_RESERVED_SPACE(rt->dst.dev);
 
 	fragheaderlen = sizeof(struct ipv6hdr) + rt->rt6i_nfheader_len +
 			(opt ? opt->opt_nflen : 0);
@@ -1224,7 +1224,7 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to,
 		}
 
 		if (proto == IPPROTO_UDP &&
-		    (rt->u.dst.dev->features & NETIF_F_UFO)) {
+		    (rt->dst.dev->features & NETIF_F_UFO)) {
 
 			err = ip6_ufo_append_data(sk, getfrag, from, length,
 						  hh_len, fragheaderlen,
@@ -1270,7 +1270,7 @@ alloc_new_skb:
 
 			fraglen = datalen + fragheaderlen;
 			if ((flags & MSG_MORE) &&
-			    !(rt->u.dst.dev->features&NETIF_F_SG))
+			    !(rt->dst.dev->features&NETIF_F_SG))
 				alloclen = mtu;
 			else
 				alloclen = datalen + fragheaderlen;
@@ -1281,7 +1281,7 @@ alloc_new_skb:
 			 * because we have no idea if we're the last one.
 			 */
 			if (datalen == length + fraggap)
-				alloclen += rt->u.dst.trailer_len;
+				alloclen += rt->dst.trailer_len;
 
 			/*
 			 * We just reserve space for fragment header.
@@ -1358,7 +1358,7 @@ alloc_new_skb:
 		if (copy > length)
 			copy = length;
 
-		if (!(rt->u.dst.dev->features&NETIF_F_SG)) {
+		if (!(rt->dst.dev->features&NETIF_F_SG)) {
 			unsigned int off;
 
 			off = skb->len;
@@ -1503,7 +1503,7 @@ int ip6_push_pending_frames(struct sock *sk)
 	skb->priority = sk->sk_priority;
 	skb->mark = sk->sk_mark;
 
-	skb_dst_set(skb, dst_clone(&rt->u.dst));
+	skb_dst_set(skb, dst_clone(&rt->dst));
 	IP6_UPD_PO_STATS(net, rt->rt6i_idev, IPSTATS_MIB_OUT, skb->len);
 	if (proto == IPPROTO_ICMPV6) {
 		struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb));
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 8f39893d8081..0fd027f3f47e 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -552,7 +552,7 @@ ip4ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	if (ip_route_output_key(dev_net(skb->dev), &rt, &fl))
 		goto out;
 
-	skb2->dev = rt->u.dst.dev;
+	skb2->dev = rt->dst.dev;
 
 	/* route "incoming" packet */
 	if (rt->rt_flags & RTCF_LOCAL) {
@@ -562,7 +562,7 @@ ip4ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		fl.fl4_src = eiph->saddr;
 		fl.fl4_tos = eiph->tos;
 		if (ip_route_output_key(dev_net(skb->dev), &rt, &fl) ||
-		    rt->u.dst.dev->type != ARPHRD_TUNNEL) {
+		    rt->dst.dev->type != ARPHRD_TUNNEL) {
 			ip_rt_put(rt);
 			goto out;
 		}
@@ -626,7 +626,7 @@ ip6ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		icmpv6_send(skb2, rel_type, rel_code, rel_info);
 
 		if (rt)
-			dst_release(&rt->u.dst);
+			dst_release(&rt->dst);
 
 		kfree_skb(skb2);
 	}
@@ -1135,7 +1135,7 @@ static void ip6_tnl_link_config(struct ip6_tnl *t)
 			if (dev->mtu < IPV6_MIN_MTU)
 				dev->mtu = IPV6_MIN_MTU;
 		}
-		dst_release(&rt->u.dst);
+		dst_release(&rt->dst);
 	}
 }
 
diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index 3e36d1538b6e..d1444b95ad7e 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -158,7 +158,7 @@ int ipv6_sock_mc_join(struct sock *sk, int ifindex, const struct in6_addr *addr)
 		rt = rt6_lookup(net, addr, NULL, 0, 0);
 		if (rt) {
 			dev = rt->rt6i_dev;
-			dst_release(&rt->u.dst);
+			dst_release(&rt->dst);
 		}
 	} else
 		dev = dev_get_by_index_rcu(net, ifindex);
@@ -248,7 +248,7 @@ static struct inet6_dev *ip6_mc_find_dev_rcu(struct net *net,
 		if (rt) {
 			dev = rt->rt6i_dev;
 			dev_hold(dev);
-			dst_release(&rt->u.dst);
+			dst_release(&rt->dst);
 		}
 	} else
 		dev = dev_get_by_index_rcu(net, ifindex);
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 0abdc242ddb7..1fc46fc60efd 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -1229,7 +1229,7 @@ static void ndisc_router_discovery(struct sk_buff *skb)
 			ND_PRINTK0(KERN_ERR
 				   "ICMPv6 RA: %s() got default router without neighbour.\n",
 				   __func__);
-			dst_release(&rt->u.dst);
+			dst_release(&rt->dst);
 			in6_dev_put(in6_dev);
 			return;
 		}
@@ -1244,7 +1244,7 @@ static void ndisc_router_discovery(struct sk_buff *skb)
 	if (ra_msg->icmph.icmp6_hop_limit) {
 		in6_dev->cnf.hop_limit = ra_msg->icmph.icmp6_hop_limit;
 		if (rt)
-			rt->u.dst.metrics[RTAX_HOPLIMIT-1] = ra_msg->icmph.icmp6_hop_limit;
+			rt->dst.metrics[RTAX_HOPLIMIT-1] = ra_msg->icmph.icmp6_hop_limit;
 	}
 
 skip_defrtr:
@@ -1363,7 +1363,7 @@ skip_linkparms:
 			in6_dev->cnf.mtu6 = mtu;
 
 			if (rt)
-				rt->u.dst.metrics[RTAX_MTU-1] = mtu;
+				rt->dst.metrics[RTAX_MTU-1] = mtu;
 
 			rt6_mtu_change(skb->dev, mtu);
 		}
@@ -1384,7 +1384,7 @@ skip_linkparms:
 	}
 out:
 	if (rt)
-		dst_release(&rt->u.dst);
+		dst_release(&rt->dst);
 	else if (neigh)
 		neigh_release(neigh);
 	in6_dev_put(in6_dev);
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 968b96490729..e677937a07fc 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -611,23 +611,23 @@ static int rawv6_send_hdrinc(struct sock *sk, void *from, int length,
 	int err;
 	struct rt6_info *rt = (struct rt6_info *)*dstp;
 
-	if (length > rt->u.dst.dev->mtu) {
-		ipv6_local_error(sk, EMSGSIZE, fl, rt->u.dst.dev->mtu);
+	if (length > rt->dst.dev->mtu) {
+		ipv6_local_error(sk, EMSGSIZE, fl, rt->dst.dev->mtu);
 		return -EMSGSIZE;
 	}
 	if (flags&MSG_PROBE)
 		goto out;
 
 	skb = sock_alloc_send_skb(sk,
-				  length + LL_ALLOCATED_SPACE(rt->u.dst.dev) + 15,
+				  length + LL_ALLOCATED_SPACE(rt->dst.dev) + 15,
 				  flags & MSG_DONTWAIT, &err);
 	if (skb == NULL)
 		goto error;
-	skb_reserve(skb, LL_RESERVED_SPACE(rt->u.dst.dev));
+	skb_reserve(skb, LL_RESERVED_SPACE(rt->dst.dev));
 
 	skb->priority = sk->sk_priority;
 	skb->mark = sk->sk_mark;
-	skb_dst_set(skb, &rt->u.dst);
+	skb_dst_set(skb, &rt->dst);
 	*dstp = NULL;
 
 	skb_put(skb, length);
@@ -643,7 +643,7 @@ static int rawv6_send_hdrinc(struct sock *sk, void *from, int length,
 
 	IP6_UPD_PO_STATS(sock_net(sk), rt->rt6i_idev, IPSTATS_MIB_OUT, skb->len);
 	err = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, skb, NULL,
-		      rt->u.dst.dev, dst_output);
+		      rt->dst.dev, dst_output);
 	if (err > 0)
 		err = net_xmit_errno(err);
 	if (err)
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 252d76199c41..f7702850d45c 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -126,16 +126,14 @@ static struct dst_ops ip6_dst_blackhole_ops = {
 };
 
 static struct rt6_info ip6_null_entry_template = {
-	.u = {
-		.dst = {
-			.__refcnt	= ATOMIC_INIT(1),
-			.__use		= 1,
-			.obsolete	= -1,
-			.error		= -ENETUNREACH,
-			.metrics	= { [RTAX_HOPLIMIT - 1] = 255, },
-			.input		= ip6_pkt_discard,
-			.output		= ip6_pkt_discard_out,
-		}
+	.dst = {
+		.__refcnt	= ATOMIC_INIT(1),
+		.__use		= 1,
+		.obsolete	= -1,
+		.error		= -ENETUNREACH,
+		.metrics	= { [RTAX_HOPLIMIT - 1] = 255, },
+		.input		= ip6_pkt_discard,
+		.output		= ip6_pkt_discard_out,
 	},
 	.rt6i_flags	= (RTF_REJECT | RTF_NONEXTHOP),
 	.rt6i_protocol  = RTPROT_KERNEL,
@@ -149,16 +147,14 @@ static int ip6_pkt_prohibit(struct sk_buff *skb);
 static int ip6_pkt_prohibit_out(struct sk_buff *skb);
 
 static struct rt6_info ip6_prohibit_entry_template = {
-	.u = {
-		.dst = {
-			.__refcnt	= ATOMIC_INIT(1),
-			.__use		= 1,
-			.obsolete	= -1,
-			.error		= -EACCES,
-			.metrics	= { [RTAX_HOPLIMIT - 1] = 255, },
-			.input		= ip6_pkt_prohibit,
-			.output		= ip6_pkt_prohibit_out,
-		}
+	.dst = {
+		.__refcnt	= ATOMIC_INIT(1),
+		.__use		= 1,
+		.obsolete	= -1,
+		.error		= -EACCES,
+		.metrics	= { [RTAX_HOPLIMIT - 1] = 255, },
+		.input		= ip6_pkt_prohibit,
+		.output		= ip6_pkt_prohibit_out,
 	},
 	.rt6i_flags	= (RTF_REJECT | RTF_NONEXTHOP),
 	.rt6i_protocol  = RTPROT_KERNEL,
@@ -167,16 +163,14 @@ static struct rt6_info ip6_prohibit_entry_template = {
 };
 
 static struct rt6_info ip6_blk_hole_entry_template = {
-	.u = {
-		.dst = {
-			.__refcnt	= ATOMIC_INIT(1),
-			.__use		= 1,
-			.obsolete	= -1,
-			.error		= -EINVAL,
-			.metrics	= { [RTAX_HOPLIMIT - 1] = 255, },
-			.input		= dst_discard,
-			.output		= dst_discard,
-		}
+	.dst = {
+		.__refcnt	= ATOMIC_INIT(1),
+		.__use		= 1,
+		.obsolete	= -1,
+		.error		= -EINVAL,
+		.metrics	= { [RTAX_HOPLIMIT - 1] = 255, },
+		.input		= dst_discard,
+		.output		= dst_discard,
 	},
 	.rt6i_flags	= (RTF_REJECT | RTF_NONEXTHOP),
 	.rt6i_protocol  = RTPROT_KERNEL,
@@ -249,7 +243,7 @@ static inline struct rt6_info *rt6_device_match(struct net *net,
 	if (!oif && ipv6_addr_any(saddr))
 		goto out;
 
-	for (sprt = rt; sprt; sprt = sprt->u.dst.rt6_next) {
+	for (sprt = rt; sprt; sprt = sprt->dst.rt6_next) {
 		struct net_device *dev = sprt->rt6i_dev;
 
 		if (oif) {
@@ -407,10 +401,10 @@ static struct rt6_info *find_rr_leaf(struct fib6_node *fn,
 
 	match = NULL;
 	for (rt = rr_head; rt && rt->rt6i_metric == metric;
-	     rt = rt->u.dst.rt6_next)
+	     rt = rt->dst.rt6_next)
 		match = find_match(rt, oif, strict, &mpri, match);
 	for (rt = fn->leaf; rt && rt != rr_head && rt->rt6i_metric == metric;
-	     rt = rt->u.dst.rt6_next)
+	     rt = rt->dst.rt6_next)
 		match = find_match(rt, oif, strict, &mpri, match);
 
 	return match;
@@ -432,7 +426,7 @@ static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict)
 
 	if (!match &&
 	    (strict & RT6_LOOKUP_F_REACHABLE)) {
-		struct rt6_info *next = rt0->u.dst.rt6_next;
+		struct rt6_info *next = rt0->dst.rt6_next;
 
 		/* no entries matched; do round-robin */
 		if (!next || next->rt6i_metric != rt0->rt6i_metric)
@@ -517,7 +511,7 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
 			rt->rt6i_expires = jiffies + HZ * lifetime;
 			rt->rt6i_flags |= RTF_EXPIRES;
 		}
-		dst_release(&rt->u.dst);
+		dst_release(&rt->dst);
 	}
 	return 0;
 }
@@ -555,7 +549,7 @@ restart:
 	rt = rt6_device_match(net, rt, &fl->fl6_src, fl->oif, flags);
 	BACKTRACK(net, &fl->fl6_src);
 out:
-	dst_use(&rt->u.dst, jiffies);
+	dst_use(&rt->dst, jiffies);
 	read_unlock_bh(&table->tb6_lock);
 	return rt;
 
@@ -643,7 +637,7 @@ static struct rt6_info *rt6_alloc_cow(struct rt6_info *ort, struct in6_addr *dad
 		ipv6_addr_copy(&rt->rt6i_dst.addr, daddr);
 		rt->rt6i_dst.plen = 128;
 		rt->rt6i_flags |= RTF_CACHE;
-		rt->u.dst.flags |= DST_HOST;
+		rt->dst.flags |= DST_HOST;
 
 #ifdef CONFIG_IPV6_SUBTREES
 		if (rt->rt6i_src.plen && saddr) {
@@ -677,7 +671,7 @@ static struct rt6_info *rt6_alloc_cow(struct rt6_info *ort, struct in6_addr *dad
 			if (net_ratelimit())
 				printk(KERN_WARNING
 				       "Neighbour table overflow.\n");
-			dst_free(&rt->u.dst);
+			dst_free(&rt->dst);
 			return NULL;
 		}
 		rt->rt6i_nexthop = neigh;
@@ -694,7 +688,7 @@ static struct rt6_info *rt6_alloc_clone(struct rt6_info *ort, struct in6_addr *d
 		ipv6_addr_copy(&rt->rt6i_dst.addr, daddr);
 		rt->rt6i_dst.plen = 128;
 		rt->rt6i_flags |= RTF_CACHE;
-		rt->u.dst.flags |= DST_HOST;
+		rt->dst.flags |= DST_HOST;
 		rt->rt6i_nexthop = neigh_clone(ort->rt6i_nexthop);
 	}
 	return rt;
@@ -726,7 +720,7 @@ restart:
 	    rt->rt6i_flags & RTF_CACHE)
 		goto out;
 
-	dst_hold(&rt->u.dst);
+	dst_hold(&rt->dst);
 	read_unlock_bh(&table->tb6_lock);
 
 	if (!rt->rt6i_nexthop && !(rt->rt6i_flags & RTF_NONEXTHOP))
@@ -739,10 +733,10 @@ restart:
 #endif
 	}
 
-	dst_release(&rt->u.dst);
+	dst_release(&rt->dst);
 	rt = nrt ? : net->ipv6.ip6_null_entry;
 
-	dst_hold(&rt->u.dst);
+	dst_hold(&rt->dst);
 	if (nrt) {
 		err = ip6_ins_rt(nrt);
 		if (!err)
@@ -756,7 +750,7 @@ restart:
 	 * Race condition! In the gap, when table->tb6_lock was
 	 * released someone could insert this route.  Relookup.
 	 */
-	dst_release(&rt->u.dst);
+	dst_release(&rt->dst);
 	goto relookup;
 
 out:
@@ -764,11 +758,11 @@ out:
 		reachable = 0;
 		goto restart_2;
 	}
-	dst_hold(&rt->u.dst);
+	dst_hold(&rt->dst);
 	read_unlock_bh(&table->tb6_lock);
 out2:
-	rt->u.dst.lastuse = jiffies;
-	rt->u.dst.__use++;
+	rt->dst.lastuse = jiffies;
+	rt->dst.__use++;
 
 	return rt;
 }
@@ -835,15 +829,15 @@ int ip6_dst_blackhole(struct sock *sk, struct dst_entry **dstp, struct flowi *fl
 	struct dst_entry *new = NULL;
 
 	if (rt) {
-		new = &rt->u.dst;
+		new = &rt->dst;
 
 		atomic_set(&new->__refcnt, 1);
 		new->__use = 1;
 		new->input = dst_discard;
 		new->output = dst_discard;
 
-		memcpy(new->metrics, ort->u.dst.metrics, RTAX_MAX*sizeof(u32));
-		new->dev = ort->u.dst.dev;
+		memcpy(new->metrics, ort->dst.metrics, RTAX_MAX*sizeof(u32));
+		new->dev = ort->dst.dev;
 		if (new->dev)
 			dev_hold(new->dev);
 		rt->rt6i_idev = ort->rt6i_idev;
@@ -912,7 +906,7 @@ static void ip6_link_failure(struct sk_buff *skb)
 	rt = (struct rt6_info *) skb_dst(skb);
 	if (rt) {
 		if (rt->rt6i_flags&RTF_CACHE) {
-			dst_set_expires(&rt->u.dst, 0);
+			dst_set_expires(&rt->dst, 0);
 			rt->rt6i_flags |= RTF_EXPIRES;
 		} else if (rt->rt6i_node && (rt->rt6i_flags & RTF_DEFAULT))
 			rt->rt6i_node->fn_sernum = -1;
@@ -986,14 +980,14 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
 	rt->rt6i_dev	  = dev;
 	rt->rt6i_idev     = idev;
 	rt->rt6i_nexthop  = neigh;
-	atomic_set(&rt->u.dst.__refcnt, 1);
-	rt->u.dst.metrics[RTAX_HOPLIMIT-1] = 255;
-	rt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(rt->rt6i_dev);
-	rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, dst_mtu(&rt->u.dst));
-	rt->u.dst.output  = ip6_output;
+	atomic_set(&rt->dst.__refcnt, 1);
+	rt->dst.metrics[RTAX_HOPLIMIT-1] = 255;
+	rt->dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(rt->rt6i_dev);
+	rt->dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, dst_mtu(&rt->dst));
+	rt->dst.output  = ip6_output;
 
 #if 0	/* there's no chance to use these for ndisc */
-	rt->u.dst.flags   = ipv6_addr_type(addr) & IPV6_ADDR_UNICAST
+	rt->dst.flags   = ipv6_addr_type(addr) & IPV6_ADDR_UNICAST
 				? DST_HOST
 				: 0;
 	ipv6_addr_copy(&rt->rt6i_dst.addr, addr);
@@ -1001,14 +995,14 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev,
 #endif
 
 	spin_lock_bh(&icmp6_dst_lock);
-	rt->u.dst.next = icmp6_dst_gc_list;
-	icmp6_dst_gc_list = &rt->u.dst;
+	rt->dst.next = icmp6_dst_gc_list;
+	icmp6_dst_gc_list = &rt->dst;
 	spin_unlock_bh(&icmp6_dst_lock);
 
 	fib6_force_start_gc(net);
 
 out:
-	return &rt->u.dst;
+	return &rt->dst;
 }
 
 int icmp6_dst_gc(void)
@@ -1159,7 +1153,7 @@ int ip6_route_add(struct fib6_config *cfg)
 		goto out;
 	}
 
-	rt->u.dst.obsolete = -1;
+	rt->dst.obsolete = -1;
 	rt->rt6i_expires = (cfg->fc_flags & RTF_EXPIRES) ?
 				jiffies + clock_t_to_jiffies(cfg->fc_expires) :
 				0;
@@ -1171,16 +1165,16 @@ int ip6_route_add(struct fib6_config *cfg)
 	addr_type = ipv6_addr_type(&cfg->fc_dst);
 
 	if (addr_type & IPV6_ADDR_MULTICAST)
-		rt->u.dst.input = ip6_mc_input;
+		rt->dst.input = ip6_mc_input;
 	else
-		rt->u.dst.input = ip6_forward;
+		rt->dst.input = ip6_forward;
 
-	rt->u.dst.output = ip6_output;
+	rt->dst.output = ip6_output;
 
 	ipv6_addr_prefix(&rt->rt6i_dst.addr, &cfg->fc_dst, cfg->fc_dst_len);
 	rt->rt6i_dst.plen = cfg->fc_dst_len;
 	if (rt->rt6i_dst.plen == 128)
-	       rt->u.dst.flags = DST_HOST;
+	       rt->dst.flags = DST_HOST;
 
 #ifdef CONFIG_IPV6_SUBTREES
 	ipv6_addr_prefix(&rt->rt6i_src.addr, &cfg->fc_src, cfg->fc_src_len);
@@ -1208,9 +1202,9 @@ int ip6_route_add(struct fib6_config *cfg)
 				goto out;
 			}
 		}
-		rt->u.dst.output = ip6_pkt_discard_out;
-		rt->u.dst.input = ip6_pkt_discard;
-		rt->u.dst.error = -ENETUNREACH;
+		rt->dst.output = ip6_pkt_discard_out;
+		rt->dst.input = ip6_pkt_discard;
+		rt->dst.error = -ENETUNREACH;
 		rt->rt6i_flags = RTF_REJECT|RTF_NONEXTHOP;
 		goto install_route;
 	}
@@ -1244,7 +1238,7 @@ int ip6_route_add(struct fib6_config *cfg)
 				goto out;
 			if (dev) {
 				if (dev != grt->rt6i_dev) {
-					dst_release(&grt->u.dst);
+					dst_release(&grt->dst);
 					goto out;
 				}
 			} else {
@@ -1255,7 +1249,7 @@ int ip6_route_add(struct fib6_config *cfg)
 			}
 			if (!(grt->rt6i_flags&RTF_GATEWAY))
 				err = 0;
-			dst_release(&grt->u.dst);
+			dst_release(&grt->dst);
 
 			if (err)
 				goto out;
@@ -1294,18 +1288,18 @@ install_route:
 					goto out;
 				}
 
-				rt->u.dst.metrics[type - 1] = nla_get_u32(nla);
+				rt->dst.metrics[type - 1] = nla_get_u32(nla);
 			}
 		}
 	}
 
-	if (dst_metric(&rt->u.dst, RTAX_HOPLIMIT) == 0)
-		rt->u.dst.metrics[RTAX_HOPLIMIT-1] = -1;
-	if (!dst_mtu(&rt->u.dst))
-		rt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(dev);
-	if (!dst_metric(&rt->u.dst, RTAX_ADVMSS))
-		rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, dst_mtu(&rt->u.dst));
-	rt->u.dst.dev = dev;
+	if (dst_metric(&rt->dst, RTAX_HOPLIMIT) == 0)
+		rt->dst.metrics[RTAX_HOPLIMIT-1] = -1;
+	if (!dst_mtu(&rt->dst))
+		rt->dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(dev);
+	if (!dst_metric(&rt->dst, RTAX_ADVMSS))
+		rt->dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, dst_mtu(&rt->dst));
+	rt->dst.dev = dev;
 	rt->rt6i_idev = idev;
 	rt->rt6i_table = table;
 
@@ -1319,7 +1313,7 @@ out:
 	if (idev)
 		in6_dev_put(idev);
 	if (rt)
-		dst_free(&rt->u.dst);
+		dst_free(&rt->dst);
 	return err;
 }
 
@@ -1336,7 +1330,7 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info)
 	write_lock_bh(&table->tb6_lock);
 
 	err = fib6_del(rt, info);
-	dst_release(&rt->u.dst);
+	dst_release(&rt->dst);
 
 	write_unlock_bh(&table->tb6_lock);
 
@@ -1369,7 +1363,7 @@ static int ip6_route_del(struct fib6_config *cfg)
 			 &cfg->fc_src, cfg->fc_src_len);
 
 	if (fn) {
-		for (rt = fn->leaf; rt; rt = rt->u.dst.rt6_next) {
+		for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) {
 			if (cfg->fc_ifindex &&
 			    (rt->rt6i_dev == NULL ||
 			     rt->rt6i_dev->ifindex != cfg->fc_ifindex))
@@ -1379,7 +1373,7 @@ static int ip6_route_del(struct fib6_config *cfg)
 				continue;
 			if (cfg->fc_metric && cfg->fc_metric != rt->rt6i_metric)
 				continue;
-			dst_hold(&rt->u.dst);
+			dst_hold(&rt->dst);
 			read_unlock_bh(&table->tb6_lock);
 
 			return __ip6_del_rt(rt, &cfg->fc_nlinfo);
@@ -1421,7 +1415,7 @@ static struct rt6_info *__ip6_route_redirect(struct net *net,
 	read_lock_bh(&table->tb6_lock);
 	fn = fib6_lookup(&table->tb6_root, &fl->fl6_dst, &fl->fl6_src);
 restart:
-	for (rt = fn->leaf; rt; rt = rt->u.dst.rt6_next) {
+	for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) {
 		/*
 		 * Current route is on-link; redirect is always invalid.
 		 *
@@ -1445,7 +1439,7 @@ restart:
 		rt = net->ipv6.ip6_null_entry;
 	BACKTRACK(net, &fl->fl6_src);
 out:
-	dst_hold(&rt->u.dst);
+	dst_hold(&rt->dst);
 
 	read_unlock_bh(&table->tb6_lock);
 
@@ -1513,10 +1507,10 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *src,
 	 * Look, redirects are sent only in response to data packets,
 	 * so that this nexthop apparently is reachable. --ANK
 	 */
-	dst_confirm(&rt->u.dst);
+	dst_confirm(&rt->dst);
 
 	/* Duplicate redirect: silently ignore. */
-	if (neigh == rt->u.dst.neighbour)
+	if (neigh == rt->dst.neighbour)
 		goto out;
 
 	nrt = ip6_rt_copy(rt);
@@ -1529,20 +1523,20 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *src,
 
 	ipv6_addr_copy(&nrt->rt6i_dst.addr, dest);
 	nrt->rt6i_dst.plen = 128;
-	nrt->u.dst.flags |= DST_HOST;
+	nrt->dst.flags |= DST_HOST;
 
 	ipv6_addr_copy(&nrt->rt6i_gateway, (struct in6_addr*)neigh->primary_key);
 	nrt->rt6i_nexthop = neigh_clone(neigh);
 	/* Reset pmtu, it may be better */
-	nrt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(neigh->dev);
-	nrt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dev_net(neigh->dev),
-							dst_mtu(&nrt->u.dst));
+	nrt->dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(neigh->dev);
+	nrt->dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dev_net(neigh->dev),
+							dst_mtu(&nrt->dst));
 
 	if (ip6_ins_rt(nrt))
 		goto out;
 
-	netevent.old = &rt->u.dst;
-	netevent.new = &nrt->u.dst;
+	netevent.old = &rt->dst;
+	netevent.new = &nrt->dst;
 	call_netevent_notifiers(NETEVENT_REDIRECT, &netevent);
 
 	if (rt->rt6i_flags&RTF_CACHE) {
@@ -1551,7 +1545,7 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *src,
 	}
 
 out:
-	dst_release(&rt->u.dst);
+	dst_release(&rt->dst);
 }
 
 /*
@@ -1570,7 +1564,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
 	if (rt == NULL)
 		return;
 
-	if (pmtu >= dst_mtu(&rt->u.dst))
+	if (pmtu >= dst_mtu(&rt->dst))
 		goto out;
 
 	if (pmtu < IPV6_MIN_MTU) {
@@ -1588,7 +1582,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
 	   They are sent only in response to data packets,
 	   so that this nexthop apparently is reachable. --ANK
 	 */
-	dst_confirm(&rt->u.dst);
+	dst_confirm(&rt->dst);
 
 	/* Host route. If it is static, it would be better
 	   not to override it, but add new one, so that
@@ -1596,10 +1590,10 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
 	   would return automatically.
 	 */
 	if (rt->rt6i_flags & RTF_CACHE) {
-		rt->u.dst.metrics[RTAX_MTU-1] = pmtu;
+		rt->dst.metrics[RTAX_MTU-1] = pmtu;
 		if (allfrag)
-			rt->u.dst.metrics[RTAX_FEATURES-1] |= RTAX_FEATURE_ALLFRAG;
-		dst_set_expires(&rt->u.dst, net->ipv6.sysctl.ip6_rt_mtu_expires);
+			rt->dst.metrics[RTAX_FEATURES-1] |= RTAX_FEATURE_ALLFRAG;
+		dst_set_expires(&rt->dst, net->ipv6.sysctl.ip6_rt_mtu_expires);
 		rt->rt6i_flags |= RTF_MODIFIED|RTF_EXPIRES;
 		goto out;
 	}
@@ -1615,9 +1609,9 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
 		nrt = rt6_alloc_clone(rt, daddr);
 
 	if (nrt) {
-		nrt->u.dst.metrics[RTAX_MTU-1] = pmtu;
+		nrt->dst.metrics[RTAX_MTU-1] = pmtu;
 		if (allfrag)
-			nrt->u.dst.metrics[RTAX_FEATURES-1] |= RTAX_FEATURE_ALLFRAG;
+			nrt->dst.metrics[RTAX_FEATURES-1] |= RTAX_FEATURE_ALLFRAG;
 
 		/* According to RFC 1981, detecting PMTU increase shouldn't be
 		 * happened within 5 mins, the recommended timer is 10 mins.
@@ -1625,13 +1619,13 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr,
 		 * which is 10 mins. After 10 mins the decreased pmtu is expired
 		 * and detecting PMTU increase will be automatically happened.
 		 */
-		dst_set_expires(&nrt->u.dst, net->ipv6.sysctl.ip6_rt_mtu_expires);
+		dst_set_expires(&nrt->dst, net->ipv6.sysctl.ip6_rt_mtu_expires);
 		nrt->rt6i_flags |= RTF_DYNAMIC|RTF_EXPIRES;
 
 		ip6_ins_rt(nrt);
 	}
 out:
-	dst_release(&rt->u.dst);
+	dst_release(&rt->dst);
 }
 
 /*
@@ -1644,18 +1638,18 @@ static struct rt6_info * ip6_rt_copy(struct rt6_info *ort)
 	struct rt6_info *rt = ip6_dst_alloc(&net->ipv6.ip6_dst_ops);
 
 	if (rt) {
-		rt->u.dst.input = ort->u.dst.input;
-		rt->u.dst.output = ort->u.dst.output;
-
-		memcpy(rt->u.dst.metrics, ort->u.dst.metrics, RTAX_MAX*sizeof(u32));
-		rt->u.dst.error = ort->u.dst.error;
-		rt->u.dst.dev = ort->u.dst.dev;
-		if (rt->u.dst.dev)
-			dev_hold(rt->u.dst.dev);
+		rt->dst.input = ort->dst.input;
+		rt->dst.output = ort->dst.output;
+
+		memcpy(rt->dst.metrics, ort->dst.metrics, RTAX_MAX*sizeof(u32));
+		rt->dst.error = ort->dst.error;
+		rt->dst.dev = ort->dst.dev;
+		if (rt->dst.dev)
+			dev_hold(rt->dst.dev);
 		rt->rt6i_idev = ort->rt6i_idev;
 		if (rt->rt6i_idev)
 			in6_dev_hold(rt->rt6i_idev);
-		rt->u.dst.lastuse = jiffies;
+		rt->dst.lastuse = jiffies;
 		rt->rt6i_expires = 0;
 
 		ipv6_addr_copy(&rt->rt6i_gateway, &ort->rt6i_gateway);
@@ -1689,14 +1683,14 @@ static struct rt6_info *rt6_get_route_info(struct net *net,
 	if (!fn)
 		goto out;
 
-	for (rt = fn->leaf; rt; rt = rt->u.dst.rt6_next) {
+	for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) {
 		if (rt->rt6i_dev->ifindex != ifindex)
 			continue;
 		if ((rt->rt6i_flags & (RTF_ROUTEINFO|RTF_GATEWAY)) != (RTF_ROUTEINFO|RTF_GATEWAY))
 			continue;
 		if (!ipv6_addr_equal(&rt->rt6i_gateway, gwaddr))
 			continue;
-		dst_hold(&rt->u.dst);
+		dst_hold(&rt->dst);
 		break;
 	}
 out:
@@ -1744,14 +1738,14 @@ struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *d
 		return NULL;
 
 	write_lock_bh(&table->tb6_lock);
-	for (rt = table->tb6_root.leaf; rt; rt=rt->u.dst.rt6_next) {
+	for (rt = table->tb6_root.leaf; rt; rt=rt->dst.rt6_next) {
 		if (dev == rt->rt6i_dev &&
 		    ((rt->rt6i_flags & (RTF_ADDRCONF | RTF_DEFAULT)) == (RTF_ADDRCONF | RTF_DEFAULT)) &&
 		    ipv6_addr_equal(&rt->rt6i_gateway, addr))
 			break;
 	}
 	if (rt)
-		dst_hold(&rt->u.dst);
+		dst_hold(&rt->dst);
 	write_unlock_bh(&table->tb6_lock);
 	return rt;
 }
@@ -1790,9 +1784,9 @@ void rt6_purge_dflt_routers(struct net *net)
 
 restart:
 	read_lock_bh(&table->tb6_lock);
-	for (rt = table->tb6_root.leaf; rt; rt = rt->u.dst.rt6_next) {
+	for (rt = table->tb6_root.leaf; rt; rt = rt->dst.rt6_next) {
 		if (rt->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF)) {
-			dst_hold(&rt->u.dst);
+			dst_hold(&rt->dst);
 			read_unlock_bh(&table->tb6_lock);
 			ip6_del_rt(rt);
 			goto restart;
@@ -1930,15 +1924,15 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
 	dev_hold(net->loopback_dev);
 	in6_dev_hold(idev);
 
-	rt->u.dst.flags = DST_HOST;
-	rt->u.dst.input = ip6_input;
-	rt->u.dst.output = ip6_output;
+	rt->dst.flags = DST_HOST;
+	rt->dst.input = ip6_input;
+	rt->dst.output = ip6_output;
 	rt->rt6i_dev = net->loopback_dev;
 	rt->rt6i_idev = idev;
-	rt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(rt->rt6i_dev);
-	rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, dst_mtu(&rt->u.dst));
-	rt->u.dst.metrics[RTAX_HOPLIMIT-1] = -1;
-	rt->u.dst.obsolete = -1;
+	rt->dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(rt->rt6i_dev);
+	rt->dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, dst_mtu(&rt->dst));
+	rt->dst.metrics[RTAX_HOPLIMIT-1] = -1;
+	rt->dst.obsolete = -1;
 
 	rt->rt6i_flags = RTF_UP | RTF_NONEXTHOP;
 	if (anycast)
@@ -1947,7 +1941,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
 		rt->rt6i_flags |= RTF_LOCAL;
 	neigh = ndisc_get_neigh(rt->rt6i_dev, &rt->rt6i_gateway);
 	if (IS_ERR(neigh)) {
-		dst_free(&rt->u.dst);
+		dst_free(&rt->dst);
 
 		/* We are casting this because that is the return
 		 * value type.  But an errno encoded pointer is the
@@ -1962,7 +1956,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
 	rt->rt6i_dst.plen = 128;
 	rt->rt6i_table = fib6_get_table(net, RT6_TABLE_LOCAL);
 
-	atomic_set(&rt->u.dst.__refcnt, 1);
+	atomic_set(&rt->dst.__refcnt, 1);
 
 	return rt;
 }
@@ -2033,12 +2027,12 @@ static int rt6_mtu_change_route(struct rt6_info *rt, void *p_arg)
 	   PMTU discouvery.
 	 */
 	if (rt->rt6i_dev == arg->dev &&
-	    !dst_metric_locked(&rt->u.dst, RTAX_MTU) &&
-	    (dst_mtu(&rt->u.dst) >= arg->mtu ||
-	     (dst_mtu(&rt->u.dst) < arg->mtu &&
-	      dst_mtu(&rt->u.dst) == idev->cnf.mtu6))) {
-		rt->u.dst.metrics[RTAX_MTU-1] = arg->mtu;
-		rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, arg->mtu);
+	    !dst_metric_locked(&rt->dst, RTAX_MTU) &&
+	    (dst_mtu(&rt->dst) >= arg->mtu ||
+	     (dst_mtu(&rt->dst) < arg->mtu &&
+	      dst_mtu(&rt->dst) == idev->cnf.mtu6))) {
+		rt->dst.metrics[RTAX_MTU-1] = arg->mtu;
+		rt->dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, arg->mtu);
 	}
 	return 0;
 }
@@ -2252,20 +2246,20 @@ static int rt6_fill_node(struct net *net,
 #endif
 			NLA_PUT_U32(skb, RTA_IIF, iif);
 	} else if (dst) {
-		struct inet6_dev *idev = ip6_dst_idev(&rt->u.dst);
+		struct inet6_dev *idev = ip6_dst_idev(&rt->dst);
 		struct in6_addr saddr_buf;
 		if (ipv6_dev_get_saddr(net, idev ? idev->dev : NULL,
 				       dst, 0, &saddr_buf) == 0)
 			NLA_PUT(skb, RTA_PREFSRC, 16, &saddr_buf);
 	}
 
-	if (rtnetlink_put_metrics(skb, rt->u.dst.metrics) < 0)
+	if (rtnetlink_put_metrics(skb, rt->dst.metrics) < 0)
 		goto nla_put_failure;
 
-	if (rt->u.dst.neighbour)
-		NLA_PUT(skb, RTA_GATEWAY, 16, &rt->u.dst.neighbour->primary_key);
+	if (rt->dst.neighbour)
+		NLA_PUT(skb, RTA_GATEWAY, 16, &rt->dst.neighbour->primary_key);
 
-	if (rt->u.dst.dev)
+	if (rt->dst.dev)
 		NLA_PUT_U32(skb, RTA_OIF, rt->rt6i_dev->ifindex);
 
 	NLA_PUT_U32(skb, RTA_PRIORITY, rt->rt6i_metric);
@@ -2277,8 +2271,8 @@ static int rt6_fill_node(struct net *net,
 	else
 		expires = INT_MAX;
 
-	if (rtnl_put_cacheinfo(skb, &rt->u.dst, 0, 0, 0,
-			       expires, rt->u.dst.error) < 0)
+	if (rtnl_put_cacheinfo(skb, &rt->dst, 0, 0, 0,
+			       expires, rt->dst.error) < 0)
 		goto nla_put_failure;
 
 	return nlmsg_end(skb, nlh);
@@ -2364,7 +2358,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void
 	skb_reserve(skb, MAX_HEADER + sizeof(struct ipv6hdr));
 
 	rt = (struct rt6_info*) ip6_route_output(net, NULL, &fl);
-	skb_dst_set(skb, &rt->u.dst);
+	skb_dst_set(skb, &rt->dst);
 
 	err = rt6_fill_node(net, skb, rt, &fl.fl6_dst, &fl.fl6_src, iif,
 			    RTM_NEWROUTE, NETLINK_CB(in_skb).pid,
@@ -2416,12 +2410,12 @@ static int ip6_route_dev_notify(struct notifier_block *this,
 	struct net *net = dev_net(dev);
 
 	if (event == NETDEV_REGISTER && (dev->flags & IFF_LOOPBACK)) {
-		net->ipv6.ip6_null_entry->u.dst.dev = dev;
+		net->ipv6.ip6_null_entry->dst.dev = dev;
 		net->ipv6.ip6_null_entry->rt6i_idev = in6_dev_get(dev);
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
-		net->ipv6.ip6_prohibit_entry->u.dst.dev = dev;
+		net->ipv6.ip6_prohibit_entry->dst.dev = dev;
 		net->ipv6.ip6_prohibit_entry->rt6i_idev = in6_dev_get(dev);
-		net->ipv6.ip6_blk_hole_entry->u.dst.dev = dev;
+		net->ipv6.ip6_blk_hole_entry->dst.dev = dev;
 		net->ipv6.ip6_blk_hole_entry->rt6i_idev = in6_dev_get(dev);
 #endif
 	}
@@ -2464,8 +2458,8 @@ static int rt6_info_route(struct rt6_info *rt, void *p_arg)
 		seq_puts(m, "00000000000000000000000000000000");
 	}
 	seq_printf(m, " %08x %08x %08x %08x %8s\n",
-		   rt->rt6i_metric, atomic_read(&rt->u.dst.__refcnt),
-		   rt->u.dst.__use, rt->rt6i_flags,
+		   rt->rt6i_metric, atomic_read(&rt->dst.__refcnt),
+		   rt->dst.__use, rt->rt6i_flags,
 		   rt->rt6i_dev ? rt->rt6i_dev->name : "");
 	return 0;
 }
@@ -2646,9 +2640,9 @@ static int __net_init ip6_route_net_init(struct net *net)
 					   GFP_KERNEL);
 	if (!net->ipv6.ip6_null_entry)
 		goto out_ip6_dst_ops;
-	net->ipv6.ip6_null_entry->u.dst.path =
+	net->ipv6.ip6_null_entry->dst.path =
 		(struct dst_entry *)net->ipv6.ip6_null_entry;
-	net->ipv6.ip6_null_entry->u.dst.ops = &net->ipv6.ip6_dst_ops;
+	net->ipv6.ip6_null_entry->dst.ops = &net->ipv6.ip6_dst_ops;
 
 #ifdef CONFIG_IPV6_MULTIPLE_TABLES
 	net->ipv6.ip6_prohibit_entry = kmemdup(&ip6_prohibit_entry_template,
@@ -2656,18 +2650,18 @@ static int __net_init ip6_route_net_init(struct net *net)
 					       GFP_KERNEL);
 	if (!net->ipv6.ip6_prohibit_entry)
 		goto out_ip6_null_entry;
-	net->ipv6.ip6_prohibit_entry->u.dst.path =
+	net->ipv6.ip6_prohibit_entry->dst.path =
 		(struct dst_entry *)net->ipv6.ip6_prohibit_entry;
-	net->ipv6.ip6_prohibit_entry->u.dst.ops = &net->ipv6.ip6_dst_ops;
+	net->ipv6.ip6_prohibit_entry->dst.ops = &net->ipv6.ip6_dst_ops;
 
 	net->ipv6.ip6_blk_hole_entry = kmemdup(&ip6_blk_hole_entry_template,
 					       sizeof(*net->ipv6.ip6_blk_hole_entry),
 					       GFP_KERNEL);
 	if (!net->ipv6.ip6_blk_hole_entry)
 		goto out_ip6_prohibit_entry;
-	net->ipv6.ip6_blk_hole_entry->u.dst.path =
+	net->ipv6.ip6_blk_hole_entry->dst.path =
 		(struct dst_entry *)net->ipv6.ip6_blk_hole_entry;
-	net->ipv6.ip6_blk_hole_entry->u.dst.ops = &net->ipv6.ip6_dst_ops;
+	net->ipv6.ip6_blk_hole_entry->dst.ops = &net->ipv6.ip6_dst_ops;
 #endif
 
 	net->ipv6.sysctl.flush_delay = 0;
@@ -2742,12 +2736,12 @@ int __init ip6_route_init(void)
 	/* Registering of the loopback is done before this portion of code,
 	 * the loopback reference in rt6_info will not be taken, do it
 	 * manually for init_net */
-	init_net.ipv6.ip6_null_entry->u.dst.dev = init_net.loopback_dev;
+	init_net.ipv6.ip6_null_entry->dst.dev = init_net.loopback_dev;
 	init_net.ipv6.ip6_null_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
   #ifdef CONFIG_IPV6_MULTIPLE_TABLES
-	init_net.ipv6.ip6_prohibit_entry->u.dst.dev = init_net.loopback_dev;
+	init_net.ipv6.ip6_prohibit_entry->dst.dev = init_net.loopback_dev;
 	init_net.ipv6.ip6_prohibit_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
-	init_net.ipv6.ip6_blk_hole_entry->u.dst.dev = init_net.loopback_dev;
+	init_net.ipv6.ip6_blk_hole_entry->dst.dev = init_net.loopback_dev;
 	init_net.ipv6.ip6_blk_hole_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
   #endif
 	ret = fib6_init();
diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c
index 702c532ec21e..4699cd3c3118 100644
--- a/net/ipv6/sit.c
+++ b/net/ipv6/sit.c
@@ -712,7 +712,7 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb,
 		stats->tx_carrier_errors++;
 		goto tx_error_icmp;
 	}
-	tdev = rt->u.dst.dev;
+	tdev = rt->dst.dev;
 
 	if (tdev == dev) {
 		ip_rt_put(rt);
@@ -721,7 +721,7 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb,
 	}
 
 	if (df) {
-		mtu = dst_mtu(&rt->u.dst) - sizeof(struct iphdr);
+		mtu = dst_mtu(&rt->dst) - sizeof(struct iphdr);
 
 		if (mtu < 68) {
 			stats->collisions++;
@@ -780,7 +780,7 @@ static netdev_tx_t ipip6_tunnel_xmit(struct sk_buff *skb,
 	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
 	IPCB(skb)->flags = 0;
 	skb_dst_drop(skb);
-	skb_dst_set(skb, &rt->u.dst);
+	skb_dst_set(skb, &rt->dst);
 
 	/*
 	 *	Push down and install the IPIP header.
@@ -829,7 +829,7 @@ static void ipip6_tunnel_bind_dev(struct net_device *dev)
 				    .proto = IPPROTO_IPV6 };
 		struct rtable *rt;
 		if (!ip_route_output_key(dev_net(dev), &rt, &fl)) {
-			tdev = rt->u.dst.dev;
+			tdev = rt->dst.dev;
 			ip_rt_put(rt);
 		}
 		dev->flags |= IFF_POINTOPOINT;
-- 
cgit v1.2.3


From f6bc7d9e4760324258ad5f5d147e79db8442842e Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 14 Jun 2010 04:39:27 +0000
Subject: ipv6: avoid two atomics in ipv6_rthdr_rcv()

Use __in6_dev_get() instead of in6_dev_get()/in6_dev_put()

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/exthdrs.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c
index 853a633a94d4..262f105d23b9 100644
--- a/net/ipv6/exthdrs.c
+++ b/net/ipv6/exthdrs.c
@@ -312,6 +312,7 @@ static int ipv6_destopt_rcv(struct sk_buff *skb)
   Routing header.
  ********************************/
 
+/* called with rcu_read_lock() */
 static int ipv6_rthdr_rcv(struct sk_buff *skb)
 {
 	struct inet6_skb_parm *opt = IP6CB(skb);
@@ -324,12 +325,9 @@ static int ipv6_rthdr_rcv(struct sk_buff *skb)
 	struct net *net = dev_net(skb->dev);
 	int accept_source_route = net->ipv6.devconf_all->accept_source_route;
 
-	idev = in6_dev_get(skb->dev);
-	if (idev) {
-		if (accept_source_route > idev->cnf.accept_source_route)
-			accept_source_route = idev->cnf.accept_source_route;
-		in6_dev_put(idev);
-	}
+	idev = __in6_dev_get(skb->dev);
+	if (idev && accept_source_route > idev->cnf.accept_source_route)
+		accept_source_route = idev->cnf.accept_source_route;
 
 	if (!pskb_may_pull(skb, skb_transport_offset(skb) + 8) ||
 	    !pskb_may_pull(skb, (skb_transport_offset(skb) +
-- 
cgit v1.2.3


From c68f24cc354050415c5ea543cd19ea5424463a2f Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 14 Jun 2010 04:46:20 +0000
Subject: ipv6: RCU changes in ipv6_get_mtu() and ip6_dst_hoplimit()

Use RCU to avoid atomic ops on idev refcnt in ipv6_get_mtu()
and ip6_dst_hoplimit()

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/route.c | 19 +++++++++++--------
 1 file changed, 11 insertions(+), 8 deletions(-)

(limited to 'net/ipv6')

diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index f7702850d45c..8f2d0400cf8a 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1084,11 +1084,11 @@ static int ipv6_get_mtu(struct net_device *dev)
 	int mtu = IPV6_MIN_MTU;
 	struct inet6_dev *idev;
 
-	idev = in6_dev_get(dev);
-	if (idev) {
+	rcu_read_lock();
+	idev = __in6_dev_get(dev);
+	if (idev)
 		mtu = idev->cnf.mtu6;
-		in6_dev_put(idev);
-	}
+	rcu_read_unlock();
 	return mtu;
 }
 
@@ -1097,12 +1097,15 @@ int ip6_dst_hoplimit(struct dst_entry *dst)
 	int hoplimit = dst_metric(dst, RTAX_HOPLIMIT);
 	if (hoplimit < 0) {
 		struct net_device *dev = dst->dev;
-		struct inet6_dev *idev = in6_dev_get(dev);
-		if (idev) {
+		struct inet6_dev *idev;
+
+		rcu_read_lock();
+		idev = __in6_dev_get(dev);
+		if (idev)
 			hoplimit = idev->cnf.hop_limit;
-			in6_dev_put(idev);
-		} else
+		else
 			hoplimit = dev_net(dev)->ipv6.devconf_all->hop_limit;
+		rcu_read_unlock();
 	}
 	return hoplimit;
 }
-- 
cgit v1.2.3