Merge branch 'lwt-ipv6'

Jiri Benc says:

====================
lwtunnel: per route ipv6 support for vxlan

v3: Moved LWTUNNEL_ENCAP_IP6 definition in patch 13.
v2: Fixed issues in patch 4 pointed out by Alexei.

This series enables IPv6 tunnels based on lwtunnel infrastructure. Only
vxlan is supported for now.

Tested in all combinations of IPv4 over IPv6, IPv6 over IPv4 and IPv6 over
IPv6.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

3 files changed, 127 insertions, 31 deletions

diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index fb44d693796e..1bf328182697 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -407,10 +407,10 @@ static int ipgre_rcv(struct sk_buff *skb, const struct tnl_ptk_info *tpi)
 				return PACKET_REJECT;
 
 			info = &tun_dst->u.tun_info;
-			info->key.ipv4_src = iph->saddr;
-			info->key.ipv4_dst = iph->daddr;
-			info->key.ipv4_tos = iph->tos;
-			info->key.ipv4_ttl = iph->ttl;
+			info->key.u.ipv4.src = iph->saddr;
+			info->key.u.ipv4.dst = iph->daddr;
+			info->key.tos = iph->tos;
+			info->key.ttl = iph->ttl;
 
 			info->mode = IP_TUNNEL_INFO_RX;
 			info->key.tun_flags = tpi->flags &
@@ -521,15 +521,15 @@ static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev)
 	__be16 df, flags;
 	int err;
 
-	tun_info = skb_tunnel_info(skb, AF_INET);
+	tun_info = skb_tunnel_info(skb);
 	if (unlikely(!tun_info || tun_info->mode != IP_TUNNEL_INFO_TX))
 		goto err_free_skb;
 
 	key = &tun_info->key;
 	memset(&fl, 0, sizeof(fl));
-	fl.daddr = key->ipv4_dst;
-	fl.saddr = key->ipv4_src;
-	fl.flowi4_tos = RT_TOS(key->ipv4_tos);
+	fl.daddr = key->u.ipv4.dst;
+	fl.saddr = key->u.ipv4.src;
+	fl.flowi4_tos = RT_TOS(key->tos);
 	fl.flowi4_mark = skb->mark;
 	fl.flowi4_proto = IPPROTO_GRE;
 
@@ -564,8 +564,8 @@ static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	df = key->tun_flags & TUNNEL_DONT_FRAGMENT ?  htons(IP_DF) : 0;
 	err = iptunnel_xmit(skb->sk, rt, skb, fl.saddr,
-			    key->ipv4_dst, IPPROTO_GRE,
-			    key->ipv4_tos, key->ipv4_ttl, df, false);
+			    key->u.ipv4.dst, IPPROTO_GRE,
+			    key->tos, key->ttl, df, false);
 	iptunnel_xmit_stats(err, &dev->stats, dev->tstats);
 	return;
 
diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c
index 1c2389d582a6..289b6c26ce37 100644
--- a/net/ipv4/ip_tunnel_core.c
+++ b/net/ipv4/ip_tunnel_core.c
@@ -227,16 +227,16 @@ static int ip_tun_build_state(struct net_device *dev, struct nlattr *attr,
 		tun_info->key.tun_id = nla_get_u64(tb[LWTUNNEL_IP_ID]);
 
 	if (tb[LWTUNNEL_IP_DST])
-		tun_info->key.ipv4_dst = nla_get_be32(tb[LWTUNNEL_IP_DST]);
+		tun_info->key.u.ipv4.dst = nla_get_be32(tb[LWTUNNEL_IP_DST]);
 
 	if (tb[LWTUNNEL_IP_SRC])
-		tun_info->key.ipv4_src = nla_get_be32(tb[LWTUNNEL_IP_SRC]);
+		tun_info->key.u.ipv4.src = nla_get_be32(tb[LWTUNNEL_IP_SRC]);
 
 	if (tb[LWTUNNEL_IP_TTL])
-		tun_info->key.ipv4_ttl = nla_get_u8(tb[LWTUNNEL_IP_TTL]);
+		tun_info->key.ttl = nla_get_u8(tb[LWTUNNEL_IP_TTL]);
 
 	if (tb[LWTUNNEL_IP_TOS])
-		tun_info->key.ipv4_tos = nla_get_u8(tb[LWTUNNEL_IP_TOS]);
+		tun_info->key.tos = nla_get_u8(tb[LWTUNNEL_IP_TOS]);
 
 	if (tb[LWTUNNEL_IP_SPORT])
 		tun_info->key.tp_src = nla_get_be16(tb[LWTUNNEL_IP_SPORT]);
@@ -262,10 +262,10 @@ static int ip_tun_fill_encap_info(struct sk_buff *skb,
 	struct ip_tunnel_info *tun_info = lwt_tun_info(lwtstate);
 
 	if (nla_put_u64(skb, LWTUNNEL_IP_ID, tun_info->key.tun_id) ||
-	    nla_put_be32(skb, LWTUNNEL_IP_DST, tun_info->key.ipv4_dst) ||
-	    nla_put_be32(skb, LWTUNNEL_IP_SRC, tun_info->key.ipv4_src) ||
-	    nla_put_u8(skb, LWTUNNEL_IP_TOS, tun_info->key.ipv4_tos) ||
-	    nla_put_u8(skb, LWTUNNEL_IP_TTL, tun_info->key.ipv4_ttl) ||
+	    nla_put_be32(skb, LWTUNNEL_IP_DST, tun_info->key.u.ipv4.dst) ||
+	    nla_put_be32(skb, LWTUNNEL_IP_SRC, tun_info->key.u.ipv4.src) ||
+	    nla_put_u8(skb, LWTUNNEL_IP_TOS, tun_info->key.tos) ||
+	    nla_put_u8(skb, LWTUNNEL_IP_TTL, tun_info->key.ttl) ||
 	    nla_put_u16(skb, LWTUNNEL_IP_SPORT, tun_info->key.tp_src) ||
 	    nla_put_u16(skb, LWTUNNEL_IP_DPORT, tun_info->key.tp_dst) ||
 	    nla_put_u16(skb, LWTUNNEL_IP_FLAGS, tun_info->key.tun_flags))
@@ -299,9 +299,111 @@ static const struct lwtunnel_encap_ops ip_tun_lwt_ops = {
 	.cmp_encap = ip_tun_cmp_encap,
 };
 
+static const struct nla_policy ip6_tun_policy[LWTUNNEL_IP6_MAX + 1] = {
+	[LWTUNNEL_IP6_ID]		= { .type = NLA_U64 },
+	[LWTUNNEL_IP6_DST]		= { .len = sizeof(struct in6_addr) },
+	[LWTUNNEL_IP6_SRC]		= { .len = sizeof(struct in6_addr) },
+	[LWTUNNEL_IP6_HOPLIMIT]		= { .type = NLA_U8 },
+	[LWTUNNEL_IP6_TC]		= { .type = NLA_U8 },
+	[LWTUNNEL_IP6_SPORT]		= { .type = NLA_U16 },
+	[LWTUNNEL_IP6_DPORT]		= { .type = NLA_U16 },
+	[LWTUNNEL_IP6_FLAGS]		= { .type = NLA_U16 },
+};
+
+static int ip6_tun_build_state(struct net_device *dev, struct nlattr *attr,
+			       struct lwtunnel_state **ts)
+{
+	struct ip_tunnel_info *tun_info;
+	struct lwtunnel_state *new_state;
+	struct nlattr *tb[LWTUNNEL_IP6_MAX + 1];
+	int err;
+
+	err = nla_parse_nested(tb, LWTUNNEL_IP6_MAX, attr, ip6_tun_policy);
+	if (err < 0)
+		return err;
+
+	new_state = lwtunnel_state_alloc(sizeof(*tun_info));
+	if (!new_state)
+		return -ENOMEM;
+
+	new_state->type = LWTUNNEL_ENCAP_IP6;
+
+	tun_info = lwt_tun_info(new_state);
+
+	if (tb[LWTUNNEL_IP6_ID])
+		tun_info->key.tun_id = nla_get_u64(tb[LWTUNNEL_IP6_ID]);
+
+	if (tb[LWTUNNEL_IP6_DST])
+		tun_info->key.u.ipv6.dst = nla_get_in6_addr(tb[LWTUNNEL_IP6_DST]);
+
+	if (tb[LWTUNNEL_IP6_SRC])
+		tun_info->key.u.ipv6.src = nla_get_in6_addr(tb[LWTUNNEL_IP6_SRC]);
+
+	if (tb[LWTUNNEL_IP6_HOPLIMIT])
+		tun_info->key.ttl = nla_get_u8(tb[LWTUNNEL_IP6_HOPLIMIT]);
+
+	if (tb[LWTUNNEL_IP6_TC])
+		tun_info->key.tos = nla_get_u8(tb[LWTUNNEL_IP6_TC]);
+
+	if (tb[LWTUNNEL_IP6_SPORT])
+		tun_info->key.tp_src = nla_get_be16(tb[LWTUNNEL_IP6_SPORT]);
+
+	if (tb[LWTUNNEL_IP6_DPORT])
+		tun_info->key.tp_dst = nla_get_be16(tb[LWTUNNEL_IP6_DPORT]);
+
+	if (tb[LWTUNNEL_IP6_FLAGS])
+		tun_info->key.tun_flags = nla_get_u16(tb[LWTUNNEL_IP6_FLAGS]);
+
+	tun_info->mode = IP_TUNNEL_INFO_TX;
+	tun_info->options = NULL;
+	tun_info->options_len = 0;
+
+	*ts = new_state;
+
+	return 0;
+}
+
+static int ip6_tun_fill_encap_info(struct sk_buff *skb,
+				   struct lwtunnel_state *lwtstate)
+{
+	struct ip_tunnel_info *tun_info = lwt_tun_info(lwtstate);
+
+	if (nla_put_u64(skb, LWTUNNEL_IP6_ID, tun_info->key.tun_id) ||
+	    nla_put_in6_addr(skb, LWTUNNEL_IP6_DST, &tun_info->key.u.ipv6.dst) ||
+	    nla_put_in6_addr(skb, LWTUNNEL_IP6_SRC, &tun_info->key.u.ipv6.src) ||
+	    nla_put_u8(skb, LWTUNNEL_IP6_HOPLIMIT, tun_info->key.tos) ||
+	    nla_put_u8(skb, LWTUNNEL_IP6_TC, tun_info->key.ttl) ||
+	    nla_put_u16(skb, LWTUNNEL_IP6_SPORT, tun_info->key.tp_src) ||
+	    nla_put_u16(skb, LWTUNNEL_IP6_DPORT, tun_info->key.tp_dst) ||
+	    nla_put_u16(skb, LWTUNNEL_IP6_FLAGS, tun_info->key.tun_flags))
+		return -ENOMEM;
+
+	return 0;
+}
+
+static int ip6_tun_encap_nlsize(struct lwtunnel_state *lwtstate)
+{
+	return nla_total_size(8)	/* LWTUNNEL_IP6_ID */
+		+ nla_total_size(16)	/* LWTUNNEL_IP6_DST */
+		+ nla_total_size(16)	/* LWTUNNEL_IP6_SRC */
+		+ nla_total_size(1)	/* LWTUNNEL_IP6_HOPLIMIT */
+		+ nla_total_size(1)	/* LWTUNNEL_IP6_TC */
+		+ nla_total_size(2)	/* LWTUNNEL_IP6_SPORT */
+		+ nla_total_size(2)	/* LWTUNNEL_IP6_DPORT */
+		+ nla_total_size(2);	/* LWTUNNEL_IP6_FLAGS */
+}
+
+static const struct lwtunnel_encap_ops ip6_tun_lwt_ops = {
+	.build_state = ip6_tun_build_state,
+	.fill_encap = ip6_tun_fill_encap_info,
+	.get_encap_size = ip6_tun_encap_nlsize,
+	.cmp_encap = ip_tun_cmp_encap,
+};
+
 void __init ip_tunnel_core_init(void)
 {
 	lwtunnel_encap_add_ops(&ip_tun_lwt_ops, LWTUNNEL_ENCAP_IP);
+	lwtunnel_encap_add_ops(&ip6_tun_lwt_ops, LWTUNNEL_ENCAP_IP6);
 }
 
 struct static_key ip_tunnel_metadata_cnt = STATIC_KEY_INIT_FALSE;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 2403e85107f0..f3087aaa6dd8 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1359,7 +1359,6 @@ static void ipv4_dst_destroy(struct dst_entry *dst)
 		list_del(&rt->rt_uncached);
 		spin_unlock_bh(&ul->lock);
 	}
-	lwtstate_put(rt->rt_lwtstate);
 }
 
 void rt_flush_dev(struct net_device *dev)
@@ -1408,7 +1407,7 @@ static void rt_set_nexthop(struct rtable *rt, __be32 daddr,
 #ifdef CONFIG_IP_ROUTE_CLASSID
 		rt->dst.tclassid = nh->nh_tclassid;
 #endif
-		rt->rt_lwtstate = lwtstate_get(nh->nh_lwtstate);
+		rt->dst.lwtstate = lwtstate_get(nh->nh_lwtstate);
 		if (unlikely(fnhe))
 			cached = rt_bind_exception(rt, fnhe, daddr);
 		else if (!(rt->dst.flags & DST_NOCACHE))
@@ -1494,7 +1493,6 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 	rth->rt_gateway	= 0;
 	rth->rt_uses_gateway = 0;
 	INIT_LIST_HEAD(&rth->rt_uncached);
-	rth->rt_lwtstate = NULL;
 	if (our) {
 		rth->dst.input= ip_local_deliver;
 		rth->rt_flags |= RTCF_LOCAL;
@@ -1624,19 +1622,18 @@ static int __mkroute_input(struct sk_buff *skb,
 	rth->rt_gateway	= 0;
 	rth->rt_uses_gateway = 0;
 	INIT_LIST_HEAD(&rth->rt_uncached);
-	rth->rt_lwtstate = NULL;
 	RT_CACHE_STAT_INC(in_slow_tot);
 
 	rth->dst.input = ip_forward;
 	rth->dst.output = ip_output;
 
 	rt_set_nexthop(rth, daddr, res, fnhe, res->fi, res->type, itag);
-	if (lwtunnel_output_redirect(rth->rt_lwtstate)) {
-		rth->rt_lwtstate->orig_output = rth->dst.output;
+	if (lwtunnel_output_redirect(rth->dst.lwtstate)) {
+		rth->dst.lwtstate->orig_output = rth->dst.output;
 		rth->dst.output = lwtunnel_output;
 	}
-	if (lwtunnel_input_redirect(rth->rt_lwtstate)) {
-		rth->rt_lwtstate->orig_input = rth->dst.input;
+	if (lwtunnel_input_redirect(rth->dst.lwtstate)) {
+		rth->dst.lwtstate->orig_input = rth->dst.input;
 		rth->dst.input = lwtunnel_input;
 	}
 	skb_dst_set(skb, &rth->dst);
@@ -1695,7 +1692,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 	   by fib_lookup.
 	 */
 
-	tun_info = skb_tunnel_info(skb, AF_INET);
+	tun_info = skb_tunnel_info(skb);
 	if (tun_info && tun_info->mode == IP_TUNNEL_INFO_RX)
 		fl4.flowi4_tun_key.tun_id = tun_info->key.tun_id;
 	else
@@ -1815,7 +1812,6 @@ local_input:
 	rth->rt_gateway	= 0;
 	rth->rt_uses_gateway = 0;
 	INIT_LIST_HEAD(&rth->rt_uncached);
-	rth->rt_lwtstate = NULL;
 
 	RT_CACHE_STAT_INC(in_slow_tot);
 	if (res.type == RTN_UNREACHABLE) {
@@ -2006,7 +2002,6 @@ add:
 	rth->rt_gateway = 0;
 	rth->rt_uses_gateway = 0;
 	INIT_LIST_HEAD(&rth->rt_uncached);
-	rth->rt_lwtstate = NULL;
 	RT_CACHE_STAT_INC(out_slow_tot);
 
 	if (flags & RTCF_LOCAL)
@@ -2029,7 +2024,7 @@ add:
 	}
 
 	rt_set_nexthop(rth, fl4->daddr, res, fnhe, fi, type, 0);
-	if (lwtunnel_output_redirect(rth->rt_lwtstate))
+	if (lwtunnel_output_redirect(rth->dst.lwtstate))
 		rth->dst.output = lwtunnel_output;
 
 	return rth;
@@ -2293,7 +2288,6 @@ struct dst_entry *ipv4_blackhole_route(struct net *net, struct dst_entry *dst_or
 		rt->rt_uses_gateway = ort->rt_uses_gateway;
 
 		INIT_LIST_HEAD(&rt->rt_uncached);
-		rt->rt_lwtstate = NULL;
 		dst_free(new);
 	}