[PATCH net v2 2/2] seg6: check lwtunnel state after nf hooks in iptunnel

[Xiang Mei (Microsoft)]

seg6_input_core() and seg6_output_core() are the okfns of the
POST_ROUTING hook that seg6_input_nf() and seg6_output_nf() dispatch
through when nf_hooks_lwtunnel is enabled, and they read
skb_dst(skb)->lwtstate on the assumption fixed for seg6local in patch 1.
A hook may drop the dst, replace it with a metadata dst, or leave a route
whose lwtstate is NULL; the last is reachable with SNAT plus an XFRM
policy.

Validate the dst and the encap type before use. seg6_do_srh() reads
skb_dst(skb)->lwtstate too, but these two are its only callers and
neither touches the dst in between.

In seg6_output_core() the validated lwtstate also takes over the dst-loop
comparison, so "orig_dst" is gone, matching seg6_input_core().

Fixes: 7a3f5b0de364 ("netfilter: add netfilter hooks to SRv6 data plane")
Reported-by: Andrea Mayer <andrea.mayer@xxxxxxxxxxx>
Signed-off-by: Xiang Mei (Microsoft) <xmei5@xxxxxxx>
---
v2: new patch, split out from the seg6local fix (Andrea Mayer).

 net/ipv6/seg6_iptunnel.c | 33 +++++++++++++++++++++++++++------
 1 file changed, 27 insertions(+), 6 deletions(-)

diff --git a/net/ipv6/seg6_iptunnel.c b/net/ipv6/seg6_iptunnel.c
index 4c45c0a77d75..60883dae5ae9 100644
--- a/net/ipv6/seg6_iptunnel.c
+++ b/net/ipv6/seg6_iptunnel.c
@@ -23,6 +23,7 @@
 #include <net/addrconf.h>
 #include <net/ip6_route.h>
 #include <net/dst_cache.h>
+#include <net/dst_metadata.h>
 #ifdef CONFIG_IPV6_SEG6_HMAC
 #include <net/seg6_hmac.h>
 #endif
@@ -65,6 +66,17 @@ seg6_encap_lwtunnel(struct lwtunnel_state *lwt)
 	return seg6_lwt_lwtunnel(lwt)->tuninfo;
 }
 
+static struct lwtunnel_state *seg6_lwtst_from_skb(struct sk_buff *skb)
+{
+	struct dst_entry *dst = skb_dst(skb);
+
+	if (!skb_valid_dst(skb) || !dst->lwtstate ||
+	    dst->lwtstate->type != LWTUNNEL_ENCAP_SEG6)
+		return NULL;
+
+	return dst->lwtstate;
+}
+
 static const struct nla_policy seg6_iptunnel_policy[SEG6_IPTUNNEL_MAX + 1] = {
 	[SEG6_IPTUNNEL_SRH]	= { .type = NLA_BINARY },
 	[SEG6_IPTUNNEL_SRC]	= NLA_POLICY_EXACT_LEN(sizeof(struct in6_addr)),
@@ -488,18 +500,21 @@ static int seg6_input_finish(struct net *net, struct sock *sk,
 static int seg6_input_core(struct net *net, struct sock *sk,
 			   struct sk_buff *skb)
 {
-	struct dst_entry *orig_dst = skb_dst(skb);
 	struct dst_entry *dst = NULL;
 	struct lwtunnel_state *lwtst;
 	struct seg6_lwt *slwt;
 	int err;
 
-	/* We cannot dereference "orig_dst" once ip6_route_input() or
+	/* We cannot dereference the incoming dst once ip6_route_input() or
 	 * skb_dst_drop() is called. However, in order to detect a dst loop, we
 	 * need the address of its lwtstate. So, save the address of lwtstate
 	 * now and use it later as a comparison.
 	 */
-	lwtst = orig_dst->lwtstate;
+	lwtst = seg6_lwtst_from_skb(skb);
+	if (unlikely(!lwtst)) {
+		err = -EINVAL;
+		goto drop;
+	}
 
 	slwt = seg6_lwt_lwtunnel(lwtst);
 
@@ -581,12 +596,18 @@ static int seg6_input(struct sk_buff *skb)
 static int seg6_output_core(struct net *net, struct sock *sk,
 			    struct sk_buff *skb)
 {
-	struct dst_entry *orig_dst = skb_dst(skb);
 	struct dst_entry *dst = NULL;
+	struct lwtunnel_state *lwtst;
 	struct seg6_lwt *slwt;
 	int err;
 
-	slwt = seg6_lwt_lwtunnel(orig_dst->lwtstate);
+	lwtst = seg6_lwtst_from_skb(skb);
+	if (unlikely(!lwtst)) {
+		err = -EINVAL;
+		goto drop;
+	}
+
+	slwt = seg6_lwt_lwtunnel(lwtst);
 
 	local_bh_disable();
 	dst = dst_cache_get(&slwt->cache_output);
@@ -614,7 +635,7 @@ static int seg6_output_core(struct net *net, struct sock *sk,
 		}
 
 		/* cache only if we don't create a dst reference loop */
-		if (orig_dst->lwtstate != dst->lwtstate) {
+		if (lwtst != dst->lwtstate) {
 			local_bh_disable();
 			dst_cache_set_ip6(&slwt->cache_output, dst, &fl6.saddr);
 			local_bh_enable();
-- 
2.43.0