net: Fix vti use case with oif in dst lookups for IPv6
authorDavid Ahern <dsa@cumulusnetworks.com>
Mon, 5 Oct 2015 14:32:51 +0000 (08:32 -0600)
committerDavid S. Miller <davem@davemloft.net>
Wed, 7 Oct 2015 11:24:00 +0000 (04:24 -0700)
It occurred to me yesterday that 741a11d9e4103 ("net: ipv6: Add
RT6_LOOKUP_F_IFACE flag if oif is set") means that xfrm6_dst_lookup
needs the FLOWI_FLAG_SKIP_NH_OIF flag set. This latest commit causes
the oif to be considered in lookups which is known to break vti. This
explains why 58189ca7b274 did not the IPv6 change at the time it was
submitted.

Fixes: 42a7b32b73d6 ("xfrm: Add oif to dst lookups")
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/ipv6/xfrm6_policy.c

index 69cee4e..08c9c93 100644 (file)
@@ -37,6 +37,7 @@ static struct dst_entry *xfrm6_dst_lookup(struct net *net, int tos, int oif,
 
        memset(&fl6, 0, sizeof(fl6));
        fl6.flowi6_oif = oif;
+       fl6.flowi6_flags = FLOWI_FLAG_SKIP_NH_OIF;
        memcpy(&fl6.daddr, daddr, sizeof(fl6.daddr));
        if (saddr)
                memcpy(&fl6.saddr, saddr, sizeof(fl6.saddr));