ipv6: fix backtracking for throw routes
authorSteven Barth <cyrus@openwrt.org>
Thu, 19 Mar 2015 15:16:04 +0000 (16:16 +0100)
committerDavid S. Miller <davem@davemloft.net>
Fri, 20 Mar 2015 16:57:23 +0000 (12:57 -0400)
for throw routes to trigger evaluation of other policy rules
EAGAIN needs to be propagated up to fib_rules_lookup
similar to how its done for IPv4

A simple testcase for verification is:

ip -6 rule add lookup 33333 priority 33333
ip -6 route add throw 2001:db8::1
ip -6 route add 2001:db8::1 via fe80::1 dev wlan0 table 33333
ip route get 2001:db8::1

Signed-off-by: Steven Barth <cyrus@openwrt.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/ipv6/fib6_rules.c

index b4d5e1d97c1b2576fc02a15b7a05a773197bb8e0..27ca79682efbf681a0ab6073f50f8fa73214028e 100644 (file)
@@ -104,6 +104,7 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp,
                                goto again;
                        flp6->saddr = saddr;
                }
+               err = rt->dst.error;
                goto out;
        }
 again: