Re: [PATCH net 1/2] ipv6: enforce flowi6_oif usage in ip6_dst_lookup_tail()

[Hannes Frederic Sowa]

On 28.01.2016 13:27, Paolo Abeni wrote:
On Wed, 2016-01-27 at 19:38 +0100, Hannes Frederic Sowa wrote:
On 27.01.2016 14:45, Paolo Abeni wrote:
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 3c8834b..973cb73 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1183,11 +1183,10 @@ static struct rt6_info *ip6_pol_route_output(struct net 
*net, struct fib6_table
        return ip6_pol_route(net, table, fl6->flowi6_oif, fl6, flags);
   }
-struct dst_entry *ip6_route_output(struct net *net, const struct sock *sk,
-                                   struct flowi6 *fl6)
+struct dst_entry *ip6_route_output_flags(struct net *net, const struct sock 
*sk,
+                                        struct flowi6 *fl6, int flags)
   {
        struct dst_entry *dst;
-       int flags = 0;
        bool any_src;

        dst = l3mdev_rt6_dst_by_oif(net, fl6);
@@ -1208,6 +1207,13 @@ struct dst_entry *ip6_route_output(struct net *net, 
const struct sock *sk,

        return fib6_rule_lookup(net, fl6, flags, ip6_pol_route_output);
   }
+EXPORT_SYMBOL_GPL(ip6_route_output_flags);
+
+struct dst_entry *ip6_route_output(struct net *net, const struct sock *sk,
+                                  struct flowi6 *fl6)
+{
+       return ip6_route_output_flags(net, sk, fl6, 0);
+}
   EXPORT_SYMBOL(ip6_route_output);

I think this can just be a static inline function.

Is it a lot of work to introduce the flags argument globally? Most other
functions already have a flags parameter, maybe instead of just adding
another wrapper just bite the bullet and add it everywhere?

There are ~20 call sites for ip6_route_output(). Replacing them with
ip6_route_output_flags() should be trivial, but it sounds quite
invasive. Moving the new ip6_route_output() definition into the header
file as static inline function should be pretty much equivalent, may I
go with the latter option ?

I am not really a fan of such static inline wrappers all over the place, 
it doesnt't really help to evolve the code base.

But as a net/stable fix, go with the static inline first, IMHO.

Thanks,
Hannes