[PATCH net-next] ipv6: Always allocate pcpu memory in a fib6_nh

[PATCH net-next] ipv6: Always allocate pcpu memory in a fib6_nh

[David Ahern]

From: David Ahern <dsahern@gmail.com>

A recent commit had an unintended side effect with reject routes:
rt6i_pcpu is expected to always be initialized for all fib6_info except
the null entry. The commit mentioned below skips it for reject routes
and ends up leaking references to the loopback device. For example,

    ip netns add foo
    ip -netns foo li set lo up
    ip -netns foo -6 ro add blackhole 2001:db8:1::1
    ip netns exec foo ping6 2001:db8:1::1
    ip netns del foo

ends up spewing:
    unregister_netdevice: waiting for lo to become free. Usage count = 3

The fib_nh_common_init is not needed for reject routes (no ipv4 caching
or encaps), so move the alloc_percpu_gfp after it and adjust the goto label.

Fixes: f40b6ae2b612 ("ipv6: Move pcpu cached routes to fib6_nh")
Signed-off-by: David Ahern <dsahern@gmail.com>
---
 net/ipv6/route.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 29c2f5086116..d5777e92609d 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -3400,7 +3400,7 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
 				goto out;
 			}
 		}
-		goto set_dev;
+		goto pcpu_alloc;
 	}
 
 	if (cfg->fc_flags & RTF_GATEWAY) {
@@ -3432,17 +3432,18 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
 	    !netif_carrier_ok(dev))
 		fib6_nh->fib_nh_flags |= RTNH_F_LINKDOWN;
 
+	err = fib_nh_common_init(&fib6_nh->nh_common, cfg->fc_encap,
+				 cfg->fc_encap_type, cfg, gfp_flags, extack);
+	if (err)
+		goto out;
+
+pcpu_alloc:
 	fib6_nh->rt6i_pcpu = alloc_percpu_gfp(struct rt6_info *, gfp_flags);
 	if (!fib6_nh->rt6i_pcpu) {
 		err = -ENOMEM;
 		goto out;
 	}
 
-	err = fib_nh_common_init(&fib6_nh->nh_common, cfg->fc_encap,
-				 cfg->fc_encap_type, cfg, gfp_flags, extack);
-	if (err)
-		goto out;
-set_dev:
 	fib6_nh->fib_nh_dev = dev;
 	fib6_nh->fib_nh_oif = dev->ifindex;
 	err = 0;
-- 
2.11.0

Re: [PATCH net-next] ipv6: Always allocate pcpu memory in a fib6_nh

[David Miller]

From: David Ahern <dsahern@kernel.org>
Date: Mon,  3 Jun 2019 18:37:03 -0700

> From: David Ahern <dsahern@gmail.com>
> 
> A recent commit had an unintended side effect with reject routes:
> rt6i_pcpu is expected to always be initialized for all fib6_info except
> the null entry. The commit mentioned below skips it for reject routes
> and ends up leaking references to the loopback device. For example,
> 
>     ip netns add foo
>     ip -netns foo li set lo up
>     ip -netns foo -6 ro add blackhole 2001:db8:1::1
>     ip netns exec foo ping6 2001:db8:1::1
>     ip netns del foo
> 
> ends up spewing:
>     unregister_netdevice: waiting for lo to become free. Usage count = 3
> 
> The fib_nh_common_init is not needed for reject routes (no ipv4 caching
> or encaps), so move the alloc_percpu_gfp after it and adjust the goto label.
> 
> Fixes: f40b6ae2b612 ("ipv6: Move pcpu cached routes to fib6_nh")
> Signed-off-by: David Ahern <dsahern@gmail.com>

Applied.