diff options
| author | Eric Dumazet <edumazet@google.com> | 2014-06-24 10:05:11 -0700 |
|---|---|---|
| committer | Greg Kroah-Hartman <gregkh@linuxfoundation.org> | 2014-07-28 08:05:59 -0700 |
| commit | df90819dafcd6b97fc665f63a15752a570e227a2 (patch) | |
| tree | a95d89aa04b90bad0bca0e32922457599457b0eb | |
| parent | 10bc26ecaf0e7e80c14b98375bbfe85f977a53f2 (diff) | |
ipv4: fix dst race in sk_dst_get()
[ Upstream commit f88649721268999bdff09777847080a52004f691 ]
When IP route cache had been removed in linux-3.6, we broke assumption
that dst entries were all freed after rcu grace period. DST_NOCACHE
dst were supposed to be freed from dst_release(). But it appears
we want to keep such dst around, either in UDP sockets or tunnels.
In sk_dst_get() we need to make sure dst refcount is not 0
before incrementing it, or else we might end up freeing a dst
twice.
DST_NOCACHE set on a dst does not mean this dst can not be attached
to a socket or a tunnel.
Then, before actual freeing, we need to observe a rcu grace period
to make sure all other cpus can catch the fact the dst is no longer
usable.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Dormando <dormando@rydia.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
| -rw-r--r-- | include/net/sock.h | 4 | ||||
| -rw-r--r-- | net/core/dst.c | 16 | ||||
| -rw-r--r-- | net/ipv4/ip_tunnel.c | 14 |
3 files changed, 18 insertions, 16 deletions
diff --git a/include/net/sock.h b/include/net/sock.h index 57c31dd15e6..a803e77024b 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1755,8 +1755,8 @@ sk_dst_get(struct sock *sk) rcu_read_lock(); dst = rcu_dereference(sk->sk_dst_cache); - if (dst) - dst_hold(dst); + if (dst && !atomic_inc_not_zero(&dst->__refcnt)) + dst = NULL; rcu_read_unlock(); return dst; } diff --git a/net/core/dst.c b/net/core/dst.c index ca4231ec734..15b6792e6eb 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -267,6 +267,15 @@ again: } EXPORT_SYMBOL(dst_destroy); +static void dst_destroy_rcu(struct rcu_head *head) +{ + struct dst_entry *dst = container_of(head, struct dst_entry, rcu_head); + + dst = dst_destroy(dst); + if (dst) + __dst_free(dst); +} + void dst_release(struct dst_entry *dst) { if (dst) { @@ -274,11 +283,8 @@ void dst_release(struct dst_entry *dst) newrefcnt = atomic_dec_return(&dst->__refcnt); WARN_ON(newrefcnt < 0); - if (unlikely(dst->flags & DST_NOCACHE) && !newrefcnt) { - dst = dst_destroy(dst); - if (dst) - __dst_free(dst); - } + if (unlikely(dst->flags & DST_NOCACHE) && !newrefcnt) + call_rcu(&dst->rcu_head, dst_destroy_rcu); } } EXPORT_SYMBOL(dst_release); diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index 719dd634da9..62cd9e0ae35 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -73,12 +73,7 @@ static void __tunnel_dst_set(struct ip_tunnel_dst *idst, { struct dst_entry *old_dst; - if (dst) { - if (dst->flags & DST_NOCACHE) - dst = NULL; - else - dst_clone(dst); - } + dst_clone(dst); old_dst = xchg((__force struct dst_entry **)&idst->dst, dst); dst_release(old_dst); } @@ -108,13 +103,14 @@ static struct rtable *tunnel_rtable_get(struct ip_tunnel *t, u32 cookie) rcu_read_lock(); dst = rcu_dereference(this_cpu_ptr(t->dst_cache)->dst); + if (dst && !atomic_inc_not_zero(&dst->__refcnt)) + dst = NULL; if (dst) { if (dst->obsolete && dst->ops->check(dst, cookie) == NULL) { - rcu_read_unlock(); tunnel_dst_reset(t); - return NULL; + dst_release(dst); + dst = NULL; } - dst_hold(dst); } rcu_read_unlock(); return (struct rtable *)dst; |