From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 6 Dec 2010 17:29:43 +0000 (-0800)
Subject: filter: fix sk_filter rcu handling
X-Git-Tag: v2.6.37-rc6~14^2~39
X-Git-Url: https://openfabrics.org/gitweb/?a=commitdiff_plain;h=46bcf14f44d8f31ecfdc8b6708ec15a3b33316d9;p=~shefty%2Frdma-dev.git

filter: fix sk_filter rcu handling

Pavel Emelyanov tried to fix a race between sk_filter_(de|at)tach and
sk_clone() in commit 47e958eac280c263397

Problem is we can have several clones sharing a common sk_filter, and
these clones might want to sk_filter_attach() their own filters at the
same time, and can overwrite old_filter->rcu, corrupting RCU queues.

We can not use filter->rcu without being sure no other thread could do
the same thing.

Switch code to a more conventional ref-counting technique : Do the
atomic decrement immediately and queue one rcu call back when last
reference is released.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---

diff --git a/include/net/sock.h b/include/net/sock.h
index a6338d03985..659d968d95c 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1155,6 +1155,8 @@ extern void sk_common_release(struct sock *sk);
 /* Initialise core socket variables */
 extern void sock_init_data(struct socket *sock, struct sock *sk);
 
+extern void sk_filter_release_rcu(struct rcu_head *rcu);
+
 /**
  *	sk_filter_release - release a socket filter
  *	@fp: filter to remove
@@ -1165,7 +1167,7 @@ extern void sock_init_data(struct socket *sock, struct sock *sk);
 static inline void sk_filter_release(struct sk_filter *fp)
 {
 	if (atomic_dec_and_test(&fp->refcnt))
-		kfree(fp);
+		call_rcu_bh(&fp->rcu, sk_filter_release_rcu);
 }
 
 static inline void sk_filter_uncharge(struct sock *sk, struct sk_filter *fp)
diff --git a/net/core/filter.c b/net/core/filter.c
index c1ee800bc08..ae21a0d3c4a 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -589,23 +589,16 @@ int sk_chk_filter(struct sock_filter *filter, int flen)
 EXPORT_SYMBOL(sk_chk_filter);
 
 /**
- * 	sk_filter_rcu_release - Release a socket filter by rcu_head
+ * 	sk_filter_release_rcu - Release a socket filter by rcu_head
  *	@rcu: rcu_head that contains the sk_filter to free
  */
-static void sk_filter_rcu_release(struct rcu_head *rcu)
+void sk_filter_release_rcu(struct rcu_head *rcu)
 {
 	struct sk_filter *fp = container_of(rcu, struct sk_filter, rcu);
 
-	sk_filter_release(fp);
-}
-
-static void sk_filter_delayed_uncharge(struct sock *sk, struct sk_filter *fp)
-{
-	unsigned int size = sk_filter_len(fp);
-
-	atomic_sub(size, &sk->sk_omem_alloc);
-	call_rcu_bh(&fp->rcu, sk_filter_rcu_release);
+	kfree(fp);
 }
+EXPORT_SYMBOL(sk_filter_release_rcu);
 
 /**
  *	sk_attach_filter - attach a socket filter
@@ -649,7 +642,7 @@ int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk)
 	rcu_assign_pointer(sk->sk_filter, fp);
 
 	if (old_fp)
-		sk_filter_delayed_uncharge(sk, old_fp);
+		sk_filter_uncharge(sk, old_fp);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(sk_attach_filter);
@@ -663,7 +656,7 @@ int sk_detach_filter(struct sock *sk)
 					   sock_owned_by_user(sk));
 	if (filter) {
 		rcu_assign_pointer(sk->sk_filter, NULL);
-		sk_filter_delayed_uncharge(sk, filter);
+		sk_filter_uncharge(sk, filter);
 		ret = 0;
 	}
 	return ret;