net: Clone skb before setting peeked flag

commit 738ac1ebb96d02e0d23bc320302a6ea94c612dec upstream.

Shared skbs must not be modified and this is crucial for broadcast
and/or multicast paths where we use it as an optimisation to avoid
unnecessary cloning.

The function skb_recv_datagram breaks this rule by setting peeked
without cloning the skb first.  This causes funky races which leads
to double-free.

This patch fixes this by cloning the skb and replacing the skb
in the list when setting skb->peeked.

Fixes: a59322be07c9 ("[UDP]: Only increment counter on first peek/recv")
Reported-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
[bwh: Backported to 3.2: adjust context]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
(cherry picked from commit 72e6f0680249f5e0a87f2b282d033baefd90d84e)
[wt: adjusted context for 2.6.32. Introduces a bug, see next commit]
Signed-off-by: Willy Tarreau <w@1wt.eu>

diff --git a/net/core/datagram.c b/net/core/datagram.c
index 4ade3011bb3c2bd4d270cfb10cdcf5e112402ce9..cbb3100c37dc8de7017db186681240dadbb694b5 100644 (file)
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -127,6 +127,35 @@ out_noerr:
        goto out;
 }
 
+static int skb_set_peeked(struct sk_buff *skb)
+{
+       struct sk_buff *nskb;
+
+       if (skb->peeked)
+               return 0;
+
+       /* We have to unshare an skb before modifying it. */
+       if (!skb_shared(skb))
+               goto done;
+
+       nskb = skb_clone(skb, GFP_ATOMIC);
+       if (!nskb)
+               return -ENOMEM;
+
+       skb->prev->next = nskb;
+       skb->next->prev = nskb;
+       nskb->prev = skb->prev;
+       nskb->next = skb->next;
+
+       consume_skb(skb);
+       skb = nskb;
+
+done:
+       skb->peeked = 1;
+
+       return 0;
+}
+
 /**
  *     __skb_recv_datagram - Receive a datagram skbuff
  *     @sk: socket
@@ -160,6 +189,7 @@ struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned flags,
                                    int *peeked, int *err)
 {
        struct sk_buff *skb;
+       unsigned long cpu_flags;
        long timeo;
        /*
         * Caller is allowed not to check sk->sk_err before skb_recv_datagram()
@@ -178,14 +208,16 @@ struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned flags,
                 * Look at current nfs client by the way...
                 * However, this function was corrent in any case. 8)
                 */
-               unsigned long cpu_flags;
-
                spin_lock_irqsave(&sk->sk_receive_queue.lock, cpu_flags);
                skb = skb_peek(&sk->sk_receive_queue);
                if (skb) {
                        *peeked = skb->peeked;
                        if (flags & MSG_PEEK) {
-                               skb->peeked = 1;
+
+                               error = skb_set_peeked(skb);
+                               if (error)
+                                       goto unlock_err;
+
                                atomic_inc(&skb->users);
                        } else
                                __skb_unlink(skb, &sk->sk_receive_queue);
@@ -204,6 +236,8 @@ struct sk_buff *__skb_recv_datagram(struct sock *sk, unsigned flags,
 
        return NULL;
 
+unlock_err:
+       spin_unlock_irqrestore(&sk->sk_receive_queue.lock, cpu_flags);
 no_packet:
        *err = error;
        return NULL;