af_unix: Revert 'lock_interruptible' in stream receive code
authorRainer Weikusat <rweikusat@mobileactivedefense.com>
Wed, 16 Dec 2015 20:09:25 +0000 (20:09 +0000)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Sat, 23 Jan 2016 03:47:56 +0000 (19:47 -0800)
[ Upstream commit 3822b5c2fc62e3de8a0f33806ff279fb7df92432 ]

With b3ca9b02b00704053a38bfe4c31dbbb9c13595d0, the AF_UNIX SOCK_STREAM
receive code was changed from using mutex_lock(&u->readlock) to
mutex_lock_interruptible(&u->readlock) to prevent signals from being
delayed for an indefinite time if a thread sleeping on the mutex
happened to be selected for handling the signal. But this was never a
problem with the stream receive code (as opposed to its datagram
counterpart) as that never went to sleep waiting for new messages with the
mutex held and thus, wouldn't cause secondary readers to block on the
mutex waiting for the sleeping primary reader. As the interruptible
locking makes the code more complicated in exchange for no benefit,
change it back to using mutex_lock.

Signed-off-by: Rainer Weikusat <rweikusat@mobileactivedefense.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
net/unix/af_unix.c

index 95291fee38affb2d876c7de52b9ca56a72dbc7ff..f934e7ba5eb833c45098c59cb8bb440d9f50b52d 100644 (file)
@@ -1934,14 +1934,7 @@ static int unix_dgram_recvmsg(struct kiocb *iocb, struct socket *sock,
        if (flags&MSG_OOB)
                goto out;
 
-       err = mutex_lock_interruptible(&u->readlock);
-       if (unlikely(err)) {
-               /* recvmsg() in non blocking mode is supposed to return -EAGAIN
-                * sk_rcvtimeo is not honored by mutex_lock_interruptible()
-                */
-               err = noblock ? -EAGAIN : -ERESTARTSYS;
-               goto out;
-       }
+       mutex_lock(&u->readlock);
 
        skip = sk_peek_offset(sk, flags);
 
@@ -2133,12 +2126,12 @@ again:
 
                        timeo = unix_stream_data_wait(sk, timeo, last);
 
-                       if (signal_pending(current)
-                           ||  mutex_lock_interruptible(&u->readlock)) {
+                       if (signal_pending(current)) {
                                err = sock_intr_errno(timeo);
                                goto out;
                        }
 
+                       mutex_lock(&u->readlock);
                        continue;
  unlock:
                        unix_state_unlock(sk);