[PATCH net 3/5] rxrpc: Fix notification vs call-release vs recvmsg

[David Howells]

When a call is released, rxrpc takes the spinlock and removes it from
->recvmsg_q in an effort to prevent racing recvmsg() invocations from
seeing the same call.  Now, rxrpc_recvmsg() only takes the spinlock when
actually removing a call from the queue; it doesn't, however, take it in
the lead up to that when it checks to see if the queue is empty.  It *does*
hold the socket lock, which prevents a recvmsg/recvmsg race - but this
doesn't prevent sendmsg from ending the call because sendmsg() drops the
socket lock and relies on the call->user_mutex.

Fix this by firstly removing the bit in rxrpc_release_call() that dequeues
the released call and, instead, rely on recvmsg() to simply discard
released calls (done in a preceding fix).

Secondly, rxrpc_notify_socket() is abandoned if the call is already marked
as released rather than trying to be clever by setting both pointers in
call->recvmsg_link to NULL to trick list_empty().  This isn't perfect and
can still race, resulting in a released call on the queue, but recvmsg()
will now clean that up.

Fixes: 17926a79320a ("[AF_RXRPC]: Provide secure RxRPC sockets for use by userspace and kernel both")
Signed-off-by: David Howells <dhowells at redhat.com>
Reviewed-by: Jeffrey Altman <jaltman at auristor.com>
cc: Marc Dionne <marc.dionne at auristor.com>
cc: Junvyyang, Tencent Zhuque Lab <zhuque at tencent.com>
cc: LePremierHomme <kwqcheii at proton.me>
cc: Jakub Kicinski <kuba at kernel.org>
cc: Paolo Abeni <pabeni at redhat.com>
cc: "David S. Miller" <davem at davemloft.net>
cc: Eric Dumazet <edumazet at google.com>
cc: Simon Horman <horms at kernel.org>
cc: linux-afs at lists.infradead.org
cc: netdev at vger.kernel.org
cc: stable at vger.kernel.org
---
 include/trace/events/rxrpc.h |  2 +-
 net/rxrpc/call_object.c      | 28 ++++++++++++----------------
 net/rxrpc/recvmsg.c          |  4 ++++
 3 files changed, 17 insertions(+), 17 deletions(-)

diff --git a/include/trace/events/rxrpc.h b/include/trace/events/rxrpc.h
index e7dcfb1369b6..8e5a73eb5268 100644
--- a/include/trace/events/rxrpc.h
+++ b/include/trace/events/rxrpc.h
@@ -325,7 +325,6 @@
 	EM(rxrpc_call_put_release_sock,		"PUT rls-sock") \
 	EM(rxrpc_call_put_release_sock_tba,	"PUT rls-sk-a") \
 	EM(rxrpc_call_put_sendmsg,		"PUT sendmsg ") \
-	EM(rxrpc_call_put_unnotify,		"PUT unnotify") \
 	EM(rxrpc_call_put_userid_exists,	"PUT u-exists") \
 	EM(rxrpc_call_put_userid,		"PUT user-id ") \
 	EM(rxrpc_call_see_accept,		"SEE accept  ") \
@@ -338,6 +337,7 @@
 	EM(rxrpc_call_see_disconnected,		"SEE disconn ") \
 	EM(rxrpc_call_see_distribute_error,	"SEE dist-err") \
 	EM(rxrpc_call_see_input,		"SEE input   ") \
+	EM(rxrpc_call_see_notify_released,	"SEE nfy-rlsd") \
 	EM(rxrpc_call_see_recvmsg,		"SEE recvmsg ") \
 	EM(rxrpc_call_see_release,		"SEE release ") \
 	EM(rxrpc_call_see_userid_exists,	"SEE u-exists") \
diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c
index 15067ff7b1f2..918f41d97a2f 100644
--- a/net/rxrpc/call_object.c
+++ b/net/rxrpc/call_object.c
@@ -561,7 +561,7 @@ static void rxrpc_cleanup_rx_buffers(struct rxrpc_call *call)
 void rxrpc_release_call(struct rxrpc_sock *rx, struct rxrpc_call *call)
 {
 	struct rxrpc_connection *conn = call->conn;
-	bool put = false, putu = false;
+	bool putu = false;
 
 	_enter("{%d,%d}", call->debug_id, refcount_read(&call->ref));
 
@@ -573,23 +573,13 @@ void rxrpc_release_call(struct rxrpc_sock *rx, struct rxrpc_call *call)
 
 	rxrpc_put_call_slot(call);
 
-	/* Make sure we don't get any more notifications */
+	/* Note that at this point, the call may still be on or may have been
+	 * added back on to the socket receive queue.  recvmsg() must discard
+	 * released calls.  The CALL_RELEASED flag should prevent further
+	 * notifications.
+	 */
 	spin_lock_irq(&rx->recvmsg_lock);
-
-	if (!list_empty(&call->recvmsg_link)) {
-		_debug("unlinking once-pending call %p { e=%lx f=%lx }",
-		       call, call->events, call->flags);
-		list_del(&call->recvmsg_link);
-		put = true;
-	}
-
-	/* list_empty() must return false in rxrpc_notify_socket() */
-	call->recvmsg_link.next = NULL;
-	call->recvmsg_link.prev = NULL;
-
 	spin_unlock_irq(&rx->recvmsg_lock);
-	if (put)
-		rxrpc_put_call(call, rxrpc_call_put_unnotify);
 
 	write_lock(&rx->call_lock);
 
@@ -638,6 +628,12 @@ void rxrpc_release_calls_on_socket(struct rxrpc_sock *rx)
 		rxrpc_put_call(call, rxrpc_call_put_release_sock);
 	}
 
+	while ((call = list_first_entry_or_null(&rx->recvmsg_q,
+						struct rxrpc_call, recvmsg_link))) {
+		list_del_init(&call->recvmsg_link);
+		rxrpc_put_call(call, rxrpc_call_put_release_recvmsg_q);
+	}
+
 	_leave("");
 }
 
diff --git a/net/rxrpc/recvmsg.c b/net/rxrpc/recvmsg.c
index 6990e37697de..7fa7e77f6bb9 100644
--- a/net/rxrpc/recvmsg.c
+++ b/net/rxrpc/recvmsg.c
@@ -29,6 +29,10 @@ void rxrpc_notify_socket(struct rxrpc_call *call)
 
 	if (!list_empty(&call->recvmsg_link))
 		return;
+	if (test_bit(RXRPC_CALL_RELEASED, &call->flags)) {
+		rxrpc_see_call(call, rxrpc_call_see_notify_released);
+		return;
+	}
 
 	rcu_read_lock();