IB/mad: Fix ib_post_send_mad() returning 0 with no generate send comp

If ib_post_send_mad() returns 0, the API guarantees that there will be
a callback to send_buf->mad_agent->send_handler() so that the sender
can call ib_free_send_mad().  Otherwise, the ib_mad_send_buf will be
leaked and the mad_agent reference count will never go to zero and the
IB device module cannot be unloaded.  The above can happen without
this patch if process_mad() returns (IB_MAD_RESULT_SUCCESS |
IB_MAD_RESULT_CONSUMED).

If process_mad() returns IB_MAD_RESULT_SUCCESS and there is no agent
registered to receive the mad being sent, handle_outgoing_dr_smp()
returns zero which causes a MAD packet which is at the end of the
directed route to be incorrectly sent on the wire but doesn't cause a
hang since the HCA generates a send completion.

Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c
index dbcd285..de922a0 100644
--- a/drivers/infiniband/core/mad.c
+++ b/drivers/infiniband/core/mad.c
@@ -742,9 +742,7 @@
 		break;
 	case IB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_CONSUMED:
 		kmem_cache_free(ib_mad_cache, mad_priv);
-		kfree(local);
-		ret = 1;
-		goto out;
+		break;
 	case IB_MAD_RESULT_SUCCESS:
 		/* Treat like an incoming receive MAD */
 		port_priv = ib_get_mad_port(mad_agent_priv->agent.device,
@@ -755,10 +753,12 @@
 						        &mad_priv->mad.mad);
 		}
 		if (!port_priv || !recv_mad_agent) {
+			/*
+			 * No receiving agent so drop packet and
+			 * generate send completion.
+			 */
 			kmem_cache_free(ib_mad_cache, mad_priv);
-			kfree(local);
-			ret = 0;
-			goto out;
+			break;
 		}
 		local->mad_priv = mad_priv;
 		local->recv_mad_agent = recv_mad_agent;