libiscsi: Fix host busy blocking during connection teardown

commit 660d0831 upstream.

In case of hw iscsi offload, an host can have N-number of active
connections. There can be IO's running on some connections which
make host->host_busy always TRUE. Now if logout from a connection
is tried then the code gets into an infinite loop as host->host_busy
is always TRUE.

 iscsi_conn_teardown(....)
 {
   .........
    /*
     * Block until all in-progress commands for this connection
     * time out or fail.
     */
     for (;;) {
      spin_lock_irqsave(session->host->host_lock, flags);
      if (!atomic_read(&session->host->host_busy)) { /* OK for ERL == 0 */
	      spin_unlock_irqrestore(session->host->host_lock, flags);
              break;
      }
     spin_unlock_irqrestore(session->host->host_lock, flags);
     msleep_interruptible(500);
     iscsi_conn_printk(KERN_INFO, conn, "iscsi conn_destroy(): "
                 "host_busy %d host_failed %d\n",
	          atomic_read(&session->host->host_busy),
	          session->host->host_failed);

	................
	...............
     }
  }

This is not an issue with software-iscsi/iser as each cxn is a separate
host.

Fix:
Acquiring eh_mutex in iscsi_conn_teardown() before setting
session->state = ISCSI_STATE_TERMINATE.
Signed-off-by: John Soni Jose <sony.john@avagotech.com>
Reviewed-by: Mike Christie <michaelc@cs.wisc.edu>
Reviewed-by: Chris Leech <cleech@redhat.com>
Signed-off-by: James Bottomley <JBottomley@Odin.com>
[ luis: backported to 3.16: adjusted context ]
Signed-off-by: Luis Henriques <luis.henriques@canonical.com>

drivers/scsi/libiscsi.c

@@ -2960,10 +2960,10 @@ void iscsi_conn_teardown(struct iscsi_cls_conn *cls_conn)
 {
 	struct iscsi_conn *conn = cls_conn->dd_data;
 	struct iscsi_session *session = conn->session;
-	unsigned long flags;
 
 	del_timer_sync(&conn->transport_timer);
 
+	mutex_lock(&session->eh_mutex);
 	spin_lock_bh(&session->frwd_lock);
 	conn->c_stage = ISCSI_CONN_CLEANUP_WAIT;
 	if (session->leadconn == conn) {
@@ -2975,28 +2975,6 @@ void iscsi_conn_teardown(struct iscsi_cls_conn *cls_conn)
 	}
 	spin_unlock_bh(&session->frwd_lock);
 
-	/*
-	 * Block until all in-progress commands for this connection
-	 * time out or fail.
-	 */
-	for (;;) {
-		spin_lock_irqsave(session->host->host_lock, flags);
-		if (!session->host->host_busy) { /* OK for ERL == 0 */
-			spin_unlock_irqrestore(session->host->host_lock, flags);
-			break;
-		}
-		spin_unlock_irqrestore(session->host->host_lock, flags);
-		msleep_interruptible(500);
-		iscsi_conn_printk(KERN_INFO, conn, "iscsi conn_destroy(): "
-				  "host_busy %d host_failed %d\n",
-				  session->host->host_busy,
-				  session->host->host_failed);
-		/*
-		 * force eh_abort() to unblock
-		 */
-		wake_up(&conn->ehwait);
-	}
-
 	/* flush queued up work because we free the connection below */
 	iscsi_suspend_tx(conn);
 
@@ -3013,6 +2991,7 @@ void iscsi_conn_teardown(struct iscsi_cls_conn *cls_conn)
 	if (session->leadconn == conn)
 		session->leadconn = NULL;
 	spin_unlock_bh(&session->frwd_lock);
+	mutex_unlock(&session->eh_mutex);
 
 	iscsi_destroy_conn(cls_conn);
 }