Commit 
7ea0ed2b5be8 ("ipmi: Make the message handler easier to use for
SMI interfaces") changed handle_new_recv_msgs() to call handle_one_recv_msg()
for a smi_msg while the smi_msg is still connected to waiting_rcv_msgs list.
That could lead to following list corruption problems:
1) low-level function treats smi_msg as not connected to list
  handle_one_recv_msg() could end up calling smi_send(), which
  assumes the msg is not connected to list.
  For example, the following sequence could corrupt list by
  doing list_add_tail() for the entry still connected to other list.
    handle_new_recv_msgs()
      msg = list_entry(waiting_rcv_msgs)
      handle_one_recv_msg(msg)
        handle_ipmb_get_msg_cmd(msg)
          smi_send(msg)
            spin_lock(xmit_msgs_lock)
            list_add_tail(msg)
            spin_unlock(xmit_msgs_lock)
2) race between multiple handle_new_recv_msgs() instances
  handle_new_recv_msgs() once releases waiting_rcv_msgs_lock before calling
  handle_one_recv_msg() then retakes the lock and list_del() it.
  If others call handle_new_recv_msgs() during the window shown below
  list_del() will be done twice for the same smi_msg.
  handle_new_recv_msgs()
    spin_lock(waiting_rcv_msgs_lock)
    msg = list_entry(waiting_rcv_msgs)
    spin_unlock(waiting_rcv_msgs_lock)
  |
  | handle_one_recv_msg(msg)
  |
    spin_lock(waiting_rcv_msgs_lock)
    list_del(msg)
    spin_unlock(waiting_rcv_msgs_lock)
Fixes: 7ea0ed2b5be8 ("ipmi: Make the message handler easier to use for SMI interfaces")
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
[Added a comment to describe why this works.]
Signed-off-by: Corey Minyard <cminyard@mvista.com>
Cc: stable@vger.kernel.org # 3.19
Tested-by: Ye Feng <yefeng.yl@alibaba-inc.com>
        while (!list_empty(&intf->waiting_rcv_msgs)) {
                smi_msg = list_entry(intf->waiting_rcv_msgs.next,
                                     struct ipmi_smi_msg, link);
+               list_del(&smi_msg->link);
                if (!run_to_completion)
                        spin_unlock_irqrestore(&intf->waiting_rcv_msgs_lock,
                                               flags);
                if (rv > 0) {
                        /*
                         * To preserve message order, quit if we
-                        * can't handle a message.
+                        * can't handle a message.  Add the message
+                        * back at the head, this is safe because this
+                        * tasklet is the only thing that pulls the
+                        * messages.
                         */
+                       list_add(&smi_msg->link, &intf->waiting_rcv_msgs);
                        break;
                } else {
-                       list_del(&smi_msg->link);
                        if (rv == 0)
                                /* Message handled */
                                ipmi_free_smi_msg(smi_msg);