RDMA/iwcm: Don't touch cm_id after deref in rem_ref

rem_ref() calls iwcm_deref_id(), which will wake up any blockers on
cm_id_priv->destroy_comp if the refcnt hits 0.  That will unblock
someone in iw_destroy_cm_id() which will free the cmid.  If that
happens before rem_ref() calls test_bit(IWCM_F_CALLBACK_DESTROY,
&cm_id_priv->flags), then the test_bit() will touch freed memory.

The fix is to read the bit first, then deref.  We should never be in
iw_destroy_cm_id() with IWCM_F_CALLBACK_DESTROY set, and there is a
BUG_ON() to make sure of that.

Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>

drivers/infiniband/core/iwcm.c

@@ -181,9 +181,16 @@ static void add_ref(struct iw_cm_id *cm_id)
 static void rem_ref(struct iw_cm_id *cm_id)
 {
 	struct iwcm_id_private *cm_id_priv;
+	int cb_destroy;
+
 	cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
-	if (iwcm_deref_id(cm_id_priv) &&
-	    test_bit(IWCM_F_CALLBACK_DESTROY, &cm_id_priv->flags)) {
+
+	/*
+	 * Test bit before deref in case the cm_id gets freed on another
+	 * thread.
+	 */
+	cb_destroy = test_bit(IWCM_F_CALLBACK_DESTROY, &cm_id_priv->flags);
+	if (iwcm_deref_id(cm_id_priv) && cb_destroy) {
 		BUG_ON(!list_empty(&cm_id_priv->work_list));
 		free_cm_id(cm_id_priv);
 	}