diff --git a/mysql-test/r/null_key.result b/mysql-test/r/null_key.result
index 4b7400e5e60..9f11a0129cd 100644
--- a/mysql-test/r/null_key.result
+++ b/mysql-test/r/null_key.result
@@ -364,3 +364,34 @@ select * from t1;
 id	id2
 1	1
 drop table t1;
+CREATE TABLE t1 (a int);
+CREATE TABLE t2 (a int, b int, INDEX idx(a));
+CREATE TABLE t3 (b int, INDEX idx(b));
+INSERT INTO t1 VALUES (1), (2), (3), (4);
+INSERT INTO t2 VALUES (1, 1), (3, 1);
+INSERT INTO t3 VALUES 
+(NULL), (NULL), (NULL), (NULL), (NULL),
+(NULL), (NULL), (NULL), (NULL), (NULL),
+(2);
+ANALYZE table t1, t2, t3;
+Table	Op	Msg_type	Msg_text
+test.t1	analyze	status	OK
+test.t2	analyze	status	OK
+test.t3	analyze	status	OK
+EXPLAIN SELECT SQL_CALC_FOUND_ROWS * FROM t1 LEFT JOIN t2 ON t1.a=t2.a
+LEFT JOIN t3 ON t2.b=t3.b;
+id	select_type	table	type	possible_keys	key	key_len	ref	rows	Extra
+1	SIMPLE	t1	ALL	NULL	NULL	NULL	NULL	4	
+1	SIMPLE	t2	ref	idx	idx	5	test.t1.a	1	
+1	SIMPLE	t3	ref	idx	idx	5	test.t2.b	1	Using index
+SELECT SQL_CALC_FOUND_ROWS * FROM t1 LEFT JOIN t2 ON t1.a=t2.a
+LEFT JOIN t3 ON t2.b=t3.b;
+a	a	b	b
+1	1	1	NULL
+2	NULL	NULL	NULL
+3	3	1	NULL
+4	NULL	NULL	NULL
+SELECT FOUND_ROWS();
+FOUND_ROWS()
+4
+DROP TABLE t1,t2,t3;
diff --git a/mysql-test/t/null_key.test b/mysql-test/t/null_key.test
index 9b346a181bf..0eff53ac371 100644
--- a/mysql-test/t/null_key.test
+++ b/mysql-test/t/null_key.test
@@ -191,3 +191,29 @@ select * from t1 where id2 is null or id2 > 0;
 delete from t1 where id <=> NULL;
 select * from t1;
 drop table t1;
+
+#
+# Test for bug #12144: optimizations for key access with null keys 
+#                      used for outer joins
+#
+
+CREATE TABLE t1 (a int);
+CREATE TABLE t2 (a int, b int, INDEX idx(a));
+CREATE TABLE t3 (b int, INDEX idx(b));
+INSERT INTO t1 VALUES (1), (2), (3), (4);
+INSERT INTO t2 VALUES (1, 1), (3, 1);
+INSERT INTO t3 VALUES 
+  (NULL), (NULL), (NULL), (NULL), (NULL),
+  (NULL), (NULL), (NULL), (NULL), (NULL),
+  (2);
+ANALYZE table t1, t2, t3;
+
+EXPLAIN SELECT SQL_CALC_FOUND_ROWS * FROM t1 LEFT JOIN t2 ON t1.a=t2.a
+                                             LEFT JOIN t3 ON t2.b=t3.b;
+
+SELECT SQL_CALC_FOUND_ROWS * FROM t1 LEFT JOIN t2 ON t1.a=t2.a
+                                     LEFT JOIN t3 ON t2.b=t3.b;
+
+SELECT FOUND_ROWS();
+
+DROP TABLE t1,t2,t3;
diff --git a/sql/sql_select.cc b/sql/sql_select.cc
index 29560e8701b..6f95ca1390e 100644
--- a/sql/sql_select.cc
+++ b/sql/sql_select.cc
@@ -2255,7 +2255,9 @@ add_key_field(KEY_FIELD **key_fields,uint and_level, Item_func *cond,
   */
   (*key_fields)->null_rejecting= (cond->functype() == Item_func::EQ_FUNC) &&
                                  ((*value)->type() == Item::FIELD_ITEM) &&
-                                 ((Item_field*)*value)->field->maybe_null();
+
+                        (((Item_field*)*value)->field->maybe_null() ||
+			 ((Item_field *)*value)->field->table->maybe_null);
   (*key_fields)++;
 }
 
@@ -6310,6 +6312,11 @@ join_read_always_key(JOIN_TAB *tab)
   int error;
   TABLE *table= tab->table;
 
+  for (uint i= 0 ; i < tab->ref.key_parts ; i++)
+  {
+    if ((tab->ref.null_rejecting & 1 << i) && tab->ref.items[i]->is_null())
+        return -1;
+  } 
   if (!table->file->inited)
     table->file->ha_index_init(tab->ref.key);
   if (cp_buffer_from_ref(tab->join->thd, &tab->ref))