MDEV-17382 Hash join algorithm should not be used to join materialized

           derived table / view by equality

Now rows of a materialized derived table are always put into a
temporary table before join operation. If BNLH is used to join this
table with the result of a partial join then both operands of the
join are actually put into main memory. In most cases this is not
efficient.
We could avoid this by sending the rows of the derived table directly
to the join operation. However this kind of data flow is not supported
yet.
Fixed by not allowing usage of hash join algorithm to join a materialized
derived table if it's joined by an equality predicate of the form
f=e where f is a field of the derived table.

1 files changed, 8 insertions, 0 deletions

diff --git a/sql/sql_select.cc b/sql/sql_select.cc
index fa0be81ff3d..62f40eeb99c 100644
--- a/sql/sql_select.cc
+++ b/sql/sql_select.cc
@@ -11005,7 +11005,15 @@ uint check_join_cache_usage(JOIN_TAB *tab,
     effort now.
   */
   if (tab->table->pos_in_table_list->is_materialized_derived())
+  {
     no_bka_cache= true;
+    /*
+      Don't use hash join algorithm if the temporary table for the rows
+      of the derived table will be created with an equi-join key.
+    */
+    if (tab->table->s->keys)
+      no_hashed_cache= true;
+  }
 
   /*
     Don't use join buffering if we're dictated not to by no_jbuf_after