MDEV-20468: Allocating more space than required for JOIN_TAB array for a query with SJM table

sql/opt_subselect.cc

@@ -3934,6 +3934,39 @@ void fix_semijoin_strategies_for_picked_join_order(JOIN *join)
 }
 
 
+/*
+  Return the number of tables at the top-level of the JOIN
+
+  SYNOPSIS
+    get_number_of_tables_at_top_level()
+      join  The join with the picked join order
+
+  DESCRIPTION
+    The number of tables in the JOIN currently include all the inner tables of the
+    mergeable semi-joins. The function would make sure that we only count the semi-join
+    nest and not the inner tables of teh semi-join nest.
+*/
+
+uint get_number_of_tables_at_top_level(JOIN *join)
+{
+  uint j= 0, tables= 0;
+  while(j < join->table_count)
+  {
+    POSITION *cur_pos= &join->best_positions[j];
+    tables++;
+    if (cur_pos->sj_strategy == SJ_OPT_MATERIALIZE ||
+        cur_pos->sj_strategy == SJ_OPT_MATERIALIZE_SCAN)
+    {
+      SJ_MATERIALIZATION_INFO *sjm= cur_pos->table->emb_sj_nest->sj_mat_info;
+      j= j + sjm->tables;
+    }
+    else
+      j++;
+  }
+  return tables;
+}
+
+
 /*
   Setup semi-join materialization strategy for one semi-join nest
   

sql/opt_subselect.h

@@ -323,6 +323,7 @@ void fix_semijoin_strategies_for_picked_join_order(JOIN *join);
 
 bool setup_sj_materialization_part1(JOIN_TAB *sjm_tab);
 bool setup_sj_materialization_part2(JOIN_TAB *sjm_tab);
+uint get_number_of_tables_at_top_level(JOIN *join);
 
 
 /*

sql/sql_select.cc

@@ -3600,7 +3600,7 @@ bool JOIN::make_aggr_tables_info()
         unit->select_limit_cnt == 1 (we only need one row in the result set)
       */
       sort_tab->filesort->limit=
-        (has_group_by || (join_tab + table_count > curr_tab + 1)) ?
+        (has_group_by || (join_tab + top_join_tab_count > curr_tab + 1)) ?
          select_limit : unit->select_limit_cnt;
     }
     if (!only_const_tables() &&
@@ -10204,14 +10204,16 @@ bool JOIN::get_best_combination()
 
   if (aggr_tables > 2)
     aggr_tables= 2;
-  if (!(join_tab= (JOIN_TAB*) thd->alloc(sizeof(JOIN_TAB)*
-                                        (top_join_tab_count + aggr_tables))))
-    DBUG_RETURN(TRUE);
 
   full_join=0;
   hash_join= FALSE;
 
   fix_semijoin_strategies_for_picked_join_order(this);
+  top_join_tab_count= get_number_of_tables_at_top_level(this);
+
+  if (!(join_tab= (JOIN_TAB*) thd->alloc(sizeof(JOIN_TAB)*
+                                        (top_join_tab_count + aggr_tables))))
+    DBUG_RETURN(TRUE);
    
   JOIN_TAB_RANGE *root_range;
   if (!(root_range= new (thd->mem_root) JOIN_TAB_RANGE))
@@ -13923,7 +13925,7 @@ remove_const(JOIN *join,ORDER *first_order, COND *cond,
       ORDER BY and GROUP BY
     */
     for (JOIN_TAB *tab= join->join_tab + join->const_tables;
-         tab < join->join_tab + join->table_count;
+         tab < join->join_tab + join->top_join_tab_count;
          tab++)
       tab->cached_eq_ref_table= FALSE;
 
@@ -19772,8 +19774,7 @@ do_select(JOIN *join, Procedure *procedure)
 
     if (join->pushdown_query->store_data_in_temp_table)
     {
-      JOIN_TAB *last_tab= join->join_tab + join->table_count -
-                          join->exec_join_tab_cnt();      
+      JOIN_TAB *last_tab= join->join_tab + join->exec_join_tab_cnt();
       last_tab->next_select= end_send;
 
       enum_nested_loop_state state= last_tab->aggr->end_send();