Skip to content

M
MariaDB
Commit bc7369b7 authored by unknown's avatar unknown
Browse files
Options

MWL#89: Cost-based choice between Materialization and IN->EXISTS transformation 

Merge 5.3-mwl89 into 5.3 main.

There is one remaining test failure in this merge:
innodb_mysql_lock2. All other tests have been checked to
deliver the same results/explains as 5.3-mwl89, including
the few remaining wrong results.
parents 2dc2098e 9f2bddbd
Expand all Show whitespace changes
Inline Side-by-side
Showing with 5804 additions and 791 deletions
mysql-test/include/subselect_mat_cost.inc 0 → 100644
View file @ bc7369b7
-- echo
-- echo /* A. Subqueries in the SELECT clause. */
explain
select a1, a1 in (select b1 from t2 where b1 > '0') from t1;
select a1, a1 in (select b1 from t2 where b1 > '0') from t1;
-- echo
explain
select a1, a2, (a1, a2) in (select b1, b2 from t2 where b1 > '0') from t1;
select a1, a2, (a1, a2) in (select b1, b2 from t2 where b1 > '0') from t1;
-- echo
explain
select a1, a2, (a1, a2) in (select b1, b2 from t2 where b1 > '0' and b1 < '9') from t1;
select a1, a2, (a1, a2) in (select b1, b2 from t2 where b1 > '0' and b1 < '9') from t1;
-- echo
-- echo /*
-- echo B. "Natural" examples of subqueries without grouping that
-- echo cannot be flattened into semijoin.
-- echo */
explain
select a1 from t1 where a1 in (select b2 from t2) or a2 < '9';
select a1 from t1 where a1 in (select b2 from t2) or a2 < '9';
-- echo
explain
select a1, a2 from t1 where (a1, a2) in (select b1, b2 from t2 where b1 > '0') or a2 < '9';
select a1, a2 from t1 where (a1, a2) in (select b1, b2 from t2 where b1 > '0') or a2 < '9';
-- echo UNION subqueries are currently limited to only use IN-TO-EXISTS.
explain
select a2 from t1 where a2 in (select b2 from t2 UNION select b3 from t2 as t3);
select a2 from t1 where a2 in (select b2 from t2 UNION select b3 from t2 as t3);
-- echo
explain
select a1 from t1 where a1 = '1 - 02' and a1 in (select max(b1) from t2 where b2 = '2 - 02');
select a1 from t1 where a1 = '1 - 02' and a1 in (select max(b1) from t2 where b2 = '2 - 02');
-- echo
explain
select a1, a2 from t1 where (a1, a2) in (select b1, b2 from t2 order by b3);
select a1, a2 from t1 where (a1, a2) in (select b1, b2 from t2 order by b3);
-- echo
-- echo /* C. Subqueries in the WHERE clause with GROUP BY. */
explain
select * from t1 where a1 in (select b1 from t2 where b1 > '0' group by b1);
select * from t1 where a1 in (select b1 from t2 where b1 > '0' group by b1);
-- echo
explain
select * from t1 where (a1, a2) in (select b1, b2 from t2 where b1 > '0' group by b1, b2);
select * from t1 where (a1, a2) in (select b1, b2 from t2 where b1 > '0' group by b1, b2);
-- echo
explain
select * from t1 where (a1, a2) in (select b1, b2 from t2 where b1 > '0' group by b1, b2 having b2 < '2 - 04');
select * from t1 where (a1, a2) in (select b1, b2 from t2 where b1 > '0' group by b1, b2 having b2 < '2 - 04');
-- echo
explain
select * from t1 where (a1, a2, a3) in (select b1, b2, b3 from t2 group by b1, b2, b3);
select * from t1 where (a1, a2, a3) in (select b1, b2, b3 from t2 group by b1, b2, b3);
-- echo
explain
select * from t1 where (a1, a2, a3) in (select b1, b2, b3 from t2 where b3 = '3 - 02' group by b1, b2);
select * from t1 where (a1, a2, a3) in (select b1, b2, b3 from t2 where b3 = '3 - 02' group by b1, b2);
-- echo
explain
select * from t1 where (a1,a2,a3) in (select b1,b2,b3 from t2 where b1 = '1 - 01' group by b1,b2,b3);
select * from t1 where (a1,a2,a3) in (select b1,b2,b3 from t2 where b1 = '1 - 01' group by b1,b2,b3);
-- echo
-- echo /*
-- echo D. Subqueries for which materialization is not possible, and the
-- echo optimizer reverts to in-to-exists.
-- echo */
# The first two cases are rejected during the prepare phase by the procedure
# subquery_types_allow_materialization().
explain
select left(a1,7), left(a2,7) from t1_1024 where a1 in (select b1 from t2_1024 where b1 > '0') or a2 < '9';
select left(a1,7), left(a2,7) from t1_1024 where a1 in (select b1 from t2_1024 where b1 > '0') or a2 < '9';
explain
select left(a1,7), left(a2,7) from t1_1024 where (a1,a2) in (select b1, b2 from t2_1024 where b1 > '0') or a2 < '9';
select left(a1,7), left(a2,7) from t1_1024 where (a1,a2) in (select b1, b2 from t2_1024 where b1 > '0') or a2 < '9';
-- echo
# The following two subqueries return the result of a string function with a
# blob argument, where the return type may be != blob. These are rejected during
# cost-based optimization when attempting to create a temporary table.
explain
select left(a1,7), left(a2,7) from t1_1024 where a1 in (select substring(b1,1,1024) from t2_1024 where b1 > '0') or a2 < '9';
select left(a1,7), left(a2,7) from t1_1024 where a1 in (select substring(b1,1,1024) from t2_1024 where b1 > '0') or a2 < '9';
explain
select left(a1,7), left(a2,7) from t1_1024 where (a1,a2) in (select substring(b1,1,1024), substring(b2,1,1024) from t2_1024 where b1 > '0') or a2 < '9';
select left(a1,7), left(a2,7) from t1_1024 where (a1,a2) in (select substring(b1,1,1024), substring(b2,1,1024) from t2_1024 where b1 > '0') or a2 < '9';
-- echo
-- echo
-- echo /* E. Edge cases. */
-- echo
-- echo /* E.1 Both materialization and in_to_exists cannot be off. */
set @save_optimizer_switch=@@optimizer_switch;
set @@optimizer_switch = 'materialization=off,in_to_exists=off';
--error ER_ILLEGAL_SUBQUERY_OPTIMIZER_SWITCHES
select * from t1 where a1 in (select b1 from t2 where b1 > '0' group by b1);
set @@optimizer_switch = @save_optimizer_switch;
-- echo /* E.2 Outer query without tables, always uses IN-TO-EXISTS. */
explain
select '1 - 03' in (select b1 from t2 where b1 > '0');
select '1 - 03' in (select b1 from t2 where b1 > '0');
-- echo /* E.3 Subqueries without tables. */
explain
select a1 from t1 where a1 in (select '1 - 03') or a2 < '9';
select a1 from t1 where a1 in (select '1 - 03') or a2 < '9';
-- echo UNION subqueries are currently limited to only use IN-TO-EXISTS.
explain
select a1 from t1 where a1 in (select '1 - 03' UNION select '1 - 02');
select a1 from t1 where a1 in (select '1 - 03' UNION select '1 - 02');
-- echo /* E.4 optimize_cond detects FALSE where/having clause. */
explain
select a1 from t1 where a1 in (select b1 from t2 where b1 = b2 and b2 = '1 - 03' and b1 = '1 - 02' ) or a2 < '9';
select a1 from t1 where a1 in (select b1 from t2 where b1 = b2 and b2 = '1 - 03' and b1 = '1 - 02' ) or a2 < '9';
-- echo /* E.5 opt_sum_query detects no matching min/max row or substitutes MIN/MAX with a const. */
-- echo TODO this test produces wrong result due to missing logic to handle the case
-- echo when JOIN::optimize detects an empty subquery result.
explain
select a1 from t1 where a1 in (select max(b1) from t2);
select a1 from t1 where a1 in (select max(b1) from t2);
-- echo
explain
select a1 from t1 where a1 in (select max(b1) from t2 where b1 = '7 - 02');
select a1 from t1 where a1 in (select max(b1) from t2 where b1 = '7 - 02');
-- echo /* E.6 make_join_select detects impossible WHERE. *
-- echo TODO
-- echo /* E.7 constant optimization detects "no matching row in const table". */
-- echo TODO
-- echo /* E.8 Impossible WHERE noticed after reading const tables. */
explain
select '1 - 03' in (select b1 from t2 where b1 > '0' and b1 < '0');
select '1 - 03' in (select b1 from t2 where b1 > '0' and b1 < '0');
-- echo
-- echo /* F. UPDATE/DELETE with subqueries. */
-- echo
-- echo TODO
-- echo
mysql-test/r/subselect3.result
View file @ bc7369b7
......@@ -840,11 +840,16 @@ x ROW(11, 12) = (SELECT MAX(x), 22) ROW(11, 12) IN (SELECT MAX(x), 22)
1 0 0
2 0 0
11 0 0
# 2nd and 3rd columns should be same for x == 11 only
# 2nd and 3rd columns should be same
EXPLAIN SELECT a AS x, ROW(11, 12) = (SELECT MAX(x), 12), ROW(11, 12) IN (SELECT MAX(x), 12) FROM t1;
id select_type table type possible_keys key key_len ref rows Extra
1 PRIMARY t1 ALL NULL NULL NULL NULL 3
3 DEPENDENT SUBQUERY NULL NULL NULL NULL NULL NULL NULL No tables used
2 DEPENDENT SUBQUERY NULL NULL NULL NULL NULL NULL NULL No tables used
SELECT a AS x, ROW(11, 12) = (SELECT MAX(x), 12), ROW(11, 12) IN (SELECT MAX(x), 12) FROM t1;
x ROW(11, 12) = (SELECT MAX(x), 12) ROW(11, 12) IN (SELECT MAX(x), 12)
1 0 1
2 0 1
1 0 0
2 0 0
11 1 1
DROP TABLE t1;
# both columns should be same
......
mysql-test/r/subselect3_jcl6.result
View file @ bc7369b7
......@@ -844,11 +844,16 @@ x ROW(11, 12) = (SELECT MAX(x), 22) ROW(11, 12) IN (SELECT MAX(x), 22)
1 0 0
2 0 0
11 0 0
# 2nd and 3rd columns should be same for x == 11 only
# 2nd and 3rd columns should be same
EXPLAIN SELECT a AS x, ROW(11, 12) = (SELECT MAX(x), 12), ROW(11, 12) IN (SELECT MAX(x), 12) FROM t1;
id select_type table type possible_keys key key_len ref rows Extra
1 PRIMARY t1 ALL NULL NULL NULL NULL 3
3 DEPENDENT SUBQUERY NULL NULL NULL NULL NULL NULL NULL No tables used
2 DEPENDENT SUBQUERY NULL NULL NULL NULL NULL NULL NULL No tables used
SELECT a AS x, ROW(11, 12) = (SELECT MAX(x), 12), ROW(11, 12) IN (SELECT MAX(x), 12) FROM t1;
x ROW(11, 12) = (SELECT MAX(x), 12) ROW(11, 12) IN (SELECT MAX(x), 12)
1 0 1
2 0 1
1 0 0
2 0 0
11 1 1
DROP TABLE t1;
# both columns should be same
......
mysql-test/r/subselect_mat.result
View file @ bc7369b7
......@@ -30,7 +30,7 @@ create index it3i3 on t3i (c1, c2);
insert into t1i select * from t1;
insert into t2i select * from t2;
insert into t3i select * from t3;
set @@optimizer_switch='semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
/******************************************************************************
* Simple tests.
******************************************************************************/
......@@ -176,33 +176,33 @@ a1 a2
1 - 02 2 - 02
select * from t1 where (a1, a2) in (select b1, min(b2) from t2i limit 1,1);
ERROR 42000: This version of MySQL doesn't yet support 'LIMIT & IN/ALL/ANY/SOME subquery'
set @@optimizer_switch='default,semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=on';
prepare st1 from
"select * from t1 where (a1, a2) in (select b1, min(b2) from t2 where b1 > '0' group by b1)";
set @@optimizer_switch='default,materialization=off';
set @@optimizer_switch='materialization=off,in_to_exists=on,semijoin=on';
execute st1;
a1 a2
1 - 01 2 - 01
1 - 02 2 - 02
set @@optimizer_switch='default,semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=on';
execute st1;
a1 a2
1 - 01 2 - 01
1 - 02 2 - 02
set @@optimizer_switch='default,materialization=off';
set @@optimizer_switch='materialization=off,in_to_exists=on,semijoin=on';
prepare st1 from
"select * from t1 where (a1, a2) in (select b1, min(b2) from t2 where b1 > '0' group by b1)";
set @@optimizer_switch='default,semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
execute st1;
a1 a2
1 - 01 2 - 01
1 - 02 2 - 02
set @@optimizer_switch='default,materialization=off';
set @@optimizer_switch='materialization=off,in_to_exists=on,semijoin=on';
execute st1;
a1 a2
1 - 01 2 - 01
1 - 02 2 - 02
set @@optimizer_switch='default,semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
explain extended
select * from t1 where (a1, a2) in (select b1, b2 from t2 order by b1, b2);
id select_type table type possible_keys key key_len ref rows filtered Extra
......@@ -549,7 +549,7 @@ a1 a2
Test that BLOBs are not materialized (except when arguments of some functions).
*/
# force materialization to be always considered
set @@optimizer_switch='semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
set @prefix_len = 6;
set @blob_len = 16;
set @suffix_len = @blob_len - @prefix_len;
......@@ -951,7 +951,7 @@ insert into t1bit values (b'010', b'110');
insert into t2bit values (b'001', b'101');
insert into t2bit values (b'010', b'110');
insert into t2bit values (b'110', b'111');
set @@optimizer_switch='semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
explain extended select bin(a1), bin(a2)
from t1bit
where (a1, a2) in (select b1, b2 from t2bit);
......@@ -994,7 +994,7 @@ drop table t1, t2, t3, t1i, t2i, t3i, columns;
/******************************************************************************
* Test the cache of the left operand of IN.
******************************************************************************/
set @@optimizer_switch='semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
create table t1 (s1 int);
create table t2 (s2 int);
insert into t1 values (5),(1),(0);
......@@ -1136,27 +1136,40 @@ drop table t2;
create table t1 (a1 int key);
create table t2 (b1 int);
insert into t1 values (5);
Only the last query returns correct result. Filed as BUG#40037.
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
explain select min(a1) from t1 where 7 in (select b1 from t2 group by b1);
id select_type table type possible_keys key key_len ref rows Extra
1 PRIMARY NULL NULL NULL NULL NULL NULL NULL Select tables optimized away
2 SUBQUERY t2 system NULL NULL NULL NULL 0 const row not found
select min(a1) from t1 where 7 in (select b1 from t2 group by b1);
min(a1)
set @@optimizer_switch='default,materialization=off';
NULL
set @@optimizer_switch='materialization=off,in_to_exists=on,semijoin=off';
explain select min(a1) from t1 where 7 in (select b1 from t2 group by b1);
id select_type table type possible_keys key key_len ref rows Extra
1 PRIMARY NULL NULL NULL NULL NULL NULL NULL Select tables optimized away
2 DEPENDENT SUBQUERY t2 system NULL NULL NULL NULL 0 const row not found
select min(a1) from t1 where 7 in (select b1 from t2 group by b1);
min(a1)
set @@optimizer_switch='default,semijoin=off';
NULL
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
explain select min(a1) from t1 where 7 in (select b1 from t2);
id select_type table type possible_keys key key_len ref rows Extra
1 PRIMARY NULL NULL NULL NULL NULL NULL NULL Select tables optimized away
2 SUBQUERY t2 system NULL NULL NULL NULL 0 const row not found
select min(a1) from t1 where 7 in (select b1 from t2);
min(a1)
set @@optimizer_switch='default,materialization=off';
NULL
set @@optimizer_switch='materialization=off,in_to_exists=on,semijoin=off';
explain select min(a1) from t1 where 7 in (select b1 from t2);
id select_type table type possible_keys key key_len ref rows Extra
1 PRIMARY NULL NULL NULL NULL NULL NULL NULL Select tables optimized away
2 DEPENDENT SUBQUERY NULL NULL NULL NULL NULL NULL NULL Impossible WHERE noticed after reading const tables
select min(a1) from t1 where 7 in (select b1 from t2);
min(a1)
NULL
set @@optimizer_switch='materialization=off,in_to_exists=off,semijoin=on';
explain select min(a1) from t1 where 7 in (select b1 from t2);
id select_type table type possible_keys key key_len ref rows Extra
1 PRIMARY NULL NULL NULL NULL NULL NULL NULL Impossible WHERE noticed after reading const tables
......@@ -1167,7 +1180,7 @@ drop table t1,t2;
create table t1 (a char(2), b varchar(10));
insert into t1 values ('a', 'aaa');
insert into t1 values ('aa', 'aaaa');
set @@optimizer_switch='default,semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
explain select a,b from t1 where b in (select a from t1);
id select_type table type possible_keys key key_len ref rows Extra
1 PRIMARY t1 ALL NULL NULL NULL NULL 2 Using where
......@@ -1187,7 +1200,7 @@ INSERT INTO t1 (f1, f2) VALUES (10, 1.668);
CREATE TABLE t2 LIKE t1;
INSERT INTO t2 VALUES (1, 1.789);
INSERT INTO t2 VALUES (13, 1.454);
SET @@optimizer_switch='default,semijoin=on,materialization=on';
SET @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=on';
EXPLAIN SELECT COUNT(*) FROM t1 WHERE (f1,f2) IN (SELECT f1,f2 FROM t2);
id select_type table type possible_keys key key_len ref rows Extra
1 PRIMARY subselect2 ALL unique_key NULL NULL NULL 2
......@@ -1208,7 +1221,7 @@ PRIMARY KEY (pk)
INSERT INTO t1 VALUES (1,'o','ffff','ffff','ffoo'),(2,'f','ffff','ffff','ffff');
CREATE TABLE t2 LIKE t1;
INSERT INTO t2 VALUES (1,'i','iiii','iiii','iiii'),(2,'f','ffff','ffff','ffff');
SET @@optimizer_switch='default,semijoin=on,materialization=on';
SET @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=on';
EXPLAIN SELECT pk FROM t1 WHERE (a) IN (SELECT a FROM t2 WHERE pk > 0);
id select_type table type possible_keys key key_len ref rows Extra
1 PRIMARY t1 ALL NULL NULL NULL NULL 2
......@@ -1237,7 +1250,7 @@ i
3
4
set @save_optimizer_switch=@@optimizer_switch;
set session optimizer_switch='materialization=off';
set session optimizer_switch='materialization=off,in_to_exists=on';
select * from t1 where t1.i in (select t2.i from t2 join t3 where t2.i + t3.i = 5);
i
1
......
mysql-test/r/subselect_mat_cost.result 0 → 100644
View file @ bc7369b7
This source diff could not be displayed because it is too large. You can view the blob instead.
mysql-test/t/subselect3.test
View file @ bc7369b7
......@@ -681,7 +681,8 @@ SELECT a, ROW(11, 12) = (SELECT a, 12), ROW(11, 12) IN (SELECT a, 12) FROM t1;
# The x alias is used below to workaround bug #40674.
# Regression tests for sum function on outer column in subselect from dual:
SELECT a AS x, ROW(11, 12) = (SELECT MAX(x), 22), ROW(11, 12) IN (SELECT MAX(x), 22) FROM t1;
--echo # 2nd and 3rd columns should be same for x == 11 only
--echo # 2nd and 3rd columns should be same
EXPLAIN SELECT a AS x, ROW(11, 12) = (SELECT MAX(x), 12), ROW(11, 12) IN (SELECT MAX(x), 12) FROM t1;
SELECT a AS x, ROW(11, 12) = (SELECT MAX(x), 12), ROW(11, 12) IN (SELECT MAX(x), 12) FROM t1;
DROP TABLE t1;
......
mysql-test/t/subselect_mat.test
View file @ bc7369b7
......@@ -48,7 +48,7 @@ insert into t2i select * from t2;
insert into t3i select * from t3;
# force the use of materialization
set @@optimizer_switch='semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
/******************************************************************************
* Simple tests.
......@@ -111,22 +111,22 @@ select * from t1 where (a1, a2) in (select b1, min(b2) from t2i limit 1,1);
# test re-optimization/re-execution with different execution methods
# prepare once, exec with different modes
set @@optimizer_switch='default,semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=on';
prepare st1 from
"select * from t1 where (a1, a2) in (select b1, min(b2) from t2 where b1 > '0' group by b1)";
set @@optimizer_switch='default,materialization=off';
set @@optimizer_switch='materialization=off,in_to_exists=on,semijoin=on';
execute st1;
set @@optimizer_switch='default,semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=on';
execute st1;
set @@optimizer_switch='default,materialization=off';
set @@optimizer_switch='materialization=off,in_to_exists=on,semijoin=on';
prepare st1 from
"select * from t1 where (a1, a2) in (select b1, min(b2) from t2 where b1 > '0' group by b1)";
set @@optimizer_switch='default,semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
execute st1;
set @@optimizer_switch='default,materialization=off';
set @@optimizer_switch='materialization=off,in_to_exists=on,semijoin=on';
execute st1;
set @@optimizer_switch='default,semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
# materialize the result of ORDER BY
# non-indexed fields
......@@ -327,7 +327,7 @@ select * from t1 order by (select col from columns limit 1);
Test that BLOBs are not materialized (except when arguments of some functions).
*/
# force materialization to be always considered
set @@optimizer_switch='semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
set @prefix_len = 6;
# BLOB == 16 (small blobs that could be stored in HEAP tables)
......@@ -680,7 +680,7 @@ insert into t2bit values (b'001', b'101');
insert into t2bit values (b'010', b'110');
insert into t2bit values (b'110', b'111');
set @@optimizer_switch='semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
explain extended select bin(a1), bin(a2)
from t1bit
......@@ -718,7 +718,7 @@ drop table t1, t2, t3, t1i, t2i, t3i, columns;
/******************************************************************************
* Test the cache of the left operand of IN.
******************************************************************************/
set @@optimizer_switch='semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
# Test that default values of Cached_item are not used for comparison
create table t1 (s1 int);
......@@ -812,23 +812,28 @@ drop table t2;
create table t1 (a1 int key);
create table t2 (b1 int);
insert into t1 values (5);
-- echo Only the last query returns correct result. Filed as BUG#40037.
# Query with group by, executed via materialization
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
explain select min(a1) from t1 where 7 in (select b1 from t2 group by b1);
select min(a1) from t1 where 7 in (select b1 from t2 group by b1);
# Query with group by, executed via IN=>EXISTS
set @@optimizer_switch='default,materialization=off';
set @@optimizer_switch='materialization=off,in_to_exists=on,semijoin=off';
explain select min(a1) from t1 where 7 in (select b1 from t2 group by b1);
select min(a1) from t1 where 7 in (select b1 from t2 group by b1);
# Executed with materialization
set @@optimizer_switch='default,semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
explain select min(a1) from t1 where 7 in (select b1 from t2);
select min(a1) from t1 where 7 in (select b1 from t2);
# Executed via IN=>EXISTS
set @@optimizer_switch='materialization=off,in_to_exists=on,semijoin=off';
explain select min(a1) from t1 where 7 in (select b1 from t2);
select min(a1) from t1 where 7 in (select b1 from t2);
# Executed with semi-join. Notice, this time we get a different result (NULL).
# This is the only correct result of all four queries. This difference is
# This is the only correct result of all five queries. This difference is
# filed as BUG#40037.
set @@optimizer_switch='default,materialization=off';
set @@optimizer_switch='materialization=off,in_to_exists=off,semijoin=on';
explain select min(a1) from t1 where 7 in (select b1 from t2);
select min(a1) from t1 where 7 in (select b1 from t2);
drop table t1,t2;
......@@ -840,7 +845,7 @@ create table t1 (a char(2), b varchar(10));
insert into t1 values ('a', 'aaa');
insert into t1 values ('aa', 'aaaa');
set @@optimizer_switch='default,semijoin=off';
set @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=off';
explain select a,b from t1 where b in (select a from t1);
select a,b from t1 where b in (select a from t1);
prepare st1 from "select a,b from t1 where b in (select a from t1)";
......@@ -861,7 +866,7 @@ CREATE TABLE t2 LIKE t1;
INSERT INTO t2 VALUES (1, 1.789);
INSERT INTO t2 VALUES (13, 1.454);
SET @@optimizer_switch='default,semijoin=on,materialization=on';
SET @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=on';
EXPLAIN SELECT COUNT(*) FROM t1 WHERE (f1,f2) IN (SELECT f1,f2 FROM t2);
SELECT COUNT(*) FROM t1 WHERE (f1,f2) IN (SELECT f1,f2 FROM t2);
......@@ -883,7 +888,7 @@ INSERT INTO t1 VALUES (1,'o','ffff','ffff','ffoo'),(2,'f','ffff','ffff','ffff');
CREATE TABLE t2 LIKE t1;
INSERT INTO t2 VALUES (1,'i','iiii','iiii','iiii'),(2,'f','ffff','ffff','ffff');
SET @@optimizer_switch='default,semijoin=on,materialization=on';
SET @@optimizer_switch='materialization=on,in_to_exists=off,semijoin=on';
EXPLAIN SELECT pk FROM t1 WHERE (a) IN (SELECT a FROM t2 WHERE pk > 0);
SELECT pk FROM t1 WHERE (a) IN (SELECT a FROM t2 WHERE pk > 0);
SELECT pk FROM t1 WHERE (b,c,d) IN (SELECT b,c,d FROM t2 WHERE pk > 0);
......@@ -900,7 +905,7 @@ create table t3(i int);
insert into t3 values (1), (2), (3), (4), (5), (6), (7), (8), (9), (10);
select * from t1 where t1.i in (select t2.i from t2 join t3 where t2.i + t3.i = 5);
set @save_optimizer_switch=@@optimizer_switch;
set session optimizer_switch='materialization=off';
set session optimizer_switch='materialization=off,in_to_exists=on';
select * from t1 where t1.i in (select t2.i from t2 join t3 where t2.i + t3.i = 5);
set session optimizer_switch=@save_optimizer_switch;
drop table t1, t2, t3;
......
mysql-test/t/subselect_mat_cost.test 0 → 100644
View file @ bc7369b7
#
# Tets of cost-based choice between the materialization and in-to-exists
# subquery execution strategies (MWL#89)
#
--disable_warnings
drop table if exists t1, t2, t1_1024, t2_1024;
drop procedure if exists make_t1_indexes;
drop procedure if exists make_t2_indexes;
drop procedure if exists remove_t1_indexes;
drop procedure if exists remove_t2_indexes;
drop procedure if exists add_materialization_data;
drop procedure if exists delete_materialization_data;
drop procedure if exists set_all_columns_not_null;
drop procedure if exists set_all_columns_nullable;
--enable_warnings
create table t1 (a1 char(8), a2 char(8), a3 char(8), a4 int);
insert into t1 values ('1 - 00', '2 - 00', '3 - 00', 0);
insert into t1 values ('1 - 01', '2 - 01', '3 - 01', 1);
insert into t1 values ('1 - 02', '2 - 02', '3 - 02', 2);
create table t2 (b1 char(8), b2 char(8), b3 char(8), b4 int);
insert into t2 values ('1 - 01', '2 - 01', '3 - 01', 1);
insert into t2 values ('1 - 01', '2 - 01', '3 - 02', 2);
insert into t2 values ('1 - 02', '2 - 02', '3 - 03', 3);
insert into t2 values ('1 - 02', '2 - 02', '3 - 04', 4);
insert into t2 values ('1 - 03', '2 - 03', '3 - 05', 5);
create table t1_1024 (a1 blob(1024), a2 blob(1024));
insert into t1_1024 values (concat('1 - 00', repeat('x', 1018)), concat('2 - 00', repeat('x', 1018)));
insert into t1_1024 values (concat('1 - 01', repeat('x', 1018)), concat('2 - 01', repeat('x', 1018)));
create table t2_1024 (b1 blob(1024), b2 blob(1024));
insert into t2_1024 values (concat('1 - 01', repeat('x', 1018)), concat('2 - 01', repeat('x', 1018)));
insert into t2_1024 values (concat('1 - 02', repeat('x', 1018)), concat('2 - 02', repeat('x', 1018)));
insert into t2_1024 values (concat('1 - 03', repeat('x', 1018)), concat('2 - 03', repeat('x', 1018)));
insert into t2_1024 values (concat('1 - 04', repeat('x', 1018)), concat('2 - 04', repeat('x', 1018)));
delimiter |;
create procedure make_t1_indexes()
begin
create index it1i1 on t1 (a1);
create index it1i2 on t1 (a2);
create index it1i3 on t1 (a1, a2);
create index it1_1024i1 on t1_1024 (a1(6));
create index it1_1024i2 on t1_1024 (a2(6));
create index it1_1024i3 on t1_1024 (a1(6), a2(6));
end|
create procedure make_t2_indexes()
begin
create index it2i1 on t2 (b1);
create index it2i2 on t2 (b2);
create index it2i3 on t2 (b1, b2);
create unique index it2i4 on t2 (b1, b2, b3);
create index it2_1024i1 on t2_1024 (b1(6));
create index it2_1024i2 on t2_1024 (b2(6));
create index it2_1024i3 on t2_1024 (b1(6), b2(6));
end|
create procedure remove_t1_indexes()
begin
drop index it1i1 on t1;
drop index it1i2 on t1;
drop index it1i3 on t1;
drop index it1_1024i1 on t1_1024;
drop index it1_1024i2 on t1_1024;
drop index it1_1024i3 on t1_1024;
end|
create procedure remove_t2_indexes()
begin
drop index it2i1 on t2;
drop index it2i2 on t2;
drop index it2i3 on t2;
drop index it2i4 on t2;
drop index it2_1024i1 on t2_1024;
drop index it2_1024i2 on t2_1024;
drop index it2_1024i3 on t2_1024;
end|
create procedure add_materialization_data()
begin
insert into t1 values ('1 - 03', '2 - 03', '3 - 03', 3);
insert into t1 values ('1 - 04', '2 - 04', '3 - 04', 4);
insert into t1 values ('1 - 05', '2 - 05', '3 - 05', 5);
insert into t1 values ('1 - 06', '2 - 06', '3 - 06', 6);
insert into t1 values ('1 - 07', '2 - 07', '3 - 07', 7);
insert into t1_1024 values (concat('1 - 03', repeat('x', 1018)), concat('2 - 03', repeat('x', 1018)));
end|
create procedure delete_materialization_data()
begin
delete from t1 where a1 >= '1 - 03';
delete from t1_1024 where a1 >= '1 - 03';
end|
create procedure set_all_columns_not_null()
begin
alter table t1 modify a1 char(8) not null, modify a2 char(8) not null, modify a3 char(8) not null;
alter table t2 modify b1 char(8) not null, modify b2 char(8) not null, modify b3 char(8) not null;
end|
create procedure set_all_columns_nullable()
begin
alter table t1 modify a1 char(8) null, modify a2 char(8) null, modify a3 char(8) null;
alter table t2 modify b1 char(8) null, modify b2 char(8) null, modify b3 char(8) null;
end|
delimiter ;|
-- echo
-- echo /******************************************************************************
-- echo 1. Both materialization and in-to-exists are ON, make a cost-based choice.
-- echo ******************************************************************************/
set @@optimizer_switch='materialization=on,in_to_exists=on';
-- echo
-- echo /* 1.1 In-to-exists is cheaper */
call make_t1_indexes();
-- echo /* 1.1.1 non-indexed table access */
-- source include/subselect_mat_cost.inc
-- echo /* 1.1.2 indexed table access, nullabale columns. */
call make_t2_indexes();
-- source include/subselect_mat_cost.inc
-- echo /* 1.1.3 indexed table access, non-nullabale columns. */
call set_all_columns_not_null();
-- source include/subselect_mat_cost.inc
call set_all_columns_nullable();
-- echo
-- echo /* 1.2 Materialization is cheaper */
# make materialization cheaper
call add_materialization_data();
call remove_t1_indexes();
-- echo /* 1.2.1 non-indexed table access */
call remove_t2_indexes();
-- source include/subselect_mat_cost.inc
-- echo /* 1.2.2 indexed table access, nullabale columns. */
call make_t2_indexes();
-- source include/subselect_mat_cost.inc
-- echo /* 1.2.3 indexed table access, non-nullabale columns. */
call set_all_columns_not_null();
-- source include/subselect_mat_cost.inc
call set_all_columns_nullable();
-- echo /******************************************************************************
-- echo 2. Materialization is OFF, in-to-exists is ON, materialization is cheaper.
-- echo ******************************************************************************/
set @@optimizer_switch='materialization=off,in_to_exists=on';
-- echo /* 2.1 non-indexed table access */
call remove_t2_indexes();
-- source include/subselect_mat_cost.inc
-- echo /* 2.2 indexed table access, nullabale columns. */
call make_t2_indexes();
-- source include/subselect_mat_cost.inc
-- echo /* 2.3 indexed table access, non-nullabale columns. */
call set_all_columns_not_null();
-- source include/subselect_mat_cost.inc
call set_all_columns_nullable();
-- echo /******************************************************************************
-- echo 3. Materialization is ON, in-to-exists is OFF, in-to-exists is cheaper.
-- echo ******************************************************************************/
set @@optimizer_switch='materialization=on,in_to_exists=off';
# make IN-TO-EXISTS cheaper
call delete_materialization_data();
call make_t1_indexes();
-- echo /* 3.1 non-indexed table access */
call remove_t2_indexes();
-- source include/subselect_mat_cost.inc
-- echo /* 3.2 indexed table access, nullabale columns. */
call make_t2_indexes();
-- source include/subselect_mat_cost.inc
-- echo /* 3.3 indexed table access, non-nullabale columns. */
call set_all_columns_not_null();
-- source include/subselect_mat_cost.inc
call set_all_columns_nullable();
drop procedure make_t1_indexes;
drop procedure make_t2_indexes;
drop procedure remove_t1_indexes;
drop procedure remove_t2_indexes;
drop procedure add_materialization_data;
drop procedure delete_materialization_data;
drop procedure set_all_columns_not_null;
drop procedure set_all_columns_nullable;
drop table t1, t2, t1_1024, t2_1024;
--echo #
--echo # LP BUG#643424 valgrind warning in choose_subquery_plan()
--echo #
CREATE TABLE t1 (
pk int(11) NOT NULL AUTO_INCREMENT,
c1 int(11) DEFAULT NULL,
c2 int(11) DEFAULT NULL,
PRIMARY KEY (pk),
KEY c2 (c2));
INSERT INTO t1 VALUES (1,NULL,2);
INSERT INTO t1 VALUES (2,7,9);
INSERT INTO t1 VALUES (9,NULL,8);
CREATE TABLE t2 (
pk int(11) NOT NULL AUTO_INCREMENT,
c1 int(11) DEFAULT NULL,
c2 int(11) DEFAULT NULL,
PRIMARY KEY (pk),
KEY c2 (c2));
INSERT INTO t2 VALUES (1,1,7);
set @save_optimizer_switch=@@optimizer_switch;
set @@optimizer_switch='materialization=on,in_to_exists=on,semijoin=off';
SELECT pk FROM t1 WHERE (c2, c1) IN (SELECT c2, c2 FROM t2);
set session optimizer_switch=@save_optimizer_switch;
drop table t1, t2;
--echo #
--echo # LP BUG#652727 Crash in create_ref_for_key()
--echo #
CREATE TABLE t2 (
pk int(11) NOT NULL AUTO_INCREMENT,
c1 int(11) DEFAULT NULL,
PRIMARY KEY (pk));
INSERT INTO t2 VALUES (10,7);
INSERT INTO t2 VALUES (11,1);
INSERT INTO t2 VALUES (17,NULL);
CREATE TABLE t1 (
pk int(11) NOT NULL AUTO_INCREMENT,
c1 int(11) DEFAULT NULL,
PRIMARY KEY (pk));
INSERT INTO t1 VALUES (15,1);
INSERT INTO t1 VALUES (19,NULL);
CREATE TABLE t3 (c2 int(11) DEFAULT NULL, KEY c2 (c2));
INSERT INTO t3 VALUES (1);
set @save_optimizer_switch=@@optimizer_switch;
set @@optimizer_switch='materialization=on,in_to_exists=on,semijoin=off';
SELECT c2
FROM t3
WHERE (2, 6) IN (SELECT t1.c1, t1.c1 FROM t1 STRAIGHT_JOIN t2 ON t2.pk = t1.pk);
set session optimizer_switch=@save_optimizer_switch;
drop table t1, t2, t3;
sql/item.h
View file @ bc7369b7
......@@ -948,6 +948,8 @@ class Item {
virtual bool register_field_in_read_map(uchar *arg) { return 0; }
virtual bool enumerate_field_refs_processor(uchar *arg) { return 0; }
virtual bool mark_as_eliminated_processor(uchar *arg) { return 0; }
virtual bool eliminate_subselect_processor(uchar *arg) { return 0; }
virtual bool set_fake_select_as_master_processor(uchar *arg) { return 0; }
/* To call bool function for all arguments */
struct bool_func_call_args
......
sql/item_cmpfunc.cc
View file @ bc7369b7
......@@ -1985,6 +1985,18 @@ Item *Item_in_optimizer::transform(Item_transformer transformer, uchar *argument
}
bool Item_in_optimizer::is_expensive_processor(uchar *arg)
{
return args[1]->is_expensive_processor(arg);
}
bool Item_in_optimizer::is_expensive()
{
return args[1]->is_expensive();
}
longlong Item_func_eq::val_int()
{
DBUG_ASSERT(fixed == 1);
......@@ -4650,12 +4662,6 @@ Item *and_expressions(Item *a, Item *b, Item **org_item)
longlong Item_func_isnull::val_int()
{
DBUG_ASSERT(fixed == 1);
/*
Handle optimization if the argument can't be null
This has to be here because of the test in update_used_tables().
*/
if (!used_tables_cache && !with_subselect)
return cached_value;
return args[0]->is_null() ? 1: 0;
}
......@@ -4663,12 +4669,6 @@ longlong Item_is_not_null_test::val_int()
{
DBUG_ASSERT(fixed == 1);
DBUG_ENTER("Item_is_not_null_test::val_int");
if (!used_tables_cache && !with_subselect)
{
owner->was_null|= (!cached_value);
DBUG_PRINT("info", ("cached: %ld", (long) cached_value));
DBUG_RETURN(cached_value);
}
if (args[0]->is_null())
{
DBUG_PRINT("info", ("null"));
......@@ -4685,19 +4685,9 @@ longlong Item_is_not_null_test::val_int()
void Item_is_not_null_test::update_used_tables()
{
if (!args[0]->maybe_null)
{
used_tables_cache= 0; /* is always true */
cached_value= (longlong) 1;
}
else
{
args[0]->update_used_tables();
if (!(used_tables_cache=args[0]->used_tables()) && !with_subselect)
{
/* Remember if the value is always NULL or never NULL */
cached_value= (longlong) !args[0]->is_null();
}
}
}
......@@ -5373,7 +5363,7 @@ Item *Item_func_nop_all::neg_transformer(THD *thd)
/* "NOT (e $cmp$ ANY (SELECT ...)) -> e $rev_cmp$" ALL (SELECT ...) */
Item_func_not_all *new_item= new Item_func_not_all(args[0]);
Item_allany_subselect *allany= (Item_allany_subselect*)args[0];
allany->func= allany->func_creator(FALSE);
allany->create_comp_func(FALSE);
allany->all= !allany->all;
allany->upper_item= new_item;
return new_item;
......@@ -5385,7 +5375,7 @@ Item *Item_func_not_all::neg_transformer(THD *thd)
Item_func_nop_all *new_item= new Item_func_nop_all(args[0]);
Item_allany_subselect *allany= (Item_allany_subselect*)args[0];
allany->all= !allany->all;
allany->func= allany->func_creator(TRUE);
allany->create_comp_func(TRUE);
allany->upper_item= new_item;
return new_item;
}
......@@ -5668,6 +5658,9 @@ longlong Item_equal::val_int()
Item_field *item_field;
if (cond_false)
return 0;
/* If there is a single constant and no fields, the equality is TRUE. */
if (const_item && !fields.elements)
return 1;
List_iterator_fast<Item_field> it(fields);
Item *item= const_item ? const_item : it++;
if ((null_value= item->is_null()))
......@@ -5688,6 +5681,15 @@ longlong Item_equal::val_int()
void Item_equal::fix_length_and_dec()
{
Item *item= get_first(NULL);
if (!item)
{
/*
If there are no fields, there must be at least a constant, in which
case Item_equal::val_int evaluates to TRUE.
*/
DBUG_ASSERT(const_item);
return;
}
eval_item= cmp_item::get_comparator(item->result_type(),
item->collation.collation);
}
......
sql/item_cmpfunc.h
View file @ bc7369b7
......@@ -266,6 +266,8 @@ class Item_in_optimizer: public Item_bool_func
void keep_top_level_cache();
Item *transform(Item_transformer transformer, uchar *arg);
virtual Item *expr_cache_insert_transformer(uchar *thd_arg);
bool is_expensive_processor(uchar *arg);
bool is_expensive();
};
class Comp_creator
......@@ -1303,8 +1305,6 @@ class in_row :public in_vector
class Item_func_isnull :public Item_bool_func
{
protected:
longlong cached_value;
public:
Item_func_isnull(Item *a) :Item_bool_func(a) {}
longlong val_int();
......@@ -1322,18 +1322,9 @@ class Item_func_isnull :public Item_bool_func
{
used_tables_cache= 0; /* is always false */
const_item_cache= 1;
cached_value= (longlong) 0;
}
else
{
args[0]->update_used_tables();
if ((const_item_cache= !(used_tables_cache= args[0]->used_tables()) &&
!with_subselect))
{
/* Remember if the value is always NULL or never NULL */
cached_value= (longlong) args[0]->is_null();
}
}
}
table_map not_null_tables() const { return 0; }
optimize_type select_optimize() const { return OPTIMIZE_NULL; }
......
sql/item_subselect.cc
View file @ bc7369b7
This diff is collapsed.
sql/item_subselect.h
View file @ bc7369b7
......@@ -71,6 +71,13 @@ class Item_subselect :public Item_result_field
bool inside_first_fix_fields;
bool done_first_fix_fields;
/*
Set to TRUE if at optimization or execution time we determine that this
item's value is a constant. We need this member because it is not possible
to substitute 'this' with a constant item.
*/
bool forced_const;
public:
/* A reference from inside subquery predicate to somewhere outside of it */
class Ref_to_outside : public Sql_alloc
......@@ -119,6 +126,12 @@ class Item_subselect :public Item_result_field
Item_subselect();
virtual subs_type substype() { return UNKNOWN_SUBS; }
bool is_in_predicate()
{
return (substype() == Item_subselect::IN_SUBS ||
substype() == Item_subselect::ALL_SUBS ||
substype() == Item_subselect::ANY_SUBS);
}
/*
We need this method, because some compilers do not allow 'this'
......@@ -149,12 +162,21 @@ class Item_subselect :public Item_result_field
void fix_after_pullout(st_select_lex *new_parent, Item **ref);
void recalc_used_tables(st_select_lex *new_parent, bool after_pullout);
virtual bool exec();
/*
If subquery optimization or execution determines that the subquery has
an empty result, mark the subquery predicate as a constant value.
*/
void make_const()
{
used_tables_cache= 0;
const_item_cache= 0;
forced_const= TRUE;
}
virtual void fix_length_and_dec();
table_map used_tables() const;
table_map not_null_tables() const { return 0; }
bool const_item() const;
inline table_map get_used_tables_cache() { return used_tables_cache; }
inline bool get_const_item_cache() { return const_item_cache; }
Item *get_tmp_table_item(THD *thd);
void update_used_tables();
virtual void print(String *str, enum_query_type query_type);
......@@ -181,6 +203,8 @@ class Item_subselect :public Item_result_field
enum_parsing_place place() { return parsing_place; }
bool walk(Item_processor processor, bool walk_subquery, uchar *arg);
bool mark_as_eliminated_processor(uchar *arg);
bool eliminate_subselect_processor(uchar *arg);
bool set_fake_select_as_master_processor(uchar *arg);
bool enumerate_field_refs_processor(uchar *arg);
bool check_vcol_func_processor(uchar *int_arg)
{
......@@ -313,6 +337,18 @@ class Item_exists_subselect :public Item_subselect
};
/*
Possible methods to execute an IN predicate. These are set by the optimizer
based on user-set optimizer switches, semantic analysis and cost comparison.
*/
#define SUBS_NOT_TRANSFORMED 0 /* No execution method was chosen for this IN. */
#define SUBS_SEMI_JOIN 1 /* IN was converted to semi-join. */
#define SUBS_IN_TO_EXISTS 2 /* IN was converted to correlated EXISTS. */
#define SUBS_MATERIALIZATION 4 /* Execute IN via subquery materialization. */
/* Partial matching substrategies of MATERIALIZATION. */
#define SUBS_PARTIAL_MATCH_ROWID_MERGE 8
#define SUBS_PARTIAL_MATCH_TABLE_SCAN 16
/**
Representation of IN subquery predicates of the form
"left_expr IN (SELECT ...)".
......@@ -330,8 +366,6 @@ class Item_exists_subselect :public Item_subselect
class Item_in_subselect :public Item_exists_subselect
{
public:
Item *left_expr;
protected:
/*
Cache of the left operand of the subquery predicate. Allocated in the
......@@ -339,12 +373,6 @@ class Item_in_subselect :public Item_exists_subselect
*/
List<Cached_item> *left_expr_cache;
bool first_execution;
/*
Set to TRUE if at query execution time we determine that this item's
value is a constant during this execution. We need this member because
it is not possible to substitute 'this' with a constant item.
*/
bool is_constant;
/*
expr & optimizer used in subselect rewriting to store Item for
......@@ -354,10 +382,24 @@ class Item_in_subselect :public Item_exists_subselect
Item_in_optimizer *optimizer;
bool was_null;
bool abort_on_null;
public:
/* Used to trigger on/off conditions that were pushed down to subselect */
bool *pushed_cond_guards;
Comp_creator *func;
protected:
bool init_cond_guards();
trans_res select_in_like_transformer(JOIN *join);
trans_res single_value_transformer(JOIN *join);
trans_res row_value_transformer(JOIN * join);
bool fix_having(Item *having, st_select_lex *select_lex);
trans_res create_single_in_to_exists_cond(JOIN * join,
Item **where_item,
Item **having_item);
trans_res create_row_in_to_exists_cond(JOIN * join,
Item **where_item,
Item **having_item);
public:
Item *left_expr;
/* Priority of this predicate in the convert-to-semi-join-nest process. */
int sj_convert_priority;
/*
......@@ -388,14 +430,8 @@ class Item_in_subselect :public Item_exists_subselect
*/
bool sjm_scan_allowed;
/* The method chosen to execute the IN predicate. */
enum enum_exec_method {
NOT_TRANSFORMED, /* No execution method was chosen for this IN. */
SEMI_JOIN, /* IN was converted to semi-join nest and should be removed. */
IN_TO_EXISTS, /* IN was converted to correlated EXISTS. */
MATERIALIZATION /* IN will be executed via subquery materialization. */
};
enum_exec_method exec_method;
/* A bitmap of possible execution strategies for an IN predicate. */
uchar in_strategy;
bool *get_cond_guard(int i)
{
......@@ -413,8 +449,9 @@ class Item_in_subselect :public Item_exists_subselect
Item_in_subselect(Item * left_expr, st_select_lex *select_lex);
Item_in_subselect()
:Item_exists_subselect(), left_expr_cache(0), first_execution(TRUE),
is_constant(FALSE), optimizer(0), abort_on_null(0),
pushed_cond_guards(NULL), exec_method(NOT_TRANSFORMED), upper_item(0)
optimizer(0), abort_on_null(0),
pushed_cond_guards(NULL), func(NULL), in_strategy(0),
upper_item(0)
{}
void cleanup();
subs_type substype() { return IN_SUBS; }
......@@ -426,12 +463,9 @@ class Item_in_subselect :public Item_exists_subselect
was_null= 0;
}
trans_res select_transformer(JOIN *join);
trans_res select_in_like_transformer(JOIN *join, Comp_creator *func);
trans_res single_value_transformer(JOIN *join, Comp_creator *func);
trans_res row_value_transformer(JOIN * join);
trans_res single_value_in_to_exists_transformer(JOIN * join,
Comp_creator *func);
trans_res row_value_in_to_exists_transformer(JOIN * join);
bool create_in_to_exists_cond(JOIN *join_arg);
bool inject_in_to_exists_cond(JOIN *join_arg);
virtual bool exec();
longlong val_int();
double val_real();
......@@ -446,11 +480,12 @@ class Item_in_subselect :public Item_exists_subselect
bool fix_fields(THD *thd, Item **ref);
void fix_after_pullout(st_select_lex *new_parent, Item **ref);
void update_used_tables();
bool setup_engine();
bool setup_mat_engine();
bool init_left_expr_cache();
/* Inform 'this' that it was computed, and contains a valid result. */
void set_first_execution() { if (first_execution) first_execution= FALSE; }
bool is_expensive_processor(uchar *arg);
bool is_expensive() { return TRUE; }
bool expr_cache_is_needed(THD *thd);
/*
......@@ -472,7 +507,6 @@ class Item_allany_subselect :public Item_in_subselect
{
public:
chooser_compare_func_creator func_creator;
Comp_creator *func;
bool all;
Item_allany_subselect(Item * left_expr, chooser_compare_func_creator fc,
......@@ -481,6 +515,7 @@ class Item_allany_subselect :public Item_in_subselect
// only ALL subquery has upper not
subs_type substype() { return all?ALL_SUBS:ANY_SUBS; }
trans_res select_transformer(JOIN *join);
void create_comp_func(bool invert) { func= func_creator(invert); }
virtual void print(String *str, enum_query_type query_type);
};
......@@ -821,10 +856,9 @@ class subselect_hash_sj_engine : public subselect_engine
}
~subselect_hash_sj_engine();
bool init_permanent(List<Item> *tmp_columns);
bool init_runtime();
bool init(List<Item> *tmp_columns);
void cleanup();
int prepare() { return 0; } /* Override virtual function in base class. */
int prepare();
int exec();
virtual void print(String *str, enum_query_type query_type);
uint cols()
......
sql/mysql_priv.h
View file @ bc7369b7
......@@ -562,24 +562,29 @@ class Default_object_creation_ctx : public Object_creation_ctx
#define OPTIMIZER_SWITCH_INDEX_MERGE_SORT_UNION 4
#define OPTIMIZER_SWITCH_INDEX_MERGE_INTERSECT 8
#define OPTIMIZER_SWITCH_INDEX_COND_PUSHDOWN 16
#define OPTIMIZER_SWITCH_FIRSTMATCH 32
#define OPTIMIZER_SWITCH_LOOSE_SCAN 64
#define OPTIMIZER_SWITCH_MATERIALIZATION 128
#define OPTIMIZER_SWITCH_SEMIJOIN 256
#define OPTIMIZER_SWITCH_PARTIAL_MATCH_ROWID_MERGE 512
#define OPTIMIZER_SWITCH_PARTIAL_MATCH_TABLE_SCAN 1024
#define OPTIMIZER_SWITCH_SUBQUERY_CACHE (1<<11)
#define OPTIMIZER_SWITCH_IN_TO_EXISTS 256
#define OPTIMIZER_SWITCH_SEMIJOIN 512
#define OPTIMIZER_SWITCH_PARTIAL_MATCH_ROWID_MERGE 1024
#define OPTIMIZER_SWITCH_PARTIAL_MATCH_TABLE_SCAN (1<<11)
#define OPTIMIZER_SWITCH_SUBQUERY_CACHE (1<<12)
#ifdef DBUG_OFF
# define OPTIMIZER_SWITCH_LAST (1<<12)
#else
# define OPTIMIZER_SWITCH_TABLE_ELIMINATION (1<<12)
# define OPTIMIZER_SWITCH_LAST (1<<13)
#else
# define OPTIMIZER_SWITCH_TABLE_ELIMINATION (1<<13)
# define OPTIMIZER_SWITCH_LAST (1<<14)
#endif
#ifdef DBUG_OFF
/* The following must be kept in sync with optimizer_switch_str in mysqld.cc */
/*
TODO: Materialization is off by default to mimic 5.1/5.2 behavior.
Once cost based choice between materialization and in-to-exists should be
enabled by default, add OPTIMIZER_SWITCH_MATERIALIZATION
*/
# define OPTIMIZER_SWITCH_DEFAULT (OPTIMIZER_SWITCH_INDEX_MERGE | \
OPTIMIZER_SWITCH_INDEX_MERGE_UNION | \
OPTIMIZER_SWITCH_INDEX_MERGE_SORT_UNION | \
......@@ -587,7 +592,7 @@ class Default_object_creation_ctx : public Object_creation_ctx
OPTIMIZER_SWITCH_INDEX_COND_PUSHDOWN | \
OPTIMIZER_SWITCH_FIRSTMATCH | \
OPTIMIZER_SWITCH_LOOSE_SCAN | \
OPTIMIZER_SWITCH_MATERIALIZATION | \
OPTIMIZER_SWITCH_IN_TO_EXISTS | \
OPTIMIZER_SWITCH_SEMIJOIN | \
OPTIMIZER_SWITCH_PARTIAL_MATCH_ROWID_MERGE|\
OPTIMIZER_SWITCH_PARTIAL_MATCH_TABLE_SCAN|\
......@@ -601,7 +606,7 @@ class Default_object_creation_ctx : public Object_creation_ctx
OPTIMIZER_SWITCH_TABLE_ELIMINATION | \
OPTIMIZER_SWITCH_FIRSTMATCH | \
OPTIMIZER_SWITCH_LOOSE_SCAN | \
OPTIMIZER_SWITCH_MATERIALIZATION | \
OPTIMIZER_SWITCH_IN_TO_EXISTS | \
OPTIMIZER_SWITCH_SEMIJOIN | \
OPTIMIZER_SWITCH_PARTIAL_MATCH_ROWID_MERGE|\
OPTIMIZER_SWITCH_PARTIAL_MATCH_TABLE_SCAN|\
......
sql/mysqld.cc
View file @ bc7369b7
......@@ -341,7 +341,7 @@ static const char *optimizer_switch_names[]=
"index_merge","index_merge_union","index_merge_sort_union",
"index_merge_intersection",
"index_condition_pushdown",
"firstmatch","loosescan","materialization", "semijoin",
"firstmatch","loosescan","materialization","in_to_exists","semijoin",
"partial_match_rowid_merge",
"partial_match_table_scan",
"subquery_cache",
......@@ -362,6 +362,7 @@ static const unsigned int optimizer_switch_names_len[]=
sizeof("firstmatch") - 1,
sizeof("loosescan") - 1,
sizeof("materialization") - 1,
sizeof("in_to_exists") - 1,
sizeof("semijoin") - 1,
sizeof("partial_match_rowid_merge") - 1,
sizeof("partial_match_table_scan") - 1,
......@@ -460,7 +461,8 @@ static const char *optimizer_switch_str="index_merge=on,index_merge_union=on,"
"index_condition_pushdown=on,"
"firstmatch=on,"
"loosescan=on,"
"materialization=on,"
"materialization=off,"
"in_to_exists=on,"
"semijoin=on,"
"partial_match_rowid_merge=on,"
"partial_match_table_scan=on,"
......@@ -7375,7 +7377,7 @@ thread is in the relay logs.",
{"optimizer_switch", OPT_OPTIMIZER_SWITCH,
"optimizer_switch=option=val[,option=val...], where option={index_merge, "
"index_merge_union, index_merge_sort_union, index_merge_intersection, "
"index_condition_pushdown, firstmatch, loosescan, materialization, "
"index_condition_pushdown, firstmatch, loosescan, materialization, in_to_exists, "
"semijoin, partial_match_rowid_merge, partial_match_table_scan, "
"subquery_cache"
#ifndef DBUG_OFF
......
sql/opt_subselect.cc
View file @ bc7369b7
This diff is collapsed.
sql/share/errmsg.txt
View file @ bc7369b7
......@@ -6247,3 +6247,5 @@ ER_UNKNOWN_OPTION
eng "Unknown option '%-.64s'"
ER_BAD_OPTION_VALUE
eng "Incorrect value '%-.64s' for option '%-.64s'"
ER_ILLEGAL_SUBQUERY_OPTIMIZER_SWITCHES
eng "At least one of the 'in_to_exists' or 'materialization' optimizer_switch flags must be 'on'."
sql/sql_class.cc
View file @ bc7369b7
......@@ -2656,6 +2656,7 @@ void Query_arena::free_items()
for (; free_list; free_list= next)
{
next= free_list->next;
DBUG_ASSERT(free_list != next);
free_list->delete_self();
}
/* Postcondition: free_list is 0 */
......@@ -3092,6 +3093,7 @@ void TMP_TABLE_PARAM::init()
table_charset= 0;
precomputed_group_by= 0;
bit_fields_as_long= 0;
materialized_subquery= 0;
skip_create_table= 0;
DBUG_VOID_RETURN;
}
......
sql/sql_class.h
View file @ bc7369b7
......@@ -2926,6 +2926,8 @@ class TMP_TABLE_PARAM :public Sql_alloc
uint convert_blob_length;
CHARSET_INFO *table_charset;
bool schema_table;
/* TRUE if the temp table is created for subquery materialization. */
bool materialized_subquery;
/*
True if GROUP BY and its aggregate functions are already computed
by a table access method (e.g. by loose index scan). In this case
......@@ -2949,8 +2951,8 @@ class TMP_TABLE_PARAM :public Sql_alloc
TMP_TABLE_PARAM()
:copy_field(0), group_parts(0),
group_length(0), group_null_parts(0), convert_blob_length(0),
schema_table(0), precomputed_group_by(0), force_copy_fields(0),
bit_fields_as_long(0), skip_create_table(0)
schema_table(0), materialized_subquery(0), precomputed_group_by(0),
force_copy_fields(0), bit_fields_as_long(0), skip_create_table(0)
{}
~TMP_TABLE_PARAM()
{
......@@ -2983,6 +2985,7 @@ class select_union :public select_result_interceptor
virtual bool create_result_table(THD *thd, List<Item> *column_types,
bool is_distinct, ulonglong options,
const char *alias, bool bit_fields_as_long);
TMP_TABLE_PARAM *get_tmp_table_param() { return &tmp_table_param; }
};
/* Base subselect interface class */
......@@ -3046,7 +3049,7 @@ class select_materialize_with_stats : public select_union
void reset();
public:
select_materialize_with_stats() {}
select_materialize_with_stats() { tmp_table_param.init(); }
virtual bool create_result_table(THD *thd, List<Item> *column_types,
bool is_distinct, ulonglong options,
const char *alias, bool bit_fields_as_long);
......
sql/sql_delete.cc
View file @ bc7369b7
......@@ -92,6 +92,10 @@ bool mysql_delete(THD *thd, TABLE_LIST *table_list, COND *conds,
}
}
/* Apply the IN=>EXISTS transformation to all subqueries and optimize them. */
if (select_lex->optimize_unflattened_subqueries())
DBUG_RETURN(TRUE);
const_cond= (!conds || conds->const_item());
safe_update=test(thd->options & OPTION_SAFE_UPDATES);
if (safe_update && const_cond)
......
sql/sql_lex.cc
View file @ bc7369b7
......@@ -1735,17 +1735,29 @@ void st_select_lex_node::fast_exclude()
}
/*
Exclude a node from the tree lex structure, but leave it in the global
list of nodes.
*/
void st_select_lex_node::exclude_from_tree()
{
if ((*prev= next))
next->prev= prev;
}
/*
excluding select_lex structure (except first (first select can't be
Exclude select_lex structure (except first (first select can't be
deleted, because it is most upper select))
*/
void st_select_lex_node::exclude()
{
//exclude from global list
/* exclude from global list */
fast_exclude();
//exclude from other structures
if ((*prev= next))
next->prev= prev;
/* exclude from other structures */
exclude_from_tree();
/*
We do not need following statements, because prev pointer of first
list element point to master->slave
......@@ -2145,8 +2157,8 @@ void st_select_lex::print_limit(THD *thd,
select_limit == 1, and there should be no offset_limit.
*/
(((subs_type == Item_subselect::IN_SUBS) &&
((Item_in_subselect*)item)->exec_method ==
Item_in_subselect::MATERIALIZATION) ?
((Item_in_subselect*)item)->in_strategy &
SUBS_MATERIALIZATION) ?
TRUE :
(select_limit->val_int() == 1LL) &&
offset_limit == 0));
......@@ -3076,6 +3088,70 @@ bool st_select_lex::add_index_hint (THD *thd, char *str, uint length)
str, length));
}
bool st_select_lex::optimize_unflattened_subqueries()
{
for (SELECT_LEX_UNIT *un= first_inner_unit(); un; un= un->next_unit())
{
Item_subselect *subquery_predicate= un->item;
if (subquery_predicate)
{
for (SELECT_LEX *sl= un->first_select(); sl; sl= sl->next_select())
{
JOIN *inner_join= sl->join;
SELECT_LEX *save_select= un->thd->lex->current_select;
ulonglong save_options;
int res;
/* We need only 1 row to determine existence */
un->set_limit(un->global_parameters);
un->thd->lex->current_select= sl;
save_options= inner_join->select_options;
if (un->outer_select()->options & SELECT_DESCRIBE)
{
/* Optimize the subquery in the context of EXPLAIN. */
set_explain_type();
inner_join->select_options= options;
}
res= inner_join->optimize();
inner_join->select_options= save_options;
un->thd->lex->current_select= save_select;
if (res)
return TRUE;
}
}
}
return FALSE;
}
/**
Set the EXPLAIN type for this subquery.
*/
void st_select_lex::set_explain_type()
{
SELECT_LEX *first= master_unit()->first_select();
/* drop UNCACHEABLE_EXPLAIN, because it is for internal usage only */
uint8 is_uncacheable= (uncacheable & ~UNCACHEABLE_EXPLAIN);
type= ((&master_unit()->thd->lex->select_lex == this) ?
(first_inner_unit() || next_select() ?
"PRIMARY" : "SIMPLE") :
((this == first) ?
((linkage == DERIVED_TABLE_TYPE) ?
"DERIVED" :
((is_uncacheable & UNCACHEABLE_DEPENDENT) ?
"DEPENDENT SUBQUERY" :
(is_uncacheable ? "UNCACHEABLE SUBQUERY" :
"SUBQUERY"))) :
((is_uncacheable & UNCACHEABLE_DEPENDENT) ?
"DEPENDENT UNION":
is_uncacheable ? "UNCACHEABLE UNION":
"UNION")));
options|= SELECT_DESCRIBE;
}
/**
A routine used by the parser to decide whether we are specifying a full
partitioning or if only partitions to add or to split.
......@@ -3093,4 +3169,3 @@ bool st_lex::is_partition_management() const
(alter_info.flags == ALTER_ADD_PARTITION ||
alter_info.flags == ALTER_REORGANIZE_PARTITION));
}
sql/sql_lex.h
View file @ bc7369b7
......@@ -439,6 +439,7 @@ class st_select_lex_node {
st_select_lex_node(): linkage(UNSPECIFIED_TYPE) {}
virtual ~st_select_lex_node() {}
inline st_select_lex_node* get_master() { return master; }
inline void set_master(st_select_lex_node* master_arg) { master= master_arg; }
virtual void init_query();
virtual void init_select();
void include_down(st_select_lex_node *upper);
......@@ -446,6 +447,7 @@ class st_select_lex_node {
void include_standalone(st_select_lex_node *sel, st_select_lex_node **ref);
void include_global(st_select_lex_node **plink);
void exclude();
void exclude_from_tree();
virtual st_select_lex_unit* master_unit()= 0;
virtual st_select_lex* outer_select()= 0;
......@@ -846,6 +848,15 @@ class st_select_lex: public st_select_lex_node
void clear_index_hints(void) { index_hints= NULL; }
bool is_part_of_union() { return master_unit()->is_union(); }
/*
Optimize all subqueries that have not been flattened into semi-joins.
This functionality is a method of SELECT_LEX instead of JOIN because
some SQL statements as DELETE do not have a corresponding JOIN object.
*/
bool optimize_unflattened_subqueries();
/* Set the EXPLAIN type for this subquery. */
void set_explain_type();
private:
/* current index hint kind. used in filling up index_hints */
enum index_hint_type current_index_hint_type;
......
sql/sql_select.cc
View file @ bc7369b7
This diff is collapsed.
sql/sql_select.h
View file @ bc7369b7
......@@ -1369,8 +1369,30 @@ inline bool sj_is_materialize_strategy(uint strategy)
class JOIN :public Sql_alloc
{
private:
JOIN(const JOIN &rhs); /**< not implemented */
JOIN& operator=(const JOIN &rhs); /**< not implemented */
protected:
/* Results of reoptimizing a JOIN via JOIN::reoptimize(). */
enum enum_reopt_result {
REOPT_NEW_PLAN, /* there is a new reoptimized plan */
REOPT_OLD_PLAN, /* no new improved plan can be found, use the old one */
REOPT_ERROR, /* an irrecovarable error occured during reoptimization */
REOPT_NONE /* not yet reoptimized */
};
/* Support for plan reoptimization with rewritten conditions. */
enum_reopt_result reoptimize(Item *added_where, table_map join_tables);
int save_query_plan(DYNAMIC_ARRAY *save_keyuse, POSITION *save_positions,
KEYUSE **save_join_tab_keyuse,
key_map *save_join_tab_checked_keys);
void restore_query_plan(DYNAMIC_ARRAY *save_keyuse, POSITION *save_positions,
KEYUSE **save_join_tab_keyuse,
key_map *save_join_tab_checked_keys);
/* Choose a subquery plan for a table-less subquery. */
bool choose_tableless_subquery_plan();
public:
JOIN_TAB *join_tab,**best_ref;
JOIN_TAB **map2table; ///< mapping between table indexes and JOIN_TABs
......@@ -1581,8 +1603,15 @@ class JOIN :public Sql_alloc
bool union_part; ///< this subselect is part of union
bool optimized; ///< flag to avoid double optimization in EXPLAIN
bool initialized; ///< flag to avoid double init_execution calls
Array<Item_in_subselect> sj_subselects;
/*
Additional WHERE and HAVING predicates to be considered for IN=>EXISTS
subquery transformation of a JOIN object.
*/
Item *in_to_exists_where;
Item *in_to_exists_having;
/* Temporary tables used to weed-out semi-join duplicates */
List<TABLE> sj_tmp_tables;
......@@ -1657,6 +1686,7 @@ class JOIN :public Sql_alloc
ref_pointer_array_size= 0;
zero_result_cause= 0;
optimized= 0;
initialized= 0;
cond_equal= 0;
having_equal= 0;
group_optimized_away= 0;
......@@ -1672,6 +1702,8 @@ class JOIN :public Sql_alloc
no_const_tables= FALSE;
first_select= sub_select;
in_to_exists_where= NULL;
in_to_exists_having= NULL;
}
int prepare(Item ***rref_pointer_array, TABLE_LIST *tables, uint wind_num,
......@@ -1680,12 +1712,13 @@ class JOIN :public Sql_alloc
SELECT_LEX_UNIT *unit);
int optimize();
int reinit();
int init_execution();
void exec();
int destroy();
void restore_tmp();
bool alloc_func_list();
bool flatten_subqueries();
bool setup_subquery_materialization();
bool optimize_unflattened_subqueries();
bool make_sum_func_list(List<Item> &all_fields, List<Item> &send_fields,
bool before_group_by, bool recompute= FALSE);
......@@ -1745,6 +1778,10 @@ class JOIN :public Sql_alloc
NULL : join_tab+const_tables;
}
bool setup_subquery_caches();
bool choose_subquery_plan(table_map join_tables);
void get_partial_join_cost(uint n_tables,
double *read_time_arg, double *record_count_arg);
private:
/**
TRUE if the query contains an aggregate function but has no GROUP
......@@ -1977,8 +2014,6 @@ inline Item * and_items(Item* cond, Item *item)
return (cond? (new Item_cond_and(cond, item)) : item);
}
bool choose_plan(JOIN *join,table_map join_tables);
void get_partial_join_cost(JOIN *join, uint n_tables, double *read_time_arg,
double *record_count_arg);
void optimize_wo_join_buffering(JOIN *join, uint first_tab, uint last_tab,
table_map last_remaining_tables,
bool first_alt, uint no_jbuf_before,
......
sql/sql_union.cc
View file @ bc7369b7
......@@ -185,6 +185,8 @@ st_select_lex_unit::init_prepare_fake_select_lex(THD *thd_arg)
{
(*order->item)->walk(&Item::change_context_processor, 0,
(uchar*) &fake_select_lex->context);
(*order->item)->walk(&Item::set_fake_select_as_master_processor, 0,
(uchar*) fake_select_lex);
}
}
......@@ -270,6 +272,18 @@ bool st_select_lex_unit::prepare(THD *thd_arg, select_result *sel_result,
can_skip_order_by= is_union_select && !(sl->braces && sl->explicit_limit);
/*
Remove all references from the select_lex_units to the subqueries that
are inside the ORDER BY clause.
*/
if (can_skip_order_by)
{
for (ORDER *ord= (ORDER *)sl->order_list.first; ord; ord= ord->next)
{
(*ord->item)->walk(&Item::eliminate_subselect_processor, FALSE, NULL);
}
}
saved_error= join->prepare(&sl->ref_pointer_array,
sl->table_list.first,
sl->with_wild,
......
sql/sql_update.cc
View file @ bc7369b7
......@@ -290,6 +290,10 @@ int mysql_update(THD *thd,
DBUG_RETURN(1); /* purecov: inspected */
}
/* Apply the IN=>EXISTS transformation to all subqueries and optimize them. */
if (select_lex->optimize_unflattened_subqueries())
DBUG_RETURN(TRUE);
if (select_lex->inner_refs_list.elements &&
fix_inner_refs(thd, all_fields, select_lex, select_lex->ref_pointer_array))
DBUG_RETURN(1);
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7