Details
-
Type:
Bug
-
Status: Closed
-
Resolution: Not a Bug
-
Affects Version/s: None
-
Fix Version/s: None
-
Component/s: None
-
Labels:
Description
The following query:
SELECT table2 . `col_varchar_key` AS field1
FROM ( C AS table1 INNER JOIN C AS table2 ON (table2 . `col_int_key` = table1 . `col_int_key` ) )
GROUP BY field1 HAVING field1 < 1 ORDER BY field1 ;
produces the following valgrind warning:
==18898== Use of uninitialised value of size 4
==18898== at 0x8768836: my_strnncollsp_simple (ctype-simple.c:166)
==18898== by 0x825776D: Field_varstring::cmp_max(unsigned char const*, unsigned char const*, unsigned int) (field.cc:7140)
==18898== by 0x8261955: Field_varstring::cmp(unsigned char const*, unsigned char const*) (field.h:1669)
==18898== by 0x81E0E46: Field::cmp(unsigned char const*) (field.h:314)
==18898== by 0x81E0C37: Cached_item_field::cmp() (item_buff.cc:122)
==18898== by 0x8329D34: test_if_group_changed(List<Cached_item>&) (sql_select.cc:17230)
==18898== by 0x83228A8: end_send_group(JOIN*, st_join_table*, bool) (sql_select.cc:14324)
==18898== by 0x8320A19: evaluate_join_record(JOIN*, st_join_table*, int) (sql_select.cc:13490)
==18898== by 0x832050E: sub_select(JOIN*, st_join_table*, bool) (sql_select.cc:13338)
==18898== by 0x8320A19: evaluate_join_record(JOIN*, st_join_table*, int) (sql_select.cc:13490)
==18898== by 0x83203C0: sub_select(JOIN*, st_join_table*, bool) (sql_select.cc:13298)
==18898== by 0x831F7E1: do_select(JOIN*, List<Item>, st_table, Procedure*) (sql_select.cc:12842)
==18898== by 0x83057B6: JOIN::exec() (sql_select.cc:2412)
==18898== by 0x8305EDF: mysql_select(THD*, Item**, TABLE_LIST, unsigned int, List<Item>&, Item*, unsigned int, st_order*, st_order*, Item*, st_order*, unsigned long long, select_result*, st_select_lex_unit*, st_select_lex*) (sql_select.cc:2613)
==18898== by 0x82FE59A: handle_select(THD*, st_lex*, select_result*, unsigned long) (sql_select.cc:277)
==18898== by 0x829B753: execute_sqlcom_select(THD*, TABLE_LIST*) (sql_parse.cc:5081)
==18898==
in maria-5.3-mwl128 and not in 5.3-main. The EXPLAIN plan does not show any join_cache involvement.
Gliffy Diagrams
Attachments
Activity
- All
- Comments
- Work Log
- History
- Activity
- Transitions
Re: Valgrind warning "Use of uninitialised value of size 4" at my_strnncollsp_simple in maria-5.3-mwl128
Also reproducible with maria-5.3
Re: Valgrind warning "Use of uninitialised value of size 4" at my_strnncollsp_simple
No longer reproducible with the latest 5.3 as of 14-06-2011, tip:
------------------------------------------------------------
revno: 3041
committer: Sergey Petrunya <psergey@askmonty.org>
branch nick: 5.3-push3
timestamp: Mon 2011-06-13 12:41:19 +0400
message:
Remove redundant code that is a result of a wrong merge.
(Changeset sp1r-igor@olga.mysql.com-20070526173301-38848 moved this loop from one place
to another, then the merge of sp1r-gshchepa/uchum@gleb.loc-20070527192244-26330 have
kept both copies).
------------------------------------------------------------
Launchpad bug id: 666322
Re: Valgrind warning "Use of uninitialised value of size 4" at my_strnncollsp_simple in maria-5.3-mwl128
Test case:
CREATE TABLE `C` (
`col_int_key` int(11) DEFAULT NULL,
`col_varchar_key` varchar(1) DEFAULT NULL,
KEY `col_int_key` (`col_int_key`),
KEY `col_varchar_key` (`col_varchar_key`,`col_int_key`)
) ENGINE=MyISAM DEFAULT CHARSET=latin1;
INSERT INTO `C` VALUES (0,NULL);
INSERT INTO `C` VALUES (1,'c');
INSERT INTO `C` VALUES (2,'d');
INSERT INTO `C` VALUES (166,'e');
INSERT INTO `C` VALUES (6,'f');
INSERT INTO `C` VALUES (8,'h');
INSERT INTO `C` VALUES (3,'j');
INSERT INTO `C` VALUES (5,'k');
INSERT INTO `C` VALUES (9,'k');
INSERT INTO `C` VALUES (3,'m');
INSERT INTO `C` VALUES (9,'m');
INSERT INTO `C` VALUES (9,'m');
INSERT INTO `C` VALUES (3,'n');
INSERT INTO `C` VALUES (53,'o');
INSERT INTO `C` VALUES (NULL,'r');
INSERT INTO `C` VALUES (0,'t');
INSERT INTO `C` VALUES (9,'t');
INSERT INTO `C` VALUES (8,'u');
INSERT INTO `C` VALUES (2,'w');
INSERT INTO `C` VALUES (5,'y');
SELECT table2 . `col_varchar_key` AS field1
FROM ( C AS table1 INNER JOIN C AS table2 ON (table2 . `col_int_key` = table1 . `col_int_key` ) )
GROUP BY field1 HAVING field1 < 1 ORDER BY field1 ;