June 2016 - developers - lists.mariadb.org

Re: [Maria-developers] [MariaDB/server] MDEV-10303 - rsync_sst_rsync: ignore lost+found during database directory transfer (#196)
by Kristian Nielsen 29 Jun '16

29 Jun '16

Sergey Vojtovich <notifications(a)github.com> writes: > Hi Nicolas, > > Thanks for your contribution. JIRA task has been created to track this > pull request: https://jira.mariadb.org/browse/MDEV-10303 > > This task was added to 10.2.2 backlog, which hasn't been scheduled yet. > > Similar to other open source projects, the MariaDB Foundation needs to > have shared ownership of all code that is included in the MariaDB > distribution. The easiest way to achieve this is by submitting your > code under the BSD-new license. > > The other alternative is to sign the code contribution agreement which > can be found here: https://mariadb.com/kb/en/mariadb/mca/ As was concluded in this mailing thread, https://lists.launchpad.net/maria-developers/msg09606.html the third option is to contribute the code under GPLv2. Personally, I would strongly recommend to contribute under GPLv2 only. There are good and important reason why the GPL exists and why it is used by many of the most successful free software projects. Most of the code by core developers is contributed to the project under GPLv2 only, as is most of the major external contributions. - Kristian.

1 0

Re: [Maria-developers] eb9fab7: MDEV-9531: GROUP_CONCAT with ORDER BY inside takes a lot of memory while it's executed
by Sergei Golubchik 28 Jun '16

28 Jun '16

Hi, Sanja! On May 04, Olaksandr Byslkin wrote: still haven't fixed the typo? :) > revision-id: eb9fab76fb4322a7821e6502e6303ed381624429 (mariadb-10.1.13-18-geb9fab7) > parent(s): 732adec0a4c75d99389230feeb0deca0ad668de7 > committer: Oleksandr Byelkin > timestamp: 2016-05-04 19:15:31 +0200 > message: > > MDEV-9531: GROUP_CONCAT with ORDER BY inside takes a lot of memory while it's executed > > Limitation added to Red-Black tree. > > diff --git a/include/my_tree.h b/include/my_tree.h > index f8be55f..f1916b9 100644 > --- a/include/my_tree.h > +++ b/include/my_tree.h > @@ -57,11 +57,14 @@ typedef struct st_tree_element { > } TREE_ELEMENT; > > #define ELEMENT_CHILD(element, offs) (*(TREE_ELEMENT**)((char*)element + offs)) > +#define R_ELEMENT_CHILD(element, offs) ((TREE_ELEMENT**)((char*)element + offs)) what does "R_ELEMENT_CHILD" mean? What's "R"? > > typedef struct st_tree { > TREE_ELEMENT *root,null_element; > TREE_ELEMENT **parents[MAX_TREE_HEIGHT]; > + TREE_ELEMENT *free_element; why do you need a free_element pointer? > uint offset_to_key,elements_in_tree,size_of_element; > + uint elements_limit, del_direction; why do you need to support different directions here? > size_t memory_limit, allocated; > qsort_cmp2 compare; > void *custom_arg; > diff --git a/mysql-test/r/mysqld--help.result b/mysql-test/r/mysqld--help.result > index bef0390..3699b18 100644 > --- a/mysql-test/r/mysqld--help.result > +++ b/mysql-test/r/mysqld--help.result > @@ -240,6 +240,9 @@ The following options may be given as the first argument: > --group-concat-max-len=# > The maximum length of the result of function > GROUP_CONCAT() > + --group-concat-max-mem=# > + The maximum memory used to calculate result of function > + GROUP_CONCAT() No, please. There is group_concat_max_len already, it's all that user needs. There is no reason why one might need to configure a second limit on for the GROUP_CONCAT. > --gtid-domain-id=# Used with global transaction ID to identify logically > independent replication streams. When events can > propagate through multiple parallel paths (for example Regards, Sergei Chief Architect MariaDB and security(a)mariadb.org

2 3

Re: [Maria-developers] 0752333: MDEV-10161: wsrep_sync_wait not enabled when set to 1 in config file
by Sergei Golubchik 27 Jun '16

27 Jun '16

Hi, Nirbhay! On Jun 08, Nirbhay Choubey wrote: > revision-id: 075233312cc5bc6c242f4629f8293d4c67937f02 (mariadb-10.1.14-10-g0752333) > parent(s): c9f05974e618a869563f2360ef1ca910974a4f1c > author: Nirbhay Choubey > committer: Nirbhay Choubey > timestamp: 2016-06-08 16:03:01 -0400 > message: > > MDEV-10161: wsrep_sync_wait not enabled when set to 1 in config file > > Since wsrep_sync_wait & wsrep_causal_reads variables are related, > they are always kept in sync whenever one of them changes. > Same is tried on server start, where wsrep_sync_wait get updated > based on wsrep_causal_reads' value. But, since wsrep_causal_reads > is OFF by default, wsrep_sync_wait's value gets modified and loses > its WSREP_SYNC_WAIT_BEFORE_READ bit. > > Fixed by giving higher precedence to one that's explicitly set to > a non-default value and later to wsrep_sync_wait in case both are > set. > > diff --git a/sql/wsrep_mysqld.cc b/sql/wsrep_mysqld.cc > index d0798ef..fffb0cc 100644 > --- a/sql/wsrep_mysqld.cc > +++ b/sql/wsrep_mysqld.cc > @@ -567,7 +567,39 @@ int wsrep_init() > > wsrep_sst_auth_init(wsrep_sst_auth); > > - wsrep_causal_reads_update(&global_system_variables); > + /* > + On server start, there is no way to detect whether the default value > + for a system variable was set explicitly by the user. > + Now, since wsrep_causal_reads & wsrep_sync_wait are used for same > + purpose, a rule of precedence must be defined to keep their values > + in sync. It goes as following: This is wrong. First, in 10.1 there is a way to detect whether the value was set by a user. I_S.SYSTEM_VARIABLES can show that. But even if you'd detected that, it won't help you do distinguish between mysqld --wsrep-sync-wait=1 --wsrep-causal-reads=0 mysqld --wsrep-sync-wait=0 --wsrep-causal-reads=1 The only correct solution is to handle this in the mysqld_get_one_option switch, updating the dependent option when one of these two is set. Regards, Sergei Chief Architect MariaDB and security(a)mariadb.org

2 1

[Maria-developers] Please review MDEV-10236 Where expression with NOT function gives incorrect result
by Alexander Barkov 27 Jun '16

27 Jun '16

Hi Serg. Please review a patch for MDEV-10236. Thanks.

2 3

Re: [Maria-developers] 3e61493: MDEV-10010 - Recursive call to mysql_rwlock_rdlock for LOCK_system_variables_hash
by Sergei Golubchik 27 Jun '16

27 Jun '16

Hi, Sergey! On Jun 27, Sergey Vojtovich wrote: > revision-id: 3e61493dff068c86b37f2e2ec0d0b6cbe9a255b6 (mariadb-10.1.14-28-g3e61493) > parent(s): 305bbbce6152b0909937659ecba4d51e5b21c094 > committer: Sergey Vojtovich > timestamp: 2016-06-27 15:01:22 +0400 > message: > > MDEV-10010 - Recursive call to mysql_rwlock_rdlock for LOCK_system_variables_hash > > Avoid recursive LOCK_system_variables_hash acquisition in > intern_sys_var_ptr() by pre-syncing dynamic session variables. Looks good, thanks! Regards, Sergei Chief Architect MariaDB and security(a)mariadb.org

1 0

Re: [Maria-developers] 79716e0: [MDEV-9127] Crash reporter often fails to show the query that crashed
by Sergei Golubchik 27 Jun '16

27 Jun '16

Hi, Vicentiu! On May 30, Vicentiu Ciorbaru wrote: > revision-id: 79716e07f15e8dabca4d2647ff5dbf189935581a (mariadb-10.0.25-6-g79716e0) > parent(s): eaf5376f808b3d71c442fb6b95048c4dcd684b08 > author: Vicențiu Ciorbaru > committer: Vicențiu Ciorbaru > timestamp: 2016-05-30 23:21:22 +0300 > message: > > [MDEV-9127] Crash reporter often fails to show the query that crashed > > Addreses are not necessarily between heap_start && heap_end. Malloc > calls using mmap can place pointers outside these bounds. In this case, > we'll warn the user that the query pointer is potentially invalid. > However, we'll attempt to print the data anyway after we're done > printing everything else. ok to push Regards, Sergei Chief Architect MariaDB and security(a)mariadb.org

1 0

Re: [Maria-developers] d1a7ae2: MDEV-9618 solaris sparc build fails on 10.1.
by Sergei Golubchik 26 Jun '16

26 Jun '16

Hi, Alexey! On Jun 26, Alexey Botchkov wrote: > revision-id: d1a7ae232ef6e35e4f81871e3c9d0d318a5dcef1 (mariadb-10.1.14-28-gd1a7ae2) > parent(s): 305bbbce6152b0909937659ecba4d51e5b21c094 > committer: Alexey Botchkov > timestamp: 2016-06-26 00:11:30 +0400 > message: > > MDEV-9618 solaris sparc build fails on 10.1. > > Compiler on Sparc is strict about the 'const' modifiers > in function declarations and definitions. Meaning > they should be exactly same. ok to push Interestingly enough, I've just read about it few days ago - that const modifiers do *not* have to be exactly the same, the compiler can be lax about it, because these const modifiers do not mean anything for the caller. So, this strictness could be considered a bug in that Solaris C++ compiler. Regards, Sergei Chief Architect MariaDB and security(a)mariadb.org

1 0

[Maria-developers] When SP DEFINER is empty.
by Alexey Botchkov 25 Jun '16

25 Jun '16

Hi, Sergey, all. I'm in doubts about how this one should be fixed: https://jira.mariadb.org/browse/MDEV-10119 The story goes like this: - mysql_install_db script runs 'mysqld --skip-grant-tables' In this case the 'current_user()' seems to be empty. - so the CREATE PROCEDURE command in the bootstrap creates the procedure with the empty DEFINER. - as a result, the 'SHOW CREATE PROCEDURE' returns query started with 'CREATE DEFINER=`` PROCEDURE...', - that DEFINER=`` gives an error when feed to the server. That can be fixed on any stage. We can do any of these: - set some 'current_user()' to be not empty even with the --skip-grant-tables option - specify some non-empty DEFINER for the CREATE PROCEDURE statement (in both options it's not that clear what user could that be) - fix the SHOW CREATE PROCEDURE statement so it doesn't add the errorneous DEFINER=`` to the query. - make server handling the 'DEFINER=``' with no error. Maybe assigning the 'current_user()' in this case. So, what can You recommend as a fix in this case? Best regards. HF

2 1

Re: [Maria-developers] [Commits] 773ce40: MDEV-10045: Server crashes in Time_and_counter_tracker::incr_loops
by Sergey Petrunia 24 Jun '16

24 Jun '16

Hi Sanja, O the patch fixes the crash. However when I am debugging it, I see the execution to go like this: (gdb) wher #0 JOIN::optimize_inner (this=0x7fff900087d0) at /home/psergey/dev-git/10.1-dbg6/sql/sql_select.cc:1095 #1 0x0000555555a8e5f8 in JOIN::optimize (this=0x7fff900087d0) at /home/psergey/dev-git/10.1-dbg6/sql/sql_select.cc:1036 #2 0x0000555555b277ce in st_select_lex_unit::optimize (this=0x7fff900068a8) at /home/psergey/dev-git/10.1-dbg6/sql/sql_union.cc:747 #3 0x0000555555b279cb in st_select_lex_unit::exec (this=0x7fff900068a8) at /home/psergey/dev-git/10.1-dbg6/sql/sql_union.cc:780 #4 0x0000555555d3855f in subselect_union_engine::exec (this=0x7fff90007970) at /home/psergey/dev-git/10.1-dbg6/sql/item_subselect.cc:3785 #5 0x0000555555d2eb12 in Item_subselect::exec (this=0x7fff9000b028) at /home/psergey/dev-git/10.1-dbg6/sql/item_subselect.cc:684 #6 0x0000555555d3057e in Item_singlerow_subselect::val_str (this=0x7fff9000b028, str=0x7ffff43b3a40) at /home/psergey/dev-git/10.1-dbg6/sql/item_subselect.cc:1295 #7 0x0000555555cbca5e in Item_func_conv_charset::Item_func_conv_charset (this=0x7fff90021aa0, thd=0x555557ffda80, a=0x7fff9000b028, cs=0x555556c18b80, cache_if_const=true) at /home/psergey/dev-git/10.1-dbg6/sql/item_strfunc.h:975 #8 0x0000555555ca1a7b in Item::safe_charset_converter (this=0x7fff9000b028, thd=0x555557ffda80, tocs=0x555556c18b80) at /home/psergey/dev-git/10.1-dbg6/sql/item.cc:1084 #9 0x0000555555ca4452 in Item_func_or_sum::agg_item_set_converter (this=0x7fff90021828, coll=..., fname=0x555556460377 "<", args=0x7fff900218b0, nargs=2, flags=7, item_sep=1) at /home/psergey/dev-git/10.1-dbg6/sql/item.cc:2141 #10 0x0000555555cd4106 in Item_func_or_sum::agg_arg_charsets (this=0x7fff90021828, c=..., items=0x7fff900218b0, nitems=2, flags=7, item_sep=1) at /home/psergey/dev-git/10.1-dbg6/sql/item.h:3714 #11 0x0000555555cd414e in Item_func_or_sum::agg_arg_charsets_for_comparison (this=0x7fff90021828, c=..., items=0x7fff900218b0, nitems=2, item_sep=1) at /home/psergey/dev-git/10.1-dbg6/sql/item.h:3760 #12 0x0000555555cc10ac in Item_func::setup_args_and_comparator (this=0x7fff90021828, thd=0x555557ffda80, cmp=0x7fff900218e0) at /home/psergey/dev-git/10.1-dbg6/sql/item_cmpfunc.cc:509 #13 0x0000555555cc11b5 in Item_bool_rowready_func2::fix_length_and_dec (this=0x7fff90021828) at /home/psergey/dev-git/10.1-dbg6/sql/item_cmpfunc.cc:531 #14 0x0000555555cf022b in Item_func::fix_fields (this=0x7fff90021828, thd=0x555557ffda80, ref=0x7ffff43b3eb0) at /home/psergey/dev-git/10.1-dbg6/sql/item_func.cc:234 #15 0x0000555555d322fb in Item_allany_subselect::transform_into_max_min (this=0x7fff900077a8, join=0x7fff900087d0) at /home/psergey/dev-git/10.1-dbg6/sql/item_subselect.cc:1969 #16 0x0000555555bd557d in JOIN::transform_max_min_subquery (this=0x7fff900087d0) at /home/psergey/dev-git/10.1-dbg6/sql/opt_subselect.cc:901 #17 0x0000555555a8e9c6 in JOIN::optimize_inner (this=0x7fff900087d0) at /home/psergey/dev-git/10.1-dbg6/sql/sql_select.cc:1131 #18 0x0000555555a8e5f8 in JOIN::optimize (this=0x7fff900087d0) at /home/psergey/dev-git/10.1-dbg6/sql/sql_select.cc:1036 #19 0x0000555555a44f98 in st_select_lex::optimize_unflattened_subqueries (this=0x555558001b98, const_only=false) at /home/psergey/dev-git/10.1-dbg6/sql/sql_lex.cc:3760 #20 0x0000555555bdeaee in JOIN::optimize_unflattened_subqueries (this=0x7fff90007ca8) at /home/psergey/dev-git/10.1-dbg6/sql/opt_subselect.cc:5051 #21 0x0000555555a91dd7 in JOIN::optimize_inner (this=0x7fff90007ca8) at /home/psergey/dev-git/10.1-dbg6/sql/sql_select.cc:2043 #22 0x0000555555a8e5f8 in JOIN::optimize (this=0x7fff90007ca8) at /home/psergey/dev-git/10.1-dbg6/sql/sql_select.cc:1036 #23 0x0000555555a96bf2 in mysql_select (thd=0x555557ffda80, rref_pointer_array=0x555558001e10, tables=0x7fff900054b0, wild_num=1, fields=..., conds=0x7fff90007a80, og_num=0, order=0x0, group=0x0, having=0x0, proc_param=0x0, select_options=2147748608, result=0x7fff90007c88, unit=0x555558001498, select_lex=0x555558001b98) at /home/psergey/dev-git/10.1-dbg6/sql/sql_select.cc:3437 #24 0x0000555555a8c7fd in handle_select (thd=0x555557ffda80, lex=0x5555580013d0, result=0x7fff90007c88, setup_tables_done_option=0) at /home/psergey/dev-git/10.1-dbg6/sql/sql_select.cc:384 #25 0x0000555555a5cd61 in execute_sqlcom_select (thd=0x555557ffda80, all_tables=0x7fff900054b0) at /home/psergey/dev-git/10.1-dbg6/sql/sql_parse.cc:5894 #26 0x0000555555a52a21 in mysql_execute_command (thd=0x555557ffda80) at /home/psergey/dev-git/10.1-dbg6/sql/sql_parse.cc:2960 #27 0x0000555555a604b1 in mysql_parse (thd=0x555557ffda80, rawbuf=0x7fff90005258 "SELECT * FROM t1, t2 WHERE f3 = f2 AND f1 > ANY ( SELECT 'foo' UNION SELECT 'bar' )", length=83, parser_state=0x7ffff43b54f0) at /home/psergey/dev-git/10.1-dbg6/sql/sql_parse.cc:7314 #28 0x0000555555a4ec23 in dispatch_command (command=COM_QUERY, thd=0x555557ffda80, packet=0x555558004031 "SELECT * FROM t1, t2 WHERE f3 = f2 AND f1 > ANY ( SELECT 'foo' UNION SELECT 'bar' )", packet_length=83) at /home/psergey/dev-git/10.1-dbg6/sql/sql_parse.cc:1486 #29 0x0000555555a4d944 in do_command (thd=0x555557ffda80) at /home/psergey/dev-git/10.1-dbg6/sql/sql_parse.cc:1107 Note frame #0 and frame #17: JOIN::optimize_inner() has invoked a few functions which in turn invoked JOIN::optimize (and JOIN::optimize_inner()) for the same join. Before your patch this didn't happen, as the inner call to JOIN::optimize would find join->optimized= true and return immediately. In this particular example, the double-optimized join a degenerate "SELECT 'foo'", double-optimization has no effect. I tried to create an example with a non-degenerate join but didn't succeed. Looking at the code, I have no certainity that it's impossible to have two JOIN::optimize calls nest for a non-degenerate join. If you can explain why this is not possible, please do. If you can't prove this, I would suggest adding JOIN::inside_optimize_call and have "if (inside_optimize_call) return 0;" as the first line of JOIN::optimize. On Wed, Jun 22, 2016 at 11:17:44AM +0200, Oleksandr Byelkin wrote: > revision-id: 773ce408762b5f8256d4053b6d0d418d15657b92 (mariadb-10.1.14-24-g773ce40) > parent(s): 63120090f994cc78876944e9f7a76f53337fa46e > committer: Oleksandr Byelkin > timestamp: 2016-06-22 11:17:44 +0200 > message: > > MDEV-10045: Server crashes in Time_and_counter_tracker::incr_loops > > Do not set 'optimized' flag until whole optimization procedure is finished. > > --- > mysql-test/r/subselect.result | 11 +++++++++++ > mysql-test/r/subselect_no_exists_to_in.result | 11 +++++++++++ > mysql-test/r/subselect_no_mat.result | 11 +++++++++++ > mysql-test/r/subselect_no_opts.result | 11 +++++++++++ > mysql-test/r/subselect_no_scache.result | 11 +++++++++++ > mysql-test/r/subselect_no_semijoin.result | 11 +++++++++++ > mysql-test/t/subselect.test | 15 +++++++++++++++ > sql/sql_select.cc | 6 +++--- > 8 files changed, 84 insertions(+), 3 deletions(-) > > diff --git a/mysql-test/r/subselect.result b/mysql-test/r/subselect.result > index 6a53199..1ea7000 100644 > --- a/mysql-test/r/subselect.result > +++ b/mysql-test/r/subselect.result > @@ -7159,3 +7159,14 @@ f > foo > drop table t1; > SET NAMES default; > +# > +# MDEV-10045: Server crashes in Time_and_counter_tracker::incr_loops > +# > +SET NAMES utf8; > +CREATE TABLE t1 (f1 VARCHAR(3), f2 INT UNSIGNED) ENGINE=MyISAM; > +CREATE TABLE t2 (f3 INT) ENGINE=MyISAM; > +SELECT * FROM t1, t2 WHERE f3 = f2 AND f1 > ANY ( SELECT 'foo' UNION SELECT 'bar' ); > +f1 f2 f3 > +DROP TABLE t1, t2; > +SET NAMES default; > +End of 10.1 tests > diff --git a/mysql-test/r/subselect_no_exists_to_in.result b/mysql-test/r/subselect_no_exists_to_in.result > index aa68434..be5e5b6 100644 > --- a/mysql-test/r/subselect_no_exists_to_in.result > +++ b/mysql-test/r/subselect_no_exists_to_in.result > @@ -7159,6 +7159,17 @@ f > foo > drop table t1; > SET NAMES default; > +# > +# MDEV-10045: Server crashes in Time_and_counter_tracker::incr_loops > +# > +SET NAMES utf8; > +CREATE TABLE t1 (f1 VARCHAR(3), f2 INT UNSIGNED) ENGINE=MyISAM; > +CREATE TABLE t2 (f3 INT) ENGINE=MyISAM; > +SELECT * FROM t1, t2 WHERE f3 = f2 AND f1 > ANY ( SELECT 'foo' UNION SELECT 'bar' ); > +f1 f2 f3 > +DROP TABLE t1, t2; > +SET NAMES default; > +End of 10.1 tests > set optimizer_switch=default; > select @@optimizer_switch like '%exists_to_in=off%'; > @@optimizer_switch like '%exists_to_in=off%' > diff --git a/mysql-test/r/subselect_no_mat.result b/mysql-test/r/subselect_no_mat.result > index 754aec1..6dd227c 100644 > --- a/mysql-test/r/subselect_no_mat.result > +++ b/mysql-test/r/subselect_no_mat.result > @@ -7152,6 +7152,17 @@ f > foo > drop table t1; > SET NAMES default; > +# > +# MDEV-10045: Server crashes in Time_and_counter_tracker::incr_loops > +# > +SET NAMES utf8; > +CREATE TABLE t1 (f1 VARCHAR(3), f2 INT UNSIGNED) ENGINE=MyISAM; > +CREATE TABLE t2 (f3 INT) ENGINE=MyISAM; > +SELECT * FROM t1, t2 WHERE f3 = f2 AND f1 > ANY ( SELECT 'foo' UNION SELECT 'bar' ); > +f1 f2 f3 > +DROP TABLE t1, t2; > +SET NAMES default; > +End of 10.1 tests > set optimizer_switch=default; > select @@optimizer_switch like '%materialization=on%'; > @@optimizer_switch like '%materialization=on%' > diff --git a/mysql-test/r/subselect_no_opts.result b/mysql-test/r/subselect_no_opts.result > index e05dd4d..389c0f5 100644 > --- a/mysql-test/r/subselect_no_opts.result > +++ b/mysql-test/r/subselect_no_opts.result > @@ -7150,4 +7150,15 @@ f > foo > drop table t1; > SET NAMES default; > +# > +# MDEV-10045: Server crashes in Time_and_counter_tracker::incr_loops > +# > +SET NAMES utf8; > +CREATE TABLE t1 (f1 VARCHAR(3), f2 INT UNSIGNED) ENGINE=MyISAM; > +CREATE TABLE t2 (f3 INT) ENGINE=MyISAM; > +SELECT * FROM t1, t2 WHERE f3 = f2 AND f1 > ANY ( SELECT 'foo' UNION SELECT 'bar' ); > +f1 f2 f3 > +DROP TABLE t1, t2; > +SET NAMES default; > +End of 10.1 tests > set @optimizer_switch_for_subselect_test=null; > diff --git a/mysql-test/r/subselect_no_scache.result b/mysql-test/r/subselect_no_scache.result > index 71ade62..a90278c 100644 > --- a/mysql-test/r/subselect_no_scache.result > +++ b/mysql-test/r/subselect_no_scache.result > @@ -7165,6 +7165,17 @@ f > foo > drop table t1; > SET NAMES default; > +# > +# MDEV-10045: Server crashes in Time_and_counter_tracker::incr_loops > +# > +SET NAMES utf8; > +CREATE TABLE t1 (f1 VARCHAR(3), f2 INT UNSIGNED) ENGINE=MyISAM; > +CREATE TABLE t2 (f3 INT) ENGINE=MyISAM; > +SELECT * FROM t1, t2 WHERE f3 = f2 AND f1 > ANY ( SELECT 'foo' UNION SELECT 'bar' ); > +f1 f2 f3 > +DROP TABLE t1, t2; > +SET NAMES default; > +End of 10.1 tests > set optimizer_switch=default; > select @@optimizer_switch like '%subquery_cache=on%'; > @@optimizer_switch like '%subquery_cache=on%' > diff --git a/mysql-test/r/subselect_no_semijoin.result b/mysql-test/r/subselect_no_semijoin.result > index 43d191b..16aa7b8 100644 > --- a/mysql-test/r/subselect_no_semijoin.result > +++ b/mysql-test/r/subselect_no_semijoin.result > @@ -7150,5 +7150,16 @@ f > foo > drop table t1; > SET NAMES default; > +# > +# MDEV-10045: Server crashes in Time_and_counter_tracker::incr_loops > +# > +SET NAMES utf8; > +CREATE TABLE t1 (f1 VARCHAR(3), f2 INT UNSIGNED) ENGINE=MyISAM; > +CREATE TABLE t2 (f3 INT) ENGINE=MyISAM; > +SELECT * FROM t1, t2 WHERE f3 = f2 AND f1 > ANY ( SELECT 'foo' UNION SELECT 'bar' ); > +f1 f2 f3 > +DROP TABLE t1, t2; > +SET NAMES default; > +End of 10.1 tests > set @optimizer_switch_for_subselect_test=null; > set @join_cache_level_for_subselect_test=NULL; > diff --git a/mysql-test/t/subselect.test b/mysql-test/t/subselect.test > index 3599b52..c9c89a2 100644 > --- a/mysql-test/t/subselect.test > +++ b/mysql-test/t/subselect.test > @@ -6011,3 +6011,18 @@ INSERT INTO t1 VALUES ('foo'); > SELECT f FROM t1 WHERE f > ALL ( SELECT 'bar' UNION SELECT 'baz' ); > drop table t1; > SET NAMES default; > + > +--echo # > +--echo # MDEV-10045: Server crashes in Time_and_counter_tracker::incr_loops > +--echo # > +SET NAMES utf8; > + > +CREATE TABLE t1 (f1 VARCHAR(3), f2 INT UNSIGNED) ENGINE=MyISAM; > +CREATE TABLE t2 (f3 INT) ENGINE=MyISAM; > + > +SELECT * FROM t1, t2 WHERE f3 = f2 AND f1 > ANY ( SELECT 'foo' UNION SELECT 'bar' ); > + > +DROP TABLE t1, t2; > +SET NAMES default; > + > +--echo End of 10.1 tests > diff --git a/sql/sql_select.cc b/sql/sql_select.cc > index 4825726..ded59eb 100644 > --- a/sql/sql_select.cc > +++ b/sql/sql_select.cc > @@ -1040,7 +1040,7 @@ int JOIN::optimize() > and deleted. The second call will not produce a valid query plan, it will > short-circuit because optimized==TRUE. > > - "was_optimized != optimized" is here to handle this case: > + "!was_optimized" is here to handle this case: > - first optimization starts, gets an error (from a const. cheap > subquery), returns 1 > - another JOIN::optimize() call made, and now join->optimize() will > @@ -1049,7 +1049,7 @@ int JOIN::optimize() > Can have QEP_NOT_PRESENT_YET for degenerate queries (for example, > SELECT * FROM tbl LIMIT 0) > */ > - if (was_optimized != optimized && !res && have_query_plan != QEP_DELETED) > + if (!was_optimized && !res && have_query_plan != QEP_DELETED) > { > create_explain_query_if_not_exists(thd->lex, thd->mem_root); > have_query_plan= QEP_AVAILABLE; > @@ -1058,6 +1058,7 @@ int JOIN::optimize() > !skip_sort_order && !no_order && (order || group_list), > select_distinct); > } > + optimized= 1; > return res; > } > > @@ -1086,7 +1087,6 @@ JOIN::optimize_inner() > // to prevent double initialization on EXPLAIN > if (optimized) > DBUG_RETURN(0); > - optimized= 1; > DEBUG_SYNC(thd, "before_join_optimize"); > > THD_STAGE_INFO(thd, stage_optimizing); > _______________________________________________ > commits mailing list > commits(a)mariadb.org > https://lists.askmonty.org/cgi-bin/mailman/listinfo/commits -- BR Sergei -- Sergei Petrunia, Software Developer MariaDB Corporation | Skype: sergefp | Blog: http://s.petrunia.net/blog

1 0

Re: [Maria-developers] [Commits] 475cab8: MDEV-10050: Crash in subselect
by Sergey Petrunia 24 Jun '16

24 Jun '16

Hi Sanja, On Wed, Jun 22, 2016 at 02:17:06PM +0200, Oleksandr Byelkin wrote: > revision-id: 475cab835fb48c91d5cca649ab93917ec1718d75 (mariadb-5.5.50-6-g475cab8) > parent(s): a482e76e65a4fee70479e877929381c86b1ec62f > committer: Oleksandr Byelkin > timestamp: 2016-06-22 14:17:06 +0200 > message: > > MDEV-10050: Crash in subselect > > thd should not be taken earlier then fix_field and reset on fix_fields if it is needed. > > --- > sql/item_subselect.cc | 15 ++++++++++++--- > 1 file changed, 12 insertions(+), 3 deletions(-) > > diff --git a/sql/item_subselect.cc b/sql/item_subselect.cc > index ba67474..60cdd3f 100644 > --- a/sql/item_subselect.cc > +++ b/sql/item_subselect.cc > @@ -79,7 +79,9 @@ void Item_subselect::init(st_select_lex *select_lex, > DBUG_PRINT("enter", ("select_lex: 0x%lx this: 0x%lx", > (ulong) select_lex, (ulong) this)); > unit= select_lex->master_unit(); > - thd= unit->thd; > +#ifndef DBUG_OFF > + thd= 0; > +#endif So I've applied the patch, and I'm debugging this statement: prepare s from 'select a, (select max(one_k.a) from one_k where one_k.a <ten.a) from ten'; (this is just the first subquery I tried, nothing special about it). We arrive at the above #ifndef, and I have thd=0xa5a5a5a5a5, that is, it's uninitialized data. I let it execute further... > if (unit->item) > { > /* > Item can be changed in JOIN::prepare while engine in JOIN::optimize > => we do not copy old_engine here > */ > engine= unit->item->engine; > own_engine= FALSE; > parsing_place= unit->item->parsing_place; > unit->thd->change_item_tree((Item**)&unit->item, this); > engine->change_result(this, result, TRUE); > } > else > { > SELECT_LEX *outer_select= unit->outer_select(); > /* > do not take into account expression inside aggregate functions because > they can access original table fields > */ > parsing_place= (outer_select->in_sum_expr ? > NO_MATTER : > outer_select->parsing_place); > if (unit->is_union()) > engine= new subselect_union_engine(thd, unit, result, this); > else > engine= new subselect_single_select_engine(thd, select_lex, result, this); and it arrives here. Stepping into this call, I eventually reach where this stack trace shows: (gdb) wher #0 select_result::set_thd (this=0x7fffb002c130, thd_arg=0x0) at /home/psergey/dev-git/5.5/sql/sql_class.h:3316 #1 0x000000000086fd30 in subselect_engine::set_thd (this=0x7fffb002c150, thd_arg=0x0) at /home/psergey/dev-git/5.5/sql/item_subselect.cc:2892 #2 0x00000000007648e9 in subselect_engine::subselect_engine (this=0x7fffb002c150, thd_arg=0x0, si=0x7fffb002c010, res=0x7fffb002c130) at /home/psergey/dev-git/5.5/sql/item_subselect.h:726 #3 0x000000000086fd66 in subselect_single_select_engine::subselect_single_select_engine (this=0x7fffb002c150, thd_arg=0x0, select=0x7fffb002a8e0, result_arg=0x7fffb002c130, item_arg=0x7fffb002c010) at /home/psergey/dev-git/5.5/sql/item_subselect.cc:2902 #4 0x0000000000868232 in Item_subselect::init (this=0x7fffb002c010, select_lex=0x7fffb002a8e0, result=0x7fffb002c130) at /home/psergey/dev-git/5.5/sql/item_subselect.cc:111 #5 0x000000000086a0d1 in Item_singlerow_subselect::Item_singlerow_subselect (this=0x7fffb002c010, select_lex=0x7fffb002a8e0) at /home/psergey/dev-git/5.5/sql/item_subselect.cc:906 #6 0x000000000077d36f in MYSQLparse (thd=0x2fd99e0) at /home/psergey/dev-git/5.5/sql/sql_yacc.yy:8366 #7 0x00000000006214f2 in parse_sql (thd=0x2fd99e0, parser_state=0x7ffff7f81800, creation_ctx=0x0) at /home/psergey/dev-git/5.5/sql/sql_parse.cc:7818 #8 0x0000000000633dda in Prepared_statement::prepare (this=0x7fffb00167f0, packet=0x7fffb00063f8 "select a, (select max(one_k.a) from one_k where one_k.a <ten.a) from ten", packet_len=72) at /home/psergey/dev-git/5.5/sql/sql_prepare.cc:3353 #9 0x0000000000631a89 in mysql_sql_stmt_prepare (thd=0x2fd99e0) at /home/psergey/dev-git/5.5/sql/sql_prepare.cc:2458 #10 0x000000000061324d in mysql_execute_command (thd=0x2fd99e0) at /home/psergey/dev-git/5.5/sql/sql_parse.cc:2239 #11 0x000000000061d58d in mysql_parse (thd=0x2fd99e0, rawbuf=0x7fffb00062c8 "prepare s from 'select a, (select max(one_k.a) from one_k where one_k.a <ten.a) from ten'", length=89, parser_state=0x7ffff7f82540) at /home/psergey/dev-git/5.5/sql/sql_parse.cc:5934 Note that thd_arg is passing around un-initialized data all of this time. Is this how it is intended to work? This looks scary, a lot of functions are passed garbage as parameter. I assume the test suite passed with this patch, so these function don't do anything with the parameter (they probably save away the THD pointer, and then somebody assigns the pointer a valid value before it is used). Still, this looks really scary, we need to either fix this (strongly preferable), or add a lot of comments about it (acceptable as a last resort measure). > if (unit->item) > { > @@ -90,7 +92,7 @@ void Item_subselect::init(st_select_lex *select_lex, > engine= unit->item->engine; > own_engine= FALSE; > parsing_place= unit->item->parsing_place; > - thd->change_item_tree((Item**)&unit->item, this); > + unit->thd->change_item_tree((Item**)&unit->item, this); > engine->change_result(this, result, TRUE); > } > else > @@ -220,6 +222,10 @@ bool Item_subselect::fix_fields(THD *thd_param, Item **ref) > uint8 uncacheable; > bool res; > > + thd= thd_param; > + > + DBUG_ASSERT(unit->thd == thd); > + > status_var_increment(thd_param->status_var.feature_subquery); > > DBUG_ASSERT(fixed == 0); > @@ -2651,7 +2657,10 @@ bool Item_in_subselect::fix_fields(THD *thd_arg, Item **ref) > { > uint outer_cols_num; > List<Item> *inner_cols; > - char const *save_where= thd->where; > + char const *save_where= thd_arg->where; > + > + thd= thd_arg; > + DBUG_ASSERT(unit->thd == thd); > > if (test_strategy(SUBS_SEMI_JOIN)) > return !( (*ref)= new Item_int(1)); > _______________________________________________ > commits mailing list > commits(a)mariadb.org > https://lists.askmonty.org/cgi-bin/mailman/listinfo/commits -- BR Sergei -- Sergei Petrunia, Software Developer MariaDB Corporation | Skype: sergefp | Blog: http://s.petrunia.net/blog

2 1