March 2010 - developers - lists.mariadb.org

[Maria-developers] Updated (by Serg): INSTALL PLUGIN * (78)
by worklog-noreply＠askmonty.org 08 Mar '10

08 Mar '10

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: INSTALL PLUGIN * CREATION DATE..: Tue, 09 Feb 2010, 18:10 SUPERVISOR.....: Sergei IMPLEMENTOR....: COPIES TO......: CATEGORY.......: Server-RawIdeaBin TASK ID........: 78 (http://askmonty.org/worklog/?tid=78) VERSION........: WorkLog-3.4 STATUS.........: Cancelled PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 10 (hours remain) ORIG. ESTIMATE.: 10 PROGRESS NOTES: -=-=(Serg - Mon, 08 Mar 2010, 20:12)=-=- Status updated. --- /tmp/wklog.78.old.32035 2010-03-08 20:12:39.000000000 +0000 +++ /tmp/wklog.78.new.32035 2010-03-08 20:12:39.000000000 +0000 @@ -1 +1 @@ -Un-Assigned +Cancelled DESCRIPTION: InnoDB, XtraDB, PBXT (at least) come with a storage engine plugin and many information_schema plugins in one .so file. Currently one needs to install them all one by one. INSTALL PLUGIN * SONAME xxx would be a more convenient way to install everything at once. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] Updated (by Serg): INSTALL PLUGIN * (78)
by worklog-noreply＠askmonty.org 08 Mar '10

08 Mar '10

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: INSTALL PLUGIN * CREATION DATE..: Tue, 09 Feb 2010, 18:10 SUPERVISOR.....: Sergei IMPLEMENTOR....: COPIES TO......: CATEGORY.......: Server-RawIdeaBin TASK ID........: 78 (http://askmonty.org/worklog/?tid=78) VERSION........: WorkLog-3.4 STATUS.........: Cancelled PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 10 (hours remain) ORIG. ESTIMATE.: 10 PROGRESS NOTES: -=-=(Serg - Mon, 08 Mar 2010, 20:12)=-=- Status updated. --- /tmp/wklog.78.old.32035 2010-03-08 20:12:39.000000000 +0000 +++ /tmp/wklog.78.new.32035 2010-03-08 20:12:39.000000000 +0000 @@ -1 +1 @@ -Un-Assigned +Cancelled DESCRIPTION: InnoDB, XtraDB, PBXT (at least) come with a storage engine plugin and many information_schema plugins in one .so file. Currently one needs to install them all one by one. INSTALL PLUGIN * SONAME xxx would be a more convenient way to install everything at once. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] [Branch ~maria-captains/maria/5.1] Rev 2824: do not take LOCK_plugin for built-in plugins
by noreply＠launchpad.net 08 Mar '10

08 Mar '10

------------------------------------------------------------ revno: 2824 committer: Sergei Golubchik <sergii(a)pisem.net> branch nick: maria-5.1 timestamp: Mon 2010-03-08 18:05:09 +0100 message: do not take LOCK_plugin for built-in plugins modified: sql/sql_plugin.cc -- lp:maria https://code.launchpad.net/~maria-captains/maria/5.1 Your team Maria developers is subscribed to branch lp:maria. To unsubscribe from this branch go to https://code.launchpad.net/~maria-captains/maria/5.1/+edit-subscription.

1 0

[Maria-developers] [Branch ~maria-captains/maria/5.1] Rev 2823: small code cleanup - "good ifdef is no ifdef"
by noreply＠launchpad.net 08 Mar '10

08 Mar '10

------------------------------------------------------------ revno: 2823 committer: Sergei Golubchik <sergii(a)pisem.net> branch nick: maria-5.1 timestamp: Mon 2010-03-08 14:57:32 +0100 message: small code cleanup - "good ifdef is no ifdef" modified: sql/handler.cc sql/set_var.cc sql/sql_plugin.cc sql/sql_plugin.h sql/sql_select.cc sql/table.cc -- lp:maria https://code.launchpad.net/~maria-captains/maria/5.1 Your team Maria developers is subscribed to branch lp:maria. To unsubscribe from this branch go to https://code.launchpad.net/~maria-captains/maria/5.1/+edit-subscription.

1 0

[Maria-developers] Updated (by Hakan): Benchmark suite for sysbench (100)
by worklog-noreply＠askmonty.org 08 Mar '10

08 Mar '10

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Benchmark suite for sysbench CREATION DATE..: Thu, 04 Mar 2010, 17:46 SUPERVISOR.....: Igor IMPLEMENTOR....: Hakan COPIES TO......: CATEGORY.......: Other TASK ID........: 100 (http://askmonty.org/worklog/?tid=100) VERSION........: Benchmarks-3.0 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 40 (hours remain) ORIG. ESTIMATE.: 40 PROGRESS NOTES: -=-=(Hakan - Mon, 08 Mar 2010, 12:27)=-=- Low Level Design modified. --- /tmp/wklog.100.old.21404 2010-03-08 12:27:08.000000000 +0000 +++ /tmp/wklog.100.new.21404 2010-03-08 12:27:08.000000000 +0000 @@ -36,6 +36,7 @@ DATA_DIR of the database for faster starts. ** Between each run, run sync and clear file system caches with echo 3 > /proc/sys/vm/drop_caches (http://linux.die.net/man/5/proc) + ** Write out mysqld and sysbench options for reference. * Analyze numbers This is implemented in analyze-sysbench.php -=-=(Guest - Thu, 04 Mar 2010, 18:15)=-=- Low Level Design modified. --- /tmp/wklog.100.old.2695 2010-03-04 18:15:46.000000000 +0000 +++ /tmp/wklog.100.new.2695 2010-03-04 18:15:46.000000000 +0000 @@ -1 +1,112 @@ +All scripts can be found at lp:mariadb-tools/sysbench +* Run sysbench tests on a machine and collect numbers +This is implemented in run-sysbench.sh. Currently it supports: + ** Optionally pull of latest source from Launchpad and compile + ** Starting the server + ** Running each sysbench test for $LOOP_COUNT times and + $RUN_TIME time. + ** Number of concurrent clients is hardcoded + ** Machine specific configuration like location of binaries and + directories needed are in separate config files located at conf/<hostname>.inc + +The main loop of run-sysbench.sh is: + +start_mysqld +for SYSBENCH_TEST in $SYSBENCH_TESTS + for THREADS in $NUM_THREADS + while [ $k -lt $LOOP_COUNT ] + drop schema sbtest + create schema sbtest + + $SYSBENCH $SYSBENCH_OPTIONS prepare + $SYSBENCH $SYSBENCH_OPTIONS run + done + done +done + +Open items: + ** OProfile hook + ** iostat hook + ** sar -u (CPU utilization) hook + ** Crash detection + ** Error detection + ** Set random seed of sysbench to have better comparision + ** Restart mysqld from scratch for each run and copy away + DATA_DIR of the database for faster starts. + ** Between each run, run sync and clear file system caches with + echo 3 > /proc/sys/vm/drop_caches (http://linux.die.net/man/5/proc) + +* Analyze numbers +This is implemented in analyze-sysbench.php + +Open items: + ** Read result files from + ${RESULT_DIR}/${TODAY}/${PRODUCT}/${SYSBENCH_TEST}/${THREADS}/results.txt + + ** Collect OProfile, iostat, cpu utilization, and machine info + ** Detect errors and crashes. + ** Generate SQL INSERT strings for presentation usage + +The layout for storing the numbers is: + CREATE TABLE sysbench_run ( + id int unsigned NOT NULL auto_increment, + host varchar(80), -- Hostname we ran the test on. + run_date date, -- The day we ran the test. + sysbench_version varchar(32), -- Version of sysbench we used. + test_name varchar(32), -- Name of the sysbench test. + run_time int unsigned, -- Run time in seconds. + runs int unsigned, -- Number of iterations of the test. + PRIMARY KEY (id), + KEY (host), + KEY (run_date + ); + + CREATE TABLE sysbench_comment ( + id int unsigned NOT NULL auto_increment, + sysbench_run_id int unsigned NOT NULL, -- FK pointing to sysbench_run. + compile_info text, -- Compile options we used. + machine_info text, -- Details about the hardware. + sysbench_options text, -- The sysbench options we used. + PRIMARY KEY (id), + KEY (sysbench_run_id) + ); + + CREATE TABLE sysbench_result ( + id int unsigned NOT NULL auto_increment, + sysbench_run_id int unsigned NOT NULL, -- FK pointing to sysbench_run. + concurrency int unsigned, -- Concurrency level we used. + result decimal(7,2), -- The actual result. + io varchar(80), -- The IO from iostat. + cpu varchar(80), -- CPU utilization. + profile text, -- Profiling information. + error text, -- Error messages and stack traces. + PRIMARY KEY (id), + KEY (sysbench_run_id + ); + +* Generate a report out of the numbers +This script will generate a HTML version for putting up on the web and a +txt version for email usage. + +Open items: + ** Generate an overview table in the form of + Number of threads + 1 4 8 16 32 64 128 +sysbench test + delete 121.52 144.77 117.70 115.15 100.48 75.39 66.56 + mean value of runs + 1 first run + 2 second run + 3 third run + STDEV + STDEV in % of mean + CPU utilization (usr/sys/wait/idle) + IO (read/write) + +For HTML version additionally generate a graph with JPGraph. + +* Get machine(s) and run the test on a weekly basis and for each release +comparing with the prior release. + +* Email weekly reports and blog about it. DESCRIPTION: Create a benchmark suite for running sysbench * Run sysbench tests on a machine and collect numbers * Analyze numbers * Generate a report out of the numbers LOW-LEVEL DESIGN: All scripts can be found at lp:mariadb-tools/sysbench * Run sysbench tests on a machine and collect numbers This is implemented in run-sysbench.sh. Currently it supports: ** Optionally pull of latest source from Launchpad and compile ** Starting the server ** Running each sysbench test for $LOOP_COUNT times and $RUN_TIME time. ** Number of concurrent clients is hardcoded ** Machine specific configuration like location of binaries and directories needed are in separate config files located at conf/<hostname>.inc The main loop of run-sysbench.sh is: start_mysqld for SYSBENCH_TEST in $SYSBENCH_TESTS for THREADS in $NUM_THREADS while [ $k -lt $LOOP_COUNT ] drop schema sbtest create schema sbtest $SYSBENCH $SYSBENCH_OPTIONS prepare $SYSBENCH $SYSBENCH_OPTIONS run done done done Open items: ** OProfile hook ** iostat hook ** sar -u (CPU utilization) hook ** Crash detection ** Error detection ** Set random seed of sysbench to have better comparision ** Restart mysqld from scratch for each run and copy away DATA_DIR of the database for faster starts. ** Between each run, run sync and clear file system caches with echo 3 > /proc/sys/vm/drop_caches (http://linux.die.net/man/5/proc) ** Write out mysqld and sysbench options for reference. * Analyze numbers This is implemented in analyze-sysbench.php Open items: ** Read result files from ${RESULT_DIR}/${TODAY}/${PRODUCT}/${SYSBENCH_TEST}/${THREADS}/results.txt ** Collect OProfile, iostat, cpu utilization, and machine info ** Detect errors and crashes. ** Generate SQL INSERT strings for presentation usage The layout for storing the numbers is: CREATE TABLE sysbench_run ( id int unsigned NOT NULL auto_increment, host varchar(80), -- Hostname we ran the test on. run_date date, -- The day we ran the test. sysbench_version varchar(32), -- Version of sysbench we used. test_name varchar(32), -- Name of the sysbench test. run_time int unsigned, -- Run time in seconds. runs int unsigned, -- Number of iterations of the test. PRIMARY KEY (id), KEY (host), KEY (run_date ); CREATE TABLE sysbench_comment ( id int unsigned NOT NULL auto_increment, sysbench_run_id int unsigned NOT NULL, -- FK pointing to sysbench_run. compile_info text, -- Compile options we used. machine_info text, -- Details about the hardware. sysbench_options text, -- The sysbench options we used. PRIMARY KEY (id), KEY (sysbench_run_id) ); CREATE TABLE sysbench_result ( id int unsigned NOT NULL auto_increment, sysbench_run_id int unsigned NOT NULL, -- FK pointing to sysbench_run. concurrency int unsigned, -- Concurrency level we used. result decimal(7,2), -- The actual result. io varchar(80), -- The IO from iostat. cpu varchar(80), -- CPU utilization. profile text, -- Profiling information. error text, -- Error messages and stack traces. PRIMARY KEY (id), KEY (sysbench_run_id ); * Generate a report out of the numbers This script will generate a HTML version for putting up on the web and a txt version for email usage. Open items: ** Generate an overview table in the form of Number of threads 1 4 8 16 32 64 128 sysbench test delete 121.52 144.77 117.70 115.15 100.48 75.39 66.56 mean value of runs 1 first run 2 second run 3 third run STDEV STDEV in % of mean CPU utilization (usr/sys/wait/idle) IO (read/write) For HTML version additionally generate a graph with JPGraph. * Get machine(s) and run the test on a weekly basis and for each release comparing with the prior release. * Email weekly reports and blog about it. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] Updated (by Hakan): Benchmark suite for sysbench (100)
by worklog-noreply＠askmonty.org 08 Mar '10

08 Mar '10

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Benchmark suite for sysbench CREATION DATE..: Thu, 04 Mar 2010, 17:46 SUPERVISOR.....: Igor IMPLEMENTOR....: Hakan COPIES TO......: CATEGORY.......: Other TASK ID........: 100 (http://askmonty.org/worklog/?tid=100) VERSION........: Benchmarks-3.0 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 40 (hours remain) ORIG. ESTIMATE.: 40 PROGRESS NOTES: -=-=(Hakan - Mon, 08 Mar 2010, 12:27)=-=- Low Level Design modified. --- /tmp/wklog.100.old.21404 2010-03-08 12:27:08.000000000 +0000 +++ /tmp/wklog.100.new.21404 2010-03-08 12:27:08.000000000 +0000 @@ -36,6 +36,7 @@ DATA_DIR of the database for faster starts. ** Between each run, run sync and clear file system caches with echo 3 > /proc/sys/vm/drop_caches (http://linux.die.net/man/5/proc) + ** Write out mysqld and sysbench options for reference. * Analyze numbers This is implemented in analyze-sysbench.php -=-=(Guest - Thu, 04 Mar 2010, 18:15)=-=- Low Level Design modified. --- /tmp/wklog.100.old.2695 2010-03-04 18:15:46.000000000 +0000 +++ /tmp/wklog.100.new.2695 2010-03-04 18:15:46.000000000 +0000 @@ -1 +1,112 @@ +All scripts can be found at lp:mariadb-tools/sysbench +* Run sysbench tests on a machine and collect numbers +This is implemented in run-sysbench.sh. Currently it supports: + ** Optionally pull of latest source from Launchpad and compile + ** Starting the server + ** Running each sysbench test for $LOOP_COUNT times and + $RUN_TIME time. + ** Number of concurrent clients is hardcoded + ** Machine specific configuration like location of binaries and + directories needed are in separate config files located at conf/<hostname>.inc + +The main loop of run-sysbench.sh is: + +start_mysqld +for SYSBENCH_TEST in $SYSBENCH_TESTS + for THREADS in $NUM_THREADS + while [ $k -lt $LOOP_COUNT ] + drop schema sbtest + create schema sbtest + + $SYSBENCH $SYSBENCH_OPTIONS prepare + $SYSBENCH $SYSBENCH_OPTIONS run + done + done +done + +Open items: + ** OProfile hook + ** iostat hook + ** sar -u (CPU utilization) hook + ** Crash detection + ** Error detection + ** Set random seed of sysbench to have better comparision + ** Restart mysqld from scratch for each run and copy away + DATA_DIR of the database for faster starts. + ** Between each run, run sync and clear file system caches with + echo 3 > /proc/sys/vm/drop_caches (http://linux.die.net/man/5/proc) + +* Analyze numbers +This is implemented in analyze-sysbench.php + +Open items: + ** Read result files from + ${RESULT_DIR}/${TODAY}/${PRODUCT}/${SYSBENCH_TEST}/${THREADS}/results.txt + + ** Collect OProfile, iostat, cpu utilization, and machine info + ** Detect errors and crashes. + ** Generate SQL INSERT strings for presentation usage + +The layout for storing the numbers is: + CREATE TABLE sysbench_run ( + id int unsigned NOT NULL auto_increment, + host varchar(80), -- Hostname we ran the test on. + run_date date, -- The day we ran the test. + sysbench_version varchar(32), -- Version of sysbench we used. + test_name varchar(32), -- Name of the sysbench test. + run_time int unsigned, -- Run time in seconds. + runs int unsigned, -- Number of iterations of the test. + PRIMARY KEY (id), + KEY (host), + KEY (run_date + ); + + CREATE TABLE sysbench_comment ( + id int unsigned NOT NULL auto_increment, + sysbench_run_id int unsigned NOT NULL, -- FK pointing to sysbench_run. + compile_info text, -- Compile options we used. + machine_info text, -- Details about the hardware. + sysbench_options text, -- The sysbench options we used. + PRIMARY KEY (id), + KEY (sysbench_run_id) + ); + + CREATE TABLE sysbench_result ( + id int unsigned NOT NULL auto_increment, + sysbench_run_id int unsigned NOT NULL, -- FK pointing to sysbench_run. + concurrency int unsigned, -- Concurrency level we used. + result decimal(7,2), -- The actual result. + io varchar(80), -- The IO from iostat. + cpu varchar(80), -- CPU utilization. + profile text, -- Profiling information. + error text, -- Error messages and stack traces. + PRIMARY KEY (id), + KEY (sysbench_run_id + ); + +* Generate a report out of the numbers +This script will generate a HTML version for putting up on the web and a +txt version for email usage. + +Open items: + ** Generate an overview table in the form of + Number of threads + 1 4 8 16 32 64 128 +sysbench test + delete 121.52 144.77 117.70 115.15 100.48 75.39 66.56 + mean value of runs + 1 first run + 2 second run + 3 third run + STDEV + STDEV in % of mean + CPU utilization (usr/sys/wait/idle) + IO (read/write) + +For HTML version additionally generate a graph with JPGraph. + +* Get machine(s) and run the test on a weekly basis and for each release +comparing with the prior release. + +* Email weekly reports and blog about it. DESCRIPTION: Create a benchmark suite for running sysbench * Run sysbench tests on a machine and collect numbers * Analyze numbers * Generate a report out of the numbers LOW-LEVEL DESIGN: All scripts can be found at lp:mariadb-tools/sysbench * Run sysbench tests on a machine and collect numbers This is implemented in run-sysbench.sh. Currently it supports: ** Optionally pull of latest source from Launchpad and compile ** Starting the server ** Running each sysbench test for $LOOP_COUNT times and $RUN_TIME time. ** Number of concurrent clients is hardcoded ** Machine specific configuration like location of binaries and directories needed are in separate config files located at conf/<hostname>.inc The main loop of run-sysbench.sh is: start_mysqld for SYSBENCH_TEST in $SYSBENCH_TESTS for THREADS in $NUM_THREADS while [ $k -lt $LOOP_COUNT ] drop schema sbtest create schema sbtest $SYSBENCH $SYSBENCH_OPTIONS prepare $SYSBENCH $SYSBENCH_OPTIONS run done done done Open items: ** OProfile hook ** iostat hook ** sar -u (CPU utilization) hook ** Crash detection ** Error detection ** Set random seed of sysbench to have better comparision ** Restart mysqld from scratch for each run and copy away DATA_DIR of the database for faster starts. ** Between each run, run sync and clear file system caches with echo 3 > /proc/sys/vm/drop_caches (http://linux.die.net/man/5/proc) ** Write out mysqld and sysbench options for reference. * Analyze numbers This is implemented in analyze-sysbench.php Open items: ** Read result files from ${RESULT_DIR}/${TODAY}/${PRODUCT}/${SYSBENCH_TEST}/${THREADS}/results.txt ** Collect OProfile, iostat, cpu utilization, and machine info ** Detect errors and crashes. ** Generate SQL INSERT strings for presentation usage The layout for storing the numbers is: CREATE TABLE sysbench_run ( id int unsigned NOT NULL auto_increment, host varchar(80), -- Hostname we ran the test on. run_date date, -- The day we ran the test. sysbench_version varchar(32), -- Version of sysbench we used. test_name varchar(32), -- Name of the sysbench test. run_time int unsigned, -- Run time in seconds. runs int unsigned, -- Number of iterations of the test. PRIMARY KEY (id), KEY (host), KEY (run_date ); CREATE TABLE sysbench_comment ( id int unsigned NOT NULL auto_increment, sysbench_run_id int unsigned NOT NULL, -- FK pointing to sysbench_run. compile_info text, -- Compile options we used. machine_info text, -- Details about the hardware. sysbench_options text, -- The sysbench options we used. PRIMARY KEY (id), KEY (sysbench_run_id) ); CREATE TABLE sysbench_result ( id int unsigned NOT NULL auto_increment, sysbench_run_id int unsigned NOT NULL, -- FK pointing to sysbench_run. concurrency int unsigned, -- Concurrency level we used. result decimal(7,2), -- The actual result. io varchar(80), -- The IO from iostat. cpu varchar(80), -- CPU utilization. profile text, -- Profiling information. error text, -- Error messages and stack traces. PRIMARY KEY (id), KEY (sysbench_run_id ); * Generate a report out of the numbers This script will generate a HTML version for putting up on the web and a txt version for email usage. Open items: ** Generate an overview table in the form of Number of threads 1 4 8 16 32 64 128 sysbench test delete 121.52 144.77 117.70 115.15 100.48 75.39 66.56 mean value of runs 1 first run 2 second run 3 third run STDEV STDEV in % of mean CPU utilization (usr/sys/wait/idle) IO (read/write) For HTML version additionally generate a graph with JPGraph. * Get machine(s) and run the test on a weekly basis and for each release comparing with the prior release. * Email weekly reports and blog about it. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] Valgrind error in table elimination in MySQL 5.1.44 merge
by Kristian Nielsen 08 Mar '10

08 Mar '10

Hi Sergey, When I merged MySQL 5.1.44, there was a new testcase introduced: CREATE TABLE t1 (a VARCHAR(10), FULLTEXT KEY a (a)); INSERT INTO t1 VALUES (1),(2); CREATE TABLE t2 (b INT); INSERT INTO t2 VALUES (1),(2); EXPLAIN EXTENDED SELECT * FROM t1 UNION SELECT * FROM t1 ORDER BY (SELECT a FROM t2 WHERE b = 12); DROP TABLE t1,t2; This crashes the server in MariaDB. This is Bug#49734, but this bug is marked private. After my merge, the server no longer crashes, but there is a Valgrind error about uninitialised value: ==14946== Thread 9: ==14946== Conditional jump or move depends on uninitialised value(s) ==14946== at 0x71E89D: select_describe(JOIN*, bool, bool, bool, char const*) (sql_select.cc:16883) ==14946== by 0x71F585: JOIN::exec() (sql_select.cc:1837) ==14946== by 0x8467D1: st_select_lex_unit::exec() (sql_union.cc:513) ==14946== by 0x71C016: mysql_explain_union(THD*, st_select_lex_unit*, select_result*) (sql_select.cc:16927) ==14946== by 0x6880BC: execute_sqlcom_select(THD*, TABLE_LIST*) (sql_parse.cc:5091) ==14946== by 0x68A260: mysql_execute_command(THD*) (sql_parse.cc:2299) ==14946== by 0x692EB3: mysql_parse(THD*, char const*, unsigned, char const**) (sql_parse.cc:6034) ==14946== by 0x693CC5: dispatch_command(enum_server_command, THD*, char*, unsigned) (sql_parse.cc:1247) ==14946== by 0x6951C3: do_command(THD*) (sql_parse.cc:886) ==14946== by 0x6805A0: handle_one_connection (sql_connect.cc:1132) ==14946== by 0x50463F6: start_thread (pthread_create.c:297) ==14946== by 0x6026B4C: clone (in /usr/lib/debug/libc-2.7.so) The tree on Launchpad containing the merge and showing the error is lp:~maria-captains/maria/mariadb-5.1-knielsen Problem is seen in this code in sql/sql_select.cc, static void select_describe(): if (!(unit->item && unit->item->eliminated)) { if (mysql_explain_union(thd, unit, result)) DBUG_VOID_RETURN; } Here unit->item->eliminated is uninitialised. Checking the code in item_subselect.cc, I see that item->eliminated is not initialised in the constructor, only in fix_fields() and mark_as_eliminated_processor(). However, I do not know the code enough to know if the problem is missing initialisation in the constructor, or missing call of fix_fields() (or maybe something else). Can you help? Looks related to the table elimination patch. Since the bug report is privat, I have instead appended the patch in MySQL for the bug. - Kristian. ----------------------------------------------------------------------- revno: 2502.884.63 committer: Georgi Kodinov <joro(a)sun.com> branch nick: B49734-5.1-bugteam timestamp: Tue 2009-12-22 17:52:15 +0200 message: Bug #49734: Crash on EXPLAIN EXTENDED UNION ... ORDER BY <any non-const-function> Several problems fixed : 1. Non constant expressions in UNION ... ORDER BY were not correctly cleaned up in st_select_lex_unit::cleanup() causing crashes in EXPLAIN EXTENDED because of fields quoted by these expressions pointing to the already freed temporary table used to calculate the UNION. Fixed by correctly cleaning up expressions of any depth. 2. Subqueries in the order by part of UNION ... ORDER BY ... caused a crash in EXPLAIN EXTENDED because of a transformation attempt made during EXPLAIN EXTENDED execution. Fixed by not doing the transformation when in EXPLAIN. 3. Fulltext functions caused crash when in the ORDER BY part of an un-parenthesized UNION that gets "promoted" to be valid for the whole union, e.g. SELECT * FROM t1 UNION SELECT * FROM t2 ORDER BY MATCHES (a) AGAINST ('abc' IN BOOLEAN MODE). This is a case that demonstrates a more general problem of parts of the query being moved to another level. When doing such transformation late in the optimization run when most of the flags about the contents of the query are already aggregated it's possible to "split" the flags so that they correctly reflect the new queries after the transformation. In specific the ST_SELECT_LEX::ftfunc_list is holding all the free text function for all the parts of the second SELECT in the UNION and we don't know what part of that is in the ORDER BY that we're to move to the UNION level and what part is about the other parts of the second SELECT. Fixed by throwing and error when such statements are about to be processed by adding a check for the presence of MATCH() inside the ORDER BY clause that's going to get promoted to UNION. To workaround this new limitation one must parenthesize the UNION SELECTs and provide a real global ORDER BY for the UNION outside of the parenthesis. diff: === modified file 'mysql-test/r/fulltext_order_by.result' --- mysql-test/r/fulltext_order_by.result 2005-08-12 16:27:54 +0000 +++ mysql-test/r/fulltext_order_by.result 2009-12-22 15:52:15 +0000 @@ -126,7 +126,7 @@ a.text, b.id, b.betreff order by match(b.betreff) against ('+abc' in boolean mode) desc; -ERROR 42S22: Unknown column 'b.betreff' in 'order clause' +ERROR 42000: Incorrect usage/placement of 'MATCH()' select a.text, b.id, b.betreff from t2 a inner join t3 b on a.id = b.forum inner join @@ -142,7 +142,7 @@ match(c.beitrag) against ('+abc' in boolean mode) order by match(b.betreff) against ('+abc' in boolean mode) desc; -ERROR 42S22: Unknown column 'b.betreff' in 'order clause' +ERROR 42000: Incorrect usage/placement of 'MATCH()' select a.text, b.id, b.betreff from t2 a inner join t3 b on a.id = b.forum inner join @@ -158,7 +158,7 @@ match(c.beitrag) against ('+abc' in boolean mode) order by match(betreff) against ('+abc' in boolean mode) desc; -text id betreff +ERROR 42000: Incorrect usage/placement of 'MATCH()' (select b.id, b.betreff from t3 b) union (select b.id, b.betreff from t3 b) order by match(betreff) against ('+abc' in boolean mode) desc; === modified file 'mysql-test/r/union.result' --- mysql-test/r/union.result 2009-05-15 07:11:07 +0000 +++ mysql-test/r/union.result 2009-12-22 15:52:15 +0000 @@ -1588,3 +1588,66 @@ Note 1003 select '0' AS `a` from `test`.`t1` union select '0' AS `a` from `test`.`t1` order by `a` DROP TABLE t1; End of 5.0 tests +# +# Bug #49734: Crash on EXPLAIN EXTENDED UNION ... ORDER BY +# <any non-const-function> +# +CREATE TABLE t1 (a VARCHAR(10), FULLTEXT KEY a (a)); +INSERT INTO t1 VALUES (1),(2); +CREATE TABLE t2 (b INT); +INSERT INTO t2 VALUES (1),(2); +# Should not crash +EXPLAIN EXTENDED +SELECT * FROM t1 UNION SELECT * FROM t1 ORDER BY a + 12; +id select_type table type possible_keys key key_len ref rows filtered Extra +1 PRIMARY t1 ALL NULL NULL NULL NULL 2 100.00 +2 UNION t1 ALL NULL NULL NULL NULL 2 100.00 +NULL UNION RESULT <union1,2> ALL NULL NULL NULL NULL NULL NULL Using filesort +Warnings: +Note 1003 select `test`.`t1`.`a` AS `a` from `test`.`t1` union select `test`.`t1`.`a` AS `a` from `test`.`t1` order by (`a` + 12) +# Should not crash +SELECT * FROM t1 UNION SELECT * FROM t1 ORDER BY a + 12; +a +1 +2 +# Should not crash +EXPLAIN EXTENDED +SELECT * FROM t1 UNION SELECT * FROM t1 +ORDER BY MATCH(a) AGAINST ('+abc' IN BOOLEAN MODE); +ERROR 42000: Incorrect usage/placement of 'MATCH()' +# Should not crash +SELECT * FROM t1 UNION SELECT * FROM t1 +ORDER BY MATCH(a) AGAINST ('+abc' IN BOOLEAN MODE); +ERROR 42000: Incorrect usage/placement of 'MATCH()' +# Should not crash +(SELECT * FROM t1) UNION (SELECT * FROM t1) +ORDER BY MATCH(a) AGAINST ('+abc' IN BOOLEAN MODE); +a +1 +2 +# Should not crash +EXPLAIN EXTENDED +SELECT * FROM t1 UNION SELECT * FROM t1 +ORDER BY (SELECT a FROM t2 WHERE b = 12); +id select_type table type possible_keys key key_len ref rows filtered Extra +1 PRIMARY t1 ALL NULL NULL NULL NULL 2 100.00 +2 UNION t1 ALL NULL NULL NULL NULL 2 100.00 +3 SUBQUERY t2 ALL NULL NULL NULL NULL 2 100.00 Using where +NULL UNION RESULT <union1,2> ALL NULL NULL NULL NULL NULL NULL Using filesort +Warnings: +Note 1276 Field or reference 'test.t1.a' of SELECT #3 was resolved in SELECT #2 +Note 1003 select `test`.`t1`.`a` AS `a` from `test`.`t1` union select `test`.`t1`.`a` AS `a` from `test`.`t1` order by (select `test`.`t1`.`a` AS `a` from `test`.`t2` where (`test`.`t2`.`b` = 12)) +# Should not crash +SELECT * FROM t1 UNION SELECT * FROM t1 +ORDER BY (SELECT a FROM t2 WHERE b = 12); +a +1 +2 +# Should not crash +SELECT * FROM t2 UNION SELECT * FROM t2 +ORDER BY (SELECT * FROM t1 WHERE MATCH(a) AGAINST ('+abc' IN BOOLEAN MODE)); +b +1 +2 +DROP TABLE t1,t2; +End of 5.1 tests === modified file 'mysql-test/t/fulltext_order_by.test' --- mysql-test/t/fulltext_order_by.test 2005-08-12 16:27:54 +0000 +++ mysql-test/t/fulltext_order_by.test 2009-12-22 15:52:15 +0000 @@ -80,7 +80,7 @@ FULLTEXT KEY betreff (betreff) ) ENGINE=MyISAM DEFAULT CHARSET=utf8 AUTO_INCREMENT=996 ; ---error 1054 +--error ER_CANT_USE_OPTION_HERE select a.text, b.id, b.betreff from t2 a inner join t3 b on a.id = b.forum inner join @@ -100,7 +100,7 @@ order by match(b.betreff) against ('+abc' in boolean mode) desc; ---error 1054 +--error ER_CANT_USE_OPTION_HERE select a.text, b.id, b.betreff from t2 a inner join t3 b on a.id = b.forum inner join @@ -117,6 +117,7 @@ order by match(b.betreff) against ('+abc' in boolean mode) desc; +--error ER_CANT_USE_OPTION_HERE select a.text, b.id, b.betreff from t2 a inner join t3 b on a.id = b.forum inner join === modified file 'mysql-test/t/union.test' --- mysql-test/t/union.test 2009-05-15 07:11:07 +0000 +++ mysql-test/t/union.test 2009-12-22 15:52:15 +0000 @@ -1102,3 +1102,56 @@ --echo End of 5.0 tests + + +--echo # +--echo # Bug #49734: Crash on EXPLAIN EXTENDED UNION ... ORDER BY +--echo # <any non-const-function> +--echo # + +CREATE TABLE t1 (a VARCHAR(10), FULLTEXT KEY a (a)); +INSERT INTO t1 VALUES (1),(2); +CREATE TABLE t2 (b INT); +INSERT INTO t2 VALUES (1),(2); + +--echo # Should not crash +EXPLAIN EXTENDED +SELECT * FROM t1 UNION SELECT * FROM t1 ORDER BY a + 12; + +--echo # Should not crash +SELECT * FROM t1 UNION SELECT * FROM t1 ORDER BY a + 12; + + +--echo # Should not crash +--error ER_CANT_USE_OPTION_HERE +EXPLAIN EXTENDED +SELECT * FROM t1 UNION SELECT * FROM t1 + ORDER BY MATCH(a) AGAINST ('+abc' IN BOOLEAN MODE); + +--echo # Should not crash +--error ER_CANT_USE_OPTION_HERE +SELECT * FROM t1 UNION SELECT * FROM t1 + ORDER BY MATCH(a) AGAINST ('+abc' IN BOOLEAN MODE); + +--echo # Should not crash +(SELECT * FROM t1) UNION (SELECT * FROM t1) + ORDER BY MATCH(a) AGAINST ('+abc' IN BOOLEAN MODE); + + +--echo # Should not crash +EXPLAIN EXTENDED +SELECT * FROM t1 UNION SELECT * FROM t1 + ORDER BY (SELECT a FROM t2 WHERE b = 12); + +--echo # Should not crash +SELECT * FROM t1 UNION SELECT * FROM t1 + ORDER BY (SELECT a FROM t2 WHERE b = 12); + +--echo # Should not crash +SELECT * FROM t2 UNION SELECT * FROM t2 + ORDER BY (SELECT * FROM t1 WHERE MATCH(a) AGAINST ('+abc' IN BOOLEAN MODE)); + +DROP TABLE t1,t2; + + +--echo End of 5.1 tests === modified file 'sql/item.h' --- sql/item.h 2009-12-13 20:57:57 +0000 +++ sql/item.h 2009-12-22 15:52:15 +0000 @@ -967,6 +967,23 @@ return FALSE; } + /** + Find a function of a given type + + @param arg the function type to search (enum Item_func::Functype) + @return + @retval TRUE the function type we're searching for is found + @retval FALSE the function type wasn't found + + @description + This function can be used (together with Item::walk()) to find functions + in an item tree fragment. + */ + virtual bool find_function_processor (uchar *arg) + { + return FALSE; + } + /* For SP local variable returns pointer to Item representing its current value and pointer to current Item otherwise. === modified file 'sql/item_func.h' --- sql/item_func.h 2009-12-13 20:57:57 +0000 +++ sql/item_func.h 2009-12-22 15:52:15 +0000 @@ -212,6 +212,11 @@ { return has_timestamp_args(); } + + virtual bool find_function_processor (uchar *arg) + { + return functype() == *(Functype *) arg; + } }; === modified file 'sql/sql_select.cc' --- sql/sql_select.cc 2009-12-17 09:55:18 +0000 +++ sql/sql_select.cc 2009-12-22 15:52:15 +0000 @@ -519,7 +519,7 @@ thd->lex->allow_sum_func= save_allow_sum_func; } - if (!thd->lex->view_prepare_mode) + if (!thd->lex->view_prepare_mode && !(select_options & SELECT_DESCRIBE)) { Item_subselect *subselect; /* Is it subselect? */ === modified file 'sql/sql_union.cc' --- sql/sql_union.cc 2009-05-15 07:11:07 +0000 +++ sql/sql_union.cc 2009-12-22 15:52:15 +0000 @@ -335,6 +335,35 @@ } } + /* + Disable the usage of fulltext searches in the last union branch. + This is a temporary 5.x limitation because of the way the fulltext + search functions are handled by the optimizer. + This is manifestation of the more general problems of "taking away" + parts of a SELECT statement post-fix_fields(). This is generally not + doable since various flags are collected in various places (e.g. + SELECT_LEX) that carry information about the presence of certain + expressions or constructs in the parts of the query. + When part of the query is taken away it's not clear how to "divide" + the meaning of these accumulated flags and what to carry over to the + recipient query (SELECT_LEX). + */ + if (global_parameters->ftfunc_list->elements && + global_parameters->order_list.elements && + global_parameters != fake_select_lex) + { + ORDER *ord; + Item_func::Functype ft= Item_func::FT_FUNC; + for (ord= (ORDER*)global_parameters->order_list.first; ord; ord= ord->next) + if ((*ord->item)->walk (&Item::find_function_processor, FALSE, + (uchar *) &ft)) + { + my_error (ER_CANT_USE_OPTION_HERE, MYF(0), "MATCH()"); + goto err; + } + } + + create_options= (first_sl->options | thd_arg->options | TMP_TABLE_ALL_COLUMNS); /* @@ -669,7 +698,7 @@ { ORDER *ord; for (ord= (ORDER*)global_parameters->order_list.first; ord; ord= ord->next) - (*ord->item)->cleanup(); + (*ord->item)->walk (&Item::cleanup_processor, 0, 0); } }

1 0

[Maria-developers] [Branch ~maria-captains/maria/5.1] Rev 2822: LPBUG#524025 Running RQG outer_join test leads to crash
by noreply＠launchpad.net 07 Mar '10

07 Mar '10

------------------------------------------------------------ revno: 2822 committer: Sergey Petrunya <psergey(a)askmonty.org> branch nick: maria-5.1 timestamp: Sat 2010-03-06 15:09:02 +0300 message: LPBUG#524025 Running RQG outer_join test leads to crash Save no-records constant tables in JOIN::const_table_map before we invoke eliminate_tables(). Failure to do so caused crash when the same table was marked as constant two times modified: mysql-test/r/table_elim.result mysql-test/t/table_elim.test sql/sql_select.cc -- lp:maria https://code.launchpad.net/~maria-captains/maria/5.1 Your team Maria developers is subscribed to branch lp:maria. To unsubscribe from this branch go to https://code.launchpad.net/~maria-captains/maria/5.1/+edit-subscription.

1 0

[Maria-developers] Rev 2773: BUG#49129: Wrong result with IN-subquery with join_cache_level=6 and firstmatch=off in file:///home/psergey/dev/maria-5.3-subqueries-r7/
by Sergey Petrunya 07 Mar '10

07 Mar '10

At file:///home/psergey/dev/maria-5.3-subqueries-r7/ ------------------------------------------------------------ revno: 2773 revision-id: psergey(a)askmonty.org-20100307154145-ksby2b1l0sqm1xne parent: igor(a)askmonty.org-20100306191455-g2srime08w9u6l49 committer: Sergey Petrunya <psergey(a)askmonty.org> branch nick: maria-5.3-subqueries-r7 timestamp: Sun 2010-03-07 18:41:45 +0300 message: BUG#49129: Wrong result with IN-subquery with join_cache_level=6 and firstmatch=off - The problem was that DuplicateWeedout strategy setup code wasn't aware of the fact that join buffering will be used and applied optimization that doesn't work together with join buffering. Fixed by making DuplicateWeedout setup code to have a pessimistic check about whether there is a chance that join buffering will be used. - Make JOIN_CACHE_BKA::init() correctly process Copy_field elements that denote saving current rowids in the join buffer. === modified file 'mysql-test/r/subselect_sj2.result' --- a/mysql-test/r/subselect_sj2.result 2010-02-17 10:47:55 +0000 +++ b/mysql-test/r/subselect_sj2.result 2010-03-07 15:41:45 +0000 @@ -264,8 +264,8 @@ from t0 where a in (select t2.a+t3.a from t1 left join (t2 join t3) on t2.a=t1.a and t3.a=t1.a); id select_type table type possible_keys key key_len ref rows Extra -1 PRIMARY t0 ALL NULL NULL NULL NULL 10 -1 PRIMARY t1 index NULL a 5 NULL 10 Using index; Start temporary; Using join buffer +1 PRIMARY t0 ALL NULL NULL NULL NULL 10 Start temporary +1 PRIMARY t1 index NULL a 5 NULL 10 Using index; Using join buffer 1 PRIMARY t2 ref a a 5 test.t1.a 1 Using index 1 PRIMARY t3 ref a a 5 test.t1.a 1 Using where; Using index; End temporary drop table t0, t1,t2,t3; === modified file 'mysql-test/r/subselect_sj2_jcl6.result' --- a/mysql-test/r/subselect_sj2_jcl6.result 2010-03-05 18:54:48 +0000 +++ b/mysql-test/r/subselect_sj2_jcl6.result 2010-03-07 15:41:45 +0000 @@ -268,8 +268,8 @@ from t0 where a in (select t2.a+t3.a from t1 left join (t2 join t3) on t2.a=t1.a and t3.a=t1.a); id select_type table type possible_keys key key_len ref rows Extra -1 PRIMARY t0 ALL NULL NULL NULL NULL 10 -1 PRIMARY t1 index NULL a 5 NULL 10 Using index; Start temporary; Using join buffer +1 PRIMARY t0 ALL NULL NULL NULL NULL 10 Start temporary +1 PRIMARY t1 index NULL a 5 NULL 10 Using index; Using join buffer 1 PRIMARY t2 ref a a 5 test.t1.a 1 Using index 1 PRIMARY t3 ref a a 5 test.t1.a 1 Using where; Using index; End temporary drop table t0, t1,t2,t3; @@ -421,20 +421,23 @@ where t0.a in ( select t1.a from t1,t2 where t2.a=t0.a and t1.b=t2.b); id select_type table type possible_keys key key_len ref rows filtered Extra -1 PRIMARY t0 ALL NULL NULL NULL NULL 5 100.00 -1 PRIMARY t1 ref a a 5 test.t0.a 1 100.00 Start temporary; Using join buffer +1 PRIMARY t0 ALL NULL NULL NULL NULL 5 100.00 Start temporary +1 PRIMARY t1 ref a a 5 test.t0.a 1 100.00 Using join buffer 1 PRIMARY t2 eq_ref PRIMARY PRIMARY 4 test.t0.a 1 100.00 Using where; End temporary; Using join buffer Warnings: Note 1276 Field or reference 'test.t0.a' of SELECT #2 was resolved in SELECT #1 Note 1003 select `test`.`t0`.`a` AS `a` from `test`.`t2` semi join (`test`.`t1`) join `test`.`t0` where ((`test`.`t2`.`b` = `test`.`t1`.`b`) and (`test`.`t1`.`a` = `test`.`t0`.`a`) and (`test`.`t2`.`a` = `test`.`t0`.`a`)) update t1 set a=3, b=11 where a=4; update t2 set b=11 where a=3; - +# Not anymore: # The following query gives wrong result due to Bug#49129 select * from t0 where t0.a in (select t1.a from t1, t2 where t2.a=t0.a and t1.b=t2.b); a 0 +1 +2 +3 drop table t0, t1, t2; CREATE TABLE t1 ( id int(11) NOT NULL, === modified file 'mysql-test/r/subselect_sj_jcl6.result' --- a/mysql-test/r/subselect_sj_jcl6.result 2010-03-05 18:54:48 +0000 +++ b/mysql-test/r/subselect_sj_jcl6.result 2010-03-07 15:41:45 +0000 @@ -875,6 +875,37 @@ DROP TABLE t1, t2, t3; DROP VIEW v2, v3; # End of Bug#49198 +# +# BUG#49129: Wrong result with IN-subquery with join_cache_level=6 and firstmatch=off +# +CREATE TABLE t0 (a INT); +INSERT INTO t0 VALUES (0),(1),(2),(3),(4); +CREATE TABLE t1 (a INT, b INT, KEY(a)); +INSERT INTO t1 SELECT a, a from t0; +CREATE TABLE t2 (a INT, b INT, PRIMARY KEY(a)); +INSERT INTO t2 SELECT * FROM t1; +UPDATE t1 SET a=3, b=11 WHERE a=4; +UPDATE t2 SET b=11 WHERE a=3; +set @save_optimizer_switch=@@optimizer_switch; +set optimizer_switch='firstmatch=off'; +The following should use a join order of t0,t1,t2, with DuplicateElimination: +explain +SELECT * FROM t0 WHERE t0.a IN +(SELECT t1.a FROM t1, t2 WHERE t2.a=t0.a AND t1.b=t2.b); +id select_type table type possible_keys key key_len ref rows Extra +1 PRIMARY t0 ALL NULL NULL NULL NULL 5 Start temporary +1 PRIMARY t1 ref a a 5 test.t0.a 1 Using join buffer +1 PRIMARY t2 eq_ref PRIMARY PRIMARY 4 test.t0.a 1 Using where; End temporary; Using join buffer +SELECT * FROM t0 WHERE t0.a IN +(SELECT t1.a FROM t1, t2 WHERE t2.a=t0.a AND t1.b=t2.b); +a +0 +1 +2 +3 +set optimizer_switch=@save_optimizer_switch; +drop table t0, t1, t2; +# End set join_cache_level=default; show variables like 'join_cache_level'; Variable_name Value === modified file 'mysql-test/t/subselect_sj2.test' --- a/mysql-test/t/subselect_sj2.test 2010-01-17 14:51:10 +0000 +++ b/mysql-test/t/subselect_sj2.test 2010-03-07 15:41:45 +0000 @@ -583,7 +583,7 @@ if (`select @@join_cache_level=6`) { - --echo + --echo # Not anymore: --echo # The following query gives wrong result due to Bug#49129 } select * from t0 where t0.a in === modified file 'mysql-test/t/subselect_sj_jcl6.test' --- a/mysql-test/t/subselect_sj_jcl6.test 2010-01-17 14:51:10 +0000 +++ b/mysql-test/t/subselect_sj_jcl6.test 2010-03-07 15:41:45 +0000 @@ -7,5 +7,33 @@ --source t/subselect_sj.test +--echo # +--echo # BUG#49129: Wrong result with IN-subquery with join_cache_level=6 and firstmatch=off +--echo # +CREATE TABLE t0 (a INT); +INSERT INTO t0 VALUES (0),(1),(2),(3),(4); +CREATE TABLE t1 (a INT, b INT, KEY(a)); +INSERT INTO t1 SELECT a, a from t0; +CREATE TABLE t2 (a INT, b INT, PRIMARY KEY(a)); +INSERT INTO t2 SELECT * FROM t1; +UPDATE t1 SET a=3, b=11 WHERE a=4; +UPDATE t2 SET b=11 WHERE a=3; + +set @save_optimizer_switch=@@optimizer_switch; +set optimizer_switch='firstmatch=off'; + +--echo The following should use a join order of t0,t1,t2, with DuplicateElimination: +explain +SELECT * FROM t0 WHERE t0.a IN + (SELECT t1.a FROM t1, t2 WHERE t2.a=t0.a AND t1.b=t2.b); + +SELECT * FROM t0 WHERE t0.a IN + (SELECT t1.a FROM t1, t2 WHERE t2.a=t0.a AND t1.b=t2.b); + +set optimizer_switch=@save_optimizer_switch; +drop table t0, t1, t2; + +--echo # End + set join_cache_level=default; show variables like 'join_cache_level'; === modified file 'sql/opt_subselect.cc' --- a/sql/opt_subselect.cc 2010-02-25 08:09:10 +0000 +++ b/sql/opt_subselect.cc 2010-03-07 15:41:45 +0000 @@ -3035,10 +3035,24 @@ forwards, but do not destroy other duplicate elimination methods. */ uint first_table= i; + uint join_cache_level= join->thd->variables.join_cache_level; for (uint j= i; j < i + pos->n_sj_tables; j++) { - if (join->best_positions[j].use_join_buffer && j <= no_jbuf_after) + /* + When we'll properly take join buffering into account during + join optimization, the below check should be changed to + "if (join->best_positions[j].use_join_buffer && + j <= no_jbuf_after)". + For now, use a rough criteria: + */ + JOIN_TAB *js_tab=join->join_tab + j; + if (j != join->const_tables && js_tab->use_quick != 2 && + j <= no_jbuf_after && + ((js_tab->type == JT_ALL && join_cache_level != 0) || + (join_cache_level > 4 && (tab->type == JT_REF || + tab->type == JT_EQ_REF)))) { + /* Looks like we'll be using join buffer */ first_table= join->const_tables; break; } @@ -3116,7 +3130,12 @@ JOIN_TAB *j, *jump_to= tab-1; for (j= tab; j != tab + pos->n_sj_tables; j++) { - if (!tab->emb_sj_nest) + /* + NOTE: this loop probably doesn't do the right thing for the case + where FirstMatch's duplicate-generating range is interleaved with + "unrelated" tables (as specified in WL#3750, section 2.2). + */ + if (!j->emb_sj_nest) jump_to= tab; else { === modified file 'sql/sql_join_cache.cc' --- a/sql/sql_join_cache.cc 2010-03-06 19:14:55 +0000 +++ b/sql/sql_join_cache.cc 2010-03-07 15:41:45 +0000 @@ -608,7 +608,12 @@ copy_end= cache->field_descr+cache->fields; for (copy= cache->field_descr+cache->flag_fields; copy < copy_end; copy++) { - if (copy->field->table == tab->table && + /* + (1) - when we store rowids for DuplicateWeedout, they have + copy->field==NULL + */ + if (copy->field && // (1) + copy->field->table == tab->table && bitmap_is_set(key_read_set, copy->field->field_index)) { *copy_ptr++= copy; === modified file 'sql/sql_select.cc' --- a/sql/sql_select.cc 2010-03-05 18:54:48 +0000 +++ b/sql/sql_select.cc 2010-03-07 15:41:45 +0000 @@ -5632,7 +5632,11 @@ uint blob_length=(uint) (join_tab->table->file->stats.mean_rec_length- (join_tab->table->s->reclength-rec_length)); rec_length+=(uint) max(4,blob_length); - } + } + /* + psergey-todo: why we don't count here rowid that we might need to store + when using DuplicateElimination? + */ join_tab->used_fields=fields; join_tab->used_fieldlength=rec_length; join_tab->used_blobs=blobs;

1 0

[Maria-developers] bzr commit into Mariadb 5.2, with Maria 2.0:maria/5.2 branch (igor:2772) Bug#51092
by Igor Babaev 06 Mar '10

06 Mar '10

#At lp:maria/5.2 based on revid:igor@askmonty.org-20100305185448-satt36ksg5xxkwji 2772 Igor Babaev 2010-03-06 Fixed bug #51092. The function JOIN_CACHE::read_all_record_fields could return 0 for an incremental join cache in two cases: 1. there were no more records in the associated join buffer 2. there was no table fields stored in the join buffer. As a result the function JOIN_CACHE::get_record() could return prematurely and did not read all needed fields from join buffers into the record buffer. Now the function JOIN_CACHE::read_all_record_fields returns -1 if there are no more records in the associated join buffer. modified: mysql-test/r/join_cache.result mysql-test/t/join_cache.test sql/sql_join_cache.cc === modified file 'mysql-test/r/join_cache.result' --- a/mysql-test/r/join_cache.result 2010-02-11 21:59:32 +0000 +++ b/mysql-test/r/join_cache.result 2010-03-06 19:14:55 +0000 @@ -4142,3 +4142,46 @@ c1 c2 c1 c2 LENGTH(t2.c1) LENGTH(t2.c2) 2 2 tt uu 2 2 set join_cache_level=default; DROP TABLE t1,t2; +# +# Bug #51092: linked join buffer is used for a 3-way cross join query +# that selects only records of the first table +# +create table t1 (a int, b int); +insert into t1 values (1,1),(2,2); +create table t2 (a int, b int); +insert into t2 values (1,1),(2,2); +create table t3 (a int, b int); +insert into t3 values (1,1),(2,2); +explain select t1.* from t1,t2,t3; +id select_type table type possible_keys key key_len ref rows Extra +1 SIMPLE t1 ALL NULL NULL NULL NULL 2 +1 SIMPLE t2 ALL NULL NULL NULL NULL 2 Using join buffer +1 SIMPLE t3 ALL NULL NULL NULL NULL 2 Using join buffer +select t1.* from t1,t2,t3; +a b +1 1 +2 2 +1 1 +2 2 +1 1 +2 2 +1 1 +2 2 +set join_cache_level=2; +explain select t1.* from t1,t2,t3; +id select_type table type possible_keys key key_len ref rows Extra +1 SIMPLE t1 ALL NULL NULL NULL NULL 2 +1 SIMPLE t2 ALL NULL NULL NULL NULL 2 Using join buffer +1 SIMPLE t3 ALL NULL NULL NULL NULL 2 Using join buffer +select t1.* from t1,t2,t3; +a b +1 1 +2 2 +1 1 +2 2 +1 1 +2 2 +1 1 +2 2 +set join_cache_level=default; +drop table t1,t2,t3; === modified file 'mysql-test/t/join_cache.test' --- a/mysql-test/t/join_cache.test 2009-12-21 02:26:15 +0000 +++ b/mysql-test/t/join_cache.test 2010-03-06 19:14:55 +0000 @@ -1823,3 +1823,27 @@ SELECT t1.*, t2.*, LENGTH(t2.c1), LENGTH set join_cache_level=default; DROP TABLE t1,t2; + +--echo # +--echo # Bug #51092: linked join buffer is used for a 3-way cross join query +--echo # that selects only records of the first table +--echo # + +create table t1 (a int, b int); +insert into t1 values (1,1),(2,2); +create table t2 (a int, b int); +insert into t2 values (1,1),(2,2); +create table t3 (a int, b int); +insert into t3 values (1,1),(2,2); + +explain select t1.* from t1,t2,t3; +select t1.* from t1,t2,t3; + +set join_cache_level=2; + +explain select t1.* from t1,t2,t3; +select t1.* from t1,t2,t3; + +set join_cache_level=default; + +drop table t1,t2,t3; === modified file 'sql/sql_join_cache.cc' --- a/sql/sql_join_cache.cc 2010-03-05 18:54:48 +0000 +++ b/sql/sql_join_cache.cc 2010-03-06 19:14:55 +0000 @@ -31,6 +31,8 @@ #include "sql_select.h" #include "opt_subselect.h" +#define NO_MORE_RECORDS_IN_BUFFER (uint)(-1) + /***************************************************************************** * Join cache module @@ -1237,7 +1239,7 @@ bool JOIN_CACHE::get_record() prev_rec_ptr= prev_cache->get_rec_ref(pos); } curr_rec_pos= pos; - if (!(res= read_all_record_fields() == 0)) + if (!(res= read_all_record_fields() == NO_MORE_RECORDS_IN_BUFFER)) { pos+= referenced_fields*size_of_fld_ofs; if (prev_cache) @@ -1326,7 +1328,8 @@ bool JOIN_CACHE::get_match_flag_by_pos(u read data. RETURN - length of the data read from the join buffer + (-1) - if there is no more records in the join buffer + length of the data read from the join buffer - otherwise */ uint JOIN_CACHE::read_all_record_fields() @@ -1334,7 +1337,7 @@ uint JOIN_CACHE::read_all_record_fields( uchar *init_pos= pos; if (pos > last_rec_pos || !records) - return 0; + return NO_MORE_RECORDS_IN_BUFFER; /* First match flag, read null bitmaps and null_row flag for each table */ read_flag_fields();

1 0