mariadb/storage/innobase/include/que0que.h

511 lines
17 KiB
C
Raw Normal View History

2001-02-17 14:19:19 +02:00
/******************************************************
Query graph
(c) 1996 Innobase Oy
Created 5/27/1996 Heikki Tuuri
*******************************************************/
#ifndef que0que_h
#define que0que_h
#include "univ.i"
#include "data0data.h"
#include "dict0types.h"
#include "trx0trx.h"
#include "srv0srv.h"
#include "usr0types.h"
#include "que0types.h"
#include "row0types.h"
#include "pars0types.h"
/* If the following flag is set TRUE, the module will print trace info
of SQL execution in the UNIV_SQL_DEBUG version */
extern ibool que_trace_on;
/***************************************************************************
Adds a query graph to the session's list of graphs. */
void
que_graph_publish(
/*==============*/
que_t* graph, /* in: graph */
sess_t* sess); /* in: session */
/***************************************************************************
Creates a query graph fork node. */
que_fork_t*
que_fork_create(
/*============*/
/* out, own: fork node */
que_t* graph, /* in: graph, if NULL then this
fork node is assumed to be the
graph root */
que_node_t* parent, /* in: parent node */
ulint fork_type, /* in: fork type */
mem_heap_t* heap); /* in: memory heap where created */
/***************************************************************************
Gets the first thr in a fork. */
UNIV_INLINE
que_thr_t*
que_fork_get_first_thr(
/*===================*/
Applied innodb-5.1-ss269 snapshot. Fixed BUGS: #3300: "UPDATE statement with no index column in where condition locks all rows" Implement semi-consistent read to reduce lock conflicts at the cost of breaking serializability. ha_innobase::unlock_row(): reset the "did semi consistent read" flag ha_innobase::was_semi_consistent_read(), ha_innobase::try_semi_consistent_read(): new methods row_prebuilt_t, row_create_prebuilt(): add field row_read_type for keeping track of semi-consistent reads row_vers_build_for_semi_consistent_read(), row_sel_build_committed_vers_for_mysql(): new functions row_search_for_mysql(): implement semi-consistent reads #9802: "Foreign key checks disallow alter table". Added test cases. #12456: "Cursor shows incorrect data - DML does not affect, probably caching" This patch implements a high-granularity read view to be used with cursors. In this high-granularity consistent read view modifications done by the creating transaction after the cursor is created or future transactions are not visible. But those modifications that transaction did before the cursor was created are visible. #12701: "Support >4GB buffer pool and log files on 64-bit Windows" Do not call os_file_create_tmpfile() at runtime. Instead, create all tempfiles at startup and guard access to them with mutexes. #13778: "If FOREIGN_KEY_CHECKS=0, one can create inconsistent FOREIGN KEYs". When FOREIGN_KEY_CHECKS=0 we still need to check that datatypes between foreign key references are compatible. #14189: "VARBINARY and BINARY variables: trailing space ignored with InnoDB" innobase_init(): Assert that DATA_MYSQL_BINARY_CHARSET_COLL == my_charset_bin.number. dtype_get_pad_char(): Do not pad VARBINARY or BINARY columns. row_ins_cascade_calc_update_vec(): Refuse ON UPDATE CASCADE when trying to change the length of a VARBINARY column that refers to or is referenced by a BINARY column. BINARY columns are no longer padded on comparison, and thus they cannot be padded on storage either. #14747: "Race condition can cause btr_search_drop_page_hash_index() to crash" Note that buf_block_t::index should be protected by btr_search_latch or an s-latch or x-latch on the index page. btr_search_drop_page_hash_index(): Read block->index while holding btr_search_latch and use the cached value in the loop. Remove some redundant assertions. #15108: "mysqld crashes when innodb_log_file_size is set > 4G" #15308: "Problem of Order with Enum Column in Primary Key" #15550: "mysqld crashes in printing a FOREIGN KEY error in InnoDB" row_ins_foreign_report_add_err(): When printing the parent record, use the index in the parent table rather than the index in the child table. #15653: "Slow inserts to InnoDB if many thousands of .ibd files" Keep track on unflushed modifications to file spaces. When there are tens of thousands of file spaces, flushing all files in fil_flush_file_spaces() would be very slow. fil_flush_file_spaces(): Only flush unflushed file spaces. fil_space_t, fil_system_t: Add a list of unflushed spaces. #15991: "innodb-file-per-table + symlink database + rename = cr" os_file_handle_error(): Map the error codes EXDEV, ENOTDIR, and EISDIR to the new code OS_FILE_PATH_ERROR. Treat this code as OS_FILE_PATH_ERROR. This fixes the crash on RENAME TABLE when the .ibd file is a symbolic link to a different file system. #16157: "InnoDB crashes when main location settings are empty" This patch is from Heikki. #16298: "InnoDB segfaults in INSERTs in upgrade of 4.0 -> 5.0 tables with VARCHAR BINARY" dict_load_columns(): Set the charset-collation code DATA_MYSQL_BINARY_CHARSET_COLL for those binary string columns that lack a charset-collation code, i.e., the tables were created with an older version of MySQL/InnoDB than 4.1.2. #16229: "MySQL/InnoDB uses full explicit table locks in trigger processing" Take a InnoDB table lock only if user has explicitly requested a table lock. Added some additional comments to store_lock() and external_lock(). #16387: "InnoDB crash when dropping a foreign key <table>_ibfk_0" Do not mistake TABLENAME_ibfk_0 for auto-generated id. dict_table_get_highest_foreign_id(): Ignore foreign constraint identifiers starting with the pattern TABLENAME_ibfk_0. #16582: "InnoDB: Error in an adaptive hash index pointer to page" Account for a race condition when dropping the adaptive hash index for a B-tree page. btr_search_drop_page_hash_index(): Retry the operation if a hash index with different parameters was built meanwhile. Add diagnostics for the case that hash node pointers to the page remain. btr_search_info_update_hash(), btr_search_info_update_slow(): Document the parameter "info" as in/out. #16814: "SHOW INNODB STATUS format error in LATEST FOREIGN KEY ERROR section" Add a missing newline to the LAST FOREIGN KEY ERROR section in SHOW INNODB STATUS output. dict_foreign_error_report(): Always print a newline after invoking dict_print_info_on_foreign_key_in_create_format(). #16827: "Better InnoDB error message if ibdata files omitted from my.cnf" #17126: "CHECK TABLE on InnoDB causes a short hang during check of adaptive hash" CHECK TABLE blocking other queries, by releasing the btr_search_latch periodically during the adaptive hash table validation. #17405: "Valgrind: conditional jump or move depends on unititialised values" buf_block_init(): Reset magic_n, buf_fix_count and io_fix to avoid testing uninitialized variables.
2006-03-10 19:22:21 +03:00
que_fork_t* fork); /* in: query fork */
2001-02-17 14:19:19 +02:00
/***************************************************************************
Gets the child node of the first thr in a fork. */
UNIV_INLINE
que_node_t*
que_fork_get_child(
/*===============*/
que_fork_t* fork); /* in: query fork */
/***************************************************************************
Sets the parent of a graph node. */
UNIV_INLINE
void
que_node_set_parent(
/*================*/
que_node_t* node, /* in: graph node */
que_node_t* parent);/* in: parent */
/***************************************************************************
Creates a query graph thread node. */
que_thr_t*
que_thr_create(
/*===========*/
/* out, own: query thread node */
que_fork_t* parent, /* in: parent node, i.e., a fork node */
mem_heap_t* heap); /* in: memory heap where created */
/**************************************************************************
Checks if the query graph is in a state where it should be freed, and
frees it in that case. If the session is in a state where it should be
closed, also this is done. */
ibool
que_graph_try_free(
/*===============*/
/* out: TRUE if freed */
que_t* graph); /* in: query graph */
/**************************************************************************
Frees a query graph, but not the heap where it was created. Does not free
explicit cursor declarations, they are freed in que_graph_free. */
void
que_graph_free_recursive(
/*=====================*/
que_node_t* node); /* in: query graph node */
/**************************************************************************
Frees a query graph. */
void
que_graph_free(
/*===========*/
que_t* graph); /* in: query graph; we assume that the memory
heap where this graph was created is private
to this graph: if not, then use
que_graph_free_recursive and free the heap
afterwards! */
/**************************************************************************
Stops a query thread if graph or trx is in a state requiring it. The
conditions are tested in the order (1) graph, (2) trx. The kernel mutex has
to be reserved. */
ibool
que_thr_stop(
/*=========*/
/* out: TRUE if stopped */
que_thr_t* thr); /* in: query thread */
/**************************************************************************
Moves a thread from another state to the QUE_THR_RUNNING state. Increments
the n_active_thrs counters of the query graph and transaction. */
2001-02-17 14:19:19 +02:00
void
que_thr_move_to_run_state_for_mysql(
/*================================*/
que_thr_t* thr, /* in: an query thread */
trx_t* trx); /* in: transaction */
/**************************************************************************
A patch for MySQL used to 'stop' a dummy query thread used in MySQL
select, when there is no error or lock wait. */
2001-02-17 14:19:19 +02:00
void
que_thr_stop_for_mysql_no_error(
/*============================*/
que_thr_t* thr, /* in: query thread */
trx_t* trx); /* in: transaction */
/**************************************************************************
A patch for MySQL used to 'stop' a dummy query thread used in MySQL. The
query thread is stopped and made inactive, except in the case where
it was put to the lock wait state in lock0lock.c, but the lock has already
been granted or the transaction chosen as a victim in deadlock resolution. */
2001-02-17 14:19:19 +02:00
void
que_thr_stop_for_mysql(
/*===================*/
que_thr_t* thr); /* in: query thread */
/**************************************************************************
Run a query thread. Handles lock waits. */
2001-02-17 14:19:19 +02:00
void
que_run_threads(
/*============*/
que_thr_t* thr); /* in: query thread */
2001-02-17 14:19:19 +02:00
/**************************************************************************
After signal handling is finished, returns control to a query graph error
handling routine. (Currently, just returns the control to the root of the
graph so that the graph can communicate an error message to the client.) */
void
que_fork_error_handle(
/*==================*/
trx_t* trx, /* in: trx */
que_t* fork); /* in: query graph which was run before signal
handling started, NULL not allowed */
/**************************************************************************
2001-02-17 14:19:19 +02:00
Moves a suspended query thread to the QUE_THR_RUNNING state and releases
a single worker thread to execute it. This function should be used to end
the wait state of a query thread waiting for a lock or a stored procedure
completion. */
void
2001-02-17 14:19:19 +02:00
que_thr_end_wait(
/*=============*/
que_thr_t* thr, /* in: query thread in the
2001-02-17 14:19:19 +02:00
QUE_THR_LOCK_WAIT,
or QUE_THR_PROCEDURE_WAIT, or
QUE_THR_SIG_REPLY_WAIT state */
que_thr_t** next_thr); /* in/out: next query thread to run;
if the value which is passed in is
a pointer to a NULL pointer, then the
calling function can start running
a new query thread */
2001-02-17 14:19:19 +02:00
/**************************************************************************
Same as que_thr_end_wait, but no parameter next_thr available. */
void
que_thr_end_wait_no_next_thr(
/*=========================*/
que_thr_t* thr); /* in: query thread in the
QUE_THR_LOCK_WAIT,
or QUE_THR_PROCEDURE_WAIT, or
QUE_THR_SIG_REPLY_WAIT state */
/**************************************************************************
Starts execution of a command in a query fork. Picks a query thread which
is not in the QUE_THR_RUNNING state and moves it to that state. If none
can be chosen, a situation which may arise in parallelized fetches, NULL
is returned. */
que_thr_t*
que_fork_start_command(
/*===================*/
/* out: a query thread of the graph moved to
QUE_THR_RUNNING state, or NULL; the query
thread should be executed by que_run_threads
by the caller */
Applied innodb-5.1-ss269 snapshot. Fixed BUGS: #3300: "UPDATE statement with no index column in where condition locks all rows" Implement semi-consistent read to reduce lock conflicts at the cost of breaking serializability. ha_innobase::unlock_row(): reset the "did semi consistent read" flag ha_innobase::was_semi_consistent_read(), ha_innobase::try_semi_consistent_read(): new methods row_prebuilt_t, row_create_prebuilt(): add field row_read_type for keeping track of semi-consistent reads row_vers_build_for_semi_consistent_read(), row_sel_build_committed_vers_for_mysql(): new functions row_search_for_mysql(): implement semi-consistent reads #9802: "Foreign key checks disallow alter table". Added test cases. #12456: "Cursor shows incorrect data - DML does not affect, probably caching" This patch implements a high-granularity read view to be used with cursors. In this high-granularity consistent read view modifications done by the creating transaction after the cursor is created or future transactions are not visible. But those modifications that transaction did before the cursor was created are visible. #12701: "Support >4GB buffer pool and log files on 64-bit Windows" Do not call os_file_create_tmpfile() at runtime. Instead, create all tempfiles at startup and guard access to them with mutexes. #13778: "If FOREIGN_KEY_CHECKS=0, one can create inconsistent FOREIGN KEYs". When FOREIGN_KEY_CHECKS=0 we still need to check that datatypes between foreign key references are compatible. #14189: "VARBINARY and BINARY variables: trailing space ignored with InnoDB" innobase_init(): Assert that DATA_MYSQL_BINARY_CHARSET_COLL == my_charset_bin.number. dtype_get_pad_char(): Do not pad VARBINARY or BINARY columns. row_ins_cascade_calc_update_vec(): Refuse ON UPDATE CASCADE when trying to change the length of a VARBINARY column that refers to or is referenced by a BINARY column. BINARY columns are no longer padded on comparison, and thus they cannot be padded on storage either. #14747: "Race condition can cause btr_search_drop_page_hash_index() to crash" Note that buf_block_t::index should be protected by btr_search_latch or an s-latch or x-latch on the index page. btr_search_drop_page_hash_index(): Read block->index while holding btr_search_latch and use the cached value in the loop. Remove some redundant assertions. #15108: "mysqld crashes when innodb_log_file_size is set > 4G" #15308: "Problem of Order with Enum Column in Primary Key" #15550: "mysqld crashes in printing a FOREIGN KEY error in InnoDB" row_ins_foreign_report_add_err(): When printing the parent record, use the index in the parent table rather than the index in the child table. #15653: "Slow inserts to InnoDB if many thousands of .ibd files" Keep track on unflushed modifications to file spaces. When there are tens of thousands of file spaces, flushing all files in fil_flush_file_spaces() would be very slow. fil_flush_file_spaces(): Only flush unflushed file spaces. fil_space_t, fil_system_t: Add a list of unflushed spaces. #15991: "innodb-file-per-table + symlink database + rename = cr" os_file_handle_error(): Map the error codes EXDEV, ENOTDIR, and EISDIR to the new code OS_FILE_PATH_ERROR. Treat this code as OS_FILE_PATH_ERROR. This fixes the crash on RENAME TABLE when the .ibd file is a symbolic link to a different file system. #16157: "InnoDB crashes when main location settings are empty" This patch is from Heikki. #16298: "InnoDB segfaults in INSERTs in upgrade of 4.0 -> 5.0 tables with VARCHAR BINARY" dict_load_columns(): Set the charset-collation code DATA_MYSQL_BINARY_CHARSET_COLL for those binary string columns that lack a charset-collation code, i.e., the tables were created with an older version of MySQL/InnoDB than 4.1.2. #16229: "MySQL/InnoDB uses full explicit table locks in trigger processing" Take a InnoDB table lock only if user has explicitly requested a table lock. Added some additional comments to store_lock() and external_lock(). #16387: "InnoDB crash when dropping a foreign key <table>_ibfk_0" Do not mistake TABLENAME_ibfk_0 for auto-generated id. dict_table_get_highest_foreign_id(): Ignore foreign constraint identifiers starting with the pattern TABLENAME_ibfk_0. #16582: "InnoDB: Error in an adaptive hash index pointer to page" Account for a race condition when dropping the adaptive hash index for a B-tree page. btr_search_drop_page_hash_index(): Retry the operation if a hash index with different parameters was built meanwhile. Add diagnostics for the case that hash node pointers to the page remain. btr_search_info_update_hash(), btr_search_info_update_slow(): Document the parameter "info" as in/out. #16814: "SHOW INNODB STATUS format error in LATEST FOREIGN KEY ERROR section" Add a missing newline to the LAST FOREIGN KEY ERROR section in SHOW INNODB STATUS output. dict_foreign_error_report(): Always print a newline after invoking dict_print_info_on_foreign_key_in_create_format(). #16827: "Better InnoDB error message if ibdata files omitted from my.cnf" #17126: "CHECK TABLE on InnoDB causes a short hang during check of adaptive hash" CHECK TABLE blocking other queries, by releasing the btr_search_latch periodically during the adaptive hash table validation. #17405: "Valgrind: conditional jump or move depends on unititialised values" buf_block_init(): Reset magic_n, buf_fix_count and io_fix to avoid testing uninitialized variables.
2006-03-10 19:22:21 +03:00
que_fork_t* fork); /* in: a query fork */
2001-02-17 14:19:19 +02:00
/***************************************************************************
Gets the trx of a query thread. */
UNIV_INLINE
trx_t*
thr_get_trx(
/*========*/
que_thr_t* thr); /* in: query thread */
/***************************************************************************
Gets the type of a graph node. */
UNIV_INLINE
ulint
que_node_get_type(
/*==============*/
que_node_t* node); /* in: graph node */
/***************************************************************************
Gets pointer to the value data type field of a graph node. */
UNIV_INLINE
dtype_t*
que_node_get_data_type(
/*===================*/
que_node_t* node); /* in: graph node */
/***************************************************************************
Gets pointer to the value dfield of a graph node. */
UNIV_INLINE
dfield_t*
que_node_get_val(
/*=============*/
que_node_t* node); /* in: graph node */
/***************************************************************************
Gets the value buffer size of a graph node. */
UNIV_INLINE
ulint
que_node_get_val_buf_size(
/*======================*/
/* out: val buffer size, not defined if
val.data == NULL in node */
que_node_t* node); /* in: graph node */
/***************************************************************************
Sets the value buffer size of a graph node. */
UNIV_INLINE
void
que_node_set_val_buf_size(
/*======================*/
que_node_t* node, /* in: graph node */
ulint size); /* in: size */
/*************************************************************************
Gets the next list node in a list of query graph nodes. */
UNIV_INLINE
que_node_t*
que_node_get_next(
/*==============*/
que_node_t* node); /* in: node in a list */
/*************************************************************************
Gets the parent node of a query graph node. */
UNIV_INLINE
que_node_t*
que_node_get_parent(
/*================*/
/* out: parent node or NULL */
que_node_t* node); /* in: node */
Applied innodb-5.1-ss475 snapshot. * Fix BUG#15650: "DELETE with LEFT JOIN crashes server with innodb_locks_unsafe_for binlog" * Fix BUG#17134: "Partitions: uncommitted changes are visible" * Fix BUG#17992: "Partitions: InnoDB, somehow rotten table after UPDATE" row0ins.c: MySQL's partitioned table code does not set preduilt->sql_stat_start right if it does an insert in the same statement after doing a search first in the same partition table. We now write trx id always to the buffer, not just when flag sql_stat_start is on. This will waste CPU time very sightly. * Fix BUG#18077: "InnoDB uses full explicit table locks in stored FUNCTION" * Fix BUG#18238: "When locks exhaust the buffer pool, InnoDB does not roll back the trx" * Fix BUG#18252" "Disk space leak in updates of InnoDB BLOB rows in 5.0 and 5.1" * Fix BUG#18283: "When InnoDB returns error 'lock table full', MySQL can write to binlog too much" * Fix BUG#18350: "Use consistent read in CREATE ... SELECT ... if innodb_locks_unsafe_for_binlog" * Fix BUG#18384: "InnoDB memory leak on duplicate key errors in 5.0 if row has many columns" * Fix BUG#18934: "InnoDB crashes when table uses column names like DB_ROW_ID" Refuse tables that use reserved column names. * InnoDB's SQL parser: - Add support for UNSIGNED types, EXIT keyword, quoted identifiers, user-function callbacks for processing results of FETCH statements, bound literals, DATA_VARCHAR for bound literals. - Allow bound literals of type non-INTEGER to be of length 0. - Add make_flex.sh and update lexer/parser generation documentation. - Add comment clarifying the difference between 'alias' and 'indirection' fields in sym_node_t. - Remove never reached duplicate code in pars_set_dfield_type(). - Rewrite pars_info datatypes and APIs, add a few helper functions. - Since the functions definitions in pars_info_t are accessed after pars_sql() returns in the query graph execution stage, we can't free pars_info_t in pars_sql(). Instead, make pars_sql() transfer ownership of pars_info_t to the created query graph, and make que_graph_free() free it if needed. - Allow access to system columns like DB_ROW_ID. * Use bound literals in row_truncate_table_for_mysql, row_drop_table_for_mysql, row_discard_tablespace_for_mysql, and row_rename_table_for_mysql. * Setting an isolation level of the transaction to read committed weakens the locks for this session similarly like the option innodb_locks_unsafe_for binlog. This patch removes alnost all gap locking (used in next-key locking) and makes MySQL to release the row locks on the rows which does not belong to result set. Additionally, nonlocking selects on INSERT INTO SELECT, UPDATE ... (SELECT ...), and CREATE ... SELECT ... use a nonlocking consistent read. If a binlog is used, then binlog format should be set to row based binloging to make the execution of the complex SQL statements. * Disable the statistic variables btr_search_n_hash_fail and n_hash_succ, n_hash_fail, n_patt_succ, and n_searches of btr_search_t in builds without #ifdef UNIV_SEARCH_PERF_STAT. * Make innodb.test faster. Group all consistent read test cases to a one test case and wait their lock timeout after all have been send to the server. Decrease amount of rows inserted in a certain test - this has no effect on the effectiveness of the test and reduces the running time by ~10 sec. Remove temporary work-arounds from innodb.result now that ALTER TABLE DROP FOREIGN KEY works once again. * Make innodb_unsafe_binlog.test faster. Grout all consistent read test cases to a one test case amd wait their lock timeout after all have been sent to the server. Remove unnecessary option --loose_innodb_lock_wait_timeout. * Print dictionary memory size in SHOW INNODB STATUS. * Fix memory leaks in row_create_table_for_mysql() in rare corner cases. * Remove code related to clustered tables. They were never implemented, and the implementation would be challenging with ROW_FORMAT=COMPACT. Remove the table types DICT_TABLE_CLUSTER_MEMBER and DICT_TABLE_CLUSTER and all related tests and functions. dict_table_t: Remove mix_id, mix_len, mix_id_len, mix_id_buf, and cluster_name. plan_t: Remove mixed_index. dict_create_sys_tables_tuple(): Set MIX_ID=0, MIX_LEN=0, CLUSTER_NAME=NULL when inserting into SYS_TABLES. dict_tree_check_search_tuple(): Enclose in #ifdef UNIV_DEBUG. * Move calling of thr_local_free() from trx_free_for_mysql() to innobase_close_connection().
2006-04-23 12:48:31 +04:00
/********************************************************************
Get the first containing loop node (e.g. while_node_t or for_node_t) for the
given node, or NULL if the node is not within a loop. */
que_node_t*
que_node_get_containing_loop_node(
/*==============================*/
/* out: containing loop node, or NULL. */
que_node_t* node); /* in: node */
2001-02-17 14:19:19 +02:00
/*************************************************************************
Catenates a query graph node to a list of them, possible empty list. */
UNIV_INLINE
que_node_t*
que_node_list_add_last(
/*===================*/
/* out: one-way list of nodes */
que_node_t* node_list, /* in: node list, or NULL */
que_node_t* node); /* in: node */
/*************************************************************************
Gets a query graph node list length. */
UNIV_INLINE
ulint
que_node_list_get_len(
/*==================*/
/* out: length, for NULL list 0 */
que_node_t* node_list); /* in: node list, or NULL */
/**************************************************************************
Checks if graph, trx, or session is in a state where the query thread should
be stopped. */
UNIV_INLINE
ibool
que_thr_peek_stop(
/*==============*/
/* out: TRUE if should be stopped; NOTE that
if the peek is made without reserving the
kernel mutex, then another peek with the
mutex reserved is necessary before deciding
the actual stopping */
que_thr_t* thr); /* in: query thread */
/***************************************************************************
Returns TRUE if the query graph is for a SELECT statement. */
UNIV_INLINE
ibool
que_graph_is_select(
/*================*/
/* out: TRUE if a select */
que_t* graph); /* in: graph */
/**************************************************************************
Prints info of an SQL query graph node. */
void
que_node_print_info(
/*================*/
que_node_t* node); /* in: query graph node */
Applied innodb-5.1-ss475 snapshot. * Fix BUG#15650: "DELETE with LEFT JOIN crashes server with innodb_locks_unsafe_for binlog" * Fix BUG#17134: "Partitions: uncommitted changes are visible" * Fix BUG#17992: "Partitions: InnoDB, somehow rotten table after UPDATE" row0ins.c: MySQL's partitioned table code does not set preduilt->sql_stat_start right if it does an insert in the same statement after doing a search first in the same partition table. We now write trx id always to the buffer, not just when flag sql_stat_start is on. This will waste CPU time very sightly. * Fix BUG#18077: "InnoDB uses full explicit table locks in stored FUNCTION" * Fix BUG#18238: "When locks exhaust the buffer pool, InnoDB does not roll back the trx" * Fix BUG#18252" "Disk space leak in updates of InnoDB BLOB rows in 5.0 and 5.1" * Fix BUG#18283: "When InnoDB returns error 'lock table full', MySQL can write to binlog too much" * Fix BUG#18350: "Use consistent read in CREATE ... SELECT ... if innodb_locks_unsafe_for_binlog" * Fix BUG#18384: "InnoDB memory leak on duplicate key errors in 5.0 if row has many columns" * Fix BUG#18934: "InnoDB crashes when table uses column names like DB_ROW_ID" Refuse tables that use reserved column names. * InnoDB's SQL parser: - Add support for UNSIGNED types, EXIT keyword, quoted identifiers, user-function callbacks for processing results of FETCH statements, bound literals, DATA_VARCHAR for bound literals. - Allow bound literals of type non-INTEGER to be of length 0. - Add make_flex.sh and update lexer/parser generation documentation. - Add comment clarifying the difference between 'alias' and 'indirection' fields in sym_node_t. - Remove never reached duplicate code in pars_set_dfield_type(). - Rewrite pars_info datatypes and APIs, add a few helper functions. - Since the functions definitions in pars_info_t are accessed after pars_sql() returns in the query graph execution stage, we can't free pars_info_t in pars_sql(). Instead, make pars_sql() transfer ownership of pars_info_t to the created query graph, and make que_graph_free() free it if needed. - Allow access to system columns like DB_ROW_ID. * Use bound literals in row_truncate_table_for_mysql, row_drop_table_for_mysql, row_discard_tablespace_for_mysql, and row_rename_table_for_mysql. * Setting an isolation level of the transaction to read committed weakens the locks for this session similarly like the option innodb_locks_unsafe_for binlog. This patch removes alnost all gap locking (used in next-key locking) and makes MySQL to release the row locks on the rows which does not belong to result set. Additionally, nonlocking selects on INSERT INTO SELECT, UPDATE ... (SELECT ...), and CREATE ... SELECT ... use a nonlocking consistent read. If a binlog is used, then binlog format should be set to row based binloging to make the execution of the complex SQL statements. * Disable the statistic variables btr_search_n_hash_fail and n_hash_succ, n_hash_fail, n_patt_succ, and n_searches of btr_search_t in builds without #ifdef UNIV_SEARCH_PERF_STAT. * Make innodb.test faster. Group all consistent read test cases to a one test case and wait their lock timeout after all have been send to the server. Decrease amount of rows inserted in a certain test - this has no effect on the effectiveness of the test and reduces the running time by ~10 sec. Remove temporary work-arounds from innodb.result now that ALTER TABLE DROP FOREIGN KEY works once again. * Make innodb_unsafe_binlog.test faster. Grout all consistent read test cases to a one test case amd wait their lock timeout after all have been sent to the server. Remove unnecessary option --loose_innodb_lock_wait_timeout. * Print dictionary memory size in SHOW INNODB STATUS. * Fix memory leaks in row_create_table_for_mysql() in rare corner cases. * Remove code related to clustered tables. They were never implemented, and the implementation would be challenging with ROW_FORMAT=COMPACT. Remove the table types DICT_TABLE_CLUSTER_MEMBER and DICT_TABLE_CLUSTER and all related tests and functions. dict_table_t: Remove mix_id, mix_len, mix_id_len, mix_id_buf, and cluster_name. plan_t: Remove mixed_index. dict_create_sys_tables_tuple(): Set MIX_ID=0, MIX_LEN=0, CLUSTER_NAME=NULL when inserting into SYS_TABLES. dict_tree_check_search_tuple(): Enclose in #ifdef UNIV_DEBUG. * Move calling of thr_local_free() from trx_free_for_mysql() to innobase_close_connection().
2006-04-23 12:48:31 +04:00
/*************************************************************************
Evaluate the given SQL */
Applied innodb-5.1-ss475 snapshot. * Fix BUG#15650: "DELETE with LEFT JOIN crashes server with innodb_locks_unsafe_for binlog" * Fix BUG#17134: "Partitions: uncommitted changes are visible" * Fix BUG#17992: "Partitions: InnoDB, somehow rotten table after UPDATE" row0ins.c: MySQL's partitioned table code does not set preduilt->sql_stat_start right if it does an insert in the same statement after doing a search first in the same partition table. We now write trx id always to the buffer, not just when flag sql_stat_start is on. This will waste CPU time very sightly. * Fix BUG#18077: "InnoDB uses full explicit table locks in stored FUNCTION" * Fix BUG#18238: "When locks exhaust the buffer pool, InnoDB does not roll back the trx" * Fix BUG#18252" "Disk space leak in updates of InnoDB BLOB rows in 5.0 and 5.1" * Fix BUG#18283: "When InnoDB returns error 'lock table full', MySQL can write to binlog too much" * Fix BUG#18350: "Use consistent read in CREATE ... SELECT ... if innodb_locks_unsafe_for_binlog" * Fix BUG#18384: "InnoDB memory leak on duplicate key errors in 5.0 if row has many columns" * Fix BUG#18934: "InnoDB crashes when table uses column names like DB_ROW_ID" Refuse tables that use reserved column names. * InnoDB's SQL parser: - Add support for UNSIGNED types, EXIT keyword, quoted identifiers, user-function callbacks for processing results of FETCH statements, bound literals, DATA_VARCHAR for bound literals. - Allow bound literals of type non-INTEGER to be of length 0. - Add make_flex.sh and update lexer/parser generation documentation. - Add comment clarifying the difference between 'alias' and 'indirection' fields in sym_node_t. - Remove never reached duplicate code in pars_set_dfield_type(). - Rewrite pars_info datatypes and APIs, add a few helper functions. - Since the functions definitions in pars_info_t are accessed after pars_sql() returns in the query graph execution stage, we can't free pars_info_t in pars_sql(). Instead, make pars_sql() transfer ownership of pars_info_t to the created query graph, and make que_graph_free() free it if needed. - Allow access to system columns like DB_ROW_ID. * Use bound literals in row_truncate_table_for_mysql, row_drop_table_for_mysql, row_discard_tablespace_for_mysql, and row_rename_table_for_mysql. * Setting an isolation level of the transaction to read committed weakens the locks for this session similarly like the option innodb_locks_unsafe_for binlog. This patch removes alnost all gap locking (used in next-key locking) and makes MySQL to release the row locks on the rows which does not belong to result set. Additionally, nonlocking selects on INSERT INTO SELECT, UPDATE ... (SELECT ...), and CREATE ... SELECT ... use a nonlocking consistent read. If a binlog is used, then binlog format should be set to row based binloging to make the execution of the complex SQL statements. * Disable the statistic variables btr_search_n_hash_fail and n_hash_succ, n_hash_fail, n_patt_succ, and n_searches of btr_search_t in builds without #ifdef UNIV_SEARCH_PERF_STAT. * Make innodb.test faster. Group all consistent read test cases to a one test case and wait their lock timeout after all have been send to the server. Decrease amount of rows inserted in a certain test - this has no effect on the effectiveness of the test and reduces the running time by ~10 sec. Remove temporary work-arounds from innodb.result now that ALTER TABLE DROP FOREIGN KEY works once again. * Make innodb_unsafe_binlog.test faster. Grout all consistent read test cases to a one test case amd wait their lock timeout after all have been sent to the server. Remove unnecessary option --loose_innodb_lock_wait_timeout. * Print dictionary memory size in SHOW INNODB STATUS. * Fix memory leaks in row_create_table_for_mysql() in rare corner cases. * Remove code related to clustered tables. They were never implemented, and the implementation would be challenging with ROW_FORMAT=COMPACT. Remove the table types DICT_TABLE_CLUSTER_MEMBER and DICT_TABLE_CLUSTER and all related tests and functions. dict_table_t: Remove mix_id, mix_len, mix_id_len, mix_id_buf, and cluster_name. plan_t: Remove mixed_index. dict_create_sys_tables_tuple(): Set MIX_ID=0, MIX_LEN=0, CLUSTER_NAME=NULL when inserting into SYS_TABLES. dict_tree_check_search_tuple(): Enclose in #ifdef UNIV_DEBUG. * Move calling of thr_local_free() from trx_free_for_mysql() to innobase_close_connection().
2006-04-23 12:48:31 +04:00
ulint
que_eval_sql(
/*=========*/
/* out: error code or DB_SUCCESS */
pars_info_t* info, /* in: info struct, or NULL */
const char* sql, /* in: SQL string */
ibool reserve_dict_mutex,
/* in: if TRUE, acquire/release
dict_sys->mutex around call to pars_sql. */
Applied innodb-5.1-ss475 snapshot. * Fix BUG#15650: "DELETE with LEFT JOIN crashes server with innodb_locks_unsafe_for binlog" * Fix BUG#17134: "Partitions: uncommitted changes are visible" * Fix BUG#17992: "Partitions: InnoDB, somehow rotten table after UPDATE" row0ins.c: MySQL's partitioned table code does not set preduilt->sql_stat_start right if it does an insert in the same statement after doing a search first in the same partition table. We now write trx id always to the buffer, not just when flag sql_stat_start is on. This will waste CPU time very sightly. * Fix BUG#18077: "InnoDB uses full explicit table locks in stored FUNCTION" * Fix BUG#18238: "When locks exhaust the buffer pool, InnoDB does not roll back the trx" * Fix BUG#18252" "Disk space leak in updates of InnoDB BLOB rows in 5.0 and 5.1" * Fix BUG#18283: "When InnoDB returns error 'lock table full', MySQL can write to binlog too much" * Fix BUG#18350: "Use consistent read in CREATE ... SELECT ... if innodb_locks_unsafe_for_binlog" * Fix BUG#18384: "InnoDB memory leak on duplicate key errors in 5.0 if row has many columns" * Fix BUG#18934: "InnoDB crashes when table uses column names like DB_ROW_ID" Refuse tables that use reserved column names. * InnoDB's SQL parser: - Add support for UNSIGNED types, EXIT keyword, quoted identifiers, user-function callbacks for processing results of FETCH statements, bound literals, DATA_VARCHAR for bound literals. - Allow bound literals of type non-INTEGER to be of length 0. - Add make_flex.sh and update lexer/parser generation documentation. - Add comment clarifying the difference between 'alias' and 'indirection' fields in sym_node_t. - Remove never reached duplicate code in pars_set_dfield_type(). - Rewrite pars_info datatypes and APIs, add a few helper functions. - Since the functions definitions in pars_info_t are accessed after pars_sql() returns in the query graph execution stage, we can't free pars_info_t in pars_sql(). Instead, make pars_sql() transfer ownership of pars_info_t to the created query graph, and make que_graph_free() free it if needed. - Allow access to system columns like DB_ROW_ID. * Use bound literals in row_truncate_table_for_mysql, row_drop_table_for_mysql, row_discard_tablespace_for_mysql, and row_rename_table_for_mysql. * Setting an isolation level of the transaction to read committed weakens the locks for this session similarly like the option innodb_locks_unsafe_for binlog. This patch removes alnost all gap locking (used in next-key locking) and makes MySQL to release the row locks on the rows which does not belong to result set. Additionally, nonlocking selects on INSERT INTO SELECT, UPDATE ... (SELECT ...), and CREATE ... SELECT ... use a nonlocking consistent read. If a binlog is used, then binlog format should be set to row based binloging to make the execution of the complex SQL statements. * Disable the statistic variables btr_search_n_hash_fail and n_hash_succ, n_hash_fail, n_patt_succ, and n_searches of btr_search_t in builds without #ifdef UNIV_SEARCH_PERF_STAT. * Make innodb.test faster. Group all consistent read test cases to a one test case and wait their lock timeout after all have been send to the server. Decrease amount of rows inserted in a certain test - this has no effect on the effectiveness of the test and reduces the running time by ~10 sec. Remove temporary work-arounds from innodb.result now that ALTER TABLE DROP FOREIGN KEY works once again. * Make innodb_unsafe_binlog.test faster. Grout all consistent read test cases to a one test case amd wait their lock timeout after all have been sent to the server. Remove unnecessary option --loose_innodb_lock_wait_timeout. * Print dictionary memory size in SHOW INNODB STATUS. * Fix memory leaks in row_create_table_for_mysql() in rare corner cases. * Remove code related to clustered tables. They were never implemented, and the implementation would be challenging with ROW_FORMAT=COMPACT. Remove the table types DICT_TABLE_CLUSTER_MEMBER and DICT_TABLE_CLUSTER and all related tests and functions. dict_table_t: Remove mix_id, mix_len, mix_id_len, mix_id_buf, and cluster_name. plan_t: Remove mixed_index. dict_create_sys_tables_tuple(): Set MIX_ID=0, MIX_LEN=0, CLUSTER_NAME=NULL when inserting into SYS_TABLES. dict_tree_check_search_tuple(): Enclose in #ifdef UNIV_DEBUG. * Move calling of thr_local_free() from trx_free_for_mysql() to innobase_close_connection().
2006-04-23 12:48:31 +04:00
trx_t* trx); /* in: trx */
2001-02-17 14:19:19 +02:00
/* Query graph query thread node: the fields are protected by the kernel
mutex with the exceptions named below */
struct que_thr_struct{
que_common_t common; /* type: QUE_NODE_THR */
2002-01-28 22:18:49 +02:00
ulint magic_n; /* magic number to catch memory
corruption */
2001-02-17 14:19:19 +02:00
que_node_t* child; /* graph child node */
que_t* graph; /* graph where this node belongs */
ibool is_active; /* TRUE if the thread has been set
to the run state in
que_thr_move_to_run_state, but not
deactivated in
que_thr_dec_reference_count */
ulint state; /* state of the query thread */
UT_LIST_NODE_T(que_thr_t)
thrs; /* list of thread nodes of the fork
node */
UT_LIST_NODE_T(que_thr_t)
trx_thrs; /* lists of threads in wait list of
the trx */
UT_LIST_NODE_T(que_thr_t)
queue; /* list of runnable thread nodes in
the server task queue */
/*------------------------------*/
/* The following fields are private to the OS thread executing the
query thread, and are not protected by the kernel mutex: */
que_node_t* run_node; /* pointer to the node where the
subgraph down from this node is
currently executed */
que_node_t* prev_node; /* pointer to the node from which
the control came */
ulint resource; /* resource usage of the query thread
thus far */
Applied innodb-5.1-ss269 snapshot. Fixed BUGS: #3300: "UPDATE statement with no index column in where condition locks all rows" Implement semi-consistent read to reduce lock conflicts at the cost of breaking serializability. ha_innobase::unlock_row(): reset the "did semi consistent read" flag ha_innobase::was_semi_consistent_read(), ha_innobase::try_semi_consistent_read(): new methods row_prebuilt_t, row_create_prebuilt(): add field row_read_type for keeping track of semi-consistent reads row_vers_build_for_semi_consistent_read(), row_sel_build_committed_vers_for_mysql(): new functions row_search_for_mysql(): implement semi-consistent reads #9802: "Foreign key checks disallow alter table". Added test cases. #12456: "Cursor shows incorrect data - DML does not affect, probably caching" This patch implements a high-granularity read view to be used with cursors. In this high-granularity consistent read view modifications done by the creating transaction after the cursor is created or future transactions are not visible. But those modifications that transaction did before the cursor was created are visible. #12701: "Support >4GB buffer pool and log files on 64-bit Windows" Do not call os_file_create_tmpfile() at runtime. Instead, create all tempfiles at startup and guard access to them with mutexes. #13778: "If FOREIGN_KEY_CHECKS=0, one can create inconsistent FOREIGN KEYs". When FOREIGN_KEY_CHECKS=0 we still need to check that datatypes between foreign key references are compatible. #14189: "VARBINARY and BINARY variables: trailing space ignored with InnoDB" innobase_init(): Assert that DATA_MYSQL_BINARY_CHARSET_COLL == my_charset_bin.number. dtype_get_pad_char(): Do not pad VARBINARY or BINARY columns. row_ins_cascade_calc_update_vec(): Refuse ON UPDATE CASCADE when trying to change the length of a VARBINARY column that refers to or is referenced by a BINARY column. BINARY columns are no longer padded on comparison, and thus they cannot be padded on storage either. #14747: "Race condition can cause btr_search_drop_page_hash_index() to crash" Note that buf_block_t::index should be protected by btr_search_latch or an s-latch or x-latch on the index page. btr_search_drop_page_hash_index(): Read block->index while holding btr_search_latch and use the cached value in the loop. Remove some redundant assertions. #15108: "mysqld crashes when innodb_log_file_size is set > 4G" #15308: "Problem of Order with Enum Column in Primary Key" #15550: "mysqld crashes in printing a FOREIGN KEY error in InnoDB" row_ins_foreign_report_add_err(): When printing the parent record, use the index in the parent table rather than the index in the child table. #15653: "Slow inserts to InnoDB if many thousands of .ibd files" Keep track on unflushed modifications to file spaces. When there are tens of thousands of file spaces, flushing all files in fil_flush_file_spaces() would be very slow. fil_flush_file_spaces(): Only flush unflushed file spaces. fil_space_t, fil_system_t: Add a list of unflushed spaces. #15991: "innodb-file-per-table + symlink database + rename = cr" os_file_handle_error(): Map the error codes EXDEV, ENOTDIR, and EISDIR to the new code OS_FILE_PATH_ERROR. Treat this code as OS_FILE_PATH_ERROR. This fixes the crash on RENAME TABLE when the .ibd file is a symbolic link to a different file system. #16157: "InnoDB crashes when main location settings are empty" This patch is from Heikki. #16298: "InnoDB segfaults in INSERTs in upgrade of 4.0 -> 5.0 tables with VARCHAR BINARY" dict_load_columns(): Set the charset-collation code DATA_MYSQL_BINARY_CHARSET_COLL for those binary string columns that lack a charset-collation code, i.e., the tables were created with an older version of MySQL/InnoDB than 4.1.2. #16229: "MySQL/InnoDB uses full explicit table locks in trigger processing" Take a InnoDB table lock only if user has explicitly requested a table lock. Added some additional comments to store_lock() and external_lock(). #16387: "InnoDB crash when dropping a foreign key <table>_ibfk_0" Do not mistake TABLENAME_ibfk_0 for auto-generated id. dict_table_get_highest_foreign_id(): Ignore foreign constraint identifiers starting with the pattern TABLENAME_ibfk_0. #16582: "InnoDB: Error in an adaptive hash index pointer to page" Account for a race condition when dropping the adaptive hash index for a B-tree page. btr_search_drop_page_hash_index(): Retry the operation if a hash index with different parameters was built meanwhile. Add diagnostics for the case that hash node pointers to the page remain. btr_search_info_update_hash(), btr_search_info_update_slow(): Document the parameter "info" as in/out. #16814: "SHOW INNODB STATUS format error in LATEST FOREIGN KEY ERROR section" Add a missing newline to the LAST FOREIGN KEY ERROR section in SHOW INNODB STATUS output. dict_foreign_error_report(): Always print a newline after invoking dict_print_info_on_foreign_key_in_create_format(). #16827: "Better InnoDB error message if ibdata files omitted from my.cnf" #17126: "CHECK TABLE on InnoDB causes a short hang during check of adaptive hash" CHECK TABLE blocking other queries, by releasing the btr_search_latch periodically during the adaptive hash table validation. #17405: "Valgrind: conditional jump or move depends on unititialised values" buf_block_init(): Reset magic_n, buf_fix_count and io_fix to avoid testing uninitialized variables.
2006-03-10 19:22:21 +03:00
ulint lock_state; /* lock state of thread (table or
2005-03-16 00:34:15 +02:00
row) */
2001-02-17 14:19:19 +02:00
};
2002-01-28 22:18:49 +02:00
#define QUE_THR_MAGIC_N 8476583
#define QUE_THR_MAGIC_FREED 123461526
2001-02-17 14:19:19 +02:00
/* Query graph fork node: its fields are protected by the kernel mutex */
struct que_fork_struct{
que_common_t common; /* type: QUE_NODE_FORK */
que_t* graph; /* query graph of this node */
ulint fork_type; /* fork type */
ulint n_active_thrs; /* if this is the root of a graph, the
number query threads that have been
started in que_thr_move_to_run_state
but for which que_thr_dec_refer_count
has not yet been called */
trx_t* trx; /* transaction: this is set only in
the root node */
ulint state; /* state of the fork node */
que_thr_t* caller; /* pointer to a possible calling query
thread */
UT_LIST_BASE_NODE_T(que_thr_t)
thrs; /* list of query threads */
/*------------------------------*/
/* The fields in this section are defined only in the root node */
sym_tab_t* sym_tab; /* symbol table of the query,
generated by the parser, or NULL
if the graph was created 'by hand' */
Applied innodb-5.1-ss475 snapshot. * Fix BUG#15650: "DELETE with LEFT JOIN crashes server with innodb_locks_unsafe_for binlog" * Fix BUG#17134: "Partitions: uncommitted changes are visible" * Fix BUG#17992: "Partitions: InnoDB, somehow rotten table after UPDATE" row0ins.c: MySQL's partitioned table code does not set preduilt->sql_stat_start right if it does an insert in the same statement after doing a search first in the same partition table. We now write trx id always to the buffer, not just when flag sql_stat_start is on. This will waste CPU time very sightly. * Fix BUG#18077: "InnoDB uses full explicit table locks in stored FUNCTION" * Fix BUG#18238: "When locks exhaust the buffer pool, InnoDB does not roll back the trx" * Fix BUG#18252" "Disk space leak in updates of InnoDB BLOB rows in 5.0 and 5.1" * Fix BUG#18283: "When InnoDB returns error 'lock table full', MySQL can write to binlog too much" * Fix BUG#18350: "Use consistent read in CREATE ... SELECT ... if innodb_locks_unsafe_for_binlog" * Fix BUG#18384: "InnoDB memory leak on duplicate key errors in 5.0 if row has many columns" * Fix BUG#18934: "InnoDB crashes when table uses column names like DB_ROW_ID" Refuse tables that use reserved column names. * InnoDB's SQL parser: - Add support for UNSIGNED types, EXIT keyword, quoted identifiers, user-function callbacks for processing results of FETCH statements, bound literals, DATA_VARCHAR for bound literals. - Allow bound literals of type non-INTEGER to be of length 0. - Add make_flex.sh and update lexer/parser generation documentation. - Add comment clarifying the difference between 'alias' and 'indirection' fields in sym_node_t. - Remove never reached duplicate code in pars_set_dfield_type(). - Rewrite pars_info datatypes and APIs, add a few helper functions. - Since the functions definitions in pars_info_t are accessed after pars_sql() returns in the query graph execution stage, we can't free pars_info_t in pars_sql(). Instead, make pars_sql() transfer ownership of pars_info_t to the created query graph, and make que_graph_free() free it if needed. - Allow access to system columns like DB_ROW_ID. * Use bound literals in row_truncate_table_for_mysql, row_drop_table_for_mysql, row_discard_tablespace_for_mysql, and row_rename_table_for_mysql. * Setting an isolation level of the transaction to read committed weakens the locks for this session similarly like the option innodb_locks_unsafe_for binlog. This patch removes alnost all gap locking (used in next-key locking) and makes MySQL to release the row locks on the rows which does not belong to result set. Additionally, nonlocking selects on INSERT INTO SELECT, UPDATE ... (SELECT ...), and CREATE ... SELECT ... use a nonlocking consistent read. If a binlog is used, then binlog format should be set to row based binloging to make the execution of the complex SQL statements. * Disable the statistic variables btr_search_n_hash_fail and n_hash_succ, n_hash_fail, n_patt_succ, and n_searches of btr_search_t in builds without #ifdef UNIV_SEARCH_PERF_STAT. * Make innodb.test faster. Group all consistent read test cases to a one test case and wait their lock timeout after all have been send to the server. Decrease amount of rows inserted in a certain test - this has no effect on the effectiveness of the test and reduces the running time by ~10 sec. Remove temporary work-arounds from innodb.result now that ALTER TABLE DROP FOREIGN KEY works once again. * Make innodb_unsafe_binlog.test faster. Grout all consistent read test cases to a one test case amd wait their lock timeout after all have been sent to the server. Remove unnecessary option --loose_innodb_lock_wait_timeout. * Print dictionary memory size in SHOW INNODB STATUS. * Fix memory leaks in row_create_table_for_mysql() in rare corner cases. * Remove code related to clustered tables. They were never implemented, and the implementation would be challenging with ROW_FORMAT=COMPACT. Remove the table types DICT_TABLE_CLUSTER_MEMBER and DICT_TABLE_CLUSTER and all related tests and functions. dict_table_t: Remove mix_id, mix_len, mix_id_len, mix_id_buf, and cluster_name. plan_t: Remove mixed_index. dict_create_sys_tables_tuple(): Set MIX_ID=0, MIX_LEN=0, CLUSTER_NAME=NULL when inserting into SYS_TABLES. dict_tree_check_search_tuple(): Enclose in #ifdef UNIV_DEBUG. * Move calling of thr_local_free() from trx_free_for_mysql() to innobase_close_connection().
2006-04-23 12:48:31 +04:00
pars_info_t* info; /* in: info struct, or NULL */
2001-02-17 14:19:19 +02:00
/* The following cur_... fields are relevant only in a select graph */
ulint cur_end; /* QUE_CUR_NOT_DEFINED, QUE_CUR_START,
QUE_CUR_END */
ulint cur_pos; /* if there are n rows in the result
set, values 0 and n + 1 mean before
first row, or after last row, depending
on cur_end; values 1...n mean a row
index */
ibool cur_on_row; /* TRUE if cursor is on a row, i.e.,
it is not before the first row or
after the last row */
dulint n_inserts; /* number of rows inserted */
dulint n_updates; /* number of rows updated */
dulint n_deletes; /* number of rows deleted */
sel_node_t* last_sel_node; /* last executed select node, or NULL
if none */
UT_LIST_NODE_T(que_fork_t)
graphs; /* list of query graphs of a session
or a stored procedure */
/*------------------------------*/
mem_heap_t* heap; /* memory heap where the fork was
created */
Applied innodb-5.1-ss269 snapshot. Fixed BUGS: #3300: "UPDATE statement with no index column in where condition locks all rows" Implement semi-consistent read to reduce lock conflicts at the cost of breaking serializability. ha_innobase::unlock_row(): reset the "did semi consistent read" flag ha_innobase::was_semi_consistent_read(), ha_innobase::try_semi_consistent_read(): new methods row_prebuilt_t, row_create_prebuilt(): add field row_read_type for keeping track of semi-consistent reads row_vers_build_for_semi_consistent_read(), row_sel_build_committed_vers_for_mysql(): new functions row_search_for_mysql(): implement semi-consistent reads #9802: "Foreign key checks disallow alter table". Added test cases. #12456: "Cursor shows incorrect data - DML does not affect, probably caching" This patch implements a high-granularity read view to be used with cursors. In this high-granularity consistent read view modifications done by the creating transaction after the cursor is created or future transactions are not visible. But those modifications that transaction did before the cursor was created are visible. #12701: "Support >4GB buffer pool and log files on 64-bit Windows" Do not call os_file_create_tmpfile() at runtime. Instead, create all tempfiles at startup and guard access to them with mutexes. #13778: "If FOREIGN_KEY_CHECKS=0, one can create inconsistent FOREIGN KEYs". When FOREIGN_KEY_CHECKS=0 we still need to check that datatypes between foreign key references are compatible. #14189: "VARBINARY and BINARY variables: trailing space ignored with InnoDB" innobase_init(): Assert that DATA_MYSQL_BINARY_CHARSET_COLL == my_charset_bin.number. dtype_get_pad_char(): Do not pad VARBINARY or BINARY columns. row_ins_cascade_calc_update_vec(): Refuse ON UPDATE CASCADE when trying to change the length of a VARBINARY column that refers to or is referenced by a BINARY column. BINARY columns are no longer padded on comparison, and thus they cannot be padded on storage either. #14747: "Race condition can cause btr_search_drop_page_hash_index() to crash" Note that buf_block_t::index should be protected by btr_search_latch or an s-latch or x-latch on the index page. btr_search_drop_page_hash_index(): Read block->index while holding btr_search_latch and use the cached value in the loop. Remove some redundant assertions. #15108: "mysqld crashes when innodb_log_file_size is set > 4G" #15308: "Problem of Order with Enum Column in Primary Key" #15550: "mysqld crashes in printing a FOREIGN KEY error in InnoDB" row_ins_foreign_report_add_err(): When printing the parent record, use the index in the parent table rather than the index in the child table. #15653: "Slow inserts to InnoDB if many thousands of .ibd files" Keep track on unflushed modifications to file spaces. When there are tens of thousands of file spaces, flushing all files in fil_flush_file_spaces() would be very slow. fil_flush_file_spaces(): Only flush unflushed file spaces. fil_space_t, fil_system_t: Add a list of unflushed spaces. #15991: "innodb-file-per-table + symlink database + rename = cr" os_file_handle_error(): Map the error codes EXDEV, ENOTDIR, and EISDIR to the new code OS_FILE_PATH_ERROR. Treat this code as OS_FILE_PATH_ERROR. This fixes the crash on RENAME TABLE when the .ibd file is a symbolic link to a different file system. #16157: "InnoDB crashes when main location settings are empty" This patch is from Heikki. #16298: "InnoDB segfaults in INSERTs in upgrade of 4.0 -> 5.0 tables with VARCHAR BINARY" dict_load_columns(): Set the charset-collation code DATA_MYSQL_BINARY_CHARSET_COLL for those binary string columns that lack a charset-collation code, i.e., the tables were created with an older version of MySQL/InnoDB than 4.1.2. #16229: "MySQL/InnoDB uses full explicit table locks in trigger processing" Take a InnoDB table lock only if user has explicitly requested a table lock. Added some additional comments to store_lock() and external_lock(). #16387: "InnoDB crash when dropping a foreign key <table>_ibfk_0" Do not mistake TABLENAME_ibfk_0 for auto-generated id. dict_table_get_highest_foreign_id(): Ignore foreign constraint identifiers starting with the pattern TABLENAME_ibfk_0. #16582: "InnoDB: Error in an adaptive hash index pointer to page" Account for a race condition when dropping the adaptive hash index for a B-tree page. btr_search_drop_page_hash_index(): Retry the operation if a hash index with different parameters was built meanwhile. Add diagnostics for the case that hash node pointers to the page remain. btr_search_info_update_hash(), btr_search_info_update_slow(): Document the parameter "info" as in/out. #16814: "SHOW INNODB STATUS format error in LATEST FOREIGN KEY ERROR section" Add a missing newline to the LAST FOREIGN KEY ERROR section in SHOW INNODB STATUS output. dict_foreign_error_report(): Always print a newline after invoking dict_print_info_on_foreign_key_in_create_format(). #16827: "Better InnoDB error message if ibdata files omitted from my.cnf" #17126: "CHECK TABLE on InnoDB causes a short hang during check of adaptive hash" CHECK TABLE blocking other queries, by releasing the btr_search_latch periodically during the adaptive hash table validation. #17405: "Valgrind: conditional jump or move depends on unititialised values" buf_block_init(): Reset magic_n, buf_fix_count and io_fix to avoid testing uninitialized variables.
2006-03-10 19:22:21 +03:00
2001-02-17 14:19:19 +02:00
};
/* Query fork (or graph) types */
#define QUE_FORK_SELECT_NON_SCROLL 1 /* forward-only cursor */
#define QUE_FORK_SELECT_SCROLL 2 /* scrollable cursor */
#define QUE_FORK_INSERT 3
#define QUE_FORK_UPDATE 4
#define QUE_FORK_ROLLBACK 5
/* This is really the undo graph used in rollback,
no signal-sending roll_node in this graph */
#define QUE_FORK_PURGE 6
#define QUE_FORK_EXECUTE 7
#define QUE_FORK_PROCEDURE 8
#define QUE_FORK_PROCEDURE_CALL 9
#define QUE_FORK_MYSQL_INTERFACE 10
#define QUE_FORK_RECOVERY 11
/* Query fork (or graph) states */
#define QUE_FORK_ACTIVE 1
#define QUE_FORK_COMMAND_WAIT 2
#define QUE_FORK_INVALID 3
#define QUE_FORK_BEING_FREED 4
/* Flag which is ORed to control structure statement node types */
#define QUE_NODE_CONTROL_STAT 1024
/* Query graph node types */
#define QUE_NODE_LOCK 1
#define QUE_NODE_INSERT 2
#define QUE_NODE_UPDATE 4
#define QUE_NODE_CURSOR 5
#define QUE_NODE_SELECT 6
#define QUE_NODE_AGGREGATE 7
#define QUE_NODE_FORK 8
#define QUE_NODE_THR 9
#define QUE_NODE_UNDO 10
#define QUE_NODE_COMMIT 11
#define QUE_NODE_ROLLBACK 12
#define QUE_NODE_PURGE 13
#define QUE_NODE_CREATE_TABLE 14
#define QUE_NODE_CREATE_INDEX 15
#define QUE_NODE_SYMBOL 16
#define QUE_NODE_RES_WORD 17
#define QUE_NODE_FUNC 18
#define QUE_NODE_ORDER 19
#define QUE_NODE_PROC (20 + QUE_NODE_CONTROL_STAT)
#define QUE_NODE_IF (21 + QUE_NODE_CONTROL_STAT)
#define QUE_NODE_WHILE (22 + QUE_NODE_CONTROL_STAT)
#define QUE_NODE_ASSIGNMENT 23
#define QUE_NODE_FETCH 24
#define QUE_NODE_OPEN 25
#define QUE_NODE_COL_ASSIGNMENT 26
#define QUE_NODE_FOR (27 + QUE_NODE_CONTROL_STAT)
#define QUE_NODE_RETURN 28
#define QUE_NODE_ROW_PRINTF 29
#define QUE_NODE_ELSIF 30
#define QUE_NODE_CALL 31
Applied innodb-5.1-ss475 snapshot. * Fix BUG#15650: "DELETE with LEFT JOIN crashes server with innodb_locks_unsafe_for binlog" * Fix BUG#17134: "Partitions: uncommitted changes are visible" * Fix BUG#17992: "Partitions: InnoDB, somehow rotten table after UPDATE" row0ins.c: MySQL's partitioned table code does not set preduilt->sql_stat_start right if it does an insert in the same statement after doing a search first in the same partition table. We now write trx id always to the buffer, not just when flag sql_stat_start is on. This will waste CPU time very sightly. * Fix BUG#18077: "InnoDB uses full explicit table locks in stored FUNCTION" * Fix BUG#18238: "When locks exhaust the buffer pool, InnoDB does not roll back the trx" * Fix BUG#18252" "Disk space leak in updates of InnoDB BLOB rows in 5.0 and 5.1" * Fix BUG#18283: "When InnoDB returns error 'lock table full', MySQL can write to binlog too much" * Fix BUG#18350: "Use consistent read in CREATE ... SELECT ... if innodb_locks_unsafe_for_binlog" * Fix BUG#18384: "InnoDB memory leak on duplicate key errors in 5.0 if row has many columns" * Fix BUG#18934: "InnoDB crashes when table uses column names like DB_ROW_ID" Refuse tables that use reserved column names. * InnoDB's SQL parser: - Add support for UNSIGNED types, EXIT keyword, quoted identifiers, user-function callbacks for processing results of FETCH statements, bound literals, DATA_VARCHAR for bound literals. - Allow bound literals of type non-INTEGER to be of length 0. - Add make_flex.sh and update lexer/parser generation documentation. - Add comment clarifying the difference between 'alias' and 'indirection' fields in sym_node_t. - Remove never reached duplicate code in pars_set_dfield_type(). - Rewrite pars_info datatypes and APIs, add a few helper functions. - Since the functions definitions in pars_info_t are accessed after pars_sql() returns in the query graph execution stage, we can't free pars_info_t in pars_sql(). Instead, make pars_sql() transfer ownership of pars_info_t to the created query graph, and make que_graph_free() free it if needed. - Allow access to system columns like DB_ROW_ID. * Use bound literals in row_truncate_table_for_mysql, row_drop_table_for_mysql, row_discard_tablespace_for_mysql, and row_rename_table_for_mysql. * Setting an isolation level of the transaction to read committed weakens the locks for this session similarly like the option innodb_locks_unsafe_for binlog. This patch removes alnost all gap locking (used in next-key locking) and makes MySQL to release the row locks on the rows which does not belong to result set. Additionally, nonlocking selects on INSERT INTO SELECT, UPDATE ... (SELECT ...), and CREATE ... SELECT ... use a nonlocking consistent read. If a binlog is used, then binlog format should be set to row based binloging to make the execution of the complex SQL statements. * Disable the statistic variables btr_search_n_hash_fail and n_hash_succ, n_hash_fail, n_patt_succ, and n_searches of btr_search_t in builds without #ifdef UNIV_SEARCH_PERF_STAT. * Make innodb.test faster. Group all consistent read test cases to a one test case and wait their lock timeout after all have been send to the server. Decrease amount of rows inserted in a certain test - this has no effect on the effectiveness of the test and reduces the running time by ~10 sec. Remove temporary work-arounds from innodb.result now that ALTER TABLE DROP FOREIGN KEY works once again. * Make innodb_unsafe_binlog.test faster. Grout all consistent read test cases to a one test case amd wait their lock timeout after all have been sent to the server. Remove unnecessary option --loose_innodb_lock_wait_timeout. * Print dictionary memory size in SHOW INNODB STATUS. * Fix memory leaks in row_create_table_for_mysql() in rare corner cases. * Remove code related to clustered tables. They were never implemented, and the implementation would be challenging with ROW_FORMAT=COMPACT. Remove the table types DICT_TABLE_CLUSTER_MEMBER and DICT_TABLE_CLUSTER and all related tests and functions. dict_table_t: Remove mix_id, mix_len, mix_id_len, mix_id_buf, and cluster_name. plan_t: Remove mixed_index. dict_create_sys_tables_tuple(): Set MIX_ID=0, MIX_LEN=0, CLUSTER_NAME=NULL when inserting into SYS_TABLES. dict_tree_check_search_tuple(): Enclose in #ifdef UNIV_DEBUG. * Move calling of thr_local_free() from trx_free_for_mysql() to innobase_close_connection().
2006-04-23 12:48:31 +04:00
#define QUE_NODE_EXIT 32
2001-02-17 14:19:19 +02:00
/* Query thread states */
#define QUE_THR_RUNNING 1
#define QUE_THR_PROCEDURE_WAIT 2
#define QUE_THR_COMPLETED 3 /* in selects this means that the
thread is at the end of its result set
(or start, in case of a scroll cursor);
in other statements, this means the
thread has done its task */
#define QUE_THR_COMMAND_WAIT 4
#define QUE_THR_LOCK_WAIT 5
#define QUE_THR_SIG_REPLY_WAIT 6
#define QUE_THR_SUSPENDED 7
#define QUE_THR_ERROR 8
/* Query thread lock states */
Applied innodb-5.1-ss269 snapshot. Fixed BUGS: #3300: "UPDATE statement with no index column in where condition locks all rows" Implement semi-consistent read to reduce lock conflicts at the cost of breaking serializability. ha_innobase::unlock_row(): reset the "did semi consistent read" flag ha_innobase::was_semi_consistent_read(), ha_innobase::try_semi_consistent_read(): new methods row_prebuilt_t, row_create_prebuilt(): add field row_read_type for keeping track of semi-consistent reads row_vers_build_for_semi_consistent_read(), row_sel_build_committed_vers_for_mysql(): new functions row_search_for_mysql(): implement semi-consistent reads #9802: "Foreign key checks disallow alter table". Added test cases. #12456: "Cursor shows incorrect data - DML does not affect, probably caching" This patch implements a high-granularity read view to be used with cursors. In this high-granularity consistent read view modifications done by the creating transaction after the cursor is created or future transactions are not visible. But those modifications that transaction did before the cursor was created are visible. #12701: "Support >4GB buffer pool and log files on 64-bit Windows" Do not call os_file_create_tmpfile() at runtime. Instead, create all tempfiles at startup and guard access to them with mutexes. #13778: "If FOREIGN_KEY_CHECKS=0, one can create inconsistent FOREIGN KEYs". When FOREIGN_KEY_CHECKS=0 we still need to check that datatypes between foreign key references are compatible. #14189: "VARBINARY and BINARY variables: trailing space ignored with InnoDB" innobase_init(): Assert that DATA_MYSQL_BINARY_CHARSET_COLL == my_charset_bin.number. dtype_get_pad_char(): Do not pad VARBINARY or BINARY columns. row_ins_cascade_calc_update_vec(): Refuse ON UPDATE CASCADE when trying to change the length of a VARBINARY column that refers to or is referenced by a BINARY column. BINARY columns are no longer padded on comparison, and thus they cannot be padded on storage either. #14747: "Race condition can cause btr_search_drop_page_hash_index() to crash" Note that buf_block_t::index should be protected by btr_search_latch or an s-latch or x-latch on the index page. btr_search_drop_page_hash_index(): Read block->index while holding btr_search_latch and use the cached value in the loop. Remove some redundant assertions. #15108: "mysqld crashes when innodb_log_file_size is set > 4G" #15308: "Problem of Order with Enum Column in Primary Key" #15550: "mysqld crashes in printing a FOREIGN KEY error in InnoDB" row_ins_foreign_report_add_err(): When printing the parent record, use the index in the parent table rather than the index in the child table. #15653: "Slow inserts to InnoDB if many thousands of .ibd files" Keep track on unflushed modifications to file spaces. When there are tens of thousands of file spaces, flushing all files in fil_flush_file_spaces() would be very slow. fil_flush_file_spaces(): Only flush unflushed file spaces. fil_space_t, fil_system_t: Add a list of unflushed spaces. #15991: "innodb-file-per-table + symlink database + rename = cr" os_file_handle_error(): Map the error codes EXDEV, ENOTDIR, and EISDIR to the new code OS_FILE_PATH_ERROR. Treat this code as OS_FILE_PATH_ERROR. This fixes the crash on RENAME TABLE when the .ibd file is a symbolic link to a different file system. #16157: "InnoDB crashes when main location settings are empty" This patch is from Heikki. #16298: "InnoDB segfaults in INSERTs in upgrade of 4.0 -> 5.0 tables with VARCHAR BINARY" dict_load_columns(): Set the charset-collation code DATA_MYSQL_BINARY_CHARSET_COLL for those binary string columns that lack a charset-collation code, i.e., the tables were created with an older version of MySQL/InnoDB than 4.1.2. #16229: "MySQL/InnoDB uses full explicit table locks in trigger processing" Take a InnoDB table lock only if user has explicitly requested a table lock. Added some additional comments to store_lock() and external_lock(). #16387: "InnoDB crash when dropping a foreign key <table>_ibfk_0" Do not mistake TABLENAME_ibfk_0 for auto-generated id. dict_table_get_highest_foreign_id(): Ignore foreign constraint identifiers starting with the pattern TABLENAME_ibfk_0. #16582: "InnoDB: Error in an adaptive hash index pointer to page" Account for a race condition when dropping the adaptive hash index for a B-tree page. btr_search_drop_page_hash_index(): Retry the operation if a hash index with different parameters was built meanwhile. Add diagnostics for the case that hash node pointers to the page remain. btr_search_info_update_hash(), btr_search_info_update_slow(): Document the parameter "info" as in/out. #16814: "SHOW INNODB STATUS format error in LATEST FOREIGN KEY ERROR section" Add a missing newline to the LAST FOREIGN KEY ERROR section in SHOW INNODB STATUS output. dict_foreign_error_report(): Always print a newline after invoking dict_print_info_on_foreign_key_in_create_format(). #16827: "Better InnoDB error message if ibdata files omitted from my.cnf" #17126: "CHECK TABLE on InnoDB causes a short hang during check of adaptive hash" CHECK TABLE blocking other queries, by releasing the btr_search_latch periodically during the adaptive hash table validation. #17405: "Valgrind: conditional jump or move depends on unititialised values" buf_block_init(): Reset magic_n, buf_fix_count and io_fix to avoid testing uninitialized variables.
2006-03-10 19:22:21 +03:00
#define QUE_THR_LOCK_NOLOCK 0
#define QUE_THR_LOCK_ROW 1
#define QUE_THR_LOCK_TABLE 2
2001-02-17 14:19:19 +02:00
/* From where the cursor position is counted */
#define QUE_CUR_NOT_DEFINED 1
#define QUE_CUR_START 2
#define QUE_CUR_END 3
#ifndef UNIV_NONINL
#include "que0que.ic"
#endif
#endif