| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |\ |
|
| | |\ |
|
| | | |
| | |
| | |
| | | |
(Bug #4446)
|
| | | |
| | |
| | |
| | | |
Do not add + 1 to the InnoDB index cardinality estimate if the B-tree just contains one page; the fix made in March 2004 caused InnoDB systematically to overestimate the cardinality of empty or small tables by 1
|
| | | |
| | |
| | |
| | |
| | |
| | | |
Decrement n_lock_table_exp in lock_table_remove_low()
instead of lock_table_dequeue().
Do not empty lock_heap in lock_release_tables_off_kernel().
|
| | | | |
|
| | | | |
|
| | | |
| | |
| | |
| | | |
(Bug #2694)
|
| | | |
| | |
| | |
| | |
| | | |
innobase_start_or_create_for_mysql(): Rename innodb.status.<pid>
to innodb_status.<pid> to avoid problems on VMS
|
| |\ \ \
| |/ / |
|
| | |\ \
| | |/ |
|
| | | |
| | |
| | |
| | | |
Document the handling of quoted strings
|
| | | |\
| | | |
| | | |
| | | | |
into hundin.mysql.fi:/home/marko/k/mysql-4.0
|
| | | | |
| | | |
| | | |
| | | | |
InnoDB with UNIV_MEM_DEBUG did not compile because a global variable was forgotten when we put back mem_print_info() functions 2 days ago; add that variable
|
| | | |/
| | |
| | |
| | |
| | |
| | | |
Improve documentation of mach_*write_*compressed()
mtr0log.ic, mtr0log.h:
Improve documentation of mlog_write_initial_log_record_fast()
|
| | |\ \
| | | |
| | | |
| | | | |
into mysql.com:/home/mysql_src/mysql-4.1-1717
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Won't be pushed as is - separate email sent for internal review.
WL#1717 "binlog-innodb consistency".
Now when mysqld starts, if InnoDB does a crash recovery, we use the binlog name
and position retrieved from InnoDB (corresponding to the last transaction
successfully committed by InnoDB) to cut any rolled back transaction from
the binary log. This is triggered by the --innodb-safe-binlog option.
Provided you configure mysqld to fsync() InnoDB at every commit (using
flush_log_at_trx_commit) and to fsync() the binlog at every write
(using --sync-binlog=1), this behaviour guarantees that a master always has
consistency between binlog and InnoDB, whenever the crash happens.
6 tests to verify that it works.
|
| | |\ \ \
| | |/ /
| |/| | |
|
| | | | |
| | | |
| | | |
| | | | |
Return NULL if a time argument is given to date_add(). (Warning will be shown after Dimitri's timezone patch is pushed)
|
| | | |\ \
| | | |/ |
|
| | | | |
| | | |
| | | |
| | | | |
Put back mem_print_info() functions that are used to analyze memory leaks; Marko removed them by accident in his Feb 20th, 2004 patch; these functions are very important for debugging, and should always be kept in the source code base
|
| | | | |
| | | |
| | | |
| | | | |
Put asserts that make sure srv_que_task_enqueue() is never called under MySQL
|
| | | | |
| | | |
| | | |
| | | | |
Revert the patch that put log debug code behind UNIV_LOG_DEBUG
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Revert the patch that put log debug code behind UNIV_LOG_DEBUG
log0log.ic, log0recv.c, log0log.c, log0log.h:
Revert 1.1730.18.1 about UNIV_LOG_DEBUG: debug code is often needed in debugging the production version
|
| | | | | |
|
| | | | | |
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Revert Marko's untested changeset 1.1844.1.6 to the stable 4.0 version; stable 4.0 must NOT BE modified except in case of bugs
srv0que.c, srv0que.h:
new file
|
| | | | | |
|
| | | | |
| | | |
| | | |
| | | | |
Remove printf's that were accidentally pushed in the last push
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Implement ut_realloc
lexyy.c, pars0lex.l:
Start using ut_malloc, ut_free, ut_realloc in pars0lex.l and lexyy.c; mem_realloc was broken; eliminate possible memcpy(NULL, ..., 0) from string_append()
|
| | | | |
| | | |
| | | |
| | | | |
Remove broken mem_realloc(); a patch to remove its uses in pars0lex.l and fil0fil.c will soon follow
|
| | | | | |
|
| | | | |
| | | |
| | | |
| | | | |
be needed when testing production releases
|
| | | | | |
|
| | | | |
| | | |
| | | |
| | | |
| | | | |
cmp_debug_dtuple_rec_with_match()
unless #ifdef UNIV_DEBUG
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
high-byte-first. (Bug #4173)
Fixed problem with NULL and derived tables (Bug #4097)
Cleanup of new pushed code
|
| | | | |
| | | |
| | | |
| | | | |
Remove the use of mem_realloc in fil0fil.c; mem_realloc() was broken and the patch to remove its definition will propagate from the 4.0 tree; crash recovery with innodb_file_per_table has to be tested after this patch, not tested yet
|
| | | | |
| | | |
| | | |
| | | | |
Return the code of row_ins_foreign_check_on_constraint() as close to 4.0 as possible, except that we use mem_strdup() to allocate the memory for the modified table name
|
| | |/ /
| | |
| | |
| | |
| | |
| | |
| | | |
introduced in ChangeSet@1.1843.1.25
InnoDB: fix bug in the error exit of
fil_create_new_single_table_tablespace(), introduced in
ChangeSet@1.1843.1.11
|
| | | | |
|
| | | |
| | |
| | |
| | | |
Make allocation for file path more uniform: always use mem_alloc(), not ut_malloc()
|
| | | |
| | |
| | |
| | | |
Align file i/o buffers for DIRECT_IO; fix mem_alloc()/mem_free() crash bugs that came from Marko's latest cleanup
|
| | | |
| | |
| | |
| | | |
Fix typo spotted by Paul DuBois: 'no continue' -> 'not continue'
|
| | |\ \
| | | |
| | | |
| | | | |
into mysql.com:/home/my/mysql-4.1
|
| | | | |
| | | |
| | | |
| | | | |
Stop InnoDB crash recovery if an .ibd file for a table exists in a database directory, but we cannot access it.
|
| | | | | |
|
| | |\ \ \
| | |/ /
| |/| /
| | |/ |
|
| | | |
| | |
| | |
| | | |
Inside LOCK TABLES, use either LOCK_S or LOCK_X in locking reads; an improvent over the previous patch
|
| | | | |
|
| | | |
| | |
| | |
| | | |
Do not lock raw devices or files opened for read only
|
