mirror/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2025-01-21 06:22:28 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions
23884 commits 2781 branches 343 tags 1.6 GiB
Commit graph

104 commits

Author SHA1 Message Date
marko@hundin.mysql.fi
e992adcb3f Merge hundin.mysql.fi:/home/marko/mysql-5.0 
into hundin.mysql.fi:/home/marko/mysql-5.0-current
 2005-06-21 07:36:18 +03:00
marko@hundin.mysql.fi
85515b579e InnoDB: remove flag lock_print_waits from non-debug builds. 2005-06-17 15:29:03 +03:00
marko@hundin.mysql.fi
a5025b8ebe InnoDB: After review fixes 2005-06-16 16:59:49 +03:00
marko@hundin.mysql.fi
c3da5b0b7b InnoDB: Do not distinguish explicitly requested table locks. 
Since UNLOCK TABLES will commit the transaction (and thus release
all InnoDB locks held by it), it is unnecessary to release the
user-requested table locks separately.
 2005-06-08 15:45:56 +03:00
marko@hundin.mysql.fi
2f0424cd62 Merge 2005-05-06 11:45:59 +03:00
marko@hundin.mysql.fi
b3d6f51705 InnoDB: Performance optimizations based on OProfile analysis 2005-04-25 10:14:35 +03:00
marko@hundin.mysql.fi
ba4f386765 row0upd.c: 
Add UNIV_UNLIKELY hint.
lock0lock.c:
  Correct the prototypes of lock_deadlock_recursive() and
  lock_deadlock_occurs().
data0data.c:
  dtuple_convert_big_rec(): Add UNIV_UNLIKELY hint.
ha_innodb.cc:
  ha_innobase::write_row(): Correct the type of 'mode'.
 2005-04-22 15:15:18 +03:00
marko@hundin.mysql.fi
c470116981 InnoDB: Define some macros around GCC's __builtin_expect() 
and __builtin_prefetch().
Add UNIV_LIKELY_NULL to if (heap) mem_free_heap(heap) tests.
 2005-04-21 14:23:26 +03:00
marko@hundin.mysql.fi
f9d145ef04 InnoDB: Truncate SHOW INNODB STATUS output at the start of the list 
of active transactions, if necessary and possible.  (Bug #5436)
 2005-04-19 14:35:47 +03:00
marko@hundin.mysql.fi
c4ace2a471 InnoDB: Introduce the symbols REC_OFFS_NORMAL_SIZE and 
REC_OFFS_SMALL_SIZE for the initial allocation sizes of
arrays passed to rec_get_offsets().
 2005-03-10 15:16:16 +02:00
marko@hundin.mysql.fi
6920a6d546 InnoDB: Improve performance by about 10% by removing implicit 
memcpy() calls, by not initializing the offsets_[] arrays.
InnoDB: Remove a Microsoft compiler warning in page0page.c.
 2005-03-09 22:04:55 +02:00
marko@hundin.mysql.fi
6c43fb504b InnoDB: Exclude code from InnoDB Hot Backup builds 
by adding #ifndef UNIV_HOTBACKUP and adjusting some function
declarations.
 2005-02-04 16:25:13 +02:00
jan@hundin.mysql.fi
a5a1f7d7c3 Fixed problems with deadlocks on LOCK TABLES...WHERE ENGINE = InnoDB. 2004-12-13 14:22:18 +02:00
jan@hundin.mysql.fi
1d3a32c719 Merge jlindstrom@bk-internal.mysql.com:/home/bk/mysql-5.0 
into hundin.mysql.fi:/home/jan/mysql-5.0
 2004-12-13 08:59:52 +02:00
marko@hundin.mysql.fi
9dc1f5146a InnoDB: Reduce the stack space consumption of ON UPDATE CASCADE 
operations.
 2004-12-10 12:55:56 +02:00
jan@hundin.mysql.fi
9f5b8f9b4a Merge jlindstrom@bk-internal.mysql.com:/home/bk/mysql-5.0 
into hundin.mysql.fi:/home/jan/mysql-5.0
 2004-12-10 08:49:31 +02:00
marko@hundin.mysql.fi
7506ca1be1 InnoDB: Changed interface to rec_get_offsets(), to reduce the use of 
memory heaps.  This changeset plugs also a few memory leaks that
were introduced with the compact InnoDB table format.
 2004-12-09 15:29:55 +02:00
jan@hundin.mysql.fi
d6effde5d0 Added support for a LOCK TABLES...WHERE ENGINE = InnoDB query which sets 
transactional table locks to tables mentioned in the query. These locks
are released at the end of the transaction automatically.
This is fix for bugs #5655, #5998 and issue #3762.
 2004-12-09 11:10:45 +02:00
marko@hundin.mysql.fi
dd48953956 Many files: 
Implement more compact InnoDB record format.
  Old format is available as CREATE TABLE ... ROW_FORMAT=DYNAMIC.
 2004-12-02 19:45:07 +02:00
brian@avenger.(none)
b85f4677d5 Merge for Matt for Innodb bug. 2004-11-30 10:10:40 -08:00
marko@hundin.mysql.fi
d365446f62 Merge marko@bk-internal.mysql.com:/home/bk/mysql-4.1 
into hundin.mysql.fi:/home/marko/k/mysql-4.1
 2004-11-30 17:52:30 +02:00
heikki@hundin.mysql.fi
12738cd23a lock0lock.c: 
If UNIV_DEBUG is defined, lock_get_type() was called before function definition: fix this
 2004-11-30 17:51:10 +02:00
marko@hundin.mysql.fi
a35a93aa3c InnoDB: Allow ALTER TABLE to do intermediate COMMIT also when the table 
contains auto_increment columns.  (Bug #6633)
 2004-11-30 17:34:37 +02:00
marko@hundin.mysql.fi
2e7cb4d011 InnoDB: Make intermediate COMMITs in ALTER TABLE more robust (Bug #6633) 2004-11-27 00:45:01 +02:00
serg@serg.mylan
ee933cd198 merged 2004-11-22 21:33:15 +01:00
heikki@hundin.mysql.fi
7f5661ae44 trx0undo.c, trx0purge.c: 
Print a warning to the .err log if the InnoDB history list length is > 20 000 even though purge reaches the list head; this is to track corruption reported on the MySQL mailing list Nov 9, 2004
lock0lock.c:
  Let SHOW INNODB STATUS print the history list length
 2004-11-21 17:38:15 +02:00
monty@mysql.com
87af3d9113 merge on pull 2004-11-09 04:06:44 +02:00
jan@hundin.mysql.fi
a3f4bd28ac Fixed a bug in UPDATE statement with no index column in where condition 
locks all rows (BUG #3300). When using innobase_locks_unsafe_for_binlog
option InnoDB does not take locks for those rows which do not
belong to the result set or werent changed by the query. This fix removes
unnecessary locks also from SELECT and DELETE queries.
 2004-11-08 14:52:15 +02:00
marko@hundin.mysql.fi
fd069e2bb3 InnoDB: commit after every 10000 rows in ALTER TABLE 2004-11-03 21:32:48 +02:00
joerg@mysql.com
a9fc6cfd31 Correct an 'unresolved identifier' problem caused by an "inline" 
function being used before it was defined - "forward" declaration
was insufficient.
 2004-10-07 12:02:22 +02:00
heikki@hundin.mysql.fi
e7500f435f lock0lock.c: 
Fix compiler warning
 2004-10-04 19:48:41 +03:00
marko@hundin.mysql.fi
76cda983a9 Merge marko@build.mysql.com:/home/bk/mysql-4.1 
into hundin.mysql.fi:/home/marko/j/mysql-4.1
 2004-10-01 14:17:05 +03:00
marko@hundin.mysql.fi
d7ea8e7c6c InnoDB: quote identifiers according to MySQL settings (Bug #5292) 2004-10-01 11:51:59 +03:00
jan@hundin.mysql.fi
9cae7b0885 Merge jlindstrom@build.mysql.com:/home/bk/mysql-4.1 
into hundin.mysql.fi:/home/jan/mysql-4.1
 2004-08-19 11:37:46 +03:00
serg@serg.mylan
e1237cbc53 manually merged 2004-08-18 19:57:55 +02:00
jan@hundin.mysql.fi
656a376285 Relaxed lock wait rules so that record lock does not need to wait for a 
gap type lock.
 2004-08-17 09:45:54 +03:00
jan@hundin.mysql.fi
38255a2cad This change includes optimization for a lock wait rules and a new behaviour for 
a REPLACE command.
 2004-08-12 13:38:10 +03:00
jan@hundin.mysql.fi
664cce3f3c Take X-lock for duplicate keys in REPLACE command. 2004-08-11 13:01:30 +03:00
marko@hundin.mysql.fi
3cb09c9817 InnoDB: Add option for disabling innodb_status.<pid> files. 
InnoDB: Implement tmpfile() differently on Windows (Bug #3998)
 2004-08-06 15:55:50 +03:00
serg@serg.mylan
1f0cd873c5 manual merged (blame me!) 2004-07-20 00:53:24 +02:00
marko@hundin.mysql.fi
b92d70b231 lock0lock.c: 
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().
 2004-07-13 12:30:34 +03:00
marko@hundin.mysql.fi
88369368f5 InnoDB: LOCK TABLES clean-up, part 2 2004-07-12 17:14:13 +03:00
marko@hundin.mysql.fi
a323342728 InnoDB: LOCK TABLE clean-up 2004-07-12 16:47:22 +03:00
monty@mysql.com
b11d258835 Merge with 4.0.21 2004-06-18 04:38:58 +03:00
marko@hundin.mysql.fi
d2724e66db InnoDB: Revert most of ChangeSet@1.1772, as the debug functions may 
be needed when testing production releases
 2004-06-16 13:41:14 +03:00
monty@mysql.com
70c2256d2f Merge with 4.0 to get fixes for netware 2004-06-02 00:09:14 +03:00
marko@hundin.mysql.fi
fb9257dc45 InnoDB: Add diagnostics when tmpfile() fails at start (Bug #3919) 2004-05-28 11:15:59 +03:00
monty@mysql.com
cd3dab49cc Merge with 4.0.20 2004-05-17 01:52:13 +03:00
marko@hundin.mysql.fi
a5cca9df64 InnoDB cleanup: eliminate IB__FILE__ 2004-05-14 16:06:21 +03:00
heikki@hundin.mysql.fi
d2649c110a ha_innodb.cc, trx0trx.h, lock0lock.c, trx0trx.c: 
Reserve the MySQL LOCK_thread_count mutex when printing thd->query of
an arbitrary transaction; if we are printing thd->query of a transaction that
we know is currently executing inside InnoDB, then we know that MySQL cannot
meanwhile change thd->query, and no need to reserve the MySQL mutex; note
that this patch still leaves open the possibility of races in MySQL's
thd->query_len
 2004-05-13 22:07:51 +03:00