into sinisa.nasamreza.org:/mnt/work/mysql


Docs/manual.texi:
  Auto merged

Docs/manual.texi:
  removing lines that do not follow texi rules correctly.
  Hope Arjen can put this together properly ...

and removed the 3.23.53 News section for now until 3.23.52 is officially
tagged.


Docs/manual.texi:
  Moved this changelog entry into the 3.23.52 section and removed the
  3.23.53 News section for now until 3.23.52 is officially tagged.

 - Only define alloca as a compiler builtin, if we use GCC and there
   is no alloca.h installed


include/my_sys.h:
   - Only define alloca as a compiler builtin, if we use GCC and there
     is no alloca.h installed

  Add some more safety margin to buffer sizes in prints


innobase/trx/trx0trx.c:
  Add some more safety margin to buffer sizes in prints
innobase/lock/lock0lock.c:
  Add some more safety margin to buffer sizes in prints
innobase/include/trx0trx.h:
  Add some more safety margin to buffer sizes in prints
sql/ha_innobase.cc:
  Add some more safety margin to buffer sizes in prints

  Found the bug in modified innobase_mysqld_print_thd; the person who modifies code should test that his change works


sql/ha_innobase.cc:
  Found the bug in modified innobase_mysqld_print_thd; the person who modifies code should test that his change works

  Revert changes to sprintf until we know why control characters scrambled the output


sql/ha_innobase.cc:
  Revert changes to sprintf until we know why control characters scrambled the output

include/my_sys.h:
  Removed redefine of alloca with GCC

Remove not needed locks in SHOW OPEN TABLES


sql/ha_innobase.cc:
  Reverted Heikkis patch as his was not portable.
sql/sql_show.cc:
  Remove not needed locks in SHOW OPEN TABLES

   compiler builtin when using gcc


include/my_sys.h:
   - Portability fix: define alloca as a compiler builtin on systems
     that have gcc but no alloca.h (e.g. some IBM AIX flavors), use
     "#pragma alloca" when using IBMs compiler (xlr_C)

  Merge InnoDB-3.23.52d


innobase/btr/btr0sea.c:
  Merge InnoDB-3.23.52d
innobase/buf/buf0buf.c:
  Merge InnoDB-3.23.52d
innobase/buf/buf0lru.c:
  Merge InnoDB-3.23.52d
innobase/include/buf0buf.h:
  Merge InnoDB-3.23.52d
innobase/include/ha0ha.h:
  Merge InnoDB-3.23.52d
innobase/include/log0log.h:
  Merge InnoDB-3.23.52d
innobase/include/os0file.h:
  Merge InnoDB-3.23.52d
innobase/include/os0thread.h:
  Merge InnoDB-3.23.52d
innobase/include/ha0ha.ic:
  Merge InnoDB-3.23.52d
innobase/include/os0sync.ic:
  Merge InnoDB-3.23.52d
innobase/include/srv0start.h:
  Merge InnoDB-3.23.52d
innobase/include/sync0rw.ic:
  Merge InnoDB-3.23.52d
innobase/include/sync0sync.ic:
  Merge InnoDB-3.23.52d
innobase/include/ut0dbg.h:
  Merge InnoDB-3.23.52d
innobase/include/univ.i:
  Merge InnoDB-3.23.52d
innobase/lock/lock0lock.c:
  Merge InnoDB-3.23.52d
innobase/log/log0log.c:
  Merge InnoDB-3.23.52d
innobase/mem/mem0pool.c:
  Merge InnoDB-3.23.52d
innobase/os/os0file.c:
  Merge InnoDB-3.23.52d
innobase/os/os0thread.c:
  Merge InnoDB-3.23.52d
innobase/srv/srv0srv.c:
  Merge InnoDB-3.23.52d
innobase/srv/srv0start.c:
  Merge InnoDB-3.23.52d
innobase/sync/sync0arr.c:
  Merge InnoDB-3.23.52d
innobase/sync/sync0rw.c:
  Merge InnoDB-3.23.52d
innobase/sync/sync0sync.c:
  Merge InnoDB-3.23.52d
innobase/thr/thr0loc.c:
  Merge InnoDB-3.23.52d
innobase/trx/trx0trx.c:
  Merge InnoDB-3.23.52d
innobase/configure.in:
  Merge InnoDB-3.23.52d
sql/ha_innobase.cc:
  Merge InnoDB-3.23.52d

Fixed bug in automatic repair of MyISAM tables where table cache was not locked properly


Docs/manual.texi:
  Changelog
libmysql/Makefile.am:
  Fixed problem with make distcheck
libmysql/Makefile.shared:
  Fixed problem with make distcheck
sql/sql_base.cc:
  Fixed bug in automatic repair where table cache was not locked properly.
strings/Makefile.am:
  Fixed problem with make distcheck

Removed Heikki's patch for handling this.

Added option --host to mysqlhotcopy


mysql-test/mysql-test-run.sh:
  Added error message if the server doesn't start.
  Increase connect timeout a bit (for running under purify).
mysql-test/t/rpl000001.test:
  Longer sleep time (for running under purify)
scripts/mysqlhotcopy.sh:
  Added option --host for usage with TCP/IP connections
sql/gen_lex_hash.cc:
  Fixed typo

  In HP-UX-10.20, but not in HP-UX-11.0, the return value of pthread_mutex_trylock is inverted


configure.in:
  In HP-UX-10.20, but not in HP-UX-11.0, the return value of pthread_mutex_trylock is inverted
include/my_pthread.h:
  In HP-UX-10.20, but not in HP-UX-11.0, the return value of pthread_mutex_trylock is inverted

Use MYSQL_CONFIG_NAME instead of "my" as the configuration file name.


Docs/manual.texi:
  Changelog
include/mysql_version.h.in:
  Use MYSQL_CONFIG_NAME instead of "my" as the configuration file name.

BitKeeper/etc/logging_ok:
  auto-union
sql/sql_base.cc:
  Auto merged

This should fix some possible table cache corruptions when doing
OPTIMIZE or REPAIR table when other threads are opening new tables.


sql/sql_base.cc:
  Added missing mutex unlock on error condition
sql/sql_insert.cc:
  Added TODO item
sql/sql_show.cc:
  Added missing pthread_mutex_lock(&LOCK_open) when calling
  query_table_status().
sql/sql_table.cc:
  Added missing pthread_mutex_lock(&LOCK_open) when calling
  hash_delete(), unlock_table_name() and remove_table_from_cache().
BitKeeper/etc/logging_ok:
  Logging to logging@openlogging.org accepted

(Merge of code from 4.0)


Docs/manual.texi:
  Changelog
include/my_base.h:
  Fixed bug with GROUP BY on NULL fields.
include/mysql_com.h:
  Fixed bug with GROUP BY on NULL fields.
myisam/mi_search.c:
  Fixed bug with GROUP BY on NULL fields.
myisam/mi_write.c:
  Fixed bug with GROUP BY on NULL fields.
mysql-test/r/group_by.result:
  Fixed bug with GROUP BY on NULL fields.
mysql-test/t/group_by.test:
  Fixed bug with GROUP BY on NULL fields.
sql/sql_base.cc:
  Removed some not-needed comments in the trace file
sql/sql_select.cc:
  Fixed bug with GROUP BY on NULL fields.

Docs/manual.texi:
  Changelog
sql/gen_lex_hash.cc:
  Merged --verbose option from 4.0
  Smaller hash table
sql/slave.cc:
  Added some mutex locks to make SLAVE START and SLAVE STOP thread safe.

  Remove the complex typecast used to convert a HP-UX pthread struct to os_thread_id_t; the typecast seemed to work wrong in gcc-3.1 on HP-UX-10.20


innobase/os/os0thread.c:
  Remove the complex typecast used to convert a HP-UX pthread struct to os_thread_id_t; the typecast seemed to work wrong in gcc-3.1 on HP-UX-10.20

into geopard.online.bg:/storage/bk/mysql-3.23

include/sslopt-usage.h:
  Paul's patch for --ssl-xxx options for the --help message
scripts/mysqlhotcopy.sh:
  Paul's patch for --port option
BitKeeper/etc/logging_ok:
  Logging to logging@openlogging.org accepted

to 3.23

  Merge InnoDB-3.23.52c
ha_innobase.cc:
  Test the ref length sanity also in the production version


sql/ha_innobase.cc:
  Test the ref length sanity also in the production version
innobase/btr/btr0cur.c:
  Merge InnoDB-3.23.52c
innobase/buf/buf0buf.c:
  Merge InnoDB-3.23.52c
innobase/buf/buf0lru.c:
  Merge InnoDB-3.23.52c
innobase/ha/ha0ha.c:
  Merge InnoDB-3.23.52c
innobase/log/log0recv.c:
  Merge InnoDB-3.23.52c
innobase/mtr/mtr0log.c:
  Merge InnoDB-3.23.52c
innobase/os/os0file.c:
  Merge InnoDB-3.23.52c
innobase/page/page0cur.c:
  Merge InnoDB-3.23.52c
innobase/include/btr0btr.h:
  Merge InnoDB-3.23.52c
innobase/include/dyn0dyn.h:
  Merge InnoDB-3.23.52c
innobase/include/log0recv.h:
  Merge InnoDB-3.23.52c
innobase/include/buf0buf.ic:
  Merge InnoDB-3.23.52c
innobase/include/log0log.ic:
  Merge InnoDB-3.23.52c

Added back 'static inline', as not having this caused more problems than having it
Fixed wrong arguments to printf()


innobase/buf/buf0buf.c:
  Fixed wrong argument to printf()
innobase/fsp/fsp0fsp.c:
  Fixed wrong argument to printf()
innobase/include/btr0btr.h:
  Removed UNIV_INLINE for functions that should not be inlined
innobase/include/dyn0dyn.h:
  Removed UNIV_INLINE for functions that should not be inlined
innobase/include/univ.i:
  Added back 'static inline', as not having this caused more problems than having it

Removed some compiler warnings regarding 'static' from InnoDB


Docs/manual.texi:
  Fixed wrong manual entry
innobase/btr/btr0cur.c:
  Fixed wrong printf() string
innobase/include/univ.i:
  Removed compiler warnings (static inline produces a lot of warnings and is probably not portable)

Docs/manual.texi:
  SCCS merged

Docs/manual.texi:
  Changelog

  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log


sql/ha_innobase.h:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
sql/ha_innobase.cc:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/buf/buf0buf.c:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/dict/dict0dict.c:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/fil/fil0fil.c:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/fsp/fsp0fsp.c:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/ha/ha0ha.c:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/include/dict0dict.h:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/include/dict0mem.h:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/include/dyn0dyn.h:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/include/fsp0fsp.h:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/include/log0log.h:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/include/log0recv.h:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/include/trx0sys.h:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/include/trx0trx.h:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/include/log0log.ic:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/lock/lock0lock.c:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/log/log0log.c:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/log/log0recv.c:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/mem/mem0dbg.c:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/row/row0mysql.c:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/srv/srv0srv.c:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/srv/srv0start.c:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/trx/trx0sys.c:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log
innobase/trx/trx0trx.c:
  Merge InnoDB-3.23.52c; improve AUTO-INC algorithm with SHOW TABLE STATUS; new checksum in log

 - overhauled the RPM spec file (more macros, package descriptions)


support-files/mysql.server.sh:
   - Added LSB-compliant init script header
support-files/mysql.spec.sh:
   - Use some more macros (_mandir and _infodir)
   - Updated package description
   - Install binaries stripped to save disk space
   - Rearranged file list (make sure man pages are in
     the same package as the binaries)
   - clean up the BuildRoot directory afterwards
   - added mysqldumpslow to the server package

  Since MySQL commits the stmt always at the end of an INSERT, it is enough to release auto-inc lock at innobase_commit and innobase_rollback; add also more comments to code


sql/ha_innobase.cc:
  Since MySQL commits the stmt always at the end of an INSERT, it is enough to release auto-inc lock at innobase_commit and innobase_rollback; add also more comments to code