metadata locks that were added at the MySQL level as part of the fix for
Bug#45225 Locking: hang if drop table with no timeout

inherited from the InnoDB Plugin and is not required anymore.

(different error number).

(different error number).

(different error number).

of the mysql server - data dictionaries do not go out of sync now.

(different error number).

imported InnoDB provides more values for "innodb_change_buffering".

InnoDB provides a few more config variables.

to mysql-test/

  ------------------------------------------------------------
  revno: 2661.581.1
  revision-id: sven.sandberg@sun.com-20090714193119-4693witmsqcaf28q
  parent: staale.smedseng@sun.com-20090615160325-miaxz8z9rjgm78h4
  committer: Sven Sandberg <sven.sandberg@sun.com>
  branch nick: 5.1-bugteam
  timestamp: Tue 2009-07-14 21:31:19 +0200
  message:
    BUG#39934: Slave stops for engine that only support row-based logging
  ...


  ------------------------------------------------------------
  revno: 3092.5.1
  revision-id: luis.soares@sun.com-20090924145252-8vvsnbvwo9l8v4vc
  parent: anurag.shekhar@sun.com-20090831075609-tkpqu41hxtupdeip
  committer: Luis Soares <luis.soares@sun.com>
  branch nick: mysql-5.1-bugteam
  timestamp: Thu 2009-09-24 15:52:52 +0100
  message:
    BUG#42829: binlogging enabled for all schemas regardless of
    binlog-db-db / binlog-ignore-db
  ...

when trying to build innodb as plugin.

The reason for the error is mismatch in mysql_temp_dir_list
declaration between mysqld.h and usage in ha_innodb.cc

Add missing MYSQL_PLUGIN_IMPORT to mysql_tmpdir_list
(variables exported by the server and used by plugin need it).

Fix tons of warnings about mismatch struct and class usage.

Tree cleaup after the last major merges in mysql-trunk:

The files sql/lex_hash.h and sql/sql_yacc.h are automatically
generated, and should not be checked in the configuration management system.
These files are now removed.
No changes are required for .bzrignore, which already listed these files
(and similar files in libmysqld/).

The file storage/perfschema/unittest/pfs_timer-t.cc did not build
after the header files refactoring affecting mysql_priv.h
The file now builds properly using sql_priv.h

assertions that were removed when introducing the flush list mutex,
and add comments that explain why these assertions are valid.

Discussed with Inaam Rana.

Adding my_global.h first in all files using
NO_EMBEDDED_ACCESS_CHECKS.

Correcting a merge problem resulting from a changed definition
of check_some_access compared to the original patches.

be excluded from performance schema instrumentation. Exclude
buffer block mutex/rwlock from performance schema instrumentation
by default.

Adding include to build on Windows.

Adding includes of my_global.h that seems to be necessary on Windows.

  ------------------------------------------------------------------------
  r6949 | mmakela | 2010-04-06 16:11:46 +0300 (Tue, 06 Apr 2010) | 7 lines

  branches/zip: Add debug checks to track down Issue #461.

  dict_table_check_for_dup_indexes(): Add the flag tmp_ok.  If !tmp_ok,
  check that no index name starts with TEMP_INDEX_PREFIX.

  ha_innobase::add_index(), ha_innobase::prepare_drop_index(),
  ha_innobase::final_drop_index(): Call dict_table_check_for_dup_indexes().
  ------------------------------------------------------------------------

Moving an extern "C" to not cover #includes.

- Adding include/mysql/innodb_priv.h which was missing before.
- Updated .bzrignore to ignore CMake generated cruft.

Revert r6931 because it introduced following bugs:
http://bugs.mysql.com/bug.php?id=52588
http://bugs.mysql.com/bug.php?id=52590

Minor fixes required because of changes since patches were developed.

Fixing merge problem that causes CSV engine to misbehave.

When a transaction joins we check if there are any other transactions
waiting on its locks. If there aren't any waiting then no deadlock can
occur.  This patch however has additional changes.

 1. Count leading zeros
 2. Count trailing zeros

There are two version of both these utility functions. One is hand
coded and the other will use the GCC builtin when available. The
changes to configure have yet to be made.

Simplify the next record lock fetch in the deadlock check code.

Pass the heap number as a parameter to the deadlock check code.

Written by: Sunny

changes to mtr_commit:

1) grab log mutex (and do initial mtr commit stuff)
for each dirty block
 2) grab flush_list mutex
 3) Insert into flush list
 4) release flush_list mutex
5) release log_sys mutex

Changed version:

1) grab log mutex (and do initial mtr commit stuff)
2) grab a new flushList order mutex
3) release log mutex
for each dirty block
 4) grab flush_list mutex
 5) Insert into flush list
 6) release flush_list mutex
7) Release new flush list order mutex 

Approved by: Marko has given the initial nod.

Non-functional change. Take our the start up message about ignoring
innodb_use_native_aio.

two new configuration parameters.

  1. innodb-purge-threads := [01] -- default is 0
  2. innodb-purge-batch-size := 20 ... 5000 -- default is 20

rb://271

This patch:

- Moves all definitions from the mysql_priv.h file into
  header files for the component where the variable is
  defined
- Creates header files if the component lacks one
- Eliminates all include directives from mysql_priv.h
- Eliminates all circular include cycles
- Rename time.cc to sql_time.cc
- Rename mysql_priv.h to sql_priv.h

Skip r6900, which was backported from branches/innodb+ 6899.
  ------------------------------------------------------------------------
  r6919 | mmakela | 2010-03-31 11:34:22 +0300 (Wed, 31 Mar 2010) | 54 lines
  Changed paths:
     M /branches/zip/ChangeLog
     M /branches/zip/handler/ha_innodb.cc
     M /branches/zip/mysql-test/innodb_bug38231.test
     A /branches/zip/mysql-test/innodb_bug51920.result
     A /branches/zip/mysql-test/innodb_bug51920.test
     M /branches/zip/row/row0sel.c
     M /branches/zip/srv/srv0srv.c

  branches/zip: Merge revisions 6788:6918 from branches/5.1:

    ------------------------------------------------------------------------
    r6822 | vasil | 2010-03-15 10:17:31 +0200 (Mon, 15 Mar 2010) | 12 lines
    Changed paths:
       M /branches/5.1/row/row0sel.c

    branches/5.1:

    Typecast to silence a compiler warning:

    row/row0sel.c: 4548
            C4244: '=' : conversion from 'float' to 'ib_ulonglong', possible loss of data
    row/row0sel.c: 4553
            C4244: '=' : conversion from 'double' to 'ib_ulonglong', possible loss of data

    Reported by:	Jonas Oreland <Jonas.Oreland@Sun.COM>
    Discussed with:	Sunny Bains <sunny.bains@oracle.com>
    ------------------------------------------------------------------------
    r6884 | vdimov | 2010-03-26 13:05:03 +0200 (Fri, 26 Mar 2010) | 6 lines
    Changed paths:
       M /branches/5.1/mysql-test/innodb_bug38231.test

    branches/5.1:

    Fix a non-determinism in innodb_bug38231.

    Reported by:	Sergey Vojtovich <svoj@Sun.COM>
    ------------------------------------------------------------------------
    r6911 | vdimov | 2010-03-30 11:39:02 +0300 (Tue, 30 Mar 2010) | 2 lines
    Changed paths:
       M /branches/5.1/handler/ha_innodb.cc

    branches/5.1: Whitespace fixup
    ------------------------------------------------------------------------
    r6912 | vdimov | 2010-03-30 12:18:46 +0300 (Tue, 30 Mar 2010) | 2 lines
    Changed paths:
       M /branches/5.1/handler/ha_innodb.cc

    branches/5.1: Whitespace fixup on line 354
    ------------------------------------------------------------------------
    r6918 | mmakela | 2010-03-31 11:14:51 +0300 (Wed, 31 Mar 2010) | 6 lines
    Changed paths:
       A /branches/5.1/mysql-test/innodb_bug51920.result
       A /branches/5.1/mysql-test/innodb_bug51920.test
       M /branches/5.1/srv/srv0srv.c

    branches/5.1: Obey KILL during a lock wait (Bug #51920).

    srv_suspend_mysql_thread(), srv_lock_timeout_and_monitor_thread():
    Check trx_is_interrupted() in addition to checking the lock wait timeout.

    rb://279 approved by Sunny Bains
    ------------------------------------------------------------------------
  ------------------------------------------------------------------------
  r6920 | mmakela | 2010-03-31 11:49:08 +0300 (Wed, 31 Mar 2010) | 1 line
  Changed paths:
     M /branches/zip/row/row0sel.c

  branches/zip: Fix a compilation error that sneaked in in r6919.
  ------------------------------------------------------------------------
  r6922 | mmakela | 2010-03-31 14:54:30 +0300 (Wed, 31 Mar 2010) | 11 lines
  Changed paths:
     M /branches/zip/mysql-test/innodb_bug51920.result
     M /branches/zip/mysql-test/innodb_bug51920.test

  branches/zip: Merge revisions 6918:6921 from branches/5.1:

    ------------------------------------------------------------------------
    r6921 | mmakela | 2010-03-31 14:33:04 +0300 (Wed, 31 Mar 2010) | 2 lines
    Changed paths:
       M /branches/5.1/mysql-test/innodb_bug51920.result
       M /branches/5.1/mysql-test/innodb_bug51920.test

    branches/5.1: innodb_bug51920.test: Make the test quicker and more
    deterministic.  Suggested by Vasil Dimov.
    ------------------------------------------------------------------------
  ------------------------------------------------------------------------
  r6925 | mmakela | 2010-03-31 15:30:56 +0300 (Wed, 31 Mar 2010) | 9 lines

  branches/zip: Merge revisions 6921:6924 from branches/5.1:

    ------------------------------------------------------------------------
    r6924 | mmakela | 2010-03-31 15:28:25 +0300 (Wed, 31 Mar 2010) | 1 line
    Changed paths:
       M /branches/5.1/mysql-test/innodb_bug51920.test

    branches/5.1: innodb_bug51920.test: Fix a race condition.
    ------------------------------------------------------------------------
  ------------------------------------------------------------------------

the upper limit of total rollback segments from 256 to 128. This is because
we can't use the sign bit. It has not caused problems in the past because we
only created one segment. InnoDB has always had the capability to use the
additional rollback segments therefore this patch is backwards compatible.
The only requirement to maintain backward compatibility has been to ensure
that the additional segments are created after the double write buffer.
This is to avoid breaking assumptions in the existing code.

Fix Bug#26590 MySQL does not allow more than 1023 open transactions

and rearrange some code in order to eliminate an unwarranted
"possibly uninitialized" warning that was reported by Sunny.

rb://276 approved by Inaam Rana

Don't allow master thread to flush pages if it has been less then
a second since last iteration.

Approved by: Marko