This ChangeSet will be null-merged to mysql-5.1.

Bugs fixed:
- Bug #21638:  InnoDB: possible segfault in page_copy_rec_list_end_no_locks
  If prefix_len is specified, write it to the insert buffer instead of type->len
- bug #10746:  InnoDB: Error: stored_select_lock_type is 0 inside ::start_stmt()!
  INSERT ... SELECT uses LOCK_NONE in innodb_locks_unsafe_for_binlog mode, so do
  not check for LOCK_X or LOCK_S in start_stmt()


innobase/ibuf/ibuf0ibuf.c:
  Applied innodb snapshot ss792 and ss854.
  
  ibuf_entry_build(): Write prefix_len to the insert buffer instead of type->len
  when prefix_len is specified.  Otherwise, btr_page_reorganize() during
  insert buffer merge would fail on ROW_FORMAT=COMPACT tables.  (Bug #21638)
innobase/include/data0type.h:
  Applied innodb snapshot ss792 and ss854.
  
  ibuf_entry_build(): Write prefix_len to the insert buffer instead of type->len
  when prefix_len is specified.  Otherwise, btr_page_reorganize() during
  insert buffer merge would fail on ROW_FORMAT=COMPACT tables.  (Bug #21638)
innobase/include/data0type.ic:
  Applied innodb snapshot ss792 and ss854.
  
  ibuf_entry_build(): Write prefix_len to the insert buffer instead of type->len
  when prefix_len is specified.  Otherwise, btr_page_reorganize() during
  insert buffer merge would fail on ROW_FORMAT=COMPACT tables.  (Bug #21638)
sql/ha_innodb.cc:
  Applied innodb snapshot ss792 and ss854.
  
  ha_innobase::start_stmt(): Remove the check for
  prebuilt->stored_select_lock_type being LOCK_X or LOCK_S.
  This would cause false alarms with INSERT ... SELECT, which would use
  LOCK_NONE in innodb_locks_unsafe_for_binlog mode.  (Bug #10746)

into  mysql.com:/usr/home/ram/work/5.0.b18267


zlib/Makefile.am:
  Auto merged

into  shellback.(none):/home/msvensson/mysql/mysql-5.0-maint

into  shellback.(none):/home/msvensson/mysql/mysql-5.0-maint

into  shellback.(none):/home/msvensson/mysql/mysql-5.0-maint

into  shellback.(none):/home/msvensson/mysql/work/my50-work


configure.in:
  Auto merged

into  shellback.(none):/home/msvensson/mysql/work/my50-work

into  shellback.(none):/home/msvensson/mysql/mysql-5.0-maint

into  shellback.(none):/home/msvensson/mysql/work/my50-work


configure.in:
  Auto merged

into  shellback.(none):/home/msvensson/mysql/mysql-5.0-maint

into  shellback.(none):/home/msvensson/mysql/work/my50-work

into  shellback.(none):/home/msvensson/mysql/mysql-5.0-maint


tests/Makefile.am:
  Auto merged

into  mysql.com:/home/hartmut/projects/mysql/dev/teamtrees/mysql-5.0-maint

into  shellback.(none):/home/msvensson/mysql/mysql-5.0-maint

into  shellback.(none):/home/msvensson/mysql/mysql-5.0-maint

extra/yassl/testsuite/test.hpp:
  Remove define for __MACH__
  http://bugs.debian.org/387369

into  zippy.cornsilk.net:/home/cmiller/work/mysql/mysql-5.0-maint

into  shellback.(none):/home/msvensson/mysql/mysql-5.0-maint

into  shellback.(none):/home/msvensson/mysql/mysql-5.0-maint


vio/viosocket.c:
  Auto merged

into  shellback.(none):/home/msvensson/mysql/mysql-5.0-maint

into  ted.mysql.internal:/home/ted/src/mysql/mysql-5.0-5.1-gca

Bug#21011 The [client] group header was overwritten by "password" option in upgrade_defaults file which confused mysqlcheck thus used to prevent the whole thing from running correctly when the "--password=" option is specified on mysql_upgrade command line with no corresponding "--user=" option


client/mysql_upgrade.c:
  The "password" option should be issued after group header to upgrade_defaults file whenever it is used

into  rolltop.ignatz42.dyndns.org:/mnt/storeage/mysql-5.0-maint

Bug#21424 mysqldump failing to export/import views.                                                                                                                                                                  

Dumps are created for the tables in each specified database then for the views in each specified database. This bug occurs when any database's views depend on the mysql database's table data while being restored. 
Added command line option --flush-privileges to the mysqldump utility which causes a FLUSH PRIVILIGES statement to be written to the dump after the mysql database.


client/mysqldump.c:
  When the flush-privileges command line option is specified, make sure to FLUSH PRIVILEGES after only the mysql database's tables' data is restored so that grant tables are current.
mysql-test/r/mysqldump.result:
  Added Results.
mysql-test/t/mysqldump.test:
  Expanded existing test case for bug 21527 to also test for bug 21424. 
  Exercises new --flush-priviliges command line option.
  Test ignores log tables.

into  rolltop.ignatz42.dyndns.org:/mnt/storeage/mysql-5.0-maint

BitKeeper/triggers/post-commit:
  Attempt to default to /usr/sbin/sendmail if it exists, otherwise use sendmail.

into  zippy.cornsilk.net:/home/cmiller/work/mysql/mysql-5.0-maint


client/mysql.cc:
  Auto merged
mysql-test/r/mysql.result:
  Manual merge.
mysql-test/t/mysql.test:
  Manual merge.

into  neptunus.(none):/home/msvensson/mysql/mysql-5.0-maint


mysql-test/mysql-test-run.pl:
  Auto merged

into  rolltop.ignatz42.dyndns.org:/mnt/storeage/mysql-5.0-maint


BitKeeper/etc/ignore:
  auto-union

BitKeeper/etc/ignore:
  Exclude files created during the Windows build process.
BitKeeper/triggers/post-commit:
  post-commit trigger writes emails to file in the BitKeeper/tmp directory before sending in case of failure.
  Now uses bk sendmail instead of hardcoded /usr/sbin/sendmail to send message.

into  polly.local:/home/kaa/src/maint/m50-maint--07OGt

The column's NOT NULL flag doesn't affect what we should print.  Remove the
wrong logic that does check it.

Also, verify that this and the previous two tests print the same data as 
other output formats.


client/mysql.cc:
  Don't check the column's NOT NULL flag when determining whether to print 
  "NULL" or not.
mysql-test/r/mysql.result:
  Add results and confirm that the other output forms give the same results
  for other similar tests.
mysql-test/t/mysql.test:
  Add regression tests for this bug and two similar bugs.

into  zippy.cornsilk.net:/home/cmiller/work/mysql/mysql-5.0-maint

into  zippy.cornsilk.net:/home/cmiller/work/mysql/mysql-5.0-maint


vio/viosocket.c:
  Auto merged

into  maint1.mysql.com:/data/localhome/cmiller/mysql-5.0-maint


sql/item_strfunc.cc:
  Auto merged

into  lamia.home:/home/timka/mysql/src/5.0-dbg


sql/sql_base.cc:
  Auto merged
sql/sql_parse.cc:
  Auto merged