When statement to be prepared contained CREATE PROCEDURE, CREATE FUNCTION
or CREATE TRIGGER statements with a syntax error in it, the preparation
would fail with syntax error message, but the memory could be corrupted.

The problem occurred because we switch memroot when parse stored
routine or trigger definitions, and on parse error we restored the
original memroot only after performing some memory operations.  In more
detail:
 - prepared statement would activate its own memory root to parse
   the definition of the stored procedure.
 - SP would reset this memory root with its own memory root to
   parse SP statements
 - a syntax error would happen
 - prepared statement would restore the original memory root
 - stored procedure would restore what it thinks was the original
   memory root, but actually was the statement memory root.
That led to double free - in destruction of the statement and in
a next call to mysql_parse().

The solution is to restore memroot right after the failed parsing.


mysql-test/r/ps.result:
  Add result for bug#21856: Prepared Statements: crash if bad create.
mysql-test/t/ps.test:
  Add test case for bug#21856: Prepared Statements: crash if bad create.
sql/sql_parse.cc:
  On parse error if thd->lex->sphead is set we have to free sp_head object
  to restore statement memroot, if it was switched during parsing.
  
  The change here is for safety, currently query_cache_abort() and
  lex->unit.cleanup() calls do not use current memroot.
sql/sql_prepare.cc:
  On parse error if thd->lex->sphead is set we have to free sp_head object
  to restore statement memroot, if it was switched during parsing.

into  alik.:/mnt/raid/alik/MySQL/devel/5.0-rt

into  perch.ndb.mysql.com:/home/jonas/src/mysql-5.0-ndb

into  alik.:/mnt/raid/alik/MySQL/devel/5.0-rt


mysql-test/lib/mtr_process.pl:
  Auto merged

into  perch.ndb.mysql.com:/home/jonas/src/mysql-5.0-ndb


sql/log.cc:
  Auto merged
sql/sql_class.h:
  Auto merged

into  perch.ndb.mysql.com:/home/jonas/src/50-work


sql/log.cc:
  Auto merged
sql/sql_class.h:
  Auto merged
sql/sql_repl.cc:
  merge

  fix deadlock if master switches log file in parallell with "show master logs"
  


sql/log.cc:
  add raw_get_current_log which dont take log mutex
sql/sql_class.h:
  add raw_get_current_log which dont take log mutex
sql/sql_repl.cc:
  Fix lock order i.e first log_lock then index_lock

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


sql/sql_select.cc:
  Auto merged

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

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


sql/sql_lex.cc:
  Auto merged
sql/sql_lex.h:
  Auto merged
sql/sql_yacc.yy:
  Auto merged

Fix problem where mysql-test-run.pl fails to start up the mysqld after a failed test. Shows up on win hosts where one failed test case make subsequent ones fails with error "Could not open connection 'default': 2003 Can't connect to MySQL server on 'localhost'" 


mysql-test/lib/mtr_process.pl:
  Reset the associated "pid" of masters and slave when building the list of processes to shutdown.

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

into  maint1.mysql.com:/data/localhome/tnurnberg/my50-21913


mysql-test/r/func_time.result:
  Auto merged
mysql-test/t/func_time.test:
  Auto merged

into  salvation.intern.azundris.com:/home/tnurnberg/21913/my50-21913

21913: DATE_FORMAT() Crashes mysql server if I use it through mysql-connector-j driver.

Variable character_set_results can legally be NULL (for "no conversion.")
This could result in a NULL deref that crashed the server.  Fixed.

(Although ran some additional precursory tests to see whether I could break
anything else, but no breakage so far.)


mysql-test/r/func_time.result:
  Bug#21913: DATE_FORMAT() Crashes mysql server if I use it through mysql-connector-j driver.
  
  Prove DATE_FORMAT() no longer crashes the server when character_set_results is
  NULL (which is a legal value and means, "no conversion").
mysql-test/t/func_time.test:
  Bug#21913: DATE_FORMAT() Crashes mysql server if I use it through mysql-connector-j driver.
  
  Prove DATE_FORMAT() no longer crashes the server when character_set_results is
  NULL (which is a legal value and means, "no conversion").
sql/sql_string.cc:
  Bug#21913: DATE_FORMAT() Crashes mysql server if I use it through mysql-connector-j driver.
  
  Avoid NULL deref in my_charset_same() -- if !to_cs, we won't need to compare
  because it is magic for, "no conversion."

Variable character_set_results can legally be NULL (for "no conversion.")
This could result in a NULL deref that crashed the server.  Fixed.

(Although ran some additional precursory tests to see whether I could break
anything else, but no breakage so far.)


mysql-test/r/func_time.result:
  Bug#21913: DATE_FORMAT() Crashes mysql server if I use it through mysql-connector-j driver.
  
  Prove DATE_FORMAT() no longer crashes the server when character_set_results is
  NULL (which is a legal value and means, "no conversion").
mysql-test/t/func_time.test:
  Bug#21913: DATE_FORMAT() Crashes mysql server if I use it through mysql-connector-j driver.
  
  Prove DATE_FORMAT() no longer crashes the server when character_set_results is
  NULL (which is a legal value and means, "no conversion").
sql/sql_string.cc:
  Bug#21913: DATE_FORMAT() Crashes mysql server if I use it through mysql-connector-j driver.
  
  Avoid NULL deref in my_charset_same() -- if !to_cs, we won't need to compare
  because it is magic for, "no conversion."

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

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

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

into  bk-internal.mysql.com:/data0/bk/mysql-5.0-cmake

into  maint2.mysql.com:/data/localhome/tsmith/bk/50

into  bk-internal.mysql.com:/data0/bk/mysql-5.0-cmake

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

  Provide a more extensible, easier-to-change way of reordering
  test cases.


mysql-test/lib/mtr_cases.pl:
  Provide a more extensible, easier-to-change way of reordering
  test cases.

start instance; kill mysqlmanager; show ...

The problem was that Instance Manager didn't close client
sockets (sockets for client connections) on execing mysqld
instance. So, mysqld-instance inherits these descriptors.

The fix is to set close-on-exec flag for each client socket.


mysql-test/r/im_daemon_life_cycle.result:
  Updated result file.
mysql-test/t/im_daemon_life_cycle.imtest:
  Test for BUG#12751.
server-tools/instance-manager/listener.cc:
  Set close-on-exec flag for each client socket.

configure.in:
  added win subdir

scripts/make_win_bin_dist:
  BitKeeper file /home/georg/work/mysql/prod/mysql-5.0-win/scripts/make_win_bin_dist
client/mysqlbinlog.cc:
  Fix for cmake build: Cmake doesn't use the VC++ files
extra/comp_err.c:
  fixed windows crash (debug): We can't call DBUG_RETURN after my_end.
include/my_dbug.h:
  added missing empty define for DBUG_LEAVE to prevent precompiler errors when
  compiling in non_debug mode
mysql-test/mysql-test-run.pl:
  Added support for new cmake release and debug paths.
mysql-test/t/system_mysql_db_fix.test:
  This test requires unix shell script mysql_fix_previleges_tables -> skip under windows
mysys/my_seek.c:
  Fix for windows debug crash. However this solution is bad: we should never
  call lseek with an invalid file pointer.
sql/ha_archive.cc:
  Fixed windows crash: We need dup in gzdopen to keep the filehandle open,
  otherwise subsequent calls to mysql_close will fail/crash.

into  dl145s.mysql.com:/data/tkatchaounov/5.0-bug-21787


mysql-test/r/limit.result:
  Auto merged
mysql-test/t/limit.test:
  Auto merged
sql/sql_select.cc:
  Adjust the fix for BUG#21787 for 5.0

  
The problem was due to a prior fix for BUG 9676, which limited
the rows stored in a temporary table to the LIMIT clause. This
optimization is not applicable to non-group queries with aggregate
functions. The fix disables the optimization in this case.


mysql-test/r/limit.result:
  Test case for BUG#21787
mysql-test/t/limit.test:
  Test case for BUG#21787
sql/sql_select.cc:
  If there is an aggregate function in a non-group query,
  materialize all rows in the temporary table no matter if
  there is a LIMIT clause. This is necessary, since the
  aggregate functions must be computed over all result rows,
  not just the first LIMIT rows.

client/mysqldump.c:
  fflush stderr after writing to it.

into  mysql.com:/home/psergey/mysql-5.0-bug21477-look-64bit

make st_select_lex::setup_ref_array() take into account that 
Item_sum-descendant objects located within descendant SELECTs
may be added into ref_pointer_array.


sql/item_sum.cc:
  BUG#21477 "memory overruns for certain kinds of subqueries":
  Make SELECT_LEX::n_sum_items contain # of Item_sum-derived objects
  that exist within this SELECT.
sql/sql_lex.h:
  BUG#21477 "memory overruns for certain kinds of subqueries":
  Add SELECT_LEX::n_sum_items and SELECT_LEXT::n_child_sum_items.
sql/sql_yacc.yy:
  BUG#21477 "memory overruns for certain kinds of subqueries":
  Make SELECT_LEX::n_sum_items contain # of Item_sum-derived objects
  that exist within this SELECT.

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

libmysql/mytest.c:
  Rename: BitKeeper/deleted/.del-mytest.c -> libmysql/mytest.c
sql/message.mc:
  BitKeeper file /home/georg/work/mysql/prod/mysql-5.0-win/sql/message.mc

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

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


BitKeeper/etc/ignore:
  auto-union
mysql-test/Makefile.am:
  Manual merge 4.1 -> 5.0

BitKeeper/etc/ignore:
  Added mysql-test/mtr to the ignore list

into  perch.ndb.mysql.com:/home/jonas/src/mysql-5.0-ndb