Final-review fixes per Monty, pre-push. OK'd for 
push. Please see each file's comments.



mysql-test/r/federated.result:
  BUG #19773
      
  Results for multi-table deletes, updates
mysql-test/t/federated.test:
  BUG #19773
      
  Test multi table update and delete. Added drop table to end of previous test.
sql/ha_federated.cc:
  BUG #19773 
      
  Post-review changes, per Monty. 3rd patch, OK'd for push.
  - Added index_read_idx_with_result_set, which uses the result set passed to it
  - Hash by entire connection scheme
  - Protected store_result result set for table scan by adding a method result set
    to index_read_idx and index_read which is passed to index_read_with_result, which
    in turn iterates over the single record via read_next.
    This is a change from having two result sets in the first two patches. 
    This keeps the code clean and avoids the need for yet another result set.
  - Rewrote ::position and ::rnd_pos to store position - if primary key use 
    primary key, if not, use record buffer.
  - Rewrote get_share to store hash with connect string vs. table name
  - delete_row added subtration of "records" by affected->rows
  - Added read_next to handle what rnd_next used to do (converting raw record
     to query and vice versa)
  - Removed many DBUG_PRINT lines
  - Removed memset initialisation since subsequent loop accomplishes
  - Removed un-necessary mysql_free_result lines
sql/ha_federated.h:
   BUG #19773
      
      Fixed "SET " to " SET " to make sure built statements are built with 
      "UPDATE `t1` SET .." instead of "UPDATE `t1`SET"

into  xiphis.org:/home/antony/work2/p4-bug12096.2


sql/mysqld.cc:
  Auto merged

include/my_global.h:
  Remove cast to ssize_t, since there is no ssize_t type on Windows.

            read buffer
Setting read buffer to values greater than SSIZE_MAX results in
unexpected behavior.

According to read(2) manual:
If count is greater than SSIZE_MAX, the result is unspecified.

Set upper limit for read_buffer_size and read_rnd_buffer_size to
SSIZE_MAX.


include/my_global.h:
  Define SSIZE_MAX if not defined.
sql/mysqld.cc:
  Set upper limit for read_buffer_size and read_rnd_buffer_size to
  SSIZE_MAX.

into  mysql.com:/home/mydev/mysql-5.0-bug19604

CHECK TABLE did temporarily clear the auto_increment value.
It runs with a read lock, allowing other readers and
conurrent INSERTs. The latter could grab the wrong value
in this moment.

CHECK TABLE does no longer modify the auto_increment value.
Not even for a short moment.


myisam/mi_check.c:
  Bug#19604 - CHECK TABLE with concurrent INSERT can reset auto_increment
  In chk_key() and update_auto_increment_key() in the repair_only
  case, do not touch info->s->state.auto_increment. Especially
  chk_key() can be called from CHECK TABLE with a read lock.
  Concurrent inserts could grab a temporarily changed value.
  Added minor style fixes.
myisam/mi_key.c:
  Bug#19604 - CHECK TABLE with concurrent INSERT can reset auto_increment
  Changed update_auto_increment() to retrieve_auto_increment()
  to reflect that it does not change the auto_increment by
  itself any more. This must now be done externally if needed.
myisam/mi_update.c:
  Bug#19604 - CHECK TABLE with concurrent INSERT can reset auto_increment
  Added explicit update of info->s->state.auto_increment
  after the change from update_auto_increment() to
  retrieve_auto_increment().
myisam/mi_write.c:
  Bug#19604 - CHECK TABLE with concurrent INSERT can reset auto_increment
  Added explicit update of info->s->state.auto_increment
  after the change from update_auto_increment() to
  retrieve_auto_increment().
myisam/myisamdef.h:
  Bug#19604 - CHECK TABLE with concurrent INSERT can reset auto_increment
  Changed update_auto_increment() to retrieve_auto_increment()
  to reflect that it does not change the auto_increment by
  itself any more. This must now be done externally if needed.

  "Add line for non-executable stack in .s files"
  Fix so that configure will use "--noexecstack" for assembler if gcc supports
  option and compiled C doesn't need executable stack.


config/ac-macros/compiler_flag.m4:
  Bug#12096
    Add macro to check if "--noexecstack" should be used when compiling assembler
configure.in:
  Bug#12096
    Add macro to check if "--noexecstack" should be used when compiling assembler
strings/Makefile.am:
  Bug#12096
    Automake knows how to handle assembler

into  xiphis.org:/home/antony/work2/p4-bug12096.1

  "Merge table does not work with bit types"
   MERGE should have HA_CAN_BIT_FIELD feature bit set or else table row is
   formatted incorrectly.


mysql-test/r/merge.result:
  Bug#19648
    Test for fix
mysql-test/t/merge.test:
  Bug#19648
    Test for fix
sql/ha_myisammrg.h:
  Bug#19648
    Must have HA_CAN_BIT_FIELD for BIT type support

The order of acquiring LOCK_mysql_create_db
and wait_if_global_read_lock() was wrong. It could happen
that a thread held LOCK_mysql_create_db while waiting for
the global read lock to be released. The thread with the
global read lock could try to administrate a database too.
It would first try to lock LOCK_mysql_create_db and hang...

The check if the current thread has the global read lock
is done in wait_if_global_read_lock(), which could not be
reached because of the hang in LOCK_mysql_create_db.

Now I exchanged the order of acquiring LOCK_mysql_create_db
and wait_if_global_read_lock(). This makes 
wait_if_global_read_lock() fail with an error message for
the thread with the global read lock. No deadlock happens.


mysql-test/r/lock_multi.result:
  Bug#19815 - CREATE/RENAME/DROP DATABASE can deadlock on a global read lock
  The test result
mysql-test/t/lock_multi.test:
  Bug#19815 - CREATE/RENAME/DROP DATABASE can deadlock on a global read lock
  The test case
sql/sql_db.cc:
  Bug#19815 - CREATE/RENAME/DROP DATABASE can deadlock on a global read lock
  Exchanged the order of acquiring LOCK_mysql_create_db
  and wait_if_global_read_lock().

into moonbone.local:/work/19077-bug-5.0-mysql

into  april.(none):/home/svoj/devel/mysql/BUG18233/mysql-5.0

The convert_constant_item() function converts constant items to ints on
prepare phase to optimize execution speed. In this case it tries to evaluate
subselect which contains a derived table and is contained in a derived table. 
All derived tables are filled only after all derived tables are prepared.
So evaluation of subselect with derived table at the prepare phase will
return a wrong result.

A new flag with_subselect is added to the Item class. It indicates that
expression which this item represents is a subselect or contains a subselect.
It is set to 0 by default. It is set to 1 in the Item_subselect constructor
for subselects.
For Item_func and Item_cond derived classes it is set after fixing any argument
in Item_func::fix_fields() and Item_cond::fix_fields accordingly.
The convert_constant_item() function now doesn't convert a constant item
if the with_subselect flag set in it. 


mysql-test/t/view.test:
  Added test case for bug#19077: A nested materialized derived table is used before being populated.
mysql-test/t/subselect.test:
  Added test case for bug#19077: A nested materialized derived table is used before being populated.
mysql-test/r/view.result:
  Added test case for bug#19077: A nested materialized derived table is used before being populated.
mysql-test/r/subselect.result:
  Added test case for bug#19077: A nested materialized derived table is used before being populated.
sql/item_subselect.cc:
  Fixed bug#19077: A nested materialized derived table is used before being populated.
  The Item_subselect class constructor sets new with_subselect flag to 1.
sql/item_func.cc:
  Fixed bug#19077: A nested materialized derived table is used before being populated.
  
  The Item_func::fix_fields() sets new with_subselect flag from with_subselect flags of its arguments.
sql/item_cmpfunc.cc:
  Fixed bug#19077: A nested materialized derived table is used before being populated.
  The convert_constant_item() function now doesn't convert a constant item
  with the with_subselect flag set.
  The Item_cond::fix_fields() sets new with_subselect flag from with_subselect flags of its arguments.
sql/item.cc:
  Fixed bug#19077: A nested materialized derived table is used before being populated.
  Set new with_subselect flag to default value - 0 in the Item constructor.
sql/item.h:
  Fixed bug#19077: A nested materialized derived table is used before being populated.
  A new flag with_subselect is added to the Item class. It indicates that
  expression which this item represents is a subselect or contains a subselect.
  It is set to 0 by default.

Fixes bug#19040 "yaSSL does not compile on AIX".


extra/yassl/taocrypt/include/runtime.hpp:
  Include "config.h" (if it exists) in all yaSSL files. This is needed to ensure the same 
  configure settings are used for yaSSL as for the other modules linked together.
  Example: the settings for "large file" on AIX.
  Fixes bug#19040 "yaSSL does not compile on AIX".

into  neptunus.(none):/home/msvensson/mysql/bug18818/my50-bug18818

into  poseidon.ndb.mysql.com:/home/tomas/mysql-5.0-main

into  poseidon.ndb.mysql.com:/home/tomas/mysql-5.0-main

into  poseidon.ndb.mysql.com:/home/tomas/mysql-5.0


ndb/src/mgmapi/mgmapi.cpp:
  Auto merged
ndb/tools/waiter.cpp:
  Auto merged

    ndb: ndb_waiter disconnect/reconnect on get status error


ndb/src/mgmapi/mgmapi.cpp:
  ndb: added some missing error codes to mgmapi
ndb/tools/waiter.cpp:
  ndb: ndb_waiter disconnect/reconnect on get status error

into  production.mysql.com:/usersnfs/tulin/mysql-5.0

into  mysql.com:/usr_rh9/home/elkin.rh9/MySQL/Merge/5.0


mysql-test/t/rpl_temporary.test:
  Auto merged
ndb/src/kernel/blocks/dbtup/DbtupExecQuery.cpp:
  Auto merged
ndb/test/ndbapi/Makefile.am:
  Auto merged

into  mysql.com:/opt/local/work/mysql-5.0-runtime-merge

fixing encoding example because of table names can not be in koi8r 
on some platforms.


mysql-test/t/rpl_temporary.test:
  koi8r encoding to name tables is changed for latin1 for the former does not work
  on some platforms.
  The effect of the test remains: a test table is named to have multibyte utf8 representation,
  which is the essence of the bug.

into mysql.com:/home/psergey/mysql-5.0-bug19618

into  poseidon.ndb.mysql.com:/home/tomas/mysql-5.0-main

into  mysql.com:/opt/local/work/mysql-5.0-runtime-merge

into  poseidon.ndb.mysql.com:/home/tomas/mysql-5.0


ndb/src/mgmsrv/MgmtSrvr.cpp:
  Auto merged
ndb/src/ndbapi/SignalSender.cpp:
  Auto merged
support-files/mysql.spec.sh:
  Auto merged

- not having lock can cause "block number list" to become corrupt => crash
- also may result in 2 threads receiving same block no => possble cause of
Bug #13987 Cluster: Loss of data nodes can cause high CPU usage from ndb_mgmd


ndb/src/mgmsrv/MgmtSrvr.cpp:
  ndb: added missing lock of mutex

into mysql.com:/home/psergey/mysql-5.0-bug19618

into  poseidon.ndb.mysql.com:/home/tomas/mysql-4.1


support-files/mysql.spec.sh:
  Auto merged

into  poseidon.ndb.mysql.com:/home/tomas/mysql-5.0


support-files/mysql.spec.sh:
  Auto merged

into  mysql.com:/data0/knielsen/tmp-5.0

mysql-test/r/rpl_temporary.result:
  Fixed result file

into  mysql.com:/opt/local/work/mysql-5.0-runtime-merge


sql/item.cc:
  Auto merged
sql/item.h:
  Auto merged

into  mysql.com:/data0/knielsen/tmp-5.0

into  mysql.com:/users/lthalmann/bk/mysql-5.0-rpl