pluse remove "first_batch_size" and instead put
batch_size on next_scan_req


ndb/include/kernel/signaldata/ScanFrag.hpp:
  Rename
  * concurrency -> batch_size_bytes
  * batch_byte_size -> batch_size_bytes
  
  Removed first_batch_size and
  added size parameters to SCAN_NEXTREQ
ndb/src/kernel/blocks/backup/Backup.cpp:
  Adopt to SCAN_FRAG_CHANGES
ndb/src/kernel/blocks/backup/Backup.hpp:
  Adopt to SCAN_FRAG_CHANGES
ndb/src/kernel/blocks/dblqh/Dblqh.hpp:
  Renamed
  scanCompletedOperations -> m_curr_batch_size_rows
  scanConcurrentOperations -> m_max_batch_size_rows
ndb/src/kernel/blocks/dblqh/DblqhMain.cpp:
  Renamed
  scanCompletedOperations -> m_curr_batch_size_rows
  scanConcurrentOperations -> m_max_batch_size_rows
ndb/src/kernel/blocks/dbtc/DbtcMain.cpp:
  Adopt to SCAN_FRAG_CHANGES
ndb/src/kernel/blocks/suma/Suma.cpp:
  Adopt to SCAN_FRAG_CHANGES

into mysql.com:/home/jonas/src/wl2025


ndb/src/common/mgmcommon/ConfigInfo.cpp:
  Auto merged
ndb/src/kernel/blocks/dblqh/DblqhMain.cpp:
  Auto merged
ndb/src/ndbapi/TransporterFacade.cpp:
  Auto merged

and max scan batch size
Removed some debug printouts

into mysql.com:/home/jonas/src/wl2025


BitKeeper/etc/logging_ok:
  auto-union
ndb/src/kernel/blocks/dblqh/Dblqh.hpp:
  Auto merged
ndb/src/kernel/blocks/dblqh/DblqhInit.cpp:
  Auto merged
ndb/src/kernel/blocks/dblqh/DblqhMain.cpp:
  Auto merged
ndb/src/kernel/blocks/dbtc/Dbtc.hpp:
  Auto merged
ndb/src/kernel/blocks/dbtc/DbtcMain.cpp:
  Auto merged
ndb/src/kernel/blocks/suma/Suma.cpp:
  Auto merged

into mysql.com:/home/jonas/src/mysql-4.1-ndb

into mysql.com:/home/jonas/src/mysql-4.1-ndb

into mysql.com:/home/jonas/src/mysql-4.1-ndb


ndb/src/ndbapi/TransporterFacade.cpp:
  Auto merged

into mysql.com:/dbdata/psergey/mysql-4.1-root

(old code mixed the 3 alternatives: native, [lo, hi], [hi, lo])

This fixes problem with auto increment on different byte order

(This also make adding of signed ints really simple...)


ndb/src/kernel/blocks/dbtup/DbtupExecQuery.cpp:
  Use native format in interpreter
  (old code mixed the 3 alternatives: native, [lo, hi], [hi, lo])
ndb/src/ndbapi/Ndb.cpp:
  Use native format in interpreter
  (old code mixed the 3 alternatives: native, [lo, hi], [hi, lo])
ndb/src/ndbapi/NdbOperationInt.cpp:
  Use native format in interpreter
  (old code mixed the 3 alternatives: native, [lo, hi], [hi, lo])

use of first_batch_size


ndb/include/kernel/ndb_limits.h:
  New maximum size
ndb/include/kernel/signaldata/ScanFrag.hpp:
  New error code
ndb/include/kernel/signaldata/ScanTab.hpp:
  Need to go to Uint16 when batch size > 255
ndb/src/kernel/blocks/dblqh/Dblqh.hpp:
  Uint8 => Uint16 when batch_size > 255
  New and changed methods for acc ptr's and
  checking end of scan batch (incl. check of batch_byte_size
ndb/src/kernel/blocks/dblqh/DblqhMain.cpp:
  Uint8 => Uint16 when batch_size > 255
  New and changed methods for acc ptr's and
  checking end of scan batch (incl. check of batch_byte_size

into mysql.com:/home/mysql_src/mysql-4.1-874

(specific to 4.1): don't put CLIENT_LOCAL_FILES in
CLIENT_CAPABILITIES; it would make mysql_options(CLIENT_LOCAL_FILES,0)
have no effect.


libmysql/client_settings.h:
  Don't enforce that we client have CLIENT_LOCAL_FILES;
  indeed this CLIENT_CAPABILITIES is used as minimum
  client options in mysql_real_connect(), which is *after*
  mysql_options(). Having CLIENT_LOCAL_FILES in
  CLIENT_CAPABILITIES defeats disabling it in mysql_options().

BitKeeper/etc/logging_ok:
  Logging to logging@openlogging.org accepted

size array.
Added a number of subroutines for this

into mysql.com:/home/jonas/src/mysql-4.1

ndb/include/ndb_global.h:
  Include my_alarm (to get signal.h)
ndb/src/cw/cpcd/Process.cpp:
  Include my_alarm (to get signal.h)
ndb/src/cw/cpcd/main.cpp:
  Include my_alarm (to get signal.h)
ndb/src/kernel/main.cpp:
  Include my_alarm (to get signal.h)
ndb/src/kernel/vm/Emulator.cpp:
  Include my_alarm (to get signal.h)
ndb/src/mgmclient/main.cpp:
  Include my_alarm (to get signal.h)
ndb/src/mgmsrv/Services.cpp:
  Include ndb_global instead of string.h
ndb/src/ndbapi/NdbOperationInt.cpp:
  Keep consistent order of 64 bits in interpreter
ndb/src/ndbapi/TransporterFacade.cpp:
  Include my_alarm (to get signal.h)

into avenger.(none):/export/brian/mysql/4.1

sql/examples/ha_tina.cc:
  Adding in pieces to remove from compile (aka ifdef).

into brandersnatch.localdomain:/home/dlenev/src/mysql-4.1-hpux

hpux11 compiler dislikes empty array initializers.

Limit testBackup BackupOne

ndb/test/ndbapi/testBackup.cpp:
  Drop table after testBackup
ndb/test/src/NdbBackup.cpp:
  Remove node id

ndb/src/ndbapi/TransporterFacade.cpp:
  Retry alloc of node id

mysql-test/r/range.result:
  Fix for bug#4488: more tests
mysql-test/t/range.test:
  Fix for bug#4488: more tests

mysql-test/r/ps_1general.result:
  Updated test results for addition of CSV engine.

into mysql.com:/home/mysql_src/mysql-4.1-874

var->save_result.ulong_value was 0 on Mac OS X probably due to positions
of members in the union. sys_var::check() only sets ulonglong_value


sql/set_var.cc:
  Of the save_result union, sys_var::check() only updates ulonglong_value;
  so other types of the union are garbage. Here we must use ulonglong_value
  in sys_var_sync_binlog_period::update(), not ulong_value (which is )

  Bug#4521: unique prefix keys, fix for HEAP HASH and HEAP BTREE case insensitive collations.


mysql-test/t/ctype_utf8.test:
  Bug#4521: unique prefix keys, fix for HEAP HASH and HEAP BTREE case insensitive collations.

into neptunus.(none):/home/magnus/mysql-4.1

sql-bench/server-cfg.sh:
  Remove the limitation for NDB+blobs

into mysql.com:/home/jonas/src/mysql-4.1

2) Fix variable scope for irix compiler


ndb/src/kernel/blocks/dblqh/DblqhMain.cpp:
  Remove part of require
ndb/test/src/NDBT_Tables.cpp:
  old compiler fix

ndb/src/kernel/blocks/dbtup/DbtupPagMan.cpp:
  Init page state
ndb/src/kernel/blocks/ndbcntr/NdbcntrMain.cpp:
  Init usr pointer for tcseize/releasereq
ndb/test/src/NDBT_Tables.cpp:
  More error printouts

into mysql.com:/home/jonas/src/mysql-4.1-ndb