Changes needed for ibbackup directory scanning fault tolerance

  Fix the OS error 2 reported by Miguel and Mark in Windows crash recovery: a * had been forgotten from the path in directory scanning
ha_innodb.cc:
  Set default directory in fil0fil.c right if we are running the Embedded Server Library, where the default dir of the process is not necessarily the MySQL datadir

  Merge with ibbackup; bug fix: .ibd files were extended 2 x the required amount; InnoDB does not create the small file inno_arch_log... any more at database creation

  Check that writes to data files always happen in to addresses divisible by 16 kB, and the chunk size is also divisible by 16 kB; a user reported 2 corrupt pages from Linux-2.4.20 where an index page seemed displaced

  Check that writes to data files always happen in to addresses divisible by 16 kB, and the chunk size is also divisible by 16 kB; a user reported 2 corrupt pages from Linux-2.4.20 where an index page seemed displaced

  Fix Visual C++ compilation errors reported by Miguel

  IMPORT TABLESPACE must reset lsns if they are too high
ha_innodb.cc:
  DISCARD/IMPORT TABLESPACE must have a TL_WRITE lock on the table

  Multiple tablespaces for InnoDB
sql_table.cc:
  Tell explicitly that InnoDB should retrieve all columns in CHECKSUM TABLE
sql_update.cc, sql_select.cc, my_base.h:
  More descriptive flag name HA_EXTRA_RETRIEVE_ALL_COLS

  strerror() does not work in Windows

  Merge
trx0trx.c:
  Print more info about a trx in SHOW INNODB status; try to find the bug reported by Plaxo
buf0buf.c:
  Check that page log sequence numbers are not in the future
log0recv.c, log0log.c:
  Fixed a bug: if you used big BLOBs, and your log files were relatively small, InnoDB could in a big BLOB operation temporarily write over the log produced AFTER the latest checkpoint. If InnoDB would crash at that moment, then the crash recovery would fail, because InnoDB would not be able to scan the log even up to the latest checkpoint. Starting from this version, InnoDB tries to ensure the latest checkpoint is young enough. If that is not possible, InnoDB prints a warning to the .err log

  Allow also O_DIRECT as innodb_flush_method; it only affects writing to data files

  Merge InnoDB-4.0.14: SAVEPOINT now implemented; InnoDB now accepts also column prefix keys; crashing bug in ON UPDATE CASCADE fixed; page checksum formula fixed

  Prevent on Windows starting of two mysqld instances on SAME InnoDB files: that could cause severe database corruption

  Put back Monty's patch which the previous push accidentally erased: print also operation type in os_file_handle_error()

  Release all event semaphores at shutdown also in Windows
srv0start.c, srv0srv.c:
  make test sometimes failed because lock timeout thread exited without decrementing the InnoDB thread counter

  Exit all threads created by innoDB at shutdown

  Print progress information if at startup InnoDB creates and writes bigger than 100 MB data file or log file

  Eliminate the LOCK_log bottleneck in group commit in binlogging: flush InnoDB log files only after it has been released

  Merge InnoDB-4.0.13; DROP FOREIGN KEY now works

  Fix bug in previous push

  Print and error message also if a read from file results in the Linux error EEXIST; may happen if the disk is broken

  The problem yesterday in ERROR_LOCK_VIOLATION was not InnoDB Hot Backup, but some file system backup tool: add to file writes 100 retries with 1 second waits

  Fix an assertion about ERROR_LOCK_VIOLATION 33 in file write if InnoDB Hot Backup is run concurrently with mysqld in Windows

  Reduce memory usage of the buffer headers
Many files:
  Merge InnoDB-4.1 with AWE support

Added rename table for BDB. This fixes a bug in ALTER TABLE with BDB tables.

  Merge InnoDB-4.0.5: new isolation levels READ COMMITTED and READ UNCOMMITTED now supported, selective deadlock resolution
mysqld.cc:
  Change MySQL default isolation level to REPEATABLE READ; note that InnoDB has always had that default, and BDB and MyISAM always run at SERIALIZABLE level anyway

  SHOW INNODB STATUS always showed average bytes read as 0 in Unix

  Modifications for query cache + trxs, fix of q.c.+ foreign keys
os0file.c:
  Use unbuffered i/o in Windows

  Change srv_flush_log_at_trx_commit to ulint, note that ibool is defined as ulint, so this is purely formal change
os0file.c:
  Start using unbuffered i/o again in Windows because sequential read using normal i/o was 4 times slower in XP

  Fix a typo: should be buf_page_print

  Put 'last millisecond' test to make sure page checksum errors do not result from memory corruption in InnoDB
os0thread.c, configure.in:
  Make sure stack size in InnoDB threads in AIX is at least 32 kB

  Merge InnoDB-3.23.52d

  Merge InnoDB-3.23.52c
ha_innobase.cc:
  Test the ref length sanity also in the production version

  Fix a bug in the previous commit change and align also the buffer in simulataled aio for possible use in i/o to raw devices

  Align the buffer used in initing a data file to zero; this may be needed if the data file is actually a raw device

  Merge InnoDB-3.23.52b

  Crash the server if file flush does not succeed: the database can get corrupt, better to crash it than let it run

  Merge 3.23.52

  Merge InnoDB-3.23.50