Port a change from MySQL.

Bug#8841 - CHECKSUM TABLE is broken in MyISAM
Fixed test results.

WL#1563 - Modify MySQL to support fast CREATE/DROP INDEX
Change "duplicate key" message to print key name
instead of key number.

testing uninitialized variables.  (Bug #17405)

chain with an equivalent switch block.

os_file_create_tmpfile(): Replace the UNIV_HOTBACKUP specific
implementation with a run-time assertion failure.

Port r146 from branches/5.0:

Make innodb_flush_log_at_trx_commit a settable global variable.

fil_extend_space_to_desired_size(): in UNIV_HOTBACKUP builds,
do not touch srv_data_file_sizes[] or srv_n_data_files.

cascading update.

Fix bug #17126, CHECK TABLE blocking other queries, by releasing the
btr_search_latch periodically during the adaptive hash table validation.

Fix bug #16827, better error message if ibdata files omitted from my.cnf.

LAST FOREIGN KEY ERROR section in SHOW INNODB STATUS output (Bug #16814).

dict_foreign_error_report(): Always print a newline after invoking
dict_print_info_on_foreign_key_in_create_format().

innodb_cache.test: save and restore query_cache_size

and other systems whose compiler pretends to be GCC 2.

univ.i: Outside __WIN__, define UNIV_INLINE as static inline.

Finalize storage engine plugins

ha_innodb.cc: remove unwanted handlerton entries; changes for show status

WL#2935 - SHOW STATUS support in plugins

ha_innodb.cc, ha_innodb.h: s/struct show_var_st/SHOW_VAR/

Remove the declaration of innodb_export_status(), as it will
from now on be invoked via a function pointer in innobase_hton.

WL#1324 table name to file name encoding

ha_innobase::create(): do not remove .frm extension

row_is_mysql_tmp_table_name(): replace '#' with '@0023'

that were not part of r154.  (Bug #3300)

innodb.test, innodb.result: Remove binlog related tests, because they
are not InnoDB specific; they merely require a transactional table.

ha_innobase::ha_innobase(): Add HA_PRIMARY_KEY_ALLOW_RANDOM_ACCESS
to table_flags.

ha_innobase::rnd_init(): Disable semi-consistent read for random access.

NetWare specific change to increase thread stack size.

of breaking serializability.  (Bug #3300)

ha_innobase::unlock_row(): reset the "did semi consistent read" flag

ha_innobase::was_semi_consistent_read(),
ha_innobase::try_semi_consistent_read(): new methods

row_prebuilt_t, row_create_prebuilt(): add field row_read_type for
keeping track of semi-consistent reads

row_vers_build_for_semi_consistent_read(),
row_sel_build_committed_vers_for_mysql(): new functions

row_search_for_mysql(): implement semi-consistent reads

adaptive hash index for a B-tree page.  (Bug #16582)

btr_search_drop_page_hash_index(): Retry the operation if a hash index
with different parameters was built meanwhile.  Add diagnostics for the
case that hash node pointers to the page remain.  This fix is from Heikki.

btr_search_info_update_hash(), btr_search_info_update_slow():
Document the parameter "info" as in/out.

Keep track on unflushed modifications to file spaces.  When there are tens
of thousands of file spaces, flushing all files in fil_flush_file_spaces()
would be very slow.  (Bug #15653)

fil_flush_file_spaces(): Only flush unflushed file spaces.

fil_space_t, fil_system_t: Add a list of unflushed spaces.

call os_event_set while holding the sync array mutex.

Idea by Heikki, implementation by Osku, bug-fixes and other changes by
Heikki, and final review and cleanup by Osku.

Make innodb_flush_log_at_trx_commit a settable global variable.

Make innodb_thread_concurrency 0 by default, and extend the useable
range from 0 to 1000 (0 disables the thread throttling).

Remove SRV_CONCURRENCY_THRESHOLD (20).  The magic value to
disable the thread throttling is 0 from now on.

since this can lead to problems.

From MySQL:

Fix bug #15108 (mysqld crashes when innodb_log_file_size is set > 4GB).

Add BUILD/compile-innodb script.

Remember to use noninlined versions of the functions
on ha_innodb.cc !

Do not mistake TABLENAME_ibfk_0 for auto-generated id (Bug #16387).

dict_table_get_highest_foreign_id(): Ignore foreign constraint identifiers
starting with the pattern TABLENAME_ibfk_0.

innodb.test, innodb.result: Add test case.

Replace goto in os_event_wait with a normal loop.

Fixed a bug #16229 MySQL/InnoDB uses full explicit table locks 
in trigger processing. Take a InnoDB table lock only if user has explicitly
requested a table lock. Added some additional comments to store_lock() and 
external_lock(). Added test cases for a bug. Fixed some code style errors.

Fixed bug #15308: Problem of Order with Enum Column in 
Primary Key. Added a new test case for bug #14056.

Added diagnostic code trx_print() to 
innobase_query_caching_of_table_permitted() to find reason
why we are holding adaptive search latch.