Do not add + 1 to the InnoDB index cardinality estimate if the B-tree just contains one page; the fix made in March 2004 caused InnoDB systematically to overestimate the cardinality of empty or small tables by 1
Correct previous bk commit comment about page splits: actually, in versions < 4.0.19 space utilization is very poor if 1500 - 8000 byte rows are inserted in the order of the primary key
Improve space utilization if we have 3 kB - 8 kB rows to insert in the order of the primary key
btr0cur.c:
Fix bug: The row count and key cardinality estimate was grossly too small if each clustered index page only contained one record
Fix bug: if one updated a secondary index column so that its alphabetical value did not change (e.g., abc -> aBc) and rolled back the update, InnoDB failed to return the value in the secondary index to its original value
row0upd.h:
Correct typing error
sync0sync.ic:
Remove inadvertently pushed sync debug code
Fix assertion failure on line 713 of row0upd.c if there is a column prefix index and the last characters in the prefix are spaces: do not assume that the length of alphabetically equal strings is the same; fix a buglet which could cause InnoDB to think that a secondary index record was not locked though it had been updated in a way which did not alpahabetically change its value, e.g., abc -> aBc
Fix bug: if the user created a prefix column key on a fixed length char column, then InnoDB claimed in CHECK TABLE that the table is corrupt
dict0dict.c:
Fix bug 1151: if the user created a prefix column primary key on a fixed length char column, then InnoDB crashed in a simple SELECT
Fix bug: if there was a 'record too long' error in an insert, InnoDB forgot to free reserved file space extents; they were only freed in mysqld restart
Fix bug reported by Dyego Souza do Carmo: if a row becomes too long, > 8000 bytes, in an update, then InnoDB simply removes the clustered index record and does not report of table handler error 139
Remove potential starvation of a full log buffer flush: only flush up to the lsn which was the largest at the time when we requested the full log buffer flush
os0sync.h, os0sync.c:
Fix a bug in os_event on Unix: even though we signaled the event, some threads could continue waiting if the event became nonsignaled quickly again; this made group commit less efficient than it should be
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
libraries always ended up in "dir..", even though they were not supposed
to be installed anyway (they are only required at link time). Fixed it by
replacing libs_LIBRARIES with noinst_LIBRARIES for all InnoDB Makefile.am
files and by removing "libsdir = " from innobase/include/Makefile.i .
Backport from 4.0: Fix a major bug in InnoDB query estimator for queries of type SELECT ... WHERE col < x and SELECT ... WHERE col > x; MySQL could pick a table scan though the result set was only a few rows in a big table
Backport bugfix from 4.0: combined log file size >= 2 GB could cause log to be written in wrong place
btr0pcur.c:
Backport bugfix from 4.0: index cursor restoration could theoretically fail
Fix bug: an index cursor can theoretically be restored in a wrong place
log0log.c:
Fix bug: if combined log file size is >= 2 GB in a 32-bit computer InnoDB can write log to a wrong position
Merge InnoDB-4.0.7. Support for ON UPDATE CASCADE
sql_select.cc:
Remove superfluous prints to .err log when a locking SELECT fails to a deadlock or a lock wait timeout
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
Do not let range estimator to return over 1 / 2 of total rows in table; use longlong in range estimation
btr0cur.h, ha_innobase.cc:
Use longlong in range estimation, in case there are > 4 billion rows
Improve mysql-test to be more robust.
Fix that GRANT doesn't delete SSL options
Change innobase_flush_log_at_trx_commit to uint.
Don't rotate logs if we read a rotate log entry from the master.
ha_innodb.cc Remove gaps in auto-inc in multi-row inserts, more space for foreign key listings in SHOW TABLE STATUS, move resetting of active_trx to amore logical place
dict0dict.h Remove gaps from auto-inc sequence if errors in multi-row insert
dict0dict.c Remove gaps from auto-inc sequence if errors in multi-row insert
