a.k.a. Bug#7975 deadlock without any locking, simple select and update

Bug#7975 was reintroduced when the storage engine API was made
pluggable in MySQL 5.1. Instead of looking at thd->lex directly, we
rely on handler::extra(). But, we were looking at the wrong extra()
flag, and we were ignoring the TRX_DUP_REPLACE flag in places where we
should obey it.

innodb_replace.test: Add tests for hopefully all affected statement
types, so that bug should never ever resurface. This kind of tests
should have been added when fixing Bug#7975 in MySQL 5.0.3 in the
first place.

rb:806 approved by Sunny Bains

No commit message

                      ALTER TABLE AND/OR PLUGIN/SEMISYNC

If a plugin was uninstalled, thread local values for plugin
variables of string type with PLUGIN_VAR_MEMALLOC flag were
not freed.

With this patch these variables are freed when thread is
done (like all other variables).

      
      A patch for this bug has already been pushed. A minor change is made here.
      The database to be used after re-enabling the disabled code is 'TEST'.
      But instead, 'MYSQL' was being used. 
      This is the minor change that is being made here.

  Un-needed replace_result accidentally messed up a variable used
  *if* not running parallel.
  Removed the bogus --replace_result

    Don't do this for echo, instead:
    1) Enable replacements also for assignment from backquoted SQL
    2) Allow replace_regex to take a variable for the *entire* argument list
  
    With this, the test can be amended, but only in its version in trunk

btr_pcur_restore_position_func(): When the cursor was positioned at
the tree infimum or supremum, initialize pos_state and latch_mode. The
assertion failed, because pos_state was BTR_PCUR_WAS_POSITIONED.  In
the test failure of WL#5874, the purge thread attempted to restore the
cursor position on the infimum record (the clustered index was empty).

btr_pcur_detach(), btr_pcur_is_detached(): Unused functions, remove.

rb:804 approved by Inaam Rana

In all callers of row_sel_convert_mysql_key_to_innobase(), assert
that the converted key is empty or nonempty when it should be.

ibuf_insert_low(), the only caller of ibuf_set_entry_counter(), will
have latched an insert buffer bitmap page in bitmap_mtr before
invoking ibuf_set_entry_counter(). The latching order forbids any
further pages to be latched.

ibuf_set_entry_counter(): Renamed to ibuf_get_entry_counter(),
simplified the code and added comments.

Added the following symbols for predefined field numbers in change
buffer records:

#define IBUF_REC_FIELD_SPACE	0	/*!< in the pre-4.1 format,
					the page number. later, the space_id */
#define IBUF_REC_FIELD_MARKER	1	/*!< starting with 4.1, a marker
					consisting of 1 byte that is 0 */
#define IBUF_REC_FIELD_PAGE	2	/*!< starting with 4.1, the
					page number */
#define IBUF_REC_FIELD_METADATA	3	/* the metadata field */
#define IBUF_REC_FIELD_USER	4	/* first user field */

rb:802 approved by Sunny Bains

This patch corrects the test mysql_plugin so that it correctly masks
the library extension of the plugin daemon_example.

  Simplified fix avoiding changes to mysys:
  Use the MY_HOLD_ORIGINAL_MODES flag when calling my_copy(),
    this also stops it from attempting to chown() the file.
  Yes this behavior is a bit confusing....
  The only case this might change the behavior is if the destination file
    exists, but since we also use MY_DONT_OVERWRITE_FILE, it would fail
    in those cases anyway.

OF 16G BUFFER POOL TAKES HOURS

rb://787 approved by Sunny

and query_cache don't play along nicely.

An alternative for RENAME DATABSE using RENAME TABLE has been used to implement this.

Parts of index_merge_innodb were disabled.These have been enabled with a few changes 
and the test is being made experimental to study it's behaviour.

Also fixed possibly uninitialized use of need_copy_table_res.

post-push fixes for show_slave_io_error= 1 of wait_for_slave_io_error.inc;
Unix and win format path specifically so few tests have to change show_slave_io_error
to zero.

The bug case is similar to one fixed earlier bug_49536.
Deadlock involving LOCK_log appears to be possible because the purge running thread
is holding LOCK_log whereas there is no sense of doing that and which fact was
exploited by the earlier bug fixes.

Fixed with small reengineering of rotate_and_purge(), adding two new methods and
setting up a policy to execute those instead of the former
rotate_and_purge(RP_LOCK_LOG_IS_ALREADY_LOCKED).
The policy for using rotate(), purge() is that if the caller acquires LOCK_log itself,
it should call rotate(), release the mutex and run purge().

Side effect of this patch is refining error message of bug@11747416 to print
the whole path.

row_rename_table_for_mysql(): Return DB_ERROR instead of DB_SUCCESS
when fil_rename_tablespace() returns an error. This bug was introduced
in the InnoDB Plugin.

Approved by Sunny Bains over IM.

TABLE ERROR, RECOVERY
            
rb://792 approved by Sunny Bains

Fixed a misplaced parenthesis, injected due to
syncing from libedit CVS head.

NEW_FRM_MEM WITHOUT NEEDING TO".

During the process of opening tables for a statement, we allocated
memory which was used only during view loading even in cases when the
statement didn't use any views. Such an unnecessary allocation (and
corresponding freeing) might have caused significant performance
overhead in some workloads. For example, it caused up to 15% slowdown
in a simple stored routine calculating Fibonacci's numbers.

This memory was pre-allocated as part of "new_frm_mem" MEM_ROOT
initialization at the beginning of open_tables(). 

This patch addresses this issue by turning off memory pre-allocation
during initialization for this MEM_ROOT. Now, memory on this root
will be allocated only at the point when the first .FRM for a view is 
opened.

The patch doesn't contain a test case since it is hard to test the
performance improvements or the absence of memory allocation in our
test framework.