Adjust the configuration options, as discussed on the
maria-developers@ mailing list.

The option to hint a transaction to not be replicated in parallel is
now called @@skip_parallel_replication, consistent with
@@skip_replication.

And the --slave-parallel-mode is now simplified to have just one of
the following values:

  none
  minimal
  conservative
  optimistic
  aggressive

This reflects successively harder efforts to find opportunities to run
things in parallel on the slave. It allows to extend the server with
more automatic heuristics in the future without having to introduce a
new configuration option for each and every one.

(Without this, it happened for me that realpath() failed returning
undef for the default vardir. This in turn caused mysql-test-run.pl to
delete the source mysql-test/ directory.)

pool->threads is freed before being reassigned the new pool.

Although not really a memory barrier I though it prudent to keep the pool
thread count to be the lower of the old/new thread list before the new threads
is allocated.

to be able disable wsrep before any further code that uses it
(e.g. initialized binlog when wsrep is enabled)

            fails in buildbot)

There was a race condition in timer functionality of query timeouts.

Race was as following:
main thread:          init_thr_timers()
timer handler thread: my_thread_init()
main thread:          end_thr_timer()/timer_thread_state= ABORTING
timer handler thread: timer_thread_state= RUNNING, continue normal op
main thread:          waits indefinitely for timer handler thread to go down

The original idea of the fix is to set RUNNING state in main thread, before
starting timer handler thread. But it didn't survive further cleanups:
- removed "timer_thread_state" and used "thr_timer_inited" for this purpose
- removed unused "timer_thread_running"
- removed code responisble for "timer handler thread" shutdown synchronization,
  use pthread_join() instead.

INFORMATION_SCHEMA.SYSTEM_VARIABLES.DEFAULT_VALUE had broken values on
big endian.

Default value is internally stored as longlong, while I_S references it's
pointer (longlong *) according to variable type (e.g. int, my_bool, etc). This
works well on little endian, but on big endian we always get 0 for such
variables.

This is temporal test fixt to avoid concurrent IST for now until
the actual issue https://github.com/codership/galera/issues/204
is fixed.

With commit 227510e0 (parser cleanup: don't store field properties in LEX,
use Create_field directly) it became possible to create columns DOUBLE(0,0) and alike

Added options for each compression type that can be disabled i.e.
bzip2, lz4, lzma and lzo.

* reset current_thd in THD::~THD, otherwise my_malloc_size_cb_func()
  might access THD after it was destroyed.
* remove now redundant set_current_thd(0) calls that follow delete thd.

This reverts commit 06c16904.

* error reporting was never needed
* avoid useless transformaton hton to db_type to hton
* in many cases the engine was guaranteed to exist, add asserts
* use ha_default_handlerton() instead of ha_checktype(DB_TYPE_DEFAULT)

but keep the code cleanup.
as it turns out, it didn't change the behavior, which is good :)

Now this works the same way as all other multi source variables.

Extend interface for 'show variables' with current scope

in case args[0] changes between PREPARE and EXECUTE

in particular, not the one that destructively changes Item's charset
(meaning different metadata for PREPARE and EXECUTE)

* fix the test to pass
* rearrange tests in a file to allow it to auto-merge in the future

use the old registry branch, even if we changed the "manufacturer" string

MDEV-7024: Assertion `! is_set()' failed in Diagnostics_area::set_eof_status on executing ANALYZE SELECT via PS

select_send::is_result_interceptor() should return FALSE because it return results to the client.

added a script to regenerate certificates easily in the future (2035!)
restored server8k-key.pem to actually be 8K key, as it was supposed to