fts_state_t, fts_slot_t::state: Remove. Replaced by fts_slot_t::running
and fts_slot_t::table_id as follows.

FTS_STATE_SUSPENDED: Removed (unused).

FTS_STATE_EMPTY: Removed. table_id=0 will denote empty slots.

FTS_STATE_RUNNING: Equivalent to running=true.

FTS_STATE_LOADED, FTS_STATE_DONE: Equivalent to running=false.

fts_slot_t::table: Remove. Tables will be identified by table_id.
After opening a table, we will check fil_table_accessible() before
accessing the data.

fts_optimize_new_table(), fts_optimize_del_table(),
fts_optimize_how_many(), fts_is_sync_needed():
Remove the parameter tables, and use the static variable fts_slots
(which was introduced in MariaDB 10.2) instead.

No functional change.

Call my_timer_init() only once and then reuse it from InnoDB and
perfschema storage engines.

This patch speeds up empty test for me like this:
./mtr -mem innodb.kevg,xtradb  1.21s user 0.84s system 34% cpu 5.999 total
./mtr -mem innodb.kevg,xtradb  1.12s user 0.60s system 31% cpu 5.385 total

The crash happens when writing into log file.
The reason is likely that the call to WriteFile() was missing a valid
parameter for lpNumberOfBytesWritten. This seems only to happen on ancient
version of Windows.

Since the fix to MDEV-16430 in 141bc58a, null pointer was passed
instead of valid pointer.

The fix is to provide a valid lpNumberOfBytesWritten parameter.

A sequel to 9180e866 and 149b7547.

ALTER TABLE ... ADD FOREIGN KEY may crash if parent table is updated
concurrently.

Block FK parent table updates even earlier, before intermediate child
table is created.

Use proper charset info for my_casedn_str() and don't update original
identifiers so that lower_cast_table_names == 2 is honoured.

Fix:
====
1) Combined innodb_ft_result_cache_limit_32.test and
innodb_ft_result_cache_limit_64.test test case in sys_vars suite.

2) Use word_size.inc for combinations of innodb_ft_result_cache_limit test case.

regression after 3db6de33

in RPMs continue creating the $datadir outside of mysql_install_db.
RPMs put the socket in there, so it cannot be chmod 0700.

Test case was not stable. Fixed also galera_parallel_autoinc_manytrx
as it has the same problem.

This reverts commit 8dc670a5.

The symbol sys_timer_info was not being exported correctly,
which caused linking failures on some platforms.

regression after 3db6de33

in RPMs continue creating the $datadir outside of mysql_install_db.
RPMs put the socket in there, so it cannot be chmod 0700.

MONITOR_PENDING_LOG_WRITE
MONITOR_PENDING_CHECKPOINT_WRITE
MONITOR_LOG_IO: read values from log_t members instead of updating own
monitor variables

No functional change.

Call my_timer_init() only once and then reuse it from InnoDB and
perfschema storage engines.

This patch speeds up empty test for me like this:
./mtr -mem innodb.kevg,xtradb  1.21s user 0.84s system 34% cpu 5.999 total
./mtr -mem innodb.kevg,xtradb  1.12s user 0.60s system 31% cpu 5.385 total

Based on pull request https://github.com/MariaDB/server/pull/999
by mkaruza@galeracluster.com

To read histograms for a table, we should check if the allocation of statistics was done or not,
if not done we should not try to read histograms for such a table.

to the patch for MDEV-17605

MDEV-14572: Assertion `! is_set()' failed in Diagnostics_area::set_eof_status upon EXPLAIN UPDATE in PS

Restore EXPAIN flag in SELECT_LEX before execution multi-update by flag in LEX
(the same but in other way made before INSERT/DELETE/SELECT)

Without it, mysql_update() didn't know that there will be EXPLAIN result set and was sending OK at the end of the update, which conflicted with the EOF sent later by EXPLAIN.

The patch for MDEV-18738 fixed this problem. Adding tests only.

copy_if_not_alloced() did not handle situations when
"from" is a constant string pointing to a substring of "to",
so this code part freed "to" but then tried to copy its old (already freed)
content to a new buffer:

  if (to->realloc(from_length))
    return from;
  if ((to->str_length=MY_MIN(from->str_length,from_length)))
    memcpy(to->Ptr,from->Ptr,to->str_length);

Adding a new code piece that catches such constant substrings
and propery reallocs "to" to preserve its important part referenced
by "from".

This problem was earlier fixed, possibly by f8a800be
and is not repeatable in 10.1-10.4 any more. Adding tests only.

if log_warnings > 1.

This makes ER_DBACCESS_DENIED_ERROR handling the same as we do for other
"access denied"

Change error message.

names

Added a call to X509_check_ip_asc() in case server_hostname represents
an IP address.

The fix for MDEV-17064 addressed this problem. Adding tests only.

           depends on uninitialised value

Initialized THD::force_read_stats introduced in the patch for MDEV-17605.
Leaving this field uninitialized in the constructor of the THD class may
trigger reading statistical data that is not needed.

Item_func_group_concat stores values in `tree`, which is often, but not
always the same as `&tree_base`.

move privilege specific part of gis2.test to gis_notembedded.test
and the rest to gis.test

Fixed the assert by making sure that not to use EITS if the column statistics was not allocated.

MDEV-11015 Assertion failed: precision > 0 in decimal_bin_size upon SELECT with DISTINCT, CAST and other functions

Item_func_min_max::fix_length_and_dec() erroneously set max_length
to UINT32_MAX.

Merge notes:
In 10.3 this problem had been fixed earlier.
During merge to 10.3, do a "null merge" in item_func.cc