GCC crashes during compilation, on a code fragment that was added  to
workaround code generation bug on GCC.

To fix, I randomly moved pragmas around, which seems to make GCC happy.

MDEV-19706 RPM no longer installs init script on systemd systems, but preun script still tries to erase it

MDEV-16249 CHECKSUM TABLE for a spider table is not parallel and saves all data in memory in the spider head by default (#1328)

followup for be5c432a

ha_partition::calculate_checksum() has to invoke calculate_checksum()
for partitions unconditionally, not under (HA_HAS_OLD_CHECKSUM | HA_HAS_NEW_CHECKSUM).
Because the server uses ::info() to ask for a live checksum, while
calculate_checksum() must, precisely, calculate it the slow way,
also for tables that don't have the live checksum at all.

Also, fix the compilation on Windows (ha_checksum/ulonglong type mix).

Introduced a print_key_value function to makes sure that the trace prints data in readable format
for readable characters and the rest of the characters are printed as hexadecimal.

Older GCC generates SSE instruction on not-128-bit-aligned data in
Bitmap<128>::buffer

Workaround by forcing GCC not to use SSE on Bitmap<N> template.

MDEV-16249 CHECKSUM TABLE for a spider table is not parallel and saves all data in memory in the spider head by default (#1328)

add checksum_null for setting null value of checksum

Some I/O functions and macros that are declared in os0file.h used to
return a Boolean status code (nonzero on success). In MySQL 5.7, they
were changed to return dberr_t instead. Alas, in MariaDB Server 10.2,
some uses of functions were not adjusted to the changed return value.

Until MDEV-19231, the valid values of dberr_t were always nonzero.
This means that some code that was incorrectly checking for a zero
return value from the functions would never detect a failure.

After MDEV-19231, some tests for ALTER ONLINE TABLE would fail with
cmake -DPLUGIN_PERFSCHEMA=NO. It turned out that the wrappers
pfs_os_file_read_no_error_handling_int_fd_func() and
pfs_os_file_write_int_fd_func() were wrongly returning
bool instead of dberr_t. Also the callers of these functions were
wrongly expecting bool (nonzero on success) instead of dberr_t.

This mistake had been made when the addition of these functions was
merged from MySQL 5.6.36 and 5.7.18 into MariaDB Server 10.2.7.

This fix also reverts commit 40becbc3
which attempted to work around the problem.

Make it visible

MDEV-18788 Live upgrade from MySQL 5.6/5.7 to MariaDB 10.4 fails with "Event Scheduler: An error occurred when initializing system tables"

if columns or indexes are modified/renamed/dropped in an ALTER TABLE,
stat tables must be updated accordingly (e.g. all statistics for a column
should be dropped). But if a stat table doesn't exist, it's not a reason
to fail the whole ALTER TABLE operation - such an error should be ignored.

comments, rename a function to reflect its usage better

https://github.com/MariaDB/server/pull/777#issuecomment-496469366

option values now have to be edited in-place,
instead of replacing an option with another one
that has the same name but a new value.

- --disable-plugin-load or --skip-plugin-load will now reset plugin load
  list. Needed for future changes to mtr tests

- multiple usage of plugin-load-add and optimizer-switch are now included
  in generated config files

Problem:
=======
fil_iterate() writes imported tablespace page0 as it is to discarded
tablespace. Space id wasn't even changed. While opening the tablespace,
tablespace fails with space id mismatch error.

Fix:
====
fil_iterate() copies the page0 with discarded space id to imported
tablespace.

fix MDEV-18750: failed to flashback large-size binlog file

fix mysqlbinlog flashback failure caused by reading io_cache without MY_FULL_IO flag

fix MDEV-18750: mysqlbinlog flashback failure on large binlog

- Use corrupt page id instead of whole block after releasing it from
LRU list.

MDEV-19600: The optimizer should be able to produce rows=1 estimate for unique index with NULLable columns

Modify best_access_path() to produce rows=1 estimate for null-rejecting
lookups on unique NULL keys.

MDEV-18015: Assertion `global_status_var.global_memory_used == 0' failed when using UDF , window functions and views

Adding destructor for Group_bound_tracker to free Cached_item_str.
The Cached_item for window functions are allocated on THD:mem_root
but the Cached_item_str has value of type string which is allocated on
the heap, so we need to call free() for it

Using different recommended speedup options for WolfSSL.

- Enable  x64 assembly code on Intel.
- in my_crypt.cc, align EVP_CIPHER_CTX buffer, since some members need
alignment of 16 (for AESNI instructions), when assembler is enabled.
- Adjust MY_AES_CTX_SIZE
- Enable fastmath in wolfssl (large integer math).

os_file_write_func() and os_file_read_no_error_handling_func() returned
different result values depending on if UNIV_PFS_IO was defined or not.

Other things:
- Added some comments about return values for some functions

Wrong compiler warning from GCC:
‘snprintf’ output 2 or more bytes (assuming 4001)
into a destination of size 4000

- --default-character-set can now be disabled in mysqldump
- --skip-resolve can be be disabled in mysqld
- mysql_client_test now resets global variables it changes
- mtr couldn't handle [mysqldump] in config files (wrong regexp used)

Do not use TARGET_LINK_LIBRARIES(mysys mysys_ssl),
this means that mysys_ssl is linked to practically everything.

client-certkey.pem contains both certificate and corresponding
private key.

The issue is that two MARIA_HA instances shares the same MARIA_STATUS_INFO
object during UNION execution, so the second MARIA_HA instance state pointer
MARIA_HA::state points to the MARIA_HA::state_save of the first MARIA instance.

This happens in
thr_multi_lock(...) {
...
	for (first_lock=data, pos= data+1 ; pos < end ; pos++)
	{
...
		if (pos[0]->lock == pos[-1]->lock && pos[0]->lock->copy_status)
			(pos[0]->lock->copy_status)((*pos)->status_param,
																	(*first_lock)->status_param);
...
	}
...
}

Usually the state is restored from ha_maria::external_lock(...):

\#0  _ma_update_status (param=0x6290000e6270) at ./storage/maria/ma_state.c:309
\#1  0x00005555577ccb15 in _ma_update_status_with_lock (info=0x6290000e6270) at ./storage/maria/ma_state.c:361
\#2  0x00005555577c7dcc in maria_lock_database (info=0x6290000e6270, lock_type=2) at ./storage/maria/ma_locking.c:66
\#3  0x0000555557802ccd in ha_maria::external_lock (this=0x61d0001b1308, thd=0x62a000048270, lock_type=2) at ./storage/maria/ha_maria.cc:2727

But _ma_update_status() does not take into account the case when
MARIA_HA::status points to the MARIA_HA::state_save of the other MARIA_HA
instance.

The fix is to restore MARIA_HA::state in ha_maria::external_lock() after
maria_lock_database() call for transactional tables.