MDEV-24190 Accessing INFORMATION_SCHEMA.INNODB_TABLESPACES_ENCRYPTION may break innodb_open_files logic

If the function i_s_tablespaces_encryption_fill_table() was not able to
report all content to the SQL layer, it would fail to decrement
fil_system.freeze_space_list that it had incremented. This would
prevent the not-frequently-used logic from working, potentially causing
frequently used files to be closed and reopened whenever innodb_open_files
is exceeded.

This regression was caused by
commit 45ed9dd9 (part of MDEV-23855).

This comes from a wrong merge from 10.3

The merge commit f2a94451
included some incorrect conflict resolution and left
an unused function wsrep_abort_slave_trx().

Also, we will clean up wsrep_innobase_kill_one_trx() a little.

MDEV-23855 broke the handling of innodb_flush_sync=OFF.
That parameter is supposed to limit the page write rate
in case the log capacity is being exceeded and log checkpoints
are needed.

With this fix, the following should pass:
./mtr --mysqld=--loose-innodb-flush-sync=0

One of our best regression tests for page flushing is
encryption.innochecksum. With innodb_page_size=16k and
innodb_flush_sync=OFF it would likely hang without this fix.

log_sys.last_checkpoint_lsn: Declare as Atomic_relaxed<lsn_t>
so that we are allowed to read the value while not holding
log_sys.mutex.

buf_flush_wait_flushed(): Let the page cleaner perform the flushing
also if innodb_flush_sync=OFF. After the page cleaner has
completed, perform a checkpoint if it is needed, because
buf_flush_sync_for_checkpoint() will not be run if
innodb_flush_sync=OFF.

buf_flush_ahead(): Simplify the condition. We do not really care
whether buf_flush_page_cleaner() is running.

buf_flush_page_cleaner(): Evaluate innodb_flush_sync at the low
level. If innodb_flush_sync=OFF, rate-limit the batches to
innodb_io_capacity_max pages per second.

Reviewed by: Vladislav Vaintroub

buf_read_ahead_random(): Do not leak a tablespace reference.
The reference was already acquired in fil_space_t::get(),
and we must only check that operations were not stopped.

This error was introduced when
commit 118e258a
merged n_pending_ios, n_pending_ops into a single n_pending.

This was not noticed earlier, because innodb_random_read_ahead
is OFF by default and our regression tests did not vary that
parameter at all.

The one which is in PATH by default is MinGW perl, which uses Unix paths.
This perl does not work with mtr.

Prepared statements which were run over binary protocol crashed
a server if the statement did not have CF_PS_ARRAY_BINDING_OPTIMIZED
flag and the statement was executed in bulk mode and a BF abort occrurred.
This was because the bulk execution resulted in several statements without
calling wsrep_after_statement() between, which confused wsrep transaction
state tracking.

As a fix, call wsrep_after_statement() in bulk loop after each execution
if CF_PS_ARRAY_BINDING_OPTIMIZED is not set.
Reviewed-by: Jan Lindström <jan.lindstrom@mariadb.com>

instant_alter_column_possible(): Relax a too strict debug assertion.
The existence of an index stub or a corrupted index on virtual columns
does not imply that virtual columns exist.

C++11 is allowed only starting with MariaDB Server 10.4.

This follows up commit
commit 94a520dd and
commit 7c5519c1.

After these changes, the default test suites on a
cmake -DWITH_UBSAN=ON build no longer fail due to passing
null pointers as parameters that are declared to never be null,
but plenty of other runtime errors remain.

This was missed in commit d6ea03fa.

This will  avoid some errors on appveyor, due to outdated SDKs.

Though this is an error message task, the problem was deep in the
`mysql_prepare_create_table` implementation. The problem is described as
follows:

1. `append_system_key_parts` was called before
`mysql_prepare_create_table`, though key name generation was done close to
the latest stage of the latter.
2. We can't move `append_system_key_parts` in the end, because system keys
should be appended before some checks done.
3. If the checks from `append_system_key_parts` are moved to the end of
`mysql_prepare_create_table`, then some other inappropriate errors are
issued. like `ER_DUP_FIELDNAME`.

To have key name specified in error message, name generation should be done
before the checks, which consequenced in more changes.

The final design for key initialization in `mysql_prepare_create_table`
follows. The initialization is done in three phases:
1. Calculate a total number of keys created with respect to keys ignored.
 Allocate KEY* buffer.
2. Generate unique names; calculate a total number of key parts.
 Make early checks. Allocate KEY_PART_INFO* buffer.
3. Initialize key parts, make the rest of the checks.

`ha_heap::clone` was creating a handler by share's handlerton, which is
partition handlerton.

handler's handlerton should be used instead.

Here in particular, HEAP handlerton will be used and it will create ha_heap
handler.

The bug was fixed by MDEV-22599 bugfix, which changed `Field::cmp` call
to `Field::cmp_prefix` in `TABLE::check_period_overlaps`.

The trick is that `Field_bit::cmp` apparently calls `Field_bit::cmp_key`,
which condiders an argument an actual pointer to data, which isn't correct
for `Field_bit`, since it stores data by `bit_ptr`. which is in the
beginning of the record, and using `ptr` is incorrect (we use it through
`ptr_in_record` call)

After Sergei's cleanup this assertion is not actual anymore -- we can't
predict if the handler was used for lookup, especially in multi-update
scenario.

`position(old_data)` is made earlier in `ha_check_overlaps`, therefore it
is guaranteed that we compare right refs.

The problem here was that ha_check_overlaps internally uses ha_index_read,
which in case of fail overwrites table->status. Even though the handlers
are different, they share a common table, so the value is anyway spoiled.
This is bad, and table->status is badly designed and overweighted by
functionality, but nothing can be done with it, since the code related to
this logic is ancient and it's impossible to extract it with normal effort.

So let's just save and restore the value in ha_update_row before and after
the checks.

Other operations like INSERT and simple UPDATE are not in risk, since they
don't use this table->status approach.
DELETE does not do any unique checks, so it's also safe.

1. Subtracting table->record[0] from record is UB (non-contiguous buffers)
2. It is very popular to use move_field_offset, which changes Field::ptr,
but leaves table->record[0] unchanged. This makes a ptr_in_record result
incorrect, since it relies on table->record[0] value.
The check ensures the result is within the queried record boundaries.

Add --system={all, users, plugins, udfs, servers, stats, timezones}

This will dump system information from the server in
a logical form like:
* CREATE USER
* GRANT
* SET DEFAULT ROLE
* CREATE ROLE
* CREATE SERVER
* INSTALL PLUGIN
* CREATE FUNCTION

"stats" is the innodb statistics tables or EITS and
these are dumped as INSERT/REPLACE INTO statements
without recreating the table.

"timezones" is the collection of timezone tables
which are important to transfer to generate identical
results on restoration.

Two other options have an effect on the SQL generated by
--system=all. These are mutually exclusive of each other.
* --replace
* --insert-ignore

--replace will include "OR REPLACE" into the logical form
like:
* CREATE OR REPLACE USER ...
* DROP ROLE IF EXISTS (MySQL-8.0+)
* CREATE OR REPLACE ROLE ...
* UNINSTALL PLUGIN IF EXISTS (10.4+) ... (before INSTALL PLUGIN)
* DROP FUNCTION IF EXISTS (MySQL-5.7+)
* CREATE OR REPLACE [AGGREGATE] FUNCTION
* CREATE OR REPLACE SERVER

--insert-ignore uses the construct " IF NOT EXISTS" where
supported in the logical syntax.

'CREATE OR REPLACE USER' includes protection against
being run as the same user that is importing the mysqldump.

Includes experimental support for dumping mysql-5.7/8.0
system tables and exporting logical SQL compatible with MySQL.

Updates mysqldump man page, including this information and
(removing obsolute bug reference)

Reviewed-by: anel@mariadb.org

Per b9f3f068, mysql_system_tables_data.sql creates
a mysql_native_password with a salted hash of "invalid" so that `set password`
will detect a native password can be applied:.

SHOW CREATE USER; diligently uses this value in its output
generating the SQL:

   MariaDB [(none)]> show create user;

   +---------------------------------------------------------------------------------------------------+
   | CREATE USER for dan@localhost                                                                     |
   +---------------------------------------------------------------------------------------------------+
   | CREATE USER `dan`@`localhost` IDENTIFIED VIA mysql_native_password USING 'invalid' OR unix_socket |
   +---------------------------------------------------------------------------------------------------+

Attempting to execute this before this patch results in:

  MariaDB [(none)]>  CREATE USER `dan2`@`localhost` IDENTIFIED VIA mysql_native_password USING 'invalid' OR unix_socket;
  ERROR 1372 (HY000): Password hash should be a 41-digit hexadecimal number

As such, deep the implementation of mysql_native_password we make "invalid" valid (pun intended)
such that the above create user will succeed. We do this by storing
"*THISISNOTAVALIDPASSWORDTHATCANBEUSEDHERE" (credit: Oracle MySQL), that is of an INCORRECT
length for a scramble.

In native_password_authenticate we check the length of this cached value
and immediately fail if it is anything other than the scramble length.

native_password_get_salt is only called in the context of set_user_salt, so all setting of native
passwords to hashed content of 'invalid', quite literally create an invalid password.

So other forms of "invalid" are valid SQL in creating invalid passwords:

   MariaDB [(none)]> set password = 'invalid';
   Query OK, 0 rows affected (0.001 sec)

   MariaDB [(none)]> alter user dan@localhost IDENTIFIED BY PASSWORD 'invalid';
   Query OK, 0 rows affected (0.000 sec)

closes #1628

Reviewer: serg@mariadb.com

buf_flush_try_neighbors(): Before invoking buf_page_t::ready_for_flush(),
check that the freshly looked up buf_pool.page_hash entry actually is
a buffer page and not a buf_pool.watch[] sentinel for purge buffering.

This race condition was introduced in MDEV-15053
(commit b1ab211d).
It is rather hard to hit this bug, because
buf_flush_check_neighbors() already checked the condition.
The problem exists if buf_pool.watch_set() was invoked for
a page in the range after the check in buf_flush_check_neighbor()
had been finished.