MDEV-12353 deliberately tries to avoid memory alignment overhead in
log_phys_t, storing the stream of log records bytes straight after
a header object. But, offsetof() is not allowed on C++ data objects,
and neither are attempts to emulate it by invoking a member function
on a null pointer.

log_phys_t::len: Remove. Make it part of the byte stream that
immediately follow the object. Thanks to Eugene Kosov for this idea.

log_phys_t::start(): The start address of the following byte stream.

log_phys_t::len(): Compute len.

log_phys_t::alloc_size(): Use a simple sizeof calculation.

The InnoDB internal tables SYS_TABLESPACES and SYS_DATAFILES as well as the
INFORMATION_SCHEMA views INNODB_SYS_TABLESPACES and INNODB_SYS_DATAFILES
were introduced in MySQL 5.6 for no good reason in
mysql/mysql-server/commit/e9255a22ef16d612a8076bc0b34002bc5a784627
when the InnoDB support for the DATA DIRECTORY attribute was introduced.

The file system should be the authoritative source of information on files.
Storing information about file system paths in the file system (symlinks,
or even the .isl files that were unfortunately chosen as the solution) is
sufficient. If information is additionally stored in some hidden tables
inside the InnoDB system tablespace, everything unnecessarily becomes
more complicated, because more copies of data mean more opportunity
for the copies to be out of sync, and because modifying the data in
the system tablespace in the desired way might not be possible at all
without modifying the InnoDB source code. So, the copy in the system
tablespace basically is a redundant, non-authoritative source of
information.

We will stop creating or accessing the system tables SYS_TABLESPACES
and SYS_DATAFILES.

We will also remove the view
INFORMATION_SCHEMA.INNODB_SYS_DATAFILES along with SYS_DATAFILES.

The view
INFORMATION_SCHEMA.INNODB_SYS_TABLESPACES will be repurposed
to directly reflect fil_system.space_list. The column
PAGE_SIZE, which would always contain the value of
the GLOBAL read-only variable innodb_page_size, is
removed. The column ZIP_PAGE_SIZE, which would actually
contain the physical page size of a page, is renamed to
PAGE_SIZE. Finally, a new column FILENAME is added, as a
replacement of SYS_DATAFILES.PATH.

This will also
address MDEV-21801 (files that were created before upgrading
to MySQL 5.6 or MariaDB 10.0 or later were never registered
in SYS_TABLESPACES or SYS_DATAFILES) and
MDEV-21801 (information about the system tablespace is not stored
in SYS_TABLESPACES or SYS_DATAFILES).

Let us introduce the parameter innodb_read_only_compressed
that is ON by default, making any ROW_FORMAT=COMPRESSED tables
read-only.

I developed the ROW_FORMAT=COMPRESSED format based on
Heikki Tuuri's rough design between 2005 and 2008. It might
have been a good idea back then, but no proper benchmarks were
ever run to validate the design or the implementation.

The format has been more or less obsolete for years.
It limits innodb_page_size to 16384 bytes (the default),
and instant ALTER TABLE is not supported.

This is the first step towards deprecating and removing
write support for ROW_FORMAT=COMPRESSED tables.

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).

Adjust the testcase for MariaDB 10.3+ : prevent IN-to-subquery conversion
optimization from working.

MDEV-23672 (commit 7eda5561)
introduced a regression that can corrupt not only undo log pages,
but anything that resides in the InnoDB buffer pool.

trx_undo_left(): Add debug assertions for the assumptions.
If the pointer is out of bounds, we will return a positive
number, not a negative one. Thus, once a page overflow occurs,
further overflow to adjacent pages will be allowed.
This allows us to remove some more relaxed debug assertions
from some callers.

trx_undo_log_v_idx(): Correctly calculate the size limit.

This bug could manifest itself for a query with WHERE condition containing
top level OR formula such that each conjunct contained a single-range
condition supported by the same index. One of these range conditions must
be fully covered by another range condition that is used later in the OR
formula. Additionally at least one of these condition should be ANDed with
a sargable range condition supported by a different index.

There were several attempts to fix related problems for OR conditions after
the backport of range optimizer code from MySQL (commit
0e19f3e3). Unfortunately the first of these
fixes contained typo remained unnoticed until recently. This typo bug led
to rejection of valid range accesses. This patch fixed this typo bug.
The fix revealed another two bugs: one in a constructor for SEL_ARG,
the other in the function tree_or(). Both are fixed in this patch.

This comes from a wrong merge from 10.3

Add a testcase.

Part#1: Revert the patch that caused it:

commit 291be494
Author: Igor Babaev <igor@askmonty.org>
Date:   Thu Sep 24 22:02:00 2020 -0700

    MDEV-23811: With large number of indexes optimizer chooses an inefficient plan

- mysqlnd from PHP < 7.3
- mysql-connector-python any version
- mysql-connector-java any version

Relaxed check about garbage at the end of the packet in case of no parameters.
Added check for array binding.
Fixed test according to the new paradigm (allow junk at the end of the packet)

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.