The trx_t::xid is always allocated, so we might as well allocate it
directly in the trx_t object to improve the locality of reference.

In commit e71e6133 we
accidentally made innodb_lock_wait_timeout=100000000
a "literal" value, not the smallest special value that
would mean "infinite" timeout.

trans_rollback_to_savepoint(): Only release metadata locks (MDL)
if the storage engines agree, after the changes were already rolled back.

Ever since commit 3792693f
and mysql/mysql-server@55ceedbc3feb911505dcba6cee8080d55ce86dda
we used to cheat here and always release MDL if the binlog is disabled.

MDL are supposed to prevent race conditions between DML and DDL also
when no replication is in use. MDL are supposed to be a superset of
InnoDB table locks: InnoDB table lock may only exist if the thread
also holds MDL on the table name.

In the included test case, ROLLBACK TO SAVEPOINT would wrongly release
the MDL on both tables and let ALTER TABLE proceed, even though the DML
transaction is actually holding locks on the table.

Until commit 1bd681c8 (MDEV-25506)
InnoDB worked around the locking violation in a blatantly non-ACID way:
If locks exist on a table that is being dropped (in this case, actually
a partition of a table that is being rebuilt by ALTER TABLE), InnoDB
would move the table (or partition) into a queue, to be dropped after
the locks and references had been released.

The scenario of commit 3792693f
is unaffected by this fix, because mariadb-dump (a.k.a. mysqldump)
would use non-locking reads, and the transaction would not be holding
any InnoDB locks during the execution of ROLLBACK TO SAVEPOINT.
MVCC reads inside InnoDB are only covered by MDL and page latches,
not by any table or record locks.

FIXME: It would be nice if storage engines were specifically asked
which MDL can be released, instead of only offering a choice
between all or nothing. InnoDB should be able to release any
locks for tables that are no longer in trx_t::mod_tables, except
if another transaction had converted some implicit record locks
to explicit ones, before the ROLLBACK TO SAVEPOINT had been completed.

Reviewed by: Sergei Golubchik

The practical maximum value of the parameter innodb_lock_wait_timeout
is 100,000,000. Any value larger than that specifies an infinite timeout.

Therefore, we should make 100,000,000 the maximum value of the parameter.

Let us use innodb_lock_wait_timeout=0 for an immediate timeout.
Also, do not override the timeout in the default connection,
so that further tests will use the default setting.

This replaces 8711adb7

if a temptable field is created for some json expression (is_json_type()
returns true), make this temptable field a proper json field.

A field is a json field (see Item_field::is_json_type()) if it
has a CHECK constraint of JSON_VALID(field).

Note that it will never be actually checked for temptable fields,
so it won't cause a run-time slowdown.

Revert "fix JSON_ARRAYAGG not to over-quote json in joins"
This removes 8711adb7 but keeps the test case.
A different fix is coming up.

Because args can be Item_field's that are later
replaced by Item_direct_view_ref to the actual field.
While Item_field preserved in orig_args will stay unfixed
with item->field==NULL and no metadata

MDEV-23004 When using GROUP BY with JSON_ARRAYAGG with joint table, the square brackets are not included

make test results stable

followup for 98c7916f

use metadata (in particular is_json() property) of the original
argument item, even if the actual argument was later replaced
with an Item_temptable_field

when checking whether thd wasn't killed before this
emb_advanced_command(), take into account that it
could've been killed before the *previous*
emb_advanced_command(). That is, the previous one has
already set thd to NULL and this one only wanted a COM_STMT_RESET
after a failure.

Since 420f8e24 InnoDB uses O_DIRECT by default

see also c3a1ba0f, 068246c0, 690ae1de

trx_t::drop_table(): Relax also another assertion that would fail
due to an AUTO_INCREMENT lock that is being held by the current
test case. This should have been part of
commit 63e9a054.

MDEV-25837 Assertion `thd->locked_tables_mode == LTM_NONE' failed in Locked_tables_list::init_locked_tables.

don't do prelocking for the FLUSH command.

trx_t::drop_table(): Delete-mark the SYS_TABLES and SYS_INDEXES
record before delete-marking any SYS_COLUMNS or SYS_FIELDS records.
Otherwise, dict_load_indexes() could fail on recovery. This fixes up
commit 1bd681c8 (MDEV-25506).

row_get_prebuilt_insert_row(): Remove some fallback code that had been
added in commit 8ea923f5 (MDEV-24818).
It seems that after all, statement boundaries are being reliably
indicated by ha_innobase::start_stmt() or
(for partitioned tables) ha_innobase::external_lock().

This is based on code that was contributed by Ning Zheng and Ray Kuan
from ScaleFlux.

The MariaDB implementation of page_compressed tables for InnoDB used
sparse files. In the worst case, in the data file, every data page
will consist of some data followed by a hole. This may be extremely
inefficient in some file systems.

If the underlying storage device is thinly provisioned (can compress
data on the fly), it would be good to write regular files (with sequences
of NUL bytes at the end of each page_compressed block) and let the
storage device take care of compressing the data.

For reads, sparse file regions and regions containing NUL bytes will be
indistinguishable.

my_test_if_disable_punch_hole(): A new predicate for detecting thinly
provisioned storage. (Not implemented yet.)

innodb_atomic_writes: Correct the comment.

buf_flush_page(): Support all values of fil_node_t::punch_hole.
On a thinly provisioned storage device, we will always write
NUL-padded innodb_page_size bytes also for page_compressed tables.

buf_flush_freed_pages(): Remove a redundant condition.

fil_space_t::atomic_write_supported: Remove. (This was duplicating
fil_node_t::atomic_write.)

fil_space_t::punch_hole: Remove. (Duplicated fil_node_t::punch_hole.)

fil_node_t: Remove magic_n, and consolidate flags into bitfields.
For punch_hole we introduce a third value that indicates a
thinly provisioned storage device.

fil_node_t::find_metadata(): Detect all attributes of the file.

my_init_atomic_write(): Detect all forms of SSD, in case multiple
types of devices are installed in the same machine.
This was broken in commit ed008a74
and further in commit 70684afe.

SAME_DEV(): Match block devices, ignoring partition numbers.

Let us use stat() instead of lstat(), in case someone has a symbolic
link in /dev.

Instead of reporting errors with perror(), let us use fprintf(stderr)
with the file name, the impact of the error, and the strerror(errno).
Because this code is specific to Linux, we may depend on the
GNU libc/uClibc/musl extension %m for strerror(errno).

The bug was originally observed as hanging binlog background thread at
shutdown similar to one of MDEV-21120.
It occurred through unnessary xid logging in 1pc execution.

Two parts of the issue are fixed.  Per engine loop by involved engine
with attempt to mark a group requiring xid unlogging gets corrected in
two ways. Do not execute it when the termination event is irrelevant
for recovery, does not have xid in particular.  Do not break the loop
anymore unconditionally at the end of the 1st iteration.

Make sure to call cached item's destructors.

trx_t::drop_table(): Remove a bogus debug assertion.
The current transaction may hold an AUTO_INCREMENT
lock on the table while
CREATE TABLE t2 (pk INT AUTO_INCREMENT PRIMARY KEY) ENGINE=InnoDB SELECT...
is being rolled back due to lock wait timeout.
Remaining debug assertions will check that only this transaction
is holding locks on the table, and that one of them is an exclusive lock.

square brackets are not included.

Item_func_json_arrayagg::copy_or_same() should be implemented.

buf_flush_relocate_on_flush_list(): If we are removing the block from
buf_pool.flush_list, subtract its size from buf_pool.stat.flush_list_bytes.
This fixes a regression that was introduced in
commit 22b62eda (MDEV-25113).

In commit 5f22511e we depend on
Total Store Ordering. For correct operation on ISAs that implement
weaker memory ordering, we must explicitly use release/acquire stores
and loads on buf_page_t::oldest_modification_ to prevent a race condition
when buf_page_t::list does not happen to be on the same cache line.

buf_page_t::clear_oldest_modification(): Assert that the block is
not in buf_pool.flush_list, and use std::memory_order_release.

buf_page_t::oldest_modification_acquire(): Read oldest_modification_
with std::memory_order_acquire. In this way, if the return value is 0,
the caller may safely assume that it will not observe the buf_page_t
as being in buf_pool.flush_list, even if it is not holding
buf_pool.flush_list_mutex.

buf_flush_relocate_on_flush_list(), buf_LRU_free_page():
Invoke buf_page_t::oldest_modification_acquire().

In commit 22b62eda (MDEV-25113)
we introduced a race condition. buf_LRU_free_page() would read
buf_page_t::oldest_modification() as 0 and assume that
buf_page_t::list can be used (for attaching the block to the
buf_pool.free list). In the observed race condition,
buf_pool_t::delete_from_flush_list() had cleared the field,
and buf_pool_t::delete_from_flush_list_low() was executing
concurrently with buf_LRU_block_free_non_file_page(),
which resulted in buf_pool.flush_list.end becoming corrupted.

buf_pool_t::delete_from_flush_list(), buf_flush_relocate_on_flush_list():
First remove the block from buf_pool.flush_list, and only then
invoke buf_page_t::clear_oldest_modification(), to ensure that
reading oldest_modification()==0 really implies that the block
no longer is in buf_pool.flush_list.

buf_dblwr_t::recover(): Issue a note, not a warning, about
pages whose FIL_PAGE_LSN is in the future. This was supposed to be
part of commit 762bcb81 (MDEV-25948)
but had been accidentally omitted.

ha_innobase::commit_inplace_alter_table(): Invoke
purge_sys.resume_FTS() on all error handling paths
if purge_sys.stop_FTS() had been called.

This fixes a regression that had been introduced in
commit 1bd681c8 (MDEV-25506).

trx_t::commit_in_memory(): Do not initiate a redo log write if
the transaction has no visible effect. If anything for this
transaction had been made durable, crash recovery will roll back
the transaction just fine even if the end of ROLLBACK is not
durably written.

Rollbacks of transactions that are associated with XA identifiers
(possibly internally via the binlog) will always be persisted.
The test rpl.rpl_gtid_crash covers this.

buf_LRU_get_free_block(): Initially wait for a single block to be
freed, signaled by buf_pool.done_free. Only if that fails and no
LRU eviction flushing batch is already running, we initiate a
flushing batch that should serve all threads that are currently
waiting in buf_LRU_get_free_block().

Note: In an extreme case, this may introduce a performance regression
at larger numbers of connections. We observed this in sysbench
oltp_update_index with 512MiB buffer pool, 4GiB of data on fast NVMe,
and 1000 concurrent connections, on a 20-thread CPU. The contention point
appears to be buf_pool.mutex, and the improvement would turn into a
regression somewhere beyond 32 concurrent connections.

On slower storage, such regression was not observed; instead, the
throughput was improving and maximum latency was reduced.

The excessive waits were pointed out by Vladislav Vaintroub.

MDEV-23855 changed the way how the page cleaner is signaled by
user threads. If a threshold is exceeded, a mini-transaction commit
would invoke buf_flush_ahead() in order to initiate page flushing
before all writers would eventually grind to halt in
log_free_check(), waiting for the checkpoint age to reduce.

However, buf_flush_ahead() would always initiate 'furious flushing',
making the buf_flush_page_cleaner thread write innodb_io_capacity_max
pages per batch, and sleeping no time between batches, until the
limit LSN is reached. Because this could saturate the I/O subsystem,
system throughput could significantly reduce during these
'furious flushing' spikes.

With this change, we introduce a gentler version of flush-ahead,
which would write innodb_io_capacity_max pages per second until
the 'soft limit' is reached.

buf_flush_ahead(): Add a parameter to specify whether furious flushing
is requested.

buf_flush_async_lsn: Similar to buf_flush_sync_lsn, a limit for
the less intrusive flushing.

buf_flush_page_cleaner(): Keep working until buf_flush_async_lsn
has been reached.

log_close(): Suppress a warning message in the event that a new log
is being created during startup, when old logs did not exist.
Return what type of page cleaning will be needed.

mtr_t::finish_write(): Also when m_log.is_small(), invoke log_close().
Return what type of page cleaning will be needed.

mtr_t::commit(): Invoke buf_flush_ahead() based on the return value of
mtr_t::finish_write().

buf_page_write_complete(): Reduce the buf_pool.mutex hold time,
and do not acquire buf_pool.flush_list_mutex at all.
Instead, mark blocks clean by setting oldest_modification to 1.
Dirty pages of temporary tables will be identified by the special
value 2 instead of the previous special value 1.
(By design of the ib_logfile0 format, actual LSN values smaller
than 2048 are not possible.)

buf_LRU_free_page(), buf_pool_t::get_oldest_modification()
and many other functions will remove the garbage (clean blocks)
from buf_pool.flush_list while holding buf_pool.flush_list_mutex.

buf_pool_t::n_flush_LRU, buf_pool_t::n_flush_list:
Replaced with non-atomic variables, protected by buf_pool.mutex,
to avoid unnecessary synchronization when modifying the counts.

export_vars: Remove unnecessary indirection for
innodb_pages_created, innodb_pages_read, innodb_pages_written.