Corrects: 7803601d

Starting with commit ef3f71fa
MemorySanitizer would complain that we are writing uninitialized
data via the doublewrite buffer.

buf_dblwr_t::add_to_batch(): Zero out any unused part of the
doublewrite buffer, for PAGE_COMPRESSED and ROW_FORMAT=COMPRESSED
tables.

Reviewed by: Eugene Kosov

This prevents the clash between NOT_FIXED_DEC differing between server
and plugins if MYSQL_SERVER is not defined during plugin compilation.

It's a virtual method and it can't be inlined anyway. This allows type
plugins (mysql_json in particular) to use Type_handler_blob and / or
subclass it, without needing to explicitly expose the
vers_type_timestamp object.

Cleanup other linker errors

As part of MDEV-23855, we eliminated fil_system.LRU
and changed the way how InnoDB data files are opened.
We are also enforcing the innodb_open_files limit when new
data files are created.

The function fil_space_t::flush() would be invoked by
row_quiesce_table_start(). If the table was already in clean state,
it is possible that the data file is not open.

fil_space_t::flush_low(): If the data file is not open,
check with a debug assertion that there are no unflushed changes,
and carry on.

Reviewed by: Eugene Kosov and Thirunarayanan Balathandayuthapani

If and only if read_variable_length() returns true, the variable
blob_length will be uninitialized and not used. For some reason,
GCC 10.2.0 -Og debug builds would issue a warning.

This patch solves two key problems.
1. There is a type number clash between MySQL and MariaDB. The number
   245, used for MariaDB Virtual Fields is the same as MySQL's JSON.
   This leads to corrupt FRM errors if unhandled. The code properly
   checks frm table version number and if it matches 5.7+ (until 10.0+)
   it will assume it is dealing with a MySQL table with the JSON
   datatype.
2. MySQL JSON datatype uses a proprietary format to pack JSON data. The
   patch introduces a datatype plugin which parses the format and convers
   it to its string representation.

The intended conversion path is to only use the JSON datatype within
ALTER TABLE <table> FORCE, to force a table recreate. This happens
during mysql_upgrade or via a direct ALTER TABLE <table> FORCE.

Convert the parameter to const as the function won't modify the pointer
value.

This avoids flushing file metadata on NTFS , and writing to <drive>:\$Log
file. With heavy write workload this can consume up to 1/3 of the
server's IO bandwidth.

Reviewed by : Marko

For a correlated subquery filesort is executed multiple times.
During each execution, sortlength() computed total sort key length in
Sort_keys::sort_length, without resetting it first.

Eventually Sort_keys::sort_length got larger than @@sort_buffer_size, which
caused filesort() to be aborted with error.

Fixed by making sortlength() to compute lengths only during the first
invocation. Subsequent invocations return pre-computed values.

Probably due to the changes to page flushing in MDEV-23399
(commit 7cffb5f6) the command
CHECK TABLE would occasionally report a different number of rows
for the corrupted secondary index. (The reported number was 991
instead of 990 on one occasion.)

Let us map all numbers to 990 in the output. We only care that the
injected corruption will be detected.

With an unreasonably small innodb_log_file_size, the page cleaner
thread would frequently acquire log_sys.flush_order_mutex and spend
a significant portion of CPU time spinning on that mutex when
determining the checkpoint LSN.

Synchronous writes and calls to fdatasync(), fsync() or
FlushFileBuffers() would ruin performance. So, let us
submit asynchronous writes for the doublewrite buffer.
We submit a single request for the likely case that the
two doublewrite buffers are contiquous in the system tablespace.

buf_dblwr_t::flush_buffered_writes_completed(): The completion callback
of buf_dblwr_t::flush_buffered_writes().

os_aio_wait_until_no_pending_writes(): Also wait for doublewrite batches.

buf_dblwr_t::element::space: Remove. We can simply use
element::request.node->space instead.

Reviewed by: Vladislav Vaintroub

Author: Vladislav Vaintroub

os_aio_userdata_t: Remove. It was basically duplicating IORequest.

buf_page_write_complete(): Take only IORequest as a parameter.

os_aio_func(), pfs_os_aio_func(): Replaced with os_aio() that has
no redundant parameters. There is only one caller, so there is no
point to pass __FILE__, __LINE__ as a parameter.

Merge n_pending_ios, n_pending_ops to std::atomic<uint32_t> n_pending.
Change some more fil_space_t members to uint32_t to reduce
the memory footprint.

fil_space_t::add(), fil_ibd_create(): Attach the already opened
handle to the tablespace, and enforce the fil_system.n_open limit.

dict_boot(): Initialize fil_system.max_assigned_id.

srv_boot(): Call srv_thread_pool_init() before anything else,
so that files should be opened in the correct mode on Windows.

fil_ibd_create(): Create the file in OS_FILE_AIO mode, just like
fil_node_open_file_low() does it.

dict_table_t::is_accessible(): Replaces fil_table_accessible().

Reviewed by: Vladislav Vaintroub

Also fixes MDEV-23929: innodb_flush_neighbors is not being ignored
for system tablespace on SSD

When the maximum configured number of file is exceeded, InnoDB will
close data files. We used to maintain a fil_system.LRU list and
a counter fil_node_t::n_pending to achieve this, at the huge cost
of multiple fil_system.mutex operations per I/O operation.

fil_node_open_file_low(): Implement a FIFO replacement policy:
The last opened file will be moved to the end of fil_system.space_list,
and files will be closed from the start of the list. However, we will
not move tablespaces in fil_system.space_list while
i_s_tablespaces_encryption_fill_table() is executing
(producing output for INFORMATION_SCHEMA.INNODB_TABLESPACES_ENCRYPTION)
because it may cause information of some tablespaces to go missing.
We also avoid this in mariabackup --backup because datafiles_iter_next()
assumes that the ordering is not changed.

IORequest: Fold more parameters to IORequest::type.

fil_space_t::io(): Replaces fil_io().

fil_space_t::flush(): Replaces fil_flush().

OS_AIO_IBUF: Remove. We will always issue synchronous reads of the
change buffer pages in buf_read_page_low().

We will always ignore some errors for background reads.

This should reduce fil_system.mutex contention a little.

fil_node_t::complete_write(): Replaces fil_node_t::complete_io().
On both read and write completion, fil_space_t::release_for_io()
will have to be called.

fil_space_t::io(): Do not acquire fil_system.mutex in the normal
code path.

xb_delta_open_matching_space(): Do not try to open the system tablespace
which was already opened. This fixes a file sharing violation in
mariabackup --prepare --incremental.

Reviewed by: Vladislav Vaintroub

After MDEV-15053, MDEV-22871, MDEV-23399 shifted the scalability
bottleneck, log checkpoints became a new bottleneck.

If innodb_io_capacity is set low or innodb_max_dirty_pct_lwm is
set high and the workload fits in the buffer pool, the page cleaner
thread will perform very little flushing. When we reach the capacity
of the circular redo log file ib_logfile0 and must initiate a checkpoint,
some 'furious flushing' will be necessary. (If innodb_flush_sync=OFF,
then flushing would continue at the innodb_io_capacity rate, and
writers would be throttled.)

We have the best chance of advancing the checkpoint LSN immediately
after a page flush batch has been completed. Hence, it is best to
perform checkpoints after every batch in the page cleaner thread,
attempting to run once per second.

By initiating high-priority flushing in the page cleaner as early
as possible, we aim to make the throughput more stable.

The function buf_flush_wait_flushed() used to sleep for 10ms, hoping
that the page cleaner thread would do something during that time.
The observed end result was that a large number of threads that call
log_free_check() would end up sleeping while nothing useful is happening.

We will revise the design so that in the default innodb_flush_sync=ON
mode, buf_flush_wait_flushed() will wake up the page cleaner thread
to perform the necessary flushing, and it will wait for a signal from
the page cleaner thread.

If innodb_io_capacity is set to a low value (causing the page cleaner to
throttle its work), a write workload would initially perform well, until
the capacity of the circular ib_logfile0 is reached and log_free_check()
will trigger checkpoints. At that point, the extra waiting in
buf_flush_wait_flushed() will start reducing throughput.

The page cleaner thread will also initiate log checkpoints after each
buf_flush_lists() call, because that is the best point of time for
the checkpoint LSN to advance by the maximum amount.

Even in 'furious flushing' mode we invoke buf_flush_lists() with
innodb_io_capacity_max pages at a time, and at the start of each
batch (in the log_flush() callback function that runs in a separate
task) we will invoke os_aio_wait_until_no_pending_writes(). This
tweak allows the checkpoint to advance in smaller steps and
significantly reduces the maximum latency. On an Intel Optane 960
NVMe SSD on Linux, it reduced from 4.6 seconds to 74 milliseconds.
On Microsoft Windows with a slower SSD, it reduced from more than
180 seconds to 0.6 seconds.

We will make innodb_adaptive_flushing=OFF simply flush innodb_io_capacity
per second whenever the dirty proportion of buffer pool pages exceeds
innodb_max_dirty_pages_pct_lwm. For innodb_adaptive_flushing=ON we try
to make page_cleaner_flush_pages_recommendation() more consistent and
predictable: if we are below innodb_adaptive_flushing_lwm, let us flush
pages according to the return value of af_get_pct_for_dirty().

innodb_max_dirty_pages_pct_lwm: Revert the change of the default value
that was made in MDEV-23399. The value innodb_max_dirty_pages_pct_lwm=0
guarantees that a shutdown of an idle server will be fast. Users might
be surprised if normal shutdown suddenly became slower when upgrading
within a GA release series.

innodb_checkpoint_usec: Remove. The master task will no longer perform
periodic log checkpoints. It is the duty of the page cleaner thread.

log_sys.max_modified_age: Remove. The current span of the
buf_pool.flush_list expressed in LSN only matters for adaptive
flushing (outside the 'furious flushing' condition).
For the correctness of checkpoints, the only thing that matters is
the checkpoint age (log_sys.lsn - log_sys.last_checkpoint_lsn).
This run-time constant was also reported as log_max_modified_age_sync.

log_sys.max_checkpoint_age_async: Remove. This does not serve any
purpose, because the checkpoints will now be triggered by the page
cleaner thread. We will retain the log_sys.max_checkpoint_age limit
for engaging 'furious flushing'.

page_cleaner.slot: Remove. It turns out that
page_cleaner_slot.flush_list_time was duplicating
page_cleaner.slot.flush_time and page_cleaner.slot.flush_list_pass
was duplicating page_cleaner.flush_pass.
Likewise, there were some redundant monitor counters, because the
page cleaner thread no longer performs any buf_pool.LRU flushing, and
because there only is one buf_flush_page_cleaner thread.

buf_flush_sync_lsn: Protect writes by buf_pool.flush_list_mutex.

buf_pool_t::get_oldest_modification(): Add a parameter to specify the
return value when no persistent data pages are dirty. Require the
caller to hold buf_pool.flush_list_mutex.

log_buf_pool_get_oldest_modification(): Take the fall-back LSN
as a parameter. All callers will also invoke log_sys.get_lsn().

log_preflush_pool_modified_pages(): Replaced with buf_flush_wait_flushed().

buf_flush_wait_flushed(): Implement two limits. If not enough buffer pool
has been flushed, signal the page cleaner (unless innodb_flush_sync=OFF)
and wait for the page cleaner to complete. If the page cleaner
thread is not running (which can be the case durign shutdown),
initiate the flush and wait for it directly.

buf_flush_ahead(): If innodb_flush_sync=ON (the default),
submit a new buf_flush_sync_lsn target for the page cleaner
but do not wait for the flushing to finish.

log_get_capacity(), log_get_max_modified_age_async(): Remove, to make
it easier to see that af_get_pct_for_lsn() is not acquiring any mutexes.

page_cleaner_flush_pages_recommendation(): Protect all access to
buf_pool.flush_list with buf_pool.flush_list_mutex. Previously there
were some race conditions in the calculation.

buf_flush_sync_for_checkpoint(): New function to process
buf_flush_sync_lsn in the page cleaner thread. At the end of
each batch, we try to wake up any blocked buf_flush_wait_flushed().
If everything up to buf_flush_sync_lsn has been flushed, we will
reset buf_flush_sync_lsn=0. The page cleaner thread will keep
'furious flushing' until the limit is reached. Any threads that
are waiting in buf_flush_wait_flushed() will be able to resume
as soon as their own limit has been satisfied.

buf_flush_page_cleaner: Prioritize buf_flush_sync_lsn and do not
sleep as long as it is set. Do not update any page_cleaner statistics
for this special mode of operation. In the normal mode
(buf_flush_sync_lsn is not set for innodb_flush_sync=ON),
try to wake up once per second. No longer check whether
srv_inc_activity_count() has been called. After each batch,
try to perform a log checkpoint, because the best chances for
the checkpoint LSN to advance by the maximum amount are upon
completing a flushing batch.

log_t: Move buf_free, max_buf_free possibly to the same cache line
with log_sys.mutex.

log_margin_checkpoint_age(): Simplify the logic, and replace
a 0.1-second sleep with a call to buf_flush_wait_flushed() to
initiate flushing. Moved to the same compilation unit
with the only caller.

log_close(): Clean up the calculations. (Should be no functional
change.) Return whether flush-ahead is needed. Moved to the same
compilation unit with the only caller.

mtr_t::finish_write(): Return whether flush-ahead is needed.

mtr_t::commit(): Invoke buf_flush_ahead() when needed. Let us avoid
external calls in mtr_t::commit() and make the logic easier to follow
by having related code in a single compilation unit. Also, we will
invoke srv_stats.log_write_requests.inc() only once per
mini-transaction commit, while not holding mutexes.

log_checkpoint_margin(): Only care about log_sys.max_checkpoint_age.
Upon reaching log_sys.max_checkpoint_age where we must wait to prevent
the log from getting corrupted, let us wait for at most 1MiB of LSN
at a time, before rechecking the condition. This should allow writers
to proceed even if the redo log capacity has been reached and
'furious flushing' is in progress. We no longer care about
log_sys.max_modified_age_sync or log_sys.max_modified_age_async.
The log_sys.max_modified_age_sync could be a relic from the time when
there was a srv_master_thread that wrote dirty pages to data files.
Also, we no longer have any log_sys.max_checkpoint_age_async limit,
because log checkpoints will now be triggered by the page cleaner
thread upon completing buf_flush_lists().

log_set_capacity(): Simplify the calculations of the limit
(no functional change).

log_checkpoint_low(): Split from log_checkpoint(). Moved to the
same compilation unit with the caller.

log_make_checkpoint(): Only wait for everything to be flushed until
the current LSN.

create_log_file(): After checkpoint, invoke log_write_up_to()
to ensure that the FILE_CHECKPOINT record has been written.
This avoids ut_ad(!srv_log_file_created) in create_log_file_rename().

srv_start(): Do not call recv_recovery_from_checkpoint_start()
if the log has just been created. Set fil_system.space_id_reuse_warned
before dict_boot() has been executed, and clear it after recovery
has finished.

dict_boot(): Initialize fil_system.max_assigned_id.

srv_check_activity(): Remove. The activity count is counting transaction
commits and therefore mostly interesting for the purge of history.

BtrBulk::insert(): Do not explicitly wake up the page cleaner,
but do invoke srv_inc_activity_count(), because that counter is
still being used in buf_load_throttle_if_needed() for some
heuristics. (It might be cleaner to execute buf_load() in the
page cleaner thread!)

Reviewed by: Vladislav Vaintroub

buf_flush_remove_pages(), buf_flush_dirty_pages(): Because
buf_page_t::state() is protected by buf_pool.mutex, which we
are not holding, the state may be BUF_BLOCK_REMOVE_HASH when
the page is being relocated. Let us relax these assertions
similar to buf_flush_validate_low().

The other in_file() assertions in buf0flu.cc look valid.

srv_start(): Avoid trx_lists_init_at_db_start() for normal
mariabackup --prepare without --export.

innodb_preshutdown(): Terminate the encryption threads before
the page cleaner thread can be shut down.

innodb_shutdown(): Always wait for the encryption threads and
page cleaner to shut down.

srv_shutdown_all_bg_threads(): Wait for the encryption threads and
the page cleaner to shut down. (After an aborted startup,
innodb_shutdown() would not be called.)

row_get_background_drop_list_len_low(): Remove.

os_thread_count: Remove. Alternatively, at the end of
srv_shutdown_all_bg_threads() we could try to wait longer
for the count to reach 0. On some platforms, an assertion
os_thread_count==0 could fail even after a small delay,
even though in the core dump all threads would have exited.

srv_shutdown_threads(): Renamed from srv_shutdown_all_bg_threads().
Do not wait for the page cleaner to shut down, because the later
innodb_shutdown(), which may invoke
logs_empty_and_mark_files_at_shutdown(), assumes that it exists.

storage engines are generally initialized in some random order
(by iterating the hash of plugin names).

S3 fails if it's initialized before Aria.

But it looks that while S3 needs Aria, it does not need
Aria to be initialized before S3. S3 copies maria_hton and then
overwrites every single member of it, so it can handle Aria being
initialized after S3.

fix TokuDB to compile with the perfschema

view_pseudo_hton cannot be dereferenced

first try to resolve the function name as a native function,
and only then look within pluggable data types

Fix the Item_func_json_objectagg::fix_fields to save the orig_args.

Ubuntu Xenial

Fixes 7f613ebd (MDEV-7284 INDEX: CREATE OR REPLACE).