- Instant alter should change the metadata alone when table is
discarded. It shouldn't try to add metadata record in clustered index.
Also make the clustered index to non-instant format.

- ALTER_ALGORITHM should be substituted when there is no mention of
algorithm in alter statement.
- Introduced algorithm(thd) in Alter_info. It returns the
user requested algorithm. If user doesn't specify algorithm explicitly then
it returns alter_algorithm variable.
- changed algorithm() to get_algorithm(thd) to return algorithm name for
displaying the error.
- set_requested_algorithm(algo_value) to avoid direct assignment on
requested_algorithm variable.
- Avoid direct access of requested_algorithm to encapsulate
requested_algorithm variable

- Added the test case to prove that alter_algorithm variable doesn't
affect when ALTER statement during PREPARE PHASE or CREATE PROCEDURE.
Only when execution/call happens, alter uses the alter_algorithm
variable when user does not mention algorithm explicitly

For a unique key if all the keyparts are NOT NULL or the predicates involving
the keyparts is NULL rejecting, then we can use EQ_REF access instead of ref
access with the unique key

Maybe this patch will help catch problems like buffer overflow.

log_t::first_in_use: removed

log_t::buf: this is where mtr_t are supposed to append data
log_t::flush_buf: this is from server writes to a file

Those two buffers are std::swap()ped when some thread is gonna write
to a file

During the UPDATE of PRIMARY KEY columns, we may miscalculate the
size of the clustered index record.

row_upd_clust_rec_by_insert(): Pass the total number of off-page columns,
which may include such columns that were inherited from the record
and not created as part of the UPDATE operation.

This is based on
mysql/mysql-server@490c45e8c8e07197958dbb21214fd45ed668b559
which is a follow-up to
mysql/mysql-server@1fa475b85d24de4b9ce2958c0eed738c221fc82c
which we filed and fixed as MDEV-21511.

No test case was provided by Oracle.

INSERT...SELECT reading from an InnoDB table is slow due to
creating explicit record locks. Use the sequence engine instead.

Also, remove the space before rtr_page_need_second_split
to actually make the debug injection work.

when detecting running mysqld processes to stop,
ignore processes in different namespaces.

Several MYSQL_SYSVAR_STR parameters that employ both a validate
function callback fail to copy the string for saving the
validated value. The affected variables include the following:

innodb_ft_aux_table
innodb_ft_server_stopword_table
innodb_ft_user_stopword_table
innodb_buffer_pool_filename

The test case is an enhanced version of
mysql/mysql-server@0b0c30641fd66336e87394ac28587e40864f8af9
and the code changes are inspired by their fixes.

We are also importing and adjusting the test innodb_fts.stopword
to get coverage for the variable innodb_ft_user_stopword_table.

buf_dump(), buf_load(): Protect srv_buf_dump_filename with
LOCK_global_system_variables.

fts_load_user_stopword(): Minor cleanup

fts_load_stopword(): Remove the parameter global_stopword_table.

innobase_fts_load_stopword(): Protect innodb_server_stopword_table
against concurrent SET GLOBAL.

innodb_buffer_pool_evict_uncompressed(): Restart the loop when
prev_block might not enjoy mutex protection.

This is based on
mysql/mysql-server@eccaecac070b6747ecf14d6b9150791f8c3e8f6d

- Inplace alter shouldn't set default date column as '0000-00-00' when
table is not empty. So mysql_inplace_alter_table() copied
alter_ctx.error_if_not_empty to a new field of Alter_inplace_info.
In ha_innobase::check_if_supported_inplace_alter() should check the
error_if_not_empty flag and return INPLACE_NOT_SUPPORTED if the table
is not empty

lock

--ftwrl-wait-timeout does not finish mariabackup execution when acquired
backup lock can't be grabbed for the certain amount of time, it just
waits for a long queries finishing before acquiring the lock to avoid
unnecessary locking.

This commit extends --ftwrl-wait-timeout so, that mariabackup execution
is finished if it waits for backup lock during certain amount of time.

dict_stats_update_if_needed(): Replace the parameter THD*
with const trx_t& so that trx_t::is_wsrep() can be invoked
instead of the more expensive wsrep_on().

Replace also other occurrences of wsrep_on() with trx_t::is_wsrep().

The only change is a change of the version number.
As noted in commit 02af6278
there were no changes to InnoDB between MySQL 5.6.46 and 5.6.47
either.

The function wsrep_on() was being called rather frequently
in InnoDB and XtraDB. Let us cache it in trx_t and invoke
trx_t::is_wsrep() instead.

innobase_trx_init(): Cache trx->wsrep = wsrep_on(thd).

ha_innobase::write_row(): Replace many repeated calls to current_thd,
and test the cheapest condition first.

fix embedded innodb_plugin tests

followup for 7198c6ab

Backport INSALL_SYMLINK cmake macro from 10.3.
It'll make libmysqlclient_r.* symlinks to link to the actual
shared library file, not to another symlink.

As a bonus it'll fix cmake warning about deprecated LOCATION property

This is a backport of the applicable part of
commit 93475aff and
commit 2c39f69d
from 10.4.

Before 10.4 and Galera 4, WSREP_ON is a macro that points to
a global Boolean variable, so it is not that expensive to
evaluate, but we will add an unlikely() hint around it.

WSREP_ON_NEW: Remove. This macro was introduced in
commit c863159c
when reverting WSREP_ON to its previous definition.

We replace some use of WSREP_ON with WSREP(thd), like it was done
in 93475aff. Note: the macro
WSREP() in 10.1 is equivalent to WSREP_NNULL() in 10.4.

Item_func_rand::seed_random(): Avoid invoking current_thd
when WSREP is not enabled.

_ma_fetch_keypage(): Correct an assertion that used to always hold.
Thanks to clang -Wint-in-bool-context for flagging this.

double_to_datetime_with_warn(): Suppress -Wimplicit-int-float-conversion
by adding a cast. LONGLONG_MAX converted to double will actually be
LONGLONG_MAX+1.

Since commit 7198c6ab
the ./mtr --embedded tests would fail to start innodb_plugin
because of an undefined reference to the symbol wsrep_log().

Let us define a stub for that function. The embedded server
is never built WITH_WSREP, but there are no separate storage
engine builds for the embedded server. Hence, by default,
the dynamic InnoDB storage engine plugin would be built WITH_WSREP
and it would fail to load into the embedded server library due to
a reference to the undefined symbol.

read TLS with my_thread_var
write TLS with set_mysys_var()

my_thread_var is no longer __attribute__ ((const)): this attribute
is simply incorrect here. Read gcc manual for more information.
sql/threadpool_generic.cc fails with that attribute.

row_vers_vc_matches_cluster(): Remove the parameter in_purge,
which was always passed as in_purge=true.

This parameter became constant in
mysql/mysql-server@1dec14d346ac55fe72989dccb071f84b3b0d3bd6
and it always was constant in MariaDB starting from the
introduction of the function in
commit 2e814d47 (MariaDB 10.2.2).

row_prebuilt_free(): Do not attempt to drop orphan indexes
that might have been left behind by a failed ADD UNIQUE INDEX.
This avoids the execution of unwanted transactions during shutdown.

mysqld_exit(): Change the assertion failure on
global_status_var.global_memory_used == 0
to fprintf, like in 0bcb65d3

It appears that in some cases, that variable may be nonzero
even when LeakSanitizer (WITH_ASAN) would not report errors.
This was observed in 10.4 88cf6f1c
with the MDEV-22348 test case (Aria startup failure when running
main.default_storage_engine).