- InnoDB fails to reset the after_apply variable before applying
the redo log in last batch during multi-batch recovery.

- During non-last batch of multi-batch recovery, InnoDB holds
log_sys.mutex and preallocates the block which may intiate
page flush, which may initiate log flush, which requires
log_sys.mutex to acquire again. This leads to assert failure.
So InnoDB recovery should release log_sys.mutex before
preallocating the block.

As requested to the UN the country formerly known as Turkey is
to be refered to as Türkiye.

Reviewer: Alexander Barkov

MDEV-30608: rpl.rpl_delayed_parallel_slave_sbm sometimes fails with Seconds_Behind_Master should not have used second transaction timestamp

One of the constraints added in the MDEV-29639 patch, is that only
the first event after idling should update last_master_timestamp;
and as long as the replica has more events to execute, the variable
should not be updated. The corresponding test,
rpl_delayed_parallel_slave_sbm.test, aims to verify this; however,
if the IO thread takes too long to queue events, the SQL thread can
appear to catch up too fast.

This fix ensures that the relay log has been fully written before
executing the events.

Note that the underlying cause of this test failure needs to be
addressed as a bug-fix, this is a temporary fix to stop test
failures. To track work on the bug-fix for the underlying issue,
please see MDEV-30619.

The parser code for single-table DELETE missed the call of the function
LEX::check_main_unit_semantics(). As a result the the field nested level
of SELECT_LEX structures remained set 0 for all non-top level selects.
This could lead to different kind of problems. In particular this did not
allow to determine properly the selects where set functions had to be
aggregated when they were used in inner subqueries.

Approved by Oleksandr Byelkin <sanja@mariadb.com>

This patch is the result of running
run-clang-tidy -fix -header-filter=.* -checks='-*,modernize-use-equals-default' .

Code style changes have been done on top. The result of this change
leads to the following improvements:

1. Binary size reduction.
* For a -DBUILD_CONFIG=mysql_release build, the binary size is reduced by
  ~400kb.
* A raw -DCMAKE_BUILD_TYPE=Release reduces the binary size by ~1.4kb.

2. Compiler can better understand the intent of the code, thus it leads
   to more optimization possibilities. Additionally it enabled detecting
   unused variables that had an empty default constructor but not marked
   so explicitly.

   Particular change required following this patch in sql/opt_range.cc

   result_keys, an unused template class Bitmap now correctly issues
   unused variable warnings.

   Setting Bitmap template class constructor to default allows the compiler
   to identify that there are no side-effects when instantiating the class.
   Previously the compiler could not issue the warning as it assumed Bitmap
   class (being a template) would not be performing a NO-OP for its default
   constructor. This prevented the "unused variable warning".

The -D flag was not passed to asm compiler, despite SET_PROPERTY(COMPILE_OPTIONS)
The exact reason for that remains unknown.  It was not seen with gcc, as
nor was be reproduced on newer CMake.

Debian 9 has EOL July 6th, 2020. This commit cleans it up
from debian/autobake-deb.sh which is used to build official
versions of MariaDB

Correct error in to only say "continuing to smaller size" if it really
is.

The existing storage/rocksdb/CMakeCache.txt defined
ATOMIC_EXTRA_LIBS when atomics where required. This was
determined by the toplevel configure.cmake test
(HAVE_GCC_C11_ATOMICS_WITH_LIBATOMIC).

As build_rocksdb.cmake is included after ATOMIC_EXTRA_LIBS
was set, we just need to use it. As such no riscv64
specific macro is needed in build_rocksdb.cmake.

As highlighted by Gianfranco Costamagna (@LocutusOfBorg)
in #2472 overwriting SYSTEM_LIBS was problematic.
This is corrected in case in future SYSTEM_LIBS is changed
elsewhere.

Closes #2472.

Normalize innodb_flush_method, the same as the service, before
attempting to print it.

The error string from ER_KILL_QUERY_DENIED_ERROR took a different
type to ER_KILL_DENIED_ERROR for the thread id. This shows
up in differences on 32 big endian arches like powerpc (Deb notation).

Normalize the passing of the THD->id to its real type of my_thread_id,
and cast to (long long) on output. As such normalize the
ER_KILL_QUERY_DENIED_ERROR to that convention too.

Note for upwards merge, convert the type to %lld on new translations
of ER_KILL_QUERY_DENIED_ERROR.

recv_sys_t::parse(): Discard old page-level redo log when parsing
a TRIM_PAGES record.

recv_sys_t::apply(): trim() was invoked in parse() already.

recv_sys_t::truncated_undo_spaces[]: Only store the size, no LSN.

page_recv_t::trim(): Do remove log records for mini-transactions
that end right at the threshold LSN. This will avoid an inconsistency
where a dirty page had been evicted from the buffer pool during
undo tablespace truncation, and recovery would attempt to apply
log records for which the last available copy in the data file is
too new. These changes would be discarded anyway.

Warning on a normal graceful disconnnect is excessive, so lets not do
it.

MDEV-19282 restructed the code from 10.3 so applying this as a 10.4+
fix.

__gcov_flush was never an external symbol in the documentation.

It was removed in gcc-11. The correct function to use is __gcov_dump
which is defined in the gcov.h header.

This patch allowed transformation of EXISTS subqueries into equivalent
IN predicands at the top level of WHERE conditions for multi-table UPDATE
and DELETE statements. There was no reason to prohibit the transformation
for such statements. The transformation provides more opportunities of
using semi-join optimizations.

Approved by Oleksandr Byelkin <sanja@mariadb.com>

When running with DBUG trace enabled, print all current active signals.

The output looks like this with the signals present in curr:
T@6 ... debug_sync_exec: wait for 'nothing'  at: 'now', curr: 'something,from_function,from_myvar'

After introducing multiple signals possible for debug_sync, the test
changes are no longer necessary. Revert them to the original state.

One can now do:
set DEBUG_SYNC='... SIGNAL s1,s2,s3...'

Multiple signals can be fired, they need to be split by commas.

One shouldn't have to instantiate a Binary_string every time a strstr
call is needed.

When waiting on a signal, NO_CLEAR_EVENT allows one to not clear the
signal, effectively allowing other threads to wait for the same signal.

The patch is inspired from MySQL. Instead of using a single String to
hold the current active debug_sync signal, use a Hash_set to store
LEX_STRINGS. This patch ensures that a signal can not be lost, by being
overwritten by another thread via set DEBUG_SYNC = '... SIGNAL ...';

All signals are kepts "alive" until they are consumed by a wait event.

This requires updating test cases that assume the GLOBAL signal is never
consumed.

Follow-up work needed:
Port the additional syntax that allows one to set multiple signals
and also conditionally deactivate signals when waiting.

If the Hash_set is not initialized, one can not call find() on it before
at least one element has been inserted into it.