storage/rocksdb/rdb_datadic.cc: In member function 'int myrocks::Rdb_key_def::unpack_integer(myrocks::Rdb_field_packing*, Field*, uchar*, myrocks::Rdb_string_reader*, myrocks::Rdb_string_reader*) const'
storage/rocksdb/rdb_datadic.cc:1781:1: internal compiler error: Segmentation fault
 }

on ppc64le, ubuntu bionic gcc 7.3.0 and debian stretch gcc 6.3.0

The error happens with -ftree-loop-vectorize when trying to vectorize
a particular loop (see Rdb_key_def::unpack_integer())

Compiler gets confused by __attribute__((optimize("O0")) that comes from
ha_rocksdb_proto.h. The intention of this __attribute__ was to prevent
function from being inlined (see ha_rocksdb.cc). Let's use a more
specific attribute that prevents inlining but does not confuse
loop vectorizer.

An error in "group commit with MariaDB's binlog" code: we would flush
the WAL even when the transaction did not do any writes (and so the logic
in myrocks::Rdb_transaction::commit caused it to rollback).

Reset query cache after every test case and add wait after
load infile.

to guarantee that it's destroyed when plugin deinit is called, not after

Use RedHat-recommended /var/lib/rpm-state/mariadb/ for restart flag.
This also fixes SUSE where /var/lib/rpm-state doesn't necessarily exist.

cover ALTER TABLE

MDEV-16792 Assertion `m_status == DA_ERROR' failed in Diagnostics_area::sql_errno or wrong result upon SHOW TABLE STATUS after adding partition under ANSI_QUOTES

if opening a table fails with a syntax error, the error shouldn't
be suppressed completely as if it didn't happen at all.

Unary minus operation for the smallest possible signed long long value
(LONLONG_MIN) is undefined in C++. Because of this, func_time.test
failed on ppc64 buildbot machines.

Fixing the code to avod using undefined operations.

This is fix is similar to "MDEV-7973 bigint fail with gcc 5.0"

RPM solution:

Make all server plugins to restart the server when installed.
To avoid multiple server restarts, do it only once in posttrans scriptlet.
Add support for CPACK_RPM_<component>_POST_TRANS_SCRIPT_FILE

alloc_root(): unpoison only requested amount of bytes instead of a
possible bigger aligned-sized buffer.

For the original test in 10.0 it was not really important if
find_user_wild() or find_user_exact() is used in sp_grant_privileges().
sp-security.test passed with either of them.

Fixing the test so it reliably fails with find_user_wild()
and pass with find_user_exact().

Make different threads that are running data_copy_thread_func()
use the same pthread_mutex for locking (not the copies of the same
pthread_mutex_t).

Simplify, and make it work with system tablespace outside of
innodb data home.

Also, do not reread TRX_SYS page  in endless loop,
if it appears to be corrupted.

Use finite number of attempts.

		if custom undo tablespace is defined

- In case of multiple undo tablespace, mariabackup have to open system
tablespace to find the list of undo tablespace present in TRX_SYS page.
For opening system tablespace, mariabackup should fetch the file name
from already initialized system tablespace object.

Add wsrep_sync_wait as we want INSERT to fail.

Wait in second node until tables with databases are created.

Test changes only.

Test changes only.

Added forgotten code signing command.

             table->pos_in_locked_tables->table == table'
             failed in mark_used_tables_as_free_for_reuse

Assertion failure can be triggered by some DDL executed under LOCK TABLES
that holds lock for DDL target table multiple times (either explicitly or
implcitly).

When closing all table instances for given table (e.g. when preparing for
table removal during CREATE OR REPLACE), only one instance was removed
from m_locked_tables list.

Later we attempt to re-insert one of the instances in mysql_create_table()/
add_back_last_deleted_lock(), which wasn't actually removed. This leads
to m_locks_tables corruption, specifically loss of all following elements.

Then UNLOCK TABLE won't reset some table instances properly (specifically
pos_in_locked_tables), since they're not present in m_locked_tables.

Eventually such table instance gets released to table cache and then
re-used by subsequent statement, which triggers this assertion failure.

The function st_select_lex_unit::exec_recursive() missed resetting of
select_limit_cnt and offset_limit_cnt before execution of union parts.
As a result recursive CTEs specified by UNIONs whose SELECTs contained
LIMIT/OFFSET could return wrong sets of records.