10.1 man pages

MDEV-8923: port innodb_buffer_pool_dump_pct from MySQL

WL#6504 InnoDB buffer pool dump/load enchantments

This patch consists of two parts:

1. Dump only the hottest N% of the buffer pool(s)
2. Prevent hogging the server duing BP load

From MySQL - commit b409342c43ce2edb68807100a77001367c7e6b8e

Add testcases for innodb_buffer_pool_dump_pct.

Part of the code authored by Daniel Black.

- Added missning setting of table->rpl_write_set in record_gtid(), required by galera
- Removed output of WSREP_PATCH_VERSION from galera_defaults, as this can change over time
- Limit galera_many_tables_pk and galera_many_tables_nopk to 900, as
  on many systems the default open table limit is 1024

As 10.1 has systemd support, there is no longer need to set
_SYSTEMCTL_SKIP_REDIRECT/SYSTEMD_NO_WRAP.

Extended my_timer_cycles() to support s390.
Some compiler tunings for RHEL/SLES RPM packages on s390.

myisam_recover is only an alias for myisam_recover_options.
Use the option name instead of the alias is more correct.
This also avoids an note about the use of an alias instead of the option
name on every server start.

by C runtime debug check.

Add testcases for innodb_buffer_pool_dump_pct

WL#6504 InnoDB buffer pool dump/load enchantments

This patch consists of two parts:

1. Dump only the hottest N% of the buffer pool(s)
2. Prevent hogging the server duing BP load

From MySQL - commit b409342c43ce2edb68807100a77001367c7e6b8e

no engines that don't support HA_READ_PREV)

Also, fixing compilation warnings in ctype-mb.ic (Windows).

Post-fix: The test case pushed with the fix had each node
acting as slave to the other two nodes with different set
of filters on server_id's. The slave's gtid_slave_pos is
updated after it processes the events received from master
nodes irrespective of whether the events were filtered
or not. Thus, sync_with_master_gtid.inc could unblock even
on filtered events.
As a result, sync_with_master_gtid.inc would fail to block
until the desired changes have been replicated.  Fixed by
simplifying the topology.
Also, modified CHANGE MASTER commands to ignore based
on gtid_domain_id instead of server_id.

Provided IBM System Z have outdated compiler version, which supports gcc sync
builtins but not gcc atomic builtins. It also has weak memory model.

InnoDB attempted to verify if __sync_lock_test_and_set() is available by
checking IB_STRONG_MEMORY_MODEL. This macro has nothing to do with availability
of __sync_lock_test_and_set(), the right one is HAVE_ATOMIC_BUILTINS.

thread_pool_size is auto-sized before my_getopt(). But my_getopt starts from
resetting all options to their default values. So the auto-sized value is lost.

Fixed by moving thread_pool_size auto-sizing after my_getopt().

wsrep.cmake should be included before cpack_rpm.cmake as it defines
WITH_WSREP, required by cpack_rpm.cmake to conditionally include galera
helper tools in server package's requires list.

--with-plugin-name=xxxx
--with-plugins= now uses =AUTO instead of =1

1. define connect_EXPORTS, this causes the engine to use MariaDB
   versions for timestamp<->struct tm conversion instead of
   TZ-dependent libc versions.
2. remove check_access() that was removed once, but re-appeared
   during a complex merge.
3. disable a totally broken test
4. update test results
5. skip odbc_firebird test when no firebird DSN is available

Apply MySQL fix for bug#76727:
https://github.com/mysql/mysql-server/commit/69d4e72c

  Bug#20901025: SLAVE ASSERTION IN UNPACK_ROW WITH ROLLBACK TO
  SAVEPOINT IN ERROR HANDLER

feedback was bzero-ing thd->status_vars.memory_used,
but server was asserting the correct value of it

RENAME TABLE code tries to update EITS statistics. It hung, because
it used an index on (db_name,table_name) to find the table, and attempted
to update these values at the same time. The fix is do what SQL UPDATE
statement does when updating index that it's used for scanning:
- First, buffer the rowids of rows to be updated,
- then make the second pass to actually update the rows

Also fixed the call to rename_table_in_stat_tables() in sql_rename.cc
to pass the correct new database (before, it passed old db_name so cross-
database renames were not handled correctly).

Variant #2, with review feedback addressed.

rpl/rpl_mdev382     ; Wrong replace in show_binlog_events2.inc
binlog/database     ; Different error on Solaris if file exists
mroonga/repair_table_no_index_file ; Different system error on Solaris
partition_not_blackhole ; Different error on Solaris
partition_myisam     ; Different error on Solaris

Some other failures in mroonga was because have_32bit.inc didn't correctly
detect 64 bits on Solaris. Fixed using DEFAULT_MACHINE instead of MACHINE_TYPE
for Sys_version_compile_machine.