There was 2 problems:
  1) coping/moving of the same type (usually casting) as sizeof() (solved in different ways depends on the cause);
  2) using 'const' in SSL_CTX::getVerifyCallback() which return object (not reference) and so copy of the object will be created and 'const' has no sens.

Analysis:
st_select_lex_unit::prepare() computes can_skip_order_by as TRUE.
As a result join->prepare() gets called with order == NULL, and
doesn't do name resolution for the inner ORDER clause. Due to this
the prepare phase doesn't detect that the query references non-exiting
function and field.
  
Later join->optimize() calls update_used_tables() for a non-resolved
Item_field, which understandably has no Field object. This call results
in a crash.

Solution:
Resolve unnecessary ORDER BY clauses to detect if they reference non-exising
objects. Then remove such clauses from the JOIN object.

  GRANT ... IDENTIFIED BY [ PASSWORD ] ...
  GRANT ... IDENTIFIED VIA ... [ USING ... ]
  GRANT ... REQUIRE ...
  GRANT ... MAX_xxx ...
  SET PASSWORD FOR ... = ...

- Backport MySQL's fix: do set ha_partition::m_pkey_is_clustered for ha_partition 
  objects created with handler->clone() call.
- Also, include a testcase.

MDEV-5176 Server crashes in fill_schema_applicable_roles on select from APPLICABLE_ROLES after a suicide

Don't assume that thd->security_ctx->priv_user is an actually existing user account

MDEV-5170 Assertion `(&(&acl_cache->lock)->m_mutex)->count > 0 && pthread_equal(pthread_self(), (&(&acl_cache->lock)->m_mutex)->thread)' fails after restarting server with a pre-created role grants

lock acl_cache->lock mutex for the duration of acl_load

test time increased to be working on all timezones.

MDEV-5172 safe_mutex: Trying to lock mutex when the mutex was already locked on using a role and I_S role tables

don't forget to unlock if the current role isn't found

only remove this role from the arrays where it is present.

Slackware install docs to different folder, which wasn't handled
by file_contents test properly. E.g. on Redhat/Debian docs are
expected to be in /usr/share/doc/mariadb-server-(version), but
on Slackware they are in /usr/doc/mariadb-(version).

For details see Slackware mariadb build script:
http://slackbuilds.org/slackbuilds/14.0/system/mariadb/mariadb.SlackBuild

MDEV-5143: update of a joined table with a nested subquery with a syntax error crashes mysqld with signal 11

Added check of SELECT_LEX::handle_derived() result.

FLUSH PRIVILEGES no longer returns an error, when it was successful.
LOCK_grant is no longed unlocked/relocked between tables_priv and procs_priv

Only allow NONE instead of a role name in SET ROLE.
Don't allow PUBLIC as a role name anywhere (to be fixed later)
Fix db_access calculations on SET ROLE
Reduce the size of role_grants and parent_grantee per-user/role arrays.
Fix the wording and specify the correct sqlstate for ER_INVALID_ROLE

functions for traversing the role graph in either direction.
merging of global, database, table, column, routine privileges.
debug status variables for counting number of privilege merges.
tests.

Move key function from template parameter to the constructor

mainly to avoid the pattern of
* get username/hostname/rolename
* optionally find the corresponding ACL_USER and ACL_ROLE
* allocate memory, concatenate username/hostname/rolename
* call a function passing only this memory as an argument
** use concatenated username/etc to find ACL_USER and ACL_ROLE again
** do something
* free the object

Also to undo push_dynamic we use pop_dynamic now,
not a linear search/scan through the dynamic array.

as a bonus, role@ is now an invalid way to refer to a role.