revno: 4301
committer: Alexey Botchkov <holyfoot@askmonty.org>
branch nick: 10exp
timestamp: Tue 2014-07-22 15:28:15 +0500
message:
  gis-precise.test fixed to work on Power8.
------------------------------------------------------------
revno: 4295
committer: Alexey Botchkov <holyfoot@askmonty.org>
branch nick: 10exp
timestamp: Mon 2014-07-21 13:07:48 +0500
message:
  gis-precise test fixed to pass on Power8.

The data base lenth passed to invalidator fixed

* skip debian 44_scripts__mysql_config__libs.dpatch it does not apply anymore
  (and anyway it would not work for a static library)
* fix the path for install(mariadb.pc)

MDEV-5553 A view or procedure with a non existing definer can block "SHOW TABLE STATUS" with an unclear error message

Don't double-check privileges for a column in the GROUP BY that refers to
the same column in SELECT clause. Privileges were already checked for SELECT clause.

skip NULL VARCHAR key parts like it's done elsewhere

typo fixed. use correct constant.

mysqlimport had code for multi-threaded import. By mistake it was disabled for
many years, at least since 5.5 (more likely even in 5.1), but mysqlimport happily
accepted (and ignored) --use-threads option ever since.

After Mroonga merge HAVE_LIBPTHREAD became defined and multi-threaded import
suddenly came to life. As it exit() the program brutally on any error (never mind
that some import threads may be running) mysys rightfully complains. Safemalloc
complains too in debug builds.

Fix: don't try a clean exit on an error, don't shutdown mysys and tell safemalloc
to shut up. Yes, and remove #ifdef HAVE_LIBPTHREAD, since 5.5 the client library
is always multi-threaded.

and compiler warnings

backport the new fix from 10.0

1. move cflags/libs cleanup from mysql_config.sh (runtime)
   to for_client.cmake (build time). Include/library paths are
   still calculated at runtime (they depend on the location of mysql_config)

2. Use cleaned-up cflags/libs to generate mariadb.pc

3. remove obsolete @expansions@ from mysql_config (for variables that are
   never set in cmake files)

add bsdutils dependency, because mariadb-server-5.5 deb package uses logger

apply debian's patch to workaround kfreebsd bug

and apply whatever was reasonable

typo (thanks, Brad Smith!)

stat structure (from <sys/stat.h>) is conditionally defined
to have different layout and size depending on the defined macros.
The correct macro is defined in my_config.h, which means it MUST be
included first (or, at least before <features.h> - so, practically,
before including any system headers).

  SQLCOM_ASSIGN_TO_KEYCACHE should not be CF_AUTO_COMMIT_TRANS
  SQLCOM_PRELOAD_KEYS should not be CF_AUTO_COMMIT_TRANS
  SQLCOM_INSTALL_PLUGIN should need CF_AUTO_COMMIT_TRANS
  SQLCOM_UNINSTALL_PLUGIN should need CF_AUTO_COMMIT_TRANS

merge from MySQL-5.6, revision:

revno: 3677.2.1
committer: Alfranio Correia <alfranio.correia@oracle.com>
timestamp: Tue 2012-02-28 16:26:37 +0000
message:
  BUG#13627921 - MISSING FLAGS IN SQL_COMMAND_FLAGS MAY LEAD TO REPLICATION PROBLEMS

  Flags in sql_command_flags[command] are not correctly set for the following
  commands:

    . SQLCOM_SET_OPTION is missing CF_CAN_GENERATE_ROW_EVENTS;
    . SQLCOM_BINLOG_BASE64_EVENT is missing CF_CAN_GENERATE_ROW_EVENTS;
    . SQLCOM_REVOKE_ALL is missing CF_CHANGES_DATA;
    . SQLCOM_CREATE_FUNCTION is missing CF_AUTO_COMMIT_TRANS;

  This may lead to a wrong sequence of events in the binary log. To fix
  the problem, we correctly set the flags in sql_command_flags[command].

This patch is suggested by Stewart Smith.
Define CACHE_LINE_SIZE to 128 on POWER.

- removing libmysql/libmysql.version from .bzrignore,
  as we don't have this file any more

(Except Instantiation notices from oqgraph that I don't know what to do with)