Before the fix: 
  ha_partition objects had ha_partition::m_part_info==NULL and that caused
  crash
After: 
- The new ha_partition::clone() function makes the clones use parent's
  m_part_info value.
- The parent ha_partition object remains responsible for deallocation of
  m_part_info.

into  rakia.gmz:/home/kgeorge/mysql/autopush/B20420-5.1-opt

 When checking if an IN predicate can be evaluated using a key
 the optimizer makes sure that all the arguments of IN are of
 the same result type. To assure that it check whether 
 Item_func_in::array is filled in. 
 However Item_func_in::array is set if the types are
 the same AND all the arguments are compile time constants.
 Fixed by introducing Item_func_in::arg_types_compatible
 flag to allow correct checking of the desired condition.

into  olga.mysql.com:/home/igor/mysql-5.1-opt

into  olga.mysql.com:/home/igor/mysql-5.1-opt

into  bk-internal.mysql.com:/data0/bk/mysql-5.1-opt

into  bk-internal.mysql.com:/data0/bk/mysql-5.0-opt

into  olga.mysql.com:/home/igor/mysql-5.1-opt

into  zippy.cornsilk.net:/home/cmiller/work/mysql/tmpmerge

  Rename: BitKeeper/deleted/.del-README.txt -> BitKeeper/deleted/.del-README.txt-renamed

into  olga.mysql.com:/home/igor/mysql-5.1-opt

into  pilot.mysql.com:/home/msvensson/mysql/mysql-5.1-new-maint

into  pilot.mysql.com:/home/msvensson/mysql/mysql-5.1-new-maint

into  pilot.mysql.com:/home/msvensson/mysql/mysql-5.0-maint

into  pilot.mysql.com:/home/msvensson/mysql/mysql-5.0-maint

into  olga.mysql.com:/home/igor/dev-opt/mysql-5.0-opt-bug26209

into  bk-internal.mysql.com:/data0/bk/mysql-5.1-opt

into  mysql.com:/home/kent/bk/tmp3/mysql-5.1

into  mysql.com:/home/hf/work/25492/my51-25492

into  mysql.com:/home/hf/work/25492/my50-25492

into  mysql.com:/home/hf/work/20691/my51-20691

into  mysql.com:/home/hf/work/25492/my50-25492

into  mysql.com:/home/hf/work/25492/my51-25492

into  mysql.com:/home/hf/work/25492/my51-25492

into  mysql.com:/home/hf/work/25492/my50-25492

into  clam.ndb.mysql.com:/export/home/space/pekka/ndb/version/my51-merge

into  mysql.com:/home/gluh/MySQL/Merge/5.0-opt

into  mysql.com:/home/gluh/MySQL/Merge/5.0-opt

 Common symbols with and without initialization
 cause the apple linker to exclude then from the
 list of global symbols.

into  mysql.com:/home/tnurnberg/24660/50-24660

into  mysql.com:/home/tnurnberg/24660/41-24660

into  mysql.com:/home/tnurnberg/24660/50-24660

ENUMs weren't allowed to have character 0xff, a perfectly good character in some locales.
This was circumvented by mapping 0xff in ENUMs to ',', thereby prevent actual commas from
being used. Now if 0xff makes an appearance, we find a character not used in the enum and
use that as a separator. If no such character exists, we throw an error.

Any solution would have broken some sort of existing behaviour. This solution should
serve both fractions (those with 0xff and those with ',' in their enums), but
WILL REQUIRE A DUMP/RESTORE CYCLE FROM THOSE WITH 0xff IN THEIR ENUMS. :-/
That is, mysqldump with their current server, and restore when upgrading to one with
this patch.

The crash happens because second filling of the same I_S table happens in
case of subselect with order by. table->sort.io_cache previously allocated
in create_sort_index() is deleted during second filling
(function get_schema_tables_result). There are two places where
I_S table can be filled: JOIN::exec and create_sort_index().
To fix the bug we should check if the table was already filled
in one of these places and skip processing of the table in second.

Some fields (GEOMETRY first of all) can't be handled properly in this
case at all. So we return an error in this case