Commit 10986dba authored by unknown's avatar unknown

Merge ibabaev@bk-internal.mysql.com:/home/bk/mysql-5.0

into rurik.mysql.com:/home/igor/mysql-5.0


sql/opt_range.cc:
  Auto merged
parents 19103bf3 d2f7623d
...@@ -3703,7 +3703,7 @@ get_mm_leaf(PARAM *param, COND *conf_func, Field *field, KEY_PART *key_part, ...@@ -3703,7 +3703,7 @@ get_mm_leaf(PARAM *param, COND *conf_func, Field *field, KEY_PART *key_part,
/* /*
We need to restore the runtime mem_root of the thread in this We need to restore the runtime mem_root of the thread in this
function becuase it evaluates the value of its argument, while function because it evaluates the value of its argument, while
the argument can be any, e.g. a subselect. The subselect the argument can be any, e.g. a subselect. The subselect
items, in turn, assume that all the memory allocated during items, in turn, assume that all the memory allocated during
the evaluation has the same life span as the item itself. the evaluation has the same life span as the item itself.
......
...@@ -1474,13 +1474,6 @@ sp_lex_keeper::reset_lex_and_exec_core(THD *thd, uint *nextp, ...@@ -1474,13 +1474,6 @@ sp_lex_keeper::reset_lex_and_exec_core(THD *thd, uint *nextp,
they want to store some value in local variable, pass return value and they want to store some value in local variable, pass return value and
etc... So their life time should be longer than one instruction. etc... So their life time should be longer than one instruction.
Probably we can call destructors for most of them then we are leaving
routine. But this won't help much as they are allocated in main query
MEM_ROOT anyway. So they all go to global thd->free_list.
May be we can use some other MEM_ROOT for this purprose ???
What else should we do for cleanup ?
cleanup_items() is called in sp_head::execute() cleanup_items() is called in sp_head::execute()
*/ */
return res; return res;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment