Bug#48157: crash in Item_field::used_tables

MySQL handles the join syntax "JOIN ... USING( field1,
... )" and natural joins by building the same parse tree as
a corresponding join with an "ON t1.field1 = t2.field1 ..."
expression would produce. This parse tree was not cleaned up
properly in the following scenario. If a thread tries to
lock some tables and finds that the tables were dropped and
re-created while waiting for the lock, it cleans up column
references in the statement by means a per-statement free
list. But if the statement was part of a stored procedure,
column references on the stored procedure's free list
weren't cleaned up and thus contained pointers to freed
objects.
      
Fixed by adding a call to clean up the current prepared
statement's free list.

This is a backport from MySQL 5.1

sql/item.h

@@ -470,6 +470,13 @@ public:
   my_string name;			/* Name from select */
   /* Original item name (if it was renamed)*/
   my_string orig_name;
+  /**
+     Intrusive list pointer for free list. If not null, points to the next
+     Item on some Query_arena's free list. For instance, stored procedures
+     have their own Query_arena's.
+     
+     @see Query_arena::free_list
+  */
   Item *next;
   uint32 max_length;
   uint name_length;                     /* Length of name */

sql/sql_parse.cc

@@ -1411,8 +1411,10 @@ end:
 }
 
 
-    /* This works because items are allocated with sql_alloc() */
+/**
-
+   This works because items are allocated with sql_alloc().
+   @note The function also handles null pointers (empty list).
+*/
 void cleanup_items(Item *item)
 {
   DBUG_ENTER("cleanup_items");  

sql/sql_update.cc

@@ -908,8 +908,9 @@ reopen_tables:
       items from 'fields' list, so the cleanup above is necessary to.
     */
     cleanup_items(thd->free_list);
-
+    cleanup_items(thd->stmt_arena->free_list);
     close_tables_for_reopen(thd, &table_list);
+
     goto reopen_tables;
   }