Download | Release Notes | Changelog | Overview of 5.3

Release date: 10 Sep 2011

For the highlights of this release, see the release notes.

The revision number links will take you to the revision's page on Launchpad. On Launchpad you can view more details of the revision and view diffs of the code modified in that revision.

  • Revision #3182 Thu 2011-09-08 23:24:47 +0400
    • Bug #833600: Wrong result with view + outer join + uncorrelated subquery (non-semijoin)
      • The bug was caused by outer join being incorrectly converted into inner because of invalid return values of Item_direct_view_ref::not_null_tables().
      • Provided a correct Item_direct_view_ref::not_null_tables() function.
  • Revision #3181 [merge] Thu 2011-09-08 21:38:10 +0400
    • Automerge.
      • Revision #3173.1.9 Thu 2011-09-08 19:48:14 +0400
        • Bug #830993: Crash in end_read_record with derived table
          • Let join buffering code correctly take into account rowids needed by DuplicateElimination when it is calculating minimum record sizes.
          • In JOIN_CACHE::write_record_data, added asserts that prevent us from writing beyond the end of the buffer.
  • Revision #3180 [merge] Thu 2011-09-08 09:21:31 -0700
    • Merge.
      • Revision #3156.1.1 Mon 2011-08-15 23:18:36 -0700
        • Fixed LP bug #824463.
        • When merging a view / derived table the function SELECT_LEX::merge_subquery incorrectly updated the list SELECT_LEX::leaf_tables. Erroneously it appended the leaf_tables list of the merged object L and then removed the reference to the merged object T from the SELECT_LEX::leaf_tables list.
        • A correct implementation should insert the list L into the SELECT_LEX::leaf_tables list in place of the element of the list that refers to T.
        • The bug could lead to wrong results or even crashes for queries with nested outer joins over views / derived tables.
  • Revision #3179 Wed 2011-09-07 20:39:47 +0200
    • Bug #839387 Assertion `(Item_result)i != TIME_RESULT' failed with CASE + datetime
      • remove incorrect DBUG_ASSERT().
      • Fix incorrectly used cmp_item::get_comparator() in Item_func_case and Item_equal
  • Revision #3178 [merge] Tue 2011-09-06 20:59:29 +0400
    • Merge
      • Revision #3173.1.8 Tue 2011-09-06 20:52:36 +0400
        • Fix typo bug
      • Revision #3173.1.7 Tue 2011-09-06 17:06:04 +0400
        • Bug #823930: Wrong result with semijoin materialization and blob fields
          • Make subquery_types_allow_materialization() detect a case where create_tmp_table() would create a blob column which would make it impossible to use materialization
          • Non-semi-join materialization worked because it detected that this case and felt back to use IN->EXISTS. Semi-join Materialization cannot easily fallback, so we have to detect this case early.
      • Revision #3173.1.6 Mon 2011-09-05 20:51:37 +0400
        • Bug #834739: Wrong result with 3-way inner join, LooseScan,multipart keys
          • Don't use join buffering for tables that are within ranges that are covered by LooseScan strategy.
      • Revision #3173.1.5 Mon 2011-09-05 19:28:22 +0400
        • Bug #834758: Wrong result with innner join, LooseScan, two-column IN() predicate
          • get_bound_sj_equalities() would produce incorrect bitmap when non-first equality was bound, which resulted in invalid LooseScan plans.
      • Revision #3173.1.4 Sun 2011-09-04 16:35:37 +0400
        • Bug #836532: Crash in Item_equal_fields_iterator::get_curr_field with semijoin+materialization
          • Item_in_subselect::inject_in_to_exists_cond() should not call ((Item_cond*)join->conds)->argument_list()->concat(join->cond_equal->current_level) as that makes two lists share their tail, and the cond_equal list will end up containing non-Item_equal objects when substitute_for_best_equal_field() walks through join->conds and replaces all Item_equal objects with Item_func_eq objects.
          • So, instead of using List::concat(), manually copy entries from one list to another.
      • Revision #3173.1.3 Sat 2011-09-03 17:05:05 +0400
        • Bug #834514: Assertion `!table || (!table->read_set || bitmap_is_set(...' with aggregates
          • Make find_all_keys() not to rely on table->tmp_set remaining constant during execution quick_index_merge_select->reset() may change it.
      • Revision #3173.1.2 Fri 2011-09-02 23:44:28 +0400
        • Bug #836507: Crash in setup_sj_materialization_part1() with semijoin+materialization
          • setup_sj_materialization() code failed to take into account that it can be that the first [in join order ordering] table inside semi-join-materialization nest is also an inner table wrt an outer join (that is embedded in the semi-join). This can happen when all of the tables that are inside the semi-join but not inside the outer join are constant.
          • Made a trivial to not assume that table's embedding join nest is the semi-join nest: instead, walk up the outer join nests until we reach the semi-join nest.
      • Revision #3173.1.1 Fri 2011-09-02 22:43:35 +0400
        • Bug #836523: Crash in JOIN::get_partial_cost_and_fanout with semijoin+materialization
          • Make JOIN::get_partial_cost_and_fanout() be able to handle join plans with semi-join-materialization nests.
  • Revision #3177 [merge] Tue 2011-09-06 08:38:35 -0700
    • Merge.
      • Revision #3169.1.1 Tue 2011-09-06 07:17:39 -0700
        • Fixed LP bug #838633.
        • For any query JOIN::optimize() should call the method SELECT::save_leaf_tables after the last transformation that utilizes the statement memory rather than the execution memory.
  • Revision #3176 [merge] Mon 2011-09-05 10:14:48 +0300
    • Merge Bug #780386 5.2->5.3 (where other fix was present)
      • Revision #2732.44.28 Mon 2011-09-05 09:29:49 +0300
        • Fix of Bug #780386.
          • ALL subquery should return TRUE if subquery rowa set is empty independently of left part. The problem was that Item_func_(eq,ne,gt,ge,lt,le) do not call execution of second argument if first is NULL no in this case subquery will not be executed and when Item_func_not_all calls any_value() of the subquery or aggregation function which report that there was rows. So for NULL < ALL (SELECT...) result was FALSE instead of TRUE.
          • Fix is just swapping of arguments of Item_func_(eq,ne,gt,ge,lt,le) (with changing the operation if it is needed) so that result will be the same (for examole a < b is equal to b > a). This fix exploit the fact that first argument will be executed in any case.
  • Revision #3175 [merge] Mon 2011-09-05 08:28:08 +0300
    • merge 5.2->5.3
      • Revision #2732.44.27 Mon 2011-09-05 08:15:46 +0300
        • Fix pbxt suite to keep the same opti9misation it was before.
  • Revision #3174 Sat 2011-09-03 11:50:56 +0300
    • Fixed Bug #828514 "Assertion `! is_set()' failed in Diagnostics_area::set_ok_status with derived table + subquery + concurrent DML"
  • Revision #3173 [merge] Fri 2011-09-02 15:36:02 +0300
  • Revision #3172 [merge] Fri 2011-09-02 14:32:48 +0400
    • Merge
      • Revision #3167.1.2 Mon 2011-08-29 21:54:16 +0400
        • Bug #836491: Crash in Item_field::Item_field from add_ref_to_table_cond() with semijoin+materialization
          • Let create_tmp_table set KEY_PART_INFO::fieldnr. It is needed in add_ref_to_table_cond(), and possibly other places.
      • Revision #3167.1.1 Mon 2011-08-29 19:57:41 +0400
        • Bug #834534: Assertion `0' failed in replace_where_subcondition with semijoin subquery in HAVING
          • The problem was that the code that made the check whether the subquery is an AND-part of the WHERE clause didn't work correctly for nested subqueries. In particular, grand-child subquery in HAVING was treated as if it was in the WHERE, which eventually caused an assert when replace_where_subcondition looked for the subquery predicate in the WHERE and couldn't find it there.
          • The fix: Removed implementation of "thd_marker approach". thd->thd_marker was used to determine the location of subquery predicate: setup_conds() would set accordingly it when making the
            {where|on_expr}->fix_fields(...)
            
            call so that AND-parts of the WHERE/ON clauses can determine they are the AND-parts. Item_cond_or::fix_fields(), Item_func::fix_fields(), Item_subselect::fix_fields (this one was missed), and all other items-that-contain-items had to reset thd->thd_marker before calling fix_fields() for their children items, so that the children can see they are not AND-parts of WHERE/ON.
          • The "thd_marker approach" required that a lot of code in different locations maintains correct value of thd->thd_marker, so it was replaced with:
          • The new approach with mark_as_condition_AND_part does not keep context in thd->thd_marker. Instead, setup_conds() now calls
            {where|on_expr}->mark_as_condition_AND_part()
            
          • and implementations of that function make sure that:
            • parts of AND-expressions get the mark_as_condition_AND_part() call
            • Item_in_subselect objects record that they are AND-parts of WHERE/ON
  • Revision #3171 Thu 2011-09-01 23:53:12 +0300
    • Fix for Bug #834492
    • Analysis:
      • In the test query semi-join merges the inner-most subquery into the outer subquery, and the optimization of the merged subquery finds some new index access methods. Later the IN-EXISTS transformation is applied to the unmerged subquery. Since the optimizer is instructed to not consider materialization, it reoptimizes the plan in-place to take into account the new IN-EXISTS conditions. Just before reoptimization JOIN::choose_subquery_plan resets the query plan, which also resets the access methods found during the semi-join merge. Then reoptimization discovers there are no new access methods, but it leaves the query plan in its reset state. Later semi-join crashes because it assumes these access methods are present.
    • Solution:
      • When reoptimizing in-place, reset the query plan only after new access methods were discovered. If no new access methods were discovered, leave the current plan as it was.
  • Revision #3170 Thu 2011-09-01 14:23:03 +0400
    • sec_to_time() in the integer context was losing the sign of the result
  • Revision #3169 Mon 2011-08-29 18:16:18 +0300
    • Updated crash-me for 5.3
  • Revision #3168 Mon 2011-08-29 18:14:14 +0300
    • Added MariaDB executable comment syntax: /*M!# */
  • Revision #3167 Sat 2011-08-27 00:40:29 +0300
    • Fix Bug #827416
    • Analysis:
      • Constant table optimization of the outer query finds that the right side of the equality is a constant that can be used for an eq_ref access to fetch one row from t1, and substitute t1 with a constant. Thus constant optimization triggers evaluation of the subquery during the optimize phase of the outer query.
      • The innermost subquery requires a plan with a temporary table because with InnoDB tables the exact count of rows is not known, and the empty tables cannot be optimzied way. JOIN::exec for the innermost subquery substitutes the subquery tables with a temporary table.
      • When EXPLAIN gets to print the tables in the innermost subquery, EXPLAIN needs to print the name of each table through the corresponding TABLE_LIST object. However, the temporary table created during execution doesn't have a corresponding TABLE_LIST, so we get a null pointer exception.
    • Solution:
      • The solution is to forbid using expensive constant expressions for eq_ref access for contant table optimization. Notice that eq_ref with a subquery providing the value is still possible during regular execution.
  • Revision #3166 [merge] Tue 2011-08-23 15:51:47 +0300
    • Automatic merge.
      • Revision #3164.1.2 Tue 2011-08-23 15:39:15 +0300
        • Fixed Bug #825018
          • Analysis:
            • During the first execution of the query through the stored procedure, the optimization phase calls substitute_for_best_equal_field(), which calls Item_in_optimizer::transform(). The latter replaces Item_in_subselect::left_expr with args[0] via assignment. In this test case args[0] is an Item_outer_ref which is created/deallocated for each re-execution. As a result, during the second execution Item_in_subselect::left_expr pointed to freed memory, which resulted in a crash.
          • Solution:
            • The solution is to use change_item_tree(), so that the origianal left expression is restored after each execution.
      • Revision #3164.1.1 Tue 2011-08-23 00:00:13 +0300
        • Fix Bug #825095
          • Analysis:
            • Partial matching is used even when there are no NULLs in a materialized subquery, as long as the left NOT IN operand may contain NULL values.
            • This case was not handled correctly in two different places. First, the implementation of parital matching did not clear the set of matching columns when the merge process advanced to the next row.
            • Second, there is no need to perform partial matching at all when the left operand has no NULLs.
          • Solution:
            • First fix subselect_rowid_merge_engine::partial_match() to properly cleanup the bitmap of matching keys when advancing to the next row.
            • Second, change subselect_partial_match_engine::exec() so that when the materialized subquery doesn't contain any NULLs, and the left operand of [NOT] IN doesn't contain NULLs either, the method returns without doing any unnecessary partial matching. The correct result in this case is in Item::in_value.
  • Revision #3165 Mon 2011-08-22 13:38:32 +0200
  • Revision #3164 Fri 2011-08-19 21:02:05 -0700
    • Fixed Bug #826279.
      • When the WHERE/HAVING condition of a subquery has been transformed by the optimizer the pointer stored the 'where'/'having' field of the SELECT_LEX structure used for the subquery must be updated accordingly. Otherwise the pointer may refer to an invalid item. This can lead to the reported assertion failure for some queries with correlated subqueries
  • Revision #3163 Wed 2011-08-17 14:10:32 +0300
    • Fix Bug #813473
      • The bug is a duplicate of MySQL's bug#11764086, however MySQL's fix is incomplete for MariaDB, so this fix is slightly different.
      • In addition, this patch renames Item_func_not_all::top_level() to is_top_level_item() to make it in line with the analogous methods of Item_in_optimizer, and Item_subselect.
      • Analysis:
        • It is possible to determine whether a predicate is NULL-rejecting only if it is a top-level one. However, this was not taken into account for Item_in_optimizer. As a result, a NOT IN predicate was erroneously considered as NULL-rejecting, and the NULL-complemented rows generated by the outer join were rejected before being checked by the NOT IN predicate.
      • Solution:
        • Change Item_in_optimizer to be considered as NULL-rejecting only if it a top-level predicate.
  • Revision #3162 Wed 2011-08-17 12:32:15 +0400
    • field_conv.cc: added comments opt_range.cc: modified print_key() so that it doesn't do memory re-allocs when printing multipart keys over varchar columns. When it did, key printout in debug trace was interrupted with my_malloc/free printouts.
  • Revision #3161 [merge] Wed 2011-08-17 12:02:02 +0400
    • Merge
      • Revision #3155.1.2 Wed 2011-08-17 11:57:01 +0400
        • Bug #826935 Assertion `!table || (!table->read_set || bitmap_is_set(table->read_set, field_index))' failed
          • add_ref_to_table_cond() should not just overwrite pre_idx_push_select_cond with the contents tab->select_cond.
          • pre_idx_push_select_cond exists precisely for the reason that it may contain a condition that is a strict superset of what is in tab->select_cond.
          • The fix is to inject generated equality into pre_idx_push_select_cond.
  • Revision #3160 Tue 2011-08-16 22:48:35 -0700
    • Fixed Bug #825035.
      • The value of maybe_null flag should be saved for the second execution of a prepared statement from SELECT that uses an outer join.
  • Revision #3159 [merge] Tue 2011-08-16 23:45:08 +0300
    • Merge fix for small_blocksize.test
  • Revision #3158 [merge] Tue 2011-08-16 21:54:48 +0400
    • Merge
      • Revision #3155.1.1 Tue 2011-08-16 21:42:25 +0400
        • Bug #818280: crash in do_copy_not_null() in maria-5.3 with semijoin
          • Make simplify_joins() set maybe_null=FALSE for tables that were on the inner sides of inner joins and then were moved to the inner sides of semi-joins.
  • Revision #3157 [merge] Tue 2011-08-16 15:51:40 +0300
    • Automatic merge with 5.2
      • Revision #2732.44.20 Tue 2011-08-16 13:28:20 +0300
        • Fixed build failure in embedded library regarding that decrease_user_connections() was not declared
      • Revision #2732.44.19 Tue 2011-08-16 13:06:07 +0300
        • If mysqld --log-warnings=3 or higher, then print all check and repair warnings for MyISAM tables to the log. This is useful when trying to find out why an automatic myisam repair failes.
      • Revision #2732.44.18 Tue 2011-08-16 12:32:06 +0300
        • Fixed bug that MAX_USER_CONNECTIONS was not working properly in all situations (which could cause aborted connects) thd->user_connect is now handled in thd->clenup() which will ensure that it works in all context (including slaves). I added also some DBUG_ASSERT() to ensure that things are working correctly.
      • Revision #2732.44.17 Mon 2011-08-15 23:53:55 +0300
        • Fixed recovery crash Bug #814806 "Unclean shutdown corrupted Aria table blocking startup"
  • Revision #3156 [merge] Mon 2011-08-15 22:14:08 +0300
    • Automatic merge with 5.2
      • Revision #2732.44.16 [merge] Mon 2011-08-15 20:42:29 +0300
        • Merge in bug fix from 5.1
      • Revision #2732.44.15 Mon 2011-08-15 20:40:13 +0300
        • Increase server version
      • Revision #2732.44.14 Mon 2011-08-15 16:39:53 +0300
        • Fixes bugs found by testcase for Bug #815022 and Bug #726374 "ma_blockrec.c:3000: write_block_record: Assertion `cur_block[1].page_count == 0' failed with a multi-index Aria workload"
          • The issues were:
            • For some tables with a lot of not packed fields, we didn't allocate enough memory in head page which caused DBUG_ASSERT's
            • Removed wrong DBUG_ASSERT()
            • Fixed a problem with underflow() where it generates a key page where all keys didn't fit.
            • Max key length is now limited by block_size/3 (was block_size /2). This is required for underflow() to work with packed keys.
      • Revision #2732.44.13 [merge] Fri 2011-08-12 15:51:05 +0300
        • Autmatic merge with 5.1
      • Revision #2732.44.12 Wed 2011-08-10 22:44:39 +0300
        • Fixed Bug #814054 'Assertion `block->hash_link == hash_link && hash_link->block == block' in ma_pagecache.c:2275 with Aria'
          • Replaced old DBUG_ASSERT with a new correct one + a comment.
      • Revision #2732.44.11 Wed 2011-08-10 13:08:19 +0300
        • Fixes MySQL Bug #48972: mysqldump --insert-ignore leaves set unique_checks=0. This fixes a bug that when you use mysqldump --no-create-info to generate a dump that you want to merge with an existing table, you can get an innodb table with duplicated unique keys. Patch originally by Eric Bergen.
      • Revision #2732.44.10 Mon 2011-08-08 14:53:52 +0300
        • Optimize mutex usage.
  • Revision #3155 Fri 2011-08-12 14:31:40 +0300
    • Fixed test results after the tests adding.
  • Revision #3154 Fri 2011-08-12 13:54:41 +0300
    • Early check of subquery cache hit rate added to limit its performance impact in the worst case.
  • Revision #3153 Fri 2011-08-12 11:39:29 +0300
    • Bug #781508: Take relevant test cases from MySQL 5.6 feature preview trees
      • Identified all test cases in the MySQL file subquery.inc that are not present in MariaDB. This patch adds the test cases that are:
        • not present in MySQL 5.5, and
        • already fixed in MariaDB 5.3
      • The patch adds test cases for the following mysql-trunk bugs:
        • bug 12763207 - not a bug, mysql-trunk, added test case
        • MySQL Bug #50257 - not a bug, mysql-trunk, added test case
        • bug 11765699 - not a bug, mysql-trunk, added test case
        • bug 12616253 - not a bug, mysql-trunk, added test case
      • The comparison was based on the following version of mysql-trunk:
        revno: 3350 [merge]
            committer: Marko Mäkelä <marko.makela@oracle.com>
            branch nick: mysql-trunk
            timestamp: Mon 2011-08-08 12:42:09 +0300
            message:
              Merge mysql-5.5 to mysql-trunk.
        
  • Revision #3152 Fri 2011-08-12 11:23:50 +0300
    • Protect statistic variables of subquery cache.
  • Revision #3151 Thu 2011-08-11 22:34:41 -0700
  • Revision #3150 Thu 2011-08-11 20:24:32 -0700
    • Fixed Bug #823189.
      • The method Item_ref::not_null_tables() returned incorrect bitmap for outer references to view columns. This could cause an invalid conversion of an outer join into an inner join that could lead to a wrong result set for a query with a correlated subquery over an outer join whose where condition had an outer reference to a view.
  • Revision #3149 Thu 2011-08-11 12:11:04 -0700
    • Fixed Bug #823826.
      • The method Item_func_isnull::update_used_tables() erroneously did not update cached values stored in the fields used_tables_cache and const_item_cache of the Item_func_isnull objects. As a result the Item_func_isnull::used_tables() returned wrong bitmaps and, as a consequence, push-down predicates could be attached to wrong tables.
  • Revision #3148 Tue 2011-08-09 18:34:26 +0300
    • Bug #781508: Take relevant test cases from MySQL 5.6 feature preview trees
      • Identified all test cases in the MySQL file subquery_mat.inc that are not present in MariaDB. In total found 8 test cases for the following MySQL bugs:
      • All of these MySQL bugs are not present in MariaDB 5.3.
      • The comparison was based on the following version of mysql-trunk:
        revno: 3350 [merge]
          committer: Marko Mäkelä <marko.makela@oracle.com>
          branch nick: mysql-trunk
          timestamp: Mon 2011-08-08 12:42:09 +0300
          message:
            Merge mysql-5.5 to mysql-trunk.
        
  • Revision #3147 Tue 2011-08-09 10:28:57 +0300
    • Fix Bug #817384
      • This bug is a special case of Bug #813447.
      • Analysis:
        • Constant optimization finds that the condition t2.a = 1 can be used to access the primary key of table 't2'. As a result both outer table t1,t2 are considered as constant when we reach the execution phase. At the same time, during constant optimization, the IN predicate is not evaluated because it is expensive.
        • When execution of the outer query reaches do_select(), control flow enter the branch:
          if (join->table_count == join->const_tables)
               { ... }
          
        • This branch checks only the WHERE and HAVING clauses, but doesn't check the ON clauses of the query. Since the IN predicate was not evaluated during optimization, it is not evaluated at all, thus execution doesn't detect that the ON clause is FALSE.
      • Solution:
        • Similar to the patch for Bug #813447, exclude system tables from constant substitution based on unique key lookups if there is an expensive ON condition on the inner table.
  • Revision #3146 Mon 2011-08-08 22:02:10 -0700
    • Fixed Bug #819716. Do not optimize derived table for the second time ever.
  • Revision #3145 [merge] Tue 2011-08-09 01:57:08 +0400
    • Merge fix for Bug #822134
      • Revision #3138.1.3 Tue 2011-08-09 01:37:06 +0400
        • Bug #822134: Invalid plan and wrong result set for Q20 from DBT3 benchmark set
          • create_ref_for_key() has the code that walks KEYUSE array and tries to use maximum number of keyparts for ref (and eq_ref and ref_or_null) access. When one constructs ref access for table that is inside a SJ-Materialization nest, it is not possible to use tables that are ouside the nest (because materialization is performed before they have any "current value"). The bug was caused by this function not taking this into account.
  • Revision #3144 [merge] Mon 2011-08-08 23:12:34 +0400
    • Merge
  • Revision #3143 Mon 2011-08-08 18:40:41 +0200
    • Fix long xtradb shutdown on Windows XP
      • The reason for the long shutdown is hanging in io threads. It appears that just closing completion port on XP does not necessarily signal thread waiting in GetIOCompletionStatus() (even if this works fine on later Windows versions)
      • The fix is to wakeup background threads using PostQueuedCompletionStatus() with a special 'key' parameter indicating shutdown.
  • Revision #3142 Mon 2011-08-08 18:01:33 +0200
    • Bug #882689 - crash during startup on XP.
      • The reason for the crash is Innodb assertion after trying to load condition variables function dynamically and not finding them
      • The fix is to skip dynamic loading if srv_use_native_conditions is FALSE. srv_use_native_conditions is derived from Windows version and would be FALSE on XP and TRUE on later Windows.
      • This is the same handling as in MySQL 5.. In Maria 5.3 srv_use_native_conditions check was presumably lost in the downporting.
  • Revision #3141 [merge] Fri 2011-08-05 22:07:06 +0400
    • Merge
      • Revision #3138.1.2 Fri 2011-08-05 22:01:49 +0400
        • Backport of:
          revno: 2876.47.174
          revision-id: jorgen.loland@oracle.com-20110519120355-qn7eprkad9jqwu5j
          parent: mayank.prasad@oracle.com-20110518143645-bdxv4udzrmqsjmhq
          committer: Jorgen Loland <jorgen.loland@oracle.com>
          branch nick: mysql-trunk-11765831
          timestamp: Thu 2011-05-19 14:03:55 +0200
          message:
            BUG#11765831: 'RANGE ACCESS' MAY INCORRECTLY FILTER 
                                  AWAY QUALIFYING ROWS
          .
            The problem was that the ranges created when OR'ing two 
            conditions could be incorrect. Without the bugfix, 
            "I <> 6 OR (I <> 8 AND J = 5)" would create these ranges:
          .
            "NULL < I < 6",
            "6 <= I <= 6 AND 5 <= J <= 5",
            "6 < I < 8",
            "8 <= I <= 8 AND 5 <= J <= 5",
            "8 < I"
          .
            While the correct ranges is
            "NULL < I < 6",
            "6 <= I <= 6 AND 5 <= J <= 5",
            "6 < I"
          .
            The problem occurs when key_or() ORs
            (1) "NULL < I < 6, 6 <= I <= 6 AND 5 <= J <= 5, 6 < I" with 
            (2) "8 < I AND 5 <= J <= 5"
          .
            The reason for the bug is that in key_or(), SEL_ARG *tmp is 
            used to point to the range in (1) above that is merged with 
            (2) while key1 points to the root of the red-black tree of 
            (1). When merging (1) and (2), tmp refers to the "6 < I" 
            part whereas the root is the "6 <= ... AND 5 <= J <= 5" part. 
          .
            key_or() decides that the tmp range needs to be split into
            "6 < I < 8, 8 <= I <= 8, 8 < I", in which next_key_part of the 
            second range should be that of tmp. However, next_key_part is
            set to key1->next_key_part ("5 <= J <= 5") instead of 
            tmp->next_key_part (empty). Fixing this gives the correct but
            not optimal ranges:
            "NULL < I < 6",
            "6 <= I <= 6 AND 5 <= J <= 5",
            "6 < I < 8",
            "8 <= I <= 8",
            "8 < I"
          .
            A second problem can be seen above: key_or() may create 
            adjacent ranges that could be replaced with a single range. 
            Fixes for this is also included in the patch so that the range
            above becomes correct AND optimal:
            "NULL < I < 6",
            "6 <= I <= 6 AND 5 <= J <= 5",
            "6 < I"
          .
            Merging adjacent ranges like this gives a slightly lower cost 
            estimate for the range access.
          
      • Revision #3138.1.1 Thu 2011-08-04 18:20:02 +0400
        • Backport of:
          revno: 3363.3.16
          revision-id: jorgen.loland@oracle.com-20110506132631-5wickj6dvrh1dpj6
          parent: alexander.nozdrin@oracle.com-20110506132138-46459va9vcbd4nz0
          committer: Jorgen Loland <jorgen.loland@oracle.com>
          branch nick: mysql-trunk-11765831
          timestamp: Fri 2011-05-06 15:26:31 +0200
          message:
            BUG#11765831: 'RANGE ACCESS' MAY INCORRECTLY FILTER
                          AWAY QUALIFYING ROWS
          .
            Preparation patch (does not include fix for the bug):
          .
              - Extensively document key_or()
              - Remove tab indentations from key_or()
              - Minor code changes like using existing utility functions
                in key_or()
          
  • Revision #3140 Wed 2011-08-03 13:58:46 +0200
    • Limit query length in error log to 64K, to avoid output of full blobs
  • Revision #3139 Wed 2011-08-03 13:42:53 +0200
    • Enhance crash reporting. Fix
      • Bug #819711 : optimizer_switch must be reported on segfault
      • Bug #820169: Full query text must be reported on crash
  • Revision #3138 [merge] Mon 2011-08-01 11:05:30 +0200
    • Automerge 5.2->5.3
      • Revision #2732.44.9 Mon 2011-08-01 10:56:24 +0200
        • After-merge fix of result file (MARIA <-> Aria)
  • Revision #3137 [merge] Sun 2011-07-31 22:59:55 +0200
    • Automerge 5.2->5.3
  • Revision #3136 Fri 2011-07-29 17:09:16 -0700
    • Fixed Bug #817360.
      • This problem could be observed for queries with nested outer joins for which the not_exist optimization were applicable. The problem was caused by the code of the patch for MySQL Bug #49322 that erroneously forced the return to the previous nested loop level when the join algorithm successfully builds a partial record for an embedded outer to which the not_exist optimization could be applied.
      • Actually the immediate return to the previous nested loops level is correct only if this partial record is rejected by a predicate pushed down to one of the inner tables of this outer join. Otherwise attempts to find extensions of this record must be made.
  • Revision #3135 Thu 2011-07-28 17:10:29 +0300
    • Subquery cache going on disk management fix: Do not go on disk if hit rate is not good.

Comments

Comments loading...