- developers - lists.mariadb.org

[Maria-developers] Updated (by Guest): Table elimination (17)
by worklog-noreply＠askmonty.org 22 May '09

22 May '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Table elimination CREATION DATE..: Sun, 10 May 2009, 19:57 SUPERVISOR.....: Monty IMPLEMENTOR....: Psergey COPIES TO......: CATEGORY.......: Server-Sprint TASK ID........: 17 (http://askmonty.org/worklog/?tid=17) VERSION........: Server-5.1 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 0 (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30176 2009-05-22 17:00:35.000000000 +0300 +++ /tmp/wklog.17.new.30176 2009-05-22 17:00:35.000000000 +0300 @@ -1 +1 @@ -Maria-2.0 +Server-5.1 -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Category updated. --- /tmp/wklog.17.old.30162 2009-05-22 17:00:28.000000000 +0300 +++ /tmp/wklog.17.new.30162 2009-05-22 17:00:28.000000000 +0300 @@ -1 +1 @@ -Maria-Sprint +Server-Sprint -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30162 2009-05-22 17:00:28.000000000 +0300 +++ /tmp/wklog.17.new.30162 2009-05-22 17:00:28.000000000 +0300 @@ -1 +1 @@ -2.0 +9.x -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30162 2009-05-22 17:00:28.000000000 +0300 +++ /tmp/wklog.17.new.30162 2009-05-22 17:00:28.000000000 +0300 @@ -1 +1 @@ -9.x +Maria-2.0 -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Category updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Server-RawIdeaBin +Maria-Sprint -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Server-5.2 +2.0 -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Status updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Un-Assigned +Assigned -=-=(Guest - Thu, 21 May 2009, 22:54)=-=- Low Level Design modified. --- /tmp/wklog.17.old.28468 2009-05-21 22:54:28.000000000 +0300 +++ /tmp/wklog.17.new.28468 2009-05-21 22:54:28.000000000 +0300 @@ -77,7 +77,7 @@ - Relationship with prepared statements. On one hand, it's natural to desire to make table elimination a once-per-statement operation, like outer->inner join conversion. We'll have - to limit the applicablity by removing [MARK1] as that can change during + to limit the applicability by removing [MARK1] as that can change during lifetime of the statement. The other option is to do table elimination every time. This will require to -=-=(Psergey - Tue, 19 May 2009, 20:20)=-=- Low Level Design modified. --- /tmp/wklog.17.old.27126 2009-05-19 20:20:06.000000000 +0300 +++ /tmp/wklog.17.new.27126 2009-05-19 20:20:06.000000000 +0300 @@ -48,6 +48,10 @@ another outer join nest, that shouldn't prevent us from moving away the eliminated tables. +* Update join->table_count and all-join-tables bitmap. + +* That's it. Nothing else? + 4. User interface ----------------- * We'll add an @@optimizer switch flag for table elimination. Tentative -=-=(Psergey - Tue, 19 May 2009, 20:18)=-=- Low Level Design modified. --- /tmp/wklog.17.old.27035 2009-05-19 20:18:27.000000000 +0300 +++ /tmp/wklog.17.new.27035 2009-05-19 20:18:27.000000000 +0300 @@ -1 +1,92 @@ +<contents> +1. Conditions for removal +2. Removal operation properties +3. Removal operation +4. User interface +5. Todo, issues to resolve +5.1 To resolve +5.2 Resolved +</contents> + +It's not really about elimination of tables, it's about elimination of inner +sides of outer joins. + +1. Conditions for removal +------------------------- +We can eliminate an inner side of outer join if: +1. For each record combination of outer tables, it will always produce + exactly one record. +2. There are no references to columns of the inner tables anywhere else in + the query. + +#1 means that every table inside the outer join nest is: + - is a constant table: + = because it can be accessed via eq_ref(const) access, or + = it is a zero-rows or one-row MyISAM-like table [MARK1] + - has an eq_ref access method candidate. + +#2 means that WHERE clause, ON clauses of embedding outer joins, ORDER BY, + GROUP BY and HAVING do not refer to the inner tables of the outer join + nest. + +2. Removal operation properties +------------------------------- +* There is always one way to remove (no choice to remove either this or that) +* It is always better to remove as much tables as possible (at least within + our cost model). +Thus, no need for any cost calculations/etc. It's an unconditional rewrite. + +3. Removal operation +-------------------- +* Remove the outer join nest's nested join structure (i.e. get the + outer join's TABLE_LIST object $OJ and remove it from $OJ->embedding, + $OJ->embedding->nested_join. Update table_map's of all ancestor nested + joins). [MARK2] + +* Move the tables and their JOIN_TABs to front like it is done with const + tables, with exception that if eliminated outer join nest was within + another outer join nest, that shouldn't prevent us from moving away the + eliminated tables. + +4. User interface +----------------- +* We'll add an @@optimizer switch flag for table elimination. Tentative + name: 'table_elimination'. + +* With EXPLAIN, there are two options: + - Show removed tables in a way similar to const tables, with some + indication that they are removed. + - Do not show them altogether. +(the second one seems to be better? We're targeting a situation with VIEWs, +where the user would not care about what tables were added into his query +and then discarded from it?) + +5. Todo, issues to resolve +-------------------------- + +5.1 To resolve +~~~~~~~~~~~~~~ +- See EXPLAIN question in section #4. + +- Re-check how this works with equality propagation. + +- Relationship with prepared statements. + On one hand, it's natural to desire to make table elimination a + once-per-statement operation, like outer->inner join conversion. We'll have + to limit the applicablity by removing [MARK1] as that can change during + lifetime of the statement. + + The other option is to do table elimination every time. This will require to + rework operation [MARK2] to be undoable. + + I'm leaning towards doing the former. With anchor modeling, it is unlikely + that we'll meet outer joins which have N inner tables of which some are 1-row + MyISAM tables that do not have primary key. + +5.2 Resolved +~~~~~~~~~~~~ +- outer->inner join conversion is not a problem for table elimination. + We make outer->inner conversions based on predicates in WHERE. If the WHERE + referred to an inner table (requirement for OJ->IJ conversion) then table + elimination would not be applicable anyway. ------------------------------------------------------------ -=-=(View All Progress Notes, 12 total)=-=- http://askmonty.org/worklog/index.pl?tid=17&nolimit=1 DESCRIPTION: Eliminate not needed tables from SELECT queries.. This will speed up some views and automatically generated queries. Example: CREATE TABLE B (id int primary key); select A.colA from tableA A left outer join tableB B on B.id = A.id; In this case we can remove table B and the join from the query. HIGH-LEVEL SPECIFICATION: Here is an extended explanation of table elimination. Table elimination is a feature found in some modern query optimizers, of which Microsoft SQL Server 2005/2008 seems to have the most advanced implementation. Oracle 11g has also been confirmed to use table elimination but not to the same extent. Basically, what table elimination does, is to remove tables from the execution plan when it is unneccessary to include them. This can, of course, only happen if the right circumstances arise. Let us for example look at the following query: select A.colA from tableA A left outer join tableB B on B.id = A.id; When using A as the left table we ensure that the query will return at least as many rows as there are in that table. For rows where the join condition (B.id = A.id) is not met the selected column (A.colA) will contain a NULL value. However, the result set could actually contain more rows than what is found in tableA if there are duplicates of the column B.id in tableB. If A contains a row [1, "val1"] and B the rows [1, "other1a"],[1, "other1b"] then two rows will match in the join condition. The only way to know what the result will look like is to actually touch both tables during execution. Instead, let's say that tableB contains rows that make it possible to place a unique constraint on the column B.id, for example and often the case a primary key. In this situation we know that we will get exactly as many rows as there are in tableA, since joining with tableB cannot introduce any duplicates. If further, as in the example query, we do not select any columns from tableB, touching that table during execution is unneccessary. We can remove the whole join operation from the execution plan. Both SQL Server 2005/2008 and Oracle 11g will deploy table elimination in the case described above. Let us look at a more advanced query, where Oracle fails. select A.colA from tableA A left outer join tableB B on B.id = A.id and B.fromDate = ( select max(sub.fromDate) from tableB sub where sub.id = A.id ); In this example we have added another join condition, which ensures that we only pick the matching row from tableB having the latest fromDate. In this case tableB will contain duplicates of the column B.id, so in order to ensure uniqueness the primary key has to contain the fromDate column as well. In other words the primary key of tableB is (B.id, B.fromDate). Furthermore, since the subselect ensures that we only pick the latest B.fromDate for a given B.id we know that at most one row will match the join condition. We will again have the situation where joining with tableB cannot affect the number of rows in the result set. Since we do not select any columns from tableB, the whole join operation can be eliminated from the execution plan. SQL Server 2005/2008 will deploy table elimination in this situation as well. We have not found a way to make Oracle 11g use it for this type of query. Queries like these arise in two situations. Either when you have denormalized model consisting of a fact table with several related dimension tables, or when you have a highly normalized model where each attribute is stored in its own table. The example with the subselect is common whenever you store historized/versioned data. LOW-LEVEL DESIGN: <contents> 1. Conditions for removal 2. Removal operation properties 3. Removal operation 4. User interface 5. Todo, issues to resolve 5.1 To resolve 5.2 Resolved </contents> It's not really about elimination of tables, it's about elimination of inner sides of outer joins. 1. Conditions for removal ------------------------- We can eliminate an inner side of outer join if: 1. For each record combination of outer tables, it will always produce exactly one record. 2. There are no references to columns of the inner tables anywhere else in the query. #1 means that every table inside the outer join nest is: - is a constant table: = because it can be accessed via eq_ref(const) access, or = it is a zero-rows or one-row MyISAM-like table [MARK1] - has an eq_ref access method candidate. #2 means that WHERE clause, ON clauses of embedding outer joins, ORDER BY, GROUP BY and HAVING do not refer to the inner tables of the outer join nest. 2. Removal operation properties ------------------------------- * There is always one way to remove (no choice to remove either this or that) * It is always better to remove as much tables as possible (at least within our cost model). Thus, no need for any cost calculations/etc. It's an unconditional rewrite. 3. Removal operation -------------------- * Remove the outer join nest's nested join structure (i.e. get the outer join's TABLE_LIST object $OJ and remove it from $OJ->embedding, $OJ->embedding->nested_join. Update table_map's of all ancestor nested joins). [MARK2] * Move the tables and their JOIN_TABs to front like it is done with const tables, with exception that if eliminated outer join nest was within another outer join nest, that shouldn't prevent us from moving away the eliminated tables. * Update join->table_count and all-join-tables bitmap. * That's it. Nothing else? 4. User interface ----------------- * We'll add an @@optimizer switch flag for table elimination. Tentative name: 'table_elimination'. * With EXPLAIN, there are two options: - Show removed tables in a way similar to const tables, with some indication that they are removed. - Do not show them altogether. (the second one seems to be better? We're targeting a situation with VIEWs, where the user would not care about what tables were added into his query and then discarded from it?) 5. Todo, issues to resolve -------------------------- 5.1 To resolve ~~~~~~~~~~~~~~ - See EXPLAIN question in section #4. - Re-check how this works with equality propagation. - Relationship with prepared statements. On one hand, it's natural to desire to make table elimination a once-per-statement operation, like outer->inner join conversion. We'll have to limit the applicability by removing [MARK1] as that can change during lifetime of the statement. The other option is to do table elimination every time. This will require to rework operation [MARK2] to be undoable. I'm leaning towards doing the former. With anchor modeling, it is unlikely that we'll meet outer joins which have N inner tables of which some are 1-row MyISAM tables that do not have primary key. 5.2 Resolved ~~~~~~~~~~~~ - outer->inner join conversion is not a problem for table elimination. We make outer->inner conversions based on predicates in WHERE. If the WHERE referred to an inner table (requirement for OJ->IJ conversion) then table elimination would not be applicable anyway. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] Updated (by Guest): Table elimination (17)
by worklog-noreply＠askmonty.org 22 May '09

22 May '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Table elimination CREATION DATE..: Sun, 10 May 2009, 19:57 SUPERVISOR.....: Monty IMPLEMENTOR....: Psergey COPIES TO......: CATEGORY.......: Server-Sprint TASK ID........: 17 (http://askmonty.org/worklog/?tid=17) VERSION........: Server-5.1 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 0 (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30176 2009-05-22 17:00:35.000000000 +0300 +++ /tmp/wklog.17.new.30176 2009-05-22 17:00:35.000000000 +0300 @@ -1 +1 @@ -Maria-2.0 +Server-5.1 -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Category updated. --- /tmp/wklog.17.old.30162 2009-05-22 17:00:28.000000000 +0300 +++ /tmp/wklog.17.new.30162 2009-05-22 17:00:28.000000000 +0300 @@ -1 +1 @@ -Maria-Sprint +Server-Sprint -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30162 2009-05-22 17:00:28.000000000 +0300 +++ /tmp/wklog.17.new.30162 2009-05-22 17:00:28.000000000 +0300 @@ -1 +1 @@ -2.0 +9.x -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30162 2009-05-22 17:00:28.000000000 +0300 +++ /tmp/wklog.17.new.30162 2009-05-22 17:00:28.000000000 +0300 @@ -1 +1 @@ -9.x +Maria-2.0 -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Category updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Server-RawIdeaBin +Maria-Sprint -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Server-5.2 +2.0 -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Status updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Un-Assigned +Assigned -=-=(Guest - Thu, 21 May 2009, 22:54)=-=- Low Level Design modified. --- /tmp/wklog.17.old.28468 2009-05-21 22:54:28.000000000 +0300 +++ /tmp/wklog.17.new.28468 2009-05-21 22:54:28.000000000 +0300 @@ -77,7 +77,7 @@ - Relationship with prepared statements. On one hand, it's natural to desire to make table elimination a once-per-statement operation, like outer->inner join conversion. We'll have - to limit the applicablity by removing [MARK1] as that can change during + to limit the applicability by removing [MARK1] as that can change during lifetime of the statement. The other option is to do table elimination every time. This will require to -=-=(Psergey - Tue, 19 May 2009, 20:20)=-=- Low Level Design modified. --- /tmp/wklog.17.old.27126 2009-05-19 20:20:06.000000000 +0300 +++ /tmp/wklog.17.new.27126 2009-05-19 20:20:06.000000000 +0300 @@ -48,6 +48,10 @@ another outer join nest, that shouldn't prevent us from moving away the eliminated tables. +* Update join->table_count and all-join-tables bitmap. + +* That's it. Nothing else? + 4. User interface ----------------- * We'll add an @@optimizer switch flag for table elimination. Tentative -=-=(Psergey - Tue, 19 May 2009, 20:18)=-=- Low Level Design modified. --- /tmp/wklog.17.old.27035 2009-05-19 20:18:27.000000000 +0300 +++ /tmp/wklog.17.new.27035 2009-05-19 20:18:27.000000000 +0300 @@ -1 +1,92 @@ +<contents> +1. Conditions for removal +2. Removal operation properties +3. Removal operation +4. User interface +5. Todo, issues to resolve +5.1 To resolve +5.2 Resolved +</contents> + +It's not really about elimination of tables, it's about elimination of inner +sides of outer joins. + +1. Conditions for removal +------------------------- +We can eliminate an inner side of outer join if: +1. For each record combination of outer tables, it will always produce + exactly one record. +2. There are no references to columns of the inner tables anywhere else in + the query. + +#1 means that every table inside the outer join nest is: + - is a constant table: + = because it can be accessed via eq_ref(const) access, or + = it is a zero-rows or one-row MyISAM-like table [MARK1] + - has an eq_ref access method candidate. + +#2 means that WHERE clause, ON clauses of embedding outer joins, ORDER BY, + GROUP BY and HAVING do not refer to the inner tables of the outer join + nest. + +2. Removal operation properties +------------------------------- +* There is always one way to remove (no choice to remove either this or that) +* It is always better to remove as much tables as possible (at least within + our cost model). +Thus, no need for any cost calculations/etc. It's an unconditional rewrite. + +3. Removal operation +-------------------- +* Remove the outer join nest's nested join structure (i.e. get the + outer join's TABLE_LIST object $OJ and remove it from $OJ->embedding, + $OJ->embedding->nested_join. Update table_map's of all ancestor nested + joins). [MARK2] + +* Move the tables and their JOIN_TABs to front like it is done with const + tables, with exception that if eliminated outer join nest was within + another outer join nest, that shouldn't prevent us from moving away the + eliminated tables. + +4. User interface +----------------- +* We'll add an @@optimizer switch flag for table elimination. Tentative + name: 'table_elimination'. + +* With EXPLAIN, there are two options: + - Show removed tables in a way similar to const tables, with some + indication that they are removed. + - Do not show them altogether. +(the second one seems to be better? We're targeting a situation with VIEWs, +where the user would not care about what tables were added into his query +and then discarded from it?) + +5. Todo, issues to resolve +-------------------------- + +5.1 To resolve +~~~~~~~~~~~~~~ +- See EXPLAIN question in section #4. + +- Re-check how this works with equality propagation. + +- Relationship with prepared statements. + On one hand, it's natural to desire to make table elimination a + once-per-statement operation, like outer->inner join conversion. We'll have + to limit the applicablity by removing [MARK1] as that can change during + lifetime of the statement. + + The other option is to do table elimination every time. This will require to + rework operation [MARK2] to be undoable. + + I'm leaning towards doing the former. With anchor modeling, it is unlikely + that we'll meet outer joins which have N inner tables of which some are 1-row + MyISAM tables that do not have primary key. + +5.2 Resolved +~~~~~~~~~~~~ +- outer->inner join conversion is not a problem for table elimination. + We make outer->inner conversions based on predicates in WHERE. If the WHERE + referred to an inner table (requirement for OJ->IJ conversion) then table + elimination would not be applicable anyway. ------------------------------------------------------------ -=-=(View All Progress Notes, 12 total)=-=- http://askmonty.org/worklog/index.pl?tid=17&nolimit=1 DESCRIPTION: Eliminate not needed tables from SELECT queries.. This will speed up some views and automatically generated queries. Example: CREATE TABLE B (id int primary key); select A.colA from tableA A left outer join tableB B on B.id = A.id; In this case we can remove table B and the join from the query. HIGH-LEVEL SPECIFICATION: Here is an extended explanation of table elimination. Table elimination is a feature found in some modern query optimizers, of which Microsoft SQL Server 2005/2008 seems to have the most advanced implementation. Oracle 11g has also been confirmed to use table elimination but not to the same extent. Basically, what table elimination does, is to remove tables from the execution plan when it is unneccessary to include them. This can, of course, only happen if the right circumstances arise. Let us for example look at the following query: select A.colA from tableA A left outer join tableB B on B.id = A.id; When using A as the left table we ensure that the query will return at least as many rows as there are in that table. For rows where the join condition (B.id = A.id) is not met the selected column (A.colA) will contain a NULL value. However, the result set could actually contain more rows than what is found in tableA if there are duplicates of the column B.id in tableB. If A contains a row [1, "val1"] and B the rows [1, "other1a"],[1, "other1b"] then two rows will match in the join condition. The only way to know what the result will look like is to actually touch both tables during execution. Instead, let's say that tableB contains rows that make it possible to place a unique constraint on the column B.id, for example and often the case a primary key. In this situation we know that we will get exactly as many rows as there are in tableA, since joining with tableB cannot introduce any duplicates. If further, as in the example query, we do not select any columns from tableB, touching that table during execution is unneccessary. We can remove the whole join operation from the execution plan. Both SQL Server 2005/2008 and Oracle 11g will deploy table elimination in the case described above. Let us look at a more advanced query, where Oracle fails. select A.colA from tableA A left outer join tableB B on B.id = A.id and B.fromDate = ( select max(sub.fromDate) from tableB sub where sub.id = A.id ); In this example we have added another join condition, which ensures that we only pick the matching row from tableB having the latest fromDate. In this case tableB will contain duplicates of the column B.id, so in order to ensure uniqueness the primary key has to contain the fromDate column as well. In other words the primary key of tableB is (B.id, B.fromDate). Furthermore, since the subselect ensures that we only pick the latest B.fromDate for a given B.id we know that at most one row will match the join condition. We will again have the situation where joining with tableB cannot affect the number of rows in the result set. Since we do not select any columns from tableB, the whole join operation can be eliminated from the execution plan. SQL Server 2005/2008 will deploy table elimination in this situation as well. We have not found a way to make Oracle 11g use it for this type of query. Queries like these arise in two situations. Either when you have denormalized model consisting of a fact table with several related dimension tables, or when you have a highly normalized model where each attribute is stored in its own table. The example with the subselect is common whenever you store historized/versioned data. LOW-LEVEL DESIGN: <contents> 1. Conditions for removal 2. Removal operation properties 3. Removal operation 4. User interface 5. Todo, issues to resolve 5.1 To resolve 5.2 Resolved </contents> It's not really about elimination of tables, it's about elimination of inner sides of outer joins. 1. Conditions for removal ------------------------- We can eliminate an inner side of outer join if: 1. For each record combination of outer tables, it will always produce exactly one record. 2. There are no references to columns of the inner tables anywhere else in the query. #1 means that every table inside the outer join nest is: - is a constant table: = because it can be accessed via eq_ref(const) access, or = it is a zero-rows or one-row MyISAM-like table [MARK1] - has an eq_ref access method candidate. #2 means that WHERE clause, ON clauses of embedding outer joins, ORDER BY, GROUP BY and HAVING do not refer to the inner tables of the outer join nest. 2. Removal operation properties ------------------------------- * There is always one way to remove (no choice to remove either this or that) * It is always better to remove as much tables as possible (at least within our cost model). Thus, no need for any cost calculations/etc. It's an unconditional rewrite. 3. Removal operation -------------------- * Remove the outer join nest's nested join structure (i.e. get the outer join's TABLE_LIST object $OJ and remove it from $OJ->embedding, $OJ->embedding->nested_join. Update table_map's of all ancestor nested joins). [MARK2] * Move the tables and their JOIN_TABs to front like it is done with const tables, with exception that if eliminated outer join nest was within another outer join nest, that shouldn't prevent us from moving away the eliminated tables. * Update join->table_count and all-join-tables bitmap. * That's it. Nothing else? 4. User interface ----------------- * We'll add an @@optimizer switch flag for table elimination. Tentative name: 'table_elimination'. * With EXPLAIN, there are two options: - Show removed tables in a way similar to const tables, with some indication that they are removed. - Do not show them altogether. (the second one seems to be better? We're targeting a situation with VIEWs, where the user would not care about what tables were added into his query and then discarded from it?) 5. Todo, issues to resolve -------------------------- 5.1 To resolve ~~~~~~~~~~~~~~ - See EXPLAIN question in section #4. - Re-check how this works with equality propagation. - Relationship with prepared statements. On one hand, it's natural to desire to make table elimination a once-per-statement operation, like outer->inner join conversion. We'll have to limit the applicability by removing [MARK1] as that can change during lifetime of the statement. The other option is to do table elimination every time. This will require to rework operation [MARK2] to be undoable. I'm leaning towards doing the former. With anchor modeling, it is unlikely that we'll meet outer joins which have N inner tables of which some are 1-row MyISAM tables that do not have primary key. 5.2 Resolved ~~~~~~~~~~~~ - outer->inner join conversion is not a problem for table elimination. We make outer->inner conversions based on predicates in WHERE. If the WHERE referred to an inner table (requirement for OJ->IJ conversion) then table elimination would not be applicable anyway. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] Updated (by Guest): Table elimination (17)
by worklog-noreply＠askmonty.org 22 May '09

22 May '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Table elimination CREATION DATE..: Sun, 10 May 2009, 19:57 SUPERVISOR.....: Monty IMPLEMENTOR....: Psergey COPIES TO......: CATEGORY.......: Server-Sprint TASK ID........: 17 (http://askmonty.org/worklog/?tid=17) VERSION........: Maria-2.0 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 0 (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Category updated. --- /tmp/wklog.17.old.30162 2009-05-22 17:00:28.000000000 +0300 +++ /tmp/wklog.17.new.30162 2009-05-22 17:00:28.000000000 +0300 @@ -1 +1 @@ -Maria-Sprint +Server-Sprint -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30162 2009-05-22 17:00:28.000000000 +0300 +++ /tmp/wklog.17.new.30162 2009-05-22 17:00:28.000000000 +0300 @@ -1 +1 @@ -2.0 +9.x -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30162 2009-05-22 17:00:28.000000000 +0300 +++ /tmp/wklog.17.new.30162 2009-05-22 17:00:28.000000000 +0300 @@ -1 +1 @@ -9.x +Maria-2.0 -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Category updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Server-RawIdeaBin +Maria-Sprint -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Server-5.2 +2.0 -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Status updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Un-Assigned +Assigned -=-=(Guest - Thu, 21 May 2009, 22:54)=-=- Low Level Design modified. --- /tmp/wklog.17.old.28468 2009-05-21 22:54:28.000000000 +0300 +++ /tmp/wklog.17.new.28468 2009-05-21 22:54:28.000000000 +0300 @@ -77,7 +77,7 @@ - Relationship with prepared statements. On one hand, it's natural to desire to make table elimination a once-per-statement operation, like outer->inner join conversion. We'll have - to limit the applicablity by removing [MARK1] as that can change during + to limit the applicability by removing [MARK1] as that can change during lifetime of the statement. The other option is to do table elimination every time. This will require to -=-=(Psergey - Tue, 19 May 2009, 20:20)=-=- Low Level Design modified. --- /tmp/wklog.17.old.27126 2009-05-19 20:20:06.000000000 +0300 +++ /tmp/wklog.17.new.27126 2009-05-19 20:20:06.000000000 +0300 @@ -48,6 +48,10 @@ another outer join nest, that shouldn't prevent us from moving away the eliminated tables. +* Update join->table_count and all-join-tables bitmap. + +* That's it. Nothing else? + 4. User interface ----------------- * We'll add an @@optimizer switch flag for table elimination. Tentative -=-=(Psergey - Tue, 19 May 2009, 20:18)=-=- Low Level Design modified. --- /tmp/wklog.17.old.27035 2009-05-19 20:18:27.000000000 +0300 +++ /tmp/wklog.17.new.27035 2009-05-19 20:18:27.000000000 +0300 @@ -1 +1,92 @@ +<contents> +1. Conditions for removal +2. Removal operation properties +3. Removal operation +4. User interface +5. Todo, issues to resolve +5.1 To resolve +5.2 Resolved +</contents> + +It's not really about elimination of tables, it's about elimination of inner +sides of outer joins. + +1. Conditions for removal +------------------------- +We can eliminate an inner side of outer join if: +1. For each record combination of outer tables, it will always produce + exactly one record. +2. There are no references to columns of the inner tables anywhere else in + the query. + +#1 means that every table inside the outer join nest is: + - is a constant table: + = because it can be accessed via eq_ref(const) access, or + = it is a zero-rows or one-row MyISAM-like table [MARK1] + - has an eq_ref access method candidate. + +#2 means that WHERE clause, ON clauses of embedding outer joins, ORDER BY, + GROUP BY and HAVING do not refer to the inner tables of the outer join + nest. + +2. Removal operation properties +------------------------------- +* There is always one way to remove (no choice to remove either this or that) +* It is always better to remove as much tables as possible (at least within + our cost model). +Thus, no need for any cost calculations/etc. It's an unconditional rewrite. + +3. Removal operation +-------------------- +* Remove the outer join nest's nested join structure (i.e. get the + outer join's TABLE_LIST object $OJ and remove it from $OJ->embedding, + $OJ->embedding->nested_join. Update table_map's of all ancestor nested + joins). [MARK2] + +* Move the tables and their JOIN_TABs to front like it is done with const + tables, with exception that if eliminated outer join nest was within + another outer join nest, that shouldn't prevent us from moving away the + eliminated tables. + +4. User interface +----------------- +* We'll add an @@optimizer switch flag for table elimination. Tentative + name: 'table_elimination'. + +* With EXPLAIN, there are two options: + - Show removed tables in a way similar to const tables, with some + indication that they are removed. + - Do not show them altogether. +(the second one seems to be better? We're targeting a situation with VIEWs, +where the user would not care about what tables were added into his query +and then discarded from it?) + +5. Todo, issues to resolve +-------------------------- + +5.1 To resolve +~~~~~~~~~~~~~~ +- See EXPLAIN question in section #4. + +- Re-check how this works with equality propagation. + +- Relationship with prepared statements. + On one hand, it's natural to desire to make table elimination a + once-per-statement operation, like outer->inner join conversion. We'll have + to limit the applicablity by removing [MARK1] as that can change during + lifetime of the statement. + + The other option is to do table elimination every time. This will require to + rework operation [MARK2] to be undoable. + + I'm leaning towards doing the former. With anchor modeling, it is unlikely + that we'll meet outer joins which have N inner tables of which some are 1-row + MyISAM tables that do not have primary key. + +5.2 Resolved +~~~~~~~~~~~~ +- outer->inner join conversion is not a problem for table elimination. + We make outer->inner conversions based on predicates in WHERE. If the WHERE + referred to an inner table (requirement for OJ->IJ conversion) then table + elimination would not be applicable anyway. -=-=(Guest - Tue, 19 May 2009, 18:27)=-=- Title modified. --- /tmp/wklog.17.old.22023 2009-05-19 18:27:41.000000000 +0300 +++ /tmp/wklog.17.new.22023 2009-05-19 18:27:41.000000000 +0300 @@ -1 +1 @@ -Table ellimination +Table elimination ------------------------------------------------------------ -=-=(View All Progress Notes, 11 total)=-=- http://askmonty.org/worklog/index.pl?tid=17&nolimit=1 DESCRIPTION: Eliminate not needed tables from SELECT queries.. This will speed up some views and automatically generated queries. Example: CREATE TABLE B (id int primary key); select A.colA from tableA A left outer join tableB B on B.id = A.id; In this case we can remove table B and the join from the query. HIGH-LEVEL SPECIFICATION: Here is an extended explanation of table elimination. Table elimination is a feature found in some modern query optimizers, of which Microsoft SQL Server 2005/2008 seems to have the most advanced implementation. Oracle 11g has also been confirmed to use table elimination but not to the same extent. Basically, what table elimination does, is to remove tables from the execution plan when it is unneccessary to include them. This can, of course, only happen if the right circumstances arise. Let us for example look at the following query: select A.colA from tableA A left outer join tableB B on B.id = A.id; When using A as the left table we ensure that the query will return at least as many rows as there are in that table. For rows where the join condition (B.id = A.id) is not met the selected column (A.colA) will contain a NULL value. However, the result set could actually contain more rows than what is found in tableA if there are duplicates of the column B.id in tableB. If A contains a row [1, "val1"] and B the rows [1, "other1a"],[1, "other1b"] then two rows will match in the join condition. The only way to know what the result will look like is to actually touch both tables during execution. Instead, let's say that tableB contains rows that make it possible to place a unique constraint on the column B.id, for example and often the case a primary key. In this situation we know that we will get exactly as many rows as there are in tableA, since joining with tableB cannot introduce any duplicates. If further, as in the example query, we do not select any columns from tableB, touching that table during execution is unneccessary. We can remove the whole join operation from the execution plan. Both SQL Server 2005/2008 and Oracle 11g will deploy table elimination in the case described above. Let us look at a more advanced query, where Oracle fails. select A.colA from tableA A left outer join tableB B on B.id = A.id and B.fromDate = ( select max(sub.fromDate) from tableB sub where sub.id = A.id ); In this example we have added another join condition, which ensures that we only pick the matching row from tableB having the latest fromDate. In this case tableB will contain duplicates of the column B.id, so in order to ensure uniqueness the primary key has to contain the fromDate column as well. In other words the primary key of tableB is (B.id, B.fromDate). Furthermore, since the subselect ensures that we only pick the latest B.fromDate for a given B.id we know that at most one row will match the join condition. We will again have the situation where joining with tableB cannot affect the number of rows in the result set. Since we do not select any columns from tableB, the whole join operation can be eliminated from the execution plan. SQL Server 2005/2008 will deploy table elimination in this situation as well. We have not found a way to make Oracle 11g use it for this type of query. Queries like these arise in two situations. Either when you have denormalized model consisting of a fact table with several related dimension tables, or when you have a highly normalized model where each attribute is stored in its own table. The example with the subselect is common whenever you store historized/versioned data. LOW-LEVEL DESIGN: <contents> 1. Conditions for removal 2. Removal operation properties 3. Removal operation 4. User interface 5. Todo, issues to resolve 5.1 To resolve 5.2 Resolved </contents> It's not really about elimination of tables, it's about elimination of inner sides of outer joins. 1. Conditions for removal ------------------------- We can eliminate an inner side of outer join if: 1. For each record combination of outer tables, it will always produce exactly one record. 2. There are no references to columns of the inner tables anywhere else in the query. #1 means that every table inside the outer join nest is: - is a constant table: = because it can be accessed via eq_ref(const) access, or = it is a zero-rows or one-row MyISAM-like table [MARK1] - has an eq_ref access method candidate. #2 means that WHERE clause, ON clauses of embedding outer joins, ORDER BY, GROUP BY and HAVING do not refer to the inner tables of the outer join nest. 2. Removal operation properties ------------------------------- * There is always one way to remove (no choice to remove either this or that) * It is always better to remove as much tables as possible (at least within our cost model). Thus, no need for any cost calculations/etc. It's an unconditional rewrite. 3. Removal operation -------------------- * Remove the outer join nest's nested join structure (i.e. get the outer join's TABLE_LIST object $OJ and remove it from $OJ->embedding, $OJ->embedding->nested_join. Update table_map's of all ancestor nested joins). [MARK2] * Move the tables and their JOIN_TABs to front like it is done with const tables, with exception that if eliminated outer join nest was within another outer join nest, that shouldn't prevent us from moving away the eliminated tables. * Update join->table_count and all-join-tables bitmap. * That's it. Nothing else? 4. User interface ----------------- * We'll add an @@optimizer switch flag for table elimination. Tentative name: 'table_elimination'. * With EXPLAIN, there are two options: - Show removed tables in a way similar to const tables, with some indication that they are removed. - Do not show them altogether. (the second one seems to be better? We're targeting a situation with VIEWs, where the user would not care about what tables were added into his query and then discarded from it?) 5. Todo, issues to resolve -------------------------- 5.1 To resolve ~~~~~~~~~~~~~~ - See EXPLAIN question in section #4. - Re-check how this works with equality propagation. - Relationship with prepared statements. On one hand, it's natural to desire to make table elimination a once-per-statement operation, like outer->inner join conversion. We'll have to limit the applicability by removing [MARK1] as that can change during lifetime of the statement. The other option is to do table elimination every time. This will require to rework operation [MARK2] to be undoable. I'm leaning towards doing the former. With anchor modeling, it is unlikely that we'll meet outer joins which have N inner tables of which some are 1-row MyISAM tables that do not have primary key. 5.2 Resolved ~~~~~~~~~~~~ - outer->inner join conversion is not a problem for table elimination. We make outer->inner conversions based on predicates in WHERE. If the WHERE referred to an inner table (requirement for OJ->IJ conversion) then table elimination would not be applicable anyway. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] Updated (by Guest): Table elimination (17)
by worklog-noreply＠askmonty.org 22 May '09

22 May '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Table elimination CREATION DATE..: Sun, 10 May 2009, 19:57 SUPERVISOR.....: Monty IMPLEMENTOR....: Psergey COPIES TO......: CATEGORY.......: Server-Sprint TASK ID........: 17 (http://askmonty.org/worklog/?tid=17) VERSION........: Maria-2.0 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 0 (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Category updated. --- /tmp/wklog.17.old.30162 2009-05-22 17:00:28.000000000 +0300 +++ /tmp/wklog.17.new.30162 2009-05-22 17:00:28.000000000 +0300 @@ -1 +1 @@ -Maria-Sprint +Server-Sprint -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30162 2009-05-22 17:00:28.000000000 +0300 +++ /tmp/wklog.17.new.30162 2009-05-22 17:00:28.000000000 +0300 @@ -1 +1 @@ -2.0 +9.x -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30162 2009-05-22 17:00:28.000000000 +0300 +++ /tmp/wklog.17.new.30162 2009-05-22 17:00:28.000000000 +0300 @@ -1 +1 @@ -9.x +Maria-2.0 -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Category updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Server-RawIdeaBin +Maria-Sprint -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Server-5.2 +2.0 -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Status updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Un-Assigned +Assigned -=-=(Guest - Thu, 21 May 2009, 22:54)=-=- Low Level Design modified. --- /tmp/wklog.17.old.28468 2009-05-21 22:54:28.000000000 +0300 +++ /tmp/wklog.17.new.28468 2009-05-21 22:54:28.000000000 +0300 @@ -77,7 +77,7 @@ - Relationship with prepared statements. On one hand, it's natural to desire to make table elimination a once-per-statement operation, like outer->inner join conversion. We'll have - to limit the applicablity by removing [MARK1] as that can change during + to limit the applicability by removing [MARK1] as that can change during lifetime of the statement. The other option is to do table elimination every time. This will require to -=-=(Psergey - Tue, 19 May 2009, 20:20)=-=- Low Level Design modified. --- /tmp/wklog.17.old.27126 2009-05-19 20:20:06.000000000 +0300 +++ /tmp/wklog.17.new.27126 2009-05-19 20:20:06.000000000 +0300 @@ -48,6 +48,10 @@ another outer join nest, that shouldn't prevent us from moving away the eliminated tables. +* Update join->table_count and all-join-tables bitmap. + +* That's it. Nothing else? + 4. User interface ----------------- * We'll add an @@optimizer switch flag for table elimination. Tentative -=-=(Psergey - Tue, 19 May 2009, 20:18)=-=- Low Level Design modified. --- /tmp/wklog.17.old.27035 2009-05-19 20:18:27.000000000 +0300 +++ /tmp/wklog.17.new.27035 2009-05-19 20:18:27.000000000 +0300 @@ -1 +1,92 @@ +<contents> +1. Conditions for removal +2. Removal operation properties +3. Removal operation +4. User interface +5. Todo, issues to resolve +5.1 To resolve +5.2 Resolved +</contents> + +It's not really about elimination of tables, it's about elimination of inner +sides of outer joins. + +1. Conditions for removal +------------------------- +We can eliminate an inner side of outer join if: +1. For each record combination of outer tables, it will always produce + exactly one record. +2. There are no references to columns of the inner tables anywhere else in + the query. + +#1 means that every table inside the outer join nest is: + - is a constant table: + = because it can be accessed via eq_ref(const) access, or + = it is a zero-rows or one-row MyISAM-like table [MARK1] + - has an eq_ref access method candidate. + +#2 means that WHERE clause, ON clauses of embedding outer joins, ORDER BY, + GROUP BY and HAVING do not refer to the inner tables of the outer join + nest. + +2. Removal operation properties +------------------------------- +* There is always one way to remove (no choice to remove either this or that) +* It is always better to remove as much tables as possible (at least within + our cost model). +Thus, no need for any cost calculations/etc. It's an unconditional rewrite. + +3. Removal operation +-------------------- +* Remove the outer join nest's nested join structure (i.e. get the + outer join's TABLE_LIST object $OJ and remove it from $OJ->embedding, + $OJ->embedding->nested_join. Update table_map's of all ancestor nested + joins). [MARK2] + +* Move the tables and their JOIN_TABs to front like it is done with const + tables, with exception that if eliminated outer join nest was within + another outer join nest, that shouldn't prevent us from moving away the + eliminated tables. + +4. User interface +----------------- +* We'll add an @@optimizer switch flag for table elimination. Tentative + name: 'table_elimination'. + +* With EXPLAIN, there are two options: + - Show removed tables in a way similar to const tables, with some + indication that they are removed. + - Do not show them altogether. +(the second one seems to be better? We're targeting a situation with VIEWs, +where the user would not care about what tables were added into his query +and then discarded from it?) + +5. Todo, issues to resolve +-------------------------- + +5.1 To resolve +~~~~~~~~~~~~~~ +- See EXPLAIN question in section #4. + +- Re-check how this works with equality propagation. + +- Relationship with prepared statements. + On one hand, it's natural to desire to make table elimination a + once-per-statement operation, like outer->inner join conversion. We'll have + to limit the applicablity by removing [MARK1] as that can change during + lifetime of the statement. + + The other option is to do table elimination every time. This will require to + rework operation [MARK2] to be undoable. + + I'm leaning towards doing the former. With anchor modeling, it is unlikely + that we'll meet outer joins which have N inner tables of which some are 1-row + MyISAM tables that do not have primary key. + +5.2 Resolved +~~~~~~~~~~~~ +- outer->inner join conversion is not a problem for table elimination. + We make outer->inner conversions based on predicates in WHERE. If the WHERE + referred to an inner table (requirement for OJ->IJ conversion) then table + elimination would not be applicable anyway. -=-=(Guest - Tue, 19 May 2009, 18:27)=-=- Title modified. --- /tmp/wklog.17.old.22023 2009-05-19 18:27:41.000000000 +0300 +++ /tmp/wklog.17.new.22023 2009-05-19 18:27:41.000000000 +0300 @@ -1 +1 @@ -Table ellimination +Table elimination ------------------------------------------------------------ -=-=(View All Progress Notes, 11 total)=-=- http://askmonty.org/worklog/index.pl?tid=17&nolimit=1 DESCRIPTION: Eliminate not needed tables from SELECT queries.. This will speed up some views and automatically generated queries. Example: CREATE TABLE B (id int primary key); select A.colA from tableA A left outer join tableB B on B.id = A.id; In this case we can remove table B and the join from the query. HIGH-LEVEL SPECIFICATION: Here is an extended explanation of table elimination. Table elimination is a feature found in some modern query optimizers, of which Microsoft SQL Server 2005/2008 seems to have the most advanced implementation. Oracle 11g has also been confirmed to use table elimination but not to the same extent. Basically, what table elimination does, is to remove tables from the execution plan when it is unneccessary to include them. This can, of course, only happen if the right circumstances arise. Let us for example look at the following query: select A.colA from tableA A left outer join tableB B on B.id = A.id; When using A as the left table we ensure that the query will return at least as many rows as there are in that table. For rows where the join condition (B.id = A.id) is not met the selected column (A.colA) will contain a NULL value. However, the result set could actually contain more rows than what is found in tableA if there are duplicates of the column B.id in tableB. If A contains a row [1, "val1"] and B the rows [1, "other1a"],[1, "other1b"] then two rows will match in the join condition. The only way to know what the result will look like is to actually touch both tables during execution. Instead, let's say that tableB contains rows that make it possible to place a unique constraint on the column B.id, for example and often the case a primary key. In this situation we know that we will get exactly as many rows as there are in tableA, since joining with tableB cannot introduce any duplicates. If further, as in the example query, we do not select any columns from tableB, touching that table during execution is unneccessary. We can remove the whole join operation from the execution plan. Both SQL Server 2005/2008 and Oracle 11g will deploy table elimination in the case described above. Let us look at a more advanced query, where Oracle fails. select A.colA from tableA A left outer join tableB B on B.id = A.id and B.fromDate = ( select max(sub.fromDate) from tableB sub where sub.id = A.id ); In this example we have added another join condition, which ensures that we only pick the matching row from tableB having the latest fromDate. In this case tableB will contain duplicates of the column B.id, so in order to ensure uniqueness the primary key has to contain the fromDate column as well. In other words the primary key of tableB is (B.id, B.fromDate). Furthermore, since the subselect ensures that we only pick the latest B.fromDate for a given B.id we know that at most one row will match the join condition. We will again have the situation where joining with tableB cannot affect the number of rows in the result set. Since we do not select any columns from tableB, the whole join operation can be eliminated from the execution plan. SQL Server 2005/2008 will deploy table elimination in this situation as well. We have not found a way to make Oracle 11g use it for this type of query. Queries like these arise in two situations. Either when you have denormalized model consisting of a fact table with several related dimension tables, or when you have a highly normalized model where each attribute is stored in its own table. The example with the subselect is common whenever you store historized/versioned data. LOW-LEVEL DESIGN: <contents> 1. Conditions for removal 2. Removal operation properties 3. Removal operation 4. User interface 5. Todo, issues to resolve 5.1 To resolve 5.2 Resolved </contents> It's not really about elimination of tables, it's about elimination of inner sides of outer joins. 1. Conditions for removal ------------------------- We can eliminate an inner side of outer join if: 1. For each record combination of outer tables, it will always produce exactly one record. 2. There are no references to columns of the inner tables anywhere else in the query. #1 means that every table inside the outer join nest is: - is a constant table: = because it can be accessed via eq_ref(const) access, or = it is a zero-rows or one-row MyISAM-like table [MARK1] - has an eq_ref access method candidate. #2 means that WHERE clause, ON clauses of embedding outer joins, ORDER BY, GROUP BY and HAVING do not refer to the inner tables of the outer join nest. 2. Removal operation properties ------------------------------- * There is always one way to remove (no choice to remove either this or that) * It is always better to remove as much tables as possible (at least within our cost model). Thus, no need for any cost calculations/etc. It's an unconditional rewrite. 3. Removal operation -------------------- * Remove the outer join nest's nested join structure (i.e. get the outer join's TABLE_LIST object $OJ and remove it from $OJ->embedding, $OJ->embedding->nested_join. Update table_map's of all ancestor nested joins). [MARK2] * Move the tables and their JOIN_TABs to front like it is done with const tables, with exception that if eliminated outer join nest was within another outer join nest, that shouldn't prevent us from moving away the eliminated tables. * Update join->table_count and all-join-tables bitmap. * That's it. Nothing else? 4. User interface ----------------- * We'll add an @@optimizer switch flag for table elimination. Tentative name: 'table_elimination'. * With EXPLAIN, there are two options: - Show removed tables in a way similar to const tables, with some indication that they are removed. - Do not show them altogether. (the second one seems to be better? We're targeting a situation with VIEWs, where the user would not care about what tables were added into his query and then discarded from it?) 5. Todo, issues to resolve -------------------------- 5.1 To resolve ~~~~~~~~~~~~~~ - See EXPLAIN question in section #4. - Re-check how this works with equality propagation. - Relationship with prepared statements. On one hand, it's natural to desire to make table elimination a once-per-statement operation, like outer->inner join conversion. We'll have to limit the applicability by removing [MARK1] as that can change during lifetime of the statement. The other option is to do table elimination every time. This will require to rework operation [MARK2] to be undoable. I'm leaning towards doing the former. With anchor modeling, it is unlikely that we'll meet outer joins which have N inner tables of which some are 1-row MyISAM tables that do not have primary key. 5.2 Resolved ~~~~~~~~~~~~ - outer->inner join conversion is not a problem for table elimination. We make outer->inner conversions based on predicates in WHERE. If the WHERE referred to an inner table (requirement for OJ->IJ conversion) then table elimination would not be applicable anyway. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] Updated (by Guest): Table elimination (17)
by worklog-noreply＠askmonty.org 22 May '09

22 May '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Table elimination CREATION DATE..: Sun, 10 May 2009, 19:57 SUPERVISOR.....: Monty IMPLEMENTOR....: Psergey COPIES TO......: CATEGORY.......: Maria-Sprint TASK ID........: 17 (http://askmonty.org/worklog/?tid=17) VERSION........: 2.0 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 0 (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Category updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Server-RawIdeaBin +Maria-Sprint -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Server-5.2 +2.0 -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Status updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Un-Assigned +Assigned -=-=(Guest - Thu, 21 May 2009, 22:54)=-=- Low Level Design modified. --- /tmp/wklog.17.old.28468 2009-05-21 22:54:28.000000000 +0300 +++ /tmp/wklog.17.new.28468 2009-05-21 22:54:28.000000000 +0300 @@ -77,7 +77,7 @@ - Relationship with prepared statements. On one hand, it's natural to desire to make table elimination a once-per-statement operation, like outer->inner join conversion. We'll have - to limit the applicablity by removing [MARK1] as that can change during + to limit the applicability by removing [MARK1] as that can change during lifetime of the statement. The other option is to do table elimination every time. This will require to -=-=(Psergey - Tue, 19 May 2009, 20:20)=-=- Low Level Design modified. --- /tmp/wklog.17.old.27126 2009-05-19 20:20:06.000000000 +0300 +++ /tmp/wklog.17.new.27126 2009-05-19 20:20:06.000000000 +0300 @@ -48,6 +48,10 @@ another outer join nest, that shouldn't prevent us from moving away the eliminated tables. +* Update join->table_count and all-join-tables bitmap. + +* That's it. Nothing else? + 4. User interface ----------------- * We'll add an @@optimizer switch flag for table elimination. Tentative -=-=(Psergey - Tue, 19 May 2009, 20:18)=-=- Low Level Design modified. --- /tmp/wklog.17.old.27035 2009-05-19 20:18:27.000000000 +0300 +++ /tmp/wklog.17.new.27035 2009-05-19 20:18:27.000000000 +0300 @@ -1 +1,92 @@ +<contents> +1. Conditions for removal +2. Removal operation properties +3. Removal operation +4. User interface +5. Todo, issues to resolve +5.1 To resolve +5.2 Resolved +</contents> + +It's not really about elimination of tables, it's about elimination of inner +sides of outer joins. + +1. Conditions for removal +------------------------- +We can eliminate an inner side of outer join if: +1. For each record combination of outer tables, it will always produce + exactly one record. +2. There are no references to columns of the inner tables anywhere else in + the query. + +#1 means that every table inside the outer join nest is: + - is a constant table: + = because it can be accessed via eq_ref(const) access, or + = it is a zero-rows or one-row MyISAM-like table [MARK1] + - has an eq_ref access method candidate. + +#2 means that WHERE clause, ON clauses of embedding outer joins, ORDER BY, + GROUP BY and HAVING do not refer to the inner tables of the outer join + nest. + +2. Removal operation properties +------------------------------- +* There is always one way to remove (no choice to remove either this or that) +* It is always better to remove as much tables as possible (at least within + our cost model). +Thus, no need for any cost calculations/etc. It's an unconditional rewrite. + +3. Removal operation +-------------------- +* Remove the outer join nest's nested join structure (i.e. get the + outer join's TABLE_LIST object $OJ and remove it from $OJ->embedding, + $OJ->embedding->nested_join. Update table_map's of all ancestor nested + joins). [MARK2] + +* Move the tables and their JOIN_TABs to front like it is done with const + tables, with exception that if eliminated outer join nest was within + another outer join nest, that shouldn't prevent us from moving away the + eliminated tables. + +4. User interface +----------------- +* We'll add an @@optimizer switch flag for table elimination. Tentative + name: 'table_elimination'. + +* With EXPLAIN, there are two options: + - Show removed tables in a way similar to const tables, with some + indication that they are removed. + - Do not show them altogether. +(the second one seems to be better? We're targeting a situation with VIEWs, +where the user would not care about what tables were added into his query +and then discarded from it?) + +5. Todo, issues to resolve +-------------------------- + +5.1 To resolve +~~~~~~~~~~~~~~ +- See EXPLAIN question in section #4. + +- Re-check how this works with equality propagation. + +- Relationship with prepared statements. + On one hand, it's natural to desire to make table elimination a + once-per-statement operation, like outer->inner join conversion. We'll have + to limit the applicablity by removing [MARK1] as that can change during + lifetime of the statement. + + The other option is to do table elimination every time. This will require to + rework operation [MARK2] to be undoable. + + I'm leaning towards doing the former. With anchor modeling, it is unlikely + that we'll meet outer joins which have N inner tables of which some are 1-row + MyISAM tables that do not have primary key. + +5.2 Resolved +~~~~~~~~~~~~ +- outer->inner join conversion is not a problem for table elimination. + We make outer->inner conversions based on predicates in WHERE. If the WHERE + referred to an inner table (requirement for OJ->IJ conversion) then table + elimination would not be applicable anyway. -=-=(Guest - Tue, 19 May 2009, 18:27)=-=- Title modified. --- /tmp/wklog.17.old.22023 2009-05-19 18:27:41.000000000 +0300 +++ /tmp/wklog.17.new.22023 2009-05-19 18:27:41.000000000 +0300 @@ -1 +1 @@ -Table ellimination +Table elimination -=-=(Monty - Sun, 10 May 2009, 19:58)=-=- High-Level Specification modified. --- /tmp/wklog.17.old.5423 2009-05-10 19:58:06.000000000 +0300 +++ /tmp/wklog.17.new.5423 2009-05-10 19:58:06.000000000 +0300 @@ -1 +1,90 @@ +Here is an extended explanation of table elimination. + +Table elimination is a feature found in some modern query optimizers, of +which Microsoft SQL Server 2005/2008 seems to have the most advanced +implementation. Oracle 11g has also been confirmed to use table +elimination but not to the same extent. + +Basically, what table elimination does, is to remove tables from the +execution plan when it is unneccessary to include them. This can, of +course, only happen if the right circumstances arise. Let us for example +look at the following query: + +select + A.colA +from + tableA A +left outer join + tableB B +on + B.id = A.id; + +When using A as the left table we ensure that the query will return at +least as many rows as there are in that table. For rows where the join +condition (B.id = A.id) is not met the selected column (A.colA) will +contain a NULL value. + +However, the result set could actually contain more rows than what is +found in tableA if there are duplicates of the column B.id in tableB. If +A +contains a row [1, "val1"] and B the rows [1, "other1a"],[1, "other1b"] +then two rows will match in the join condition. The only way to know +what +the result will look like is to actually touch both tables during +execution. + +Instead, let's say that tableB contains rows that make it possible to +place a unique constraint on the column B.id, for example and often the +case a primary key. In this situation we know that we will get exactly +as +many rows as there are in tableA, since joining with tableB cannot +introduce any duplicates. If further, as in the example query, we do not +select any columns from tableB, touching that table during execution is +unneccessary. We can remove the whole join operation from the execution +plan. + +Both SQL Server 2005/2008 and Oracle 11g will deploy table elimination +in +the case described above. Let us look at a more advanced query, where +Oracle fails. + +select + A.colA +from + tableA A +left outer join + tableB B +on + B.id = A.id +and + B.fromDate = ( + select + max(sub.fromDate) + from + tableB sub + where + sub.id = A.id + ); + +In this example we have added another join condition, which ensures +that we only pick the matching row from tableB having the latest +fromDate. In this case tableB will contain duplicates of the column +B.id, so in order to ensure uniqueness the primary key has to contain +the fromDate column as well. In other words the primary key of tableB +is (B.id, B.fromDate). + +Furthermore, since the subselect ensures that we only pick the latest +B.fromDate for a given B.id we know that at most one row will match +the join condition. We will again have the situation where joining +with tableB cannot affect the number of rows in the result set. Since +we do not select any columns from tableB, the whole join operation can +be eliminated from the execution plan. + +SQL Server 2005/2008 will deploy table elimination in this situation as +well. We have not found a way to make Oracle 11g use it for this type of +query. Queries like these arise in two situations. Either when you have +denormalized model consisting of a fact table with several related +dimension tables, or when you have a highly normalized model where each +attribute is stored in its own table. The example with the subselect is +common whenever you store historized/versioned data. DESCRIPTION: Eliminate not needed tables from SELECT queries.. This will speed up some views and automatically generated queries. Example: CREATE TABLE B (id int primary key); select A.colA from tableA A left outer join tableB B on B.id = A.id; In this case we can remove table B and the join from the query. HIGH-LEVEL SPECIFICATION: Here is an extended explanation of table elimination. Table elimination is a feature found in some modern query optimizers, of which Microsoft SQL Server 2005/2008 seems to have the most advanced implementation. Oracle 11g has also been confirmed to use table elimination but not to the same extent. Basically, what table elimination does, is to remove tables from the execution plan when it is unneccessary to include them. This can, of course, only happen if the right circumstances arise. Let us for example look at the following query: select A.colA from tableA A left outer join tableB B on B.id = A.id; When using A as the left table we ensure that the query will return at least as many rows as there are in that table. For rows where the join condition (B.id = A.id) is not met the selected column (A.colA) will contain a NULL value. However, the result set could actually contain more rows than what is found in tableA if there are duplicates of the column B.id in tableB. If A contains a row [1, "val1"] and B the rows [1, "other1a"],[1, "other1b"] then two rows will match in the join condition. The only way to know what the result will look like is to actually touch both tables during execution. Instead, let's say that tableB contains rows that make it possible to place a unique constraint on the column B.id, for example and often the case a primary key. In this situation we know that we will get exactly as many rows as there are in tableA, since joining with tableB cannot introduce any duplicates. If further, as in the example query, we do not select any columns from tableB, touching that table during execution is unneccessary. We can remove the whole join operation from the execution plan. Both SQL Server 2005/2008 and Oracle 11g will deploy table elimination in the case described above. Let us look at a more advanced query, where Oracle fails. select A.colA from tableA A left outer join tableB B on B.id = A.id and B.fromDate = ( select max(sub.fromDate) from tableB sub where sub.id = A.id ); In this example we have added another join condition, which ensures that we only pick the matching row from tableB having the latest fromDate. In this case tableB will contain duplicates of the column B.id, so in order to ensure uniqueness the primary key has to contain the fromDate column as well. In other words the primary key of tableB is (B.id, B.fromDate). Furthermore, since the subselect ensures that we only pick the latest B.fromDate for a given B.id we know that at most one row will match the join condition. We will again have the situation where joining with tableB cannot affect the number of rows in the result set. Since we do not select any columns from tableB, the whole join operation can be eliminated from the execution plan. SQL Server 2005/2008 will deploy table elimination in this situation as well. We have not found a way to make Oracle 11g use it for this type of query. Queries like these arise in two situations. Either when you have denormalized model consisting of a fact table with several related dimension tables, or when you have a highly normalized model where each attribute is stored in its own table. The example with the subselect is common whenever you store historized/versioned data. LOW-LEVEL DESIGN: <contents> 1. Conditions for removal 2. Removal operation properties 3. Removal operation 4. User interface 5. Todo, issues to resolve 5.1 To resolve 5.2 Resolved </contents> It's not really about elimination of tables, it's about elimination of inner sides of outer joins. 1. Conditions for removal ------------------------- We can eliminate an inner side of outer join if: 1. For each record combination of outer tables, it will always produce exactly one record. 2. There are no references to columns of the inner tables anywhere else in the query. #1 means that every table inside the outer join nest is: - is a constant table: = because it can be accessed via eq_ref(const) access, or = it is a zero-rows or one-row MyISAM-like table [MARK1] - has an eq_ref access method candidate. #2 means that WHERE clause, ON clauses of embedding outer joins, ORDER BY, GROUP BY and HAVING do not refer to the inner tables of the outer join nest. 2. Removal operation properties ------------------------------- * There is always one way to remove (no choice to remove either this or that) * It is always better to remove as much tables as possible (at least within our cost model). Thus, no need for any cost calculations/etc. It's an unconditional rewrite. 3. Removal operation -------------------- * Remove the outer join nest's nested join structure (i.e. get the outer join's TABLE_LIST object $OJ and remove it from $OJ->embedding, $OJ->embedding->nested_join. Update table_map's of all ancestor nested joins). [MARK2] * Move the tables and their JOIN_TABs to front like it is done with const tables, with exception that if eliminated outer join nest was within another outer join nest, that shouldn't prevent us from moving away the eliminated tables. * Update join->table_count and all-join-tables bitmap. * That's it. Nothing else? 4. User interface ----------------- * We'll add an @@optimizer switch flag for table elimination. Tentative name: 'table_elimination'. * With EXPLAIN, there are two options: - Show removed tables in a way similar to const tables, with some indication that they are removed. - Do not show them altogether. (the second one seems to be better? We're targeting a situation with VIEWs, where the user would not care about what tables were added into his query and then discarded from it?) 5. Todo, issues to resolve -------------------------- 5.1 To resolve ~~~~~~~~~~~~~~ - See EXPLAIN question in section #4. - Re-check how this works with equality propagation. - Relationship with prepared statements. On one hand, it's natural to desire to make table elimination a once-per-statement operation, like outer->inner join conversion. We'll have to limit the applicability by removing [MARK1] as that can change during lifetime of the statement. The other option is to do table elimination every time. This will require to rework operation [MARK2] to be undoable. I'm leaning towards doing the former. With anchor modeling, it is unlikely that we'll meet outer joins which have N inner tables of which some are 1-row MyISAM tables that do not have primary key. 5.2 Resolved ~~~~~~~~~~~~ - outer->inner join conversion is not a problem for table elimination. We make outer->inner conversions based on predicates in WHERE. If the WHERE referred to an inner table (requirement for OJ->IJ conversion) then table elimination would not be applicable anyway. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] Updated (by Guest): Table elimination (17)
by worklog-noreply＠askmonty.org 22 May '09

22 May '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Table elimination CREATION DATE..: Sun, 10 May 2009, 19:57 SUPERVISOR.....: Monty IMPLEMENTOR....: Psergey COPIES TO......: CATEGORY.......: Maria-Sprint TASK ID........: 17 (http://askmonty.org/worklog/?tid=17) VERSION........: 2.0 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 0 (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Category updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Server-RawIdeaBin +Maria-Sprint -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Version updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Server-5.2 +2.0 -=-=(Guest - Fri, 22 May 2009, 17:00)=-=- Status updated. --- /tmp/wklog.17.old.30122 2009-05-22 17:00:02.000000000 +0300 +++ /tmp/wklog.17.new.30122 2009-05-22 17:00:02.000000000 +0300 @@ -1 +1 @@ -Un-Assigned +Assigned -=-=(Guest - Thu, 21 May 2009, 22:54)=-=- Low Level Design modified. --- /tmp/wklog.17.old.28468 2009-05-21 22:54:28.000000000 +0300 +++ /tmp/wklog.17.new.28468 2009-05-21 22:54:28.000000000 +0300 @@ -77,7 +77,7 @@ - Relationship with prepared statements. On one hand, it's natural to desire to make table elimination a once-per-statement operation, like outer->inner join conversion. We'll have - to limit the applicablity by removing [MARK1] as that can change during + to limit the applicability by removing [MARK1] as that can change during lifetime of the statement. The other option is to do table elimination every time. This will require to -=-=(Psergey - Tue, 19 May 2009, 20:20)=-=- Low Level Design modified. --- /tmp/wklog.17.old.27126 2009-05-19 20:20:06.000000000 +0300 +++ /tmp/wklog.17.new.27126 2009-05-19 20:20:06.000000000 +0300 @@ -48,6 +48,10 @@ another outer join nest, that shouldn't prevent us from moving away the eliminated tables. +* Update join->table_count and all-join-tables bitmap. + +* That's it. Nothing else? + 4. User interface ----------------- * We'll add an @@optimizer switch flag for table elimination. Tentative -=-=(Psergey - Tue, 19 May 2009, 20:18)=-=- Low Level Design modified. --- /tmp/wklog.17.old.27035 2009-05-19 20:18:27.000000000 +0300 +++ /tmp/wklog.17.new.27035 2009-05-19 20:18:27.000000000 +0300 @@ -1 +1,92 @@ +<contents> +1. Conditions for removal +2. Removal operation properties +3. Removal operation +4. User interface +5. Todo, issues to resolve +5.1 To resolve +5.2 Resolved +</contents> + +It's not really about elimination of tables, it's about elimination of inner +sides of outer joins. + +1. Conditions for removal +------------------------- +We can eliminate an inner side of outer join if: +1. For each record combination of outer tables, it will always produce + exactly one record. +2. There are no references to columns of the inner tables anywhere else in + the query. + +#1 means that every table inside the outer join nest is: + - is a constant table: + = because it can be accessed via eq_ref(const) access, or + = it is a zero-rows or one-row MyISAM-like table [MARK1] + - has an eq_ref access method candidate. + +#2 means that WHERE clause, ON clauses of embedding outer joins, ORDER BY, + GROUP BY and HAVING do not refer to the inner tables of the outer join + nest. + +2. Removal operation properties +------------------------------- +* There is always one way to remove (no choice to remove either this or that) +* It is always better to remove as much tables as possible (at least within + our cost model). +Thus, no need for any cost calculations/etc. It's an unconditional rewrite. + +3. Removal operation +-------------------- +* Remove the outer join nest's nested join structure (i.e. get the + outer join's TABLE_LIST object $OJ and remove it from $OJ->embedding, + $OJ->embedding->nested_join. Update table_map's of all ancestor nested + joins). [MARK2] + +* Move the tables and their JOIN_TABs to front like it is done with const + tables, with exception that if eliminated outer join nest was within + another outer join nest, that shouldn't prevent us from moving away the + eliminated tables. + +4. User interface +----------------- +* We'll add an @@optimizer switch flag for table elimination. Tentative + name: 'table_elimination'. + +* With EXPLAIN, there are two options: + - Show removed tables in a way similar to const tables, with some + indication that they are removed. + - Do not show them altogether. +(the second one seems to be better? We're targeting a situation with VIEWs, +where the user would not care about what tables were added into his query +and then discarded from it?) + +5. Todo, issues to resolve +-------------------------- + +5.1 To resolve +~~~~~~~~~~~~~~ +- See EXPLAIN question in section #4. + +- Re-check how this works with equality propagation. + +- Relationship with prepared statements. + On one hand, it's natural to desire to make table elimination a + once-per-statement operation, like outer->inner join conversion. We'll have + to limit the applicablity by removing [MARK1] as that can change during + lifetime of the statement. + + The other option is to do table elimination every time. This will require to + rework operation [MARK2] to be undoable. + + I'm leaning towards doing the former. With anchor modeling, it is unlikely + that we'll meet outer joins which have N inner tables of which some are 1-row + MyISAM tables that do not have primary key. + +5.2 Resolved +~~~~~~~~~~~~ +- outer->inner join conversion is not a problem for table elimination. + We make outer->inner conversions based on predicates in WHERE. If the WHERE + referred to an inner table (requirement for OJ->IJ conversion) then table + elimination would not be applicable anyway. -=-=(Guest - Tue, 19 May 2009, 18:27)=-=- Title modified. --- /tmp/wklog.17.old.22023 2009-05-19 18:27:41.000000000 +0300 +++ /tmp/wklog.17.new.22023 2009-05-19 18:27:41.000000000 +0300 @@ -1 +1 @@ -Table ellimination +Table elimination -=-=(Monty - Sun, 10 May 2009, 19:58)=-=- High-Level Specification modified. --- /tmp/wklog.17.old.5423 2009-05-10 19:58:06.000000000 +0300 +++ /tmp/wklog.17.new.5423 2009-05-10 19:58:06.000000000 +0300 @@ -1 +1,90 @@ +Here is an extended explanation of table elimination. + +Table elimination is a feature found in some modern query optimizers, of +which Microsoft SQL Server 2005/2008 seems to have the most advanced +implementation. Oracle 11g has also been confirmed to use table +elimination but not to the same extent. + +Basically, what table elimination does, is to remove tables from the +execution plan when it is unneccessary to include them. This can, of +course, only happen if the right circumstances arise. Let us for example +look at the following query: + +select + A.colA +from + tableA A +left outer join + tableB B +on + B.id = A.id; + +When using A as the left table we ensure that the query will return at +least as many rows as there are in that table. For rows where the join +condition (B.id = A.id) is not met the selected column (A.colA) will +contain a NULL value. + +However, the result set could actually contain more rows than what is +found in tableA if there are duplicates of the column B.id in tableB. If +A +contains a row [1, "val1"] and B the rows [1, "other1a"],[1, "other1b"] +then two rows will match in the join condition. The only way to know +what +the result will look like is to actually touch both tables during +execution. + +Instead, let's say that tableB contains rows that make it possible to +place a unique constraint on the column B.id, for example and often the +case a primary key. In this situation we know that we will get exactly +as +many rows as there are in tableA, since joining with tableB cannot +introduce any duplicates. If further, as in the example query, we do not +select any columns from tableB, touching that table during execution is +unneccessary. We can remove the whole join operation from the execution +plan. + +Both SQL Server 2005/2008 and Oracle 11g will deploy table elimination +in +the case described above. Let us look at a more advanced query, where +Oracle fails. + +select + A.colA +from + tableA A +left outer join + tableB B +on + B.id = A.id +and + B.fromDate = ( + select + max(sub.fromDate) + from + tableB sub + where + sub.id = A.id + ); + +In this example we have added another join condition, which ensures +that we only pick the matching row from tableB having the latest +fromDate. In this case tableB will contain duplicates of the column +B.id, so in order to ensure uniqueness the primary key has to contain +the fromDate column as well. In other words the primary key of tableB +is (B.id, B.fromDate). + +Furthermore, since the subselect ensures that we only pick the latest +B.fromDate for a given B.id we know that at most one row will match +the join condition. We will again have the situation where joining +with tableB cannot affect the number of rows in the result set. Since +we do not select any columns from tableB, the whole join operation can +be eliminated from the execution plan. + +SQL Server 2005/2008 will deploy table elimination in this situation as +well. We have not found a way to make Oracle 11g use it for this type of +query. Queries like these arise in two situations. Either when you have +denormalized model consisting of a fact table with several related +dimension tables, or when you have a highly normalized model where each +attribute is stored in its own table. The example with the subselect is +common whenever you store historized/versioned data. DESCRIPTION: Eliminate not needed tables from SELECT queries.. This will speed up some views and automatically generated queries. Example: CREATE TABLE B (id int primary key); select A.colA from tableA A left outer join tableB B on B.id = A.id; In this case we can remove table B and the join from the query. HIGH-LEVEL SPECIFICATION: Here is an extended explanation of table elimination. Table elimination is a feature found in some modern query optimizers, of which Microsoft SQL Server 2005/2008 seems to have the most advanced implementation. Oracle 11g has also been confirmed to use table elimination but not to the same extent. Basically, what table elimination does, is to remove tables from the execution plan when it is unneccessary to include them. This can, of course, only happen if the right circumstances arise. Let us for example look at the following query: select A.colA from tableA A left outer join tableB B on B.id = A.id; When using A as the left table we ensure that the query will return at least as many rows as there are in that table. For rows where the join condition (B.id = A.id) is not met the selected column (A.colA) will contain a NULL value. However, the result set could actually contain more rows than what is found in tableA if there are duplicates of the column B.id in tableB. If A contains a row [1, "val1"] and B the rows [1, "other1a"],[1, "other1b"] then two rows will match in the join condition. The only way to know what the result will look like is to actually touch both tables during execution. Instead, let's say that tableB contains rows that make it possible to place a unique constraint on the column B.id, for example and often the case a primary key. In this situation we know that we will get exactly as many rows as there are in tableA, since joining with tableB cannot introduce any duplicates. If further, as in the example query, we do not select any columns from tableB, touching that table during execution is unneccessary. We can remove the whole join operation from the execution plan. Both SQL Server 2005/2008 and Oracle 11g will deploy table elimination in the case described above. Let us look at a more advanced query, where Oracle fails. select A.colA from tableA A left outer join tableB B on B.id = A.id and B.fromDate = ( select max(sub.fromDate) from tableB sub where sub.id = A.id ); In this example we have added another join condition, which ensures that we only pick the matching row from tableB having the latest fromDate. In this case tableB will contain duplicates of the column B.id, so in order to ensure uniqueness the primary key has to contain the fromDate column as well. In other words the primary key of tableB is (B.id, B.fromDate). Furthermore, since the subselect ensures that we only pick the latest B.fromDate for a given B.id we know that at most one row will match the join condition. We will again have the situation where joining with tableB cannot affect the number of rows in the result set. Since we do not select any columns from tableB, the whole join operation can be eliminated from the execution plan. SQL Server 2005/2008 will deploy table elimination in this situation as well. We have not found a way to make Oracle 11g use it for this type of query. Queries like these arise in two situations. Either when you have denormalized model consisting of a fact table with several related dimension tables, or when you have a highly normalized model where each attribute is stored in its own table. The example with the subselect is common whenever you store historized/versioned data. LOW-LEVEL DESIGN: <contents> 1. Conditions for removal 2. Removal operation properties 3. Removal operation 4. User interface 5. Todo, issues to resolve 5.1 To resolve 5.2 Resolved </contents> It's not really about elimination of tables, it's about elimination of inner sides of outer joins. 1. Conditions for removal ------------------------- We can eliminate an inner side of outer join if: 1. For each record combination of outer tables, it will always produce exactly one record. 2. There are no references to columns of the inner tables anywhere else in the query. #1 means that every table inside the outer join nest is: - is a constant table: = because it can be accessed via eq_ref(const) access, or = it is a zero-rows or one-row MyISAM-like table [MARK1] - has an eq_ref access method candidate. #2 means that WHERE clause, ON clauses of embedding outer joins, ORDER BY, GROUP BY and HAVING do not refer to the inner tables of the outer join nest. 2. Removal operation properties ------------------------------- * There is always one way to remove (no choice to remove either this or that) * It is always better to remove as much tables as possible (at least within our cost model). Thus, no need for any cost calculations/etc. It's an unconditional rewrite. 3. Removal operation -------------------- * Remove the outer join nest's nested join structure (i.e. get the outer join's TABLE_LIST object $OJ and remove it from $OJ->embedding, $OJ->embedding->nested_join. Update table_map's of all ancestor nested joins). [MARK2] * Move the tables and their JOIN_TABs to front like it is done with const tables, with exception that if eliminated outer join nest was within another outer join nest, that shouldn't prevent us from moving away the eliminated tables. * Update join->table_count and all-join-tables bitmap. * That's it. Nothing else? 4. User interface ----------------- * We'll add an @@optimizer switch flag for table elimination. Tentative name: 'table_elimination'. * With EXPLAIN, there are two options: - Show removed tables in a way similar to const tables, with some indication that they are removed. - Do not show them altogether. (the second one seems to be better? We're targeting a situation with VIEWs, where the user would not care about what tables were added into his query and then discarded from it?) 5. Todo, issues to resolve -------------------------- 5.1 To resolve ~~~~~~~~~~~~~~ - See EXPLAIN question in section #4. - Re-check how this works with equality propagation. - Relationship with prepared statements. On one hand, it's natural to desire to make table elimination a once-per-statement operation, like outer->inner join conversion. We'll have to limit the applicability by removing [MARK1] as that can change during lifetime of the statement. The other option is to do table elimination every time. This will require to rework operation [MARK2] to be undoable. I'm leaning towards doing the former. With anchor modeling, it is unlikely that we'll meet outer joins which have N inner tables of which some are 1-row MyISAM tables that do not have primary key. 5.2 Resolved ~~~~~~~~~~~~ - outer->inner join conversion is not a problem for table elimination. We make outer->inner conversions based on predicates in WHERE. If the WHERE referred to an inner table (requirement for OJ->IJ conversion) then table elimination would not be applicable anyway. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] bzr commit into MariaDB 5.1, with Maria 1.5:maria branch (knielsen:2704)
by knielsen＠knielsen-hq.org 22 May '09

22 May '09

#At lp:maria 2704 knielsen(a)knielsen-hq.org 2009-05-22 After-merge fixes for problems seen in buildbot after merging MySQL-5.1.35. - Version number. - Valgrind false alarms in libz. - New variant of suppression for Valgrind warning in dlclose(). - Fix double free() in plugin init error case. modified: configure.in include/my_sys.h mysql-test/valgrind.supp mysys/my_compress.c sql/handler.cc sql/item_strfunc.cc storage/archive/azio.c per-file messages: configure.in Fix version number. We should reset the maria variant back to `1' when the MySQL version number increases. include/my_sys.h Fix false alarms in Valgrind for zlib. Apply same fix as for archive storage handler also to the cases of compression in the client protocol, and to the compression SQL function. mysql-test/valgrind.supp A new variant of the dlclose() suppression is needed now. mysys/my_compress.c Fix false alarms in Valgrind for zlib. Apply same fix as for archive storage handler also to the cases of compression in the client protocol, and to the compression SQL function. sql/handler.cc Fix a double free() in error case for plugin initialisation. sql/item_strfunc.cc Fix false alarms in Valgrind for zlib. Apply same fix as for archive storage handler also to the cases of compression in the client protocol, and to the compression SQL function. === modified file 'configure.in' --- a/configure.in 2009-04-25 10:05:32 +0000 +++ b/configure.in 2009-05-22 12:38:50 +0000 @@ -9,8 +9,11 @@ AC_CANONICAL_SYSTEM # remember to also update version.c in ndb # # When changing major version number please also check switch statement -# in mysqlbinlog::check_master_version(). -AM_INIT_AUTOMAKE(mysql, 5.1.35-maria-beta2) +# in mysqlbinlog.cc / check_master_version(). +# +# When merging new MySQL releases, update the version number to match the +# MySQL version number, but reset the maria subrelease (-beta1). +AM_INIT_AUTOMAKE(mysql, 5.1.35-maria-beta1) AM_CONFIG_HEADER([include/config.h:config.h.in]) PROTOCOL_VERSION=10 === modified file 'include/my_sys.h' --- a/include/my_sys.h 2009-04-25 10:05:32 +0000 +++ b/include/my_sys.h 2009-05-22 12:38:50 +0000 @@ -879,6 +879,10 @@ extern my_bool my_compress(uchar *, size extern my_bool my_uncompress(uchar *, size_t , size_t *); extern uchar *my_compress_alloc(const uchar *packet, size_t *len, size_t *complen); +extern void *my_az_allocator(void *dummy, unsigned int items, unsigned int size); +extern void my_az_free(void *dummy, void *address); +extern int my_compress_buffer(uchar *dest, size_t *destLen, + const uchar *source, size_t sourceLen); extern int packfrm(uchar *, size_t, uchar **, size_t *); extern int unpackfrm(uchar **, size_t *, const uchar *); === modified file 'mysql-test/valgrind.supp' --- a/mysql-test/valgrind.supp 2009-04-08 16:55:26 +0000 +++ b/mysql-test/valgrind.supp 2009-05-22 12:38:50 +0000 @@ -401,6 +401,20 @@ } { + dlclose memory loss from plugin variant 3 + Memcheck:Leak + fun:malloc + fun:_dl_scope_free + fun:_dl_close_worker + fun:_dl_close + fun:_dl_catch_error + fun:_dlerror_run + fun:dlclose + fun:_Z15free_plugin_memP12st_plugin_dl + fun:_Z13plugin_dl_delPK19st_mysql_lex_string +} + +{ dlopen / ptread_cancel_init memory loss on Suse Linux 10.3 32/64 bit Memcheck:Leak fun:*alloc === modified file 'mysys/my_compress.c' --- a/mysys/my_compress.c 2009-04-25 10:05:32 +0000 +++ b/mysys/my_compress.c 2009-05-22 12:38:50 +0000 @@ -57,19 +57,85 @@ my_bool my_compress(uchar *packet, size_ } +/* + Valgrind normally gives false alarms for zlib operations, in the form of + "conditional jump depends on uninitialised values" etc. The reason is + explained in the zlib FAQ (http://www.zlib.net/zlib_faq.html#faq36) + + "That is intentional for performance reasons, and the output of deflate + is not affected." + + Also discussed on a blog + (http://www.sirena.org.uk/log/2006/02/19/zlib-generating-valgrind-warnings/) + + "...loop unrolling in the zlib library causes the mentioned + “Conditional jump or move depends on uninitialised value(s)” + warnings. These are safe since the results of the comparison are + subsequently ignored..." + + "the results of the calculations are discarded by bounds checking done + after the loop exits" + + Fix by initializing the memory allocated by zlib when running under Valgrind. + + This fix is safe, since such memory is only used internally by zlib, so we + will not hide any bugs in mysql this way. +*/ +void *my_az_allocator(void *dummy, unsigned int items, unsigned int size) +{ + return my_malloc((size_t)items*(size_t)size, IF_VALGRIND(MY_ZEROFILL, MYF(0))); +} + +void my_az_free(void *dummy, void *address) +{ + my_free(address, MYF(MY_ALLOW_ZERO_PTR)); +} + +/* + This works like zlib compress(), but using custom memory allocators to work + better with my_malloc leak detection and Valgrind. +*/ +int my_compress_buffer(uchar *dest, size_t *destLen, + const uchar *source, size_t sourceLen) +{ + z_stream stream; + int err; + + stream.next_in = (Bytef*)source; + stream.avail_in = (uInt)sourceLen; + stream.next_out = (Bytef*)dest; + stream.avail_out = (uInt)*destLen; + if ((size_t)stream.avail_out != *destLen) + return Z_BUF_ERROR; + + stream.zalloc = (alloc_func)my_az_allocator; + stream.zfree = (free_func)my_az_free; + stream.opaque = (voidpf)0; + + err = deflateInit(&stream, Z_DEFAULT_COMPRESSION); + if (err != Z_OK) return err; + + err = deflate(&stream, Z_FINISH); + if (err != Z_STREAM_END) { + deflateEnd(&stream); + return err == Z_OK ? Z_BUF_ERROR : err; + } + *destLen = stream.total_out; + + err = deflateEnd(&stream); + return err; +} + uchar *my_compress_alloc(const uchar *packet, size_t *len, size_t *complen) { uchar *compbuf; - uLongf tmp_complen; int res; *complen= *len * 120 / 100 + 12; if (!(compbuf= (uchar *) my_malloc(*complen, MYF(MY_WME)))) return 0; /* Not enough memory */ - tmp_complen= (uLongf) *complen; - res= compress((Bytef*) compbuf, &tmp_complen, (Bytef*) packet, (uLong) *len); - *complen= tmp_complen; + res= my_compress_buffer(compbuf, complen, packet, *len); if (res != Z_OK) { === modified file 'sql/handler.cc' --- a/sql/handler.cc 2009-04-25 10:05:32 +0000 +++ b/sql/handler.cc 2009-05-22 12:38:50 +0000 @@ -432,9 +432,6 @@ int ha_initialize_handlerton(st_plugin_i { sql_print_error("Plugin '%s' init function returned error.", plugin->name.str); - /* Free data, so that we don't refer to it in ha_finalize_handlerton */ - my_free(hton, MYF(0)); - plugin->data= 0; goto err; } === modified file 'sql/item_strfunc.cc' --- a/sql/item_strfunc.cc 2009-04-25 10:05:32 +0000 +++ b/sql/item_strfunc.cc 2009-05-22 12:38:50 +0000 @@ -3243,7 +3243,7 @@ longlong Item_func_crc32::val_int() String *Item_func_compress::val_str(String *str) { int err= Z_OK, code; - ulong new_size; + size_t new_size; String *res; Byte *body; char *tmp, *last_char; @@ -3279,8 +3279,8 @@ String *Item_func_compress::val_str(Stri body= ((Byte*)buffer.ptr()) + 4; // As far as we have checked res->is_empty() we can use ptr() - if ((err= compress(body, &new_size, - (const Bytef*)res->ptr(), res->length())) != Z_OK) + if ((err= my_compress_buffer(body, &new_size, (const uchar *)res->ptr(), + res->length())) != Z_OK) { code= err==Z_MEM_ERROR ? ER_ZLIB_Z_MEM_ERROR : ER_ZLIB_Z_BUF_ERROR; push_warning(current_thd,MYSQL_ERROR::WARN_LEVEL_ERROR,code,ER(code)); === modified file 'storage/archive/azio.c' --- a/storage/archive/azio.c 2009-05-19 09:28:05 +0000 +++ b/storage/archive/azio.c 2009-05-22 12:38:50 +0000 @@ -16,6 +16,8 @@ #include <stdio.h> #include <string.h> +#include "my_sys.h" + static int const gz_magic[2] = {0x1f, 0x8b}; /* gzip magic header */ static int const az_magic[3] = {0xfe, 0x03, 0x01}; /* az magic header */ @@ -37,40 +39,6 @@ void putLong(File file, uLong x); uLong getLong(azio_stream *s); void read_header(azio_stream *s, unsigned char *buffer); -/* - Valgrind normally gives false alarms for zlib operations, in the form of - "conditional jump depends on uninitialised values" etc. The reason is - explained in the zlib FAQ (http://www.zlib.net/zlib_faq.html#faq36) - - "That is intentional for performance reasons, and the output of deflate - is not affected." - - Also discussed on a blog - (http://www.sirena.org.uk/log/2006/02/19/zlib-generating-valgrind-warnings/) - - "...loop unrolling in the zlib library causes the mentioned - “Conditional jump or move depends on uninitialised value(s)” - warnings. These are safe since the results of the comparison are - subsequently ignored..." - - "the results of the calculations are discarded by bounds checking done - after the loop exits" - - Fix by initializing the memory allocated by zlib when running under Valgrind. - - This fix is safe, since such memory is only used internally by zlib, so we - will not hide any bugs in mysql this way. -*/ -static void *az_allocator(void *dummy, uInt items, uInt size) -{ - return my_malloc((size_t)items*(size_t)size, IF_VALGRIND(MY_ZEROFILL, MYF(0))); -} - -static void az_free(void *dummy, void *address) -{ - my_free(address, MYF(MY_ALLOW_ZERO_PTR)); -} - /* =========================================================================== Opens a gzip (.gz) file for reading or writing. The mode parameter is as in fopen ("rb" or "wb"). The file is given either by file descriptor @@ -86,8 +54,8 @@ int az_open (azio_stream *s, const char int level = Z_DEFAULT_COMPRESSION; /* compression level */ int strategy = Z_DEFAULT_STRATEGY; /* compression strategy */ - s->stream.zalloc = az_allocator; - s->stream.zfree = az_free; + s->stream.zalloc = my_az_allocator; + s->stream.zfree = my_az_free; s->stream.opaque = (voidpf)0; memset(s->inbuf, 0, AZ_BUFSIZE_READ); memset(s->outbuf, 0, AZ_BUFSIZE_WRITE);

1 0

[Maria-developers] Updated (by Knielsen): Using the Valgrind API in mysqld (23)
by worklog-noreply＠askmonty.org 22 May '09

22 May '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Using the Valgrind API in mysqld CREATION DATE..: Fri, 22 May 2009, 11:43 SUPERVISOR.....: Monty IMPLEMENTOR....: Knielsen COPIES TO......: CATEGORY.......: Server-RawIdeaBin TASK ID........: 23 (http://askmonty.org/worklog/?tid=23) VERSION........: Benchmarks-3.0 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 40 (hours remain) ORIG. ESTIMATE.: 40 PROGRESS NOTES: -=-=(Knielsen - Fri, 22 May 2009, 14:31)=-=- Low Level Design modified. --- /tmp/wklog.23.old.24587 2009-05-22 14:31:52.000000000 +0300 +++ /tmp/wklog.23.new.24587 2009-05-22 14:31:52.000000000 +0300 @@ -1 +1,7 @@ +Two places where we call into libz, and where checking for defined parameters +would be good: + + - mysys/my_compress.c + + - sql/item_strfunc.cc (Item_func_compress). -=-=(Guest - Fri, 22 May 2009, 12:04)=-=- High-Level Specification modified. --- /tmp/wklog.23.old.18061 2009-05-22 12:04:05.000000000 +0300 +++ /tmp/wklog.23.new.18061 2009-05-22 12:04:05.000000000 +0300 @@ -26,3 +26,5 @@ initialised, it is possible to detect problems earlier, speeding up debugging. Such code can be added in more places over time as development and debugging goes on. + +See also a patch here: http://bugs.mysql.com/bug.php?id=44582 -=-=(Knielsen - Fri, 22 May 2009, 11:52)=-=- High-Level Specification modified. --- /tmp/wklog.23.old.17628 2009-05-22 11:52:33.000000000 +0300 +++ /tmp/wklog.23.new.17628 2009-05-22 11:52:33.000000000 +0300 @@ -1 +1,28 @@ +With custom memory allocators, using the Valgrind APIs we can tell Valgrind when +a memory block is allocated (so that data read from memory is marked as undefined +instead of being defined or not at random depending on prior use); and when a +memory block is freed (so that use after freeing can be reported as an error). +In some cases cheking for leaks may also be appropriate. + +Another possibility is to add an explicit check for whether memory is defined. + +One place this would be useful is when calling libz. Due to the design of that +library, Valgrind produces lots of false alarms about using undefined values +(I think the issue is that it runs a few bytes off of initialized memory to +reduce boundary checks in each loop iteration, then after the loop has checks to +avoid using the undefined part of the result). This means we have lots of libz +Valgrind suppressions and continue to add more as new warnings surface. So we +might easily miss a real problem in this area. This could be improved by adding +explicit checks at the call to libz functions that the passed memory is properly +defined. + +Another use is to improve debugging. It is often the case when debugging a +warning about using un-initialised memory that the detection happens long after +the real problem, the un-initialized value being passed along through the code +for a long time before being detected. This makes debugging the problem slow. + +By adding in strategic places code that asserts that a specific value must be +initialised, it is possible to detect problems earlier, speeding up debugging. +Such code can be added in more places over time as development and debugging +goes on. DESCRIPTION: Valgrind (the memcheck tool) has some very useful APIs that can be used in mysqld when testing with Valgrind to improve testing and/or debugging: file:///usr/share/doc/valgrind/html/mc-manual.html#mc-manual.clientreqs file:///usr/share/doc/valgrind/html/mc-manual.html#mc-manual.mempools This worklog is about adding configure checks and headers to allow to use these in a way that continues to work on machines where the Valgrind headers or functionality is missing. It also includes adding some basic Valgrind enhancements: - Adding Valgrind annotations to custom memory allocators so that Valgrind can detect leaks, use-before-init, and use-after-free problems also for these allocators. - Adding checks for definedness in appropriate places (eg. when calling libz). HIGH-LEVEL SPECIFICATION: With custom memory allocators, using the Valgrind APIs we can tell Valgrind when a memory block is allocated (so that data read from memory is marked as undefined instead of being defined or not at random depending on prior use); and when a memory block is freed (so that use after freeing can be reported as an error). In some cases cheking for leaks may also be appropriate. Another possibility is to add an explicit check for whether memory is defined. One place this would be useful is when calling libz. Due to the design of that library, Valgrind produces lots of false alarms about using undefined values (I think the issue is that it runs a few bytes off of initialized memory to reduce boundary checks in each loop iteration, then after the loop has checks to avoid using the undefined part of the result). This means we have lots of libz Valgrind suppressions and continue to add more as new warnings surface. So we might easily miss a real problem in this area. This could be improved by adding explicit checks at the call to libz functions that the passed memory is properly defined. Another use is to improve debugging. It is often the case when debugging a warning about using un-initialised memory that the detection happens long after the real problem, the un-initialized value being passed along through the code for a long time before being detected. This makes debugging the problem slow. By adding in strategic places code that asserts that a specific value must be initialised, it is possible to detect problems earlier, speeding up debugging. Such code can be added in more places over time as development and debugging goes on. See also a patch here: http://bugs.mysql.com/bug.php?id=44582 LOW-LEVEL DESIGN: Two places where we call into libz, and where checking for defined parameters would be good: - mysys/my_compress.c - sql/item_strfunc.cc (Item_func_compress). ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] Updated (by Knielsen): Using the Valgrind API in mysqld (23)
by worklog-noreply＠askmonty.org 22 May '09

22 May '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Using the Valgrind API in mysqld CREATION DATE..: Fri, 22 May 2009, 11:43 SUPERVISOR.....: Monty IMPLEMENTOR....: Knielsen COPIES TO......: CATEGORY.......: Server-RawIdeaBin TASK ID........: 23 (http://askmonty.org/worklog/?tid=23) VERSION........: Benchmarks-3.0 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 40 (hours remain) ORIG. ESTIMATE.: 40 PROGRESS NOTES: -=-=(Knielsen - Fri, 22 May 2009, 14:31)=-=- Low Level Design modified. --- /tmp/wklog.23.old.24587 2009-05-22 14:31:52.000000000 +0300 +++ /tmp/wklog.23.new.24587 2009-05-22 14:31:52.000000000 +0300 @@ -1 +1,7 @@ +Two places where we call into libz, and where checking for defined parameters +would be good: + + - mysys/my_compress.c + + - sql/item_strfunc.cc (Item_func_compress). -=-=(Guest - Fri, 22 May 2009, 12:04)=-=- High-Level Specification modified. --- /tmp/wklog.23.old.18061 2009-05-22 12:04:05.000000000 +0300 +++ /tmp/wklog.23.new.18061 2009-05-22 12:04:05.000000000 +0300 @@ -26,3 +26,5 @@ initialised, it is possible to detect problems earlier, speeding up debugging. Such code can be added in more places over time as development and debugging goes on. + +See also a patch here: http://bugs.mysql.com/bug.php?id=44582 -=-=(Knielsen - Fri, 22 May 2009, 11:52)=-=- High-Level Specification modified. --- /tmp/wklog.23.old.17628 2009-05-22 11:52:33.000000000 +0300 +++ /tmp/wklog.23.new.17628 2009-05-22 11:52:33.000000000 +0300 @@ -1 +1,28 @@ +With custom memory allocators, using the Valgrind APIs we can tell Valgrind when +a memory block is allocated (so that data read from memory is marked as undefined +instead of being defined or not at random depending on prior use); and when a +memory block is freed (so that use after freeing can be reported as an error). +In some cases cheking for leaks may also be appropriate. + +Another possibility is to add an explicit check for whether memory is defined. + +One place this would be useful is when calling libz. Due to the design of that +library, Valgrind produces lots of false alarms about using undefined values +(I think the issue is that it runs a few bytes off of initialized memory to +reduce boundary checks in each loop iteration, then after the loop has checks to +avoid using the undefined part of the result). This means we have lots of libz +Valgrind suppressions and continue to add more as new warnings surface. So we +might easily miss a real problem in this area. This could be improved by adding +explicit checks at the call to libz functions that the passed memory is properly +defined. + +Another use is to improve debugging. It is often the case when debugging a +warning about using un-initialised memory that the detection happens long after +the real problem, the un-initialized value being passed along through the code +for a long time before being detected. This makes debugging the problem slow. + +By adding in strategic places code that asserts that a specific value must be +initialised, it is possible to detect problems earlier, speeding up debugging. +Such code can be added in more places over time as development and debugging +goes on. DESCRIPTION: Valgrind (the memcheck tool) has some very useful APIs that can be used in mysqld when testing with Valgrind to improve testing and/or debugging: file:///usr/share/doc/valgrind/html/mc-manual.html#mc-manual.clientreqs file:///usr/share/doc/valgrind/html/mc-manual.html#mc-manual.mempools This worklog is about adding configure checks and headers to allow to use these in a way that continues to work on machines where the Valgrind headers or functionality is missing. It also includes adding some basic Valgrind enhancements: - Adding Valgrind annotations to custom memory allocators so that Valgrind can detect leaks, use-before-init, and use-after-free problems also for these allocators. - Adding checks for definedness in appropriate places (eg. when calling libz). HIGH-LEVEL SPECIFICATION: With custom memory allocators, using the Valgrind APIs we can tell Valgrind when a memory block is allocated (so that data read from memory is marked as undefined instead of being defined or not at random depending on prior use); and when a memory block is freed (so that use after freeing can be reported as an error). In some cases cheking for leaks may also be appropriate. Another possibility is to add an explicit check for whether memory is defined. One place this would be useful is when calling libz. Due to the design of that library, Valgrind produces lots of false alarms about using undefined values (I think the issue is that it runs a few bytes off of initialized memory to reduce boundary checks in each loop iteration, then after the loop has checks to avoid using the undefined part of the result). This means we have lots of libz Valgrind suppressions and continue to add more as new warnings surface. So we might easily miss a real problem in this area. This could be improved by adding explicit checks at the call to libz functions that the passed memory is properly defined. Another use is to improve debugging. It is often the case when debugging a warning about using un-initialised memory that the detection happens long after the real problem, the un-initialized value being passed along through the code for a long time before being detected. This makes debugging the problem slow. By adding in strategic places code that asserts that a specific value must be initialised, it is possible to detect problems earlier, speeding up debugging. Such code can be added in more places over time as development and debugging goes on. See also a patch here: http://bugs.mysql.com/bug.php?id=44582 LOW-LEVEL DESIGN: Two places where we call into libz, and where checking for defined parameters would be good: - mysys/my_compress.c - sql/item_strfunc.cc (Item_func_compress). ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] Updated (by Guest): Using the Valgrind API in mysqld (23)
by worklog-noreply＠askmonty.org 22 May '09

22 May '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Using the Valgrind API in mysqld CREATION DATE..: Fri, 22 May 2009, 11:43 SUPERVISOR.....: Monty IMPLEMENTOR....: Knielsen COPIES TO......: CATEGORY.......: Server-RawIdeaBin TASK ID........: 23 (http://askmonty.org/worklog/?tid=23) VERSION........: Benchmarks-3.0 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 40 (hours remain) ORIG. ESTIMATE.: 40 PROGRESS NOTES: -=-=(Guest - Fri, 22 May 2009, 12:04)=-=- High-Level Specification modified. --- /tmp/wklog.23.old.18061 2009-05-22 12:04:05.000000000 +0300 +++ /tmp/wklog.23.new.18061 2009-05-22 12:04:05.000000000 +0300 @@ -26,3 +26,5 @@ initialised, it is possible to detect problems earlier, speeding up debugging. Such code can be added in more places over time as development and debugging goes on. + +See also a patch here: http://bugs.mysql.com/bug.php?id=44582 -=-=(Knielsen - Fri, 22 May 2009, 11:52)=-=- High-Level Specification modified. --- /tmp/wklog.23.old.17628 2009-05-22 11:52:33.000000000 +0300 +++ /tmp/wklog.23.new.17628 2009-05-22 11:52:33.000000000 +0300 @@ -1 +1,28 @@ +With custom memory allocators, using the Valgrind APIs we can tell Valgrind when +a memory block is allocated (so that data read from memory is marked as undefined +instead of being defined or not at random depending on prior use); and when a +memory block is freed (so that use after freeing can be reported as an error). +In some cases cheking for leaks may also be appropriate. + +Another possibility is to add an explicit check for whether memory is defined. + +One place this would be useful is when calling libz. Due to the design of that +library, Valgrind produces lots of false alarms about using undefined values +(I think the issue is that it runs a few bytes off of initialized memory to +reduce boundary checks in each loop iteration, then after the loop has checks to +avoid using the undefined part of the result). This means we have lots of libz +Valgrind suppressions and continue to add more as new warnings surface. So we +might easily miss a real problem in this area. This could be improved by adding +explicit checks at the call to libz functions that the passed memory is properly +defined. + +Another use is to improve debugging. It is often the case when debugging a +warning about using un-initialised memory that the detection happens long after +the real problem, the un-initialized value being passed along through the code +for a long time before being detected. This makes debugging the problem slow. + +By adding in strategic places code that asserts that a specific value must be +initialised, it is possible to detect problems earlier, speeding up debugging. +Such code can be added in more places over time as development and debugging +goes on. DESCRIPTION: Valgrind (the memcheck tool) has some very useful APIs that can be used in mysqld when testing with Valgrind to improve testing and/or debugging: file:///usr/share/doc/valgrind/html/mc-manual.html#mc-manual.clientreqs file:///usr/share/doc/valgrind/html/mc-manual.html#mc-manual.mempools This worklog is about adding configure checks and headers to allow to use these in a way that continues to work on machines where the Valgrind headers or functionality is missing. It also includes adding some basic Valgrind enhancements: - Adding Valgrind annotations to custom memory allocators so that Valgrind can detect leaks, use-before-init, and use-after-free problems also for these allocators. - Adding checks for definedness in appropriate places (eg. when calling libz). HIGH-LEVEL SPECIFICATION: With custom memory allocators, using the Valgrind APIs we can tell Valgrind when a memory block is allocated (so that data read from memory is marked as undefined instead of being defined or not at random depending on prior use); and when a memory block is freed (so that use after freeing can be reported as an error). In some cases cheking for leaks may also be appropriate. Another possibility is to add an explicit check for whether memory is defined. One place this would be useful is when calling libz. Due to the design of that library, Valgrind produces lots of false alarms about using undefined values (I think the issue is that it runs a few bytes off of initialized memory to reduce boundary checks in each loop iteration, then after the loop has checks to avoid using the undefined part of the result). This means we have lots of libz Valgrind suppressions and continue to add more as new warnings surface. So we might easily miss a real problem in this area. This could be improved by adding explicit checks at the call to libz functions that the passed memory is properly defined. Another use is to improve debugging. It is often the case when debugging a warning about using un-initialised memory that the detection happens long after the real problem, the un-initialized value being passed along through the code for a long time before being detected. This makes debugging the problem slow. By adding in strategic places code that asserts that a specific value must be initialised, it is possible to detect problems earlier, speeding up debugging. Such code can be added in more places over time as development and debugging goes on. See also a patch here: http://bugs.mysql.com/bug.php?id=44582 ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0