The cause of regression was handling for ROWNUM() function.
For queries like
SELECT ROWNUM() FROM ... ORDER BY ...
ROWNUM() should be computed before the ORDER BY.
The computation was moved to be before the ORDER BY for any entries in
the select list that had RAND_TABLE_BIT set.
This had a negative impact on queries in form:
SELECT sp_func() FROM t1 ORDER BY ... LIMIT n
where sp_func() is NOT declared as DETERMINISTIC (and so has
RAND_TABLE_BIT set).
The fix is to require evaluation for sorting only for the ROWNUM()
function. Functions that just have RAND_TABLE_BIT() can be computed
after ORDER BY ... LIMIT is applied.
(think about a possible index that satisfies the ORDER BY clause. In
that case, the the rows would be read in the needed order and we would
stop after reading LIMIT rows, achieving the same effect).
The issue is that record_should_be_deleted() returns true in
mysql_delete() even if sub-select with join gets error from storage
engine when DELETE FROM ... WHERE ... IN (SELECT ...) statement is
executed.
The same is true for mysql_update() where select->skip_record() returns
true even if sub-select with join gets error from storage engine.
In the test case if sub-select is chosen as deadlock victim the whole
transaction is rolled back during sub-select execution, but
mysql_delete()/mysql_update() continues transaction execution and invokes
table->delete_row() as record_should_be_deleted() wrongly returns true
in mysql_delete() and table->update_row() as select->skip_record(thd)
wrongly returns 1 for mysql_update().
record_should_be_deleted() wrogly returns true because thd->is_error()
returns false SQL_SELECT::skip_record() invoked from
record_should_be_deleted().
It's supposed that THD error should be set in rr_handle_error() called
from rr_sequential() during sub-select JOIN::exec_inner() execution.
But rr_handle_error() does not set THD error because
READ_RECORD::print_error is not set in JOIN_TAB::read_record.
READ_RECORD::print_error should be initialized in
init_read_record()/init_read_record_idx(). But make_join_readinfo() does
not invoke init_read_record()/init_read_record_idx() for
JOIN_TAB::read_record.
The fix is to set JOIN_TAB::read_record.print_error in
make_join_readinfo(), i.e. in the same place where
JOIN_TAB::read_record.table is set.
Reviewed by Sergey Petrunya.
Deallocation of TABLE_LIST::dt_handler and TABLE_LIST::pushdown_derived
was performed in multiple places if code. This not only made the code
more difficult to maintain but also led to memory leaks and
ASAN heap-use-after-free errors.
This commit puts deallocation of TABLE_LIST::dt_handler and
TABLE_LIST::pushdown_derived to the single point - JOIN::cleanup()
it's incorrect to use change_item_tree() to replace arguments
of top-level AND/OR, because they (arguments) are stored in a List,
so a pointer to an argument is in the list_node, and individual
list_node's of top-level AND/OR can be deleted in Item_cond::build_equal_items().
In that case rollback_item_tree_changes() will modify the deleted object.
Luckily, it's not needed to use change_item_tree() for top-level
AND/OR, because the whole top-level item is copied and preserved
in prep_where and prep_on, and restored from there.
So, just don't.
If all elements in the list of 'IN' or 'NOT IN' clause are equal
and there are no NULLs then clause
- "a IN (e1,..,en)" can be converted to "a = e1"
- "a NOT IN (e1,..,en)" can be converted to "a <> e1".
This means an object of Item_func_in can be replaced with an object
of Item_func_eq for IN (e1,..,en) clause and Item_func_ne for
NOT IN (e1,...,en). Such a replacement allows the optimizer to choose
a better execution plan
When a range rowid filter was used with an index ref access the cost of
accessing the index entries for the records rejected by the filter was not
taken into account. For a ref access by an index with big average number
of records per key this led to poor execution plans if selectivity of the
used filter was high.
The patch resolves this problem. It also introduces a minor optimization
that skips look-ups into a filter that turns out to be empty.
With this patch the output of ANALYZE stmt reports the number of look-ups
into used rowid filters.
The patch also back-ports from 10.5 the code that properly sets the field
TABLE::file::table for opened temporary tables.
The test cases that were supposed to use rowid filters have been adjusted
in order to use similar execution plans after this fix.
Approved by Oleksandr Byelkin <sanja@mariadb.com>
Part #2: make sure we allocate space for two JOIN_TABs that
use temporary tables.
The dbug_join_tab_array_size is still set to catch cases where
we try to access more JOIN_TAB object than we thought we would have.
The problem was caused by use of COLLATION(AVG('x')). This is an
item whose value is a constant.
Name Resolution code called convert_const_to_int() which removed AVG('x').
However, the item representing COLLATION(...) still had with_sum_func=1.
This inconsistent state confused the code that handles grouping and
DISTINCT: JOIN::get_best_combination() decided to use one temporary
table and allocated one JOIN_TAB for it, but then
JOIN::make_aggr_tables_info() attempted to use two and made writes
beyond the end of the JOIN::join_tab array.
The fix:
- Do not replace constant expressions which contain aggregate functions.
- Add JOIN::dbug_join_tab_array_size to catch attempts to use more
JOIN_TAB objects than we've allocated.
This bug could cause a crash of the server when executing queries containing
ANY/ALL predicands with redundant subqueries in GROUP BY clauses.
These subqueries are eliminated by remove_redundant_subquery_clause()
together with elimination of GROUP BY list containing these subqueries.
However the references to the elements of the GROUP BY remained in the
JOIN::all_fields list of the right operand of of the ALL/ANY predicand.
Later these references confused make_aggr_tables_info() when forming
proper execution structures after ALL/ANY predicands had been replaced
with expressions containing MIN/MAX set functions.
The patch just removes these references from JOIN::all_fields list used
by the subquery of the ALL/ANY predicand when its GROUP BY clause is
eliminated.
Approved by Oleksandr Byelkin <sanja@mariadb.com>
1. For INSERT..SELECT statements: don't include table/view the data
is inserted into in the list of leaf tables
2. Remove duplicated and dead code related to table_count
optimize_semi_joins() calls update_sj_state() to update semi-join
optimization state in the JOIN class.
greedy_search() algorithm considers different join prefixes,
and then picks one table to put into the join prefix.
Most of the semi-join optimization state is in the table's entry
in the join->positions[cur_prefix_size].
However, it also needs to call update_sj_state() to update the
semi-join optimization state in the JOIN class.
There is one exception, which is the cause of this bug: when we're
inside optimize_semi_join_nests() and are optimizing a subquery,
optimize_semi_joins() does nothing, it doesn't call update_sj_state().
greedy_search() must not do that either.
The bug was that build_notnull_conds_for_range_scans() did not take into
account the join_tab is not yet sorted with constant tables first.
Fixed the bug by testing explicitely if a table is a const table.
Part of:
MDEV-28073 Slow query performance in MariaDB when using many tables
s->key_dependent has a list of tables that are compared with key fields
in the current table. However it does not take into account if a key
field could be resolved by another table.
This is because MariaDB expands 'join_tab->keyuse' to include all generated
comparisons.
For example:
SELECT * from t1,t2,t3 where t1.key=t2.key and t2.key=t3.key
In this case keyuse for t1 includes t2.key and t3.key and key_dependent
contains 't2.map | t3.map'
If we in best_extension_by_limited_search() consider t2,t1 then t1's
key is fully defined, but we cannot do any prune of plans as
s->key_dependent indicates that t3 is still needed.
Fixed by calculating in best_access_patch the current key_dependent map
of tables that is needed to satisfy all keys. This allows us to prune
more bad plans earlier as soon as all keys can be used.
We also set key_dependent to 0 if we found an EQ_REF key, as this an
optimal key for the table and there is no reason to check more keys.
best_extension_by_limited_search() assumes that tables should be sorted
according to size to be able to quickly disregard bad plans. However the
current usage of swap_variables() will change the table order to a not
sorted one for the next recursive call. This breaks the assumtion and
causes performance issues when using many tables (we have to examine
many more plans).
This patch fixes this by ensuring that the original table order is kept
for the not yet used tables when best_extension_by_limited_search() is
called.
This was done by always calling swap_variables() for each table and
restoring the original table order at exit.
Some test changed:
- In a majority of the test the change was that two "identical tables"
where swapped and the optimzer is now using the first/smaller table
- In few test the table order was changed. The new plan looks identical
or slighly better than the original.
(Try 2)
The code that updates semi-join optimization state for a join order prefix
had several bugs. The visible effect was bad optimization for FirstMatch or
LooseScan strategies: they either weren't considered when they should have
been, or considered when they shouldn't have been.
In order to hit the bug, the optimizer needs to consider several different
join prefixes in a certain order. Queries with "obvious" query plans which
prune all join orders except one are not affected.
Internally, the bugs in updates of semi-join state were:
1. restore_prev_sj_state() assumed that
"we assume remaining_tables doesnt contain @tab"
which wasn't true.
2. Another bug in this function: it did remove bits from
join->cur_sj_inner_tables but never added them.
3. greedy_search() adds tables into the join prefix but neglects to update
the semi-join optimization state. (It does update nested outer join
state, see this call:
check_interleaving_with_nj(best_table)
but there's no matching call to update the semi-join state.
(This wasn't visible because most of the state is in the POSITION
structure which is updated. But there is also state in JOIN, too)
The patch:
- Fixes all of the above
- Adds JOIN::dbug_verify_sj_inner_tables() which is used to verify the
state is correct at every step.
- Renames advance_sj_state() to optimize_semi_joins().
= Introduces update_sj_state() which ideally should have been called
"advance_sj_state" but I didn't reuse the name to not create confusion.
Main fix was replacing read_time+= with read_time
I also did updated the 'identical' code in optimize_straight_join) and
best_extension_by_limited_search() to make them eaiser to compare.
Reviewer: Sergei Petrunia <sergey@mariadb.com>
(Try 2) (Cherry-pick back into 10.3)
The code that updates semi-join optimization state for a join order prefix
had several bugs. The visible effect was bad optimization for FirstMatch or
LooseScan strategies: they either weren't considered when they should have
been, or considered when they shouldn't have been.
In order to hit the bug, the optimizer needs to consider several different
join prefixes in a certain order. Queries with "obvious" query plans which
prune all join orders except one are not affected.
Internally, the bugs in updates of semi-join state were:
1. restore_prev_sj_state() assumed that
"we assume remaining_tables doesnt contain @tab"
which wasn't true.
2. Another bug in this function: it did remove bits from
join->cur_sj_inner_tables but never added them.
3. greedy_search() adds tables into the join prefix but neglects to update
the semi-join optimization state. (It does update nested outer join
state, see this call:
check_interleaving_with_nj(best_table)
but there's no matching call to update the semi-join state.
(This wasn't visible because most of the state is in the POSITION
structure which is updated. But there is also state in JOIN, too)
The patch:
- Fixes all of the above
- Adds JOIN::dbug_verify_sj_inner_tables() which is used to verify the
state is correct at every step.
- Renames advance_sj_state() to optimize_semi_joins().
= Introduces update_sj_state() which ideally should have been called
"advance_sj_state" but I didn't reuse the name to not create confusion.
The warning comes from copying POSITION objects where 'type' is not
initialized. This does not affect any production code as when 'type'
was compared/used, it was always initialized.
Removed by initializing the type variable in the constructor
- In best_extension_by_limited_search(), do not check for
"(remaining_tables & real_table_bit)", it is guaranteed to be true.
Make it an assert.
- In (!idx || check_interleaving_with_nj())", remove the !idx part.
This check made sense only in the original version of this function.
- "micro optimization" in check_interleaving_with_nj().
The issue was that best_extension_by_limited_search() had to go through
too many plans with the same cost as there where many EQ_REF tables.
Fixed by shortcutting EQ_REF (AND REF) when the result only contains one
row. This got the optimization time down from hours to sub seconds.
The only known downside with this patch is that in some cases a table
with ref and 1 record may be used before on EQ_REF table. The faster
optimzation phase should compensate for this.
Fixed failing main.default on Windows
(to trigger an assert the test needed a debug build without
safemalloc, as 0xa5 happened to have the important bit set "correctly")
(This is the assert that was added in fix for MDEV-26047)
Table elimination may remove an ON expression from an outer join.
However SELECT_LEX::update_used_tables() will still call
item->walk(&Item::eval_not_null_tables)
for eliminated expressions. If the subquery is constant and cheap
Item_cond_and will attempt to evaluate it, which will trigger an
assert.
The fix is not to call update_used_tables() or eval_not_null_tables()
for ON expressions that were eliminated.
