It is based on the sum() function, thus much of the logic is shared.
Considering that there are 2 counters stored within the function, one
that handles the null value, while the other that handles the divisor
for the avg computation, it is possible to remove the counter from the
Item_sum_avg. I have not removed it in this patch as we may choose to
refactor the whole code into a separate class.
This remains to be dicussed.
1. Item_windowfunc must check the nullness of the field if it returns the
value from the result_field.
2. The decimal value must come from the window_func instead of the
result_field when not fetching values from the result_field.
Added a more extensive test case that highlights if these bugs occur.
The 2 select statements check behaviour for decimal and real typed
fields respectively.
Query result had 0 where it should have had NULLs:
- Make Item_window_func::val* functions honor NULL-handling conventions:
1. set null_value to indicate whether we've returned a NULL value
2. val_str and val_decimal should return NULL pointer when they're
returning SQL NULL value.
Fix assertion failure when sending results to network.
- The assert was due to window func returing SQL NULL despite having
maybe_null=false
- Fixed by settting Item_window_func::maybe_null correctly in fix_fields
The counter keeps track of the number of items added to the sum
function. It is increased when we add a value to the sum function and
decreased when it is removed.
Add support for "RANGE n PRECEDING|FOLLOWING" frame bounds.
- n is currently limited to whatever Item and Item_sum_plus/minus
can handle (i.e. no DATETIME intervals).
- Didn't check NULL value handling yet.
Part#2: Fix a couple more issues in rows-type frames.
This also has a code cleanup:
- introduce a separate Frame_rows_current_row_(top,bottom). This is
is a special case which doesn't need its cursor or partition bound check
- Split Frame_n_rows into
= Frame_n_rows_preceding (this one is now much simpler)
= Frame_n_rows_following (simpler and works but may need some work still)
Part#1: Frame_n_rows::next_partition() should not assume that the
current table->record[0] points to the first row in the partition.
Since cursor supports move_to() operation, we dont need this.
Simplified audit event dispatching call chain from:
mysql_audit_notify_connection_connect() // can be inlined
mysql_audit_notify() // can't be inlined
connection_class_handler() // can't be inlined
event_class_dispatch() // can be inlined
plugins_dispatch() // can be inlined
plugin->event_notify() // can't be inlined
to:
mysql_audit_notify_connection_connect() // can be inlined
mysql_audit_notify() // can't be inlined
plugins_dispatch() // can be inlined
plugin->event_notify() // can't be inlined
* only copy args[0] to args[2] after fix_fields (when all item
substitutions have already happened)
* change QT_ITEM_FUNC_NULLIF_TO_CASE (that allows to print NULLIF
as CASE) to QT_ITEM_ORIGINAL_FUNC_NULLIF (that prohibits it).
So that NULLIF-to-CASE is allowed by default and only disabled
explicitly for SHOW VIEW|FUNCTION|PROCEDURE and mysql_make_view.
By default it is allowed (in particular in error messages and
debug output, that can happen anytime before or after optimizer).
don't do special SUM_FUNC_ITEM treatment in NULLIF for views
(as before), but do it for derived tables (when
context_analysis_only == CONTEXT_ANALYSIS_ONLY_DERIVED)
There was a race between end_slave() and cleanup code at the end of
handle_slave_sql(). This could cause access to master_info_index and
global_rpl_thread_pool after they had been freed.
Fix by skipping that cleanup if server shutdown is in progress, as is done
in other parts of the code as well (the cleanup, which stops worker threads
that are not needed anymore, is redundant anyway when the server is shutting
down).
INTERVALS
ISSUE:
------
Some string functions return one or a combination of the
parameters as their result. Here the resultant string's
charset could be incorrectly set to that of the chosen
parameter.
This results in incorrect behavior when an ascii string is
expected.
SOLUTION:
---------
Since an ascii string is expected, val_str_ascii should
explicitly convert the string.
Part of the fix is a backport of Bug#22340858 for mysql-5.5
and mysql-5.6.
Introduce explicitly-defined cursors.
- Rowid_seq_cursor just reads the rowid sequence
- Table_read_cursor also reads rows for the rowids
- Also, we need a kind of cursor that stops at partition boundary?
FULL
Bug#21753696: MAKE SHOW SLAVE STATUS NON BLOCKING IF IO
THREAD WAITS FOR DISK SPACE
Problem:
========
Currently SHOW SLAVE STATUS blocks if IO thread waits for
disk space. This makes automation tools verifying
server health block on taking relevant action. Finally this
will create SHOW SLAVE STATUS piles.
Analysis:
=========
SHOW SLAVE STATUS hangs on mi->data_lock if relay log write
is waiting for free disk space while holding mi->data_lock.
mi->data_lock is needed to protect the format description
event (mi->format_description_event) which is accessed by
the clients running FLUSH LOGS and slave IO thread. Note
relay log writes don't need to be protected by
mi->data_lock, LOCK_log is used to protect relay log between
IO and SQL thread (see MYSQL_BIN_LOG::append_event). The
code takes mi->data_lock to protect
mi->format_description_event during relay log rotate which
might get triggered right after relay log write.
Fix:
====
Release the data_lock just for the duration of writing into
relay log.
Made change to ensure the following lock order is maintained
to avoid deadlocks.
data_lock, LOCK_log
data_lock is held during relay log rotations to protect
the description event.
REPLICATION
Problem: In RBR mode, merge table updates are not successfully applied on a cascading
replication.
Analysis & Fix: Every type of row event is preceded by one or more table_map_log_events
that gives the information about all the tables that are involved in the row
event. Server maintains the list in RPL_TABLE_LIST and it goes through all the
tables and checks for the compatibility between master and slave. Before
checking for the compatibility, it calls 'open_tables()' which takes the list
of all tables that needs to be locked and opened. In RBR, because of the
Table_map_log_event , we already have all the tables including base tables in
the list. But the open_tables() which is generic call takes care of appending
base tables if the list contains merge tables. There is an assumption in the
current replication layer logic that these tables (TABLE_LIST type objects) are always
added in the end of the list. Replication layer maintains the count of
tables(tables_to_lock_count) that needs to be verified for compatibility check
and runs through only those many tables from the list and rest of the objects
in linked list can be skipped. But this assumption is wrong.
open_tables()->..->add_children_to_list() adds base tables to the list immediately
after seeing the merge table in the list.
For eg: If the list passed to open_tables() is t1->t2->t3 where t3 is merge
table (and t1 and t2 are base tables), it adds t1'->t2' to the list after t3.
New table list looks like t1->t2->t3->t1'->t2'. It looks like it added at the
end of the list but that is not correct. If the list passed to open_tables()
is t3->t1->t2 where t3 is merge table (and t1 and t2 are base tables), the new
prepared list will be t3->t1'->t2'->t1->t2. Where t1' and t2' are of
TABLE_LIST objects which were added by add_children_to_list() call and replication
layer should not look into them. Here tables_to_lock_count will not help as the
objects are added in between the list.
Fix: After investigating add_children_list() logic (which is called from open_tables()),
there is no flag/logic in it to skip adding the children to the list even if the
children are already included in the table list. Hence to fix the issue, a
logic should be added in the replication layer to skip children in the list by
checking whether 'parent_l' is non-null or not. If it is children, we will skip 'compatibility'
check for that table.
Also this patch is not removing 'tables_to_lock_count' logic for the performance issues
if there are any children at the end of the list, those can be easily skipped directly by
stopping the loop with tables_to_lock_count check.
