This patch:
- Moves all definitions from the mysql_priv.h file into
header files for the component where the variable is
defined
- Creates header files if the component lacks one
- Eliminates all include directives from mysql_priv.h
- Eliminates all circular include cycles
- Rename time.cc to sql_time.cc
- Rename mysql_priv.h to sql_priv.h
into partitioned MyISAM table
Problem was that the ha_data structure was introduced in 5.1
and only used for partitioning first, but with the intention
of be of use for others engines as well, and when used by other
engines it would clash if it also was partitioned.
Solution is to move the partitioning specific data to a separate
structure, with its own mutex (which is used for auto_increment).
Also did rename PARTITION_INFO to PARTITION_STATS since there
already exist a class named partition_info, also cleaned up
some related variables.
mysql-test/r/partition_binlog_stmt.result:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
New result file
mysql-test/t/partition_binlog_stmt.test:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
New result file
sql/ha_ndbcluster.cc:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
Rename of PARTITION_INFO to PARTITION_STATS to better
match the use (and there is also a class named
partition_info...)
sql/ha_ndbcluster.h:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
Rename of PARTITION_INFO to PARTITION_STATS to better
match the use (and there is also a class named
partition_info...)
sql/ha_partition.cc:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
Removed the partitioning engines use of ha_data in
TABLE_SHARE and added ha_part_data instead, since
they collide if used in the same time.
Rename of PARTITION_INFO to PARTITION_STATS to better
match the use (and there is also a class named
partition_info...)
Removed some dead code.
sql/ha_partition.h:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
Removed some dead code.
Rename of PARTITION_INFO to PARTITION_STATS to better
match the use (and there is also a class named
partition_info...)
Removed the partitioning engines use of ha_data in
TABLE_SHARE and added ha_part_data instead, since
they collide if used in the same time.
sql/handler.cc:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
Rename of PARTITION_INFO to PARTITION_STATS to better
match the use (and there is also a class named
partition_info...)
sql/handler.h:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
Rename of PARTITION_INFO to PARTITION_STATS to better
match the use (and there is also a class named
partition_info...)
sql/mysql_priv.h:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
Removed the partitioning engines use of ha_data in
TABLE_SHARE and added ha_part_data instead, since
they collide if used in the same time.
Added key_PARTITION_LOCK_auto_inc for instrumentation.
sql/mysqld.cc:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
Removed the partitioning engines use of ha_data in
TABLE_SHARE and added ha_part_data instead, since
they collide if used in the same time.
Added key_PARTITION_LOCK_auto_inc for instrumentation.
sql/partition_info.h:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
Removed part_state* since it was not in use.
sql/sql_partition.cc:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
Removed part_state* since it was not in use.
sql/sql_partition.h:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
Cleaned up old commented out code.
Removed part_state* since it was not in use.
sql/sql_show.cc:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
Rename of PARTITION_INFO to PARTITION_STATS to better
match the use (and there is also a class named
partition_info...)
Renamed partition_info to partition_info_str, since
partition_info is a name of a class.
sql/sql_table.cc:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
Renamed partition_info to partition_info_str, since
partition_info is a name of a class.
sql/table.cc:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
Removed the partitioning engines use of ha_data in
TABLE_SHARE and added ha_part_data instead, since
they collide if used in the same time.
Renamed partition_info to partition_info_str, since
partition_info is a name of a class.
removed part_state* since it was not in use.
sql/table.h:
Bug#51851: Server with SBR locks mutex twice on LOAD DATA
into partitioned MyISAM table
Removed the partitioning engines use of ha_data in
TABLE_SHARE and added ha_part_data instead, since
they collide if used in the same time.
Renamed partition_info to partition_info_str, since
partition_info is a name of a class.
removed part_state* since it was not in use.
into partitioned MyISAM table
Problem was that the ha_data structure was introduced in 5.1
and only used for partitioning first, but with the intention
of be of use for others engines as well, and when used by other
engines it would clash if it also was partitioned.
Solution is to move the partitioning specific data to a separate
structure, with its own mutex (which is used for auto_increment).
Also did rename PARTITION_INFO to PARTITION_STATS since there
already exist a class named partition_info, also cleaned up
some related variables.
myisam-recover options changed from OFF to 'DEFAULT' to get less change of data loss when using MyISAM.
(The disadvantage is that changed MyISAM tables will be checked at access time; Use --myisam-recover=OFF for old behavior)
Don't call extra(HA_EXTRA_FORCE_REOPEN) in ALTER TABLE if table is locked as this will mark table as crashed!
Added assert to detect if we accidently would use MyISAM versioning in MySQL
include/my_base.h:
Mark NOT_USED as USED, as we now use this as a flag to not call extra()
mysql-test/mysql-test-run.pl:
Don't write all options when there is something wrong with the arguments
mysql-test/r/sp-destruct.result:
Add missing flush of mysql.proc (as the test copied live tables)
mysql-test/r/variables.result:
myisam-recover options changed to 'default'
mysql-test/r/view.result:
Don't show create time in result
mysql-test/suite/maria/t/maria-recovery2-master.opt:
Don't run test with myisam-recover (as this produces extra warnings during simulated death)
mysql-test/t/sp-destruct.test:
Add missing flush of mysql.proc (as the test copied live tables)
mysql-test/t/view.test:
Don't show create time in result
sql/lock.cc:
Added marker if table was deleted to argument list
sql/mysql_priv.h:
Added marker if table was deleted to argument list
sql/mysqld.cc:
myisam-recover options changed from OFF to 'DEFAULT' to get less change of data loss when using MyISAM
Allow one to specify OFF as argument to myisam-recover (was default before but one couldn't specify it)
sql/sql_base.cc:
Mark if table is going to be deleted
sql/sql_delete.cc:
Mark if table is going to be deleted
sql/sql_table.cc:
Mark if table is going to be deleted
Don't call extra(HA_EXTRA_FORCE_REOPEN) in ALTER TABLE if table is locked as this will mark table as crashed!
sql/table.cc:
Signal to handler if table is getting deleted as part of getting droped from table cache.
sql/table.h:
Added marker if table is going to be deleted.
storage/maria/ha_maria.cc:
Don't search for transaction handler if file is not transactional or outside of transaction
(Fixed possible core dump)
storage/maria/ma_blockrec.c:
Don't write changed information if table is going to be deleted.
storage/maria/ma_close.c:
Don't write changed information if table is going to be deleted.
storage/maria/ma_extra.c:
Mark tables that are deleted as crased, to ensure good behavior on restart if we suddenly crash.
storage/maria/ma_locking.c:
Cleanup
storage/maria/ma_recovery.c:
We need trnman to be inited during redo phase (to be able to open tables checked with maria_chk)
storage/maria/maria_def.h:
Added marker if table is going to be deleted.
storage/myisam/mi_close.c:
Don't write changed information if table is going to be deleted.
storage/myisam/mi_extra.c:
Mark tables that are deleted as crased, to ensure good behavior on restart if we suddenly crash.
storage/myisam/mi_open.c:
Added assert to detect if we accidently would use MyISAM versioning in MySQL
storage/myisam/myisamdef.h:
Added marker if table is going to be deleted.
Add a wait-for graph based deadlock detector to the
MDL subsystem.
Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and
bug #37346 "innodb does not detect deadlock between update and
alter table".
The first bug manifested itself as an unwarranted abort of a
transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER
statement, when this transaction tried to repeat use of a
table, which it has already used in a similar fashion before
ALTER started.
The second bug showed up as a deadlock between table-level
locks and InnoDB row locks, which was "detected" only after
innodb_lock_wait_timeout timeout.
A transaction would start using the table and modify a few
rows.
Then ALTER TABLE would come in, and start copying rows
into a temporary table. Eventually it would stumble on
the modified records and get blocked on a row lock.
The first transaction would try to do more updates, and get
blocked on thr_lock.c lock.
This situation of circular wait would only get resolved
by a timeout.
Both these bugs stemmed from inadequate solutions to the
problem of deadlocks occurring between different
locking subsystems.
In the first case we tried to avoid deadlocks between metadata
locking and table-level locking subsystems, when upgrading shared
metadata lock to exclusive one.
Transactions holding the shared lock on the table and waiting for
some table-level lock used to be aborted too aggressively.
We also allowed ALTER TABLE to start in presence of transactions
that modify the subject table. ALTER TABLE acquires
TL_WRITE_ALLOW_READ lock at start, and that block all writes
against the table (naturally, we don't want any writes to be lost
when switching the old and the new table). TL_WRITE_ALLOW_READ
lock, in turn, would block the started transaction on thr_lock.c
lock, should they do more updates. This, again, lead to the need
to abort such transactions.
The second bug occurred simply because we didn't have any
mechanism to detect deadlocks between the table-level locks
in thr_lock.c and row-level locks in InnoDB, other than
innodb_lock_wait_timeout.
This patch solves both these problems by moving lock conflicts
which are causing these deadlocks into the metadata locking
subsystem, thus making it possible to avoid or detect such
deadlocks inside MDL.
To do this we introduce new type-of-operation-aware metadata
locks, which allow MDL subsystem to know not only the fact that
transaction has used or is going to use some object but also what
kind of operation it has carried out or going to carry out on the
object.
This, along with the addition of a special kind of upgradable
metadata lock, allows ALTER TABLE to wait until all
transactions which has updated the table to go away.
This solves the second issue.
Another special type of upgradable metadata lock is acquired
by LOCK TABLE WRITE. This second lock type allows to solve the
first issue, since abortion of table-level locks in event of
DDL under LOCK TABLES becomes also unnecessary.
Below follows the list of incompatible changes introduced by
this patch:
- From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those
statements that acquire TL_WRITE_ALLOW_READ lock)
wait for all transactions which has *updated* the table to
complete.
- From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE
(i.e. all statements which acquire TL_WRITE table-level lock) wait
for all transaction which *updated or read* from the table
to complete.
As a consequence, innodb_table_locks=0 option no longer applies
to LOCK TABLES ... WRITE.
- DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort
statements or transactions which use tables being dropped or
renamed, and instead wait for these transactions to complete.
- Since LOCK TABLES WRITE now takes a special metadata lock,
not compatible with with reads or writes against the subject table
and transaction-wide, thr_lock.c deadlock avoidance algorithm
that used to ensure absence of deadlocks between LOCK TABLES
WRITE and other statements is no longer sufficient, even for
MyISAM. The wait-for graph based deadlock detector of MDL
subsystem may sometimes be necessary and is involved. This may
lead to ER_LOCK_DEADLOCK error produced for multi-statement
transactions even if these only use MyISAM:
session 1: session 2:
begin;
update t1 ... lock table t2 write, t1 write;
-- gets a lock on t2, blocks on t1
update t2 ...
(ER_LOCK_DEADLOCK)
- Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE
was abandoned.
LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same
priority as the usual LOCK TABLE ... WRITE.
SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in
the wait queue.
- We do not take upgradable metadata locks on implicitly
locked tables. So if one has, say, a view v1 that uses
table t1, and issues:
LOCK TABLE v1 WRITE;
FLUSH TABLE t1; -- (or just 'FLUSH TABLES'),
an error is produced.
In order to be able to perform DDL on a table under LOCK TABLES,
the table must be locked explicitly in the LOCK TABLES list.
mysql-test/include/handler.inc:
Adjusted test case to trigger an execution path on which bug 41110
"crash with handler command when used concurrently with alter
table" and bug 41112 "crash in mysql_ha_close_table/get_lock_data
with alter table" were originally discovered. Left old test case
which no longer triggers this execution path for the sake of
coverage.
Added test coverage for HANDLER SQL statements and type-aware
metadata locks.
Added a test for the global shared lock and HANDLER SQL.
Updated tests to take into account that the old simple deadlock
detection heuristics was replaced with a graph-based deadlock
detector.
mysql-test/r/debug_sync.result:
Updated results (see debug_sync.test).
mysql-test/r/handler_innodb.result:
Updated results (see handler.inc test).
mysql-test/r/handler_myisam.result:
Updated results (see handler.inc test).
mysql-test/r/innodb-lock.result:
Updated results (see innodb-lock.test).
mysql-test/r/innodb_mysql_lock.result:
Updated results (see innodb_mysql_lock.test).
mysql-test/r/lock.result:
Updated results (see lock.test).
mysql-test/r/lock_multi.result:
Updated results (see lock_multi.test).
mysql-test/r/lock_sync.result:
Updated results (see lock_sync.test).
mysql-test/r/mdl_sync.result:
Updated results (see mdl_sync.test).
mysql-test/r/sp-threads.result:
SHOW PROCESSLIST output has changed due to the fact that waiting
for LOCK TABLES WRITE now happens within metadata locking
subsystem.
mysql-test/r/truncate_coverage.result:
Updated results (see truncate_coverage.test).
mysql-test/suite/funcs_1/datadict/processlist_val.inc:
SELECT FROM I_S.PROCESSLIST output has changed due to fact that
waiting for LOCK TABLES WRITE now happens within metadata locking
subsystem.
mysql-test/suite/funcs_1/r/processlist_val_no_prot.result:
SELECT FROM I_S.PROCESSLIST output has changed due to fact that
waiting for LOCK TABLES WRITE now happens within metadata locking
subsystem.
mysql-test/suite/rpl/t/rpl_sp.test:
Updated to a new SHOW PROCESSLIST state name.
mysql-test/t/debug_sync.test:
Use LOCK TABLES READ instead of LOCK TABLES WRITE as the latter
no longer allows to trigger execution path involving waiting on
thr_lock.c lock and therefore reaching debug sync-point covered
by this test.
mysql-test/t/innodb-lock.test:
Adjusted test case to the fact that innodb_table_locks=0 option is
no longer supported, since LOCK TABLES WRITE handles all its
conflicts within MDL subsystem.
mysql-test/t/innodb_mysql_lock.test:
Added test for bug #37346 "innodb does not detect deadlock between
update and alter table".
mysql-test/t/lock.test:
Added test coverage which checks the fact that we no longer support
DDL under LOCK TABLES on tables which were locked implicitly.
Adjusted existing test cases accordingly.
mysql-test/t/lock_multi.test:
Added test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary
deadlock". Adjusted other test cases to take into account the
fact that waiting for LOCK TABLES ... WRITE now happens within MDL
subsystem.
mysql-test/t/lock_sync.test:
Since LOCK TABLES ... WRITE now takes SNRW metadata lock for
tables locked explicitly we have to implicitly lock InnoDB tables
(through view) to trigger the table-level lock conflict between
TL_WRITE and TL_WRITE_ALLOW_WRITE.
mysql-test/t/mdl_sync.test:
Added basic test coverage for type-of-operation-aware metadata
locks. Also covered with tests some use cases involving HANDLER
statements in which a deadlock could arise.
Adjusted existing tests to take type-of-operation-aware MDL into
account.
mysql-test/t/multi_update.test:
Update to a new SHOW PROCESSLIST state name.
mysql-test/t/truncate_coverage.test:
Adjusted test case after making LOCK TABLES WRITE to wait until
transactions that use the table to be locked are completed.
Updated to the changed name of DEBUG_SYNC point.
sql/handler.cc:
Global read lock functionality has been
moved into a class.
sql/lock.cc:
Global read lock functionality has been
moved into a class.
Updated code to use the new MDL API.
sql/mdl.cc:
Introduced new type-of-operation aware metadata locks.
To do this:
- Changed MDL_lock to use one list for waiting requests and one
list for granted requests. For each list, added a bitmap
that holds information what lock types a list contains.
Added a helper class MDL_lock::List to manipulate with granted
and waited lists while keeping the bitmaps in sync
with list contents.
- Changed lock-compatibility functions to use bitmaps that
define compatibility.
- Introduced a graph based deadlock detector inspired by
waiting_threads.c from Maria implementation.
- Now that we have a deadlock detector, and no longer have
a global lock to protect individual lock objects, but rather
use an rw lock per object, removed redundant code for upgrade,
and the global read lock. Changed the MDL API to
no longer require the caller to acquire the global
intention exclusive lock by means of a separate method.
Removed a few more methods that became redundant.
- Removed deadlock detection heuristic, it has been made
obsolete by the deadlock detector.
- With operation-type-aware metadata locks, MDL subsystem has
become aware of potential conflicts between DDL and open
transactions. This made it possible to remove calls to
mysql_abort_transactions_with_shared_lock() from acquisition
paths for exclusive lock and lock upgrade. Now we can simply
wait for these transactions to complete without fear of
deadlock. Function mysql_lock_abort() has also become
unnecessary for all conflicting cases except when a DDL
conflicts with a connection that has an open HANDLER.
sql/mdl.h:
Introduced new type-of-operation aware metadata locks.
Introduced a graph based deadlock detector and supporting
methods.
Added comments.
God rid of redundant API calls.
Renamed m_lt_or_ha_sentinel to m_trans_sentinel,
since now it guards the global read lock as well as
LOCK TABLES and HANDLER locks.
sql/mysql_priv.h:
Moved the global read lock functionality into a
class.
Added MYSQL_OPEN_FORCE_SHARED_MDL flag which forces
open_tables() to take MDL_SHARED on tables instead of
metadata locks specified in the parser. We use this to
allow PREPARE run concurrently in presence of
LOCK TABLES ... WRITE.
Added signature for find_table_for_mdl_ugprade().
sql/set_var.cc:
Global read lock functionality has been
moved into a class.
sql/sp_head.cc:
When creating TABLE_LIST elements for prelocking or
system tables set the type of request for metadata
lock according to the operation that will be performed
on the table.
sql/sql_base.cc:
- Updated code to use the new MDL API.
- In order to avoid locks starvation we take upgradable
locks all at once. As result implicitly locked tables no
longer get an upgradable lock. Consequently DDL and FLUSH
TABLES for such tables is prohibited.
find_write_locked_table() was replaced by
find_table_for_mdl_upgrade() function.
open_table() was adjusted to return TABLE instance with
upgradable ticket when necessary.
- We no longer wait for all locks on OT_WAIT back off
action -- only on the lock that caused the wait
conflict. Moreover, now we distinguish cases when we
have to wait due to conflict in MDL and old version
of table in TDC.
- Upate mysql_notify_threads_having_share_locks()
to only abort thr_lock.c waits of threads that
have open HANDLERs, since lock conflicts with only
these threads now can lead to deadlocks not detectable
by the MDL deadlock detector.
- Remove mysql_abort_transactions_with_shared_locks()
which is no longer needed.
sql/sql_class.cc:
Global read lock functionality has been moved into a class.
Re-arranged code in THD::cleanup() to simplify assert.
sql/sql_class.h:
Introduced class to incapsulate global read lock
functionality.
Now sentinel in MDL subsystem guards the global read lock
as well as LOCK TABLES and HANDLER locks. Adjusted code
accordingly.
sql/sql_db.cc:
Global read lock functionality has been moved into a class.
sql/sql_delete.cc:
We no longer acquire upgradable metadata locks on tables
which are locked by LOCK TABLES implicitly. As result
TRUNCATE TABLE is no longer allowed for such tables.
Updated code to use the new MDL API.
sql/sql_handler.cc:
Inform MDL_context about presence of open HANDLERs.
Since HANLDERs break MDL protocol by acquiring table-level
lock while holding only S metadata lock on a table MDL
subsystem should take special care about such contexts (Now
this is the only case when mysql_lock_abort() is used).
sql/sql_parse.cc:
Global read lock functionality has been moved into a class.
Do not take upgradable metadata locks when opening tables
for CREATE TABLE SELECT as it is not necessary and limits
concurrency.
When initializing TABLE_LIST objects before adding them
to the table list set the type of request for metadata lock
according to the operation that will be performed on the
table.
We no longer acquire upgradable metadata locks on tables
which are locked by LOCK TABLES implicitly. As result FLUSH
TABLES is no longer allowed for such tables.
sql/sql_prepare.cc:
Use MYSQL_OPEN_FORCE_SHARED_MDL flag when opening
tables during PREPARE. This allows PREPARE to run
concurrently in presence of LOCK TABLES ... WRITE.
sql/sql_rename.cc:
Global read lock functionality has been moved into a class.
sql/sql_show.cc:
Updated code to use the new MDL API.
sql/sql_table.cc:
Global read lock functionality has been moved into a class.
We no longer acquire upgradable metadata locks on tables
which are locked by LOCK TABLES implicitly. As result DROP
TABLE is no longer allowed for such tables.
Updated code to use the new MDL API.
sql/sql_trigger.cc:
Global read lock functionality has been moved into a class.
We no longer acquire upgradable metadata locks on tables
which are locked by LOCK TABLES implicitly. As result
CREATE/DROP TRIGGER is no longer allowed for such tables.
Updated code to use the new MDL API.
sql/sql_view.cc:
Global read lock functionality has been moved into a class.
Fixed results of wrong merge that led to misuse of GLR API.
CREATE VIEW statement is not a commit statement.
sql/table.cc:
When resetting TABLE_LIST objects for PS or SP re-execution
set the type of request for metadata lock according to the
operation that will be performed on the table. Do the same
in auxiliary function initializing metadata lock requests
in a table list.
sql/table.h:
When initializing TABLE_LIST objects set the type of request
for metadata lock according to the operation that will be
performed on the table.
sql/transaction.cc:
Global read lock functionality has been moved into a class.
Add a wait-for graph based deadlock detector to the
MDL subsystem.
Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and
bug #37346 "innodb does not detect deadlock between update and
alter table".
The first bug manifested itself as an unwarranted abort of a
transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER
statement, when this transaction tried to repeat use of a
table, which it has already used in a similar fashion before
ALTER started.
The second bug showed up as a deadlock between table-level
locks and InnoDB row locks, which was "detected" only after
innodb_lock_wait_timeout timeout.
A transaction would start using the table and modify a few
rows.
Then ALTER TABLE would come in, and start copying rows
into a temporary table. Eventually it would stumble on
the modified records and get blocked on a row lock.
The first transaction would try to do more updates, and get
blocked on thr_lock.c lock.
This situation of circular wait would only get resolved
by a timeout.
Both these bugs stemmed from inadequate solutions to the
problem of deadlocks occurring between different
locking subsystems.
In the first case we tried to avoid deadlocks between metadata
locking and table-level locking subsystems, when upgrading shared
metadata lock to exclusive one.
Transactions holding the shared lock on the table and waiting for
some table-level lock used to be aborted too aggressively.
We also allowed ALTER TABLE to start in presence of transactions
that modify the subject table. ALTER TABLE acquires
TL_WRITE_ALLOW_READ lock at start, and that block all writes
against the table (naturally, we don't want any writes to be lost
when switching the old and the new table). TL_WRITE_ALLOW_READ
lock, in turn, would block the started transaction on thr_lock.c
lock, should they do more updates. This, again, lead to the need
to abort such transactions.
The second bug occurred simply because we didn't have any
mechanism to detect deadlocks between the table-level locks
in thr_lock.c and row-level locks in InnoDB, other than
innodb_lock_wait_timeout.
This patch solves both these problems by moving lock conflicts
which are causing these deadlocks into the metadata locking
subsystem, thus making it possible to avoid or detect such
deadlocks inside MDL.
To do this we introduce new type-of-operation-aware metadata
locks, which allow MDL subsystem to know not only the fact that
transaction has used or is going to use some object but also what
kind of operation it has carried out or going to carry out on the
object.
This, along with the addition of a special kind of upgradable
metadata lock, allows ALTER TABLE to wait until all
transactions which has updated the table to go away.
This solves the second issue.
Another special type of upgradable metadata lock is acquired
by LOCK TABLE WRITE. This second lock type allows to solve the
first issue, since abortion of table-level locks in event of
DDL under LOCK TABLES becomes also unnecessary.
Below follows the list of incompatible changes introduced by
this patch:
- From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those
statements that acquire TL_WRITE_ALLOW_READ lock)
wait for all transactions which has *updated* the table to
complete.
- From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE
(i.e. all statements which acquire TL_WRITE table-level lock) wait
for all transaction which *updated or read* from the table
to complete.
As a consequence, innodb_table_locks=0 option no longer applies
to LOCK TABLES ... WRITE.
- DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort
statements or transactions which use tables being dropped or
renamed, and instead wait for these transactions to complete.
- Since LOCK TABLES WRITE now takes a special metadata lock,
not compatible with with reads or writes against the subject table
and transaction-wide, thr_lock.c deadlock avoidance algorithm
that used to ensure absence of deadlocks between LOCK TABLES
WRITE and other statements is no longer sufficient, even for
MyISAM. The wait-for graph based deadlock detector of MDL
subsystem may sometimes be necessary and is involved. This may
lead to ER_LOCK_DEADLOCK error produced for multi-statement
transactions even if these only use MyISAM:
session 1: session 2:
begin;
update t1 ... lock table t2 write, t1 write;
-- gets a lock on t2, blocks on t1
update t2 ...
(ER_LOCK_DEADLOCK)
- Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE
was abandoned.
LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same
priority as the usual LOCK TABLE ... WRITE.
SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in
the wait queue.
- We do not take upgradable metadata locks on implicitly
locked tables. So if one has, say, a view v1 that uses
table t1, and issues:
LOCK TABLE v1 WRITE;
FLUSH TABLE t1; -- (or just 'FLUSH TABLES'),
an error is produced.
In order to be able to perform DDL on a table under LOCK TABLES,
the table must be locked explicitly in the LOCK TABLES list.
Queries optimized with GROUP_MIN_MAX didn't cleanup KEYREAD
optimization properly. As a result subsequent queries may
return incomplete rows (fields are initialized to default
values).
mysql-test/r/group_min_max.result:
A test case for BUG#49902.
mysql-test/t/group_min_max.test:
A test case for BUG#49902.
sql/opt_range.cc:
Refactor of KEYREAD optimization switch so that KEYREAD
handler state is in sync with st_table::key_read flag.
All SQL code is supposed to switch KEYREAD optimization
via st_table::set_keyread().
sql/opt_sum.cc:
Refactor of KEYREAD optimization switch so that KEYREAD
handler state is in sync with st_table::key_read flag.
All SQL code is supposed to switch KEYREAD optimization
via st_table::set_keyread().
sql/sql_select.cc:
Refactor of KEYREAD optimization switch so that KEYREAD
handler state is in sync with st_table::key_read flag.
All SQL code is supposed to switch KEYREAD optimization
via st_table::set_keyread().
sql/sql_update.cc:
Refactor of KEYREAD optimization switch so that KEYREAD
handler state is in sync with st_table::key_read flag.
All SQL code is supposed to switch KEYREAD optimization
via st_table::set_keyread().
sql/table.cc:
Refactor of KEYREAD optimization switch so that KEYREAD
handler state is in sync with st_table::key_read flag.
All SQL code is supposed to switch KEYREAD optimization
via st_table::set_keyread().
sql/table.h:
Refactor of KEYREAD optimization switch so that KEYREAD
handler state is in sync with st_table::key_read flag.
All SQL code is supposed to switch KEYREAD optimization
via st_table::set_keyread().
Queries optimized with GROUP_MIN_MAX didn't cleanup KEYREAD
optimization properly. As a result subsequent queries may
return incomplete rows (fields are initialized to default
values).
- Marked a couple of tests with --big
- Fixed xtradb/handler/ha_innodb.cc to call explain_filename()
storage/xtradb/handler/ha_innodb.cc:
Call explain_filename() to get proper names for partitioned tables
Bug #46654 False deadlock on concurrent DML/DDL with partitions,
inconsistent behavior
The problem was that if one connection is running a multi-statement
transaction which involves a single partitioned table, and another
connection attempts to alter the table, the first connection gets
ER_LOCK_DEADLOCK and cannot proceed anymore, even when the ALTER TABLE
statement in another connection has timed out or failed.
The reason for this was that the prepare phase for ALTER TABLE for
partitioned tables removed all instances of the table from the table
definition cache before it started waiting on the lock. The transaction
running in the first connection would notice this and report ER_LOCK_DEADLOCK.
This patch changes the prep_alter_part_table() ALTER TABLE code so that
tdc_remove_table() is no longer called. Instead, only the TABLE instance
changed by prep_alter_part_table() is marked as needing reopen.
The patch also removes an unnecessary call to tdc_remove_table() from
mysql_unpack_partition() as the changed TABLE object is destroyed by the
caller at a later point.
Test case added in partition_sync.test.
Bug #46654 False deadlock on concurrent DML/DDL with partitions,
inconsistent behavior
The problem was that if one connection is running a multi-statement
transaction which involves a single partitioned table, and another
connection attempts to alter the table, the first connection gets
ER_LOCK_DEADLOCK and cannot proceed anymore, even when the ALTER TABLE
statement in another connection has timed out or failed.
The reason for this was that the prepare phase for ALTER TABLE for
partitioned tables removed all instances of the table from the table
definition cache before it started waiting on the lock. The transaction
running in the first connection would notice this and report ER_LOCK_DEADLOCK.
This patch changes the prep_alter_part_table() ALTER TABLE code so that
tdc_remove_table() is no longer called. Instead, only the TABLE instance
changed by prep_alter_part_table() is marked as needing reopen.
The patch also removes an unnecessary call to tdc_remove_table() from
mysql_unpack_partition() as the changed TABLE object is destroyed by the
caller at a later point.
Test case added in partition_sync.test.
Bug#42546 Backup: RESTORE fails, thinking it finds an existing table
The problem occured when a MDL locking conflict happened for a non-existent
table between a CREATE and a INSERT statement. The code for CREATE
interpreted this lock conflict to mean that the table existed,
which meant that the statement failed when it should not have.
The problem could occur for CREATE TABLE, CREATE TABLE LIKE and
ALTER TABLE RENAME.
This patch fixes the problem for CREATE TABLE and CREATE TABLE LIKE.
It is based on code backported from the mysql-6.1-fk tree written
by Dmitry Lenev. CREATE now uses normal open_and_lock_tables() code
to acquire exclusive locks. This means that for the test case in the bug
description, CREATE will wait until INSERT completes so that it can
get the exclusive lock. This resolves the reported bug.
The patch also prohibits CREATE TABLE and CREATE TABLE LIKE under
LOCK TABLES. Note that this is an incompatible change and must
be reflected in the documentation. Affected test cases have been
updated.
mdl_sync.test contains tests for CREATE TABLE and CREATE TABLE LIKE.
Fixing the issue for ALTER TABLE RENAME is beyond the scope of this
patch. ALTER TABLE cannot be prohibited from working under LOCK TABLES
as this could seriously impact customers and a proper fix would require
a significant rewrite.
Bug#42546 Backup: RESTORE fails, thinking it finds an existing table
The problem occured when a MDL locking conflict happened for a non-existent
table between a CREATE and a INSERT statement. The code for CREATE
interpreted this lock conflict to mean that the table existed,
which meant that the statement failed when it should not have.
The problem could occur for CREATE TABLE, CREATE TABLE LIKE and
ALTER TABLE RENAME.
This patch fixes the problem for CREATE TABLE and CREATE TABLE LIKE.
It is based on code backported from the mysql-6.1-fk tree written
by Dmitry Lenev. CREATE now uses normal open_and_lock_tables() code
to acquire exclusive locks. This means that for the test case in the bug
description, CREATE will wait until INSERT completes so that it can
get the exclusive lock. This resolves the reported bug.
The patch also prohibits CREATE TABLE and CREATE TABLE LIKE under
LOCK TABLES. Note that this is an incompatible change and must
be reflected in the documentation. Affected test cases have been
updated.
mdl_sync.test contains tests for CREATE TABLE and CREATE TABLE LIKE.
Fixing the issue for ALTER TABLE RENAME is beyond the scope of this
patch. ALTER TABLE cannot be prohibited from working under LOCK TABLES
as this could seriously impact customers and a proper fix would require
a significant rewrite.
------------------------------------------------------------
revno: 2617.68.25
committer: Dmitry Lenev <dlenev@mysql.com>
branch nick: mysql-next-bg-pre2-2
timestamp: Wed 2009-09-16 18:26:50 +0400
message:
Follow-up for one of pre-requisite patches for fixing bug #30977
"Concurrent statement using stored function and DROP FUNCTION
breaks SBR".
Made enum_mdl_namespace enum part of MDL_key class and removed MDL_
prefix from the names of enum members. In order to do the latter
changed name of PROCEDURE symbol to PROCEDURE_SYM (otherwise macro
which was automatically generated for this symbol conflicted with
MDL_key::PROCEDURE enum member).
------------------------------------------------------------
revno: 2617.68.25
committer: Dmitry Lenev <dlenev@mysql.com>
branch nick: mysql-next-bg-pre2-2
timestamp: Wed 2009-09-16 18:26:50 +0400
message:
Follow-up for one of pre-requisite patches for fixing bug #30977
"Concurrent statement using stored function and DROP FUNCTION
breaks SBR".
Made enum_mdl_namespace enum part of MDL_key class and removed MDL_
prefix from the names of enum members. In order to do the latter
changed name of PROCEDURE symbol to PROCEDURE_SYM (otherwise macro
which was automatically generated for this symbol conflicted with
MDL_key::PROCEDURE enum member).
A pre-requisite patch for Bug#30977 "Concurrent statement using
stored function and DROP FUNCTION breaks SBR".
This patch changes the MDL API by introducing a namespace for
lock keys: MDL_TABLE for tables and views and MDL_PROCEDURE
for stored procedures and functions. The latter is needed for
the fix for Bug#30977.
A pre-requisite patch for Bug#30977 "Concurrent statement using
stored function and DROP FUNCTION breaks SBR".
This patch changes the MDL API by introducing a namespace for
lock keys: MDL_TABLE for tables and views and MDL_PROCEDURE
for stored procedures and functions. The latter is needed for
the fix for Bug#30977.
----------------------------------------------------------
revno: 2617.69.21
committer: Konstantin Osipov <kostja@sun.com>
branch nick: 5.4-4284-1-assert
timestamp: Thu 2009-08-13 20:13:55 +0400
message:
A fix and a test case for Bug#46610 "MySQL 5.4.4: MyISAM MRG engine crash
on auto-repair of child".
Also fixes Bug#42862 "Crash on failed attempt to open a children of a
merge table".
MERGE engine needs to extend the global table list
with TABLE_LIST elements for child tables,
so that they are opened and locked.
Previously these table list elements were allocated
in memory of ha_myisammrg object (MERGE engine handler).
That would lead to access to freed memory in
recover_from_failed_open_table_attempt(), which would
try to recover a MERGE table child (MyISAM table)
and use for that TABLE_LIST of that child.
But by the time recover_from_failed_open_table_attempt()
is invoked, ha_myisammrg object that owns this
TABLE_LIST may be destroyed, and thus TABLE_LIST
memory freed.
The fix is to ensure that TABLE_LIST elements
that are added to the global table list (lex->query_tables)
are always allocated in thd->mem_root, which is not
destroyed until end of execution.
If previously TABLE_LIST elements were allocated
at ha_myisammrg::open() (i.e. when the TABLE
object was created and added to the table cache),
now they are allocated in ha_myisammrg::add_chidlren_list()
(i.e. right after "open" of the merge parent in
open_tables()).
We still create a list of children names
at ha_myisammrg::open() to use as a basis
for creation of TABLE_LISTs, that allows
to avoid reading the merge handler data
file on every execution.
mysql-test/r/merge_recover.result:
Test results for Bug#46610.
mysql-test/t/merge_recover-master.opt:
Option file for Bug#46610 test (need a new test because
of that option, which is not tested anywhere else).
mysql-test/t/merge_recover.test:
Add a test case for Bug#46610.
sql/table.h:
MERGE child child_def_version is now moved from TABLE_LIST
to MERGE engine specific data structure.
storage/myisammrg/ha_myisammrg.cc:
Introduce an auxiliary structure to keep MERGE child name
and definition version. A list of Mrg_child_def is created
in ha_myisammrg::open() and reused in ha_myisammrg::add_children_list().
----------------------------------------------------------
revno: 2617.69.21
committer: Konstantin Osipov <kostja@sun.com>
branch nick: 5.4-4284-1-assert
timestamp: Thu 2009-08-13 20:13:55 +0400
message:
A fix and a test case for Bug#46610 "MySQL 5.4.4: MyISAM MRG engine crash
on auto-repair of child".
Also fixes Bug#42862 "Crash on failed attempt to open a children of a
merge table".
MERGE engine needs to extend the global table list
with TABLE_LIST elements for child tables,
so that they are opened and locked.
Previously these table list elements were allocated
in memory of ha_myisammrg object (MERGE engine handler).
That would lead to access to freed memory in
recover_from_failed_open_table_attempt(), which would
try to recover a MERGE table child (MyISAM table)
and use for that TABLE_LIST of that child.
But by the time recover_from_failed_open_table_attempt()
is invoked, ha_myisammrg object that owns this
TABLE_LIST may be destroyed, and thus TABLE_LIST
memory freed.
The fix is to ensure that TABLE_LIST elements
that are added to the global table list (lex->query_tables)
are always allocated in thd->mem_root, which is not
destroyed until end of execution.
If previously TABLE_LIST elements were allocated
at ha_myisammrg::open() (i.e. when the TABLE
object was created and added to the table cache),
now they are allocated in ha_myisammrg::add_chidlren_list()
(i.e. right after "open" of the merge parent in
open_tables()).
We still create a list of children names
at ha_myisammrg::open() to use as a basis
for creation of TABLE_LISTs, that allows
to avoid reading the merge handler data
file on every execution.
----------------------------------------------------------
revno: 2617.69.20
committer: Konstantin Osipov <kostja@sun.com>
branch nick: 5.4-4284-1-assert
timestamp: Thu 2009-08-13 18:29:55 +0400
message:
WL#4284 "Transactional DDL locking"
A review fix.
Since WL#4284 implementation separated MDL_request and MDL_ticket,
MDL_request becamse a utility object necessary only to get a ticket.
Store it by-value in TABLE_LIST with the intent to merge
MDL_request::key with table_list->table_name and table_list->db
in future.
Change the MDL subsystem to not require MDL_requests to
stay around till close_thread_tables().
Remove the list of requests from the MDL context.
Requests for shared metadata locks acquired in open_tables()
are only used as a list in recover_from_failed_open_table_attempt(),
which calls mdl_context.wait_for_locks() for this list.
To keep such list for recover_from_failed_open_table_attempt(),
introduce a context class (Open_table_context), that collects
all requests.
A lot of minor cleanups and simplications that became possible
with this change.
sql/event_db_repository.cc:
Remove alloc_mdl_requests(). Now MDL_request instance is a member
of TABLE_LIST, and init_one_table() initializes it.
sql/ha_ndbcluster_binlog.cc:
Remove now unnecessary declaration and initialization
of binlog_mdl_request.
sql/lock.cc:
No need to allocate MDL requests in lock_table_names() now.
sql/log.cc:
Use init_one_table() method, remove alloc_mdl_requests(),
which is now unnecessary.
sql/log_event.cc:
No need to allocate mdl_request separately now.
Use init_one_table() method.
sql/log_event_old.cc:
Update to the new signature of close_tables_for_reopen().
sql/mdl.cc:
Update try_acquire_exclusive_lock() to be more easy to use.
Function lock_table_name_if_not_cached() has been removed.
Make acquire_shared_lock() signature consistent with
try_acquire_exclusive_lock() signature.
Remove methods that are no longer used.
Update comments.
sql/mdl.h:
Implement an assignment operator that doesn't
copy MDL_key (MDL_key::operator= is private and
should remain private).
This is a hack to work-around assignment of TABLE_LIST
by value in several places. Such assignments violate
encapsulation, since only perform a shallow copy.
In most cases these assignments are a hack on their own.
sql/mysql_priv.h:
Update signatures of close_thread_tables() and close_tables_for_reopen().
sql/sp.cc:
Allocate TABLE_LIST in thd->mem_root.
Use init_one_table().
sql/sp_head.cc:
Use init_one_table().
Remove thd->locked_tables_root, it's no longer needed.
sql/sql_acl.cc:
Use init_mdl_requests() and init_one_table().
sql/sql_base.cc:
Update to new signatures of try_acquire_shared_lock() and
try_acquire_exclusive_lock().
Remove lock_table_name_if_not_cached().
Fix a bug in open_ltable() that would not return ER_LOCK_DEADLOCK
in case of a failed lock_tables() and a multi-statement
transaction.
Fix a bug in open_and_lock_tables_derived() that would
not return ER_LOCK_DEADLOCK in case of a multi-statement
transaction and a failure of lock_tables().
Move assignment of enum_open_table_action to a method of Open_table_context, a new class that maintains information
for backoff actions.
Minor rearrangements of the code.
Remove alloc_mdl_requests() in functions that work with system
tables: instead the patch ensures that callers always initialize
TABLE_LIST argument.
sql/sql_class.cc:
THD::locked_tables_root is no more.
sql/sql_class.h:
THD::locked_tables_root is no more.
Add a declaration for Open_table_context class.
sql/sql_delete.cc:
Update to use the simplified MDL API.
sql/sql_handler.cc:
TABLE_LIST::mdl_request is stored by-value now.
Ensure that mdl_request.ticket is NULL for every request
that is passed into MDL, to satisfy MDL asserts.
@ sql/sql_help.cc
Function open_system_tables_for_read() no longer initializes
mdl_requests.
Move TABLE_LIST::mdl_request initialization closer to
TABLE_LIST initialization.
sql/sql_help.cc:
Function open_system_tables_for_read() no longer initializes
mdl_requests.
Move TABLE_LIST::mdl_request initialization closer to
TABLE_LIST initialization.
sql/sql_insert.cc:
Remove assignment by-value of TABLE_LIST in
TABLEOP_HOOKS. We can't carry over a granted
MDL ticket from one table list to another.
sql/sql_parse.cc:
Change alloc_mdl_requests() -> init_mdl_requests().
@todo We can remove init_mdl_requests() altogether
in some places: all places that call add_table_to_list()
already have mdl requests initialized.
sql/sql_plugin.cc:
Use init_one_table().
THD::locked_tables_root is no more.
sql/sql_servers.cc:
Use init_one_table().
sql/sql_show.cc:
Update acquire_high_priority_shared_lock() to use
TABLE_LIST::mdl_request rather than allocate an own.
Fix get_trigger_table_impl() to use init_one_table(),
check for out of memory, follow the coding style.
sql/sql_table.cc:
Update to work with TABLE_LIST::mdl_request by-value.
Remove lock_table_name_if_not_cached().
The code that used to delegate to it is quite simple and
concise without it now.
sql/sql_udf.cc:
Use init_one_table().
sql/sql_update.cc:
Update to use the new signature of close_tables_for_reopen().
sql/table.cc:
Move re-setting of mdl_requests for prepared statements
and stored procedures from close_thread_tables() to
reinit_stmt_before_use().
Change alloc_mdl_requests() to init_mdl_requests().
init_mdl_requests() is a hack that can't be deleted
until we don't have a list-aware TABLE_LIST constructor.
Hopefully its use will be minimal
sql/table.h:
Change alloc_mdl_requests() to init_mdl_requests()
TABLE_LIST::mdl_request is stored by value.
sql/tztime.cc:
We no longer initialize mdl requests in open_system_tables_for*()
functions. Move this initialization closer to initialization
of the rest of TABLE_LIST members.
storage/myisammrg/ha_myisammrg.cc:
Simplify mdl_request initialization.
----------------------------------------------------------
revno: 2617.69.20
committer: Konstantin Osipov <kostja@sun.com>
branch nick: 5.4-4284-1-assert
timestamp: Thu 2009-08-13 18:29:55 +0400
message:
WL#4284 "Transactional DDL locking"
A review fix.
Since WL#4284 implementation separated MDL_request and MDL_ticket,
MDL_request becamse a utility object necessary only to get a ticket.
Store it by-value in TABLE_LIST with the intent to merge
MDL_request::key with table_list->table_name and table_list->db
in future.
Change the MDL subsystem to not require MDL_requests to
stay around till close_thread_tables().
Remove the list of requests from the MDL context.
Requests for shared metadata locks acquired in open_tables()
are only used as a list in recover_from_failed_open_table_attempt(),
which calls mdl_context.wait_for_locks() for this list.
To keep such list for recover_from_failed_open_table_attempt(),
introduce a context class (Open_table_context), that collects
all requests.
A lot of minor cleanups and simplications that became possible
with this change.
2617.31.12, 2617.31.15, 2617.31.15, 2617.31.16, 2617.43.1
- initial changeset that introduced the fix for
Bug#989 and follow up fixes for all test suite failures
introduced in the initial changeset.
------------------------------------------------------------
revno: 2617.31.1
committer: Davi Arnaut <Davi.Arnaut@Sun.COM>
branch nick: 4284-6.0
timestamp: Fri 2009-03-06 19:17:00 -0300
message:
Bug#989: If DROP TABLE while there's an active transaction, wrong binlog order
WL#4284: Transactional DDL locking
Currently the MySQL server does not keep metadata locks on
schema objects for the duration of a transaction, thus failing
to guarantee the integrity of the schema objects being used
during the transaction and to protect then from concurrent
DDL operations. This also poses a problem for replication as
a DDL operation might be replicated even thought there are
active transactions using the object being modified.
The solution is to defer the release of metadata locks until
a active transaction is either committed or rolled back. This
prevents other statements from modifying the table for the
entire duration of the transaction. This provides commitment
ordering for guaranteeing serializability across multiple
transactions.
- Incompatible change:
If MySQL's metadata locking system encounters a lock conflict,
the usual schema is to use the try and back-off technique to
avoid deadlocks -- this schema consists in releasing all locks
and trying to acquire them all in one go.
But in a transactional context this algorithm can't be utilized
as its not possible to release locks acquired during the course
of the transaction without breaking the transaction commitments.
To avoid deadlocks in this case, the ER_LOCK_DEADLOCK will be
returned if a lock conflict is encountered during a transaction.
Let's consider an example:
A transaction has two statements that modify table t1, then table
t2, and then commits. The first statement of the transaction will
acquire a shared metadata lock on table t1, and it will be kept
utill COMMIT to ensure serializability.
At the moment when the second statement attempts to acquire a
shared metadata lock on t2, a concurrent ALTER or DROP statement
might have locked t2 exclusively. The prescription of the current
locking protocol is that the acquirer of the shared lock backs off
-- gives up all his current locks and retries. This implies that
the entire multi-statement transaction has to be rolled back.
- Incompatible change:
FLUSH commands such as FLUSH PRIVILEGES and FLUSH TABLES WITH READ
LOCK won't cause locked tables to be implicitly unlocked anymore.
mysql-test/extra/binlog_tests/drop_table.test:
Add test case for Bug#989.
mysql-test/extra/binlog_tests/mix_innodb_myisam_binlog.test:
Fix test case to reflect the fact that transactions now hold
metadata locks for the duration of a transaction.
mysql-test/include/mix1.inc:
Fix test case to reflect the fact that transactions now hold
metadata locks for the duration of a transaction.
mysql-test/include/mix2.inc:
Fix test case to reflect the fact that transactions now hold
metadata locks for the duration of a transaction.
mysql-test/r/flush_block_commit.result:
Update test case result (WL#4284).
mysql-test/r/flush_block_commit_notembedded.result:
Update test case result (WL#4284).
mysql-test/r/innodb.result:
Update test case result (WL#4284).
mysql-test/r/innodb_mysql.result:
Update test case result (WL#4284).
mysql-test/r/lock.result:
Add test case result for an effect of WL#4284/Bug#989
(all locks should be released when a connection terminates).
mysql-test/r/mix2_myisam.result:
Update test case result (effects of WL#4284/Bug#989).
mysql-test/r/not_embedded_server.result:
Update test case result (effects of WL#4284/Bug#989).
Add a test case for interaction of WL#4284 and FLUSH PRIVILEGES.
mysql-test/r/partition_innodb_semi_consistent.result:
Update test case result (effects of WL#4284/Bug#989).
mysql-test/r/partition_sync.result:
Temporarily disable the test case for Bug#43867,
which will be fixed by a subsequent backport.
mysql-test/r/ps.result:
Add a test case for effect of PREPARE on transactional
locks: we take a savepoint at beginning of PREAPRE
and release it at the end. Thus PREPARE does not
accumulate metadata locks (Bug#989/WL#4284).
mysql-test/r/read_only_innodb.result:
Update test case result (effects of WL#4284/Bug#989).
mysql-test/suite/binlog/r/binlog_row_drop_tbl.result:
Add a test case result (WL#4284/Bug#989).
mysql-test/suite/binlog/r/binlog_row_mix_innodb_myisam.result:
Update test case result (effects of WL#4284/Bug#989).
mysql-test/suite/binlog/r/binlog_stm_drop_tbl.result:
Add a test case result (WL#4284/Bug#989).
mysql-test/suite/binlog/r/binlog_stm_mix_innodb_myisam.result:
Update test case result (effects of WL#4284/Bug#989).
mysql-test/suite/binlog/r/binlog_unsafe.result:
A side effect of Bug#989 -- slightly different table map ids.
mysql-test/suite/binlog/t/binlog_row_drop_tbl.test:
Add a test case for WL#4284/Bug#989.
mysql-test/suite/binlog/t/binlog_stm_drop_tbl.test:
Add a test case for WL#4284/Bug#989.
mysql-test/suite/binlog/t/binlog_stm_row.test:
Update to the new state name. This
is actually a follow up to another patch for WL#4284,
that changes Locked thread state to Table lock.
mysql-test/suite/ndb/r/ndb_index_ordered.result:
Remove result for disabled part of the test case.
mysql-test/suite/ndb/t/disabled.def:
Temporarily disable a test case (Bug#45621).
mysql-test/suite/ndb/t/ndb_index_ordered.test:
Disable a part of a test case (needs update to
reflect semantics of Bug#989).
mysql-test/suite/rpl/t/disabled.def:
Disable tests made meaningless by transactional metadata
locking.
mysql-test/suite/sys_vars/r/autocommit_func.result:
Add a commit (Bug#989).
mysql-test/suite/sys_vars/t/autocommit_func.test:
Add a commit (Bug#989).
mysql-test/t/flush_block_commit.test:
Fix test case to reflect the fact that transactions now hold
metadata locks for the duration of a transaction.
mysql-test/t/flush_block_commit_notembedded.test:
Fix test case to reflect the fact that transactions now hold
metadata locks for the duration of a transaction.
Add a test case for transaction-scope locks and the global
read lock (Bug#989/WL#4284).
mysql-test/t/innodb.test:
Fix test case to reflect the fact that transactions now hold
metadata locks for the duration of a transaction
(effects of Bug#989/WL#4284).
mysql-test/t/lock.test:
Add a test case for Bug#989/WL#4284.
mysql-test/t/not_embedded_server.test:
Add a test case for Bug#989/WL#4284.
mysql-test/t/partition_innodb_semi_consistent.test:
Replace TRUNCATE with DELETE, to not issue
an implicit commit of a transaction, and not depend
on metadata locks.
mysql-test/t/partition_sync.test:
Temporarily disable the test case for Bug#43867,
which needs a fix to be backported from 6.0.
mysql-test/t/ps.test:
Add a test case for semantics of PREPARE and transaction-scope
locks: metadata locks on tables used in PREPARE are enclosed into a
temporary savepoint, taken at the beginning of PREPARE,
and released at the end. Thus PREPARE does not effect
what locks a transaction owns.
mysql-test/t/read_only_innodb.test:
Fix test case to reflect the fact that transactions now hold
metadata locks for the duration of a transaction
(Bug#989/WL#4284).
Wait for the read_only statement to actually flush tables before
sending other concurrent statements that depend on its state.
mysql-test/t/xa.test:
Fix test case to reflect the fact that transactions now hold
metadata locks for the duration of a transaction
(Bug#989/WL#4284).
sql/ha_ndbcluster_binlog.cc:
Backport bits of changes of ha_ndbcluster_binlog.cc
from 6.0, to fix the failing binlog test suite with
WL#4284. WL#4284 implementation does not work
with 5.1 implementation of ndbcluster binlog index.
sql/log_event.cc:
Release metadata locks after issuing a commit.
sql/mdl.cc:
Style changes (WL#4284).
sql/mysql_priv.h:
Rename parameter to match the name used in the definition (WL#4284).
sql/rpl_injector.cc:
Release metadata locks on commit (WL#4284).
sql/rpl_rli.cc:
Remove assert made meaningless, metadata locks are released
at the end of the transaction.
sql/set_var.cc:
Close tables and release locks if autocommit mode is set.
sql/slave.cc:
Release metadata locks after a rollback.
sql/sql_acl.cc:
Don't implicitly unlock locked tables. Issue a implicit commit
at the end and unlock tables.
sql/sql_base.cc:
Defer the release of metadata locks when closing tables
if not required to.
Issue a deadlock error if the locking protocol requires
that a transaction re-acquire its locks.
Release metadata locks when closing tables for reopen.
sql/sql_class.cc:
Release metadata locks if the thread is killed.
sql/sql_parse.cc:
Release metadata locks after implicitly committing a active
transaction, or after explicit commits or rollbacks.
sql/sql_plugin.cc:
Allocate MDL request on the stack as the use of the table
is contained within the function. It will be removed from
the context once close_thread_tables is called at the end
of the function.
sql/sql_prepare.cc:
The problem is that the prepare phase of the CREATE TABLE
statement takes a exclusive metadata lock lock and this can
cause a self-deadlock the thread already holds a shared lock
on the table being that should be created.
The solution is to make the prepare phase take a shared
metadata lock when preparing a CREATE TABLE statement. The
execution of the statement will still acquire a exclusive
lock, but won't cause any problem as it issues a implicit
commit.
After some discussions with stakeholders it has been decided that
metadata locks acquired during a PREPARE statement must be released
once the statement is prepared even if it is prepared within a multi
statement transaction.
sql/sql_servers.cc:
Don't implicitly unlock locked tables. Issue a implicit commit
at the end and unlock tables.
sql/sql_table.cc:
Close table and release metadata locks after a admin operation.
sql/table.h:
The problem is that the prepare phase of the CREATE TABLE
statement takes a exclusive metadata lock lock and this can
cause a self-deadlock the thread already holds a shared lock
on the table being that should be created.
The solution is to make the prepare phase take a shared
metadata lock when preparing a CREATE TABLE statement. The
execution of the statement will still acquire a exclusive
lock, but won't cause any problem as it issues a implicit
commit.
sql/transaction.cc:
Release metadata locks after the implicitly committed due
to a new transaction being started. Also, release metadata
locks acquired after a savepoint if the transaction is rolled
back to the save point.
The problem is that in some cases transaction-long metadata locks
could be released before the transaction was committed. This could
happen when a active transaction was ended by a "START TRANSACTION"
or "BEGIN" statement, in which case the metadata locks would be
released before the actual commit of the active transaction.
The solution is to defer the release of metadata locks to after the
transaction has been implicitly committed. No test case is provided
as the effort to provide one is too disproportional to the size of
the fix.
2617.31.12, 2617.31.15, 2617.31.15, 2617.31.16, 2617.43.1
- initial changeset that introduced the fix for
Bug#989 and follow up fixes for all test suite failures
introduced in the initial changeset.
------------------------------------------------------------
revno: 2617.31.1
committer: Davi Arnaut <Davi.Arnaut@Sun.COM>
branch nick: 4284-6.0
timestamp: Fri 2009-03-06 19:17:00 -0300
message:
Bug#989: If DROP TABLE while there's an active transaction, wrong binlog order
WL#4284: Transactional DDL locking
Currently the MySQL server does not keep metadata locks on
schema objects for the duration of a transaction, thus failing
to guarantee the integrity of the schema objects being used
during the transaction and to protect then from concurrent
DDL operations. This also poses a problem for replication as
a DDL operation might be replicated even thought there are
active transactions using the object being modified.
The solution is to defer the release of metadata locks until
a active transaction is either committed or rolled back. This
prevents other statements from modifying the table for the
entire duration of the transaction. This provides commitment
ordering for guaranteeing serializability across multiple
transactions.
- Incompatible change:
If MySQL's metadata locking system encounters a lock conflict,
the usual schema is to use the try and back-off technique to
avoid deadlocks -- this schema consists in releasing all locks
and trying to acquire them all in one go.
But in a transactional context this algorithm can't be utilized
as its not possible to release locks acquired during the course
of the transaction without breaking the transaction commitments.
To avoid deadlocks in this case, the ER_LOCK_DEADLOCK will be
returned if a lock conflict is encountered during a transaction.
Let's consider an example:
A transaction has two statements that modify table t1, then table
t2, and then commits. The first statement of the transaction will
acquire a shared metadata lock on table t1, and it will be kept
utill COMMIT to ensure serializability.
At the moment when the second statement attempts to acquire a
shared metadata lock on t2, a concurrent ALTER or DROP statement
might have locked t2 exclusively. The prescription of the current
locking protocol is that the acquirer of the shared lock backs off
-- gives up all his current locks and retries. This implies that
the entire multi-statement transaction has to be rolled back.
- Incompatible change:
FLUSH commands such as FLUSH PRIVILEGES and FLUSH TABLES WITH READ
LOCK won't cause locked tables to be implicitly unlocked anymore.
----------------------------------------------------------
revno: 2617.23.20
committer: Konstantin Osipov <kostja@sun.com>
branch nick: mysql-6.0-runtime
timestamp: Wed 2009-03-04 16:31:31 +0300
message:
WL#4284 "Transactional DDL locking"
Review comments: "Objectify" the MDL API.
MDL_request and MDL_context still need manual construction and
destruction, since they are used in environment that is averse
to constructors/destructors.
sql/mdl.cc:
Improve comments.
Add asserts to backup()/restore_from_backup()/merge() methods.
Fix an order bug in the error path of mdl_acquire_exclusive_locks():
we used to first free a ticket object, and only then exclude
it from the list of tickets.
----------------------------------------------------------
revno: 2617.23.20
committer: Konstantin Osipov <kostja@sun.com>
branch nick: mysql-6.0-runtime
timestamp: Wed 2009-03-04 16:31:31 +0300
message:
WL#4284 "Transactional DDL locking"
Review comments: "Objectify" the MDL API.
MDL_request and MDL_context still need manual construction and
destruction, since they are used in environment that is averse
to constructors/destructors.
------------------------------------------------------------
revno: 2617.23.18
committer: Davi Arnaut <Davi.Arnaut@Sun.COM>
branch nick: 4284-6.0
timestamp: Mon 2009-03-02 18:18:26 -0300
message:
Bug#989: If DROP TABLE while there's an active transaction, wrong binlog order
WL#4284: Transactional DDL locking
This is a prerequisite patch:
These changes are intended to split lock requests from granted
locks and to allow the memory and lifetime of granted locks to
be managed within the MDL subsystem. Furthermore, tickets can
now be shared and therefore are used to satisfy multiple lock
requests, but only shared locks can be recursive.
The problem is that the MDL subsystem morphs lock requests into
granted locks locks but does not manage the memory and lifetime
of lock requests, and hence, does not manage the memory of
granted locks either. This can be problematic because it puts the
burden of tracking references on the users of the subsystem and
it can't be easily done in transactional contexts where the locks
have to be kept around for the duration of a transaction.
Another issue is that recursive locks (when the context trying to
acquire a lock already holds a lock on the same object) requires
that each time the lock is granted, a unique lock request/granted
lock structure structure must be kept around until the lock is
released. This can lead to memory leaks in transactional contexts
as locks taken during the transaction should only be released at
the end of the transaction. This also leads to unnecessary wake
ups (broadcasts) in the MDL subsystem if the context still holds
a equivalent of the lock being released.
These issues are exacerbated due to the fact that WL#4284 low-level
design says that the implementation should "2) Store metadata locks
in transaction memory root, rather than statement memory root" but
this is not possible because a memory root, as implemented in mysys,
requires all objects allocated from it to be freed all at once.
This patch combines review input and significant code contributions
from Konstantin Osipov (kostja) and Dmitri Lenev (dlenev).
mysql-test/r/mdl_sync.result:
Add test case result.
mysql-test/t/mdl_sync.test:
Add test case for shared lock upgrade case.
sql/event_db_repository.cc:
Rename mdl_alloc_lock to mdl_request_alloc.
sql/ha_ndbcluster_binlog.cc:
Use new function names to initialize MDL lock requests.
sql/lock.cc:
Rename MDL functions.
sql/log_event.cc:
The MDL request now holds the table and database name data (MDL_KEY).
sql/mdl.cc:
Move the MDL key to the MDL_LOCK structure in order to make the
object suitable for allocation from a fixed-size allocator. This
allows the simplification of the lists in the MDL_LOCK object,
which now are just two, one for granted tickets and other for
waiting (upgraders) tickets.
Recursive requests for a shared lock on the same object can now
be granted using the same lock ticket. This schema is only used
for shared locks because that the only case that matters. This
is used to avoid waste of resources in case a context (connection)
already holds a shared lock on a object.
sql/mdl.h:
Introduce a metadata lock object key which is used to uniquely
identify lock objects.
Separate the structure used to represent pending lock requests
from the structure used to represent granted metadata locks.
Rename functions used to manipulate locks requests in order to
have a more consistent function naming schema.
sql/sp_head.cc:
Rename mdl_alloc_lock to mdl_request_alloc.
sql/sql_acl.cc:
Rename alloc_mdl_locks to alloc_mdl_requests.
sql/sql_base.cc:
Various changes to accommodate that lock requests are separated
from lock tickets (granted locks).
sql/sql_class.h:
Last acquired lock before the savepoint was set.
sql/sql_delete.cc:
Various changes to accommodate that lock requests are separated
from lock tickets (granted locks).
sql/sql_handler.cc:
Various changes to accommodate that lock requests are separated
from lock tickets (granted locks).
sql/sql_insert.cc:
Rename alloc_mdl_locks to alloc_mdl_requests.
sql/sql_parse.cc:
Rename alloc_mdl_locks to alloc_mdl_requests.
sql/sql_plist.h:
Typedef for iterator type.
sql/sql_plugin.cc:
Rename alloc_mdl_locks to alloc_mdl_requests.
sql/sql_servers.cc:
Rename alloc_mdl_locks to alloc_mdl_requests.
sql/sql_show.cc:
Various changes to accommodate that lock requests are separated
from lock tickets (granted locks).
sql/sql_table.cc:
Various changes to accommodate that lock requests are separated
from lock tickets (granted locks).
sql/sql_trigger.cc:
Save reference to the lock ticket so it can be downgraded later.
sql/sql_udf.cc:
Rename alloc_mdl_locks to alloc_mdl_requests.
sql/table.cc:
Rename mdl_alloc_lock to mdl_request_alloc.
sql/table.h:
Separate MDL lock requests from lock tickets (granted locks).
storage/myisammrg/ha_myisammrg.cc:
Rename alloc_mdl_locks to alloc_mdl_requests.
