and 'THREAD_SAFE_CLIENT'.
As of MySQL 5.5, we no longer support non-threaded
builds. This patch removes all references to the
obsolete THREAD and THREAD_SAFE_CLIENT preprocessor
symbols. These were used to distinguish between
threaded and non-threaded builds.
detector". This patch addresses performance regression in OLTP_RO/MyISAM
test on Windows introduced by the fix for bug #56405. Thus it makes
original patch acceptable as a solution for bug #56585 "Slowdown of
readonly sysbench benchmarks (e.g point_select) on Windows 5.5".
With this patch, MySQL will use native Windows condition variables and
reader-writer locks if they are supported by the OS.
This speeds up MyISAM and the effect comes mostly from using native
rwlocks. Native conditions improve scalability with higher number of
concurrent users in other situations, e.g for prlocks.
Benchmark numbers for this patch as measured on Win2008R2 quad
core machine are attached to the bug report.
( direct link http://bugs.mysql.com/file.php?id=15883 )
Note that currently we require at least Windows7/WS2008R2 for
reader-writer locks, even though native rwlock is available also on Vista.
Reason is that "trylock" APIs are missing on Vista, and trylock is used in
the server (in a single place in query cache).
While this patch could have been written differently, to enable the native
rwlock optimization also on Vista/WS2008 (e.g using native locks everywhere
but portable implementation in query cache), this would come at the
expense of the code clarity, as it would introduce a new "try-able" rwlock
type, to handle Vista case.
Another way to improve performance for the special case
(OLTP_RO/MYISAM/Vista) would be to eliminate "trylock" usage from server,
but this is outside of the scope here.
Native conditions variables are used beginning with Vista though the effect
of using condition variables alone is not measurable in this benchmark.
But when used together with native rwlocks on Win7, native conditions improve
performance in high-concurrency OLTP_RO/MyISAM (128 and more sysbench
users).
detector" that doesn't introduce bug #56715 "Concurrent
transactions + FLUSH result in sporadical unwarranted
deadlock errors".
Deadlock could have occurred when workload containing a mix
of DML, DDL and FLUSH TABLES statements affecting the same
set of tables was executed in a heavily concurrent environment.
This deadlock occurred when several connections tried to
perform deadlock detection in the metadata locking subsystem.
The first connection started traversing wait-for graph,
encountered a sub-graph representing a wait for flush, acquired
LOCK_open and dived into sub-graph inspection. Then it
encountered sub-graph corresponding to wait for metadata lock
and blocked while trying to acquire a rd-lock on
MDL_lock::m_rwlock, since some,other thread had a wr-lock on it.
When this wr-lock was released it could have happened (if there
was another pending wr-lock against this rwlock) that the rd-lock
from the first connection was left unsatisfied but at the same
time the new rd-lock request from the second connection sneaked
in and was satisfied (for this to be possible the second
rd-request should come exactly after the wr-lock is released but
before pending the wr-lock manages to grab rwlock, which is
possible both on Linux and in our own rwlock implementation).
If this second connection continued traversing the wait-for graph
and encountered a sub-graph representing a wait for flush it tried
to acquire LOCK_open and thus the deadlock was created.
The previous patch tried to workaround this problem by not
allowing the deadlock detector to lock LOCK_open mutex if
some other thread doing deadlock detection already owns it
and current search depth is greater than 0. Instead deadlock
was reported. As a result it has introduced bug #56715.
This patch solves this problem in a different way.
It introduces a new rw_pr_lock_t implementation to be used
by MDL subsystem instead of one based on Linux rwlocks or
our own rwlock implementation. This new implementation
never allows situation in which an rwlock is rd-locked and
there is a blocked pending rd-lock. Thus the situation which
has caused this bug becomes impossible with this implementation.
Due to fact that this implementation is optimized for
wr-lock/unlock scenario which is most common in the MDL
subsystem it doesn't introduce noticeable performance
regressions in sysbench tests. Moreover it significantly
improves situation for POINT_SELECT test when many
connections are used.
No test case is provided as this bug is very hard to repeat
in MTR environment but is repeatable with the help of RQG
tests.
This patch also doesn't include a test for bug #56715
"Concurrent transactions + FLUSH result in sporadical
unwarranted deadlock errors" as it takes too much time to
be run as part of normal test-suite runs.
Fix warnings flagged by the new warning option -Wunused-but-set-variable
that was added to GCC 4.6 and that is enabled by -Wunused and -Wall. The
option causes a warning whenever a local variable is assigned to but is
later unused. It also warns about meaningless pointer dereferences.
The atomic operations implementation on 5.1 has a few problems,
which might cause tests to abort randomly. Since no code in 5.1
uses atomic operations, simply remove the code.
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
on Windows".
On platforms where read-write lock implementation does not
prefer readers by default (Windows, Solaris) server might
have deadlocked while detecting MDL deadlock.
MDL deadlock detector relies on the fact that read-write
locks which are used in its implementation prefer readers
(see new comment for MDL_lock::m_rwlock for details).
So far MDL code assumed that default implementation of
read/write locks for the system has this property.
Indeed, this turned out ot be wrong, for example, for
Windows or Solaris. Thus MDL deadlock detector might have
deadlocked on these systems.
This fix simply adds portable implementation of read/write
lock which prefer readers and changes MDL code to use this
new type of synchronization primitive.
No test case is added as existing rqg_mdl_stability test can
serve as one.
This patch introduces timeouts for metadata locks.
The timeout is specified in seconds using the new dynamic system
variable "lock_wait_timeout" which has both GLOBAL and SESSION
scopes. Allowed values range from 1 to 31536000 seconds (= 1 year).
The default value is 1 year.
The new server parameter "lock-wait-timeout" can be used to set
the default value parameter upon server startup.
"lock_wait_timeout" applies to all statements that use metadata locks.
These include DML and DDL operations on tables, views, stored procedures
and stored functions. They also include LOCK TABLES, FLUSH TABLES WITH
READ LOCK and HANDLER statements.
The patch also changes thr_lock.c code (table data locks used by MyISAM
and other simplistic engines) to use the same system variable.
InnoDB row locks are unaffected.
One exception to the handling of the "lock_wait_timeout" variable
is delayed inserts. All delayed inserts are executed with a timeout
of 1 year regardless of the setting for the global variable. As the
connection issuing the delayed insert gets no notification of
delayed insert timeouts, we want to avoid unnecessary timeouts.
It's important to note that the timeout value is used for each lock
acquired and that one statement can take more than one lock.
A statement can therefore block for longer than the lock_wait_timeout
value before reporting a timeout error. When lock timeout occurs,
ER_LOCK_WAIT_TIMEOUT is reported.
Test case added to lock_multi.test.
Bug#16565 mysqld --help --verbose does not order variablesBug#20413 sql_slave_skip_counter is not shown in show variables
Bug#20415 Output of mysqld --help --verbose is incomplete
Bug#25430 variable not found in SELECT @@global.ft_max_word_len;
Bug#32902 plugin variables don't know their names
Bug#34599 MySQLD Option and Variable Reference need to be consistent in formatting!
Bug#34829 No default value for variable and setting default does not raise error
Bug#34834 ? Is accepted as a valid sql mode
Bug#34878 Few variables have default value according to documentation but error occurs
Bug#34883 ft_boolean_syntax cant be assigned from user variable to global var.
Bug#37187 `INFORMATION_SCHEMA`.`GLOBAL_VARIABLES`: inconsistent status
Bug#40988 log_output_basic.test succeeded though syntactically false.
Bug#41010 enum-style command-line options are not honoured (maria.maria-recover fails)
Bug#42103 Setting key_buffer_size to a negative value may lead to very large allocations
Bug#44691 Some plugins configured as MYSQL_PLUGIN_MANDATORY in can be disabled
Bug#44797 plugins w/o command-line options have no disabling option in --help
Bug#46314 string system variables don't support expressions
Bug#46470 sys_vars.max_binlog_cache_size_basic_32 is broken
Bug#46586 When using the plugin interface the type "set" for options caused a crash.
Bug#47212 Crash in DBUG_PRINT in mysqltest.cc when trying to print octal number
Bug#48758 mysqltest crashes on sys_vars.collation_server_basic in gcov builds
Bug#49417 some complaints about mysqld --help --verbose output
Bug#49540 DEFAULT value of binlog_format isn't the default value
Bug#49640 ambiguous option '--skip-skip-myisam' (double skip prefix)
Bug#49644 init_connect and \0
Bug#49645 init_slave and multi-byte characters
Bug#49646 mysql --show-warnings crashes when server dies
As documented in the bug report, the double checked locking
pattern has inherent issues, and cannot guarantee correct
initialization.
This patch replaces the logic in init_available_charsets()
with the use of pthread_once(3). A wrapper function,
my_pthread_once(), is introduced and is used in lieu of direct
calls to init_available_charsets(). Related defines
MY_PTHREAD_ONCE_* are also introduced.
For the Windows platform, the implementation in lp:sysbench is
ported. For single-thread use, a simple define calls the
function and sets the pthread_once control variable.
Charset initialization is modified to use my_pthread_once().
-------------------------------------------------------------
revno: 2877
committer: Davi Arnaut <Davi.Arnaut@Sun.COM>
branch nick: 35164-6.0
timestamp: Wed 2008-10-15 19:53:18 -0300
message:
Bug#35164: Large number of invalid pthread_attr_setschedparam calls
Bug#37536: Thread scheduling causes performance degradation at low thread count
Bug#12702: Long queries take 100% of CPU and freeze other applications under Windows
The problem is that although having threads with different priorities
yields marginal improvements [1] in some platforms [2], relying on some
statically defined priorities (QUERY_PRIOR and WAIT_PRIOR) to play well
(or to work at all) with different scheduling practices and disciplines
is, at best, a shot in the dark as the meaning of priority values may
change depending on the scheduling policy set for the process.
Another problem is that increasing priorities can hurt other concurrent
(running on the same hardware) applications (such as AMP) by causing
starvation problems as MySQL threads will successively preempt lower
priority processes. This can be evidenced by Bug#12702.
The solution is to not change the threads priorities and rely on the
system scheduler to perform its job. This also enables a system admin
to increase or decrease the scheduling priority of the MySQL process,
if intended.
Furthermore, the internal wrappers and code for changing the priority
of threads is being removed as they are now unused and ancient.
1. Due to unintentional side effects. On Solaris this could artificially
help benchmarks as calling the priority changing syscall millions of
times is more beneficial than the actual setting of the priority.
2. Where it actually works. It has never worked on Linux as the default
scheduling policy SCHED_OTHER only accepts the static priority 0.
http://lists.mysql.com/commits/59686
Cleanup pthread_self(), pthread_create(), pthread_join() implementation on Windows.
Prior implementation is was unnecessarily complicated and even differs in embedded
and non-embedded case.
Improvements in this patch:
* pthread_t is now the unique thread ID, instead of HANDLE returned by beginthread
This simplifies pthread_self() to be just straight GetCurrentThreadId().
prior it was much art involved in passing the beginthread() handle from the caller
to the TLS structure in the child thread ( did not work for the main thread of
course)
* remove MySQL specific my_thread_init()/my_thread_end() from pthread_create.
No automagic is done on Unix on pthread_create(). Having the same on Windows will
improve portability and avoid extra #ifdef's
* remove redefinition of getpid() - it was defined as GetCurrentThreadId()
htttp://lists.mysql.com/commits/50957?f=plain
Always use TLS functions instead of __declspec(thread) to access
thread local storage variables.
The change removes the necessity to recomplile the same source
files twice - with USE_TLS for DLLs and without USE_TLS for EXEs.
Real benefit of this change is better readability and maintainability
of TLS functions within MySQL.
There is a performance loss using TlsXXX functions compared to __declspec
but the difference is negligible in practice. In a sysbench-like benchmark
I ran with with TlsGetValue, pthread_[get|set]_specific was called 600000000
times and took 0.17sec of total 35min CPU time, or 0.008%.
Improved data typing of server variables in InnoDB to avoid bugs on
Windows.
Workaround for bug in gcc assembler on 32-bit Mac OS X as well as
on Solaris.
The problem is that the function used by the server to increase
the thread's priority (pthread_setschedparam) has the unintended
side-effect of changing the calling thread scheduling policy,
possibly overwriting a scheduling policy set by a sysadmin.
The solution is to rely on the pthread_setschedprio function, if
available, as it only changes the scheduling priority and does not
change the scheduling policy. This function is usually available on
Solaris and Linux, but it use won't work by default on Linux as the
the default scheduling policy only accepts a static priority 0 -- this
is acceptable for now as priority changing on Linux is broken anyway.
The problem is that MySQL's 'fast' mutex implementation uses the
random() routine to determine the spin delay. Unfortunately, the
routine interface is not thead-safe and some implementations (eg:
glibc) might use a internal lock to protect the RNG state, causing
excessive locking contention if lots of threads are spinning on
a MySQL's 'fast' mutex. The code was also misusing the value
of the RAND_MAX macro, this macro represents the largest value
that can be returned from the rand() function, not random().
The solution is to use the quite simple Park-Miller random number
generator. The initial seed is set to 1 because the previously used
generator wasn't being seeded -- the initial seed is 1 if srandom()
is not called.
Futhermore, the 'fast' mutex implementation has several shortcomings
and provides no measurable performance benefit. Therefore, its use is
not recommended unless it provides directly measurable results.
The problem is that unimplemented WIN32 version of pthread_kill
is returning ESRCH no matter the arguments, causing calls to
mysqld_list_processes to set the procinfo to dead because
pthread_kill returns non zero. The dead procinfo would show
up on a second invocation of show processlist.
When locking a "fast" mutex a static variable cpu_count
was used as a flag to initialize itself on the first usage
by calling sysconf() and setting non-zero value.
This is not thread and optimization safe on some
platforms. That's why the global initialization needs
to be done once in a designated function.
This will also speed up the usage (by a small bit)
because it won't have to check if it's initialized on
every call.
Fixed by moving the fast mutexes initialization out of
my_pthread_fastmutex_lock() to fastmutex_global_init()
and call it from my_init()
