Relax memory barrier for lock_word.
rw_lock_lock_word_decr() - used to acquire rw-lock, thus we only need to issue
ACQUIRE when we succeed locking.
rw_lock_x_lock_func_nowait() - same as above, but used to attempt to acquire
X-lock.
rw_lock_s_unlock_func() - used to release S-lock, RELEASE is what we need here.
rw_lock_x_unlock_func() - used to release X-lock. Ideally we'd need only RELEASE
here, but due to mess with waiters (they must be loaded after lock_word is
stored) we have to issue both ACQUIRE and RELEASE.
rw_lock_sx_unlock_func() - same as above, but used to release SX-lock.
rw_lock_s_lock_spin(), rw_lock_x_lock_func(), rw_lock_sx_lock_func() -
fetch-and-store to waiters has to issue only ACQUIRE memory barrier, so that
waiters are stored before lock_word is loaded.
Note that there is violation of RELEASE-ACQUIRE protocol here, because we do
on lock:
my_atomic_fas32_explicit((int32*) &lock->waiters, 1, MY_MEMORY_ORDER_ACQUIRE);
my_atomic_load32_explicit(&lock->lock_word, MY_MEMORY_ORDER_RELAXED);
on unlock
my_atomic_add32_explicit(&lock->lock_word, X_LOCK_DECR, MY_MEMORY_ORDER_ACQ_REL);
my_atomic_load32_explicit((int32*) &lock->waiters, MY_MEMORY_ORDER_RELAXED);
That is we kind of synchronize ACQUIRE on lock_word with ACQUIRE on waiters.
It was there before this patch. Simple fix may have negative performance impact.
Proper fix requires refactoring of lock_word.
Relax memory barrier for waiters: these 2 stores must be completed before
os_event_set() finishes. This is guaranteed by RELEASE barrier issued by
mutex.exit() of os_event_set().
Remove volatile modifier from waiters: it's not supposed for inter-thread
communication, use appropriate atomic operations instead.
Changed waiters to int32_t, my_atomic friendly type.
Define my_thread_id as an unsigned type, to avoid mismatch with
ulonglong. Change some parameters to this type.
Use size_t in a few more places.
Declare many flag constants as unsigned to avoid sign mismatch
when shifting bits or applying the unary ~ operator.
When applying the unary ~ operator to enum constants, explictly
cast the result to an unsigned type, because enum constants can
be treated as signed.
In InnoDB, change the source code line number parameters from
ulint to unsigned type. Also, make some InnoDB functions return
a narrower type (unsigned or uint32_t instead of ulint;
bool instead of ibool).
MDEV-7618 introduced configuration parameter innodb_instrument_semaphores
in MariaDB Server 10.1. The parameter seems to only affect the rw-lock
X-latch acquisition. Extra fields are added to rw_lock_t to remember one
X-latch holder or waiter. These fields are not being consulted or reported
anywhere. This is basically only adding code bloat.
If the intention is to debug hangs or deadlocks, we have better tools for
that in the debug server, and for the non-debug server, core dumps can
reveal a lot. For example, the mini-transaction memo records the
currently held buffer block or index rw-locks, to be released at
mtr_t::commit().
The configuration parameter innodb_instrument_semaphores will be
deprecated in 10.2.5 and removed in 10.3.0.
rw_lock_t: Remove the members lock_name, file_name, line, thread_id
which did not affect any output.
Simplify away recursive flag: it is not necessary for rw-locks to operate
properly. Now writer_thread != 0 means recursive.
As we only need correct value of writer_thread only in writer_thread itself
it is rather safe to load and update it non-atomically.
This patch also fixes potential reorder of "writer_thread" and "recursive"
loads (aka MDEV-7148), which was reopened along with InnoDB thread fences
simplification. Previous versions are unaffected, because they have os_rmb
in rw_lock_lock_word_decr(). It wasn't observed at the moment of writing
though.
No point to issue RELEASE memory barrier in os_thread_create_func(): thread
creation is full memory barrier.
No point to issue os_wmb in rw_lock_set_waiter_flag() and
rw_lock_reset_waiter_flag(): this is deadcode and it is unlikely operational
anyway. If atomic builtins are unavailable - memory barriers are most certainly
unavailable too.
RELEASE memory barrier is definitely abused in buf_pool_withdraw_blocks(): most
probably it was supposed to commit volatile variable update, which is not what
memory barriers actually do. To operate properly it needs corresponding ACQUIRE
barrier without an associated atomic operation anyway.
ACQUIRE memory barrier is definitely abused in log_write_up_to(): most probably
it was supposed to synchronize dirty read of log_sys->write_lsn. To operate
properly it needs corresponding RELEASE barrier without an associated atomic
operation anyway.
Removed a bunch of ACQUIRE memory barriers from InnoDB rwlocks. They're
meaningless without corresponding RELEASE memory barriers.
Valid usage example of memory barriers without an associated atomic operation:
http://en.cppreference.com/w/cpp/atomic/atomic_thread_fence
Replaced InnoDB atomic operations with server atomic operations.
Moved INNODB_RW_LOCKS_USE_ATOMICS - it is always defined (code won't compile
otherwise).
NOTE: InnoDB uses thread identifiers as a target for atomic operations.
Thread identifiers should be considered opaque: any attempt to use a
thread ID other than in pthreads calls is nonportable and can lead to
unspecified results.
Contains also
MDEV-10547: Test multi_update_innodb fails with InnoDB 5.7
The failure happened because 5.7 has changed the signature of
the bool handler::primary_key_is_clustered() const
virtual function ("const" was added). InnoDB was using the old
signature which caused the function not to be used.
MDEV-10550: Parallel replication lock waits/deadlock handling does not work with InnoDB 5.7
Fixed mutexing problem on lock_trx_handle_wait. Note that
rpl_parallel and rpl_optimistic_parallel tests still
fail.
MDEV-10156 : Group commit tests fail on 10.2 InnoDB (branch bb-10.2-jan)
Reason: incorrect merge
MDEV-10550: Parallel replication can't sync with master in InnoDB 5.7 (branch bb-10.2-jan)
Reason: incorrect merge
Re-applied lost in the merge revision:
commit ed313e8a92
Author: Sergey Vojtovich <svoj@mariadb.org>
Date: Mon Dec 1 14:58:29 2014 +0400
MDEV-7148 - Recurring: InnoDB: Failing assertion: !lock->recursive
On PPC64 high-loaded server may crash due to assertion failure in InnoDB
rwlocks code.
This happened because load order between "recursive" and "writer_thread"
wasn't properly enforced.
MDEV-7399: Add support for INFORMATION_SCHEMA.INNODB_MUTEXES
MDEV-7618: Improve semaphore instrumentation
Introduced two new information schema tables to monitor mutex waits
and semaphore waits. Added a new configuration variable
innodb_intrument_semaphores to add thread_id, file name and
line of current holder of mutex/rw_lock.
Part of this work is based on Stewart Smitch's memory barrier and lower priori
patches for power8.
- Added memory syncronization for innodb & xtradb for power8.
- Added HAVE_WINDOWS_MM_FENCE to CMakeList.txt
- Added os_isync to fix a syncronization problem on power
- Added log_get_lsn_nowait which is now used srv_error_monitor_thread to ensur
if log mutex is locked.
All changes done both for InnoDB and Xtradb
storage/innobase/include/sync0rw.ic:
Prerequisite for compiling with gcc4 on solaris: ignore result from
os_compare_and_swap_ulint
storage/myisam/mi_dynrec.c:
Prerequisite for compiling with gcc4 on solaris: cast to void*
rw_lock_x_lock_func(): Assert that the thread is not already holding
the lock in a conflicting mode (RW_LOCK_SHARED).
rw_lock_s_lock_func(): Assert that the thread is not already holding
the lock in a conflicting mode (RW_LOCK_EX).
This change is for performance optimization.
Fixed the performance schema instrumentation interface as follows:
- simplified mysql_unlock_mutex()
- simplified mysql_unlock_rwlock()
- simplified mysql_cond_signal()
- simplified mysql_cond_broadcast()
Changed the get_thread_XXX_locker apis to have one extra parameter,
to provide memory to the instrumentation implementation.
This API change allows to use memory provided by the caller,
to avoid having to use thread local storage.
Using this extra parameter will be done in a separate fix,
this change is for the interface only.
Adjusted all the code and unit tests accordingly.
layout as we always had in trees containing only the builtin
2) win\configure.js WITH_INNOBASE_STORAGE_ENGINE still works.
storage/innobase/CMakeLists.txt:
fix to new directory name (and like 5.1)
storage/innobase/Makefile.am:
fix to new directory name (and like 5.1)
storage/innobase/handler/ha_innodb.cc:
fix to new directory name (and like 5.1)
storage/innobase/plug.in:
fix to new directory name (and like 5.1)
