2014-02-26 19:11:54 +01:00
|
|
|
/*****************************************************************************
|
|
|
|
|
|
2017-01-04 14:34:38 +05:30
|
|
|
Copyright (c) 1995, 2017, Oracle and/or its affiliates. All Rights Reserved.
|
2022-01-28 08:31:57 +02:00
|
|
|
Copyright (c) 2013, 2022, MariaDB Corporation.
|
2014-02-26 19:11:54 +01:00
|
|
|
|
|
|
|
|
This program is free software; you can redistribute it and/or modify it under
|
|
|
|
|
the terms of the GNU General Public License as published by the Free Software
|
|
|
|
|
Foundation; version 2 of the License.
|
|
|
|
|
|
|
|
|
|
This program is distributed in the hope that it will be useful, but WITHOUT
|
|
|
|
|
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
|
|
|
|
|
FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
|
|
|
|
|
|
|
|
|
|
You should have received a copy of the GNU General Public License along with
|
|
|
|
|
this program; if not, write to the Free Software Foundation, Inc.,
|
2019-05-11 19:25:02 +03:00
|
|
|
51 Franklin Street, Fifth Floor, Boston, MA 02110-1335 USA
|
2014-02-26 19:11:54 +01:00
|
|
|
|
|
|
|
|
*****************************************************************************/
|
|
|
|
|
|
|
|
|
|
/**************************************************//**
|
|
|
|
|
@file include/fil0fil.h
|
|
|
|
|
The low-level file system
|
|
|
|
|
|
|
|
|
|
Created 10/25/1995 Heikki Tuuri
|
|
|
|
|
*******************************************************/
|
|
|
|
|
|
|
|
|
|
#ifndef fil0fil_h
|
|
|
|
|
#define fil0fil_h
|
|
|
|
|
|
2019-02-06 19:50:11 +02:00
|
|
|
#include "fsp0types.h"
|
2020-05-07 17:15:34 +03:00
|
|
|
#include "mach0data.h"
|
|
|
|
|
#include "assume_aligned.h"
|
2018-11-30 21:48:45 +02:00
|
|
|
|
2014-02-26 19:11:54 +01:00
|
|
|
#ifndef UNIV_INNOCHECKSUM
|
|
|
|
|
|
2020-11-20 12:30:55 +02:00
|
|
|
#include "srw_lock.h"
|
MDEV-23399: Performance regression with write workloads
The buffer pool refactoring in MDEV-15053 and MDEV-22871 shifted
the performance bottleneck to the page flushing.
The configuration parameters will be changed as follows:
innodb_lru_flush_size=32 (new: how many pages to flush on LRU eviction)
innodb_lru_scan_depth=1536 (old: 1024)
innodb_max_dirty_pages_pct=90 (old: 75)
innodb_max_dirty_pages_pct_lwm=75 (old: 0)
Note: The parameter innodb_lru_scan_depth will only affect LRU
eviction of buffer pool pages when a new page is being allocated. The
page cleaner thread will no longer evict any pages. It used to
guarantee that some pages will remain free in the buffer pool. Now, we
perform that eviction 'on demand' in buf_LRU_get_free_block().
The parameter innodb_lru_scan_depth(srv_LRU_scan_depth) is used as follows:
* When the buffer pool is being shrunk in buf_pool_t::withdraw_blocks()
* As a buf_pool.free limit in buf_LRU_list_batch() for terminating
the flushing that is initiated e.g., by buf_LRU_get_free_block()
The parameter also used to serve as an initial limit for unzip_LRU
eviction (evicting uncompressed page frames while retaining
ROW_FORMAT=COMPRESSED pages), but now we will use a hard-coded limit
of 100 or unlimited for invoking buf_LRU_scan_and_free_block().
The status variables will be changed as follows:
innodb_buffer_pool_pages_flushed: This includes also the count of
innodb_buffer_pool_pages_LRU_flushed and should work reliably,
updated one by one in buf_flush_page() to give more real-time
statistics. The function buf_flush_stats(), which we are removing,
was not called in every code path. For both counters, we will use
regular variables that are incremented in a critical section of
buf_pool.mutex. Note that show_innodb_vars() directly links to the
variables, and reads of the counters will *not* be protected by
buf_pool.mutex, so you cannot get a consistent snapshot of both variables.
The following INFORMATION_SCHEMA.INNODB_METRICS counters will be
removed, because the page cleaner no longer deals with writing or
evicting least recently used pages, and because the single-page writes
have been removed:
* buffer_LRU_batch_flush_avg_time_slot
* buffer_LRU_batch_flush_avg_time_thread
* buffer_LRU_batch_flush_avg_time_est
* buffer_LRU_batch_flush_avg_pass
* buffer_LRU_single_flush_scanned
* buffer_LRU_single_flush_num_scan
* buffer_LRU_single_flush_scanned_per_call
When moving to a single buffer pool instance in MDEV-15058, we missed
some opportunity to simplify the buf_flush_page_cleaner thread. It was
unnecessarily using a mutex and some complex data structures, even
though we always have a single page cleaner thread.
Furthermore, the buf_flush_page_cleaner thread had separate 'recovery'
and 'shutdown' modes where it was waiting to be triggered by some
other thread, adding unnecessary latency and potential for hangs in
relatively rarely executed startup or shutdown code.
The page cleaner was also running two kinds of batches in an
interleaved fashion: "LRU flush" (writing out some least recently used
pages and evicting them on write completion) and the normal batches
that aim to increase the MIN(oldest_modification) in the buffer pool,
to help the log checkpoint advance.
The buf_pool.flush_list flushing was being blocked by
buf_block_t::lock for no good reason. Furthermore, if the FIL_PAGE_LSN
of a page is ahead of log_sys.get_flushed_lsn(), that is, what has
been persistently written to the redo log, we would trigger a log
flush and then resume the page flushing. This would unnecessarily
limit the performance of the page cleaner thread and trigger the
infamous messages "InnoDB: page_cleaner: 1000ms intended loop took 4450ms.
The settings might not be optimal" that were suppressed in
commit d1ab89037a518fcffbc50c24e4bd94e4ec33aed0 unless log_warnings>2.
Our revised algorithm will make log_sys.get_flushed_lsn() advance at
the start of buf_flush_lists(), and then execute a 'best effort' to
write out all pages. The flush batches will skip pages that were modified
since the log was written, or are are currently exclusively locked.
The MDEV-13670 message "page_cleaner: 1000ms intended loop took" message
will be removed, because by design, the buf_flush_page_cleaner() should
not be blocked during a batch for extended periods of time.
We will remove the single-page flushing altogether. Related to this,
the debug parameter innodb_doublewrite_batch_size will be removed,
because all of the doublewrite buffer will be used for flushing
batches. If a page needs to be evicted from the buffer pool and all
100 least recently used pages in the buffer pool have unflushed
changes, buf_LRU_get_free_block() will execute buf_flush_lists() to
write out and evict innodb_lru_flush_size pages. At most one thread
will execute buf_flush_lists() in buf_LRU_get_free_block(); other
threads will wait for that LRU flushing batch to finish.
To improve concurrency, we will replace the InnoDB ib_mutex_t and
os_event_t native mutexes and condition variables in this area of code.
Most notably, this means that the buffer pool mutex (buf_pool.mutex)
is no longer instrumented via any InnoDB interfaces. It will continue
to be instrumented via PERFORMANCE_SCHEMA.
For now, both buf_pool.flush_list_mutex and buf_pool.mutex will be
declared with MY_MUTEX_INIT_FAST (PTHREAD_MUTEX_ADAPTIVE_NP). The critical
sections of buf_pool.flush_list_mutex should be shorter than those for
buf_pool.mutex, because in the worst case, they cover a linear scan of
buf_pool.flush_list, while the worst case of a critical section of
buf_pool.mutex covers a linear scan of the potentially much longer
buf_pool.LRU list.
mysql_mutex_is_owner(), safe_mutex_is_owner(): New predicate, usable
with SAFE_MUTEX. Some InnoDB debug assertions need this predicate
instead of mysql_mutex_assert_owner() or mysql_mutex_assert_not_owner().
buf_pool_t::n_flush_LRU, buf_pool_t::n_flush_list:
Replaces buf_pool_t::init_flush[] and buf_pool_t::n_flush[].
The number of active flush operations.
buf_pool_t::mutex, buf_pool_t::flush_list_mutex: Use mysql_mutex_t
instead of ib_mutex_t, to have native mutexes with PERFORMANCE_SCHEMA
and SAFE_MUTEX instrumentation.
buf_pool_t::done_flush_LRU: Condition variable for !n_flush_LRU.
buf_pool_t::done_flush_list: Condition variable for !n_flush_list.
buf_pool_t::do_flush_list: Condition variable to wake up the
buf_flush_page_cleaner when a log checkpoint needs to be written
or the server is being shut down. Replaces buf_flush_event.
We will keep using timed waits (the page cleaner thread will wake
_at least_ once per second), because the calculations for
innodb_adaptive_flushing depend on fixed time intervals.
buf_dblwr: Allocate statically, and move all code to member functions.
Use a native mutex and condition variable. Remove code to deal with
single-page flushing.
buf_dblwr_check_block(): Make the check debug-only. We were spending
a significant amount of execution time in page_simple_validate_new().
flush_counters_t::unzip_LRU_evicted: Remove.
IORequest: Make more members const. FIXME: m_fil_node should be removed.
buf_flush_sync_lsn: Protect by std::atomic, not page_cleaner.mutex
(which we are removing).
page_cleaner_slot_t, page_cleaner_t: Remove many redundant members.
pc_request_flush_slot(): Replaces pc_request() and pc_flush_slot().
recv_writer_thread: Remove. Recovery works just fine without it, if we
simply invoke buf_flush_sync() at the end of each batch in
recv_sys_t::apply().
recv_recovery_from_checkpoint_finish(): Remove. We can simply call
recv_sys.debug_free() directly.
srv_started_redo: Replaces srv_start_state.
SRV_SHUTDOWN_FLUSH_PHASE: Remove. logs_empty_and_mark_files_at_shutdown()
can communicate with the normal page cleaner loop via the new function
flush_buffer_pool().
buf_flush_remove(): Assert that the calling thread is holding
buf_pool.flush_list_mutex. This removes unnecessary mutex operations
from buf_flush_remove_pages() and buf_flush_dirty_pages(),
which replace buf_LRU_flush_or_remove_pages().
buf_flush_lists(): Renamed from buf_flush_batch(), with simplified
interface. Return the number of flushed pages. Clarified comments and
renamed min_n to max_n. Identify LRU batch by lsn=0. Merge all the functions
buf_flush_start(), buf_flush_batch(), buf_flush_end() directly to this
function, which was their only caller, and remove 2 unnecessary
buf_pool.mutex release/re-acquisition that we used to perform around
the buf_flush_batch() call. At the start, if not all log has been
durably written, wait for a background task to do it, or start a new
task to do it. This allows the log write to run concurrently with our
page flushing batch. Any pages that were skipped due to too recent
FIL_PAGE_LSN or due to them being latched by a writer should be flushed
during the next batch, unless there are further modifications to those
pages. It is possible that a page that we must flush due to small
oldest_modification also carries a recent FIL_PAGE_LSN or is being
constantly modified. In the worst case, all writers would then end up
waiting in log_free_check() to allow the flushing and the checkpoint
to complete.
buf_do_flush_list_batch(): Clarify comments, and rename min_n to max_n.
Cache the last looked up tablespace. If neighbor flushing is not applicable,
invoke buf_flush_page() directly, avoiding a page lookup in between.
buf_flush_space(): Auxiliary function to look up a tablespace for
page flushing.
buf_flush_page(): Defer the computation of space->full_crc32(). Never
call log_write_up_to(), but instead skip persistent pages whose latest
modification (FIL_PAGE_LSN) is newer than the redo log. Also skip
pages on which we cannot acquire a shared latch without waiting.
buf_flush_try_neighbors(): Do not bother checking buf_fix_count
because buf_flush_page() will no longer wait for the page latch.
Take the tablespace as a parameter, and only execute this function
when innodb_flush_neighbors>0. Avoid repeated calls of page_id_t::fold().
buf_flush_relocate_on_flush_list(): Declare as cold, and push down
a condition from the callers.
buf_flush_check_neighbor(): Take id.fold() as a parameter.
buf_flush_sync(): Ensure that the buf_pool.flush_list is empty,
because the flushing batch will skip pages whose modifications have
not yet been written to the log or were latched for modification.
buf_free_from_unzip_LRU_list_batch(): Remove redundant local variables.
buf_flush_LRU_list_batch(): Let the caller buf_do_LRU_batch() initialize
the counters, and report n->evicted.
Cache the last looked up tablespace. If neighbor flushing is not applicable,
invoke buf_flush_page() directly, avoiding a page lookup in between.
buf_do_LRU_batch(): Return the number of pages flushed.
buf_LRU_free_page(): Only release and re-acquire buf_pool.mutex if
adaptive hash index entries are pointing to the block.
buf_LRU_get_free_block(): Do not wake up the page cleaner, because it
will no longer perform any useful work for us, and we do not want it
to compete for I/O while buf_flush_lists(innodb_lru_flush_size, 0)
writes out and evicts at most innodb_lru_flush_size pages. (The
function buf_do_LRU_batch() may complete after writing fewer pages if
more than innodb_lru_scan_depth pages end up in buf_pool.free list.)
Eliminate some mutex release-acquire cycles, and wait for the LRU
flush batch to complete before rescanning.
buf_LRU_check_size_of_non_data_objects(): Simplify the code.
buf_page_write_complete(): Remove the parameter evict, and always
evict pages that were part of an LRU flush.
buf_page_create(): Take a pre-allocated page as a parameter.
buf_pool_t::free_block(): Free a pre-allocated block.
recv_sys_t::recover_low(), recv_sys_t::apply(): Preallocate the block
while not holding recv_sys.mutex. During page allocation, we may
initiate a page flush, which in turn may initiate a log flush, which
would require acquiring log_sys.mutex, which should always be acquired
before recv_sys.mutex in order to avoid deadlocks. Therefore, we must
not be holding recv_sys.mutex while allocating a buffer pool block.
BtrBulk::logFreeCheck(): Skip a redundant condition.
row_undo_step(): Do not invoke srv_inc_activity_count() for every row
that is being rolled back. It should suffice to invoke the function in
trx_flush_log_if_needed() during trx_t::commit_in_memory() when the
rollback completes.
sync_check_enable(): Remove. We will enable innodb_sync_debug from the
very beginning.
Reviewed by: Vladislav Vaintroub
2020-10-15 12:10:42 +03:00
|
|
|
#include "buf0dblwr.h"
|
2019-06-11 11:03:18 +03:00
|
|
|
#include "hash0hash.h"
|
2016-08-12 11:17:45 +03:00
|
|
|
#include "log0recv.h"
|
2014-02-26 19:11:54 +01:00
|
|
|
#include "dict0types.h"
|
2020-05-17 16:24:40 +03:00
|
|
|
#include "ilist.h"
|
2020-06-11 22:52:47 +05:30
|
|
|
#include <set>
|
|
|
|
|
#include <mutex>
|
2014-02-26 19:11:54 +01:00
|
|
|
|
2019-12-23 01:18:39 +08:00
|
|
|
struct unflushed_spaces_tag_t;
|
|
|
|
|
struct rotation_list_tag_t;
|
2021-03-20 01:31:01 +03:00
|
|
|
struct space_list_tag_t;
|
|
|
|
|
struct named_spaces_tag_t;
|
|
|
|
|
|
|
|
|
|
using space_list_t= ilist<fil_space_t, space_list_tag_t>;
|
2014-02-26 19:11:54 +01:00
|
|
|
|
|
|
|
|
// Forward declaration
|
2018-04-29 09:41:42 +03:00
|
|
|
extern my_bool srv_use_doublewrite_buf;
|
MDEV-15053 Reduce buf_pool_t::mutex contention
User-visible changes: The INFORMATION_SCHEMA views INNODB_BUFFER_PAGE
and INNODB_BUFFER_PAGE_LRU will report a dummy value FLUSH_TYPE=0
and will no longer report the PAGE_STATE value READY_FOR_USE.
We will remove some fields from buf_page_t and move much code to
member functions of buf_pool_t and buf_page_t, so that the access
rules of data members can be enforced consistently.
Evicting or adding pages in buf_pool.LRU will remain covered by
buf_pool.mutex.
Evicting or adding pages in buf_pool.page_hash will remain
covered by both buf_pool.mutex and the buf_pool.page_hash X-latch.
After this fix, buf_pool.page_hash lookups can entirely
avoid acquiring buf_pool.mutex, only relying on
buf_pool.hash_lock_get() S-latch.
Similarly, buf_flush_check_neighbors() can will rely solely on
buf_pool.mutex, no buf_pool.page_hash latch at all.
The buf_pool.mutex is rather contended in I/O heavy benchmarks,
especially when the workload does not fit in the buffer pool.
The first attempt to alleviate the contention was the
buf_pool_t::mutex split in
commit 4ed7082eefe56b3e97e0edefb3df76dd7ef5e858
which introduced buf_block_t::mutex, which we are now removing.
Later, multiple instances of buf_pool_t were introduced
in commit c18084f71b02ea707c6461353e6cfc15d7553bc6
and recently removed by us in
commit 1a6f708ec594ac0ae2dd30db926ab07b100fa24b (MDEV-15058).
UNIV_BUF_DEBUG: Remove. This option to enable some buffer pool
related debugging in otherwise non-debug builds has not been used
for years. Instead, we have been using UNIV_DEBUG, which is enabled
in CMAKE_BUILD_TYPE=Debug.
buf_block_t::mutex, buf_pool_t::zip_mutex: Remove. We can mainly rely on
std::atomic and the buf_pool.page_hash latches, and in some cases
depend on buf_pool.mutex or buf_pool.flush_list_mutex just like before.
We must always release buf_block_t::lock before invoking
unfix() or io_unfix(), to prevent a glitch where a block that was
added to the buf_pool.free list would apper X-latched. See
commit c5883debd6ef440a037011c11873b396923e93c5 how this glitch
was finally caught in a debug environment.
We move some buf_pool_t::page_hash specific code from the
ha and hash modules to buf_pool, for improved readability.
buf_pool_t::close(): Assert that all blocks are clean, except
on aborted startup or crash-like shutdown.
buf_pool_t::validate(): No longer attempt to validate
n_flush[] against the number of BUF_IO_WRITE fixed blocks,
because buf_page_t::flush_type no longer exists.
buf_pool_t::watch_set(): Replaces buf_pool_watch_set().
Reduce mutex contention by separating the buf_pool.watch[]
allocation and the insert into buf_pool.page_hash.
buf_pool_t::page_hash_lock<bool exclusive>(): Acquire a
buf_pool.page_hash latch.
Replaces and extends buf_page_hash_lock_s_confirm()
and buf_page_hash_lock_x_confirm().
buf_pool_t::READ_AHEAD_PAGES: Renamed from BUF_READ_AHEAD_PAGES.
buf_pool_t::curr_size, old_size, read_ahead_area, n_pend_reads:
Use Atomic_counter.
buf_pool_t::running_out(): Replaces buf_LRU_buf_pool_running_out().
buf_pool_t::LRU_remove(): Remove a block from the LRU list
and return its predecessor. Incorporates buf_LRU_adjust_hp(),
which was removed.
buf_page_get_gen(): Remove a redundant call of fsp_is_system_temporary(),
for mode == BUF_GET_IF_IN_POOL_OR_WATCH, which is only used by
BTR_DELETE_OP (purge), which is never invoked on temporary tables.
buf_free_from_unzip_LRU_list_batch(): Avoid redundant assignments.
buf_LRU_free_from_unzip_LRU_list(): Simplify the loop condition.
buf_LRU_free_page(): Clarify the function comment.
buf_flush_check_neighbor(), buf_flush_check_neighbors():
Rewrite the construction of the page hash range. We will hold
the buf_pool.mutex for up to buf_pool.read_ahead_area (at most 64)
consecutive lookups of buf_pool.page_hash.
buf_flush_page_and_try_neighbors(): Remove.
Merge to its only callers, and remove redundant operations in
buf_flush_LRU_list_batch().
buf_read_ahead_random(), buf_read_ahead_linear(): Rewrite.
Do not acquire buf_pool.mutex, and iterate directly with page_id_t.
ut_2_power_up(): Remove. my_round_up_to_next_power() is inlined
and avoids any loops.
fil_page_get_prev(), fil_page_get_next(), fil_addr_is_null(): Remove.
buf_flush_page(): Add a fil_space_t* parameter. Minimize the
buf_pool.mutex hold time. buf_pool.n_flush[] is no longer updated
atomically with the io_fix, and we will protect most buf_block_t
fields with buf_block_t::lock. The function
buf_flush_write_block_low() is removed and merged here.
buf_page_init_for_read(): Use static linkage. Initialize the newly
allocated block and acquire the exclusive buf_block_t::lock while not
holding any mutex.
IORequest::IORequest(): Remove the body. We only need to invoke
set_punch_hole() in buf_flush_page() and nowhere else.
buf_page_t::flush_type: Remove. Replaced by IORequest::flush_type.
This field is only used during a fil_io() call.
That function already takes IORequest as a parameter, so we had
better introduce for the rarely changing field.
buf_block_t::init(): Replaces buf_page_init().
buf_page_t::init(): Replaces buf_page_init_low().
buf_block_t::initialise(): Initialise many fields, but
keep the buf_page_t::state(). Both buf_pool_t::validate() and
buf_page_optimistic_get() requires that buf_page_t::in_file()
be protected atomically with buf_page_t::in_page_hash
and buf_page_t::in_LRU_list.
buf_page_optimistic_get(): Now that buf_block_t::mutex
no longer exists, we must check buf_page_t::io_fix()
after acquiring the buf_pool.page_hash lock, to detect
whether buf_page_init_for_read() has been initiated.
We will also check the io_fix() before acquiring hash_lock
in order to avoid unnecessary computation.
The field buf_block_t::modify_clock (protected by buf_block_t::lock)
allows buf_page_optimistic_get() to validate the block.
buf_page_t::real_size: Remove. It was only used while flushing
pages of page_compressed tables.
buf_page_encrypt(): Add an output parameter that allows us ot eliminate
buf_page_t::real_size. Replace a condition with debug assertion.
buf_page_should_punch_hole(): Remove.
buf_dblwr_t::add_to_batch(): Replaces buf_dblwr_add_to_batch().
Add the parameter size (to replace buf_page_t::real_size).
buf_dblwr_t::write_single_page(): Replaces buf_dblwr_write_single_page().
Add the parameter size (to replace buf_page_t::real_size).
fil_system_t::detach(): Replaces fil_space_detach().
Ensure that fil_validate() will not be violated even if
fil_system.mutex is released and reacquired.
fil_node_t::complete_io(): Renamed from fil_node_complete_io().
fil_node_t::close_to_free(): Replaces fil_node_close_to_free().
Avoid invoking fil_node_t::close() because fil_system.n_open
has already been decremented in fil_space_t::detach().
BUF_BLOCK_READY_FOR_USE: Remove. Directly use BUF_BLOCK_MEMORY.
BUF_BLOCK_ZIP_DIRTY: Remove. Directly use BUF_BLOCK_ZIP_PAGE,
and distinguish dirty pages by buf_page_t::oldest_modification().
BUF_BLOCK_POOL_WATCH: Remove. Use BUF_BLOCK_NOT_USED instead.
This state was only being used for buf_page_t that are in
buf_pool.watch.
buf_pool_t::watch[]: Remove pointer indirection.
buf_page_t::in_flush_list: Remove. It was set if and only if
buf_page_t::oldest_modification() is nonzero.
buf_page_decrypt_after_read(), buf_corrupt_page_release(),
buf_page_check_corrupt(): Change the const fil_space_t* parameter
to const fil_node_t& so that we can report the correct file name.
buf_page_monitor(): Declare as an ATTRIBUTE_COLD global function.
buf_page_io_complete(): Split to buf_page_read_complete() and
buf_page_write_complete().
buf_dblwr_t::in_use: Remove.
buf_dblwr_t::buf_block_array: Add IORequest::flush_t.
buf_dblwr_sync_datafiles(): Remove. It was a useless wrapper of
os_aio_wait_until_no_pending_writes().
buf_flush_write_complete(): Declare static, not global.
Add the parameter IORequest::flush_t.
buf_flush_freed_page(): Simplify the code.
recv_sys_t::flush_lru: Renamed from flush_type and changed to bool.
fil_read(), fil_write(): Replaced with direct use of fil_io().
fil_buffering_disabled(): Remove. Check srv_file_flush_method directly.
fil_mutex_enter_and_prepare_for_io(): Return the resolved
fil_space_t* to avoid a duplicated lookup in the caller.
fil_report_invalid_page_access(): Clean up the parameters.
fil_io(): Return fil_io_t, which comprises fil_node_t and error code.
Always invoke fil_space_t::acquire_for_io() and let either the
sync=true caller or fil_aio_callback() invoke
fil_space_t::release_for_io().
fil_aio_callback(): Rewrite to replace buf_page_io_complete().
fil_check_pending_operations(): Remove a parameter, and remove some
redundant lookups.
fil_node_close_to_free(): Wait for n_pending==0. Because we no longer
do an extra lookup of the tablespace between fil_io() and the
completion of the operation, we must give fil_node_t::complete_io() a
chance to decrement the counter.
fil_close_tablespace(): Remove unused parameter trx, and document
that this is only invoked during the error handling of IMPORT TABLESPACE.
row_import_discard_changes(): Merged with the only caller,
row_import_cleanup(). Do not lock up the data dictionary while
invoking fil_close_tablespace().
logs_empty_and_mark_files_at_shutdown(): Do not invoke
fil_close_all_files(), to avoid a !needs_flush assertion failure
on fil_node_t::close().
innodb_shutdown(): Invoke os_aio_free() before fil_close_all_files().
fil_close_all_files(): Invoke fil_flush_file_spaces()
to ensure proper durability.
thread_pool::unbind(): Fix a crash that would occur on Windows
after srv_thread_pool->disable_aio() and os_file_close().
This fix was submitted by Vladislav Vaintroub.
Thanks to Matthias Leich and Axel Schwenke for extensive testing,
Vladislav Vaintroub for helpful comments, and Eugene Kosov for a review.
2020-06-05 12:35:46 +03:00
|
|
|
|
|
|
|
|
/** Possible values of innodb_flush_method */
|
|
|
|
|
enum srv_flush_t
|
|
|
|
|
{
|
|
|
|
|
/** fsync, the default */
|
|
|
|
|
SRV_FSYNC= 0,
|
|
|
|
|
/** open log files in O_DSYNC mode */
|
|
|
|
|
SRV_O_DSYNC,
|
|
|
|
|
/** do not call os_file_flush() when writing data files, but do flush
|
|
|
|
|
after writing to log files */
|
|
|
|
|
SRV_LITTLESYNC,
|
|
|
|
|
/** do not flush after writing */
|
|
|
|
|
SRV_NOSYNC,
|
|
|
|
|
/** invoke os_file_set_nocache() on data files. This implies using
|
2021-02-20 11:58:58 +02:00
|
|
|
unbuffered I/O but still fdatasync(), because some filesystems might
|
|
|
|
|
not flush meta-data on write completion */
|
MDEV-15053 Reduce buf_pool_t::mutex contention
User-visible changes: The INFORMATION_SCHEMA views INNODB_BUFFER_PAGE
and INNODB_BUFFER_PAGE_LRU will report a dummy value FLUSH_TYPE=0
and will no longer report the PAGE_STATE value READY_FOR_USE.
We will remove some fields from buf_page_t and move much code to
member functions of buf_pool_t and buf_page_t, so that the access
rules of data members can be enforced consistently.
Evicting or adding pages in buf_pool.LRU will remain covered by
buf_pool.mutex.
Evicting or adding pages in buf_pool.page_hash will remain
covered by both buf_pool.mutex and the buf_pool.page_hash X-latch.
After this fix, buf_pool.page_hash lookups can entirely
avoid acquiring buf_pool.mutex, only relying on
buf_pool.hash_lock_get() S-latch.
Similarly, buf_flush_check_neighbors() can will rely solely on
buf_pool.mutex, no buf_pool.page_hash latch at all.
The buf_pool.mutex is rather contended in I/O heavy benchmarks,
especially when the workload does not fit in the buffer pool.
The first attempt to alleviate the contention was the
buf_pool_t::mutex split in
commit 4ed7082eefe56b3e97e0edefb3df76dd7ef5e858
which introduced buf_block_t::mutex, which we are now removing.
Later, multiple instances of buf_pool_t were introduced
in commit c18084f71b02ea707c6461353e6cfc15d7553bc6
and recently removed by us in
commit 1a6f708ec594ac0ae2dd30db926ab07b100fa24b (MDEV-15058).
UNIV_BUF_DEBUG: Remove. This option to enable some buffer pool
related debugging in otherwise non-debug builds has not been used
for years. Instead, we have been using UNIV_DEBUG, which is enabled
in CMAKE_BUILD_TYPE=Debug.
buf_block_t::mutex, buf_pool_t::zip_mutex: Remove. We can mainly rely on
std::atomic and the buf_pool.page_hash latches, and in some cases
depend on buf_pool.mutex or buf_pool.flush_list_mutex just like before.
We must always release buf_block_t::lock before invoking
unfix() or io_unfix(), to prevent a glitch where a block that was
added to the buf_pool.free list would apper X-latched. See
commit c5883debd6ef440a037011c11873b396923e93c5 how this glitch
was finally caught in a debug environment.
We move some buf_pool_t::page_hash specific code from the
ha and hash modules to buf_pool, for improved readability.
buf_pool_t::close(): Assert that all blocks are clean, except
on aborted startup or crash-like shutdown.
buf_pool_t::validate(): No longer attempt to validate
n_flush[] against the number of BUF_IO_WRITE fixed blocks,
because buf_page_t::flush_type no longer exists.
buf_pool_t::watch_set(): Replaces buf_pool_watch_set().
Reduce mutex contention by separating the buf_pool.watch[]
allocation and the insert into buf_pool.page_hash.
buf_pool_t::page_hash_lock<bool exclusive>(): Acquire a
buf_pool.page_hash latch.
Replaces and extends buf_page_hash_lock_s_confirm()
and buf_page_hash_lock_x_confirm().
buf_pool_t::READ_AHEAD_PAGES: Renamed from BUF_READ_AHEAD_PAGES.
buf_pool_t::curr_size, old_size, read_ahead_area, n_pend_reads:
Use Atomic_counter.
buf_pool_t::running_out(): Replaces buf_LRU_buf_pool_running_out().
buf_pool_t::LRU_remove(): Remove a block from the LRU list
and return its predecessor. Incorporates buf_LRU_adjust_hp(),
which was removed.
buf_page_get_gen(): Remove a redundant call of fsp_is_system_temporary(),
for mode == BUF_GET_IF_IN_POOL_OR_WATCH, which is only used by
BTR_DELETE_OP (purge), which is never invoked on temporary tables.
buf_free_from_unzip_LRU_list_batch(): Avoid redundant assignments.
buf_LRU_free_from_unzip_LRU_list(): Simplify the loop condition.
buf_LRU_free_page(): Clarify the function comment.
buf_flush_check_neighbor(), buf_flush_check_neighbors():
Rewrite the construction of the page hash range. We will hold
the buf_pool.mutex for up to buf_pool.read_ahead_area (at most 64)
consecutive lookups of buf_pool.page_hash.
buf_flush_page_and_try_neighbors(): Remove.
Merge to its only callers, and remove redundant operations in
buf_flush_LRU_list_batch().
buf_read_ahead_random(), buf_read_ahead_linear(): Rewrite.
Do not acquire buf_pool.mutex, and iterate directly with page_id_t.
ut_2_power_up(): Remove. my_round_up_to_next_power() is inlined
and avoids any loops.
fil_page_get_prev(), fil_page_get_next(), fil_addr_is_null(): Remove.
buf_flush_page(): Add a fil_space_t* parameter. Minimize the
buf_pool.mutex hold time. buf_pool.n_flush[] is no longer updated
atomically with the io_fix, and we will protect most buf_block_t
fields with buf_block_t::lock. The function
buf_flush_write_block_low() is removed and merged here.
buf_page_init_for_read(): Use static linkage. Initialize the newly
allocated block and acquire the exclusive buf_block_t::lock while not
holding any mutex.
IORequest::IORequest(): Remove the body. We only need to invoke
set_punch_hole() in buf_flush_page() and nowhere else.
buf_page_t::flush_type: Remove. Replaced by IORequest::flush_type.
This field is only used during a fil_io() call.
That function already takes IORequest as a parameter, so we had
better introduce for the rarely changing field.
buf_block_t::init(): Replaces buf_page_init().
buf_page_t::init(): Replaces buf_page_init_low().
buf_block_t::initialise(): Initialise many fields, but
keep the buf_page_t::state(). Both buf_pool_t::validate() and
buf_page_optimistic_get() requires that buf_page_t::in_file()
be protected atomically with buf_page_t::in_page_hash
and buf_page_t::in_LRU_list.
buf_page_optimistic_get(): Now that buf_block_t::mutex
no longer exists, we must check buf_page_t::io_fix()
after acquiring the buf_pool.page_hash lock, to detect
whether buf_page_init_for_read() has been initiated.
We will also check the io_fix() before acquiring hash_lock
in order to avoid unnecessary computation.
The field buf_block_t::modify_clock (protected by buf_block_t::lock)
allows buf_page_optimistic_get() to validate the block.
buf_page_t::real_size: Remove. It was only used while flushing
pages of page_compressed tables.
buf_page_encrypt(): Add an output parameter that allows us ot eliminate
buf_page_t::real_size. Replace a condition with debug assertion.
buf_page_should_punch_hole(): Remove.
buf_dblwr_t::add_to_batch(): Replaces buf_dblwr_add_to_batch().
Add the parameter size (to replace buf_page_t::real_size).
buf_dblwr_t::write_single_page(): Replaces buf_dblwr_write_single_page().
Add the parameter size (to replace buf_page_t::real_size).
fil_system_t::detach(): Replaces fil_space_detach().
Ensure that fil_validate() will not be violated even if
fil_system.mutex is released and reacquired.
fil_node_t::complete_io(): Renamed from fil_node_complete_io().
fil_node_t::close_to_free(): Replaces fil_node_close_to_free().
Avoid invoking fil_node_t::close() because fil_system.n_open
has already been decremented in fil_space_t::detach().
BUF_BLOCK_READY_FOR_USE: Remove. Directly use BUF_BLOCK_MEMORY.
BUF_BLOCK_ZIP_DIRTY: Remove. Directly use BUF_BLOCK_ZIP_PAGE,
and distinguish dirty pages by buf_page_t::oldest_modification().
BUF_BLOCK_POOL_WATCH: Remove. Use BUF_BLOCK_NOT_USED instead.
This state was only being used for buf_page_t that are in
buf_pool.watch.
buf_pool_t::watch[]: Remove pointer indirection.
buf_page_t::in_flush_list: Remove. It was set if and only if
buf_page_t::oldest_modification() is nonzero.
buf_page_decrypt_after_read(), buf_corrupt_page_release(),
buf_page_check_corrupt(): Change the const fil_space_t* parameter
to const fil_node_t& so that we can report the correct file name.
buf_page_monitor(): Declare as an ATTRIBUTE_COLD global function.
buf_page_io_complete(): Split to buf_page_read_complete() and
buf_page_write_complete().
buf_dblwr_t::in_use: Remove.
buf_dblwr_t::buf_block_array: Add IORequest::flush_t.
buf_dblwr_sync_datafiles(): Remove. It was a useless wrapper of
os_aio_wait_until_no_pending_writes().
buf_flush_write_complete(): Declare static, not global.
Add the parameter IORequest::flush_t.
buf_flush_freed_page(): Simplify the code.
recv_sys_t::flush_lru: Renamed from flush_type and changed to bool.
fil_read(), fil_write(): Replaced with direct use of fil_io().
fil_buffering_disabled(): Remove. Check srv_file_flush_method directly.
fil_mutex_enter_and_prepare_for_io(): Return the resolved
fil_space_t* to avoid a duplicated lookup in the caller.
fil_report_invalid_page_access(): Clean up the parameters.
fil_io(): Return fil_io_t, which comprises fil_node_t and error code.
Always invoke fil_space_t::acquire_for_io() and let either the
sync=true caller or fil_aio_callback() invoke
fil_space_t::release_for_io().
fil_aio_callback(): Rewrite to replace buf_page_io_complete().
fil_check_pending_operations(): Remove a parameter, and remove some
redundant lookups.
fil_node_close_to_free(): Wait for n_pending==0. Because we no longer
do an extra lookup of the tablespace between fil_io() and the
completion of the operation, we must give fil_node_t::complete_io() a
chance to decrement the counter.
fil_close_tablespace(): Remove unused parameter trx, and document
that this is only invoked during the error handling of IMPORT TABLESPACE.
row_import_discard_changes(): Merged with the only caller,
row_import_cleanup(). Do not lock up the data dictionary while
invoking fil_close_tablespace().
logs_empty_and_mark_files_at_shutdown(): Do not invoke
fil_close_all_files(), to avoid a !needs_flush assertion failure
on fil_node_t::close().
innodb_shutdown(): Invoke os_aio_free() before fil_close_all_files().
fil_close_all_files(): Invoke fil_flush_file_spaces()
to ensure proper durability.
thread_pool::unbind(): Fix a crash that would occur on Windows
after srv_thread_pool->disable_aio() and os_file_close().
This fix was submitted by Vladislav Vaintroub.
Thanks to Matthias Leich and Axel Schwenke for extensive testing,
Vladislav Vaintroub for helpful comments, and Eugene Kosov for a review.
2020-06-05 12:35:46 +03:00
|
|
|
SRV_O_DIRECT,
|
2021-02-20 11:58:58 +02:00
|
|
|
/** Like O_DIRECT, but skip fdatasync(), assuming that the data is
|
|
|
|
|
durable on write completion */
|
MDEV-15053 Reduce buf_pool_t::mutex contention
User-visible changes: The INFORMATION_SCHEMA views INNODB_BUFFER_PAGE
and INNODB_BUFFER_PAGE_LRU will report a dummy value FLUSH_TYPE=0
and will no longer report the PAGE_STATE value READY_FOR_USE.
We will remove some fields from buf_page_t and move much code to
member functions of buf_pool_t and buf_page_t, so that the access
rules of data members can be enforced consistently.
Evicting or adding pages in buf_pool.LRU will remain covered by
buf_pool.mutex.
Evicting or adding pages in buf_pool.page_hash will remain
covered by both buf_pool.mutex and the buf_pool.page_hash X-latch.
After this fix, buf_pool.page_hash lookups can entirely
avoid acquiring buf_pool.mutex, only relying on
buf_pool.hash_lock_get() S-latch.
Similarly, buf_flush_check_neighbors() can will rely solely on
buf_pool.mutex, no buf_pool.page_hash latch at all.
The buf_pool.mutex is rather contended in I/O heavy benchmarks,
especially when the workload does not fit in the buffer pool.
The first attempt to alleviate the contention was the
buf_pool_t::mutex split in
commit 4ed7082eefe56b3e97e0edefb3df76dd7ef5e858
which introduced buf_block_t::mutex, which we are now removing.
Later, multiple instances of buf_pool_t were introduced
in commit c18084f71b02ea707c6461353e6cfc15d7553bc6
and recently removed by us in
commit 1a6f708ec594ac0ae2dd30db926ab07b100fa24b (MDEV-15058).
UNIV_BUF_DEBUG: Remove. This option to enable some buffer pool
related debugging in otherwise non-debug builds has not been used
for years. Instead, we have been using UNIV_DEBUG, which is enabled
in CMAKE_BUILD_TYPE=Debug.
buf_block_t::mutex, buf_pool_t::zip_mutex: Remove. We can mainly rely on
std::atomic and the buf_pool.page_hash latches, and in some cases
depend on buf_pool.mutex or buf_pool.flush_list_mutex just like before.
We must always release buf_block_t::lock before invoking
unfix() or io_unfix(), to prevent a glitch where a block that was
added to the buf_pool.free list would apper X-latched. See
commit c5883debd6ef440a037011c11873b396923e93c5 how this glitch
was finally caught in a debug environment.
We move some buf_pool_t::page_hash specific code from the
ha and hash modules to buf_pool, for improved readability.
buf_pool_t::close(): Assert that all blocks are clean, except
on aborted startup or crash-like shutdown.
buf_pool_t::validate(): No longer attempt to validate
n_flush[] against the number of BUF_IO_WRITE fixed blocks,
because buf_page_t::flush_type no longer exists.
buf_pool_t::watch_set(): Replaces buf_pool_watch_set().
Reduce mutex contention by separating the buf_pool.watch[]
allocation and the insert into buf_pool.page_hash.
buf_pool_t::page_hash_lock<bool exclusive>(): Acquire a
buf_pool.page_hash latch.
Replaces and extends buf_page_hash_lock_s_confirm()
and buf_page_hash_lock_x_confirm().
buf_pool_t::READ_AHEAD_PAGES: Renamed from BUF_READ_AHEAD_PAGES.
buf_pool_t::curr_size, old_size, read_ahead_area, n_pend_reads:
Use Atomic_counter.
buf_pool_t::running_out(): Replaces buf_LRU_buf_pool_running_out().
buf_pool_t::LRU_remove(): Remove a block from the LRU list
and return its predecessor. Incorporates buf_LRU_adjust_hp(),
which was removed.
buf_page_get_gen(): Remove a redundant call of fsp_is_system_temporary(),
for mode == BUF_GET_IF_IN_POOL_OR_WATCH, which is only used by
BTR_DELETE_OP (purge), which is never invoked on temporary tables.
buf_free_from_unzip_LRU_list_batch(): Avoid redundant assignments.
buf_LRU_free_from_unzip_LRU_list(): Simplify the loop condition.
buf_LRU_free_page(): Clarify the function comment.
buf_flush_check_neighbor(), buf_flush_check_neighbors():
Rewrite the construction of the page hash range. We will hold
the buf_pool.mutex for up to buf_pool.read_ahead_area (at most 64)
consecutive lookups of buf_pool.page_hash.
buf_flush_page_and_try_neighbors(): Remove.
Merge to its only callers, and remove redundant operations in
buf_flush_LRU_list_batch().
buf_read_ahead_random(), buf_read_ahead_linear(): Rewrite.
Do not acquire buf_pool.mutex, and iterate directly with page_id_t.
ut_2_power_up(): Remove. my_round_up_to_next_power() is inlined
and avoids any loops.
fil_page_get_prev(), fil_page_get_next(), fil_addr_is_null(): Remove.
buf_flush_page(): Add a fil_space_t* parameter. Minimize the
buf_pool.mutex hold time. buf_pool.n_flush[] is no longer updated
atomically with the io_fix, and we will protect most buf_block_t
fields with buf_block_t::lock. The function
buf_flush_write_block_low() is removed and merged here.
buf_page_init_for_read(): Use static linkage. Initialize the newly
allocated block and acquire the exclusive buf_block_t::lock while not
holding any mutex.
IORequest::IORequest(): Remove the body. We only need to invoke
set_punch_hole() in buf_flush_page() and nowhere else.
buf_page_t::flush_type: Remove. Replaced by IORequest::flush_type.
This field is only used during a fil_io() call.
That function already takes IORequest as a parameter, so we had
better introduce for the rarely changing field.
buf_block_t::init(): Replaces buf_page_init().
buf_page_t::init(): Replaces buf_page_init_low().
buf_block_t::initialise(): Initialise many fields, but
keep the buf_page_t::state(). Both buf_pool_t::validate() and
buf_page_optimistic_get() requires that buf_page_t::in_file()
be protected atomically with buf_page_t::in_page_hash
and buf_page_t::in_LRU_list.
buf_page_optimistic_get(): Now that buf_block_t::mutex
no longer exists, we must check buf_page_t::io_fix()
after acquiring the buf_pool.page_hash lock, to detect
whether buf_page_init_for_read() has been initiated.
We will also check the io_fix() before acquiring hash_lock
in order to avoid unnecessary computation.
The field buf_block_t::modify_clock (protected by buf_block_t::lock)
allows buf_page_optimistic_get() to validate the block.
buf_page_t::real_size: Remove. It was only used while flushing
pages of page_compressed tables.
buf_page_encrypt(): Add an output parameter that allows us ot eliminate
buf_page_t::real_size. Replace a condition with debug assertion.
buf_page_should_punch_hole(): Remove.
buf_dblwr_t::add_to_batch(): Replaces buf_dblwr_add_to_batch().
Add the parameter size (to replace buf_page_t::real_size).
buf_dblwr_t::write_single_page(): Replaces buf_dblwr_write_single_page().
Add the parameter size (to replace buf_page_t::real_size).
fil_system_t::detach(): Replaces fil_space_detach().
Ensure that fil_validate() will not be violated even if
fil_system.mutex is released and reacquired.
fil_node_t::complete_io(): Renamed from fil_node_complete_io().
fil_node_t::close_to_free(): Replaces fil_node_close_to_free().
Avoid invoking fil_node_t::close() because fil_system.n_open
has already been decremented in fil_space_t::detach().
BUF_BLOCK_READY_FOR_USE: Remove. Directly use BUF_BLOCK_MEMORY.
BUF_BLOCK_ZIP_DIRTY: Remove. Directly use BUF_BLOCK_ZIP_PAGE,
and distinguish dirty pages by buf_page_t::oldest_modification().
BUF_BLOCK_POOL_WATCH: Remove. Use BUF_BLOCK_NOT_USED instead.
This state was only being used for buf_page_t that are in
buf_pool.watch.
buf_pool_t::watch[]: Remove pointer indirection.
buf_page_t::in_flush_list: Remove. It was set if and only if
buf_page_t::oldest_modification() is nonzero.
buf_page_decrypt_after_read(), buf_corrupt_page_release(),
buf_page_check_corrupt(): Change the const fil_space_t* parameter
to const fil_node_t& so that we can report the correct file name.
buf_page_monitor(): Declare as an ATTRIBUTE_COLD global function.
buf_page_io_complete(): Split to buf_page_read_complete() and
buf_page_write_complete().
buf_dblwr_t::in_use: Remove.
buf_dblwr_t::buf_block_array: Add IORequest::flush_t.
buf_dblwr_sync_datafiles(): Remove. It was a useless wrapper of
os_aio_wait_until_no_pending_writes().
buf_flush_write_complete(): Declare static, not global.
Add the parameter IORequest::flush_t.
buf_flush_freed_page(): Simplify the code.
recv_sys_t::flush_lru: Renamed from flush_type and changed to bool.
fil_read(), fil_write(): Replaced with direct use of fil_io().
fil_buffering_disabled(): Remove. Check srv_file_flush_method directly.
fil_mutex_enter_and_prepare_for_io(): Return the resolved
fil_space_t* to avoid a duplicated lookup in the caller.
fil_report_invalid_page_access(): Clean up the parameters.
fil_io(): Return fil_io_t, which comprises fil_node_t and error code.
Always invoke fil_space_t::acquire_for_io() and let either the
sync=true caller or fil_aio_callback() invoke
fil_space_t::release_for_io().
fil_aio_callback(): Rewrite to replace buf_page_io_complete().
fil_check_pending_operations(): Remove a parameter, and remove some
redundant lookups.
fil_node_close_to_free(): Wait for n_pending==0. Because we no longer
do an extra lookup of the tablespace between fil_io() and the
completion of the operation, we must give fil_node_t::complete_io() a
chance to decrement the counter.
fil_close_tablespace(): Remove unused parameter trx, and document
that this is only invoked during the error handling of IMPORT TABLESPACE.
row_import_discard_changes(): Merged with the only caller,
row_import_cleanup(). Do not lock up the data dictionary while
invoking fil_close_tablespace().
logs_empty_and_mark_files_at_shutdown(): Do not invoke
fil_close_all_files(), to avoid a !needs_flush assertion failure
on fil_node_t::close().
innodb_shutdown(): Invoke os_aio_free() before fil_close_all_files().
fil_close_all_files(): Invoke fil_flush_file_spaces()
to ensure proper durability.
thread_pool::unbind(): Fix a crash that would occur on Windows
after srv_thread_pool->disable_aio() and os_file_close().
This fix was submitted by Vladislav Vaintroub.
Thanks to Matthias Leich and Axel Schwenke for extensive testing,
Vladislav Vaintroub for helpful comments, and Eugene Kosov for a review.
2020-06-05 12:35:46 +03:00
|
|
|
SRV_O_DIRECT_NO_FSYNC
|
|
|
|
|
#ifdef _WIN32
|
|
|
|
|
/** Traditional Windows appoach to open all files without caching,
|
|
|
|
|
and do FileFlushBuffers() */
|
|
|
|
|
,SRV_ALL_O_DIRECT_FSYNC
|
|
|
|
|
#endif
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
/** innodb_flush_method */
|
|
|
|
|
extern ulong srv_file_flush_method;
|
|
|
|
|
|
|
|
|
|
/** Undo tablespaces starts with space_id. */
|
|
|
|
|
extern ulint srv_undo_space_id_start;
|
|
|
|
|
/** The number of UNDO tablespaces that are open and ready to use. */
|
|
|
|
|
extern ulint srv_undo_tablespaces_open;
|
|
|
|
|
|
|
|
|
|
/** Check whether given space id is undo tablespace id
|
|
|
|
|
@param[in] space_id space id to check
|
|
|
|
|
@return true if it is undo tablespace else false. */
|
|
|
|
|
inline bool srv_is_undo_tablespace(ulint space_id)
|
|
|
|
|
{
|
|
|
|
|
return srv_undo_space_id_start > 0 &&
|
|
|
|
|
space_id >= srv_undo_space_id_start &&
|
|
|
|
|
space_id < srv_undo_space_id_start + srv_undo_tablespaces_open;
|
|
|
|
|
}
|
|
|
|
|
|
2016-08-12 11:17:45 +03:00
|
|
|
class page_id_t;
|
2016-09-06 09:43:16 +03:00
|
|
|
|
2017-03-30 12:48:42 +02:00
|
|
|
/** Structure containing encryption specification */
|
|
|
|
|
struct fil_space_crypt_t;
|
|
|
|
|
|
2016-08-12 11:17:45 +03:00
|
|
|
/** File types */
|
|
|
|
|
enum fil_type_t {
|
|
|
|
|
/** temporary tablespace (temporary undo log or tables) */
|
|
|
|
|
FIL_TYPE_TEMPORARY,
|
|
|
|
|
/** a tablespace that is being imported (no logging until finished) */
|
|
|
|
|
FIL_TYPE_IMPORT,
|
|
|
|
|
/** persistent tablespace (for system, undo log or tables) */
|
|
|
|
|
FIL_TYPE_TABLESPACE,
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
struct fil_node_t;
|
|
|
|
|
|
2020-06-11 22:52:47 +05:30
|
|
|
/** Structure to store first and last value of range */
|
|
|
|
|
struct range_t
|
|
|
|
|
{
|
|
|
|
|
uint32_t first;
|
|
|
|
|
uint32_t last;
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
/** Sort the range based on first value of the range */
|
|
|
|
|
struct range_compare
|
|
|
|
|
{
|
|
|
|
|
bool operator() (const range_t lhs, const range_t rhs) const
|
|
|
|
|
{
|
|
|
|
|
return lhs.first < rhs.first;
|
|
|
|
|
}
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
using range_set_t= std::set<range_t, range_compare>;
|
|
|
|
|
/** Range to store the set of ranges of integers */
|
|
|
|
|
class range_set
|
|
|
|
|
{
|
|
|
|
|
private:
|
|
|
|
|
range_set_t ranges;
|
2020-07-17 19:56:33 +05:30
|
|
|
|
|
|
|
|
range_set_t::iterator find(uint32_t value) const
|
|
|
|
|
{
|
|
|
|
|
auto r_offset= ranges.lower_bound({value, value});
|
|
|
|
|
const auto r_end= ranges.end();
|
|
|
|
|
if (r_offset != r_end);
|
|
|
|
|
else if (empty())
|
|
|
|
|
return r_end;
|
|
|
|
|
else
|
|
|
|
|
r_offset= std::prev(r_end);
|
|
|
|
|
if (r_offset->first <= value && r_offset->last >= value)
|
|
|
|
|
return r_offset;
|
|
|
|
|
return r_end;
|
|
|
|
|
}
|
2020-06-11 22:52:47 +05:30
|
|
|
public:
|
|
|
|
|
/** Merge the current range with previous range.
|
|
|
|
|
@param[in] range range to be merged
|
|
|
|
|
@param[in] prev_range range to be merged with next */
|
|
|
|
|
void merge_range(range_set_t::iterator range,
|
|
|
|
|
range_set_t::iterator prev_range)
|
|
|
|
|
{
|
|
|
|
|
if (range->first != prev_range->last + 1)
|
|
|
|
|
return;
|
|
|
|
|
|
|
|
|
|
/* Merge the current range with previous range */
|
|
|
|
|
range_t new_range {prev_range->first, range->last};
|
|
|
|
|
ranges.erase(prev_range);
|
|
|
|
|
ranges.erase(range);
|
|
|
|
|
ranges.emplace(new_range);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/** Split the range and add two more ranges
|
|
|
|
|
@param[in] range range to be split
|
|
|
|
|
@param[in] value Value to be removed from range */
|
|
|
|
|
void split_range(range_set_t::iterator range, uint32_t value)
|
|
|
|
|
{
|
|
|
|
|
range_t split1{range->first, value - 1};
|
|
|
|
|
range_t split2{value + 1, range->last};
|
|
|
|
|
|
|
|
|
|
/* Remove the existing element */
|
|
|
|
|
ranges.erase(range);
|
|
|
|
|
|
|
|
|
|
/* Insert the two elements */
|
|
|
|
|
ranges.emplace(split1);
|
|
|
|
|
ranges.emplace(split2);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/** Remove the value with the given range
|
|
|
|
|
@param[in,out] range range to be changed
|
|
|
|
|
@param[in] value value to be removed */
|
|
|
|
|
void remove_within_range(range_set_t::iterator range, uint32_t value)
|
|
|
|
|
{
|
|
|
|
|
range_t new_range{range->first, range->last};
|
|
|
|
|
if (value == range->first)
|
|
|
|
|
{
|
|
|
|
|
if (range->first == range->last)
|
|
|
|
|
{
|
|
|
|
|
ranges.erase(range);
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
else
|
|
|
|
|
new_range.first++;
|
|
|
|
|
}
|
|
|
|
|
else if (value == range->last)
|
|
|
|
|
new_range.last--;
|
|
|
|
|
else if (range->first < value && range->last > value)
|
|
|
|
|
return split_range(range, value);
|
|
|
|
|
|
|
|
|
|
ranges.erase(range);
|
|
|
|
|
ranges.emplace(new_range);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/** Remove the value from the ranges.
|
|
|
|
|
@param[in] value Value to be removed. */
|
|
|
|
|
void remove_value(uint32_t value)
|
|
|
|
|
{
|
2020-07-17 19:56:33 +05:30
|
|
|
if (empty())
|
2020-06-11 22:52:47 +05:30
|
|
|
return;
|
|
|
|
|
range_t new_range {value, value};
|
|
|
|
|
range_set_t::iterator range= ranges.lower_bound(new_range);
|
|
|
|
|
if (range == ranges.end())
|
|
|
|
|
return remove_within_range(std::prev(range), value);
|
|
|
|
|
|
|
|
|
|
if (range->first > value && range != ranges.begin())
|
|
|
|
|
/* Iterate the previous ranges to delete */
|
|
|
|
|
return remove_within_range(std::prev(range), value);
|
|
|
|
|
return remove_within_range(range, value);
|
|
|
|
|
}
|
|
|
|
|
/** Add the value within the existing range
|
2020-06-12 21:57:12 +05:30
|
|
|
@param[in] range range to be modified
|
2020-06-11 22:52:47 +05:30
|
|
|
@param[in] value value to be added */
|
|
|
|
|
range_set_t::iterator add_within_range(range_set_t::iterator range,
|
|
|
|
|
uint32_t value)
|
|
|
|
|
{
|
|
|
|
|
if (range->first <= value && range->last >= value)
|
|
|
|
|
return range;
|
|
|
|
|
|
|
|
|
|
range_t new_range{range->first, range->last};
|
|
|
|
|
if (range->last + 1 == value)
|
|
|
|
|
new_range.last++;
|
|
|
|
|
else if (range->first - 1 == value)
|
|
|
|
|
new_range.first--;
|
|
|
|
|
else return ranges.end();
|
|
|
|
|
ranges.erase(range);
|
|
|
|
|
return ranges.emplace(new_range).first;
|
|
|
|
|
}
|
|
|
|
|
/** Add the range in the ranges set
|
|
|
|
|
@param[in] new_range range to be added */
|
|
|
|
|
void add_range(range_t new_range)
|
|
|
|
|
{
|
|
|
|
|
auto r_offset= ranges.lower_bound(new_range);
|
|
|
|
|
auto r_begin= ranges.begin();
|
|
|
|
|
auto r_end= ranges.end();
|
|
|
|
|
if (!ranges.size())
|
|
|
|
|
{
|
|
|
|
|
new_range:
|
|
|
|
|
ranges.emplace(new_range);
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (r_offset == r_end)
|
|
|
|
|
{
|
|
|
|
|
/* last range */
|
|
|
|
|
if (add_within_range(std::prev(r_offset), new_range.first) == r_end)
|
|
|
|
|
goto new_range;
|
|
|
|
|
}
|
|
|
|
|
else if (r_offset == r_begin)
|
|
|
|
|
{
|
|
|
|
|
/* First range */
|
|
|
|
|
if (add_within_range(r_offset, new_range.first) == r_end)
|
|
|
|
|
goto new_range;
|
|
|
|
|
}
|
|
|
|
|
else if (r_offset->first - 1 == new_range.first)
|
|
|
|
|
{
|
|
|
|
|
/* Change starting of the existing range */
|
|
|
|
|
auto r_value= add_within_range(r_offset, new_range.first);
|
|
|
|
|
if (r_value != ranges.begin())
|
|
|
|
|
merge_range(r_value, std::prev(r_value));
|
|
|
|
|
}
|
|
|
|
|
else
|
|
|
|
|
{
|
|
|
|
|
/* previous range last_value alone */
|
|
|
|
|
if (add_within_range(std::prev(r_offset), new_range.first) == r_end)
|
|
|
|
|
goto new_range;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/** Add the value in the ranges
|
|
|
|
|
@param[in] value value to be added */
|
|
|
|
|
void add_value(uint32_t value)
|
|
|
|
|
{
|
|
|
|
|
range_t new_range{value, value};
|
|
|
|
|
add_range(new_range);
|
|
|
|
|
}
|
|
|
|
|
|
2020-07-17 19:56:33 +05:30
|
|
|
bool remove_if_exists(uint32_t value)
|
|
|
|
|
{
|
|
|
|
|
auto r_offset= find(value);
|
|
|
|
|
if (r_offset != ranges.end())
|
|
|
|
|
{
|
|
|
|
|
remove_within_range(r_offset, value);
|
|
|
|
|
return true;
|
|
|
|
|
}
|
|
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
bool contains(uint32_t value) const
|
|
|
|
|
{
|
|
|
|
|
return find(value) != ranges.end();
|
|
|
|
|
}
|
|
|
|
|
|
2020-06-11 22:52:47 +05:30
|
|
|
ulint size() { return ranges.size(); }
|
|
|
|
|
void clear() { ranges.clear(); }
|
|
|
|
|
bool empty() const { return ranges.empty(); }
|
|
|
|
|
typename range_set_t::iterator begin() { return ranges.begin(); }
|
|
|
|
|
typename range_set_t::iterator end() { return ranges.end(); }
|
|
|
|
|
};
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
#endif
|
|
|
|
|
|
2016-08-12 11:17:45 +03:00
|
|
|
/** Tablespace or log data space */
|
2020-01-07 13:23:41 +02:00
|
|
|
#ifndef UNIV_INNOCHECKSUM
|
2020-10-26 15:59:30 +02:00
|
|
|
struct fil_io_t
|
|
|
|
|
{
|
|
|
|
|
/** error code */
|
|
|
|
|
dberr_t err;
|
2020-10-26 16:04:12 +02:00
|
|
|
/** file; node->space->release() must follow IORequestRead call */
|
2020-10-26 15:59:30 +02:00
|
|
|
fil_node_t *node;
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
/** Tablespace encryption mode */
|
|
|
|
|
enum fil_encryption_t
|
|
|
|
|
{
|
|
|
|
|
/** Encrypted if innodb_encrypt_tables=ON (srv_encrypt_tables) */
|
|
|
|
|
FIL_ENCRYPTION_DEFAULT,
|
|
|
|
|
/** Encrypted */
|
|
|
|
|
FIL_ENCRYPTION_ON,
|
|
|
|
|
/** Not encrypted */
|
|
|
|
|
FIL_ENCRYPTION_OFF
|
|
|
|
|
};
|
|
|
|
|
|
2021-03-20 01:31:01 +03:00
|
|
|
struct fil_space_t final : ilist_node<unflushed_spaces_tag_t>,
|
|
|
|
|
ilist_node<rotation_list_tag_t>,
|
|
|
|
|
ilist_node<space_list_tag_t>,
|
|
|
|
|
ilist_node<named_spaces_tag_t>
|
2020-01-07 13:23:41 +02:00
|
|
|
#else
|
2020-10-26 16:04:12 +02:00
|
|
|
struct fil_space_t final
|
2020-01-07 13:23:41 +02:00
|
|
|
#endif
|
2019-12-23 01:18:39 +08:00
|
|
|
{
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
#ifndef UNIV_INNOCHECKSUM
|
2020-10-26 16:04:12 +02:00
|
|
|
friend fil_node_t;
|
2020-11-20 12:30:55 +02:00
|
|
|
~fil_space_t()
|
|
|
|
|
{
|
|
|
|
|
ut_ad(!latch_owner);
|
|
|
|
|
ut_ad(!latch_count);
|
|
|
|
|
latch.destroy();
|
|
|
|
|
}
|
MDEV-25691: Simplify handlerton::drop_database for InnoDB
The implementation of handlerton::drop_database in InnoDB is
unnecessarily complex. The minimal implementation should check
that no conflicting locks or references exist on the tables,
delete all table metadata in a single transaction, and finally
delete the tablespaces.
Note: DROP DATABASE will delete each individual table that the
SQL layer knows about, one table per transaction.
The handlerton::drop_database is basically a final cleanup step
for removing any garbage that could have been left behind
in InnoDB due to some bug, or not having atomic DDL in the past.
hash_node_t: Remove. Use the proper data type name in pointers.
dict_drop_index_tree(): Do not take the table as a parameter.
Instead, return the tablespace ID if the tablespace should be dropped
(we are dropping a clustered index tree).
fil_delete_tablespace(), fil_system_t::detach(): Return a single
detached file handle. Multi-file tablespaces cannot be deleted
via this interface.
ha_innobase::delete_table(): Remove a work-around for non-atomic DDL
and do not try to drop tables with similar-looking name.
innodb_drop_database(): Complete rewrite.
innobase_drop_database(), dict_get_first_table_name_in_db(),
row_drop_database_for_mysql(), drop_all_foreign_keys_in_db(): Remove.
row_purge_remove_clust_if_poss_low(), row_undo_ins_remove_clust_rec():
If the tablespace is to be deleted, try to evict the table definition
from the cache. Failing that, set dict_table_t::space to nullptr.
lock_release_on_rollback(): On the rollback of CREATE TABLE, release all
locks that the transaction had on the table, to avoid heap-use-after-free.
2021-05-18 12:53:40 +03:00
|
|
|
|
2016-08-12 11:17:45 +03:00
|
|
|
ulint id; /*!< space id */
|
MDEV-25691: Simplify handlerton::drop_database for InnoDB
The implementation of handlerton::drop_database in InnoDB is
unnecessarily complex. The minimal implementation should check
that no conflicting locks or references exist on the tables,
delete all table metadata in a single transaction, and finally
delete the tablespaces.
Note: DROP DATABASE will delete each individual table that the
SQL layer knows about, one table per transaction.
The handlerton::drop_database is basically a final cleanup step
for removing any garbage that could have been left behind
in InnoDB due to some bug, or not having atomic DDL in the past.
hash_node_t: Remove. Use the proper data type name in pointers.
dict_drop_index_tree(): Do not take the table as a parameter.
Instead, return the tablespace ID if the tablespace should be dropped
(we are dropping a clustered index tree).
fil_delete_tablespace(), fil_system_t::detach(): Return a single
detached file handle. Multi-file tablespaces cannot be deleted
via this interface.
ha_innobase::delete_table(): Remove a work-around for non-atomic DDL
and do not try to drop tables with similar-looking name.
innodb_drop_database(): Complete rewrite.
innobase_drop_database(), dict_get_first_table_name_in_db(),
row_drop_database_for_mysql(), drop_all_foreign_keys_in_db(): Remove.
row_purge_remove_clust_if_poss_low(), row_undo_ins_remove_clust_rec():
If the tablespace is to be deleted, try to evict the table definition
from the cache. Failing that, set dict_table_t::space to nullptr.
lock_release_on_rollback(): On the rollback of CREATE TABLE, release all
locks that the transaction had on the table, to avoid heap-use-after-free.
2021-05-18 12:53:40 +03:00
|
|
|
|
|
|
|
|
/** fil_system.spaces chain node */
|
|
|
|
|
fil_space_t *hash;
|
2016-08-12 11:17:45 +03:00
|
|
|
lsn_t max_lsn;
|
|
|
|
|
/*!< LSN of the most recent
|
|
|
|
|
fil_names_write_if_was_clean().
|
|
|
|
|
Reset to 0 by fil_names_clear().
|
2018-04-27 10:06:14 +03:00
|
|
|
Protected by log_sys.mutex.
|
2016-08-12 11:17:45 +03:00
|
|
|
If and only if this is nonzero, the
|
|
|
|
|
tablespace will be in named_spaces. */
|
MDEV-13564 Mariabackup does not work with TRUNCATE
Implement undo tablespace truncation via normal redo logging.
Implement TRUNCATE TABLE as a combination of RENAME to #sql-ib name,
CREATE, and DROP.
Note: Orphan #sql-ib*.ibd may be left behind if MariaDB Server 10.2
is killed before the DROP operation is committed. If MariaDB Server 10.2
is killed during TRUNCATE, it is also possible that the old table
was renamed to #sql-ib*.ibd but the data dictionary will refer to the
table using the original name.
In MariaDB Server 10.3, RENAME inside InnoDB is transactional,
and #sql-* tables will be dropped on startup. So, this new TRUNCATE
will be fully crash-safe in 10.3.
ha_mroonga::wrapper_truncate(): Pass table options to the underlying
storage engine, now that ha_innobase::truncate() will need them.
rpl_slave_state::truncate_state_table(): Before truncating
mysql.gtid_slave_pos, evict any cached table handles from
the table definition cache, so that there will be no stale
references to the old table after truncating.
== TRUNCATE TABLE ==
WL#6501 in MySQL 5.7 introduced separate log files for implementing
atomic and crash-safe TRUNCATE TABLE, instead of using the InnoDB
undo and redo log. Some convoluted logic was added to the InnoDB
crash recovery, and some extra synchronization (including a redo log
checkpoint) was introduced to make this work. This synchronization
has caused performance problems and race conditions, and the extra
log files cannot be copied or applied by external backup programs.
In order to support crash-upgrade from MariaDB 10.2, we will keep
the logic for parsing and applying the extra log files, but we will
no longer generate those files in TRUNCATE TABLE.
A prerequisite for crash-safe TRUNCATE is a crash-safe RENAME TABLE
(with full redo and undo logging and proper rollback). This will
be implemented in MDEV-14717.
ha_innobase::truncate(): Invoke RENAME, create(), delete_table().
Because RENAME cannot be fully rolled back before MariaDB 10.3
due to missing undo logging, add some explicit rename-back in
case the operation fails.
ha_innobase::delete(): Introduce a variant that takes sqlcom as
a parameter. In TRUNCATE TABLE, we do not want to touch any
FOREIGN KEY constraints.
ha_innobase::create(): Add the parameters file_per_table, trx.
In TRUNCATE, the new table must be created in the same transaction
that renames the old table.
create_table_info_t::create_table_info_t(): Add the parameters
file_per_table, trx.
row_drop_table_for_mysql(): Replace a bool parameter with sqlcom.
row_drop_table_after_create_fail(): New function, wrapping
row_drop_table_for_mysql().
dict_truncate_index_tree_in_mem(), fil_truncate_tablespace(),
fil_prepare_for_truncate(), fil_reinit_space_header_for_table(),
row_truncate_table_for_mysql(), TruncateLogger,
row_truncate_prepare(), row_truncate_rollback(),
row_truncate_complete(), row_truncate_fts(),
row_truncate_update_system_tables(),
row_truncate_foreign_key_checks(), row_truncate_sanity_checks():
Remove.
row_upd_check_references_constraints(): Remove a check for
TRUNCATE, now that the table is no longer truncated in place.
The new test innodb.truncate_foreign uses DEBUG_SYNC to cover some
race-condition like scenarios. The test innodb-innodb.truncate does
not use any synchronization.
We add a redo log subformat to indicate backup-friendly format.
MariaDB 10.4 will remove support for the old TRUNCATE logging,
so crash-upgrade from old 10.2 or 10.3 to 10.4 will involve
limitations.
== Undo tablespace truncation ==
MySQL 5.7 implements undo tablespace truncation. It is only
possible when innodb_undo_tablespaces is set to at least 2.
The logging is implemented similar to the WL#6501 TRUNCATE,
that is, using separate log files and a redo log checkpoint.
We can simply implement undo tablespace truncation within
a single mini-transaction that reinitializes the undo log
tablespace file. Unfortunately, due to the redo log format
of some operations, currently, the total redo log written by
undo tablespace truncation will be more than the combined size
of the truncated undo tablespace. It should be acceptable
to have a little more than 1 megabyte of log in a single
mini-transaction. This will be fixed in MDEV-17138 in
MariaDB Server 10.4.
recv_sys_t: Add truncated_undo_spaces[] to remember for which undo
tablespaces a MLOG_FILE_CREATE2 record was seen.
namespace undo: Remove some unnecessary declarations.
fil_space_t::is_being_truncated: Document that this flag now
only applies to undo tablespaces. Remove some references.
fil_space_t::is_stopping(): Do not refer to is_being_truncated.
This check is for tablespaces of tables. Potentially used
tablespaces are never truncated any more.
buf_dblwr_process(): Suppress the out-of-bounds warning
for undo tablespaces.
fil_truncate_log(): Write a MLOG_FILE_CREATE2 with a nonzero
page number (new size of the tablespace in pages) to inform
crash recovery that the undo tablespace size has been reduced.
fil_op_write_log(): Relax assertions, so that MLOG_FILE_CREATE2
can be written for undo tablespaces (without .ibd file suffix)
for a nonzero page number.
os_file_truncate(): Add the parameter allow_shrink=false
so that undo tablespaces can actually be shrunk using this function.
fil_name_parse(): For undo tablespace truncation,
buffer MLOG_FILE_CREATE2 in truncated_undo_spaces[].
recv_read_in_area(): Avoid reading pages for which no redo log
records remain buffered, after recv_addr_trim() removed them.
trx_rseg_header_create(): Add a FIXME comment that we could write
much less redo log.
trx_undo_truncate_tablespace(): Reinitialize the undo tablespace
in a single mini-transaction, which will be flushed to the redo log
before the file size is trimmed.
recv_addr_trim(): Discard any redo logs for pages that were
logged after the new end of a file, before the truncation LSN.
If the rec_list becomes empty, reduce n_addrs. After removing
any affected records, actually truncate the file.
recv_apply_hashed_log_recs(): Invoke recv_addr_trim() right before
applying any log records. The undo tablespace files must be open
at this point.
buf_flush_or_remove_pages(), buf_flush_dirty_pages(),
buf_LRU_flush_or_remove_pages(): Add a parameter for specifying
the number of the first page to flush or remove (default 0).
trx_purge_initiate_truncate(): Remove the log checkpoints, the
extra logging, and some unnecessary crash points. Merge the code
from trx_undo_truncate_tablespace(). First, flush all to-be-discarded
pages (beyond the new end of the file), then trim the space->size
to make the page allocation deterministic. At the only remaining
crash injection point, flush the redo log, so that the recovery
can be tested.
2018-08-28 13:43:06 +03:00
|
|
|
/** whether undo tablespace truncation is in progress */
|
2016-08-12 11:17:45 +03:00
|
|
|
bool is_being_truncated;
|
|
|
|
|
fil_type_t purpose;/*!< purpose */
|
|
|
|
|
UT_LIST_BASE_NODE_T(fil_node_t) chain;
|
|
|
|
|
/*!< base node for the file chain */
|
2020-10-15 16:28:19 +03:00
|
|
|
uint32_t size; /*!< tablespace file size in pages;
|
2016-08-12 11:17:45 +03:00
|
|
|
0 if not known yet */
|
2020-10-15 16:28:19 +03:00
|
|
|
uint32_t size_in_header;
|
2016-08-12 11:17:45 +03:00
|
|
|
/* FSP_SIZE in the tablespace header;
|
|
|
|
|
0 if not known yet */
|
2020-10-15 16:28:19 +03:00
|
|
|
uint32_t free_len;
|
2016-08-12 11:17:45 +03:00
|
|
|
/*!< length of the FSP_FREE list */
|
2020-10-15 16:28:19 +03:00
|
|
|
uint32_t free_limit;
|
2016-08-12 11:17:45 +03:00
|
|
|
/*!< contents of FSP_FREE_LIMIT */
|
2020-10-15 16:28:19 +03:00
|
|
|
uint32_t recv_size;
|
2016-12-30 13:48:22 +02:00
|
|
|
/*!< recovered tablespace size in pages;
|
|
|
|
|
0 if no size change was read from the redo log,
|
|
|
|
|
or if the size change was implemented */
|
2020-10-26 16:04:12 +02:00
|
|
|
uint32_t n_reserved_extents;
|
2016-08-12 11:17:45 +03:00
|
|
|
/*!< number of reserved free extents for
|
|
|
|
|
ongoing operations like B-tree page split */
|
2020-09-03 14:49:11 +03:00
|
|
|
private:
|
2021-03-20 01:31:01 +03:00
|
|
|
#ifdef UNIV_DEBUG
|
|
|
|
|
fil_space_t *next_in_space_list();
|
|
|
|
|
fil_space_t *prev_in_space_list();
|
|
|
|
|
|
|
|
|
|
fil_space_t *next_in_unflushed_spaces();
|
|
|
|
|
fil_space_t *prev_in_unflushed_spaces();
|
|
|
|
|
#endif
|
|
|
|
|
|
2020-10-26 15:59:30 +02:00
|
|
|
/** the committed size of the tablespace in pages */
|
|
|
|
|
Atomic_relaxed<uint32_t> committed_size;
|
2020-10-26 16:04:12 +02:00
|
|
|
/** Number of pending operations on the file.
|
|
|
|
|
The tablespace cannot be freed while (n_pending & PENDING) != 0. */
|
|
|
|
|
std::atomic<uint32_t> n_pending;
|
|
|
|
|
/** Flag in n_pending that indicates that the tablespace is being
|
2020-09-03 14:49:11 +03:00
|
|
|
deleted, and no further operations should be performed */
|
2020-10-26 16:04:12 +02:00
|
|
|
static constexpr uint32_t STOPPING= 1U << 31;
|
|
|
|
|
/** Flag in n_pending that indicates that the tablespace is a candidate
|
2020-10-26 15:59:30 +02:00
|
|
|
for being closed, and fil_node_t::is_open() can only be trusted after
|
|
|
|
|
acquiring fil_system.mutex and resetting the flag */
|
2020-10-26 16:04:12 +02:00
|
|
|
static constexpr uint32_t CLOSING= 1U << 30;
|
|
|
|
|
/** Flag in n_pending that indicates that the tablespace needs fsync().
|
|
|
|
|
This must be the least significant flag bit; @see release_flush() */
|
|
|
|
|
static constexpr uint32_t NEEDS_FSYNC= 1U << 29;
|
|
|
|
|
/** The reference count */
|
|
|
|
|
static constexpr uint32_t PENDING= ~(STOPPING | CLOSING | NEEDS_FSYNC);
|
2020-11-20 12:30:55 +02:00
|
|
|
/** latch protecting all page allocation bitmap pages */
|
|
|
|
|
srw_lock latch;
|
2021-01-14 16:35:05 +02:00
|
|
|
os_thread_id_t latch_owner;
|
2020-11-20 12:30:55 +02:00
|
|
|
ut_d(Atomic_relaxed<uint32_t> latch_count;)
|
2020-09-03 14:49:11 +03:00
|
|
|
public:
|
2016-09-06 09:43:16 +03:00
|
|
|
/** MariaDB encryption data */
|
2017-03-30 12:48:42 +02:00
|
|
|
fil_space_crypt_t* crypt_data;
|
2016-09-06 09:43:16 +03:00
|
|
|
|
2020-05-17 16:24:40 +03:00
|
|
|
/** Checks that this tablespace in a list of unflushed tablespaces. */
|
|
|
|
|
bool is_in_unflushed_spaces;
|
|
|
|
|
|
|
|
|
|
/** Checks that this tablespace needs key rotation. */
|
2021-06-14 19:37:06 +05:30
|
|
|
bool is_in_default_encrypt;
|
2020-05-17 16:24:40 +03:00
|
|
|
|
2020-06-11 22:52:47 +05:30
|
|
|
/** mutex to protect freed ranges */
|
|
|
|
|
std::mutex freed_range_mutex;
|
|
|
|
|
|
|
|
|
|
/** Variables to store freed ranges. This can be used to write
|
|
|
|
|
zeroes/punch the hole in files. Protected by freed_mutex */
|
|
|
|
|
range_set freed_ranges;
|
|
|
|
|
|
|
|
|
|
/** Stores last page freed lsn. Protected by freed_mutex */
|
|
|
|
|
lsn_t last_freed_lsn;
|
|
|
|
|
|
2016-08-12 11:17:45 +03:00
|
|
|
ulint magic_n;/*!< FIL_SPACE_MAGIC_N */
|
2017-03-30 12:48:42 +02:00
|
|
|
|
MDEV-23399: Performance regression with write workloads
The buffer pool refactoring in MDEV-15053 and MDEV-22871 shifted
the performance bottleneck to the page flushing.
The configuration parameters will be changed as follows:
innodb_lru_flush_size=32 (new: how many pages to flush on LRU eviction)
innodb_lru_scan_depth=1536 (old: 1024)
innodb_max_dirty_pages_pct=90 (old: 75)
innodb_max_dirty_pages_pct_lwm=75 (old: 0)
Note: The parameter innodb_lru_scan_depth will only affect LRU
eviction of buffer pool pages when a new page is being allocated. The
page cleaner thread will no longer evict any pages. It used to
guarantee that some pages will remain free in the buffer pool. Now, we
perform that eviction 'on demand' in buf_LRU_get_free_block().
The parameter innodb_lru_scan_depth(srv_LRU_scan_depth) is used as follows:
* When the buffer pool is being shrunk in buf_pool_t::withdraw_blocks()
* As a buf_pool.free limit in buf_LRU_list_batch() for terminating
the flushing that is initiated e.g., by buf_LRU_get_free_block()
The parameter also used to serve as an initial limit for unzip_LRU
eviction (evicting uncompressed page frames while retaining
ROW_FORMAT=COMPRESSED pages), but now we will use a hard-coded limit
of 100 or unlimited for invoking buf_LRU_scan_and_free_block().
The status variables will be changed as follows:
innodb_buffer_pool_pages_flushed: This includes also the count of
innodb_buffer_pool_pages_LRU_flushed and should work reliably,
updated one by one in buf_flush_page() to give more real-time
statistics. The function buf_flush_stats(), which we are removing,
was not called in every code path. For both counters, we will use
regular variables that are incremented in a critical section of
buf_pool.mutex. Note that show_innodb_vars() directly links to the
variables, and reads of the counters will *not* be protected by
buf_pool.mutex, so you cannot get a consistent snapshot of both variables.
The following INFORMATION_SCHEMA.INNODB_METRICS counters will be
removed, because the page cleaner no longer deals with writing or
evicting least recently used pages, and because the single-page writes
have been removed:
* buffer_LRU_batch_flush_avg_time_slot
* buffer_LRU_batch_flush_avg_time_thread
* buffer_LRU_batch_flush_avg_time_est
* buffer_LRU_batch_flush_avg_pass
* buffer_LRU_single_flush_scanned
* buffer_LRU_single_flush_num_scan
* buffer_LRU_single_flush_scanned_per_call
When moving to a single buffer pool instance in MDEV-15058, we missed
some opportunity to simplify the buf_flush_page_cleaner thread. It was
unnecessarily using a mutex and some complex data structures, even
though we always have a single page cleaner thread.
Furthermore, the buf_flush_page_cleaner thread had separate 'recovery'
and 'shutdown' modes where it was waiting to be triggered by some
other thread, adding unnecessary latency and potential for hangs in
relatively rarely executed startup or shutdown code.
The page cleaner was also running two kinds of batches in an
interleaved fashion: "LRU flush" (writing out some least recently used
pages and evicting them on write completion) and the normal batches
that aim to increase the MIN(oldest_modification) in the buffer pool,
to help the log checkpoint advance.
The buf_pool.flush_list flushing was being blocked by
buf_block_t::lock for no good reason. Furthermore, if the FIL_PAGE_LSN
of a page is ahead of log_sys.get_flushed_lsn(), that is, what has
been persistently written to the redo log, we would trigger a log
flush and then resume the page flushing. This would unnecessarily
limit the performance of the page cleaner thread and trigger the
infamous messages "InnoDB: page_cleaner: 1000ms intended loop took 4450ms.
The settings might not be optimal" that were suppressed in
commit d1ab89037a518fcffbc50c24e4bd94e4ec33aed0 unless log_warnings>2.
Our revised algorithm will make log_sys.get_flushed_lsn() advance at
the start of buf_flush_lists(), and then execute a 'best effort' to
write out all pages. The flush batches will skip pages that were modified
since the log was written, or are are currently exclusively locked.
The MDEV-13670 message "page_cleaner: 1000ms intended loop took" message
will be removed, because by design, the buf_flush_page_cleaner() should
not be blocked during a batch for extended periods of time.
We will remove the single-page flushing altogether. Related to this,
the debug parameter innodb_doublewrite_batch_size will be removed,
because all of the doublewrite buffer will be used for flushing
batches. If a page needs to be evicted from the buffer pool and all
100 least recently used pages in the buffer pool have unflushed
changes, buf_LRU_get_free_block() will execute buf_flush_lists() to
write out and evict innodb_lru_flush_size pages. At most one thread
will execute buf_flush_lists() in buf_LRU_get_free_block(); other
threads will wait for that LRU flushing batch to finish.
To improve concurrency, we will replace the InnoDB ib_mutex_t and
os_event_t native mutexes and condition variables in this area of code.
Most notably, this means that the buffer pool mutex (buf_pool.mutex)
is no longer instrumented via any InnoDB interfaces. It will continue
to be instrumented via PERFORMANCE_SCHEMA.
For now, both buf_pool.flush_list_mutex and buf_pool.mutex will be
declared with MY_MUTEX_INIT_FAST (PTHREAD_MUTEX_ADAPTIVE_NP). The critical
sections of buf_pool.flush_list_mutex should be shorter than those for
buf_pool.mutex, because in the worst case, they cover a linear scan of
buf_pool.flush_list, while the worst case of a critical section of
buf_pool.mutex covers a linear scan of the potentially much longer
buf_pool.LRU list.
mysql_mutex_is_owner(), safe_mutex_is_owner(): New predicate, usable
with SAFE_MUTEX. Some InnoDB debug assertions need this predicate
instead of mysql_mutex_assert_owner() or mysql_mutex_assert_not_owner().
buf_pool_t::n_flush_LRU, buf_pool_t::n_flush_list:
Replaces buf_pool_t::init_flush[] and buf_pool_t::n_flush[].
The number of active flush operations.
buf_pool_t::mutex, buf_pool_t::flush_list_mutex: Use mysql_mutex_t
instead of ib_mutex_t, to have native mutexes with PERFORMANCE_SCHEMA
and SAFE_MUTEX instrumentation.
buf_pool_t::done_flush_LRU: Condition variable for !n_flush_LRU.
buf_pool_t::done_flush_list: Condition variable for !n_flush_list.
buf_pool_t::do_flush_list: Condition variable to wake up the
buf_flush_page_cleaner when a log checkpoint needs to be written
or the server is being shut down. Replaces buf_flush_event.
We will keep using timed waits (the page cleaner thread will wake
_at least_ once per second), because the calculations for
innodb_adaptive_flushing depend on fixed time intervals.
buf_dblwr: Allocate statically, and move all code to member functions.
Use a native mutex and condition variable. Remove code to deal with
single-page flushing.
buf_dblwr_check_block(): Make the check debug-only. We were spending
a significant amount of execution time in page_simple_validate_new().
flush_counters_t::unzip_LRU_evicted: Remove.
IORequest: Make more members const. FIXME: m_fil_node should be removed.
buf_flush_sync_lsn: Protect by std::atomic, not page_cleaner.mutex
(which we are removing).
page_cleaner_slot_t, page_cleaner_t: Remove many redundant members.
pc_request_flush_slot(): Replaces pc_request() and pc_flush_slot().
recv_writer_thread: Remove. Recovery works just fine without it, if we
simply invoke buf_flush_sync() at the end of each batch in
recv_sys_t::apply().
recv_recovery_from_checkpoint_finish(): Remove. We can simply call
recv_sys.debug_free() directly.
srv_started_redo: Replaces srv_start_state.
SRV_SHUTDOWN_FLUSH_PHASE: Remove. logs_empty_and_mark_files_at_shutdown()
can communicate with the normal page cleaner loop via the new function
flush_buffer_pool().
buf_flush_remove(): Assert that the calling thread is holding
buf_pool.flush_list_mutex. This removes unnecessary mutex operations
from buf_flush_remove_pages() and buf_flush_dirty_pages(),
which replace buf_LRU_flush_or_remove_pages().
buf_flush_lists(): Renamed from buf_flush_batch(), with simplified
interface. Return the number of flushed pages. Clarified comments and
renamed min_n to max_n. Identify LRU batch by lsn=0. Merge all the functions
buf_flush_start(), buf_flush_batch(), buf_flush_end() directly to this
function, which was their only caller, and remove 2 unnecessary
buf_pool.mutex release/re-acquisition that we used to perform around
the buf_flush_batch() call. At the start, if not all log has been
durably written, wait for a background task to do it, or start a new
task to do it. This allows the log write to run concurrently with our
page flushing batch. Any pages that were skipped due to too recent
FIL_PAGE_LSN or due to them being latched by a writer should be flushed
during the next batch, unless there are further modifications to those
pages. It is possible that a page that we must flush due to small
oldest_modification also carries a recent FIL_PAGE_LSN or is being
constantly modified. In the worst case, all writers would then end up
waiting in log_free_check() to allow the flushing and the checkpoint
to complete.
buf_do_flush_list_batch(): Clarify comments, and rename min_n to max_n.
Cache the last looked up tablespace. If neighbor flushing is not applicable,
invoke buf_flush_page() directly, avoiding a page lookup in between.
buf_flush_space(): Auxiliary function to look up a tablespace for
page flushing.
buf_flush_page(): Defer the computation of space->full_crc32(). Never
call log_write_up_to(), but instead skip persistent pages whose latest
modification (FIL_PAGE_LSN) is newer than the redo log. Also skip
pages on which we cannot acquire a shared latch without waiting.
buf_flush_try_neighbors(): Do not bother checking buf_fix_count
because buf_flush_page() will no longer wait for the page latch.
Take the tablespace as a parameter, and only execute this function
when innodb_flush_neighbors>0. Avoid repeated calls of page_id_t::fold().
buf_flush_relocate_on_flush_list(): Declare as cold, and push down
a condition from the callers.
buf_flush_check_neighbor(): Take id.fold() as a parameter.
buf_flush_sync(): Ensure that the buf_pool.flush_list is empty,
because the flushing batch will skip pages whose modifications have
not yet been written to the log or were latched for modification.
buf_free_from_unzip_LRU_list_batch(): Remove redundant local variables.
buf_flush_LRU_list_batch(): Let the caller buf_do_LRU_batch() initialize
the counters, and report n->evicted.
Cache the last looked up tablespace. If neighbor flushing is not applicable,
invoke buf_flush_page() directly, avoiding a page lookup in between.
buf_do_LRU_batch(): Return the number of pages flushed.
buf_LRU_free_page(): Only release and re-acquire buf_pool.mutex if
adaptive hash index entries are pointing to the block.
buf_LRU_get_free_block(): Do not wake up the page cleaner, because it
will no longer perform any useful work for us, and we do not want it
to compete for I/O while buf_flush_lists(innodb_lru_flush_size, 0)
writes out and evicts at most innodb_lru_flush_size pages. (The
function buf_do_LRU_batch() may complete after writing fewer pages if
more than innodb_lru_scan_depth pages end up in buf_pool.free list.)
Eliminate some mutex release-acquire cycles, and wait for the LRU
flush batch to complete before rescanning.
buf_LRU_check_size_of_non_data_objects(): Simplify the code.
buf_page_write_complete(): Remove the parameter evict, and always
evict pages that were part of an LRU flush.
buf_page_create(): Take a pre-allocated page as a parameter.
buf_pool_t::free_block(): Free a pre-allocated block.
recv_sys_t::recover_low(), recv_sys_t::apply(): Preallocate the block
while not holding recv_sys.mutex. During page allocation, we may
initiate a page flush, which in turn may initiate a log flush, which
would require acquiring log_sys.mutex, which should always be acquired
before recv_sys.mutex in order to avoid deadlocks. Therefore, we must
not be holding recv_sys.mutex while allocating a buffer pool block.
BtrBulk::logFreeCheck(): Skip a redundant condition.
row_undo_step(): Do not invoke srv_inc_activity_count() for every row
that is being rolled back. It should suffice to invoke the function in
trx_flush_log_if_needed() during trx_t::commit_in_memory() when the
rollback completes.
sync_check_enable(): Remove. We will enable innodb_sync_debug from the
very beginning.
Reviewed by: Vladislav Vaintroub
2020-10-15 12:10:42 +03:00
|
|
|
/** @return whether doublewrite buffering is needed */
|
2021-06-28 13:42:43 +03:00
|
|
|
inline bool use_doublewrite() const;
|
2018-03-28 09:29:14 +03:00
|
|
|
|
2018-11-27 14:49:20 +02:00
|
|
|
/** Append a file to the chain of files of a space.
|
|
|
|
|
@param[in] name file name of a file that is not open
|
|
|
|
|
@param[in] handle file handle, or OS_FILE_CLOSED
|
|
|
|
|
@param[in] size file size in entire database pages
|
|
|
|
|
@param[in] is_raw whether this is a raw device
|
|
|
|
|
@param[in] atomic_write true if atomic write could be enabled
|
|
|
|
|
@param[in] max_pages maximum number of pages in file,
|
2020-10-15 16:28:19 +03:00
|
|
|
or UINT32_MAX for unlimited
|
2018-11-27 14:49:20 +02:00
|
|
|
@return file object */
|
|
|
|
|
fil_node_t* add(const char* name, pfs_os_file_t handle,
|
2020-10-15 16:28:19 +03:00
|
|
|
uint32_t size, bool is_raw, bool atomic_write,
|
|
|
|
|
uint32_t max_pages = UINT32_MAX);
|
2019-04-08 14:41:02 +03:00
|
|
|
#ifdef UNIV_DEBUG
|
|
|
|
|
/** Assert that the mini-transaction is compatible with
|
|
|
|
|
updating an allocation bitmap page.
|
|
|
|
|
@param[in] mtr mini-transaction */
|
|
|
|
|
void modify_check(const mtr_t& mtr) const;
|
|
|
|
|
#endif /* UNIV_DEBUG */
|
2018-11-28 01:02:46 +02:00
|
|
|
|
MDEV-12266: Change dict_table_t::space to fil_space_t*
InnoDB always keeps all tablespaces in the fil_system cache.
The fil_system.LRU is only for closing file handles; the
fil_space_t and fil_node_t for all data files will remain
in main memory. Between startup to shutdown, they can only be
created and removed by DDL statements. Therefore, we can
let dict_table_t::space point directly to the fil_space_t.
dict_table_t::space_id: A numeric tablespace ID for the corner cases
where we do not have a tablespace. The most prominent examples are
ALTER TABLE...DISCARD TABLESPACE or a missing or corrupted file.
There are a few functional differences; most notably:
(1) DROP TABLE will delete matching .ibd and .cfg files,
even if they were not attached to the data dictionary.
(2) Some error messages will report file names instead of numeric IDs.
There still are many functions that use numeric tablespace IDs instead
of fil_space_t*, and many functions could be converted to fil_space_t
member functions. Also, Tablespace and Datafile should be merged with
fil_space_t and fil_node_t. page_id_t and buf_page_get_gen() could use
fil_space_t& instead of a numeric ID, and after moving to a single
buffer pool (MDEV-15058), buf_pool_t::page_hash could be moved to
fil_space_t::page_hash.
FilSpace: Remove. Only few calls to fil_space_acquire() will remain,
and gradually they should be removed.
mtr_t::set_named_space_id(ulint): Renamed from set_named_space(),
to prevent accidental calls to this slower function. Very few
callers remain.
fseg_create(), fsp_reserve_free_extents(): Take fil_space_t*
as a parameter instead of a space_id.
fil_space_t::rename(): Wrapper for fil_rename_tablespace_check(),
fil_name_write_rename(), fil_rename_tablespace(). Mariabackup
passes the parameter log=false; InnoDB passes log=true.
dict_mem_table_create(): Take fil_space_t* instead of space_id
as parameter.
dict_process_sys_tables_rec_and_mtr_commit(): Replace the parameter
'status' with 'bool cached'.
dict_get_and_save_data_dir_path(): Avoid copying the fil_node_t::name.
fil_ibd_open(): Return the tablespace.
fil_space_t::set_imported(): Replaces fil_space_set_imported().
truncate_t: Change many member function parameters to fil_space_t*,
and remove page_size parameters.
row_truncate_prepare(): Merge to its only caller.
row_drop_table_from_cache(): Assert that the table is persistent.
dict_create_sys_indexes_tuple(): Write SYS_INDEXES.SPACE=FIL_NULL
if the tablespace has been discarded.
row_import_update_discarded_flag(): Remove a constant parameter.
2018-03-27 16:31:10 +03:00
|
|
|
/** Try to reserve free extents.
|
|
|
|
|
@param[in] n_free_now current number of free extents
|
|
|
|
|
@param[in] n_to_reserve number of extents to reserve
|
|
|
|
|
@return whether the reservation succeeded */
|
2020-10-26 16:04:12 +02:00
|
|
|
bool reserve_free_extents(uint32_t n_free_now, uint32_t n_to_reserve)
|
MDEV-12266: Change dict_table_t::space to fil_space_t*
InnoDB always keeps all tablespaces in the fil_system cache.
The fil_system.LRU is only for closing file handles; the
fil_space_t and fil_node_t for all data files will remain
in main memory. Between startup to shutdown, they can only be
created and removed by DDL statements. Therefore, we can
let dict_table_t::space point directly to the fil_space_t.
dict_table_t::space_id: A numeric tablespace ID for the corner cases
where we do not have a tablespace. The most prominent examples are
ALTER TABLE...DISCARD TABLESPACE or a missing or corrupted file.
There are a few functional differences; most notably:
(1) DROP TABLE will delete matching .ibd and .cfg files,
even if they were not attached to the data dictionary.
(2) Some error messages will report file names instead of numeric IDs.
There still are many functions that use numeric tablespace IDs instead
of fil_space_t*, and many functions could be converted to fil_space_t
member functions. Also, Tablespace and Datafile should be merged with
fil_space_t and fil_node_t. page_id_t and buf_page_get_gen() could use
fil_space_t& instead of a numeric ID, and after moving to a single
buffer pool (MDEV-15058), buf_pool_t::page_hash could be moved to
fil_space_t::page_hash.
FilSpace: Remove. Only few calls to fil_space_acquire() will remain,
and gradually they should be removed.
mtr_t::set_named_space_id(ulint): Renamed from set_named_space(),
to prevent accidental calls to this slower function. Very few
callers remain.
fseg_create(), fsp_reserve_free_extents(): Take fil_space_t*
as a parameter instead of a space_id.
fil_space_t::rename(): Wrapper for fil_rename_tablespace_check(),
fil_name_write_rename(), fil_rename_tablespace(). Mariabackup
passes the parameter log=false; InnoDB passes log=true.
dict_mem_table_create(): Take fil_space_t* instead of space_id
as parameter.
dict_process_sys_tables_rec_and_mtr_commit(): Replace the parameter
'status' with 'bool cached'.
dict_get_and_save_data_dir_path(): Avoid copying the fil_node_t::name.
fil_ibd_open(): Return the tablespace.
fil_space_t::set_imported(): Replaces fil_space_set_imported().
truncate_t: Change many member function parameters to fil_space_t*,
and remove page_size parameters.
row_truncate_prepare(): Merge to its only caller.
row_drop_table_from_cache(): Assert that the table is persistent.
dict_create_sys_indexes_tuple(): Write SYS_INDEXES.SPACE=FIL_NULL
if the tablespace has been discarded.
row_import_update_discarded_flag(): Remove a constant parameter.
2018-03-27 16:31:10 +03:00
|
|
|
{
|
|
|
|
|
if (n_reserved_extents + n_to_reserve > n_free_now) {
|
|
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
n_reserved_extents += n_to_reserve;
|
|
|
|
|
return true;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/** Release the reserved free extents.
|
|
|
|
|
@param[in] n_reserved number of reserved extents */
|
2020-10-26 16:04:12 +02:00
|
|
|
void release_free_extents(uint32_t n_reserved)
|
MDEV-12266: Change dict_table_t::space to fil_space_t*
InnoDB always keeps all tablespaces in the fil_system cache.
The fil_system.LRU is only for closing file handles; the
fil_space_t and fil_node_t for all data files will remain
in main memory. Between startup to shutdown, they can only be
created and removed by DDL statements. Therefore, we can
let dict_table_t::space point directly to the fil_space_t.
dict_table_t::space_id: A numeric tablespace ID for the corner cases
where we do not have a tablespace. The most prominent examples are
ALTER TABLE...DISCARD TABLESPACE or a missing or corrupted file.
There are a few functional differences; most notably:
(1) DROP TABLE will delete matching .ibd and .cfg files,
even if they were not attached to the data dictionary.
(2) Some error messages will report file names instead of numeric IDs.
There still are many functions that use numeric tablespace IDs instead
of fil_space_t*, and many functions could be converted to fil_space_t
member functions. Also, Tablespace and Datafile should be merged with
fil_space_t and fil_node_t. page_id_t and buf_page_get_gen() could use
fil_space_t& instead of a numeric ID, and after moving to a single
buffer pool (MDEV-15058), buf_pool_t::page_hash could be moved to
fil_space_t::page_hash.
FilSpace: Remove. Only few calls to fil_space_acquire() will remain,
and gradually they should be removed.
mtr_t::set_named_space_id(ulint): Renamed from set_named_space(),
to prevent accidental calls to this slower function. Very few
callers remain.
fseg_create(), fsp_reserve_free_extents(): Take fil_space_t*
as a parameter instead of a space_id.
fil_space_t::rename(): Wrapper for fil_rename_tablespace_check(),
fil_name_write_rename(), fil_rename_tablespace(). Mariabackup
passes the parameter log=false; InnoDB passes log=true.
dict_mem_table_create(): Take fil_space_t* instead of space_id
as parameter.
dict_process_sys_tables_rec_and_mtr_commit(): Replace the parameter
'status' with 'bool cached'.
dict_get_and_save_data_dir_path(): Avoid copying the fil_node_t::name.
fil_ibd_open(): Return the tablespace.
fil_space_t::set_imported(): Replaces fil_space_set_imported().
truncate_t: Change many member function parameters to fil_space_t*,
and remove page_size parameters.
row_truncate_prepare(): Merge to its only caller.
row_drop_table_from_cache(): Assert that the table is persistent.
dict_create_sys_indexes_tuple(): Write SYS_INDEXES.SPACE=FIL_NULL
if the tablespace has been discarded.
row_import_update_discarded_flag(): Remove a constant parameter.
2018-03-27 16:31:10 +03:00
|
|
|
{
|
|
|
|
|
if (!n_reserved) return;
|
|
|
|
|
ut_a(n_reserved_extents >= n_reserved);
|
|
|
|
|
n_reserved_extents -= n_reserved;
|
|
|
|
|
}
|
|
|
|
|
|
MDEV-25312 Replace fil_space_t::name with fil_space_t::name()
A consistency check for fil_space_t::name is causing recovery failures
in MDEV-25180 (Atomic ALTER TABLE). So, we'd better remove that field
altogether.
fil_space_t::name was more or less a copy of dict_table_t::name
(except for some special cases), and it was not being used for
anything useful.
There used to be a name_hash, but it had been removed already in
commit a75dbfd7183cc96680f3e3e684fd36500dac8158 (MDEV-12266).
We will also remove os_normalize_path(), OS_PATH_SEPARATOR,
OS_PATH_SEPATOR_ALT. On Microsoft Windows, we will treat \ and /
roughly in the same way. The intention is that for per-table
tablespaces, the filenames will always follow the pattern
prefix/databasename/tablename.ibd. (Any \ in the prefix must not
be converted.)
ut_basename_noext(): Remove (unused function).
read_link_file(): Replaces RemoteDatafile::read_link_file().
We will ensure that the last two path component separators are
forward slashes (converting up to 2 trailing backslashes on
Microsoft Windows), so that everywhere else we can
assume that data file names end in "/databasename/tablename.ibd".
Note: On Microsoft Windows, path names that start with \\?\ must
not contain / as path component separators. Previously, such paths
did work in the DATA DIRECTORY argument of InnoDB tables.
Reviewed by: Vladislav Vaintroub
2021-04-07 18:01:13 +03:00
|
|
|
/** Rename a file.
|
|
|
|
|
@param[in] path tablespace file name after renaming
|
|
|
|
|
@param[in] log whether to write redo log
|
|
|
|
|
@param[in] replace whether to ignore the existence of path
|
|
|
|
|
@return error code
|
|
|
|
|
@retval DB_SUCCESS on success */
|
|
|
|
|
dberr_t rename(const char *path, bool log, bool replace= false)
|
|
|
|
|
MY_ATTRIBUTE((nonnull));
|
MDEV-12266: Change dict_table_t::space to fil_space_t*
InnoDB always keeps all tablespaces in the fil_system cache.
The fil_system.LRU is only for closing file handles; the
fil_space_t and fil_node_t for all data files will remain
in main memory. Between startup to shutdown, they can only be
created and removed by DDL statements. Therefore, we can
let dict_table_t::space point directly to the fil_space_t.
dict_table_t::space_id: A numeric tablespace ID for the corner cases
where we do not have a tablespace. The most prominent examples are
ALTER TABLE...DISCARD TABLESPACE or a missing or corrupted file.
There are a few functional differences; most notably:
(1) DROP TABLE will delete matching .ibd and .cfg files,
even if they were not attached to the data dictionary.
(2) Some error messages will report file names instead of numeric IDs.
There still are many functions that use numeric tablespace IDs instead
of fil_space_t*, and many functions could be converted to fil_space_t
member functions. Also, Tablespace and Datafile should be merged with
fil_space_t and fil_node_t. page_id_t and buf_page_get_gen() could use
fil_space_t& instead of a numeric ID, and after moving to a single
buffer pool (MDEV-15058), buf_pool_t::page_hash could be moved to
fil_space_t::page_hash.
FilSpace: Remove. Only few calls to fil_space_acquire() will remain,
and gradually they should be removed.
mtr_t::set_named_space_id(ulint): Renamed from set_named_space(),
to prevent accidental calls to this slower function. Very few
callers remain.
fseg_create(), fsp_reserve_free_extents(): Take fil_space_t*
as a parameter instead of a space_id.
fil_space_t::rename(): Wrapper for fil_rename_tablespace_check(),
fil_name_write_rename(), fil_rename_tablespace(). Mariabackup
passes the parameter log=false; InnoDB passes log=true.
dict_mem_table_create(): Take fil_space_t* instead of space_id
as parameter.
dict_process_sys_tables_rec_and_mtr_commit(): Replace the parameter
'status' with 'bool cached'.
dict_get_and_save_data_dir_path(): Avoid copying the fil_node_t::name.
fil_ibd_open(): Return the tablespace.
fil_space_t::set_imported(): Replaces fil_space_set_imported().
truncate_t: Change many member function parameters to fil_space_t*,
and remove page_size parameters.
row_truncate_prepare(): Merge to its only caller.
row_drop_table_from_cache(): Assert that the table is persistent.
dict_create_sys_indexes_tuple(): Write SYS_INDEXES.SPACE=FIL_NULL
if the tablespace has been discarded.
row_import_update_discarded_flag(): Remove a constant parameter.
2018-03-27 16:31:10 +03:00
|
|
|
|
2020-10-26 15:59:30 +02:00
|
|
|
/** Note that the tablespace has been imported.
|
|
|
|
|
Initially, purpose=FIL_TYPE_IMPORT so that no redo log is
|
|
|
|
|
written while the space ID is being updated in each page. */
|
|
|
|
|
inline void set_imported();
|
MDEV-17380 innodb_flush_neighbors=ON should be ignored on SSD
For tablespaces that do not reside on spinning storage, it does
not make sense to attempt to write nearby pages when writing out
dirty pages from the InnoDB buffer pool. It is actually detrimental
to performance and to the life span of flash ROM storage.
With this change, MariaDB will detect whether an InnoDB file resides
on solid-state storage. The detection has been implemented for Linux
and Microsoft Windows. For other systems, we will err on the safe side
and assume that files reside on SSD.
As part of this change, we will reduce the number of fstat() calls
when opening data files on POSIX systems and slightly clean up some
file I/O code.
FIXME: os_is_sparse_file_supported() on POSIX works in a destructive
manner. Thus, we can only invoke it when creating files, not when
opening them.
For diagnostics, we introduce the column ON_SSD to the table
INFORMATION_SCHEMA.INNODB_TABLESPACES_SCRUBBING. The table
INNODB_SYS_TABLESPACES might seem more appropriate, but its purpose
is to reflect the contents of the InnoDB system table SYS_TABLESPACES,
which we would like to remove at some point.
On Microsoft Windows, querying StorageDeviceSeekPenaltyProperty
sometimes returns ERROR_GEN_FAILURE instead of ERROR_INVALID_FUNCTION
or ERROR_NOT_SUPPORTED. We will silently ignore also this error,
and assume that the file does not reside on SSD.
On Linux, the detection will be based on the files
/sys/block/*/queue/rotational and /sys/block/*/dev.
Especially for USB storage, it is possible that
/sys/block/*/queue/rotational will wrongly report 1 instead of 0.
fil_node_t::on_ssd: Whether the InnoDB data file resides on
solid-state storage.
fil_system_t::ssd: Collection of Linux block devices that reside on
non-rotational storage.
fil_system_t::create(): Detect ssd on Linux based on the contents
of /sys/block/*/queue/rotational and /sys/block/*/dev.
fil_system_t::is_ssd(dev_t): Determine if a Linux block device is
non-rotational. Partitions will be identified with the containing
block device by assuming that the least significant 4 bits of the
minor number identify a partition, and that the "partition number"
of the entire device is 0.
2019-04-01 11:57:06 +03:00
|
|
|
|
2020-10-26 15:59:30 +02:00
|
|
|
/** @return whether the storage device is rotational (HDD, not SSD) */
|
|
|
|
|
inline bool is_rotational() const;
|
MDEV-12266: Change dict_table_t::space to fil_space_t*
InnoDB always keeps all tablespaces in the fil_system cache.
The fil_system.LRU is only for closing file handles; the
fil_space_t and fil_node_t for all data files will remain
in main memory. Between startup to shutdown, they can only be
created and removed by DDL statements. Therefore, we can
let dict_table_t::space point directly to the fil_space_t.
dict_table_t::space_id: A numeric tablespace ID for the corner cases
where we do not have a tablespace. The most prominent examples are
ALTER TABLE...DISCARD TABLESPACE or a missing or corrupted file.
There are a few functional differences; most notably:
(1) DROP TABLE will delete matching .ibd and .cfg files,
even if they were not attached to the data dictionary.
(2) Some error messages will report file names instead of numeric IDs.
There still are many functions that use numeric tablespace IDs instead
of fil_space_t*, and many functions could be converted to fil_space_t
member functions. Also, Tablespace and Datafile should be merged with
fil_space_t and fil_node_t. page_id_t and buf_page_get_gen() could use
fil_space_t& instead of a numeric ID, and after moving to a single
buffer pool (MDEV-15058), buf_pool_t::page_hash could be moved to
fil_space_t::page_hash.
FilSpace: Remove. Only few calls to fil_space_acquire() will remain,
and gradually they should be removed.
mtr_t::set_named_space_id(ulint): Renamed from set_named_space(),
to prevent accidental calls to this slower function. Very few
callers remain.
fseg_create(), fsp_reserve_free_extents(): Take fil_space_t*
as a parameter instead of a space_id.
fil_space_t::rename(): Wrapper for fil_rename_tablespace_check(),
fil_name_write_rename(), fil_rename_tablespace(). Mariabackup
passes the parameter log=false; InnoDB passes log=true.
dict_mem_table_create(): Take fil_space_t* instead of space_id
as parameter.
dict_process_sys_tables_rec_and_mtr_commit(): Replace the parameter
'status' with 'bool cached'.
dict_get_and_save_data_dir_path(): Avoid copying the fil_node_t::name.
fil_ibd_open(): Return the tablespace.
fil_space_t::set_imported(): Replaces fil_space_set_imported().
truncate_t: Change many member function parameters to fil_space_t*,
and remove page_size parameters.
row_truncate_prepare(): Merge to its only caller.
row_drop_table_from_cache(): Assert that the table is persistent.
dict_create_sys_indexes_tuple(): Write SYS_INDEXES.SPACE=FIL_NULL
if the tablespace has been discarded.
row_import_update_discarded_flag(): Remove a constant parameter.
2018-03-27 16:31:10 +03:00
|
|
|
|
2021-06-28 13:42:43 +03:00
|
|
|
/** whether the tablespace discovery is being deferred during crash
|
|
|
|
|
recovery due to incompletely written page 0 */
|
2021-05-17 18:12:33 +03:00
|
|
|
inline bool is_deferred() const;
|
|
|
|
|
|
2020-10-26 15:59:30 +02:00
|
|
|
/** Open each file. Never invoked on .ibd files.
|
|
|
|
|
@param create_new_db whether to skip the call to fil_node_t::read_page0()
|
|
|
|
|
@return whether all files were opened */
|
|
|
|
|
bool open(bool create_new_db);
|
|
|
|
|
/** Close each file. Only invoked on fil_system.temp_space. */
|
|
|
|
|
void close();
|
2018-04-23 13:15:54 +03:00
|
|
|
|
MDEV-25506 (3 of 3): Do not delete .ibd files before commit
This is a complete rewrite of DROP TABLE, also as part of other DDL,
such as ALTER TABLE, CREATE TABLE...SELECT, TRUNCATE TABLE.
The background DROP TABLE queue hack is removed.
If a transaction needs to drop and create a table by the same name
(like TRUNCATE TABLE does), it must first rename the table to an
internal #sql-ib name. No committed version of the data dictionary
will include any #sql-ib tables, because whenever a transaction
renames a table to a #sql-ib name, it will also drop that table.
Either the rename will be rolled back, or the drop will be committed.
Data files will be unlinked after the transaction has been committed
and a FILE_RENAME record has been durably written. The file will
actually be deleted when the detached file handle returned by
fil_delete_tablespace() will be closed, after the latches have been
released. It is possible that a purge of the delete of the SYS_INDEXES
record for the clustered index will execute fil_delete_tablespace()
concurrently with the DDL transaction. In that case, the thread that
arrives later will wait for the other thread to finish.
HTON_TRUNCATE_REQUIRES_EXCLUSIVE_USE: A new handler flag.
ha_innobase::truncate() now requires that all other references to
the table be released in advance. This was implemented by Monty.
ha_innobase::delete_table(): If CREATE TABLE..SELECT is detected,
we will "hijack" the current transaction, drop the table in
the current transaction and commit the current transaction.
This essentially fixes MDEV-21602. There is a FIXME comment about
making the check less failure-prone.
ha_innobase::truncate(), ha_innobase::delete_table():
Implement a fast path for temporary tables. We will no longer allow
temporary tables to use the adaptive hash index.
dict_table_t::mdl_name: The original table name for the purpose of
acquiring MDL in purge, to prevent a race condition between a
DDL transaction that is dropping a table, and purge processing
undo log records of DML that had executed before the DDL operation.
For #sql-backup- tables during ALTER TABLE...ALGORITHM=COPY, the
dict_table_t::mdl_name will differ from dict_table_t::name.
dict_table_t::parse_name(): Use mdl_name instead of name.
dict_table_rename_in_cache(): Update mdl_name.
For the internal FTS_ tables of FULLTEXT INDEX, purge would
acquire MDL on the FTS_ table name, but not on the main table,
and therefore it would be able to run concurrently with a
DDL transaction that is dropping the table. Previously, the
DROP TABLE queue hack prevented a race between purge and DDL.
For now, we introduce purge_sys.stop_FTS() to prevent purge from
opening any table, while a DDL transaction that may drop FTS_
tables is in progress. The function fts_lock_table(), which will
be invoked before the dictionary is locked, will wait for
purge to release any table handles.
trx_t::drop_table_statistics(): Drop statistics for the table.
This replaces dict_stats_drop_index(). We will drop or rename
persistent statistics atomically as part of DDL transactions.
On lock conflict for dropping statistics, we will fail instantly
with DB_LOCK_WAIT_TIMEOUT, because we will be holding the
exclusive data dictionary latch.
trx_t::commit_cleanup(): Separated from trx_t::commit_in_memory().
Relax an assertion around fts_commit() and allow DB_LOCK_WAIT_TIMEOUT
in addition to DB_DUPLICATE_KEY. The call to fts_commit() is
entirely misplaced here and may obviously break the consistency
of transactions that affect FULLTEXT INDEX. It needs to be fixed
separately.
dict_table_t::n_foreign_key_checks_running: Remove (MDEV-21175).
The counter was a work-around for missing meta-data locking (MDL)
on the SQL layer, and not really needed in MariaDB.
ER_TABLE_IN_FK_CHECK: Replaced with ER_UNUSED_28.
HA_ERR_TABLE_IN_FK_CHECK: Remove.
row_ins_check_foreign_constraints(): Do not acquire
dict_sys.latch either. The SQL-layer MDL will protect us.
This was reviewed by Thirunarayanan Balathandayuthapani
and tested by Matthias Leich.
2021-06-09 17:02:55 +03:00
|
|
|
/** Note that operations on the tablespace must stop.
|
|
|
|
|
@return whether the operations were already stopped */
|
2021-10-03 13:49:40 +03:00
|
|
|
inline bool set_stopping_check();
|
|
|
|
|
/** Note that operations on the tablespace must stop. */
|
|
|
|
|
inline void set_stopping();
|
MDEV-25506 (3 of 3): Do not delete .ibd files before commit
This is a complete rewrite of DROP TABLE, also as part of other DDL,
such as ALTER TABLE, CREATE TABLE...SELECT, TRUNCATE TABLE.
The background DROP TABLE queue hack is removed.
If a transaction needs to drop and create a table by the same name
(like TRUNCATE TABLE does), it must first rename the table to an
internal #sql-ib name. No committed version of the data dictionary
will include any #sql-ib tables, because whenever a transaction
renames a table to a #sql-ib name, it will also drop that table.
Either the rename will be rolled back, or the drop will be committed.
Data files will be unlinked after the transaction has been committed
and a FILE_RENAME record has been durably written. The file will
actually be deleted when the detached file handle returned by
fil_delete_tablespace() will be closed, after the latches have been
released. It is possible that a purge of the delete of the SYS_INDEXES
record for the clustered index will execute fil_delete_tablespace()
concurrently with the DDL transaction. In that case, the thread that
arrives later will wait for the other thread to finish.
HTON_TRUNCATE_REQUIRES_EXCLUSIVE_USE: A new handler flag.
ha_innobase::truncate() now requires that all other references to
the table be released in advance. This was implemented by Monty.
ha_innobase::delete_table(): If CREATE TABLE..SELECT is detected,
we will "hijack" the current transaction, drop the table in
the current transaction and commit the current transaction.
This essentially fixes MDEV-21602. There is a FIXME comment about
making the check less failure-prone.
ha_innobase::truncate(), ha_innobase::delete_table():
Implement a fast path for temporary tables. We will no longer allow
temporary tables to use the adaptive hash index.
dict_table_t::mdl_name: The original table name for the purpose of
acquiring MDL in purge, to prevent a race condition between a
DDL transaction that is dropping a table, and purge processing
undo log records of DML that had executed before the DDL operation.
For #sql-backup- tables during ALTER TABLE...ALGORITHM=COPY, the
dict_table_t::mdl_name will differ from dict_table_t::name.
dict_table_t::parse_name(): Use mdl_name instead of name.
dict_table_rename_in_cache(): Update mdl_name.
For the internal FTS_ tables of FULLTEXT INDEX, purge would
acquire MDL on the FTS_ table name, but not on the main table,
and therefore it would be able to run concurrently with a
DDL transaction that is dropping the table. Previously, the
DROP TABLE queue hack prevented a race between purge and DDL.
For now, we introduce purge_sys.stop_FTS() to prevent purge from
opening any table, while a DDL transaction that may drop FTS_
tables is in progress. The function fts_lock_table(), which will
be invoked before the dictionary is locked, will wait for
purge to release any table handles.
trx_t::drop_table_statistics(): Drop statistics for the table.
This replaces dict_stats_drop_index(). We will drop or rename
persistent statistics atomically as part of DDL transactions.
On lock conflict for dropping statistics, we will fail instantly
with DB_LOCK_WAIT_TIMEOUT, because we will be holding the
exclusive data dictionary latch.
trx_t::commit_cleanup(): Separated from trx_t::commit_in_memory().
Relax an assertion around fts_commit() and allow DB_LOCK_WAIT_TIMEOUT
in addition to DB_DUPLICATE_KEY. The call to fts_commit() is
entirely misplaced here and may obviously break the consistency
of transactions that affect FULLTEXT INDEX. It needs to be fixed
separately.
dict_table_t::n_foreign_key_checks_running: Remove (MDEV-21175).
The counter was a work-around for missing meta-data locking (MDL)
on the SQL layer, and not really needed in MariaDB.
ER_TABLE_IN_FK_CHECK: Replaced with ER_UNUSED_28.
HA_ERR_TABLE_IN_FK_CHECK: Remove.
row_ins_check_foreign_constraints(): Do not acquire
dict_sys.latch either. The SQL-layer MDL will protect us.
This was reviewed by Thirunarayanan Balathandayuthapani
and tested by Matthias Leich.
2021-06-09 17:02:55 +03:00
|
|
|
|
|
|
|
|
/** Note that operations on the tablespace can resume after truncation */
|
|
|
|
|
inline void clear_stopping();
|
2020-09-03 14:49:11 +03:00
|
|
|
|
2021-04-23 16:58:25 +03:00
|
|
|
/** Look up the tablespace and wait for pending operations to cease
|
|
|
|
|
@param id tablespace identifier
|
|
|
|
|
@return tablespace
|
|
|
|
|
@retval nullptr if no tablespace was found */
|
|
|
|
|
static fil_space_t *check_pending_operations(ulint id);
|
|
|
|
|
|
2020-10-26 16:04:12 +02:00
|
|
|
private:
|
2020-09-03 14:49:11 +03:00
|
|
|
MY_ATTRIBUTE((warn_unused_result))
|
2020-10-26 16:04:12 +02:00
|
|
|
/** Try to acquire a tablespace reference.
|
|
|
|
|
@return the old reference count (if STOPPING is set, it was not acquired) */
|
|
|
|
|
uint32_t acquire_low()
|
2020-09-03 14:49:11 +03:00
|
|
|
{
|
2020-10-26 15:59:30 +02:00
|
|
|
uint32_t n= 0;
|
2020-10-26 16:04:12 +02:00
|
|
|
while (!n_pending.compare_exchange_strong(n, n + 1,
|
|
|
|
|
std::memory_order_acquire,
|
|
|
|
|
std::memory_order_relaxed) &&
|
|
|
|
|
!(n & STOPPING));
|
|
|
|
|
return n;
|
2020-09-03 14:49:11 +03:00
|
|
|
}
|
2020-10-26 16:04:12 +02:00
|
|
|
public:
|
|
|
|
|
MY_ATTRIBUTE((warn_unused_result))
|
2020-12-18 16:53:45 +02:00
|
|
|
/** Acquire a tablespace reference.
|
|
|
|
|
@return whether a tablespace reference was successfully acquired */
|
|
|
|
|
inline bool acquire_if_not_stopped();
|
2020-10-26 15:59:30 +02:00
|
|
|
|
|
|
|
|
MY_ATTRIBUTE((warn_unused_result))
|
|
|
|
|
/** Acquire a tablespace reference for I/O.
|
|
|
|
|
@return whether the file is usable */
|
2020-10-26 16:04:12 +02:00
|
|
|
bool acquire()
|
2020-10-26 15:59:30 +02:00
|
|
|
{
|
2020-10-26 16:04:12 +02:00
|
|
|
uint32_t n= acquire_low();
|
|
|
|
|
if (UNIV_LIKELY(!(n & (STOPPING | CLOSING))))
|
|
|
|
|
return true;
|
|
|
|
|
return UNIV_LIKELY(!(n & STOPPING)) && prepare();
|
2020-10-26 15:59:30 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/** Acquire another tablespace reference for I/O. */
|
2020-10-26 16:04:12 +02:00
|
|
|
inline void reacquire();
|
2020-10-26 15:59:30 +02:00
|
|
|
|
2020-10-26 16:04:12 +02:00
|
|
|
/** Release a tablespace reference.
|
|
|
|
|
@return whether this was the last reference */
|
|
|
|
|
bool release()
|
2020-10-26 15:59:30 +02:00
|
|
|
{
|
2020-10-26 16:04:12 +02:00
|
|
|
uint32_t n= n_pending.fetch_sub(1, std::memory_order_release);
|
|
|
|
|
ut_ad(n & PENDING);
|
|
|
|
|
return (n & PENDING) == 1;
|
2020-10-26 15:59:30 +02:00
|
|
|
}
|
MDEV-15053 Reduce buf_pool_t::mutex contention
User-visible changes: The INFORMATION_SCHEMA views INNODB_BUFFER_PAGE
and INNODB_BUFFER_PAGE_LRU will report a dummy value FLUSH_TYPE=0
and will no longer report the PAGE_STATE value READY_FOR_USE.
We will remove some fields from buf_page_t and move much code to
member functions of buf_pool_t and buf_page_t, so that the access
rules of data members can be enforced consistently.
Evicting or adding pages in buf_pool.LRU will remain covered by
buf_pool.mutex.
Evicting or adding pages in buf_pool.page_hash will remain
covered by both buf_pool.mutex and the buf_pool.page_hash X-latch.
After this fix, buf_pool.page_hash lookups can entirely
avoid acquiring buf_pool.mutex, only relying on
buf_pool.hash_lock_get() S-latch.
Similarly, buf_flush_check_neighbors() can will rely solely on
buf_pool.mutex, no buf_pool.page_hash latch at all.
The buf_pool.mutex is rather contended in I/O heavy benchmarks,
especially when the workload does not fit in the buffer pool.
The first attempt to alleviate the contention was the
buf_pool_t::mutex split in
commit 4ed7082eefe56b3e97e0edefb3df76dd7ef5e858
which introduced buf_block_t::mutex, which we are now removing.
Later, multiple instances of buf_pool_t were introduced
in commit c18084f71b02ea707c6461353e6cfc15d7553bc6
and recently removed by us in
commit 1a6f708ec594ac0ae2dd30db926ab07b100fa24b (MDEV-15058).
UNIV_BUF_DEBUG: Remove. This option to enable some buffer pool
related debugging in otherwise non-debug builds has not been used
for years. Instead, we have been using UNIV_DEBUG, which is enabled
in CMAKE_BUILD_TYPE=Debug.
buf_block_t::mutex, buf_pool_t::zip_mutex: Remove. We can mainly rely on
std::atomic and the buf_pool.page_hash latches, and in some cases
depend on buf_pool.mutex or buf_pool.flush_list_mutex just like before.
We must always release buf_block_t::lock before invoking
unfix() or io_unfix(), to prevent a glitch where a block that was
added to the buf_pool.free list would apper X-latched. See
commit c5883debd6ef440a037011c11873b396923e93c5 how this glitch
was finally caught in a debug environment.
We move some buf_pool_t::page_hash specific code from the
ha and hash modules to buf_pool, for improved readability.
buf_pool_t::close(): Assert that all blocks are clean, except
on aborted startup or crash-like shutdown.
buf_pool_t::validate(): No longer attempt to validate
n_flush[] against the number of BUF_IO_WRITE fixed blocks,
because buf_page_t::flush_type no longer exists.
buf_pool_t::watch_set(): Replaces buf_pool_watch_set().
Reduce mutex contention by separating the buf_pool.watch[]
allocation and the insert into buf_pool.page_hash.
buf_pool_t::page_hash_lock<bool exclusive>(): Acquire a
buf_pool.page_hash latch.
Replaces and extends buf_page_hash_lock_s_confirm()
and buf_page_hash_lock_x_confirm().
buf_pool_t::READ_AHEAD_PAGES: Renamed from BUF_READ_AHEAD_PAGES.
buf_pool_t::curr_size, old_size, read_ahead_area, n_pend_reads:
Use Atomic_counter.
buf_pool_t::running_out(): Replaces buf_LRU_buf_pool_running_out().
buf_pool_t::LRU_remove(): Remove a block from the LRU list
and return its predecessor. Incorporates buf_LRU_adjust_hp(),
which was removed.
buf_page_get_gen(): Remove a redundant call of fsp_is_system_temporary(),
for mode == BUF_GET_IF_IN_POOL_OR_WATCH, which is only used by
BTR_DELETE_OP (purge), which is never invoked on temporary tables.
buf_free_from_unzip_LRU_list_batch(): Avoid redundant assignments.
buf_LRU_free_from_unzip_LRU_list(): Simplify the loop condition.
buf_LRU_free_page(): Clarify the function comment.
buf_flush_check_neighbor(), buf_flush_check_neighbors():
Rewrite the construction of the page hash range. We will hold
the buf_pool.mutex for up to buf_pool.read_ahead_area (at most 64)
consecutive lookups of buf_pool.page_hash.
buf_flush_page_and_try_neighbors(): Remove.
Merge to its only callers, and remove redundant operations in
buf_flush_LRU_list_batch().
buf_read_ahead_random(), buf_read_ahead_linear(): Rewrite.
Do not acquire buf_pool.mutex, and iterate directly with page_id_t.
ut_2_power_up(): Remove. my_round_up_to_next_power() is inlined
and avoids any loops.
fil_page_get_prev(), fil_page_get_next(), fil_addr_is_null(): Remove.
buf_flush_page(): Add a fil_space_t* parameter. Minimize the
buf_pool.mutex hold time. buf_pool.n_flush[] is no longer updated
atomically with the io_fix, and we will protect most buf_block_t
fields with buf_block_t::lock. The function
buf_flush_write_block_low() is removed and merged here.
buf_page_init_for_read(): Use static linkage. Initialize the newly
allocated block and acquire the exclusive buf_block_t::lock while not
holding any mutex.
IORequest::IORequest(): Remove the body. We only need to invoke
set_punch_hole() in buf_flush_page() and nowhere else.
buf_page_t::flush_type: Remove. Replaced by IORequest::flush_type.
This field is only used during a fil_io() call.
That function already takes IORequest as a parameter, so we had
better introduce for the rarely changing field.
buf_block_t::init(): Replaces buf_page_init().
buf_page_t::init(): Replaces buf_page_init_low().
buf_block_t::initialise(): Initialise many fields, but
keep the buf_page_t::state(). Both buf_pool_t::validate() and
buf_page_optimistic_get() requires that buf_page_t::in_file()
be protected atomically with buf_page_t::in_page_hash
and buf_page_t::in_LRU_list.
buf_page_optimistic_get(): Now that buf_block_t::mutex
no longer exists, we must check buf_page_t::io_fix()
after acquiring the buf_pool.page_hash lock, to detect
whether buf_page_init_for_read() has been initiated.
We will also check the io_fix() before acquiring hash_lock
in order to avoid unnecessary computation.
The field buf_block_t::modify_clock (protected by buf_block_t::lock)
allows buf_page_optimistic_get() to validate the block.
buf_page_t::real_size: Remove. It was only used while flushing
pages of page_compressed tables.
buf_page_encrypt(): Add an output parameter that allows us ot eliminate
buf_page_t::real_size. Replace a condition with debug assertion.
buf_page_should_punch_hole(): Remove.
buf_dblwr_t::add_to_batch(): Replaces buf_dblwr_add_to_batch().
Add the parameter size (to replace buf_page_t::real_size).
buf_dblwr_t::write_single_page(): Replaces buf_dblwr_write_single_page().
Add the parameter size (to replace buf_page_t::real_size).
fil_system_t::detach(): Replaces fil_space_detach().
Ensure that fil_validate() will not be violated even if
fil_system.mutex is released and reacquired.
fil_node_t::complete_io(): Renamed from fil_node_complete_io().
fil_node_t::close_to_free(): Replaces fil_node_close_to_free().
Avoid invoking fil_node_t::close() because fil_system.n_open
has already been decremented in fil_space_t::detach().
BUF_BLOCK_READY_FOR_USE: Remove. Directly use BUF_BLOCK_MEMORY.
BUF_BLOCK_ZIP_DIRTY: Remove. Directly use BUF_BLOCK_ZIP_PAGE,
and distinguish dirty pages by buf_page_t::oldest_modification().
BUF_BLOCK_POOL_WATCH: Remove. Use BUF_BLOCK_NOT_USED instead.
This state was only being used for buf_page_t that are in
buf_pool.watch.
buf_pool_t::watch[]: Remove pointer indirection.
buf_page_t::in_flush_list: Remove. It was set if and only if
buf_page_t::oldest_modification() is nonzero.
buf_page_decrypt_after_read(), buf_corrupt_page_release(),
buf_page_check_corrupt(): Change the const fil_space_t* parameter
to const fil_node_t& so that we can report the correct file name.
buf_page_monitor(): Declare as an ATTRIBUTE_COLD global function.
buf_page_io_complete(): Split to buf_page_read_complete() and
buf_page_write_complete().
buf_dblwr_t::in_use: Remove.
buf_dblwr_t::buf_block_array: Add IORequest::flush_t.
buf_dblwr_sync_datafiles(): Remove. It was a useless wrapper of
os_aio_wait_until_no_pending_writes().
buf_flush_write_complete(): Declare static, not global.
Add the parameter IORequest::flush_t.
buf_flush_freed_page(): Simplify the code.
recv_sys_t::flush_lru: Renamed from flush_type and changed to bool.
fil_read(), fil_write(): Replaced with direct use of fil_io().
fil_buffering_disabled(): Remove. Check srv_file_flush_method directly.
fil_mutex_enter_and_prepare_for_io(): Return the resolved
fil_space_t* to avoid a duplicated lookup in the caller.
fil_report_invalid_page_access(): Clean up the parameters.
fil_io(): Return fil_io_t, which comprises fil_node_t and error code.
Always invoke fil_space_t::acquire_for_io() and let either the
sync=true caller or fil_aio_callback() invoke
fil_space_t::release_for_io().
fil_aio_callback(): Rewrite to replace buf_page_io_complete().
fil_check_pending_operations(): Remove a parameter, and remove some
redundant lookups.
fil_node_close_to_free(): Wait for n_pending==0. Because we no longer
do an extra lookup of the tablespace between fil_io() and the
completion of the operation, we must give fil_node_t::complete_io() a
chance to decrement the counter.
fil_close_tablespace(): Remove unused parameter trx, and document
that this is only invoked during the error handling of IMPORT TABLESPACE.
row_import_discard_changes(): Merged with the only caller,
row_import_cleanup(). Do not lock up the data dictionary while
invoking fil_close_tablespace().
logs_empty_and_mark_files_at_shutdown(): Do not invoke
fil_close_all_files(), to avoid a !needs_flush assertion failure
on fil_node_t::close().
innodb_shutdown(): Invoke os_aio_free() before fil_close_all_files().
fil_close_all_files(): Invoke fil_flush_file_spaces()
to ensure proper durability.
thread_pool::unbind(): Fix a crash that would occur on Windows
after srv_thread_pool->disable_aio() and os_file_close().
This fix was submitted by Vladislav Vaintroub.
Thanks to Matthias Leich and Axel Schwenke for extensive testing,
Vladislav Vaintroub for helpful comments, and Eugene Kosov for a review.
2020-06-05 12:35:46 +03:00
|
|
|
|
2020-10-26 16:04:12 +02:00
|
|
|
/** Clear the NEEDS_FSYNC flag */
|
|
|
|
|
void clear_flush()
|
2021-10-03 13:49:40 +03:00
|
|
|
{
|
|
|
|
|
#if defined __GNUC__ && (defined __i386__ || defined __x86_64__)
|
|
|
|
|
static_assert(NEEDS_FSYNC == 1U << 29, "compatibility");
|
|
|
|
|
__asm__ __volatile__("lock btrl $29, %0" : "+m" (n_pending));
|
2022-01-28 16:42:37 +02:00
|
|
|
#elif defined _MSC_VER && (defined _M_IX86 || defined _M_X64)
|
2021-10-03 13:49:40 +03:00
|
|
|
static_assert(NEEDS_FSYNC == 1U << 29, "compatibility");
|
|
|
|
|
_interlockedbittestandreset(reinterpret_cast<volatile long*>
|
|
|
|
|
(&n_pending), 29);
|
|
|
|
|
#else
|
|
|
|
|
n_pending.fetch_and(~NEEDS_FSYNC, std::memory_order_release);
|
|
|
|
|
#endif
|
|
|
|
|
}
|
2020-10-26 16:04:12 +02:00
|
|
|
|
|
|
|
|
private:
|
2021-10-03 13:49:40 +03:00
|
|
|
/** Clear the CLOSING flag */
|
|
|
|
|
void clear_closing()
|
|
|
|
|
{
|
|
|
|
|
#if defined __GNUC__ && (defined __i386__ || defined __x86_64__)
|
|
|
|
|
static_assert(CLOSING == 1U << 30, "compatibility");
|
|
|
|
|
__asm__ __volatile__("lock btrl $30, %0" : "+m" (n_pending));
|
2022-01-28 16:42:37 +02:00
|
|
|
#elif defined _MSC_VER && (defined _M_IX86 || defined _M_X64)
|
2021-10-03 13:49:40 +03:00
|
|
|
static_assert(CLOSING == 1U << 30, "compatibility");
|
|
|
|
|
_interlockedbittestandreset(reinterpret_cast<volatile long*>
|
|
|
|
|
(&n_pending), 30);
|
|
|
|
|
#else
|
|
|
|
|
n_pending.fetch_and(~CLOSING, std::memory_order_relaxed);
|
|
|
|
|
#endif
|
|
|
|
|
}
|
|
|
|
|
|
2020-10-26 16:04:12 +02:00
|
|
|
/** @return pending operations (and flags) */
|
|
|
|
|
uint32_t pending()const { return n_pending.load(std::memory_order_acquire); }
|
|
|
|
|
public:
|
|
|
|
|
/** @return whether close() of the file handle has been requested */
|
|
|
|
|
bool is_closing() const { return pending() & CLOSING; }
|
|
|
|
|
/** @return whether the tablespace is going to be dropped */
|
|
|
|
|
bool is_stopping() const { return pending() & STOPPING; }
|
|
|
|
|
/** @return number of pending operations */
|
|
|
|
|
bool is_ready_to_close() const
|
|
|
|
|
{ return (pending() & (PENDING | CLOSING)) == CLOSING; }
|
|
|
|
|
/** @return whether fsync() or similar is needed */
|
|
|
|
|
bool needs_flush() const { return pending() & NEEDS_FSYNC; }
|
|
|
|
|
/** @return whether fsync() or similar is needed, and the tablespace is
|
|
|
|
|
not being dropped */
|
|
|
|
|
bool needs_flush_not_stopping() const
|
|
|
|
|
{ return (pending() & (NEEDS_FSYNC | STOPPING)) == NEEDS_FSYNC; }
|
|
|
|
|
|
|
|
|
|
uint32_t referenced() const { return pending() & PENDING; }
|
|
|
|
|
private:
|
2020-10-26 15:59:30 +02:00
|
|
|
MY_ATTRIBUTE((warn_unused_result))
|
|
|
|
|
/** Prepare to close the file handle.
|
2020-10-26 16:04:12 +02:00
|
|
|
@return number of pending operations, possibly with NEEDS_FSYNC flag */
|
2020-10-26 15:59:30 +02:00
|
|
|
uint32_t set_closing()
|
MDEV-15053 Reduce buf_pool_t::mutex contention
User-visible changes: The INFORMATION_SCHEMA views INNODB_BUFFER_PAGE
and INNODB_BUFFER_PAGE_LRU will report a dummy value FLUSH_TYPE=0
and will no longer report the PAGE_STATE value READY_FOR_USE.
We will remove some fields from buf_page_t and move much code to
member functions of buf_pool_t and buf_page_t, so that the access
rules of data members can be enforced consistently.
Evicting or adding pages in buf_pool.LRU will remain covered by
buf_pool.mutex.
Evicting or adding pages in buf_pool.page_hash will remain
covered by both buf_pool.mutex and the buf_pool.page_hash X-latch.
After this fix, buf_pool.page_hash lookups can entirely
avoid acquiring buf_pool.mutex, only relying on
buf_pool.hash_lock_get() S-latch.
Similarly, buf_flush_check_neighbors() can will rely solely on
buf_pool.mutex, no buf_pool.page_hash latch at all.
The buf_pool.mutex is rather contended in I/O heavy benchmarks,
especially when the workload does not fit in the buffer pool.
The first attempt to alleviate the contention was the
buf_pool_t::mutex split in
commit 4ed7082eefe56b3e97e0edefb3df76dd7ef5e858
which introduced buf_block_t::mutex, which we are now removing.
Later, multiple instances of buf_pool_t were introduced
in commit c18084f71b02ea707c6461353e6cfc15d7553bc6
and recently removed by us in
commit 1a6f708ec594ac0ae2dd30db926ab07b100fa24b (MDEV-15058).
UNIV_BUF_DEBUG: Remove. This option to enable some buffer pool
related debugging in otherwise non-debug builds has not been used
for years. Instead, we have been using UNIV_DEBUG, which is enabled
in CMAKE_BUILD_TYPE=Debug.
buf_block_t::mutex, buf_pool_t::zip_mutex: Remove. We can mainly rely on
std::atomic and the buf_pool.page_hash latches, and in some cases
depend on buf_pool.mutex or buf_pool.flush_list_mutex just like before.
We must always release buf_block_t::lock before invoking
unfix() or io_unfix(), to prevent a glitch where a block that was
added to the buf_pool.free list would apper X-latched. See
commit c5883debd6ef440a037011c11873b396923e93c5 how this glitch
was finally caught in a debug environment.
We move some buf_pool_t::page_hash specific code from the
ha and hash modules to buf_pool, for improved readability.
buf_pool_t::close(): Assert that all blocks are clean, except
on aborted startup or crash-like shutdown.
buf_pool_t::validate(): No longer attempt to validate
n_flush[] against the number of BUF_IO_WRITE fixed blocks,
because buf_page_t::flush_type no longer exists.
buf_pool_t::watch_set(): Replaces buf_pool_watch_set().
Reduce mutex contention by separating the buf_pool.watch[]
allocation and the insert into buf_pool.page_hash.
buf_pool_t::page_hash_lock<bool exclusive>(): Acquire a
buf_pool.page_hash latch.
Replaces and extends buf_page_hash_lock_s_confirm()
and buf_page_hash_lock_x_confirm().
buf_pool_t::READ_AHEAD_PAGES: Renamed from BUF_READ_AHEAD_PAGES.
buf_pool_t::curr_size, old_size, read_ahead_area, n_pend_reads:
Use Atomic_counter.
buf_pool_t::running_out(): Replaces buf_LRU_buf_pool_running_out().
buf_pool_t::LRU_remove(): Remove a block from the LRU list
and return its predecessor. Incorporates buf_LRU_adjust_hp(),
which was removed.
buf_page_get_gen(): Remove a redundant call of fsp_is_system_temporary(),
for mode == BUF_GET_IF_IN_POOL_OR_WATCH, which is only used by
BTR_DELETE_OP (purge), which is never invoked on temporary tables.
buf_free_from_unzip_LRU_list_batch(): Avoid redundant assignments.
buf_LRU_free_from_unzip_LRU_list(): Simplify the loop condition.
buf_LRU_free_page(): Clarify the function comment.
buf_flush_check_neighbor(), buf_flush_check_neighbors():
Rewrite the construction of the page hash range. We will hold
the buf_pool.mutex for up to buf_pool.read_ahead_area (at most 64)
consecutive lookups of buf_pool.page_hash.
buf_flush_page_and_try_neighbors(): Remove.
Merge to its only callers, and remove redundant operations in
buf_flush_LRU_list_batch().
buf_read_ahead_random(), buf_read_ahead_linear(): Rewrite.
Do not acquire buf_pool.mutex, and iterate directly with page_id_t.
ut_2_power_up(): Remove. my_round_up_to_next_power() is inlined
and avoids any loops.
fil_page_get_prev(), fil_page_get_next(), fil_addr_is_null(): Remove.
buf_flush_page(): Add a fil_space_t* parameter. Minimize the
buf_pool.mutex hold time. buf_pool.n_flush[] is no longer updated
atomically with the io_fix, and we will protect most buf_block_t
fields with buf_block_t::lock. The function
buf_flush_write_block_low() is removed and merged here.
buf_page_init_for_read(): Use static linkage. Initialize the newly
allocated block and acquire the exclusive buf_block_t::lock while not
holding any mutex.
IORequest::IORequest(): Remove the body. We only need to invoke
set_punch_hole() in buf_flush_page() and nowhere else.
buf_page_t::flush_type: Remove. Replaced by IORequest::flush_type.
This field is only used during a fil_io() call.
That function already takes IORequest as a parameter, so we had
better introduce for the rarely changing field.
buf_block_t::init(): Replaces buf_page_init().
buf_page_t::init(): Replaces buf_page_init_low().
buf_block_t::initialise(): Initialise many fields, but
keep the buf_page_t::state(). Both buf_pool_t::validate() and
buf_page_optimistic_get() requires that buf_page_t::in_file()
be protected atomically with buf_page_t::in_page_hash
and buf_page_t::in_LRU_list.
buf_page_optimistic_get(): Now that buf_block_t::mutex
no longer exists, we must check buf_page_t::io_fix()
after acquiring the buf_pool.page_hash lock, to detect
whether buf_page_init_for_read() has been initiated.
We will also check the io_fix() before acquiring hash_lock
in order to avoid unnecessary computation.
The field buf_block_t::modify_clock (protected by buf_block_t::lock)
allows buf_page_optimistic_get() to validate the block.
buf_page_t::real_size: Remove. It was only used while flushing
pages of page_compressed tables.
buf_page_encrypt(): Add an output parameter that allows us ot eliminate
buf_page_t::real_size. Replace a condition with debug assertion.
buf_page_should_punch_hole(): Remove.
buf_dblwr_t::add_to_batch(): Replaces buf_dblwr_add_to_batch().
Add the parameter size (to replace buf_page_t::real_size).
buf_dblwr_t::write_single_page(): Replaces buf_dblwr_write_single_page().
Add the parameter size (to replace buf_page_t::real_size).
fil_system_t::detach(): Replaces fil_space_detach().
Ensure that fil_validate() will not be violated even if
fil_system.mutex is released and reacquired.
fil_node_t::complete_io(): Renamed from fil_node_complete_io().
fil_node_t::close_to_free(): Replaces fil_node_close_to_free().
Avoid invoking fil_node_t::close() because fil_system.n_open
has already been decremented in fil_space_t::detach().
BUF_BLOCK_READY_FOR_USE: Remove. Directly use BUF_BLOCK_MEMORY.
BUF_BLOCK_ZIP_DIRTY: Remove. Directly use BUF_BLOCK_ZIP_PAGE,
and distinguish dirty pages by buf_page_t::oldest_modification().
BUF_BLOCK_POOL_WATCH: Remove. Use BUF_BLOCK_NOT_USED instead.
This state was only being used for buf_page_t that are in
buf_pool.watch.
buf_pool_t::watch[]: Remove pointer indirection.
buf_page_t::in_flush_list: Remove. It was set if and only if
buf_page_t::oldest_modification() is nonzero.
buf_page_decrypt_after_read(), buf_corrupt_page_release(),
buf_page_check_corrupt(): Change the const fil_space_t* parameter
to const fil_node_t& so that we can report the correct file name.
buf_page_monitor(): Declare as an ATTRIBUTE_COLD global function.
buf_page_io_complete(): Split to buf_page_read_complete() and
buf_page_write_complete().
buf_dblwr_t::in_use: Remove.
buf_dblwr_t::buf_block_array: Add IORequest::flush_t.
buf_dblwr_sync_datafiles(): Remove. It was a useless wrapper of
os_aio_wait_until_no_pending_writes().
buf_flush_write_complete(): Declare static, not global.
Add the parameter IORequest::flush_t.
buf_flush_freed_page(): Simplify the code.
recv_sys_t::flush_lru: Renamed from flush_type and changed to bool.
fil_read(), fil_write(): Replaced with direct use of fil_io().
fil_buffering_disabled(): Remove. Check srv_file_flush_method directly.
fil_mutex_enter_and_prepare_for_io(): Return the resolved
fil_space_t* to avoid a duplicated lookup in the caller.
fil_report_invalid_page_access(): Clean up the parameters.
fil_io(): Return fil_io_t, which comprises fil_node_t and error code.
Always invoke fil_space_t::acquire_for_io() and let either the
sync=true caller or fil_aio_callback() invoke
fil_space_t::release_for_io().
fil_aio_callback(): Rewrite to replace buf_page_io_complete().
fil_check_pending_operations(): Remove a parameter, and remove some
redundant lookups.
fil_node_close_to_free(): Wait for n_pending==0. Because we no longer
do an extra lookup of the tablespace between fil_io() and the
completion of the operation, we must give fil_node_t::complete_io() a
chance to decrement the counter.
fil_close_tablespace(): Remove unused parameter trx, and document
that this is only invoked during the error handling of IMPORT TABLESPACE.
row_import_discard_changes(): Merged with the only caller,
row_import_cleanup(). Do not lock up the data dictionary while
invoking fil_close_tablespace().
logs_empty_and_mark_files_at_shutdown(): Do not invoke
fil_close_all_files(), to avoid a !needs_flush assertion failure
on fil_node_t::close().
innodb_shutdown(): Invoke os_aio_free() before fil_close_all_files().
fil_close_all_files(): Invoke fil_flush_file_spaces()
to ensure proper durability.
thread_pool::unbind(): Fix a crash that would occur on Windows
after srv_thread_pool->disable_aio() and os_file_close().
This fix was submitted by Vladislav Vaintroub.
Thanks to Matthias Leich and Axel Schwenke for extensive testing,
Vladislav Vaintroub for helpful comments, and Eugene Kosov for a review.
2020-06-05 12:35:46 +03:00
|
|
|
{
|
2020-10-26 16:04:12 +02:00
|
|
|
return n_pending.fetch_or(CLOSING, std::memory_order_acquire) &
|
|
|
|
|
(PENDING | NEEDS_FSYNC);
|
MDEV-15053 Reduce buf_pool_t::mutex contention
User-visible changes: The INFORMATION_SCHEMA views INNODB_BUFFER_PAGE
and INNODB_BUFFER_PAGE_LRU will report a dummy value FLUSH_TYPE=0
and will no longer report the PAGE_STATE value READY_FOR_USE.
We will remove some fields from buf_page_t and move much code to
member functions of buf_pool_t and buf_page_t, so that the access
rules of data members can be enforced consistently.
Evicting or adding pages in buf_pool.LRU will remain covered by
buf_pool.mutex.
Evicting or adding pages in buf_pool.page_hash will remain
covered by both buf_pool.mutex and the buf_pool.page_hash X-latch.
After this fix, buf_pool.page_hash lookups can entirely
avoid acquiring buf_pool.mutex, only relying on
buf_pool.hash_lock_get() S-latch.
Similarly, buf_flush_check_neighbors() can will rely solely on
buf_pool.mutex, no buf_pool.page_hash latch at all.
The buf_pool.mutex is rather contended in I/O heavy benchmarks,
especially when the workload does not fit in the buffer pool.
The first attempt to alleviate the contention was the
buf_pool_t::mutex split in
commit 4ed7082eefe56b3e97e0edefb3df76dd7ef5e858
which introduced buf_block_t::mutex, which we are now removing.
Later, multiple instances of buf_pool_t were introduced
in commit c18084f71b02ea707c6461353e6cfc15d7553bc6
and recently removed by us in
commit 1a6f708ec594ac0ae2dd30db926ab07b100fa24b (MDEV-15058).
UNIV_BUF_DEBUG: Remove. This option to enable some buffer pool
related debugging in otherwise non-debug builds has not been used
for years. Instead, we have been using UNIV_DEBUG, which is enabled
in CMAKE_BUILD_TYPE=Debug.
buf_block_t::mutex, buf_pool_t::zip_mutex: Remove. We can mainly rely on
std::atomic and the buf_pool.page_hash latches, and in some cases
depend on buf_pool.mutex or buf_pool.flush_list_mutex just like before.
We must always release buf_block_t::lock before invoking
unfix() or io_unfix(), to prevent a glitch where a block that was
added to the buf_pool.free list would apper X-latched. See
commit c5883debd6ef440a037011c11873b396923e93c5 how this glitch
was finally caught in a debug environment.
We move some buf_pool_t::page_hash specific code from the
ha and hash modules to buf_pool, for improved readability.
buf_pool_t::close(): Assert that all blocks are clean, except
on aborted startup or crash-like shutdown.
buf_pool_t::validate(): No longer attempt to validate
n_flush[] against the number of BUF_IO_WRITE fixed blocks,
because buf_page_t::flush_type no longer exists.
buf_pool_t::watch_set(): Replaces buf_pool_watch_set().
Reduce mutex contention by separating the buf_pool.watch[]
allocation and the insert into buf_pool.page_hash.
buf_pool_t::page_hash_lock<bool exclusive>(): Acquire a
buf_pool.page_hash latch.
Replaces and extends buf_page_hash_lock_s_confirm()
and buf_page_hash_lock_x_confirm().
buf_pool_t::READ_AHEAD_PAGES: Renamed from BUF_READ_AHEAD_PAGES.
buf_pool_t::curr_size, old_size, read_ahead_area, n_pend_reads:
Use Atomic_counter.
buf_pool_t::running_out(): Replaces buf_LRU_buf_pool_running_out().
buf_pool_t::LRU_remove(): Remove a block from the LRU list
and return its predecessor. Incorporates buf_LRU_adjust_hp(),
which was removed.
buf_page_get_gen(): Remove a redundant call of fsp_is_system_temporary(),
for mode == BUF_GET_IF_IN_POOL_OR_WATCH, which is only used by
BTR_DELETE_OP (purge), which is never invoked on temporary tables.
buf_free_from_unzip_LRU_list_batch(): Avoid redundant assignments.
buf_LRU_free_from_unzip_LRU_list(): Simplify the loop condition.
buf_LRU_free_page(): Clarify the function comment.
buf_flush_check_neighbor(), buf_flush_check_neighbors():
Rewrite the construction of the page hash range. We will hold
the buf_pool.mutex for up to buf_pool.read_ahead_area (at most 64)
consecutive lookups of buf_pool.page_hash.
buf_flush_page_and_try_neighbors(): Remove.
Merge to its only callers, and remove redundant operations in
buf_flush_LRU_list_batch().
buf_read_ahead_random(), buf_read_ahead_linear(): Rewrite.
Do not acquire buf_pool.mutex, and iterate directly with page_id_t.
ut_2_power_up(): Remove. my_round_up_to_next_power() is inlined
and avoids any loops.
fil_page_get_prev(), fil_page_get_next(), fil_addr_is_null(): Remove.
buf_flush_page(): Add a fil_space_t* parameter. Minimize the
buf_pool.mutex hold time. buf_pool.n_flush[] is no longer updated
atomically with the io_fix, and we will protect most buf_block_t
fields with buf_block_t::lock. The function
buf_flush_write_block_low() is removed and merged here.
buf_page_init_for_read(): Use static linkage. Initialize the newly
allocated block and acquire the exclusive buf_block_t::lock while not
holding any mutex.
IORequest::IORequest(): Remove the body. We only need to invoke
set_punch_hole() in buf_flush_page() and nowhere else.
buf_page_t::flush_type: Remove. Replaced by IORequest::flush_type.
This field is only used during a fil_io() call.
That function already takes IORequest as a parameter, so we had
better introduce for the rarely changing field.
buf_block_t::init(): Replaces buf_page_init().
buf_page_t::init(): Replaces buf_page_init_low().
buf_block_t::initialise(): Initialise many fields, but
keep the buf_page_t::state(). Both buf_pool_t::validate() and
buf_page_optimistic_get() requires that buf_page_t::in_file()
be protected atomically with buf_page_t::in_page_hash
and buf_page_t::in_LRU_list.
buf_page_optimistic_get(): Now that buf_block_t::mutex
no longer exists, we must check buf_page_t::io_fix()
after acquiring the buf_pool.page_hash lock, to detect
whether buf_page_init_for_read() has been initiated.
We will also check the io_fix() before acquiring hash_lock
in order to avoid unnecessary computation.
The field buf_block_t::modify_clock (protected by buf_block_t::lock)
allows buf_page_optimistic_get() to validate the block.
buf_page_t::real_size: Remove. It was only used while flushing
pages of page_compressed tables.
buf_page_encrypt(): Add an output parameter that allows us ot eliminate
buf_page_t::real_size. Replace a condition with debug assertion.
buf_page_should_punch_hole(): Remove.
buf_dblwr_t::add_to_batch(): Replaces buf_dblwr_add_to_batch().
Add the parameter size (to replace buf_page_t::real_size).
buf_dblwr_t::write_single_page(): Replaces buf_dblwr_write_single_page().
Add the parameter size (to replace buf_page_t::real_size).
fil_system_t::detach(): Replaces fil_space_detach().
Ensure that fil_validate() will not be violated even if
fil_system.mutex is released and reacquired.
fil_node_t::complete_io(): Renamed from fil_node_complete_io().
fil_node_t::close_to_free(): Replaces fil_node_close_to_free().
Avoid invoking fil_node_t::close() because fil_system.n_open
has already been decremented in fil_space_t::detach().
BUF_BLOCK_READY_FOR_USE: Remove. Directly use BUF_BLOCK_MEMORY.
BUF_BLOCK_ZIP_DIRTY: Remove. Directly use BUF_BLOCK_ZIP_PAGE,
and distinguish dirty pages by buf_page_t::oldest_modification().
BUF_BLOCK_POOL_WATCH: Remove. Use BUF_BLOCK_NOT_USED instead.
This state was only being used for buf_page_t that are in
buf_pool.watch.
buf_pool_t::watch[]: Remove pointer indirection.
buf_page_t::in_flush_list: Remove. It was set if and only if
buf_page_t::oldest_modification() is nonzero.
buf_page_decrypt_after_read(), buf_corrupt_page_release(),
buf_page_check_corrupt(): Change the const fil_space_t* parameter
to const fil_node_t& so that we can report the correct file name.
buf_page_monitor(): Declare as an ATTRIBUTE_COLD global function.
buf_page_io_complete(): Split to buf_page_read_complete() and
buf_page_write_complete().
buf_dblwr_t::in_use: Remove.
buf_dblwr_t::buf_block_array: Add IORequest::flush_t.
buf_dblwr_sync_datafiles(): Remove. It was a useless wrapper of
os_aio_wait_until_no_pending_writes().
buf_flush_write_complete(): Declare static, not global.
Add the parameter IORequest::flush_t.
buf_flush_freed_page(): Simplify the code.
recv_sys_t::flush_lru: Renamed from flush_type and changed to bool.
fil_read(), fil_write(): Replaced with direct use of fil_io().
fil_buffering_disabled(): Remove. Check srv_file_flush_method directly.
fil_mutex_enter_and_prepare_for_io(): Return the resolved
fil_space_t* to avoid a duplicated lookup in the caller.
fil_report_invalid_page_access(): Clean up the parameters.
fil_io(): Return fil_io_t, which comprises fil_node_t and error code.
Always invoke fil_space_t::acquire_for_io() and let either the
sync=true caller or fil_aio_callback() invoke
fil_space_t::release_for_io().
fil_aio_callback(): Rewrite to replace buf_page_io_complete().
fil_check_pending_operations(): Remove a parameter, and remove some
redundant lookups.
fil_node_close_to_free(): Wait for n_pending==0. Because we no longer
do an extra lookup of the tablespace between fil_io() and the
completion of the operation, we must give fil_node_t::complete_io() a
chance to decrement the counter.
fil_close_tablespace(): Remove unused parameter trx, and document
that this is only invoked during the error handling of IMPORT TABLESPACE.
row_import_discard_changes(): Merged with the only caller,
row_import_cleanup(). Do not lock up the data dictionary while
invoking fil_close_tablespace().
logs_empty_and_mark_files_at_shutdown(): Do not invoke
fil_close_all_files(), to avoid a !needs_flush assertion failure
on fil_node_t::close().
innodb_shutdown(): Invoke os_aio_free() before fil_close_all_files().
fil_close_all_files(): Invoke fil_flush_file_spaces()
to ensure proper durability.
thread_pool::unbind(): Fix a crash that would occur on Windows
after srv_thread_pool->disable_aio() and os_file_close().
This fix was submitted by Vladislav Vaintroub.
Thanks to Matthias Leich and Axel Schwenke for extensive testing,
Vladislav Vaintroub for helpful comments, and Eugene Kosov for a review.
2020-06-05 12:35:46 +03:00
|
|
|
}
|
2020-10-26 16:04:12 +02:00
|
|
|
|
|
|
|
|
public:
|
|
|
|
|
/** Try to close a file to adhere to the innodb_open_files limit.
|
|
|
|
|
@param print_info whether to diagnose why a file cannot be closed
|
|
|
|
|
@return whether a file was closed */
|
|
|
|
|
static bool try_to_close(bool print_info);
|
|
|
|
|
|
|
|
|
|
/** Close all tablespace files at shutdown */
|
|
|
|
|
static void close_all();
|
2020-06-11 22:52:47 +05:30
|
|
|
|
|
|
|
|
/** @return last_freed_lsn */
|
|
|
|
|
lsn_t get_last_freed_lsn() { return last_freed_lsn; }
|
|
|
|
|
/** Update last_freed_lsn */
|
|
|
|
|
void update_last_freed_lsn(lsn_t lsn)
|
|
|
|
|
{
|
|
|
|
|
std::lock_guard<std::mutex> freed_lock(freed_range_mutex);
|
|
|
|
|
last_freed_lsn= lsn;
|
|
|
|
|
}
|
|
|
|
|
|
2020-10-26 16:04:12 +02:00
|
|
|
/** Note that the file will need fsync().
|
|
|
|
|
@return whether this needs to be added to fil_system.unflushed_spaces */
|
|
|
|
|
bool set_needs_flush()
|
|
|
|
|
{
|
|
|
|
|
uint32_t n= 1;
|
|
|
|
|
while (!n_pending.compare_exchange_strong(n, n | NEEDS_FSYNC,
|
|
|
|
|
std::memory_order_acquire,
|
|
|
|
|
std::memory_order_relaxed))
|
|
|
|
|
{
|
|
|
|
|
ut_ad(n & PENDING);
|
|
|
|
|
if (n & (NEEDS_FSYNC | STOPPING))
|
|
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return true;
|
|
|
|
|
}
|
|
|
|
|
|
2020-06-19 15:24:16 +05:30
|
|
|
/** Clear all freed ranges for undo tablespace when InnoDB
|
|
|
|
|
encounters TRIM redo log record */
|
2020-06-11 22:52:47 +05:30
|
|
|
void clear_freed_ranges()
|
|
|
|
|
{
|
|
|
|
|
std::lock_guard<std::mutex> freed_lock(freed_range_mutex);
|
|
|
|
|
freed_ranges.clear();
|
|
|
|
|
}
|
2020-01-07 13:23:41 +02:00
|
|
|
#endif /* !UNIV_INNOCHECKSUM */
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
/** FSP_SPACE_FLAGS and FSP_FLAGS_MEM_ flags;
|
|
|
|
|
check fsp0types.h to more info about flags. */
|
|
|
|
|
ulint flags;
|
|
|
|
|
|
|
|
|
|
/** Determine if full_crc32 is used for a data file
|
2020-09-07 14:23:05 +05:30
|
|
|
@param[in] flags tablespace flags (FSP_SPACE_FLAGS)
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
@return whether the full_crc32 algorithm is active */
|
|
|
|
|
static bool full_crc32(ulint flags) {
|
|
|
|
|
return flags & FSP_FLAGS_FCRC32_MASK_MARKER;
|
|
|
|
|
}
|
2021-08-10 00:54:12 +06:00
|
|
|
/** Determine if full_crc32 is used along with compression */
|
|
|
|
|
static bool is_full_crc32_compressed(ulint flags)
|
|
|
|
|
{
|
|
|
|
|
if (full_crc32(flags))
|
|
|
|
|
{
|
|
|
|
|
ulint algo= FSP_FLAGS_FCRC32_GET_COMPRESSED_ALGO(flags);
|
|
|
|
|
DBUG_ASSERT(algo <= PAGE_ALGORITHM_LAST);
|
|
|
|
|
return algo > 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return false;
|
|
|
|
|
}
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
/** @return whether innodb_checksum_algorithm=full_crc32 is active */
|
|
|
|
|
bool full_crc32() const { return full_crc32(flags); }
|
2019-02-06 19:50:11 +02:00
|
|
|
/** Determine the logical page size.
|
|
|
|
|
@param flags tablespace flags (FSP_FLAGS)
|
|
|
|
|
@return the logical page size
|
|
|
|
|
@retval 0 if the flags are invalid */
|
2019-02-20 09:07:13 +02:00
|
|
|
static unsigned logical_size(ulint flags) {
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
|
|
|
|
|
ulint page_ssize = 0;
|
|
|
|
|
|
|
|
|
|
if (full_crc32(flags)) {
|
|
|
|
|
page_ssize = FSP_FLAGS_FCRC32_GET_PAGE_SSIZE(flags);
|
|
|
|
|
} else {
|
|
|
|
|
page_ssize = FSP_FLAGS_GET_PAGE_SSIZE(flags);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
switch (page_ssize) {
|
2019-02-06 19:50:11 +02:00
|
|
|
case 3: return 4096;
|
|
|
|
|
case 4: return 8192;
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
case 5:
|
|
|
|
|
{ ut_ad(full_crc32(flags)); return 16384; }
|
|
|
|
|
case 0:
|
|
|
|
|
{ ut_ad(!full_crc32(flags)); return 16384; }
|
2019-02-06 19:50:11 +02:00
|
|
|
case 6: return 32768;
|
|
|
|
|
case 7: return 65536;
|
|
|
|
|
default: return 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
/** Determine the ROW_FORMAT=COMPRESSED page size.
|
|
|
|
|
@param flags tablespace flags (FSP_FLAGS)
|
|
|
|
|
@return the ROW_FORMAT=COMPRESSED page size
|
|
|
|
|
@retval 0 if ROW_FORMAT=COMPRESSED is not used */
|
2019-02-20 09:07:13 +02:00
|
|
|
static unsigned zip_size(ulint flags) {
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
|
|
|
|
|
if (full_crc32(flags)) {
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
2019-02-06 19:50:11 +02:00
|
|
|
ulint zip_ssize = FSP_FLAGS_GET_ZIP_SSIZE(flags);
|
|
|
|
|
return zip_ssize
|
|
|
|
|
? (UNIV_ZIP_SIZE_MIN >> 1) << zip_ssize : 0;
|
|
|
|
|
}
|
|
|
|
|
/** Determine the physical page size.
|
|
|
|
|
@param flags tablespace flags (FSP_FLAGS)
|
|
|
|
|
@return the physical page size */
|
2019-02-20 09:07:13 +02:00
|
|
|
static unsigned physical_size(ulint flags) {
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
|
|
|
|
|
if (full_crc32(flags)) {
|
|
|
|
|
return logical_size(flags);
|
|
|
|
|
}
|
|
|
|
|
|
2019-02-06 19:50:11 +02:00
|
|
|
ulint zip_ssize = FSP_FLAGS_GET_ZIP_SSIZE(flags);
|
|
|
|
|
return zip_ssize
|
|
|
|
|
? (UNIV_ZIP_SIZE_MIN >> 1) << zip_ssize
|
2019-02-20 09:07:13 +02:00
|
|
|
: unsigned(srv_page_size);
|
2019-02-06 19:50:11 +02:00
|
|
|
}
|
|
|
|
|
/** @return the ROW_FORMAT=COMPRESSED page size
|
|
|
|
|
@retval 0 if ROW_FORMAT=COMPRESSED is not used */
|
2019-02-20 09:07:13 +02:00
|
|
|
unsigned zip_size() const { return zip_size(flags); }
|
2019-02-06 19:50:11 +02:00
|
|
|
/** @return the physical page size */
|
2019-02-20 09:07:13 +02:00
|
|
|
unsigned physical_size() const { return physical_size(flags); }
|
2021-08-10 00:54:12 +06:00
|
|
|
/** Check whether the compression enabled in tablespace.
|
|
|
|
|
@param[in] flags tablespace flags */
|
|
|
|
|
static bool is_compressed(ulint flags)
|
|
|
|
|
{
|
2021-10-27 09:48:22 +03:00
|
|
|
return is_full_crc32_compressed(flags) ||
|
|
|
|
|
FSP_FLAGS_HAS_PAGE_COMPRESSION(flags);
|
2021-08-10 00:54:12 +06:00
|
|
|
}
|
2021-10-27 09:48:22 +03:00
|
|
|
/** @return whether the compression enabled for the tablespace. */
|
|
|
|
|
bool is_compressed() const { return is_compressed(flags); }
|
MDEV-18644: Support full_crc32 for page_compressed
This is a follow-up task to MDEV-12026, which introduced
innodb_checksum_algorithm=full_crc32 and a simpler page format.
MDEV-12026 did not enable full_crc32 for page_compressed tables,
which we will be doing now.
This is joint work with Thirunarayanan Balathandayuthapani.
For innodb_checksum_algorithm=full_crc32 we change the
page_compressed format as follows:
FIL_PAGE_TYPE: The most significant bit will be set to indicate
page_compressed format. The least significant bits will contain
the compressed page size, rounded up to a multiple of 256 bytes.
The checksum will be stored in the last 4 bytes of the page
(whether it is the full page or a page_compressed page whose
size is determined by FIL_PAGE_TYPE), covering all preceding
bytes of the page. If encryption is used, then the page will
be encrypted between compression and computing the checksum.
For page_compressed, FIL_PAGE_LSN will not be repeated at
the end of the page.
FSP_SPACE_FLAGS (already implemented as part of MDEV-12026):
We will store the innodb_compression_algorithm that may be used
to compress pages. Previously, the choice of algorithm was written
to each compressed data page separately, and one would be unable
to know in advance which compression algorithm(s) are used.
fil_space_t::full_crc32_page_compressed_len(): Determine if the
page_compressed algorithm of the tablespace needs to know the
exact length of the compressed data. If yes, we will reserve and
write an extra byte for this right before the checksum.
buf_page_is_compressed(): Determine if a page uses page_compressed
(in any innodb_checksum_algorithm).
fil_page_decompress(): Pass also fil_space_t::flags so that the
format can be determined.
buf_page_is_zeroes(): Check if a page is full of zero bytes.
buf_page_full_crc32_is_corrupted(): Renamed from
buf_encrypted_full_crc32_page_is_corrupted(). For full_crc32,
we always simply validate the checksum to the page contents,
while the physical page size is explicitly specified by an
unencrypted part of the page header.
buf_page_full_crc32_size(): Determine the size of a full_crc32 page.
buf_dblwr_check_page_lsn(): Make this a debug-only function, because
it involves potentially costly lookups of fil_space_t.
create_table_info_t::check_table_options(),
ha_innobase::check_if_supported_inplace_alter(): Do allow the creation
of SPATIAL INDEX with full_crc32 also when page_compressed is used.
commit_cache_norebuild(): Preserve the compression algorithm when
updating the page_compression_level.
dict_tf_to_fsp_flags(): Set the flags for page compression algorithm.
FIXME: Maybe there should be a table option page_compression_algorithm
and a session variable to back it?
2019-03-18 14:08:43 +02:00
|
|
|
|
|
|
|
|
/** Get the compression algorithm for full crc32 format.
|
|
|
|
|
@param[in] flags tablespace flags
|
|
|
|
|
@return algorithm type of tablespace */
|
|
|
|
|
static ulint get_compression_algo(ulint flags)
|
|
|
|
|
{
|
|
|
|
|
return full_crc32(flags)
|
|
|
|
|
? FSP_FLAGS_FCRC32_GET_COMPRESSED_ALGO(flags)
|
|
|
|
|
: 0;
|
|
|
|
|
}
|
|
|
|
|
/** @return the page_compressed algorithm
|
|
|
|
|
@retval 0 if not page_compressed */
|
|
|
|
|
ulint get_compression_algo() const {
|
|
|
|
|
return fil_space_t::get_compression_algo(flags);
|
|
|
|
|
}
|
|
|
|
|
/** Determine if the page_compressed page contains an extra byte
|
|
|
|
|
for exact compressed stream length
|
|
|
|
|
@param[in] flags tablespace flags
|
|
|
|
|
@return whether the extra byte is needed */
|
|
|
|
|
static bool full_crc32_page_compressed_len(ulint flags)
|
|
|
|
|
{
|
|
|
|
|
DBUG_ASSERT(full_crc32(flags));
|
|
|
|
|
switch (get_compression_algo(flags)) {
|
|
|
|
|
case PAGE_LZ4_ALGORITHM:
|
|
|
|
|
case PAGE_LZO_ALGORITHM:
|
|
|
|
|
case PAGE_SNAPPY_ALGORITHM:
|
|
|
|
|
return true;
|
|
|
|
|
}
|
|
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
/** Whether the full checksum matches with non full checksum flags.
|
|
|
|
|
@param[in] flags flags present
|
|
|
|
|
@param[in] expected expected flags
|
|
|
|
|
@return true if it is equivalent */
|
|
|
|
|
static bool is_flags_full_crc32_equal(ulint flags, ulint expected)
|
|
|
|
|
{
|
|
|
|
|
ut_ad(full_crc32(flags));
|
2020-09-07 14:23:05 +05:30
|
|
|
ulint fcrc32_psize = FSP_FLAGS_FCRC32_GET_PAGE_SSIZE(flags);
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
|
|
|
|
|
if (full_crc32(expected)) {
|
2019-09-10 17:41:26 +05:30
|
|
|
/* The data file may have been created with a
|
|
|
|
|
different innodb_compression_algorithm. But
|
|
|
|
|
we only support one innodb_page_size for all files. */
|
2020-09-07 14:23:05 +05:30
|
|
|
return fcrc32_psize
|
|
|
|
|
== FSP_FLAGS_FCRC32_GET_PAGE_SSIZE(expected);
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
}
|
|
|
|
|
|
2020-09-07 14:23:05 +05:30
|
|
|
ulint non_fcrc32_psize = FSP_FLAGS_GET_PAGE_SSIZE(expected);
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
|
2020-09-07 14:23:05 +05:30
|
|
|
if (!non_fcrc32_psize) {
|
|
|
|
|
if (fcrc32_psize != 5) {
|
MDEV-18644: Support full_crc32 for page_compressed
This is a follow-up task to MDEV-12026, which introduced
innodb_checksum_algorithm=full_crc32 and a simpler page format.
MDEV-12026 did not enable full_crc32 for page_compressed tables,
which we will be doing now.
This is joint work with Thirunarayanan Balathandayuthapani.
For innodb_checksum_algorithm=full_crc32 we change the
page_compressed format as follows:
FIL_PAGE_TYPE: The most significant bit will be set to indicate
page_compressed format. The least significant bits will contain
the compressed page size, rounded up to a multiple of 256 bytes.
The checksum will be stored in the last 4 bytes of the page
(whether it is the full page or a page_compressed page whose
size is determined by FIL_PAGE_TYPE), covering all preceding
bytes of the page. If encryption is used, then the page will
be encrypted between compression and computing the checksum.
For page_compressed, FIL_PAGE_LSN will not be repeated at
the end of the page.
FSP_SPACE_FLAGS (already implemented as part of MDEV-12026):
We will store the innodb_compression_algorithm that may be used
to compress pages. Previously, the choice of algorithm was written
to each compressed data page separately, and one would be unable
to know in advance which compression algorithm(s) are used.
fil_space_t::full_crc32_page_compressed_len(): Determine if the
page_compressed algorithm of the tablespace needs to know the
exact length of the compressed data. If yes, we will reserve and
write an extra byte for this right before the checksum.
buf_page_is_compressed(): Determine if a page uses page_compressed
(in any innodb_checksum_algorithm).
fil_page_decompress(): Pass also fil_space_t::flags so that the
format can be determined.
buf_page_is_zeroes(): Check if a page is full of zero bytes.
buf_page_full_crc32_is_corrupted(): Renamed from
buf_encrypted_full_crc32_page_is_corrupted(). For full_crc32,
we always simply validate the checksum to the page contents,
while the physical page size is explicitly specified by an
unencrypted part of the page header.
buf_page_full_crc32_size(): Determine the size of a full_crc32 page.
buf_dblwr_check_page_lsn(): Make this a debug-only function, because
it involves potentially costly lookups of fil_space_t.
create_table_info_t::check_table_options(),
ha_innobase::check_if_supported_inplace_alter(): Do allow the creation
of SPATIAL INDEX with full_crc32 also when page_compressed is used.
commit_cache_norebuild(): Preserve the compression algorithm when
updating the page_compression_level.
dict_tf_to_fsp_flags(): Set the flags for page compression algorithm.
FIXME: Maybe there should be a table option page_compression_algorithm
and a session variable to back it?
2019-03-18 14:08:43 +02:00
|
|
|
return false;
|
|
|
|
|
}
|
2020-09-07 14:23:05 +05:30
|
|
|
} else if (fcrc32_psize != non_fcrc32_psize) {
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
|
2019-09-10 17:41:26 +05:30
|
|
|
return true;
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
}
|
|
|
|
|
/** Whether old tablespace flags match full_crc32 flags.
|
|
|
|
|
@param[in] flags flags present
|
|
|
|
|
@param[in] expected expected flags
|
|
|
|
|
@return true if it is equivalent */
|
|
|
|
|
static bool is_flags_non_full_crc32_equal(ulint flags, ulint expected)
|
|
|
|
|
{
|
|
|
|
|
ut_ad(!full_crc32(flags));
|
|
|
|
|
|
MDEV-18644: Support full_crc32 for page_compressed
This is a follow-up task to MDEV-12026, which introduced
innodb_checksum_algorithm=full_crc32 and a simpler page format.
MDEV-12026 did not enable full_crc32 for page_compressed tables,
which we will be doing now.
This is joint work with Thirunarayanan Balathandayuthapani.
For innodb_checksum_algorithm=full_crc32 we change the
page_compressed format as follows:
FIL_PAGE_TYPE: The most significant bit will be set to indicate
page_compressed format. The least significant bits will contain
the compressed page size, rounded up to a multiple of 256 bytes.
The checksum will be stored in the last 4 bytes of the page
(whether it is the full page or a page_compressed page whose
size is determined by FIL_PAGE_TYPE), covering all preceding
bytes of the page. If encryption is used, then the page will
be encrypted between compression and computing the checksum.
For page_compressed, FIL_PAGE_LSN will not be repeated at
the end of the page.
FSP_SPACE_FLAGS (already implemented as part of MDEV-12026):
We will store the innodb_compression_algorithm that may be used
to compress pages. Previously, the choice of algorithm was written
to each compressed data page separately, and one would be unable
to know in advance which compression algorithm(s) are used.
fil_space_t::full_crc32_page_compressed_len(): Determine if the
page_compressed algorithm of the tablespace needs to know the
exact length of the compressed data. If yes, we will reserve and
write an extra byte for this right before the checksum.
buf_page_is_compressed(): Determine if a page uses page_compressed
(in any innodb_checksum_algorithm).
fil_page_decompress(): Pass also fil_space_t::flags so that the
format can be determined.
buf_page_is_zeroes(): Check if a page is full of zero bytes.
buf_page_full_crc32_is_corrupted(): Renamed from
buf_encrypted_full_crc32_page_is_corrupted(). For full_crc32,
we always simply validate the checksum to the page contents,
while the physical page size is explicitly specified by an
unencrypted part of the page header.
buf_page_full_crc32_size(): Determine the size of a full_crc32 page.
buf_dblwr_check_page_lsn(): Make this a debug-only function, because
it involves potentially costly lookups of fil_space_t.
create_table_info_t::check_table_options(),
ha_innobase::check_if_supported_inplace_alter(): Do allow the creation
of SPATIAL INDEX with full_crc32 also when page_compressed is used.
commit_cache_norebuild(): Preserve the compression algorithm when
updating the page_compression_level.
dict_tf_to_fsp_flags(): Set the flags for page compression algorithm.
FIXME: Maybe there should be a table option page_compression_algorithm
and a session variable to back it?
2019-03-18 14:08:43 +02:00
|
|
|
if (!full_crc32(expected)) {
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
|
2020-09-07 14:23:05 +05:30
|
|
|
ulint non_fcrc32_psize = FSP_FLAGS_GET_PAGE_SSIZE(flags);
|
|
|
|
|
ulint fcrc32_psize = FSP_FLAGS_FCRC32_GET_PAGE_SSIZE(
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
expected);
|
|
|
|
|
|
2020-09-07 14:23:05 +05:30
|
|
|
if (!non_fcrc32_psize) {
|
|
|
|
|
if (fcrc32_psize != 5) {
|
MDEV-18644: Support full_crc32 for page_compressed
This is a follow-up task to MDEV-12026, which introduced
innodb_checksum_algorithm=full_crc32 and a simpler page format.
MDEV-12026 did not enable full_crc32 for page_compressed tables,
which we will be doing now.
This is joint work with Thirunarayanan Balathandayuthapani.
For innodb_checksum_algorithm=full_crc32 we change the
page_compressed format as follows:
FIL_PAGE_TYPE: The most significant bit will be set to indicate
page_compressed format. The least significant bits will contain
the compressed page size, rounded up to a multiple of 256 bytes.
The checksum will be stored in the last 4 bytes of the page
(whether it is the full page or a page_compressed page whose
size is determined by FIL_PAGE_TYPE), covering all preceding
bytes of the page. If encryption is used, then the page will
be encrypted between compression and computing the checksum.
For page_compressed, FIL_PAGE_LSN will not be repeated at
the end of the page.
FSP_SPACE_FLAGS (already implemented as part of MDEV-12026):
We will store the innodb_compression_algorithm that may be used
to compress pages. Previously, the choice of algorithm was written
to each compressed data page separately, and one would be unable
to know in advance which compression algorithm(s) are used.
fil_space_t::full_crc32_page_compressed_len(): Determine if the
page_compressed algorithm of the tablespace needs to know the
exact length of the compressed data. If yes, we will reserve and
write an extra byte for this right before the checksum.
buf_page_is_compressed(): Determine if a page uses page_compressed
(in any innodb_checksum_algorithm).
fil_page_decompress(): Pass also fil_space_t::flags so that the
format can be determined.
buf_page_is_zeroes(): Check if a page is full of zero bytes.
buf_page_full_crc32_is_corrupted(): Renamed from
buf_encrypted_full_crc32_page_is_corrupted(). For full_crc32,
we always simply validate the checksum to the page contents,
while the physical page size is explicitly specified by an
unencrypted part of the page header.
buf_page_full_crc32_size(): Determine the size of a full_crc32 page.
buf_dblwr_check_page_lsn(): Make this a debug-only function, because
it involves potentially costly lookups of fil_space_t.
create_table_info_t::check_table_options(),
ha_innobase::check_if_supported_inplace_alter(): Do allow the creation
of SPATIAL INDEX with full_crc32 also when page_compressed is used.
commit_cache_norebuild(): Preserve the compression algorithm when
updating the page_compression_level.
dict_tf_to_fsp_flags(): Set the flags for page compression algorithm.
FIXME: Maybe there should be a table option page_compression_algorithm
and a session variable to back it?
2019-03-18 14:08:43 +02:00
|
|
|
return false;
|
|
|
|
|
}
|
2020-09-07 14:23:05 +05:30
|
|
|
} else if (fcrc32_psize != non_fcrc32_psize) {
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
|
2019-09-10 17:41:26 +05:30
|
|
|
return true;
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
}
|
|
|
|
|
/** Whether both fsp flags are equivalent */
|
|
|
|
|
static bool is_flags_equal(ulint flags, ulint expected)
|
|
|
|
|
{
|
|
|
|
|
if (!((flags ^ expected) & ~(1U << FSP_FLAGS_POS_RESERVED))) {
|
|
|
|
|
return true;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return full_crc32(flags)
|
|
|
|
|
? is_flags_full_crc32_equal(flags, expected)
|
|
|
|
|
: is_flags_non_full_crc32_equal(flags, expected);
|
|
|
|
|
}
|
|
|
|
|
/** Validate the tablespace flags for full crc32 format.
|
|
|
|
|
@param[in] flags the content of FSP_SPACE_FLAGS
|
|
|
|
|
@return whether the flags are correct in full crc32 format */
|
|
|
|
|
static bool is_fcrc32_valid_flags(ulint flags)
|
|
|
|
|
{
|
|
|
|
|
ut_ad(flags & FSP_FLAGS_FCRC32_MASK_MARKER);
|
|
|
|
|
const ulint page_ssize = physical_size(flags);
|
|
|
|
|
if (page_ssize < 3 || page_ssize & 8) {
|
|
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
flags >>= FSP_FLAGS_FCRC32_POS_COMPRESSED_ALGO;
|
|
|
|
|
|
|
|
|
|
return flags <= PAGE_ALGORITHM_LAST;
|
|
|
|
|
}
|
|
|
|
|
/** Validate the tablespace flags.
|
|
|
|
|
@param[in] flags content of FSP_SPACE_FLAGS
|
|
|
|
|
@param[in] is_ibd whether this is an .ibd file
|
|
|
|
|
(not system tablespace)
|
|
|
|
|
@return whether the flags are correct. */
|
|
|
|
|
static bool is_valid_flags(ulint flags, bool is_ibd)
|
|
|
|
|
{
|
|
|
|
|
DBUG_EXECUTE_IF("fsp_flags_is_valid_failure",
|
|
|
|
|
return false;);
|
|
|
|
|
|
|
|
|
|
if (full_crc32(flags)) {
|
|
|
|
|
return is_fcrc32_valid_flags(flags);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (flags == 0) {
|
|
|
|
|
return true;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (flags & ~FSP_FLAGS_MASK) {
|
|
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if ((flags & (FSP_FLAGS_MASK_POST_ANTELOPE
|
|
|
|
|
| FSP_FLAGS_MASK_ATOMIC_BLOBS))
|
|
|
|
|
== FSP_FLAGS_MASK_ATOMIC_BLOBS) {
|
|
|
|
|
/* If the "atomic blobs" flag (indicating
|
|
|
|
|
ROW_FORMAT=DYNAMIC or ROW_FORMAT=COMPRESSED) flag
|
|
|
|
|
is set, then the "post Antelope"
|
|
|
|
|
(ROW_FORMAT!=REDUNDANT) flag must also be set. */
|
|
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* Bits 10..14 should be 0b0000d where d is the DATA_DIR flag
|
|
|
|
|
of MySQL 5.6 and MariaDB 10.0, which we ignore.
|
|
|
|
|
In the buggy FSP_SPACE_FLAGS written by MariaDB 10.1.0 to 10.1.20,
|
|
|
|
|
bits 10..14 would be nonzero 0bsssaa where sss is
|
|
|
|
|
nonzero PAGE_SSIZE (3, 4, 6, or 7)
|
|
|
|
|
and aa is ATOMIC_WRITES (not 0b11). */
|
|
|
|
|
if (FSP_FLAGS_GET_RESERVED(flags) & ~1U) {
|
|
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
const ulint ssize = FSP_FLAGS_GET_PAGE_SSIZE(flags);
|
|
|
|
|
if (ssize == 1 || ssize == 2 || ssize == 5 || ssize & 8) {
|
|
|
|
|
/* the page_size is not between 4k and 64k;
|
|
|
|
|
16k should be encoded as 0, not 5 */
|
|
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
const ulint zssize = FSP_FLAGS_GET_ZIP_SSIZE(flags);
|
|
|
|
|
if (zssize == 0) {
|
|
|
|
|
/* not ROW_FORMAT=COMPRESSED */
|
|
|
|
|
} else if (zssize > (ssize ? ssize : 5)) {
|
|
|
|
|
/* Invalid KEY_BLOCK_SIZE */
|
|
|
|
|
return false;
|
|
|
|
|
} else if (~flags & (FSP_FLAGS_MASK_POST_ANTELOPE
|
|
|
|
|
| FSP_FLAGS_MASK_ATOMIC_BLOBS)) {
|
|
|
|
|
/* both these flags should be set for
|
|
|
|
|
ROW_FORMAT=COMPRESSED */
|
|
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* The flags do look valid. But, avoid misinterpreting
|
|
|
|
|
buggy MariaDB 10.1 format flags for
|
|
|
|
|
PAGE_COMPRESSED=1 PAGE_COMPRESSION_LEVEL={0,2,3}
|
|
|
|
|
as valid-looking PAGE_SSIZE if this is known to be
|
|
|
|
|
an .ibd file and we are using the default innodb_page_size=16k. */
|
|
|
|
|
return(ssize == 0 || !is_ibd
|
|
|
|
|
|| srv_page_size != UNIV_PAGE_SIZE_ORIG);
|
|
|
|
|
}
|
2020-06-11 22:52:47 +05:30
|
|
|
|
|
|
|
|
#ifndef UNIV_INNOCHECKSUM
|
2020-10-26 15:59:30 +02:00
|
|
|
MY_ATTRIBUTE((warn_unused_result))
|
|
|
|
|
/** Create a tablespace in fil_system.
|
|
|
|
|
@param id tablespace identifier
|
|
|
|
|
@param flags tablespace flags
|
|
|
|
|
@param purpose tablespace purpose
|
|
|
|
|
@param crypt_data encryption information
|
|
|
|
|
@param mode encryption mode
|
|
|
|
|
@return pointer to created tablespace, to be filled in with add()
|
|
|
|
|
@retval nullptr on failure (such as when the same tablespace exists) */
|
MDEV-25312 Replace fil_space_t::name with fil_space_t::name()
A consistency check for fil_space_t::name is causing recovery failures
in MDEV-25180 (Atomic ALTER TABLE). So, we'd better remove that field
altogether.
fil_space_t::name was more or less a copy of dict_table_t::name
(except for some special cases), and it was not being used for
anything useful.
There used to be a name_hash, but it had been removed already in
commit a75dbfd7183cc96680f3e3e684fd36500dac8158 (MDEV-12266).
We will also remove os_normalize_path(), OS_PATH_SEPARATOR,
OS_PATH_SEPATOR_ALT. On Microsoft Windows, we will treat \ and /
roughly in the same way. The intention is that for per-table
tablespaces, the filenames will always follow the pattern
prefix/databasename/tablename.ibd. (Any \ in the prefix must not
be converted.)
ut_basename_noext(): Remove (unused function).
read_link_file(): Replaces RemoteDatafile::read_link_file().
We will ensure that the last two path component separators are
forward slashes (converting up to 2 trailing backslashes on
Microsoft Windows), so that everywhere else we can
assume that data file names end in "/databasename/tablename.ibd".
Note: On Microsoft Windows, path names that start with \\?\ must
not contain / as path component separators. Previously, such paths
did work in the DATA DIRECTORY argument of InnoDB tables.
Reviewed by: Vladislav Vaintroub
2021-04-07 18:01:13 +03:00
|
|
|
static fil_space_t *create(ulint id, ulint flags,
|
2020-10-26 15:59:30 +02:00
|
|
|
fil_type_t purpose, fil_space_crypt_t *crypt_data,
|
|
|
|
|
fil_encryption_t mode= FIL_ENCRYPTION_DEFAULT);
|
|
|
|
|
|
2020-10-26 16:04:12 +02:00
|
|
|
MY_ATTRIBUTE((warn_unused_result))
|
|
|
|
|
/** Acquire a tablespace reference.
|
|
|
|
|
@param id tablespace identifier
|
|
|
|
|
@return tablespace
|
|
|
|
|
@retval nullptr if the tablespace is missing or inaccessible */
|
|
|
|
|
static fil_space_t *get(ulint id);
|
2020-10-26 15:59:30 +02:00
|
|
|
|
2020-06-11 22:52:47 +05:30
|
|
|
/** Add/remove the free page in the freed ranges list.
|
|
|
|
|
@param[in] offset page number to be added
|
|
|
|
|
@param[in] free true if page to be freed */
|
|
|
|
|
void free_page(uint32_t offset, bool add=true)
|
|
|
|
|
{
|
2020-10-26 16:04:12 +02:00
|
|
|
std::lock_guard<std::mutex> freed_lock(freed_range_mutex);
|
2020-06-11 22:52:47 +05:30
|
|
|
if (add)
|
|
|
|
|
return freed_ranges.add_value(offset);
|
|
|
|
|
|
|
|
|
|
if (freed_ranges.empty())
|
|
|
|
|
return;
|
|
|
|
|
|
|
|
|
|
return freed_ranges.remove_value(offset);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/** Add the range of freed pages */
|
|
|
|
|
void add_free_ranges(range_set ranges)
|
|
|
|
|
{
|
2020-10-26 16:04:12 +02:00
|
|
|
std::lock_guard<std::mutex> freed_lock(freed_range_mutex);
|
2020-06-11 22:52:47 +05:30
|
|
|
freed_ranges= std::move(ranges);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/** Add the set of freed page ranges */
|
|
|
|
|
void add_free_range(const range_t range)
|
|
|
|
|
{
|
|
|
|
|
std::lock_guard<std::mutex> freed_lock(freed_range_mutex);
|
|
|
|
|
freed_ranges.add_range(range);
|
|
|
|
|
}
|
|
|
|
|
|
2020-10-26 15:59:30 +02:00
|
|
|
/** Set the tablespace size in pages */
|
|
|
|
|
void set_sizes(uint32_t s)
|
|
|
|
|
{
|
|
|
|
|
ut_ad(id ? !size : (size >= s));
|
|
|
|
|
size= s; committed_size= s;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/** Update committed_size in mtr_t::commit() */
|
2020-11-20 12:30:55 +02:00
|
|
|
void set_committed_size() { committed_size= size; }
|
2020-10-26 15:59:30 +02:00
|
|
|
|
|
|
|
|
/** @return the last persisted page number */
|
|
|
|
|
uint32_t last_page_number() const { return committed_size - 1; }
|
|
|
|
|
|
|
|
|
|
/** @return the size in pages (0 if unreadable) */
|
|
|
|
|
inline uint32_t get_size();
|
|
|
|
|
|
|
|
|
|
/** Read or write data.
|
|
|
|
|
@param type I/O context
|
|
|
|
|
@param offset offset in bytes
|
|
|
|
|
@param len number of bytes
|
|
|
|
|
@param buf the data to be read or written
|
|
|
|
|
@param bpage buffer block (for type.is_async() completion callback)
|
|
|
|
|
@return status and file descriptor */
|
|
|
|
|
fil_io_t io(const IORequest &type, os_offset_t offset, size_t len,
|
|
|
|
|
void *buf, buf_page_t *bpage= nullptr);
|
2020-10-26 16:04:12 +02:00
|
|
|
/** Flush pending writes from the file system cache to the file. */
|
2020-12-11 09:05:26 +02:00
|
|
|
template<bool have_reference> inline void flush();
|
2020-10-26 16:04:12 +02:00
|
|
|
/** Flush pending writes from the file system cache to the file. */
|
|
|
|
|
void flush_low();
|
2020-10-26 15:59:30 +02:00
|
|
|
|
|
|
|
|
/** Read the first page of a data file.
|
|
|
|
|
@return whether the page was found valid */
|
|
|
|
|
bool read_page0();
|
|
|
|
|
|
2020-10-26 16:04:12 +02:00
|
|
|
/** Determine the next tablespace for encryption key rotation.
|
|
|
|
|
@param space current tablespace (nullptr to start from the beginning)
|
|
|
|
|
@param recheck whether the removal condition needs to be rechecked after
|
|
|
|
|
encryption parameters were changed
|
|
|
|
|
@param encrypt expected state of innodb_encrypt_tables
|
|
|
|
|
@return the next tablespace
|
|
|
|
|
@retval nullptr upon reaching the end of the iteration */
|
2021-03-20 01:31:01 +03:00
|
|
|
static space_list_t::iterator next(space_list_t::iterator space,
|
|
|
|
|
bool recheck, bool encrypt);
|
2020-10-26 16:04:12 +02:00
|
|
|
|
2020-11-20 12:30:55 +02:00
|
|
|
#ifdef UNIV_DEBUG
|
|
|
|
|
bool is_latched() const { return latch_count != 0; }
|
|
|
|
|
#endif
|
2021-01-14 16:35:05 +02:00
|
|
|
bool is_owner() const { return latch_owner == os_thread_get_curr_id(); }
|
2020-11-20 12:30:55 +02:00
|
|
|
/** Acquire the allocation latch in exclusive mode */
|
|
|
|
|
void x_lock()
|
|
|
|
|
{
|
2020-12-03 09:55:53 +02:00
|
|
|
latch.wr_lock(SRW_LOCK_CALL);
|
2020-11-20 12:30:55 +02:00
|
|
|
ut_ad(!latch_owner);
|
2021-01-14 16:35:05 +02:00
|
|
|
latch_owner= os_thread_get_curr_id();
|
2020-11-20 12:30:55 +02:00
|
|
|
ut_ad(!latch_count.fetch_add(1));
|
|
|
|
|
}
|
|
|
|
|
/** Release the allocation latch from exclusive mode */
|
|
|
|
|
void x_unlock()
|
|
|
|
|
{
|
|
|
|
|
ut_ad(latch_count.fetch_sub(1) == 1);
|
|
|
|
|
ut_ad(latch_owner == os_thread_get_curr_id());
|
2021-01-14 16:35:05 +02:00
|
|
|
latch_owner= 0;
|
2020-11-20 12:30:55 +02:00
|
|
|
latch.wr_unlock();
|
|
|
|
|
}
|
|
|
|
|
/** Acquire the allocation latch in shared mode */
|
|
|
|
|
void s_lock()
|
|
|
|
|
{
|
2021-01-14 16:35:05 +02:00
|
|
|
ut_ad(!is_owner());
|
2020-12-03 09:55:53 +02:00
|
|
|
latch.rd_lock(SRW_LOCK_CALL);
|
2020-11-20 12:30:55 +02:00
|
|
|
ut_ad(!latch_owner);
|
|
|
|
|
ut_d(latch_count.fetch_add(1));
|
|
|
|
|
}
|
|
|
|
|
/** Release the allocation latch from shared mode */
|
|
|
|
|
void s_unlock()
|
|
|
|
|
{
|
|
|
|
|
ut_ad(latch_count.fetch_sub(1));
|
|
|
|
|
ut_ad(!latch_owner);
|
|
|
|
|
latch.rd_unlock();
|
|
|
|
|
}
|
|
|
|
|
|
MDEV-25312 Replace fil_space_t::name with fil_space_t::name()
A consistency check for fil_space_t::name is causing recovery failures
in MDEV-25180 (Atomic ALTER TABLE). So, we'd better remove that field
altogether.
fil_space_t::name was more or less a copy of dict_table_t::name
(except for some special cases), and it was not being used for
anything useful.
There used to be a name_hash, but it had been removed already in
commit a75dbfd7183cc96680f3e3e684fd36500dac8158 (MDEV-12266).
We will also remove os_normalize_path(), OS_PATH_SEPARATOR,
OS_PATH_SEPATOR_ALT. On Microsoft Windows, we will treat \ and /
roughly in the same way. The intention is that for per-table
tablespaces, the filenames will always follow the pattern
prefix/databasename/tablename.ibd. (Any \ in the prefix must not
be converted.)
ut_basename_noext(): Remove (unused function).
read_link_file(): Replaces RemoteDatafile::read_link_file().
We will ensure that the last two path component separators are
forward slashes (converting up to 2 trailing backslashes on
Microsoft Windows), so that everywhere else we can
assume that data file names end in "/databasename/tablename.ibd".
Note: On Microsoft Windows, path names that start with \\?\ must
not contain / as path component separators. Previously, such paths
did work in the DATA DIRECTORY argument of InnoDB tables.
Reviewed by: Vladislav Vaintroub
2021-04-07 18:01:13 +03:00
|
|
|
typedef span<const char> name_type;
|
|
|
|
|
|
|
|
|
|
/** @return the tablespace name (databasename/tablename) */
|
|
|
|
|
name_type name() const;
|
|
|
|
|
|
2020-10-26 15:59:30 +02:00
|
|
|
private:
|
|
|
|
|
/** @return whether the file is usable for io() */
|
2020-10-26 16:04:12 +02:00
|
|
|
ATTRIBUTE_COLD bool prepare(bool have_mutex= false);
|
2020-10-26 15:59:30 +02:00
|
|
|
#endif /*!UNIV_INNOCHECKSUM */
|
2016-08-12 11:17:45 +03:00
|
|
|
};
|
|
|
|
|
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
#ifndef UNIV_INNOCHECKSUM
|
2016-08-12 11:17:45 +03:00
|
|
|
/** Value of fil_space_t::magic_n */
|
|
|
|
|
#define FIL_SPACE_MAGIC_N 89472
|
|
|
|
|
|
|
|
|
|
/** File node of a tablespace or the log data space */
|
2020-10-26 16:04:12 +02:00
|
|
|
struct fil_node_t final
|
|
|
|
|
{
|
2021-06-28 13:42:43 +03:00
|
|
|
/** tablespace containing this file */
|
|
|
|
|
fil_space_t *space;
|
|
|
|
|
/** file name; protected by fil_system.mutex and log_sys.mutex */
|
|
|
|
|
char *name;
|
|
|
|
|
/** file handle */
|
|
|
|
|
pfs_os_file_t handle;
|
|
|
|
|
/** whether the file is on non-rotational media (SSD) */
|
|
|
|
|
unsigned on_ssd:1;
|
|
|
|
|
/** how to write page_compressed tables
|
|
|
|
|
(0=do not punch holes but write minimal amount of data, 1=punch holes,
|
|
|
|
|
2=always write the same amount; thinly provisioned storage will compress) */
|
|
|
|
|
unsigned punch_hole:2;
|
|
|
|
|
/** whether this file could use atomic write */
|
|
|
|
|
unsigned atomic_write:1;
|
|
|
|
|
/** whether the file actually is a raw device or disk partition */
|
|
|
|
|
unsigned is_raw_disk:1;
|
|
|
|
|
/** whether the tablespace discovery is being deferred during crash
|
|
|
|
|
recovery due to incompletely written page 0 */
|
|
|
|
|
unsigned deferred:1;
|
|
|
|
|
|
|
|
|
|
/** size of the file in database pages (0 if not known yet);
|
|
|
|
|
the possible last incomplete megabyte may be ignored if space->id == 0 */
|
|
|
|
|
uint32_t size;
|
|
|
|
|
/** initial size of the file in database pages;
|
|
|
|
|
FIL_IBD_FILE_INITIAL_SIZE by default */
|
|
|
|
|
uint32_t init_size;
|
|
|
|
|
/** maximum size of the file in database pages (0 if unlimited) */
|
|
|
|
|
uint32_t max_size;
|
|
|
|
|
/** whether the file is currently being extended */
|
|
|
|
|
Atomic_relaxed<bool> being_extended;
|
|
|
|
|
/** link to other files in this tablespace */
|
|
|
|
|
UT_LIST_NODE_T(fil_node_t) chain;
|
|
|
|
|
|
|
|
|
|
/** Filesystem block size */
|
|
|
|
|
ulint block_size;
|
|
|
|
|
|
|
|
|
|
/** @return whether this file is open */
|
|
|
|
|
bool is_open() const { return handle != OS_FILE_CLOSED; }
|
2018-08-16 08:54:58 +03:00
|
|
|
|
2021-06-28 13:42:43 +03:00
|
|
|
/** Read the first page of a data file.
|
|
|
|
|
@return whether the page was found valid */
|
|
|
|
|
bool read_page0();
|
2018-11-28 01:02:46 +02:00
|
|
|
|
2021-06-28 13:42:43 +03:00
|
|
|
/** Determine some file metadata when creating or reading the file.
|
|
|
|
|
@param file the file that is being created, or OS_FILE_CLOSED */
|
|
|
|
|
void find_metadata(os_file_t file= OS_FILE_CLOSED
|
2021-03-18 14:43:08 +02:00
|
|
|
#ifndef _WIN32
|
2021-06-28 13:42:43 +03:00
|
|
|
, bool create= false, struct stat *statbuf= nullptr
|
MDEV-17380 innodb_flush_neighbors=ON should be ignored on SSD
For tablespaces that do not reside on spinning storage, it does
not make sense to attempt to write nearby pages when writing out
dirty pages from the InnoDB buffer pool. It is actually detrimental
to performance and to the life span of flash ROM storage.
With this change, MariaDB will detect whether an InnoDB file resides
on solid-state storage. The detection has been implemented for Linux
and Microsoft Windows. For other systems, we will err on the safe side
and assume that files reside on SSD.
As part of this change, we will reduce the number of fstat() calls
when opening data files on POSIX systems and slightly clean up some
file I/O code.
FIXME: os_is_sparse_file_supported() on POSIX works in a destructive
manner. Thus, we can only invoke it when creating files, not when
opening them.
For diagnostics, we introduce the column ON_SSD to the table
INFORMATION_SCHEMA.INNODB_TABLESPACES_SCRUBBING. The table
INNODB_SYS_TABLESPACES might seem more appropriate, but its purpose
is to reflect the contents of the InnoDB system table SYS_TABLESPACES,
which we would like to remove at some point.
On Microsoft Windows, querying StorageDeviceSeekPenaltyProperty
sometimes returns ERROR_GEN_FAILURE instead of ERROR_INVALID_FUNCTION
or ERROR_NOT_SUPPORTED. We will silently ignore also this error,
and assume that the file does not reside on SSD.
On Linux, the detection will be based on the files
/sys/block/*/queue/rotational and /sys/block/*/dev.
Especially for USB storage, it is possible that
/sys/block/*/queue/rotational will wrongly report 1 instead of 0.
fil_node_t::on_ssd: Whether the InnoDB data file resides on
solid-state storage.
fil_system_t::ssd: Collection of Linux block devices that reside on
non-rotational storage.
fil_system_t::create(): Detect ssd on Linux based on the contents
of /sys/block/*/queue/rotational and /sys/block/*/dev.
fil_system_t::is_ssd(dev_t): Determine if a Linux block device is
non-rotational. Partitions will be identified with the containing
block device by assuming that the least significant 4 bits of the
minor number identify a partition, and that the "partition number"
of the entire device is 0.
2019-04-01 11:57:06 +03:00
|
|
|
#endif
|
2021-06-28 13:42:43 +03:00
|
|
|
);
|
MDEV-17380 innodb_flush_neighbors=ON should be ignored on SSD
For tablespaces that do not reside on spinning storage, it does
not make sense to attempt to write nearby pages when writing out
dirty pages from the InnoDB buffer pool. It is actually detrimental
to performance and to the life span of flash ROM storage.
With this change, MariaDB will detect whether an InnoDB file resides
on solid-state storage. The detection has been implemented for Linux
and Microsoft Windows. For other systems, we will err on the safe side
and assume that files reside on SSD.
As part of this change, we will reduce the number of fstat() calls
when opening data files on POSIX systems and slightly clean up some
file I/O code.
FIXME: os_is_sparse_file_supported() on POSIX works in a destructive
manner. Thus, we can only invoke it when creating files, not when
opening them.
For diagnostics, we introduce the column ON_SSD to the table
INFORMATION_SCHEMA.INNODB_TABLESPACES_SCRUBBING. The table
INNODB_SYS_TABLESPACES might seem more appropriate, but its purpose
is to reflect the contents of the InnoDB system table SYS_TABLESPACES,
which we would like to remove at some point.
On Microsoft Windows, querying StorageDeviceSeekPenaltyProperty
sometimes returns ERROR_GEN_FAILURE instead of ERROR_INVALID_FUNCTION
or ERROR_NOT_SUPPORTED. We will silently ignore also this error,
and assume that the file does not reside on SSD.
On Linux, the detection will be based on the files
/sys/block/*/queue/rotational and /sys/block/*/dev.
Especially for USB storage, it is possible that
/sys/block/*/queue/rotational will wrongly report 1 instead of 0.
fil_node_t::on_ssd: Whether the InnoDB data file resides on
solid-state storage.
fil_system_t::ssd: Collection of Linux block devices that reside on
non-rotational storage.
fil_system_t::create(): Detect ssd on Linux based on the contents
of /sys/block/*/queue/rotational and /sys/block/*/dev.
fil_system_t::is_ssd(dev_t): Determine if a Linux block device is
non-rotational. Partitions will be identified with the containing
block device by assuming that the least significant 4 bits of the
minor number identify a partition, and that the "partition number"
of the entire device is 0.
2019-04-01 11:57:06 +03:00
|
|
|
|
MDEV-15053 Reduce buf_pool_t::mutex contention
User-visible changes: The INFORMATION_SCHEMA views INNODB_BUFFER_PAGE
and INNODB_BUFFER_PAGE_LRU will report a dummy value FLUSH_TYPE=0
and will no longer report the PAGE_STATE value READY_FOR_USE.
We will remove some fields from buf_page_t and move much code to
member functions of buf_pool_t and buf_page_t, so that the access
rules of data members can be enforced consistently.
Evicting or adding pages in buf_pool.LRU will remain covered by
buf_pool.mutex.
Evicting or adding pages in buf_pool.page_hash will remain
covered by both buf_pool.mutex and the buf_pool.page_hash X-latch.
After this fix, buf_pool.page_hash lookups can entirely
avoid acquiring buf_pool.mutex, only relying on
buf_pool.hash_lock_get() S-latch.
Similarly, buf_flush_check_neighbors() can will rely solely on
buf_pool.mutex, no buf_pool.page_hash latch at all.
The buf_pool.mutex is rather contended in I/O heavy benchmarks,
especially when the workload does not fit in the buffer pool.
The first attempt to alleviate the contention was the
buf_pool_t::mutex split in
commit 4ed7082eefe56b3e97e0edefb3df76dd7ef5e858
which introduced buf_block_t::mutex, which we are now removing.
Later, multiple instances of buf_pool_t were introduced
in commit c18084f71b02ea707c6461353e6cfc15d7553bc6
and recently removed by us in
commit 1a6f708ec594ac0ae2dd30db926ab07b100fa24b (MDEV-15058).
UNIV_BUF_DEBUG: Remove. This option to enable some buffer pool
related debugging in otherwise non-debug builds has not been used
for years. Instead, we have been using UNIV_DEBUG, which is enabled
in CMAKE_BUILD_TYPE=Debug.
buf_block_t::mutex, buf_pool_t::zip_mutex: Remove. We can mainly rely on
std::atomic and the buf_pool.page_hash latches, and in some cases
depend on buf_pool.mutex or buf_pool.flush_list_mutex just like before.
We must always release buf_block_t::lock before invoking
unfix() or io_unfix(), to prevent a glitch where a block that was
added to the buf_pool.free list would apper X-latched. See
commit c5883debd6ef440a037011c11873b396923e93c5 how this glitch
was finally caught in a debug environment.
We move some buf_pool_t::page_hash specific code from the
ha and hash modules to buf_pool, for improved readability.
buf_pool_t::close(): Assert that all blocks are clean, except
on aborted startup or crash-like shutdown.
buf_pool_t::validate(): No longer attempt to validate
n_flush[] against the number of BUF_IO_WRITE fixed blocks,
because buf_page_t::flush_type no longer exists.
buf_pool_t::watch_set(): Replaces buf_pool_watch_set().
Reduce mutex contention by separating the buf_pool.watch[]
allocation and the insert into buf_pool.page_hash.
buf_pool_t::page_hash_lock<bool exclusive>(): Acquire a
buf_pool.page_hash latch.
Replaces and extends buf_page_hash_lock_s_confirm()
and buf_page_hash_lock_x_confirm().
buf_pool_t::READ_AHEAD_PAGES: Renamed from BUF_READ_AHEAD_PAGES.
buf_pool_t::curr_size, old_size, read_ahead_area, n_pend_reads:
Use Atomic_counter.
buf_pool_t::running_out(): Replaces buf_LRU_buf_pool_running_out().
buf_pool_t::LRU_remove(): Remove a block from the LRU list
and return its predecessor. Incorporates buf_LRU_adjust_hp(),
which was removed.
buf_page_get_gen(): Remove a redundant call of fsp_is_system_temporary(),
for mode == BUF_GET_IF_IN_POOL_OR_WATCH, which is only used by
BTR_DELETE_OP (purge), which is never invoked on temporary tables.
buf_free_from_unzip_LRU_list_batch(): Avoid redundant assignments.
buf_LRU_free_from_unzip_LRU_list(): Simplify the loop condition.
buf_LRU_free_page(): Clarify the function comment.
buf_flush_check_neighbor(), buf_flush_check_neighbors():
Rewrite the construction of the page hash range. We will hold
the buf_pool.mutex for up to buf_pool.read_ahead_area (at most 64)
consecutive lookups of buf_pool.page_hash.
buf_flush_page_and_try_neighbors(): Remove.
Merge to its only callers, and remove redundant operations in
buf_flush_LRU_list_batch().
buf_read_ahead_random(), buf_read_ahead_linear(): Rewrite.
Do not acquire buf_pool.mutex, and iterate directly with page_id_t.
ut_2_power_up(): Remove. my_round_up_to_next_power() is inlined
and avoids any loops.
fil_page_get_prev(), fil_page_get_next(), fil_addr_is_null(): Remove.
buf_flush_page(): Add a fil_space_t* parameter. Minimize the
buf_pool.mutex hold time. buf_pool.n_flush[] is no longer updated
atomically with the io_fix, and we will protect most buf_block_t
fields with buf_block_t::lock. The function
buf_flush_write_block_low() is removed and merged here.
buf_page_init_for_read(): Use static linkage. Initialize the newly
allocated block and acquire the exclusive buf_block_t::lock while not
holding any mutex.
IORequest::IORequest(): Remove the body. We only need to invoke
set_punch_hole() in buf_flush_page() and nowhere else.
buf_page_t::flush_type: Remove. Replaced by IORequest::flush_type.
This field is only used during a fil_io() call.
That function already takes IORequest as a parameter, so we had
better introduce for the rarely changing field.
buf_block_t::init(): Replaces buf_page_init().
buf_page_t::init(): Replaces buf_page_init_low().
buf_block_t::initialise(): Initialise many fields, but
keep the buf_page_t::state(). Both buf_pool_t::validate() and
buf_page_optimistic_get() requires that buf_page_t::in_file()
be protected atomically with buf_page_t::in_page_hash
and buf_page_t::in_LRU_list.
buf_page_optimistic_get(): Now that buf_block_t::mutex
no longer exists, we must check buf_page_t::io_fix()
after acquiring the buf_pool.page_hash lock, to detect
whether buf_page_init_for_read() has been initiated.
We will also check the io_fix() before acquiring hash_lock
in order to avoid unnecessary computation.
The field buf_block_t::modify_clock (protected by buf_block_t::lock)
allows buf_page_optimistic_get() to validate the block.
buf_page_t::real_size: Remove. It was only used while flushing
pages of page_compressed tables.
buf_page_encrypt(): Add an output parameter that allows us ot eliminate
buf_page_t::real_size. Replace a condition with debug assertion.
buf_page_should_punch_hole(): Remove.
buf_dblwr_t::add_to_batch(): Replaces buf_dblwr_add_to_batch().
Add the parameter size (to replace buf_page_t::real_size).
buf_dblwr_t::write_single_page(): Replaces buf_dblwr_write_single_page().
Add the parameter size (to replace buf_page_t::real_size).
fil_system_t::detach(): Replaces fil_space_detach().
Ensure that fil_validate() will not be violated even if
fil_system.mutex is released and reacquired.
fil_node_t::complete_io(): Renamed from fil_node_complete_io().
fil_node_t::close_to_free(): Replaces fil_node_close_to_free().
Avoid invoking fil_node_t::close() because fil_system.n_open
has already been decremented in fil_space_t::detach().
BUF_BLOCK_READY_FOR_USE: Remove. Directly use BUF_BLOCK_MEMORY.
BUF_BLOCK_ZIP_DIRTY: Remove. Directly use BUF_BLOCK_ZIP_PAGE,
and distinguish dirty pages by buf_page_t::oldest_modification().
BUF_BLOCK_POOL_WATCH: Remove. Use BUF_BLOCK_NOT_USED instead.
This state was only being used for buf_page_t that are in
buf_pool.watch.
buf_pool_t::watch[]: Remove pointer indirection.
buf_page_t::in_flush_list: Remove. It was set if and only if
buf_page_t::oldest_modification() is nonzero.
buf_page_decrypt_after_read(), buf_corrupt_page_release(),
buf_page_check_corrupt(): Change the const fil_space_t* parameter
to const fil_node_t& so that we can report the correct file name.
buf_page_monitor(): Declare as an ATTRIBUTE_COLD global function.
buf_page_io_complete(): Split to buf_page_read_complete() and
buf_page_write_complete().
buf_dblwr_t::in_use: Remove.
buf_dblwr_t::buf_block_array: Add IORequest::flush_t.
buf_dblwr_sync_datafiles(): Remove. It was a useless wrapper of
os_aio_wait_until_no_pending_writes().
buf_flush_write_complete(): Declare static, not global.
Add the parameter IORequest::flush_t.
buf_flush_freed_page(): Simplify the code.
recv_sys_t::flush_lru: Renamed from flush_type and changed to bool.
fil_read(), fil_write(): Replaced with direct use of fil_io().
fil_buffering_disabled(): Remove. Check srv_file_flush_method directly.
fil_mutex_enter_and_prepare_for_io(): Return the resolved
fil_space_t* to avoid a duplicated lookup in the caller.
fil_report_invalid_page_access(): Clean up the parameters.
fil_io(): Return fil_io_t, which comprises fil_node_t and error code.
Always invoke fil_space_t::acquire_for_io() and let either the
sync=true caller or fil_aio_callback() invoke
fil_space_t::release_for_io().
fil_aio_callback(): Rewrite to replace buf_page_io_complete().
fil_check_pending_operations(): Remove a parameter, and remove some
redundant lookups.
fil_node_close_to_free(): Wait for n_pending==0. Because we no longer
do an extra lookup of the tablespace between fil_io() and the
completion of the operation, we must give fil_node_t::complete_io() a
chance to decrement the counter.
fil_close_tablespace(): Remove unused parameter trx, and document
that this is only invoked during the error handling of IMPORT TABLESPACE.
row_import_discard_changes(): Merged with the only caller,
row_import_cleanup(). Do not lock up the data dictionary while
invoking fil_close_tablespace().
logs_empty_and_mark_files_at_shutdown(): Do not invoke
fil_close_all_files(), to avoid a !needs_flush assertion failure
on fil_node_t::close().
innodb_shutdown(): Invoke os_aio_free() before fil_close_all_files().
fil_close_all_files(): Invoke fil_flush_file_spaces()
to ensure proper durability.
thread_pool::unbind(): Fix a crash that would occur on Windows
after srv_thread_pool->disable_aio() and os_file_close().
This fix was submitted by Vladislav Vaintroub.
Thanks to Matthias Leich and Axel Schwenke for extensive testing,
Vladislav Vaintroub for helpful comments, and Eugene Kosov for a review.
2020-06-05 12:35:46 +03:00
|
|
|
/** Close the file handle. */
|
|
|
|
|
void close();
|
2020-05-29 01:24:50 +03:00
|
|
|
/** Same as close() but returns file handle instead of closing it. */
|
|
|
|
|
pfs_os_file_t detach() MY_ATTRIBUTE((warn_unused_result));
|
|
|
|
|
/** Prepare to free a file from fil_system.
|
|
|
|
|
@param detach_handle whether to detach instead of closing a handle
|
|
|
|
|
@return detached handle or OS_FILE_CLOSED */
|
2020-10-26 16:04:12 +02:00
|
|
|
inline pfs_os_file_t close_to_free(bool detach_handle= false);
|
MDEV-15053 Reduce buf_pool_t::mutex contention
User-visible changes: The INFORMATION_SCHEMA views INNODB_BUFFER_PAGE
and INNODB_BUFFER_PAGE_LRU will report a dummy value FLUSH_TYPE=0
and will no longer report the PAGE_STATE value READY_FOR_USE.
We will remove some fields from buf_page_t and move much code to
member functions of buf_pool_t and buf_page_t, so that the access
rules of data members can be enforced consistently.
Evicting or adding pages in buf_pool.LRU will remain covered by
buf_pool.mutex.
Evicting or adding pages in buf_pool.page_hash will remain
covered by both buf_pool.mutex and the buf_pool.page_hash X-latch.
After this fix, buf_pool.page_hash lookups can entirely
avoid acquiring buf_pool.mutex, only relying on
buf_pool.hash_lock_get() S-latch.
Similarly, buf_flush_check_neighbors() can will rely solely on
buf_pool.mutex, no buf_pool.page_hash latch at all.
The buf_pool.mutex is rather contended in I/O heavy benchmarks,
especially when the workload does not fit in the buffer pool.
The first attempt to alleviate the contention was the
buf_pool_t::mutex split in
commit 4ed7082eefe56b3e97e0edefb3df76dd7ef5e858
which introduced buf_block_t::mutex, which we are now removing.
Later, multiple instances of buf_pool_t were introduced
in commit c18084f71b02ea707c6461353e6cfc15d7553bc6
and recently removed by us in
commit 1a6f708ec594ac0ae2dd30db926ab07b100fa24b (MDEV-15058).
UNIV_BUF_DEBUG: Remove. This option to enable some buffer pool
related debugging in otherwise non-debug builds has not been used
for years. Instead, we have been using UNIV_DEBUG, which is enabled
in CMAKE_BUILD_TYPE=Debug.
buf_block_t::mutex, buf_pool_t::zip_mutex: Remove. We can mainly rely on
std::atomic and the buf_pool.page_hash latches, and in some cases
depend on buf_pool.mutex or buf_pool.flush_list_mutex just like before.
We must always release buf_block_t::lock before invoking
unfix() or io_unfix(), to prevent a glitch where a block that was
added to the buf_pool.free list would apper X-latched. See
commit c5883debd6ef440a037011c11873b396923e93c5 how this glitch
was finally caught in a debug environment.
We move some buf_pool_t::page_hash specific code from the
ha and hash modules to buf_pool, for improved readability.
buf_pool_t::close(): Assert that all blocks are clean, except
on aborted startup or crash-like shutdown.
buf_pool_t::validate(): No longer attempt to validate
n_flush[] against the number of BUF_IO_WRITE fixed blocks,
because buf_page_t::flush_type no longer exists.
buf_pool_t::watch_set(): Replaces buf_pool_watch_set().
Reduce mutex contention by separating the buf_pool.watch[]
allocation and the insert into buf_pool.page_hash.
buf_pool_t::page_hash_lock<bool exclusive>(): Acquire a
buf_pool.page_hash latch.
Replaces and extends buf_page_hash_lock_s_confirm()
and buf_page_hash_lock_x_confirm().
buf_pool_t::READ_AHEAD_PAGES: Renamed from BUF_READ_AHEAD_PAGES.
buf_pool_t::curr_size, old_size, read_ahead_area, n_pend_reads:
Use Atomic_counter.
buf_pool_t::running_out(): Replaces buf_LRU_buf_pool_running_out().
buf_pool_t::LRU_remove(): Remove a block from the LRU list
and return its predecessor. Incorporates buf_LRU_adjust_hp(),
which was removed.
buf_page_get_gen(): Remove a redundant call of fsp_is_system_temporary(),
for mode == BUF_GET_IF_IN_POOL_OR_WATCH, which is only used by
BTR_DELETE_OP (purge), which is never invoked on temporary tables.
buf_free_from_unzip_LRU_list_batch(): Avoid redundant assignments.
buf_LRU_free_from_unzip_LRU_list(): Simplify the loop condition.
buf_LRU_free_page(): Clarify the function comment.
buf_flush_check_neighbor(), buf_flush_check_neighbors():
Rewrite the construction of the page hash range. We will hold
the buf_pool.mutex for up to buf_pool.read_ahead_area (at most 64)
consecutive lookups of buf_pool.page_hash.
buf_flush_page_and_try_neighbors(): Remove.
Merge to its only callers, and remove redundant operations in
buf_flush_LRU_list_batch().
buf_read_ahead_random(), buf_read_ahead_linear(): Rewrite.
Do not acquire buf_pool.mutex, and iterate directly with page_id_t.
ut_2_power_up(): Remove. my_round_up_to_next_power() is inlined
and avoids any loops.
fil_page_get_prev(), fil_page_get_next(), fil_addr_is_null(): Remove.
buf_flush_page(): Add a fil_space_t* parameter. Minimize the
buf_pool.mutex hold time. buf_pool.n_flush[] is no longer updated
atomically with the io_fix, and we will protect most buf_block_t
fields with buf_block_t::lock. The function
buf_flush_write_block_low() is removed and merged here.
buf_page_init_for_read(): Use static linkage. Initialize the newly
allocated block and acquire the exclusive buf_block_t::lock while not
holding any mutex.
IORequest::IORequest(): Remove the body. We only need to invoke
set_punch_hole() in buf_flush_page() and nowhere else.
buf_page_t::flush_type: Remove. Replaced by IORequest::flush_type.
This field is only used during a fil_io() call.
That function already takes IORequest as a parameter, so we had
better introduce for the rarely changing field.
buf_block_t::init(): Replaces buf_page_init().
buf_page_t::init(): Replaces buf_page_init_low().
buf_block_t::initialise(): Initialise many fields, but
keep the buf_page_t::state(). Both buf_pool_t::validate() and
buf_page_optimistic_get() requires that buf_page_t::in_file()
be protected atomically with buf_page_t::in_page_hash
and buf_page_t::in_LRU_list.
buf_page_optimistic_get(): Now that buf_block_t::mutex
no longer exists, we must check buf_page_t::io_fix()
after acquiring the buf_pool.page_hash lock, to detect
whether buf_page_init_for_read() has been initiated.
We will also check the io_fix() before acquiring hash_lock
in order to avoid unnecessary computation.
The field buf_block_t::modify_clock (protected by buf_block_t::lock)
allows buf_page_optimistic_get() to validate the block.
buf_page_t::real_size: Remove. It was only used while flushing
pages of page_compressed tables.
buf_page_encrypt(): Add an output parameter that allows us ot eliminate
buf_page_t::real_size. Replace a condition with debug assertion.
buf_page_should_punch_hole(): Remove.
buf_dblwr_t::add_to_batch(): Replaces buf_dblwr_add_to_batch().
Add the parameter size (to replace buf_page_t::real_size).
buf_dblwr_t::write_single_page(): Replaces buf_dblwr_write_single_page().
Add the parameter size (to replace buf_page_t::real_size).
fil_system_t::detach(): Replaces fil_space_detach().
Ensure that fil_validate() will not be violated even if
fil_system.mutex is released and reacquired.
fil_node_t::complete_io(): Renamed from fil_node_complete_io().
fil_node_t::close_to_free(): Replaces fil_node_close_to_free().
Avoid invoking fil_node_t::close() because fil_system.n_open
has already been decremented in fil_space_t::detach().
BUF_BLOCK_READY_FOR_USE: Remove. Directly use BUF_BLOCK_MEMORY.
BUF_BLOCK_ZIP_DIRTY: Remove. Directly use BUF_BLOCK_ZIP_PAGE,
and distinguish dirty pages by buf_page_t::oldest_modification().
BUF_BLOCK_POOL_WATCH: Remove. Use BUF_BLOCK_NOT_USED instead.
This state was only being used for buf_page_t that are in
buf_pool.watch.
buf_pool_t::watch[]: Remove pointer indirection.
buf_page_t::in_flush_list: Remove. It was set if and only if
buf_page_t::oldest_modification() is nonzero.
buf_page_decrypt_after_read(), buf_corrupt_page_release(),
buf_page_check_corrupt(): Change the const fil_space_t* parameter
to const fil_node_t& so that we can report the correct file name.
buf_page_monitor(): Declare as an ATTRIBUTE_COLD global function.
buf_page_io_complete(): Split to buf_page_read_complete() and
buf_page_write_complete().
buf_dblwr_t::in_use: Remove.
buf_dblwr_t::buf_block_array: Add IORequest::flush_t.
buf_dblwr_sync_datafiles(): Remove. It was a useless wrapper of
os_aio_wait_until_no_pending_writes().
buf_flush_write_complete(): Declare static, not global.
Add the parameter IORequest::flush_t.
buf_flush_freed_page(): Simplify the code.
recv_sys_t::flush_lru: Renamed from flush_type and changed to bool.
fil_read(), fil_write(): Replaced with direct use of fil_io().
fil_buffering_disabled(): Remove. Check srv_file_flush_method directly.
fil_mutex_enter_and_prepare_for_io(): Return the resolved
fil_space_t* to avoid a duplicated lookup in the caller.
fil_report_invalid_page_access(): Clean up the parameters.
fil_io(): Return fil_io_t, which comprises fil_node_t and error code.
Always invoke fil_space_t::acquire_for_io() and let either the
sync=true caller or fil_aio_callback() invoke
fil_space_t::release_for_io().
fil_aio_callback(): Rewrite to replace buf_page_io_complete().
fil_check_pending_operations(): Remove a parameter, and remove some
redundant lookups.
fil_node_close_to_free(): Wait for n_pending==0. Because we no longer
do an extra lookup of the tablespace between fil_io() and the
completion of the operation, we must give fil_node_t::complete_io() a
chance to decrement the counter.
fil_close_tablespace(): Remove unused parameter trx, and document
that this is only invoked during the error handling of IMPORT TABLESPACE.
row_import_discard_changes(): Merged with the only caller,
row_import_cleanup(). Do not lock up the data dictionary while
invoking fil_close_tablespace().
logs_empty_and_mark_files_at_shutdown(): Do not invoke
fil_close_all_files(), to avoid a !needs_flush assertion failure
on fil_node_t::close().
innodb_shutdown(): Invoke os_aio_free() before fil_close_all_files().
fil_close_all_files(): Invoke fil_flush_file_spaces()
to ensure proper durability.
thread_pool::unbind(): Fix a crash that would occur on Windows
after srv_thread_pool->disable_aio() and os_file_close().
This fix was submitted by Vladislav Vaintroub.
Thanks to Matthias Leich and Axel Schwenke for extensive testing,
Vladislav Vaintroub for helpful comments, and Eugene Kosov for a review.
2020-06-05 12:35:46 +03:00
|
|
|
|
2020-10-26 15:59:30 +02:00
|
|
|
/** Update the data structures on write completion */
|
|
|
|
|
inline void complete_write();
|
2020-05-29 01:24:50 +03:00
|
|
|
|
|
|
|
|
private:
|
|
|
|
|
/** Does stuff common for close() and detach() */
|
|
|
|
|
void prepare_to_close_or_detach();
|
2016-08-12 11:17:45 +03:00
|
|
|
};
|
|
|
|
|
|
2021-06-28 13:42:43 +03:00
|
|
|
inline bool fil_space_t::use_doublewrite() const
|
|
|
|
|
{
|
|
|
|
|
return !UT_LIST_GET_FIRST(chain)->atomic_write && srv_use_doublewrite_buf &&
|
|
|
|
|
buf_dblwr.is_initialised();
|
|
|
|
|
}
|
2016-08-12 11:17:45 +03:00
|
|
|
|
MDEV-17380 innodb_flush_neighbors=ON should be ignored on SSD
For tablespaces that do not reside on spinning storage, it does
not make sense to attempt to write nearby pages when writing out
dirty pages from the InnoDB buffer pool. It is actually detrimental
to performance and to the life span of flash ROM storage.
With this change, MariaDB will detect whether an InnoDB file resides
on solid-state storage. The detection has been implemented for Linux
and Microsoft Windows. For other systems, we will err on the safe side
and assume that files reside on SSD.
As part of this change, we will reduce the number of fstat() calls
when opening data files on POSIX systems and slightly clean up some
file I/O code.
FIXME: os_is_sparse_file_supported() on POSIX works in a destructive
manner. Thus, we can only invoke it when creating files, not when
opening them.
For diagnostics, we introduce the column ON_SSD to the table
INFORMATION_SCHEMA.INNODB_TABLESPACES_SCRUBBING. The table
INNODB_SYS_TABLESPACES might seem more appropriate, but its purpose
is to reflect the contents of the InnoDB system table SYS_TABLESPACES,
which we would like to remove at some point.
On Microsoft Windows, querying StorageDeviceSeekPenaltyProperty
sometimes returns ERROR_GEN_FAILURE instead of ERROR_INVALID_FUNCTION
or ERROR_NOT_SUPPORTED. We will silently ignore also this error,
and assume that the file does not reside on SSD.
On Linux, the detection will be based on the files
/sys/block/*/queue/rotational and /sys/block/*/dev.
Especially for USB storage, it is possible that
/sys/block/*/queue/rotational will wrongly report 1 instead of 0.
fil_node_t::on_ssd: Whether the InnoDB data file resides on
solid-state storage.
fil_system_t::ssd: Collection of Linux block devices that reside on
non-rotational storage.
fil_system_t::create(): Detect ssd on Linux based on the contents
of /sys/block/*/queue/rotational and /sys/block/*/dev.
fil_system_t::is_ssd(dev_t): Determine if a Linux block device is
non-rotational. Partitions will be identified with the containing
block device by assuming that the least significant 4 bits of the
minor number identify a partition, and that the "partition number"
of the entire device is 0.
2019-04-01 11:57:06 +03:00
|
|
|
inline void fil_space_t::set_imported()
|
|
|
|
|
{
|
2020-10-26 15:59:30 +02:00
|
|
|
ut_ad(purpose == FIL_TYPE_IMPORT);
|
|
|
|
|
purpose= FIL_TYPE_TABLESPACE;
|
|
|
|
|
UT_LIST_GET_FIRST(chain)->find_metadata();
|
MDEV-17380 innodb_flush_neighbors=ON should be ignored on SSD
For tablespaces that do not reside on spinning storage, it does
not make sense to attempt to write nearby pages when writing out
dirty pages from the InnoDB buffer pool. It is actually detrimental
to performance and to the life span of flash ROM storage.
With this change, MariaDB will detect whether an InnoDB file resides
on solid-state storage. The detection has been implemented for Linux
and Microsoft Windows. For other systems, we will err on the safe side
and assume that files reside on SSD.
As part of this change, we will reduce the number of fstat() calls
when opening data files on POSIX systems and slightly clean up some
file I/O code.
FIXME: os_is_sparse_file_supported() on POSIX works in a destructive
manner. Thus, we can only invoke it when creating files, not when
opening them.
For diagnostics, we introduce the column ON_SSD to the table
INFORMATION_SCHEMA.INNODB_TABLESPACES_SCRUBBING. The table
INNODB_SYS_TABLESPACES might seem more appropriate, but its purpose
is to reflect the contents of the InnoDB system table SYS_TABLESPACES,
which we would like to remove at some point.
On Microsoft Windows, querying StorageDeviceSeekPenaltyProperty
sometimes returns ERROR_GEN_FAILURE instead of ERROR_INVALID_FUNCTION
or ERROR_NOT_SUPPORTED. We will silently ignore also this error,
and assume that the file does not reside on SSD.
On Linux, the detection will be based on the files
/sys/block/*/queue/rotational and /sys/block/*/dev.
Especially for USB storage, it is possible that
/sys/block/*/queue/rotational will wrongly report 1 instead of 0.
fil_node_t::on_ssd: Whether the InnoDB data file resides on
solid-state storage.
fil_system_t::ssd: Collection of Linux block devices that reside on
non-rotational storage.
fil_system_t::create(): Detect ssd on Linux based on the contents
of /sys/block/*/queue/rotational and /sys/block/*/dev.
fil_system_t::is_ssd(dev_t): Determine if a Linux block device is
non-rotational. Partitions will be identified with the containing
block device by assuming that the least significant 4 bits of the
minor number identify a partition, and that the "partition number"
of the entire device is 0.
2019-04-01 11:57:06 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
inline bool fil_space_t::is_rotational() const
|
|
|
|
|
{
|
2020-10-26 15:59:30 +02:00
|
|
|
for (const fil_node_t *node= UT_LIST_GET_FIRST(chain); node;
|
|
|
|
|
node= UT_LIST_GET_NEXT(chain, node))
|
|
|
|
|
if (!node->on_ssd)
|
|
|
|
|
return true;
|
|
|
|
|
return false;
|
MDEV-17380 innodb_flush_neighbors=ON should be ignored on SSD
For tablespaces that do not reside on spinning storage, it does
not make sense to attempt to write nearby pages when writing out
dirty pages from the InnoDB buffer pool. It is actually detrimental
to performance and to the life span of flash ROM storage.
With this change, MariaDB will detect whether an InnoDB file resides
on solid-state storage. The detection has been implemented for Linux
and Microsoft Windows. For other systems, we will err on the safe side
and assume that files reside on SSD.
As part of this change, we will reduce the number of fstat() calls
when opening data files on POSIX systems and slightly clean up some
file I/O code.
FIXME: os_is_sparse_file_supported() on POSIX works in a destructive
manner. Thus, we can only invoke it when creating files, not when
opening them.
For diagnostics, we introduce the column ON_SSD to the table
INFORMATION_SCHEMA.INNODB_TABLESPACES_SCRUBBING. The table
INNODB_SYS_TABLESPACES might seem more appropriate, but its purpose
is to reflect the contents of the InnoDB system table SYS_TABLESPACES,
which we would like to remove at some point.
On Microsoft Windows, querying StorageDeviceSeekPenaltyProperty
sometimes returns ERROR_GEN_FAILURE instead of ERROR_INVALID_FUNCTION
or ERROR_NOT_SUPPORTED. We will silently ignore also this error,
and assume that the file does not reside on SSD.
On Linux, the detection will be based on the files
/sys/block/*/queue/rotational and /sys/block/*/dev.
Especially for USB storage, it is possible that
/sys/block/*/queue/rotational will wrongly report 1 instead of 0.
fil_node_t::on_ssd: Whether the InnoDB data file resides on
solid-state storage.
fil_system_t::ssd: Collection of Linux block devices that reside on
non-rotational storage.
fil_system_t::create(): Detect ssd on Linux based on the contents
of /sys/block/*/queue/rotational and /sys/block/*/dev.
fil_system_t::is_ssd(dev_t): Determine if a Linux block device is
non-rotational. Partitions will be identified with the containing
block device by assuming that the least significant 4 bits of the
minor number identify a partition, and that the "partition number"
of the entire device is 0.
2019-04-01 11:57:06 +03:00
|
|
|
}
|
|
|
|
|
|
2021-05-17 18:12:33 +03:00
|
|
|
inline bool fil_space_t::is_deferred() const
|
|
|
|
|
{
|
|
|
|
|
return UT_LIST_GET_FIRST(chain)->deferred;
|
|
|
|
|
}
|
|
|
|
|
|
2018-12-21 17:06:08 -03:00
|
|
|
/** Common InnoDB file extensions */
|
2016-08-12 11:17:45 +03:00
|
|
|
enum ib_extention {
|
|
|
|
|
NO_EXT = 0,
|
|
|
|
|
IBD = 1,
|
|
|
|
|
ISL = 2,
|
2017-01-16 11:11:14 +02:00
|
|
|
CFG = 3
|
2016-08-12 11:17:45 +03:00
|
|
|
};
|
|
|
|
|
extern const char* dot_ext[];
|
|
|
|
|
#define DOT_IBD dot_ext[IBD]
|
|
|
|
|
#define DOT_ISL dot_ext[ISL]
|
|
|
|
|
#define DOT_CFG dot_ext[CFG]
|
|
|
|
|
|
2021-03-18 13:17:30 +05:30
|
|
|
/** When mariadbd is run, the default directory "." is the mysqld datadir,
|
|
|
|
|
but in the MariaDB Embedded Server Library and mysqlbackup it is not the default
|
2014-02-26 19:11:54 +01:00
|
|
|
directory, and we must set the base file path explicitly */
|
|
|
|
|
extern const char* fil_path_to_mysql_datadir;
|
2018-11-30 09:23:23 +02:00
|
|
|
#else
|
|
|
|
|
# include "univ.i"
|
2017-08-04 13:57:26 +03:00
|
|
|
#endif /* !UNIV_INNOCHECKSUM */
|
|
|
|
|
|
|
|
|
|
/** Initial size of a single-table tablespace in pages */
|
2018-04-27 14:26:43 +03:00
|
|
|
#define FIL_IBD_FILE_INITIAL_SIZE 4U
|
2017-08-04 13:57:26 +03:00
|
|
|
|
|
|
|
|
/** 'null' (undefined) page offset in the context of file spaces */
|
|
|
|
|
#define FIL_NULL ULINT32_UNDEFINED
|
|
|
|
|
|
2015-01-19 12:39:17 +02:00
|
|
|
|
2018-04-27 14:26:43 +03:00
|
|
|
#define FIL_ADDR_PAGE 0U /* first in address is the page offset */
|
|
|
|
|
#define FIL_ADDR_BYTE 4U /* then comes 2-byte byte offset within page*/
|
|
|
|
|
#define FIL_ADDR_SIZE 6U /* address size is 6 bytes */
|
2014-02-26 19:11:54 +01:00
|
|
|
|
|
|
|
|
/** File space address */
|
2016-08-12 11:17:45 +03:00
|
|
|
struct fil_addr_t {
|
2019-11-28 11:44:40 +02:00
|
|
|
/** page number within a tablespace */
|
|
|
|
|
uint32_t page;
|
|
|
|
|
/** byte offset within the page */
|
|
|
|
|
uint16_t boffset;
|
2014-02-26 19:11:54 +01:00
|
|
|
};
|
|
|
|
|
|
|
|
|
|
/** The byte offsets on a file page for various variables @{ */
|
|
|
|
|
#define FIL_PAGE_SPACE_OR_CHKSUM 0 /*!< in < MySQL-4.0.14 space id the
|
|
|
|
|
page belongs to (== 0) but in later
|
|
|
|
|
versions the 'new' checksum of the
|
|
|
|
|
page */
|
2018-04-27 14:26:43 +03:00
|
|
|
#define FIL_PAGE_OFFSET 4U /*!< page offset inside space */
|
|
|
|
|
#define FIL_PAGE_PREV 8U /*!< if there is a 'natural'
|
2014-02-26 19:11:54 +01:00
|
|
|
predecessor of the page, its
|
|
|
|
|
offset. Otherwise FIL_NULL.
|
|
|
|
|
This field is not set on BLOB
|
|
|
|
|
pages, which are stored as a
|
|
|
|
|
singly-linked list. See also
|
|
|
|
|
FIL_PAGE_NEXT. */
|
2018-04-27 14:26:43 +03:00
|
|
|
#define FIL_PAGE_NEXT 12U /*!< if there is a 'natural' successor
|
2014-02-26 19:11:54 +01:00
|
|
|
of the page, its offset.
|
|
|
|
|
Otherwise FIL_NULL.
|
|
|
|
|
B-tree index pages
|
|
|
|
|
(FIL_PAGE_TYPE contains FIL_PAGE_INDEX)
|
|
|
|
|
on the same PAGE_LEVEL are maintained
|
|
|
|
|
as a doubly linked list via
|
|
|
|
|
FIL_PAGE_PREV and FIL_PAGE_NEXT
|
|
|
|
|
in the collation order of the
|
|
|
|
|
smallest user record on each page. */
|
2018-04-27 14:26:43 +03:00
|
|
|
#define FIL_PAGE_LSN 16U /*!< lsn of the end of the newest
|
2014-02-26 19:11:54 +01:00
|
|
|
modification log record to the page */
|
2018-04-27 14:26:43 +03:00
|
|
|
#define FIL_PAGE_TYPE 24U /*!< file page type: FIL_PAGE_INDEX,...,
|
2014-02-26 19:11:54 +01:00
|
|
|
2 bytes.
|
|
|
|
|
|
|
|
|
|
The contents of this field can only
|
|
|
|
|
be trusted in the following case:
|
|
|
|
|
if the page is an uncompressed
|
|
|
|
|
B-tree index page, then it is
|
|
|
|
|
guaranteed that the value is
|
|
|
|
|
FIL_PAGE_INDEX.
|
|
|
|
|
The opposite does not hold.
|
|
|
|
|
|
|
|
|
|
In tablespaces created by
|
|
|
|
|
MySQL/InnoDB 5.1.7 or later, the
|
|
|
|
|
contents of this field is valid
|
|
|
|
|
for all uncompressed pages. */
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
|
|
|
|
|
/** For the first page in a system tablespace data file(ibdata*, not *.ibd):
|
|
|
|
|
the file has been flushed to disk at least up to this lsn
|
|
|
|
|
For other pages: 32-bit key version used to encrypt the page + 32-bit checksum
|
|
|
|
|
or 64 bites of zero if no encryption */
|
|
|
|
|
#define FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION 26U
|
2016-08-12 11:17:45 +03:00
|
|
|
|
|
|
|
|
/** This overloads FIL_PAGE_FILE_FLUSH_LSN for RTREE Split Sequence Number */
|
|
|
|
|
#define FIL_RTREE_SPLIT_SEQ_NUM FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION
|
|
|
|
|
|
MDEV-18644: Support full_crc32 for page_compressed
This is a follow-up task to MDEV-12026, which introduced
innodb_checksum_algorithm=full_crc32 and a simpler page format.
MDEV-12026 did not enable full_crc32 for page_compressed tables,
which we will be doing now.
This is joint work with Thirunarayanan Balathandayuthapani.
For innodb_checksum_algorithm=full_crc32 we change the
page_compressed format as follows:
FIL_PAGE_TYPE: The most significant bit will be set to indicate
page_compressed format. The least significant bits will contain
the compressed page size, rounded up to a multiple of 256 bytes.
The checksum will be stored in the last 4 bytes of the page
(whether it is the full page or a page_compressed page whose
size is determined by FIL_PAGE_TYPE), covering all preceding
bytes of the page. If encryption is used, then the page will
be encrypted between compression and computing the checksum.
For page_compressed, FIL_PAGE_LSN will not be repeated at
the end of the page.
FSP_SPACE_FLAGS (already implemented as part of MDEV-12026):
We will store the innodb_compression_algorithm that may be used
to compress pages. Previously, the choice of algorithm was written
to each compressed data page separately, and one would be unable
to know in advance which compression algorithm(s) are used.
fil_space_t::full_crc32_page_compressed_len(): Determine if the
page_compressed algorithm of the tablespace needs to know the
exact length of the compressed data. If yes, we will reserve and
write an extra byte for this right before the checksum.
buf_page_is_compressed(): Determine if a page uses page_compressed
(in any innodb_checksum_algorithm).
fil_page_decompress(): Pass also fil_space_t::flags so that the
format can be determined.
buf_page_is_zeroes(): Check if a page is full of zero bytes.
buf_page_full_crc32_is_corrupted(): Renamed from
buf_encrypted_full_crc32_page_is_corrupted(). For full_crc32,
we always simply validate the checksum to the page contents,
while the physical page size is explicitly specified by an
unencrypted part of the page header.
buf_page_full_crc32_size(): Determine the size of a full_crc32 page.
buf_dblwr_check_page_lsn(): Make this a debug-only function, because
it involves potentially costly lookups of fil_space_t.
create_table_info_t::check_table_options(),
ha_innobase::check_if_supported_inplace_alter(): Do allow the creation
of SPATIAL INDEX with full_crc32 also when page_compressed is used.
commit_cache_norebuild(): Preserve the compression algorithm when
updating the page_compression_level.
dict_tf_to_fsp_flags(): Set the flags for page compression algorithm.
FIXME: Maybe there should be a table option page_compression_algorithm
and a session variable to back it?
2019-03-18 14:08:43 +02:00
|
|
|
/** Start of the page_compressed content */
|
|
|
|
|
#define FIL_PAGE_COMP_ALGO FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION
|
|
|
|
|
|
2016-08-12 11:17:45 +03:00
|
|
|
/** starting from 4.1.x this contains the space id of the page */
|
2018-04-27 14:26:43 +03:00
|
|
|
#define FIL_PAGE_ARCH_LOG_NO_OR_SPACE_ID 34U
|
2016-08-12 11:17:45 +03:00
|
|
|
|
2014-05-05 18:20:28 +02:00
|
|
|
#define FIL_PAGE_SPACE_ID FIL_PAGE_ARCH_LOG_NO_OR_SPACE_ID
|
|
|
|
|
|
2016-08-12 11:17:45 +03:00
|
|
|
#define FIL_PAGE_DATA 38U /*!< start of the data on the page */
|
2016-09-06 09:43:16 +03:00
|
|
|
|
MDEV-18644: Support full_crc32 for page_compressed
This is a follow-up task to MDEV-12026, which introduced
innodb_checksum_algorithm=full_crc32 and a simpler page format.
MDEV-12026 did not enable full_crc32 for page_compressed tables,
which we will be doing now.
This is joint work with Thirunarayanan Balathandayuthapani.
For innodb_checksum_algorithm=full_crc32 we change the
page_compressed format as follows:
FIL_PAGE_TYPE: The most significant bit will be set to indicate
page_compressed format. The least significant bits will contain
the compressed page size, rounded up to a multiple of 256 bytes.
The checksum will be stored in the last 4 bytes of the page
(whether it is the full page or a page_compressed page whose
size is determined by FIL_PAGE_TYPE), covering all preceding
bytes of the page. If encryption is used, then the page will
be encrypted between compression and computing the checksum.
For page_compressed, FIL_PAGE_LSN will not be repeated at
the end of the page.
FSP_SPACE_FLAGS (already implemented as part of MDEV-12026):
We will store the innodb_compression_algorithm that may be used
to compress pages. Previously, the choice of algorithm was written
to each compressed data page separately, and one would be unable
to know in advance which compression algorithm(s) are used.
fil_space_t::full_crc32_page_compressed_len(): Determine if the
page_compressed algorithm of the tablespace needs to know the
exact length of the compressed data. If yes, we will reserve and
write an extra byte for this right before the checksum.
buf_page_is_compressed(): Determine if a page uses page_compressed
(in any innodb_checksum_algorithm).
fil_page_decompress(): Pass also fil_space_t::flags so that the
format can be determined.
buf_page_is_zeroes(): Check if a page is full of zero bytes.
buf_page_full_crc32_is_corrupted(): Renamed from
buf_encrypted_full_crc32_page_is_corrupted(). For full_crc32,
we always simply validate the checksum to the page contents,
while the physical page size is explicitly specified by an
unencrypted part of the page header.
buf_page_full_crc32_size(): Determine the size of a full_crc32 page.
buf_dblwr_check_page_lsn(): Make this a debug-only function, because
it involves potentially costly lookups of fil_space_t.
create_table_info_t::check_table_options(),
ha_innobase::check_if_supported_inplace_alter(): Do allow the creation
of SPATIAL INDEX with full_crc32 also when page_compressed is used.
commit_cache_norebuild(): Preserve the compression algorithm when
updating the page_compression_level.
dict_tf_to_fsp_flags(): Set the flags for page compression algorithm.
FIXME: Maybe there should be a table option page_compression_algorithm
and a session variable to back it?
2019-03-18 14:08:43 +02:00
|
|
|
/** 32-bit key version used to encrypt the page in full_crc32 format.
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
For non-encrypted page, it contains 0. */
|
|
|
|
|
#define FIL_PAGE_FCRC32_KEY_VERSION 0
|
|
|
|
|
|
MDEV-18644: Support full_crc32 for page_compressed
This is a follow-up task to MDEV-12026, which introduced
innodb_checksum_algorithm=full_crc32 and a simpler page format.
MDEV-12026 did not enable full_crc32 for page_compressed tables,
which we will be doing now.
This is joint work with Thirunarayanan Balathandayuthapani.
For innodb_checksum_algorithm=full_crc32 we change the
page_compressed format as follows:
FIL_PAGE_TYPE: The most significant bit will be set to indicate
page_compressed format. The least significant bits will contain
the compressed page size, rounded up to a multiple of 256 bytes.
The checksum will be stored in the last 4 bytes of the page
(whether it is the full page or a page_compressed page whose
size is determined by FIL_PAGE_TYPE), covering all preceding
bytes of the page. If encryption is used, then the page will
be encrypted between compression and computing the checksum.
For page_compressed, FIL_PAGE_LSN will not be repeated at
the end of the page.
FSP_SPACE_FLAGS (already implemented as part of MDEV-12026):
We will store the innodb_compression_algorithm that may be used
to compress pages. Previously, the choice of algorithm was written
to each compressed data page separately, and one would be unable
to know in advance which compression algorithm(s) are used.
fil_space_t::full_crc32_page_compressed_len(): Determine if the
page_compressed algorithm of the tablespace needs to know the
exact length of the compressed data. If yes, we will reserve and
write an extra byte for this right before the checksum.
buf_page_is_compressed(): Determine if a page uses page_compressed
(in any innodb_checksum_algorithm).
fil_page_decompress(): Pass also fil_space_t::flags so that the
format can be determined.
buf_page_is_zeroes(): Check if a page is full of zero bytes.
buf_page_full_crc32_is_corrupted(): Renamed from
buf_encrypted_full_crc32_page_is_corrupted(). For full_crc32,
we always simply validate the checksum to the page contents,
while the physical page size is explicitly specified by an
unencrypted part of the page header.
buf_page_full_crc32_size(): Determine the size of a full_crc32 page.
buf_dblwr_check_page_lsn(): Make this a debug-only function, because
it involves potentially costly lookups of fil_space_t.
create_table_info_t::check_table_options(),
ha_innobase::check_if_supported_inplace_alter(): Do allow the creation
of SPATIAL INDEX with full_crc32 also when page_compressed is used.
commit_cache_norebuild(): Preserve the compression algorithm when
updating the page_compression_level.
dict_tf_to_fsp_flags(): Set the flags for page compression algorithm.
FIXME: Maybe there should be a table option page_compression_algorithm
and a session variable to back it?
2019-03-18 14:08:43 +02:00
|
|
|
/** page_compressed without innodb_checksum_algorithm=full_crc32 @{ */
|
|
|
|
|
/** Number of bytes used to store actual payload data size on
|
|
|
|
|
page_compressed pages when not using full_crc32. */
|
|
|
|
|
#define FIL_PAGE_COMP_SIZE 0
|
|
|
|
|
|
|
|
|
|
/** Number of bytes for FIL_PAGE_COMP_SIZE */
|
|
|
|
|
#define FIL_PAGE_COMP_METADATA_LEN 2
|
|
|
|
|
|
|
|
|
|
/** Number of bytes used to store actual compression method
|
|
|
|
|
for encrypted tables when not using full_crc32. */
|
|
|
|
|
#define FIL_PAGE_ENCRYPT_COMP_ALGO 2
|
|
|
|
|
|
|
|
|
|
/** Extra header size for encrypted page_compressed pages when
|
|
|
|
|
not using full_crc32 */
|
|
|
|
|
#define FIL_PAGE_ENCRYPT_COMP_METADATA_LEN 4
|
2014-02-26 19:11:54 +01:00
|
|
|
/* @} */
|
MDEV-18644: Support full_crc32 for page_compressed
This is a follow-up task to MDEV-12026, which introduced
innodb_checksum_algorithm=full_crc32 and a simpler page format.
MDEV-12026 did not enable full_crc32 for page_compressed tables,
which we will be doing now.
This is joint work with Thirunarayanan Balathandayuthapani.
For innodb_checksum_algorithm=full_crc32 we change the
page_compressed format as follows:
FIL_PAGE_TYPE: The most significant bit will be set to indicate
page_compressed format. The least significant bits will contain
the compressed page size, rounded up to a multiple of 256 bytes.
The checksum will be stored in the last 4 bytes of the page
(whether it is the full page or a page_compressed page whose
size is determined by FIL_PAGE_TYPE), covering all preceding
bytes of the page. If encryption is used, then the page will
be encrypted between compression and computing the checksum.
For page_compressed, FIL_PAGE_LSN will not be repeated at
the end of the page.
FSP_SPACE_FLAGS (already implemented as part of MDEV-12026):
We will store the innodb_compression_algorithm that may be used
to compress pages. Previously, the choice of algorithm was written
to each compressed data page separately, and one would be unable
to know in advance which compression algorithm(s) are used.
fil_space_t::full_crc32_page_compressed_len(): Determine if the
page_compressed algorithm of the tablespace needs to know the
exact length of the compressed data. If yes, we will reserve and
write an extra byte for this right before the checksum.
buf_page_is_compressed(): Determine if a page uses page_compressed
(in any innodb_checksum_algorithm).
fil_page_decompress(): Pass also fil_space_t::flags so that the
format can be determined.
buf_page_is_zeroes(): Check if a page is full of zero bytes.
buf_page_full_crc32_is_corrupted(): Renamed from
buf_encrypted_full_crc32_page_is_corrupted(). For full_crc32,
we always simply validate the checksum to the page contents,
while the physical page size is explicitly specified by an
unencrypted part of the page header.
buf_page_full_crc32_size(): Determine the size of a full_crc32 page.
buf_dblwr_check_page_lsn(): Make this a debug-only function, because
it involves potentially costly lookups of fil_space_t.
create_table_info_t::check_table_options(),
ha_innobase::check_if_supported_inplace_alter(): Do allow the creation
of SPATIAL INDEX with full_crc32 also when page_compressed is used.
commit_cache_norebuild(): Preserve the compression algorithm when
updating the page_compression_level.
dict_tf_to_fsp_flags(): Set the flags for page compression algorithm.
FIXME: Maybe there should be a table option page_compression_algorithm
and a session variable to back it?
2019-03-18 14:08:43 +02:00
|
|
|
|
2014-02-26 19:11:54 +01:00
|
|
|
/** File page trailer @{ */
|
|
|
|
|
#define FIL_PAGE_END_LSN_OLD_CHKSUM 8 /*!< the low 4 bytes of this are used
|
|
|
|
|
to store the page checksum, the
|
|
|
|
|
last 4 bytes should be identical
|
|
|
|
|
to the last 4 bytes of FIL_PAGE_LSN */
|
|
|
|
|
#define FIL_PAGE_DATA_END 8 /*!< size of the page trailer */
|
MDEV-12026: Implement innodb_checksum_algorithm=full_crc32
MariaDB data-at-rest encryption (innodb_encrypt_tables)
had repurposed the same unused data field that was repurposed
in MySQL 5.7 (and MariaDB 10.2) for the Split Sequence Number (SSN)
field of SPATIAL INDEX. Because of this, MariaDB was unable to
support encryption on SPATIAL INDEX pages.
Furthermore, InnoDB page checksums skipped some bytes, and there
are multiple variations and checksum algorithms. By default,
InnoDB accepts all variations of all algorithms that ever existed.
This unnecessarily weakens the page checksums.
We hereby introduce two more innodb_checksum_algorithm variants
(full_crc32, strict_full_crc32) that are special in a way:
When either setting is active, newly created data files will
carry a flag (fil_space_t::full_crc32()) that indicates that
all pages of the file will use a full CRC-32C checksum over the
entire page contents (excluding the bytes where the checksum
is stored, at the very end of the page). Such files will always
use that checksum, no matter what the parameter
innodb_checksum_algorithm is assigned to.
For old files, the old checksum algorithms will continue to be
used. The value strict_full_crc32 will be equivalent to strict_crc32
and the value full_crc32 will be equivalent to crc32.
ROW_FORMAT=COMPRESSED tables will only use the old format.
These tables do not support new features, such as larger
innodb_page_size or instant ADD/DROP COLUMN. They may be
deprecated in the future. We do not want an unnecessary
file format change for them.
The new full_crc32() format also cleans up the MariaDB tablespace
flags. We will reserve flags to store the page_compressed
compression algorithm, and to store the compressed payload length,
so that checksum can be computed over the compressed (and
possibly encrypted) stream and can be validated without
decrypting or decompressing the page.
In the full_crc32 format, there no longer are separate before-encryption
and after-encryption checksums for pages. The single checksum is
computed on the page contents that is written to the file.
We do not make the new algorithm the default for two reasons.
First, MariaDB 10.4.2 was a beta release, and the default values
of parameters should not change after beta. Second, we did not
yet implement the full_crc32 format for page_compressed pages.
This will be fixed in MDEV-18644.
This is joint work with Marko Mäkelä.
2019-02-19 21:00:00 +02:00
|
|
|
|
|
|
|
|
/** Store the last 4 bytes of FIL_PAGE_LSN */
|
|
|
|
|
#define FIL_PAGE_FCRC32_END_LSN 8
|
|
|
|
|
|
|
|
|
|
/** Store crc32 checksum at the end of the page */
|
|
|
|
|
#define FIL_PAGE_FCRC32_CHECKSUM 4
|
2014-02-26 19:11:54 +01:00
|
|
|
/* @} */
|
|
|
|
|
|
|
|
|
|
/** File page types (values of FIL_PAGE_TYPE) @{ */
|
MDEV-18644: Support full_crc32 for page_compressed
This is a follow-up task to MDEV-12026, which introduced
innodb_checksum_algorithm=full_crc32 and a simpler page format.
MDEV-12026 did not enable full_crc32 for page_compressed tables,
which we will be doing now.
This is joint work with Thirunarayanan Balathandayuthapani.
For innodb_checksum_algorithm=full_crc32 we change the
page_compressed format as follows:
FIL_PAGE_TYPE: The most significant bit will be set to indicate
page_compressed format. The least significant bits will contain
the compressed page size, rounded up to a multiple of 256 bytes.
The checksum will be stored in the last 4 bytes of the page
(whether it is the full page or a page_compressed page whose
size is determined by FIL_PAGE_TYPE), covering all preceding
bytes of the page. If encryption is used, then the page will
be encrypted between compression and computing the checksum.
For page_compressed, FIL_PAGE_LSN will not be repeated at
the end of the page.
FSP_SPACE_FLAGS (already implemented as part of MDEV-12026):
We will store the innodb_compression_algorithm that may be used
to compress pages. Previously, the choice of algorithm was written
to each compressed data page separately, and one would be unable
to know in advance which compression algorithm(s) are used.
fil_space_t::full_crc32_page_compressed_len(): Determine if the
page_compressed algorithm of the tablespace needs to know the
exact length of the compressed data. If yes, we will reserve and
write an extra byte for this right before the checksum.
buf_page_is_compressed(): Determine if a page uses page_compressed
(in any innodb_checksum_algorithm).
fil_page_decompress(): Pass also fil_space_t::flags so that the
format can be determined.
buf_page_is_zeroes(): Check if a page is full of zero bytes.
buf_page_full_crc32_is_corrupted(): Renamed from
buf_encrypted_full_crc32_page_is_corrupted(). For full_crc32,
we always simply validate the checksum to the page contents,
while the physical page size is explicitly specified by an
unencrypted part of the page header.
buf_page_full_crc32_size(): Determine the size of a full_crc32 page.
buf_dblwr_check_page_lsn(): Make this a debug-only function, because
it involves potentially costly lookups of fil_space_t.
create_table_info_t::check_table_options(),
ha_innobase::check_if_supported_inplace_alter(): Do allow the creation
of SPATIAL INDEX with full_crc32 also when page_compressed is used.
commit_cache_norebuild(): Preserve the compression algorithm when
updating the page_compression_level.
dict_tf_to_fsp_flags(): Set the flags for page compression algorithm.
FIXME: Maybe there should be a table option page_compression_algorithm
and a session variable to back it?
2019-03-18 14:08:43 +02:00
|
|
|
/** page_compressed, encrypted=YES (not used for full_crc32) */
|
2019-11-08 06:28:29 +02:00
|
|
|
constexpr uint16_t FIL_PAGE_PAGE_COMPRESSED_ENCRYPTED= 37401;
|
MDEV-18644: Support full_crc32 for page_compressed
This is a follow-up task to MDEV-12026, which introduced
innodb_checksum_algorithm=full_crc32 and a simpler page format.
MDEV-12026 did not enable full_crc32 for page_compressed tables,
which we will be doing now.
This is joint work with Thirunarayanan Balathandayuthapani.
For innodb_checksum_algorithm=full_crc32 we change the
page_compressed format as follows:
FIL_PAGE_TYPE: The most significant bit will be set to indicate
page_compressed format. The least significant bits will contain
the compressed page size, rounded up to a multiple of 256 bytes.
The checksum will be stored in the last 4 bytes of the page
(whether it is the full page or a page_compressed page whose
size is determined by FIL_PAGE_TYPE), covering all preceding
bytes of the page. If encryption is used, then the page will
be encrypted between compression and computing the checksum.
For page_compressed, FIL_PAGE_LSN will not be repeated at
the end of the page.
FSP_SPACE_FLAGS (already implemented as part of MDEV-12026):
We will store the innodb_compression_algorithm that may be used
to compress pages. Previously, the choice of algorithm was written
to each compressed data page separately, and one would be unable
to know in advance which compression algorithm(s) are used.
fil_space_t::full_crc32_page_compressed_len(): Determine if the
page_compressed algorithm of the tablespace needs to know the
exact length of the compressed data. If yes, we will reserve and
write an extra byte for this right before the checksum.
buf_page_is_compressed(): Determine if a page uses page_compressed
(in any innodb_checksum_algorithm).
fil_page_decompress(): Pass also fil_space_t::flags so that the
format can be determined.
buf_page_is_zeroes(): Check if a page is full of zero bytes.
buf_page_full_crc32_is_corrupted(): Renamed from
buf_encrypted_full_crc32_page_is_corrupted(). For full_crc32,
we always simply validate the checksum to the page contents,
while the physical page size is explicitly specified by an
unencrypted part of the page header.
buf_page_full_crc32_size(): Determine the size of a full_crc32 page.
buf_dblwr_check_page_lsn(): Make this a debug-only function, because
it involves potentially costly lookups of fil_space_t.
create_table_info_t::check_table_options(),
ha_innobase::check_if_supported_inplace_alter(): Do allow the creation
of SPATIAL INDEX with full_crc32 also when page_compressed is used.
commit_cache_norebuild(): Preserve the compression algorithm when
updating the page_compression_level.
dict_tf_to_fsp_flags(): Set the flags for page compression algorithm.
FIXME: Maybe there should be a table option page_compression_algorithm
and a session variable to back it?
2019-03-18 14:08:43 +02:00
|
|
|
/** page_compressed (not used for full_crc32) */
|
2019-11-08 06:28:29 +02:00
|
|
|
constexpr uint16_t FIL_PAGE_PAGE_COMPRESSED= 34354;
|
|
|
|
|
/** B-tree index page */
|
|
|
|
|
constexpr uint16_t FIL_PAGE_INDEX= 17855;
|
|
|
|
|
/** R-tree index page (SPATIAL INDEX) */
|
|
|
|
|
constexpr uint16_t FIL_PAGE_RTREE= 17854;
|
|
|
|
|
/** Undo log page */
|
|
|
|
|
constexpr uint16_t FIL_PAGE_UNDO_LOG= 2;
|
|
|
|
|
/** Index node (of file-in-file metadata) */
|
|
|
|
|
constexpr uint16_t FIL_PAGE_INODE= 3;
|
|
|
|
|
/** Insert buffer free list */
|
|
|
|
|
constexpr uint16_t FIL_PAGE_IBUF_FREE_LIST= 4;
|
|
|
|
|
/** Freshly allocated page */
|
|
|
|
|
constexpr uint16_t FIL_PAGE_TYPE_ALLOCATED= 0;
|
|
|
|
|
/** Change buffer bitmap (pages n*innodb_page_size+1) */
|
|
|
|
|
constexpr uint16_t FIL_PAGE_IBUF_BITMAP= 5;
|
|
|
|
|
/** System page */
|
|
|
|
|
constexpr uint16_t FIL_PAGE_TYPE_SYS= 6;
|
|
|
|
|
/** Transaction system data */
|
|
|
|
|
constexpr uint16_t FIL_PAGE_TYPE_TRX_SYS= 7;
|
|
|
|
|
/** Tablespace header (page 0) */
|
|
|
|
|
constexpr uint16_t FIL_PAGE_TYPE_FSP_HDR= 8;
|
|
|
|
|
/** Extent descriptor page (pages n*innodb_page_size, except 0) */
|
|
|
|
|
constexpr uint16_t FIL_PAGE_TYPE_XDES= 9;
|
|
|
|
|
/** Uncompressed BLOB page */
|
|
|
|
|
constexpr uint16_t FIL_PAGE_TYPE_BLOB= 10;
|
|
|
|
|
/** First ROW_FORMAT=COMPRESSED BLOB page */
|
|
|
|
|
constexpr uint16_t FIL_PAGE_TYPE_ZBLOB= 11;
|
|
|
|
|
/** Subsequent ROW_FORMAT=COMPRESSED BLOB page */
|
|
|
|
|
constexpr uint16_t FIL_PAGE_TYPE_ZBLOB2= 12;
|
|
|
|
|
/** In old tablespaces, garbage in FIL_PAGE_TYPE is replaced with this
|
|
|
|
|
value when flushing pages. */
|
|
|
|
|
constexpr uint16_t FIL_PAGE_TYPE_UNKNOWN= 13;
|
2017-01-16 14:03:36 +02:00
|
|
|
|
|
|
|
|
/* File page types introduced in MySQL 5.7, not supported in MariaDB */
|
2019-11-08 06:28:29 +02:00
|
|
|
//constexpr uint16_t FIL_PAGE_COMPRESSED = 14;
|
|
|
|
|
//constexpr uint16_t FIL_PAGE_ENCRYPTED = 15;
|
|
|
|
|
//constexpr uint16_t FIL_PAGE_COMPRESSED_AND_ENCRYPTED = 16;
|
|
|
|
|
//constexpr FIL_PAGE_ENCRYPTED_RTREE = 17;
|
MDEV-11369 Instant ADD COLUMN for InnoDB
For InnoDB tables, adding, dropping and reordering columns has
required a rebuild of the table and all its indexes. Since MySQL 5.6
(and MariaDB 10.0) this has been supported online (LOCK=NONE), allowing
concurrent modification of the tables.
This work revises the InnoDB ROW_FORMAT=REDUNDANT, ROW_FORMAT=COMPACT
and ROW_FORMAT=DYNAMIC so that columns can be appended instantaneously,
with only minor changes performed to the table structure. The counter
innodb_instant_alter_column in INFORMATION_SCHEMA.GLOBAL_STATUS
is incremented whenever a table rebuild operation is converted into
an instant ADD COLUMN operation.
ROW_FORMAT=COMPRESSED tables will not support instant ADD COLUMN.
Some usability limitations will be addressed in subsequent work:
MDEV-13134 Introduce ALTER TABLE attributes ALGORITHM=NOCOPY
and ALGORITHM=INSTANT
MDEV-14016 Allow instant ADD COLUMN, ADD INDEX, LOCK=NONE
The format of the clustered index (PRIMARY KEY) is changed as follows:
(1) The FIL_PAGE_TYPE of the root page will be FIL_PAGE_TYPE_INSTANT,
and a new field PAGE_INSTANT will contain the original number of fields
in the clustered index ('core' fields).
If instant ADD COLUMN has not been used or the table becomes empty,
or the very first instant ADD COLUMN operation is rolled back,
the fields PAGE_INSTANT and FIL_PAGE_TYPE will be reset
to 0 and FIL_PAGE_INDEX.
(2) A special 'default row' record is inserted into the leftmost leaf,
between the page infimum and the first user record. This record is
distinguished by the REC_INFO_MIN_REC_FLAG, and it is otherwise in the
same format as records that contain values for the instantly added
columns. This 'default row' always has the same number of fields as
the clustered index according to the table definition. The values of
'core' fields are to be ignored. For other fields, the 'default row'
will contain the default values as they were during the ALTER TABLE
statement. (If the column default values are changed later, those
values will only be stored in the .frm file. The 'default row' will
contain the original evaluated values, which must be the same for
every row.) The 'default row' must be completely hidden from
higher-level access routines. Assertions have been added to ensure
that no 'default row' is ever present in the adaptive hash index
or in locked records. The 'default row' is never delete-marked.
(3) In clustered index leaf page records, the number of fields must
reside between the number of 'core' fields (dict_index_t::n_core_fields
introduced in this work) and dict_index_t::n_fields. If the number
of fields is less than dict_index_t::n_fields, the missing fields
are replaced with the column value of the 'default row'.
Note: The number of fields in the record may shrink if some of the
last instantly added columns are updated to the value that is
in the 'default row'. The function btr_cur_trim() implements this
'compression' on update and rollback; dtuple::trim() implements it
on insert.
(4) In ROW_FORMAT=COMPACT and ROW_FORMAT=DYNAMIC records, the new
status value REC_STATUS_COLUMNS_ADDED will indicate the presence of
a new record header that will encode n_fields-n_core_fields-1 in
1 or 2 bytes. (In ROW_FORMAT=REDUNDANT records, the record header
always explicitly encodes the number of fields.)
We introduce the undo log record type TRX_UNDO_INSERT_DEFAULT for
covering the insert of the 'default row' record when instant ADD COLUMN
is used for the first time. Subsequent instant ADD COLUMN can use
TRX_UNDO_UPD_EXIST_REC.
This is joint work with Vin Chen (陈福荣) from Tencent. The design
that was discussed in April 2017 would not have allowed import or
export of data files, because instead of the 'default row' it would
have introduced a data dictionary table. The test
rpl.rpl_alter_instant is exactly as contributed in pull request #408.
The test innodb.instant_alter is based on a contributed test.
The redo log record format changes for ROW_FORMAT=DYNAMIC and
ROW_FORMAT=COMPACT are as contributed. (With this change present,
crash recovery from MariaDB 10.3.1 will fail in spectacular ways!)
Also the semantics of higher-level redo log records that modify the
PAGE_INSTANT field is changed. The redo log format version identifier
was already changed to LOG_HEADER_FORMAT_CURRENT=103 in MariaDB 10.3.1.
Everything else has been rewritten by me. Thanks to Elena Stepanova,
the code has been tested extensively.
When rolling back an instant ADD COLUMN operation, we must empty the
PAGE_FREE list after deleting or shortening the 'default row' record,
by calling either btr_page_empty() or btr_page_reorganize(). We must
know the size of each entry in the PAGE_FREE list. If rollback left a
freed copy of the 'default row' in the PAGE_FREE list, we would be
unable to determine its size (if it is in ROW_FORMAT=COMPACT or
ROW_FORMAT=DYNAMIC) because it would contain more fields than the
rolled-back definition of the clustered index.
UNIV_SQL_DEFAULT: A new special constant that designates an instantly
added column that is not present in the clustered index record.
len_is_stored(): Check if a length is an actual length. There are
two magic length values: UNIV_SQL_DEFAULT, UNIV_SQL_NULL.
dict_col_t::def_val: The 'default row' value of the column. If the
column is not added instantly, def_val.len will be UNIV_SQL_DEFAULT.
dict_col_t: Add the accessors is_virtual(), is_nullable(), is_instant(),
instant_value().
dict_col_t::remove_instant(): Remove the 'instant ADD' status of
a column.
dict_col_t::name(const dict_table_t& table): Replaces
dict_table_get_col_name().
dict_index_t::n_core_fields: The original number of fields.
For secondary indexes and if instant ADD COLUMN has not been used,
this will be equal to dict_index_t::n_fields.
dict_index_t::n_core_null_bytes: Number of bytes needed to
represent the null flags; usually equal to UT_BITS_IN_BYTES(n_nullable).
dict_index_t::NO_CORE_NULL_BYTES: Magic value signalling that
n_core_null_bytes was not initialized yet from the clustered index
root page.
dict_index_t: Add the accessors is_instant(), is_clust(),
get_n_nullable(), instant_field_value().
dict_index_t::instant_add_field(): Adjust clustered index metadata
for instant ADD COLUMN.
dict_index_t::remove_instant(): Remove the 'instant ADD' status
of a clustered index when the table becomes empty, or the very first
instant ADD COLUMN operation is rolled back.
dict_table_t: Add the accessors is_instant(), is_temporary(),
supports_instant().
dict_table_t::instant_add_column(): Adjust metadata for
instant ADD COLUMN.
dict_table_t::rollback_instant(): Adjust metadata on the rollback
of instant ADD COLUMN.
prepare_inplace_alter_table_dict(): First create the ctx->new_table,
and only then decide if the table really needs to be rebuilt.
We must split the creation of table or index metadata from the
creation of the dictionary table records and the creation of
the data. In this way, we can transform a table-rebuilding operation
into an instant ADD COLUMN operation. Dictionary objects will only
be added to cache when table rebuilding or index creation is needed.
The ctx->instant_table will never be added to cache.
dict_table_t::add_to_cache(): Modified and renamed from
dict_table_add_to_cache(). Do not modify the table metadata.
Let the callers invoke dict_table_add_system_columns() and if needed,
set can_be_evicted.
dict_create_sys_tables_tuple(), dict_create_table_step(): Omit the
system columns (which will now exist in the dict_table_t object
already at this point).
dict_create_table_step(): Expect the callers to invoke
dict_table_add_system_columns().
pars_create_table(): Before creating the table creation execution
graph, invoke dict_table_add_system_columns().
row_create_table_for_mysql(): Expect all callers to invoke
dict_table_add_system_columns().
create_index_dict(): Replaces row_merge_create_index_graph().
innodb_update_n_cols(): Renamed from innobase_update_n_virtual().
Call my_error() if an error occurs.
btr_cur_instant_init(), btr_cur_instant_init_low(),
btr_cur_instant_root_init():
Load additional metadata from the clustered index and set
dict_index_t::n_core_null_bytes. This is invoked
when table metadata is first loaded into the data dictionary.
dict_boot(): Initialize n_core_null_bytes for the four hard-coded
dictionary tables.
dict_create_index_step(): Initialize n_core_null_bytes. This is
executed as part of CREATE TABLE.
dict_index_build_internal_clust(): Initialize n_core_null_bytes to
NO_CORE_NULL_BYTES if table->supports_instant().
row_create_index_for_mysql(): Initialize n_core_null_bytes for
CREATE TEMPORARY TABLE.
commit_cache_norebuild(): Call the code to rename or enlarge columns
in the cache only if instant ADD COLUMN is not being used.
(Instant ADD COLUMN would copy all column metadata from
instant_table to old_table, including the names and lengths.)
PAGE_INSTANT: A new 13-bit field for storing dict_index_t::n_core_fields.
This is repurposing the 16-bit field PAGE_DIRECTION, of which only the
least significant 3 bits were used. The original byte containing
PAGE_DIRECTION will be accessible via the new constant PAGE_DIRECTION_B.
page_get_instant(), page_set_instant(): Accessors for the PAGE_INSTANT.
page_ptr_get_direction(), page_get_direction(),
page_ptr_set_direction(): Accessors for PAGE_DIRECTION.
page_direction_reset(): Reset PAGE_DIRECTION, PAGE_N_DIRECTION.
page_direction_increment(): Increment PAGE_N_DIRECTION
and set PAGE_DIRECTION.
rec_get_offsets(): Use the 'leaf' parameter for non-debug purposes,
and assume that heap_no is always set.
Initialize all dict_index_t::n_fields for ROW_FORMAT=REDUNDANT records,
even if the record contains fewer fields.
rec_offs_make_valid(): Add the parameter 'leaf'.
rec_copy_prefix_to_dtuple(): Assert that the tuple is only built
on the core fields. Instant ADD COLUMN only applies to the
clustered index, and we should never build a search key that has
more than the PRIMARY KEY and possibly DB_TRX_ID,DB_ROLL_PTR.
All these columns are always present.
dict_index_build_data_tuple(): Remove assertions that would be
duplicated in rec_copy_prefix_to_dtuple().
rec_init_offsets(): Support ROW_FORMAT=REDUNDANT records whose
number of fields is between n_core_fields and n_fields.
cmp_rec_rec_with_match(): Implement the comparison between two
MIN_REC_FLAG records.
trx_t::in_rollback: Make the field available in non-debug builds.
trx_start_for_ddl_low(): Remove dangerous error-tolerance.
A dictionary transaction must be flagged as such before it has generated
any undo log records. This is because trx_undo_assign_undo() will mark
the transaction as a dictionary transaction in the undo log header
right before the very first undo log record is being written.
btr_index_rec_validate(): Account for instant ADD COLUMN
row_undo_ins_remove_clust_rec(): On the rollback of an insert into
SYS_COLUMNS, revert instant ADD COLUMN in the cache by removing the
last column from the table and the clustered index.
row_search_on_row_ref(), row_undo_mod_parse_undo_rec(), row_undo_mod(),
trx_undo_update_rec_get_update(): Handle the 'default row'
as a special case.
dtuple_t::trim(index): Omit a redundant suffix of an index tuple right
before insert or update. After instant ADD COLUMN, if the last fields
of a clustered index tuple match the 'default row', there is no
need to store them. While trimming the entry, we must hold a page latch,
so that the table cannot be emptied and the 'default row' be deleted.
btr_cur_optimistic_update(), btr_cur_pessimistic_update(),
row_upd_clust_rec_by_insert(), row_ins_clust_index_entry_low():
Invoke dtuple_t::trim() if needed.
row_ins_clust_index_entry(): Restore dtuple_t::n_fields after calling
row_ins_clust_index_entry_low().
rec_get_converted_size(), rec_get_converted_size_comp(): Allow the number
of fields to be between n_core_fields and n_fields. Do not support
infimum,supremum. They are never supposed to be stored in dtuple_t,
because page creation nowadays uses a lower-level method for initializing
them.
rec_convert_dtuple_to_rec_comp(): Assign the status bits based on the
number of fields.
btr_cur_trim(): In an update, trim the index entry as needed. For the
'default row', handle rollback specially. For user records, omit
fields that match the 'default row'.
btr_cur_optimistic_delete_func(), btr_cur_pessimistic_delete():
Skip locking and adaptive hash index for the 'default row'.
row_log_table_apply_convert_mrec(): Replace 'default row' values if needed.
In the temporary file that is applied by row_log_table_apply(),
we must identify whether the records contain the extra header for
instantly added columns. For now, we will allocate an additional byte
for this for ROW_T_INSERT and ROW_T_UPDATE records when the source table
has been subject to instant ADD COLUMN. The ROW_T_DELETE records are
fine, as they will be converted and will only contain 'core' columns
(PRIMARY KEY and some system columns) that are converted from dtuple_t.
rec_get_converted_size_temp(), rec_init_offsets_temp(),
rec_convert_dtuple_to_temp(): Add the parameter 'status'.
REC_INFO_DEFAULT_ROW = REC_INFO_MIN_REC_FLAG | REC_STATUS_COLUMNS_ADDED:
An info_bits constant for distinguishing the 'default row' record.
rec_comp_status_t: An enum of the status bit values.
rec_leaf_format: An enum that replaces the bool parameter of
rec_init_offsets_comp_ordinary().
2017-10-06 07:00:05 +03:00
|
|
|
/** Clustered index root page after instant ADD COLUMN */
|
2019-11-08 06:28:29 +02:00
|
|
|
constexpr uint16_t FIL_PAGE_TYPE_INSTANT= 18;
|
2016-08-12 11:17:45 +03:00
|
|
|
|
MDEV-11369 Instant ADD COLUMN for InnoDB
For InnoDB tables, adding, dropping and reordering columns has
required a rebuild of the table and all its indexes. Since MySQL 5.6
(and MariaDB 10.0) this has been supported online (LOCK=NONE), allowing
concurrent modification of the tables.
This work revises the InnoDB ROW_FORMAT=REDUNDANT, ROW_FORMAT=COMPACT
and ROW_FORMAT=DYNAMIC so that columns can be appended instantaneously,
with only minor changes performed to the table structure. The counter
innodb_instant_alter_column in INFORMATION_SCHEMA.GLOBAL_STATUS
is incremented whenever a table rebuild operation is converted into
an instant ADD COLUMN operation.
ROW_FORMAT=COMPRESSED tables will not support instant ADD COLUMN.
Some usability limitations will be addressed in subsequent work:
MDEV-13134 Introduce ALTER TABLE attributes ALGORITHM=NOCOPY
and ALGORITHM=INSTANT
MDEV-14016 Allow instant ADD COLUMN, ADD INDEX, LOCK=NONE
The format of the clustered index (PRIMARY KEY) is changed as follows:
(1) The FIL_PAGE_TYPE of the root page will be FIL_PAGE_TYPE_INSTANT,
and a new field PAGE_INSTANT will contain the original number of fields
in the clustered index ('core' fields).
If instant ADD COLUMN has not been used or the table becomes empty,
or the very first instant ADD COLUMN operation is rolled back,
the fields PAGE_INSTANT and FIL_PAGE_TYPE will be reset
to 0 and FIL_PAGE_INDEX.
(2) A special 'default row' record is inserted into the leftmost leaf,
between the page infimum and the first user record. This record is
distinguished by the REC_INFO_MIN_REC_FLAG, and it is otherwise in the
same format as records that contain values for the instantly added
columns. This 'default row' always has the same number of fields as
the clustered index according to the table definition. The values of
'core' fields are to be ignored. For other fields, the 'default row'
will contain the default values as they were during the ALTER TABLE
statement. (If the column default values are changed later, those
values will only be stored in the .frm file. The 'default row' will
contain the original evaluated values, which must be the same for
every row.) The 'default row' must be completely hidden from
higher-level access routines. Assertions have been added to ensure
that no 'default row' is ever present in the adaptive hash index
or in locked records. The 'default row' is never delete-marked.
(3) In clustered index leaf page records, the number of fields must
reside between the number of 'core' fields (dict_index_t::n_core_fields
introduced in this work) and dict_index_t::n_fields. If the number
of fields is less than dict_index_t::n_fields, the missing fields
are replaced with the column value of the 'default row'.
Note: The number of fields in the record may shrink if some of the
last instantly added columns are updated to the value that is
in the 'default row'. The function btr_cur_trim() implements this
'compression' on update and rollback; dtuple::trim() implements it
on insert.
(4) In ROW_FORMAT=COMPACT and ROW_FORMAT=DYNAMIC records, the new
status value REC_STATUS_COLUMNS_ADDED will indicate the presence of
a new record header that will encode n_fields-n_core_fields-1 in
1 or 2 bytes. (In ROW_FORMAT=REDUNDANT records, the record header
always explicitly encodes the number of fields.)
We introduce the undo log record type TRX_UNDO_INSERT_DEFAULT for
covering the insert of the 'default row' record when instant ADD COLUMN
is used for the first time. Subsequent instant ADD COLUMN can use
TRX_UNDO_UPD_EXIST_REC.
This is joint work with Vin Chen (陈福荣) from Tencent. The design
that was discussed in April 2017 would not have allowed import or
export of data files, because instead of the 'default row' it would
have introduced a data dictionary table. The test
rpl.rpl_alter_instant is exactly as contributed in pull request #408.
The test innodb.instant_alter is based on a contributed test.
The redo log record format changes for ROW_FORMAT=DYNAMIC and
ROW_FORMAT=COMPACT are as contributed. (With this change present,
crash recovery from MariaDB 10.3.1 will fail in spectacular ways!)
Also the semantics of higher-level redo log records that modify the
PAGE_INSTANT field is changed. The redo log format version identifier
was already changed to LOG_HEADER_FORMAT_CURRENT=103 in MariaDB 10.3.1.
Everything else has been rewritten by me. Thanks to Elena Stepanova,
the code has been tested extensively.
When rolling back an instant ADD COLUMN operation, we must empty the
PAGE_FREE list after deleting or shortening the 'default row' record,
by calling either btr_page_empty() or btr_page_reorganize(). We must
know the size of each entry in the PAGE_FREE list. If rollback left a
freed copy of the 'default row' in the PAGE_FREE list, we would be
unable to determine its size (if it is in ROW_FORMAT=COMPACT or
ROW_FORMAT=DYNAMIC) because it would contain more fields than the
rolled-back definition of the clustered index.
UNIV_SQL_DEFAULT: A new special constant that designates an instantly
added column that is not present in the clustered index record.
len_is_stored(): Check if a length is an actual length. There are
two magic length values: UNIV_SQL_DEFAULT, UNIV_SQL_NULL.
dict_col_t::def_val: The 'default row' value of the column. If the
column is not added instantly, def_val.len will be UNIV_SQL_DEFAULT.
dict_col_t: Add the accessors is_virtual(), is_nullable(), is_instant(),
instant_value().
dict_col_t::remove_instant(): Remove the 'instant ADD' status of
a column.
dict_col_t::name(const dict_table_t& table): Replaces
dict_table_get_col_name().
dict_index_t::n_core_fields: The original number of fields.
For secondary indexes and if instant ADD COLUMN has not been used,
this will be equal to dict_index_t::n_fields.
dict_index_t::n_core_null_bytes: Number of bytes needed to
represent the null flags; usually equal to UT_BITS_IN_BYTES(n_nullable).
dict_index_t::NO_CORE_NULL_BYTES: Magic value signalling that
n_core_null_bytes was not initialized yet from the clustered index
root page.
dict_index_t: Add the accessors is_instant(), is_clust(),
get_n_nullable(), instant_field_value().
dict_index_t::instant_add_field(): Adjust clustered index metadata
for instant ADD COLUMN.
dict_index_t::remove_instant(): Remove the 'instant ADD' status
of a clustered index when the table becomes empty, or the very first
instant ADD COLUMN operation is rolled back.
dict_table_t: Add the accessors is_instant(), is_temporary(),
supports_instant().
dict_table_t::instant_add_column(): Adjust metadata for
instant ADD COLUMN.
dict_table_t::rollback_instant(): Adjust metadata on the rollback
of instant ADD COLUMN.
prepare_inplace_alter_table_dict(): First create the ctx->new_table,
and only then decide if the table really needs to be rebuilt.
We must split the creation of table or index metadata from the
creation of the dictionary table records and the creation of
the data. In this way, we can transform a table-rebuilding operation
into an instant ADD COLUMN operation. Dictionary objects will only
be added to cache when table rebuilding or index creation is needed.
The ctx->instant_table will never be added to cache.
dict_table_t::add_to_cache(): Modified and renamed from
dict_table_add_to_cache(). Do not modify the table metadata.
Let the callers invoke dict_table_add_system_columns() and if needed,
set can_be_evicted.
dict_create_sys_tables_tuple(), dict_create_table_step(): Omit the
system columns (which will now exist in the dict_table_t object
already at this point).
dict_create_table_step(): Expect the callers to invoke
dict_table_add_system_columns().
pars_create_table(): Before creating the table creation execution
graph, invoke dict_table_add_system_columns().
row_create_table_for_mysql(): Expect all callers to invoke
dict_table_add_system_columns().
create_index_dict(): Replaces row_merge_create_index_graph().
innodb_update_n_cols(): Renamed from innobase_update_n_virtual().
Call my_error() if an error occurs.
btr_cur_instant_init(), btr_cur_instant_init_low(),
btr_cur_instant_root_init():
Load additional metadata from the clustered index and set
dict_index_t::n_core_null_bytes. This is invoked
when table metadata is first loaded into the data dictionary.
dict_boot(): Initialize n_core_null_bytes for the four hard-coded
dictionary tables.
dict_create_index_step(): Initialize n_core_null_bytes. This is
executed as part of CREATE TABLE.
dict_index_build_internal_clust(): Initialize n_core_null_bytes to
NO_CORE_NULL_BYTES if table->supports_instant().
row_create_index_for_mysql(): Initialize n_core_null_bytes for
CREATE TEMPORARY TABLE.
commit_cache_norebuild(): Call the code to rename or enlarge columns
in the cache only if instant ADD COLUMN is not being used.
(Instant ADD COLUMN would copy all column metadata from
instant_table to old_table, including the names and lengths.)
PAGE_INSTANT: A new 13-bit field for storing dict_index_t::n_core_fields.
This is repurposing the 16-bit field PAGE_DIRECTION, of which only the
least significant 3 bits were used. The original byte containing
PAGE_DIRECTION will be accessible via the new constant PAGE_DIRECTION_B.
page_get_instant(), page_set_instant(): Accessors for the PAGE_INSTANT.
page_ptr_get_direction(), page_get_direction(),
page_ptr_set_direction(): Accessors for PAGE_DIRECTION.
page_direction_reset(): Reset PAGE_DIRECTION, PAGE_N_DIRECTION.
page_direction_increment(): Increment PAGE_N_DIRECTION
and set PAGE_DIRECTION.
rec_get_offsets(): Use the 'leaf' parameter for non-debug purposes,
and assume that heap_no is always set.
Initialize all dict_index_t::n_fields for ROW_FORMAT=REDUNDANT records,
even if the record contains fewer fields.
rec_offs_make_valid(): Add the parameter 'leaf'.
rec_copy_prefix_to_dtuple(): Assert that the tuple is only built
on the core fields. Instant ADD COLUMN only applies to the
clustered index, and we should never build a search key that has
more than the PRIMARY KEY and possibly DB_TRX_ID,DB_ROLL_PTR.
All these columns are always present.
dict_index_build_data_tuple(): Remove assertions that would be
duplicated in rec_copy_prefix_to_dtuple().
rec_init_offsets(): Support ROW_FORMAT=REDUNDANT records whose
number of fields is between n_core_fields and n_fields.
cmp_rec_rec_with_match(): Implement the comparison between two
MIN_REC_FLAG records.
trx_t::in_rollback: Make the field available in non-debug builds.
trx_start_for_ddl_low(): Remove dangerous error-tolerance.
A dictionary transaction must be flagged as such before it has generated
any undo log records. This is because trx_undo_assign_undo() will mark
the transaction as a dictionary transaction in the undo log header
right before the very first undo log record is being written.
btr_index_rec_validate(): Account for instant ADD COLUMN
row_undo_ins_remove_clust_rec(): On the rollback of an insert into
SYS_COLUMNS, revert instant ADD COLUMN in the cache by removing the
last column from the table and the clustered index.
row_search_on_row_ref(), row_undo_mod_parse_undo_rec(), row_undo_mod(),
trx_undo_update_rec_get_update(): Handle the 'default row'
as a special case.
dtuple_t::trim(index): Omit a redundant suffix of an index tuple right
before insert or update. After instant ADD COLUMN, if the last fields
of a clustered index tuple match the 'default row', there is no
need to store them. While trimming the entry, we must hold a page latch,
so that the table cannot be emptied and the 'default row' be deleted.
btr_cur_optimistic_update(), btr_cur_pessimistic_update(),
row_upd_clust_rec_by_insert(), row_ins_clust_index_entry_low():
Invoke dtuple_t::trim() if needed.
row_ins_clust_index_entry(): Restore dtuple_t::n_fields after calling
row_ins_clust_index_entry_low().
rec_get_converted_size(), rec_get_converted_size_comp(): Allow the number
of fields to be between n_core_fields and n_fields. Do not support
infimum,supremum. They are never supposed to be stored in dtuple_t,
because page creation nowadays uses a lower-level method for initializing
them.
rec_convert_dtuple_to_rec_comp(): Assign the status bits based on the
number of fields.
btr_cur_trim(): In an update, trim the index entry as needed. For the
'default row', handle rollback specially. For user records, omit
fields that match the 'default row'.
btr_cur_optimistic_delete_func(), btr_cur_pessimistic_delete():
Skip locking and adaptive hash index for the 'default row'.
row_log_table_apply_convert_mrec(): Replace 'default row' values if needed.
In the temporary file that is applied by row_log_table_apply(),
we must identify whether the records contain the extra header for
instantly added columns. For now, we will allocate an additional byte
for this for ROW_T_INSERT and ROW_T_UPDATE records when the source table
has been subject to instant ADD COLUMN. The ROW_T_DELETE records are
fine, as they will be converted and will only contain 'core' columns
(PRIMARY KEY and some system columns) that are converted from dtuple_t.
rec_get_converted_size_temp(), rec_init_offsets_temp(),
rec_convert_dtuple_to_temp(): Add the parameter 'status'.
REC_INFO_DEFAULT_ROW = REC_INFO_MIN_REC_FLAG | REC_STATUS_COLUMNS_ADDED:
An info_bits constant for distinguishing the 'default row' record.
rec_comp_status_t: An enum of the status bit values.
rec_leaf_format: An enum that replaces the bool parameter of
rec_init_offsets_comp_ordinary().
2017-10-06 07:00:05 +03:00
|
|
|
/** Used by i_s.cc to index into the text description.
|
|
|
|
|
Note: FIL_PAGE_TYPE_INSTANT maps to the same as FIL_PAGE_INDEX. */
|
2019-11-08 06:28:29 +02:00
|
|
|
constexpr uint16_t FIL_PAGE_TYPE_LAST= FIL_PAGE_TYPE_UNKNOWN;
|
|
|
|
|
|
|
|
|
|
/** Set in FIL_PAGE_TYPE for full_crc32 pages in page_compressed format.
|
MDEV-18644: Support full_crc32 for page_compressed
This is a follow-up task to MDEV-12026, which introduced
innodb_checksum_algorithm=full_crc32 and a simpler page format.
MDEV-12026 did not enable full_crc32 for page_compressed tables,
which we will be doing now.
This is joint work with Thirunarayanan Balathandayuthapani.
For innodb_checksum_algorithm=full_crc32 we change the
page_compressed format as follows:
FIL_PAGE_TYPE: The most significant bit will be set to indicate
page_compressed format. The least significant bits will contain
the compressed page size, rounded up to a multiple of 256 bytes.
The checksum will be stored in the last 4 bytes of the page
(whether it is the full page or a page_compressed page whose
size is determined by FIL_PAGE_TYPE), covering all preceding
bytes of the page. If encryption is used, then the page will
be encrypted between compression and computing the checksum.
For page_compressed, FIL_PAGE_LSN will not be repeated at
the end of the page.
FSP_SPACE_FLAGS (already implemented as part of MDEV-12026):
We will store the innodb_compression_algorithm that may be used
to compress pages. Previously, the choice of algorithm was written
to each compressed data page separately, and one would be unable
to know in advance which compression algorithm(s) are used.
fil_space_t::full_crc32_page_compressed_len(): Determine if the
page_compressed algorithm of the tablespace needs to know the
exact length of the compressed data. If yes, we will reserve and
write an extra byte for this right before the checksum.
buf_page_is_compressed(): Determine if a page uses page_compressed
(in any innodb_checksum_algorithm).
fil_page_decompress(): Pass also fil_space_t::flags so that the
format can be determined.
buf_page_is_zeroes(): Check if a page is full of zero bytes.
buf_page_full_crc32_is_corrupted(): Renamed from
buf_encrypted_full_crc32_page_is_corrupted(). For full_crc32,
we always simply validate the checksum to the page contents,
while the physical page size is explicitly specified by an
unencrypted part of the page header.
buf_page_full_crc32_size(): Determine the size of a full_crc32 page.
buf_dblwr_check_page_lsn(): Make this a debug-only function, because
it involves potentially costly lookups of fil_space_t.
create_table_info_t::check_table_options(),
ha_innobase::check_if_supported_inplace_alter(): Do allow the creation
of SPATIAL INDEX with full_crc32 also when page_compressed is used.
commit_cache_norebuild(): Preserve the compression algorithm when
updating the page_compression_level.
dict_tf_to_fsp_flags(): Set the flags for page compression algorithm.
FIXME: Maybe there should be a table option page_compression_algorithm
and a session variable to back it?
2019-03-18 14:08:43 +02:00
|
|
|
If the flag is set, then the following holds for the remaining bits
|
|
|
|
|
of FIL_PAGE_TYPE:
|
|
|
|
|
Bits 0..7 will contain the compressed page size in bytes.
|
|
|
|
|
Bits 8..14 are reserved and must be 0. */
|
2019-11-08 06:28:29 +02:00
|
|
|
constexpr uint16_t FIL_PAGE_COMPRESS_FCRC32_MARKER= 15;
|
2014-02-26 19:11:54 +01:00
|
|
|
/* @} */
|
|
|
|
|
|
MDEV-11369 Instant ADD COLUMN for InnoDB
For InnoDB tables, adding, dropping and reordering columns has
required a rebuild of the table and all its indexes. Since MySQL 5.6
(and MariaDB 10.0) this has been supported online (LOCK=NONE), allowing
concurrent modification of the tables.
This work revises the InnoDB ROW_FORMAT=REDUNDANT, ROW_FORMAT=COMPACT
and ROW_FORMAT=DYNAMIC so that columns can be appended instantaneously,
with only minor changes performed to the table structure. The counter
innodb_instant_alter_column in INFORMATION_SCHEMA.GLOBAL_STATUS
is incremented whenever a table rebuild operation is converted into
an instant ADD COLUMN operation.
ROW_FORMAT=COMPRESSED tables will not support instant ADD COLUMN.
Some usability limitations will be addressed in subsequent work:
MDEV-13134 Introduce ALTER TABLE attributes ALGORITHM=NOCOPY
and ALGORITHM=INSTANT
MDEV-14016 Allow instant ADD COLUMN, ADD INDEX, LOCK=NONE
The format of the clustered index (PRIMARY KEY) is changed as follows:
(1) The FIL_PAGE_TYPE of the root page will be FIL_PAGE_TYPE_INSTANT,
and a new field PAGE_INSTANT will contain the original number of fields
in the clustered index ('core' fields).
If instant ADD COLUMN has not been used or the table becomes empty,
or the very first instant ADD COLUMN operation is rolled back,
the fields PAGE_INSTANT and FIL_PAGE_TYPE will be reset
to 0 and FIL_PAGE_INDEX.
(2) A special 'default row' record is inserted into the leftmost leaf,
between the page infimum and the first user record. This record is
distinguished by the REC_INFO_MIN_REC_FLAG, and it is otherwise in the
same format as records that contain values for the instantly added
columns. This 'default row' always has the same number of fields as
the clustered index according to the table definition. The values of
'core' fields are to be ignored. For other fields, the 'default row'
will contain the default values as they were during the ALTER TABLE
statement. (If the column default values are changed later, those
values will only be stored in the .frm file. The 'default row' will
contain the original evaluated values, which must be the same for
every row.) The 'default row' must be completely hidden from
higher-level access routines. Assertions have been added to ensure
that no 'default row' is ever present in the adaptive hash index
or in locked records. The 'default row' is never delete-marked.
(3) In clustered index leaf page records, the number of fields must
reside between the number of 'core' fields (dict_index_t::n_core_fields
introduced in this work) and dict_index_t::n_fields. If the number
of fields is less than dict_index_t::n_fields, the missing fields
are replaced with the column value of the 'default row'.
Note: The number of fields in the record may shrink if some of the
last instantly added columns are updated to the value that is
in the 'default row'. The function btr_cur_trim() implements this
'compression' on update and rollback; dtuple::trim() implements it
on insert.
(4) In ROW_FORMAT=COMPACT and ROW_FORMAT=DYNAMIC records, the new
status value REC_STATUS_COLUMNS_ADDED will indicate the presence of
a new record header that will encode n_fields-n_core_fields-1 in
1 or 2 bytes. (In ROW_FORMAT=REDUNDANT records, the record header
always explicitly encodes the number of fields.)
We introduce the undo log record type TRX_UNDO_INSERT_DEFAULT for
covering the insert of the 'default row' record when instant ADD COLUMN
is used for the first time. Subsequent instant ADD COLUMN can use
TRX_UNDO_UPD_EXIST_REC.
This is joint work with Vin Chen (陈福荣) from Tencent. The design
that was discussed in April 2017 would not have allowed import or
export of data files, because instead of the 'default row' it would
have introduced a data dictionary table. The test
rpl.rpl_alter_instant is exactly as contributed in pull request #408.
The test innodb.instant_alter is based on a contributed test.
The redo log record format changes for ROW_FORMAT=DYNAMIC and
ROW_FORMAT=COMPACT are as contributed. (With this change present,
crash recovery from MariaDB 10.3.1 will fail in spectacular ways!)
Also the semantics of higher-level redo log records that modify the
PAGE_INSTANT field is changed. The redo log format version identifier
was already changed to LOG_HEADER_FORMAT_CURRENT=103 in MariaDB 10.3.1.
Everything else has been rewritten by me. Thanks to Elena Stepanova,
the code has been tested extensively.
When rolling back an instant ADD COLUMN operation, we must empty the
PAGE_FREE list after deleting or shortening the 'default row' record,
by calling either btr_page_empty() or btr_page_reorganize(). We must
know the size of each entry in the PAGE_FREE list. If rollback left a
freed copy of the 'default row' in the PAGE_FREE list, we would be
unable to determine its size (if it is in ROW_FORMAT=COMPACT or
ROW_FORMAT=DYNAMIC) because it would contain more fields than the
rolled-back definition of the clustered index.
UNIV_SQL_DEFAULT: A new special constant that designates an instantly
added column that is not present in the clustered index record.
len_is_stored(): Check if a length is an actual length. There are
two magic length values: UNIV_SQL_DEFAULT, UNIV_SQL_NULL.
dict_col_t::def_val: The 'default row' value of the column. If the
column is not added instantly, def_val.len will be UNIV_SQL_DEFAULT.
dict_col_t: Add the accessors is_virtual(), is_nullable(), is_instant(),
instant_value().
dict_col_t::remove_instant(): Remove the 'instant ADD' status of
a column.
dict_col_t::name(const dict_table_t& table): Replaces
dict_table_get_col_name().
dict_index_t::n_core_fields: The original number of fields.
For secondary indexes and if instant ADD COLUMN has not been used,
this will be equal to dict_index_t::n_fields.
dict_index_t::n_core_null_bytes: Number of bytes needed to
represent the null flags; usually equal to UT_BITS_IN_BYTES(n_nullable).
dict_index_t::NO_CORE_NULL_BYTES: Magic value signalling that
n_core_null_bytes was not initialized yet from the clustered index
root page.
dict_index_t: Add the accessors is_instant(), is_clust(),
get_n_nullable(), instant_field_value().
dict_index_t::instant_add_field(): Adjust clustered index metadata
for instant ADD COLUMN.
dict_index_t::remove_instant(): Remove the 'instant ADD' status
of a clustered index when the table becomes empty, or the very first
instant ADD COLUMN operation is rolled back.
dict_table_t: Add the accessors is_instant(), is_temporary(),
supports_instant().
dict_table_t::instant_add_column(): Adjust metadata for
instant ADD COLUMN.
dict_table_t::rollback_instant(): Adjust metadata on the rollback
of instant ADD COLUMN.
prepare_inplace_alter_table_dict(): First create the ctx->new_table,
and only then decide if the table really needs to be rebuilt.
We must split the creation of table or index metadata from the
creation of the dictionary table records and the creation of
the data. In this way, we can transform a table-rebuilding operation
into an instant ADD COLUMN operation. Dictionary objects will only
be added to cache when table rebuilding or index creation is needed.
The ctx->instant_table will never be added to cache.
dict_table_t::add_to_cache(): Modified and renamed from
dict_table_add_to_cache(). Do not modify the table metadata.
Let the callers invoke dict_table_add_system_columns() and if needed,
set can_be_evicted.
dict_create_sys_tables_tuple(), dict_create_table_step(): Omit the
system columns (which will now exist in the dict_table_t object
already at this point).
dict_create_table_step(): Expect the callers to invoke
dict_table_add_system_columns().
pars_create_table(): Before creating the table creation execution
graph, invoke dict_table_add_system_columns().
row_create_table_for_mysql(): Expect all callers to invoke
dict_table_add_system_columns().
create_index_dict(): Replaces row_merge_create_index_graph().
innodb_update_n_cols(): Renamed from innobase_update_n_virtual().
Call my_error() if an error occurs.
btr_cur_instant_init(), btr_cur_instant_init_low(),
btr_cur_instant_root_init():
Load additional metadata from the clustered index and set
dict_index_t::n_core_null_bytes. This is invoked
when table metadata is first loaded into the data dictionary.
dict_boot(): Initialize n_core_null_bytes for the four hard-coded
dictionary tables.
dict_create_index_step(): Initialize n_core_null_bytes. This is
executed as part of CREATE TABLE.
dict_index_build_internal_clust(): Initialize n_core_null_bytes to
NO_CORE_NULL_BYTES if table->supports_instant().
row_create_index_for_mysql(): Initialize n_core_null_bytes for
CREATE TEMPORARY TABLE.
commit_cache_norebuild(): Call the code to rename or enlarge columns
in the cache only if instant ADD COLUMN is not being used.
(Instant ADD COLUMN would copy all column metadata from
instant_table to old_table, including the names and lengths.)
PAGE_INSTANT: A new 13-bit field for storing dict_index_t::n_core_fields.
This is repurposing the 16-bit field PAGE_DIRECTION, of which only the
least significant 3 bits were used. The original byte containing
PAGE_DIRECTION will be accessible via the new constant PAGE_DIRECTION_B.
page_get_instant(), page_set_instant(): Accessors for the PAGE_INSTANT.
page_ptr_get_direction(), page_get_direction(),
page_ptr_set_direction(): Accessors for PAGE_DIRECTION.
page_direction_reset(): Reset PAGE_DIRECTION, PAGE_N_DIRECTION.
page_direction_increment(): Increment PAGE_N_DIRECTION
and set PAGE_DIRECTION.
rec_get_offsets(): Use the 'leaf' parameter for non-debug purposes,
and assume that heap_no is always set.
Initialize all dict_index_t::n_fields for ROW_FORMAT=REDUNDANT records,
even if the record contains fewer fields.
rec_offs_make_valid(): Add the parameter 'leaf'.
rec_copy_prefix_to_dtuple(): Assert that the tuple is only built
on the core fields. Instant ADD COLUMN only applies to the
clustered index, and we should never build a search key that has
more than the PRIMARY KEY and possibly DB_TRX_ID,DB_ROLL_PTR.
All these columns are always present.
dict_index_build_data_tuple(): Remove assertions that would be
duplicated in rec_copy_prefix_to_dtuple().
rec_init_offsets(): Support ROW_FORMAT=REDUNDANT records whose
number of fields is between n_core_fields and n_fields.
cmp_rec_rec_with_match(): Implement the comparison between two
MIN_REC_FLAG records.
trx_t::in_rollback: Make the field available in non-debug builds.
trx_start_for_ddl_low(): Remove dangerous error-tolerance.
A dictionary transaction must be flagged as such before it has generated
any undo log records. This is because trx_undo_assign_undo() will mark
the transaction as a dictionary transaction in the undo log header
right before the very first undo log record is being written.
btr_index_rec_validate(): Account for instant ADD COLUMN
row_undo_ins_remove_clust_rec(): On the rollback of an insert into
SYS_COLUMNS, revert instant ADD COLUMN in the cache by removing the
last column from the table and the clustered index.
row_search_on_row_ref(), row_undo_mod_parse_undo_rec(), row_undo_mod(),
trx_undo_update_rec_get_update(): Handle the 'default row'
as a special case.
dtuple_t::trim(index): Omit a redundant suffix of an index tuple right
before insert or update. After instant ADD COLUMN, if the last fields
of a clustered index tuple match the 'default row', there is no
need to store them. While trimming the entry, we must hold a page latch,
so that the table cannot be emptied and the 'default row' be deleted.
btr_cur_optimistic_update(), btr_cur_pessimistic_update(),
row_upd_clust_rec_by_insert(), row_ins_clust_index_entry_low():
Invoke dtuple_t::trim() if needed.
row_ins_clust_index_entry(): Restore dtuple_t::n_fields after calling
row_ins_clust_index_entry_low().
rec_get_converted_size(), rec_get_converted_size_comp(): Allow the number
of fields to be between n_core_fields and n_fields. Do not support
infimum,supremum. They are never supposed to be stored in dtuple_t,
because page creation nowadays uses a lower-level method for initializing
them.
rec_convert_dtuple_to_rec_comp(): Assign the status bits based on the
number of fields.
btr_cur_trim(): In an update, trim the index entry as needed. For the
'default row', handle rollback specially. For user records, omit
fields that match the 'default row'.
btr_cur_optimistic_delete_func(), btr_cur_pessimistic_delete():
Skip locking and adaptive hash index for the 'default row'.
row_log_table_apply_convert_mrec(): Replace 'default row' values if needed.
In the temporary file that is applied by row_log_table_apply(),
we must identify whether the records contain the extra header for
instantly added columns. For now, we will allocate an additional byte
for this for ROW_T_INSERT and ROW_T_UPDATE records when the source table
has been subject to instant ADD COLUMN. The ROW_T_DELETE records are
fine, as they will be converted and will only contain 'core' columns
(PRIMARY KEY and some system columns) that are converted from dtuple_t.
rec_get_converted_size_temp(), rec_init_offsets_temp(),
rec_convert_dtuple_to_temp(): Add the parameter 'status'.
REC_INFO_DEFAULT_ROW = REC_INFO_MIN_REC_FLAG | REC_STATUS_COLUMNS_ADDED:
An info_bits constant for distinguishing the 'default row' record.
rec_comp_status_t: An enum of the status bit values.
rec_leaf_format: An enum that replaces the bool parameter of
rec_init_offsets_comp_ordinary().
2017-10-06 07:00:05 +03:00
|
|
|
/** @return whether the page type is B-tree or R-tree index */
|
2019-11-08 06:28:29 +02:00
|
|
|
inline bool fil_page_type_is_index(uint16_t page_type)
|
MDEV-11369 Instant ADD COLUMN for InnoDB
For InnoDB tables, adding, dropping and reordering columns has
required a rebuild of the table and all its indexes. Since MySQL 5.6
(and MariaDB 10.0) this has been supported online (LOCK=NONE), allowing
concurrent modification of the tables.
This work revises the InnoDB ROW_FORMAT=REDUNDANT, ROW_FORMAT=COMPACT
and ROW_FORMAT=DYNAMIC so that columns can be appended instantaneously,
with only minor changes performed to the table structure. The counter
innodb_instant_alter_column in INFORMATION_SCHEMA.GLOBAL_STATUS
is incremented whenever a table rebuild operation is converted into
an instant ADD COLUMN operation.
ROW_FORMAT=COMPRESSED tables will not support instant ADD COLUMN.
Some usability limitations will be addressed in subsequent work:
MDEV-13134 Introduce ALTER TABLE attributes ALGORITHM=NOCOPY
and ALGORITHM=INSTANT
MDEV-14016 Allow instant ADD COLUMN, ADD INDEX, LOCK=NONE
The format of the clustered index (PRIMARY KEY) is changed as follows:
(1) The FIL_PAGE_TYPE of the root page will be FIL_PAGE_TYPE_INSTANT,
and a new field PAGE_INSTANT will contain the original number of fields
in the clustered index ('core' fields).
If instant ADD COLUMN has not been used or the table becomes empty,
or the very first instant ADD COLUMN operation is rolled back,
the fields PAGE_INSTANT and FIL_PAGE_TYPE will be reset
to 0 and FIL_PAGE_INDEX.
(2) A special 'default row' record is inserted into the leftmost leaf,
between the page infimum and the first user record. This record is
distinguished by the REC_INFO_MIN_REC_FLAG, and it is otherwise in the
same format as records that contain values for the instantly added
columns. This 'default row' always has the same number of fields as
the clustered index according to the table definition. The values of
'core' fields are to be ignored. For other fields, the 'default row'
will contain the default values as they were during the ALTER TABLE
statement. (If the column default values are changed later, those
values will only be stored in the .frm file. The 'default row' will
contain the original evaluated values, which must be the same for
every row.) The 'default row' must be completely hidden from
higher-level access routines. Assertions have been added to ensure
that no 'default row' is ever present in the adaptive hash index
or in locked records. The 'default row' is never delete-marked.
(3) In clustered index leaf page records, the number of fields must
reside between the number of 'core' fields (dict_index_t::n_core_fields
introduced in this work) and dict_index_t::n_fields. If the number
of fields is less than dict_index_t::n_fields, the missing fields
are replaced with the column value of the 'default row'.
Note: The number of fields in the record may shrink if some of the
last instantly added columns are updated to the value that is
in the 'default row'. The function btr_cur_trim() implements this
'compression' on update and rollback; dtuple::trim() implements it
on insert.
(4) In ROW_FORMAT=COMPACT and ROW_FORMAT=DYNAMIC records, the new
status value REC_STATUS_COLUMNS_ADDED will indicate the presence of
a new record header that will encode n_fields-n_core_fields-1 in
1 or 2 bytes. (In ROW_FORMAT=REDUNDANT records, the record header
always explicitly encodes the number of fields.)
We introduce the undo log record type TRX_UNDO_INSERT_DEFAULT for
covering the insert of the 'default row' record when instant ADD COLUMN
is used for the first time. Subsequent instant ADD COLUMN can use
TRX_UNDO_UPD_EXIST_REC.
This is joint work with Vin Chen (陈福荣) from Tencent. The design
that was discussed in April 2017 would not have allowed import or
export of data files, because instead of the 'default row' it would
have introduced a data dictionary table. The test
rpl.rpl_alter_instant is exactly as contributed in pull request #408.
The test innodb.instant_alter is based on a contributed test.
The redo log record format changes for ROW_FORMAT=DYNAMIC and
ROW_FORMAT=COMPACT are as contributed. (With this change present,
crash recovery from MariaDB 10.3.1 will fail in spectacular ways!)
Also the semantics of higher-level redo log records that modify the
PAGE_INSTANT field is changed. The redo log format version identifier
was already changed to LOG_HEADER_FORMAT_CURRENT=103 in MariaDB 10.3.1.
Everything else has been rewritten by me. Thanks to Elena Stepanova,
the code has been tested extensively.
When rolling back an instant ADD COLUMN operation, we must empty the
PAGE_FREE list after deleting or shortening the 'default row' record,
by calling either btr_page_empty() or btr_page_reorganize(). We must
know the size of each entry in the PAGE_FREE list. If rollback left a
freed copy of the 'default row' in the PAGE_FREE list, we would be
unable to determine its size (if it is in ROW_FORMAT=COMPACT or
ROW_FORMAT=DYNAMIC) because it would contain more fields than the
rolled-back definition of the clustered index.
UNIV_SQL_DEFAULT: A new special constant that designates an instantly
added column that is not present in the clustered index record.
len_is_stored(): Check if a length is an actual length. There are
two magic length values: UNIV_SQL_DEFAULT, UNIV_SQL_NULL.
dict_col_t::def_val: The 'default row' value of the column. If the
column is not added instantly, def_val.len will be UNIV_SQL_DEFAULT.
dict_col_t: Add the accessors is_virtual(), is_nullable(), is_instant(),
instant_value().
dict_col_t::remove_instant(): Remove the 'instant ADD' status of
a column.
dict_col_t::name(const dict_table_t& table): Replaces
dict_table_get_col_name().
dict_index_t::n_core_fields: The original number of fields.
For secondary indexes and if instant ADD COLUMN has not been used,
this will be equal to dict_index_t::n_fields.
dict_index_t::n_core_null_bytes: Number of bytes needed to
represent the null flags; usually equal to UT_BITS_IN_BYTES(n_nullable).
dict_index_t::NO_CORE_NULL_BYTES: Magic value signalling that
n_core_null_bytes was not initialized yet from the clustered index
root page.
dict_index_t: Add the accessors is_instant(), is_clust(),
get_n_nullable(), instant_field_value().
dict_index_t::instant_add_field(): Adjust clustered index metadata
for instant ADD COLUMN.
dict_index_t::remove_instant(): Remove the 'instant ADD' status
of a clustered index when the table becomes empty, or the very first
instant ADD COLUMN operation is rolled back.
dict_table_t: Add the accessors is_instant(), is_temporary(),
supports_instant().
dict_table_t::instant_add_column(): Adjust metadata for
instant ADD COLUMN.
dict_table_t::rollback_instant(): Adjust metadata on the rollback
of instant ADD COLUMN.
prepare_inplace_alter_table_dict(): First create the ctx->new_table,
and only then decide if the table really needs to be rebuilt.
We must split the creation of table or index metadata from the
creation of the dictionary table records and the creation of
the data. In this way, we can transform a table-rebuilding operation
into an instant ADD COLUMN operation. Dictionary objects will only
be added to cache when table rebuilding or index creation is needed.
The ctx->instant_table will never be added to cache.
dict_table_t::add_to_cache(): Modified and renamed from
dict_table_add_to_cache(). Do not modify the table metadata.
Let the callers invoke dict_table_add_system_columns() and if needed,
set can_be_evicted.
dict_create_sys_tables_tuple(), dict_create_table_step(): Omit the
system columns (which will now exist in the dict_table_t object
already at this point).
dict_create_table_step(): Expect the callers to invoke
dict_table_add_system_columns().
pars_create_table(): Before creating the table creation execution
graph, invoke dict_table_add_system_columns().
row_create_table_for_mysql(): Expect all callers to invoke
dict_table_add_system_columns().
create_index_dict(): Replaces row_merge_create_index_graph().
innodb_update_n_cols(): Renamed from innobase_update_n_virtual().
Call my_error() if an error occurs.
btr_cur_instant_init(), btr_cur_instant_init_low(),
btr_cur_instant_root_init():
Load additional metadata from the clustered index and set
dict_index_t::n_core_null_bytes. This is invoked
when table metadata is first loaded into the data dictionary.
dict_boot(): Initialize n_core_null_bytes for the four hard-coded
dictionary tables.
dict_create_index_step(): Initialize n_core_null_bytes. This is
executed as part of CREATE TABLE.
dict_index_build_internal_clust(): Initialize n_core_null_bytes to
NO_CORE_NULL_BYTES if table->supports_instant().
row_create_index_for_mysql(): Initialize n_core_null_bytes for
CREATE TEMPORARY TABLE.
commit_cache_norebuild(): Call the code to rename or enlarge columns
in the cache only if instant ADD COLUMN is not being used.
(Instant ADD COLUMN would copy all column metadata from
instant_table to old_table, including the names and lengths.)
PAGE_INSTANT: A new 13-bit field for storing dict_index_t::n_core_fields.
This is repurposing the 16-bit field PAGE_DIRECTION, of which only the
least significant 3 bits were used. The original byte containing
PAGE_DIRECTION will be accessible via the new constant PAGE_DIRECTION_B.
page_get_instant(), page_set_instant(): Accessors for the PAGE_INSTANT.
page_ptr_get_direction(), page_get_direction(),
page_ptr_set_direction(): Accessors for PAGE_DIRECTION.
page_direction_reset(): Reset PAGE_DIRECTION, PAGE_N_DIRECTION.
page_direction_increment(): Increment PAGE_N_DIRECTION
and set PAGE_DIRECTION.
rec_get_offsets(): Use the 'leaf' parameter for non-debug purposes,
and assume that heap_no is always set.
Initialize all dict_index_t::n_fields for ROW_FORMAT=REDUNDANT records,
even if the record contains fewer fields.
rec_offs_make_valid(): Add the parameter 'leaf'.
rec_copy_prefix_to_dtuple(): Assert that the tuple is only built
on the core fields. Instant ADD COLUMN only applies to the
clustered index, and we should never build a search key that has
more than the PRIMARY KEY and possibly DB_TRX_ID,DB_ROLL_PTR.
All these columns are always present.
dict_index_build_data_tuple(): Remove assertions that would be
duplicated in rec_copy_prefix_to_dtuple().
rec_init_offsets(): Support ROW_FORMAT=REDUNDANT records whose
number of fields is between n_core_fields and n_fields.
cmp_rec_rec_with_match(): Implement the comparison between two
MIN_REC_FLAG records.
trx_t::in_rollback: Make the field available in non-debug builds.
trx_start_for_ddl_low(): Remove dangerous error-tolerance.
A dictionary transaction must be flagged as such before it has generated
any undo log records. This is because trx_undo_assign_undo() will mark
the transaction as a dictionary transaction in the undo log header
right before the very first undo log record is being written.
btr_index_rec_validate(): Account for instant ADD COLUMN
row_undo_ins_remove_clust_rec(): On the rollback of an insert into
SYS_COLUMNS, revert instant ADD COLUMN in the cache by removing the
last column from the table and the clustered index.
row_search_on_row_ref(), row_undo_mod_parse_undo_rec(), row_undo_mod(),
trx_undo_update_rec_get_update(): Handle the 'default row'
as a special case.
dtuple_t::trim(index): Omit a redundant suffix of an index tuple right
before insert or update. After instant ADD COLUMN, if the last fields
of a clustered index tuple match the 'default row', there is no
need to store them. While trimming the entry, we must hold a page latch,
so that the table cannot be emptied and the 'default row' be deleted.
btr_cur_optimistic_update(), btr_cur_pessimistic_update(),
row_upd_clust_rec_by_insert(), row_ins_clust_index_entry_low():
Invoke dtuple_t::trim() if needed.
row_ins_clust_index_entry(): Restore dtuple_t::n_fields after calling
row_ins_clust_index_entry_low().
rec_get_converted_size(), rec_get_converted_size_comp(): Allow the number
of fields to be between n_core_fields and n_fields. Do not support
infimum,supremum. They are never supposed to be stored in dtuple_t,
because page creation nowadays uses a lower-level method for initializing
them.
rec_convert_dtuple_to_rec_comp(): Assign the status bits based on the
number of fields.
btr_cur_trim(): In an update, trim the index entry as needed. For the
'default row', handle rollback specially. For user records, omit
fields that match the 'default row'.
btr_cur_optimistic_delete_func(), btr_cur_pessimistic_delete():
Skip locking and adaptive hash index for the 'default row'.
row_log_table_apply_convert_mrec(): Replace 'default row' values if needed.
In the temporary file that is applied by row_log_table_apply(),
we must identify whether the records contain the extra header for
instantly added columns. For now, we will allocate an additional byte
for this for ROW_T_INSERT and ROW_T_UPDATE records when the source table
has been subject to instant ADD COLUMN. The ROW_T_DELETE records are
fine, as they will be converted and will only contain 'core' columns
(PRIMARY KEY and some system columns) that are converted from dtuple_t.
rec_get_converted_size_temp(), rec_init_offsets_temp(),
rec_convert_dtuple_to_temp(): Add the parameter 'status'.
REC_INFO_DEFAULT_ROW = REC_INFO_MIN_REC_FLAG | REC_STATUS_COLUMNS_ADDED:
An info_bits constant for distinguishing the 'default row' record.
rec_comp_status_t: An enum of the status bit values.
rec_leaf_format: An enum that replaces the bool parameter of
rec_init_offsets_comp_ordinary().
2017-10-06 07:00:05 +03:00
|
|
|
{
|
|
|
|
|
switch (page_type) {
|
|
|
|
|
case FIL_PAGE_TYPE_INSTANT:
|
|
|
|
|
case FIL_PAGE_INDEX:
|
|
|
|
|
case FIL_PAGE_RTREE:
|
|
|
|
|
return(true);
|
|
|
|
|
}
|
|
|
|
|
return(false);
|
|
|
|
|
}
|
2014-02-26 19:11:54 +01:00
|
|
|
|
2016-08-12 11:17:45 +03:00
|
|
|
/** Check whether the page is index page (either regular Btree index or Rtree
|
|
|
|
|
index */
|
|
|
|
|
#define fil_page_index_page_check(page) \
|
|
|
|
|
fil_page_type_is_index(fil_page_get_type(page))
|
2015-01-19 12:39:17 +02:00
|
|
|
|
2020-05-07 17:15:34 +03:00
|
|
|
/** Get the file page type.
|
|
|
|
|
@param[in] page file page
|
|
|
|
|
@return page type */
|
|
|
|
|
inline uint16_t fil_page_get_type(const byte *page)
|
|
|
|
|
{
|
|
|
|
|
return mach_read_from_2(my_assume_aligned<2>(page + FIL_PAGE_TYPE));
|
|
|
|
|
}
|
|
|
|
|
|
2018-11-29 12:53:44 +02:00
|
|
|
#ifndef UNIV_INNOCHECKSUM
|
|
|
|
|
|
2014-02-26 19:11:54 +01:00
|
|
|
/** Number of pending tablespace flushes */
|
2020-10-26 16:04:12 +02:00
|
|
|
extern Atomic_counter<ulint> fil_n_pending_tablespace_flushes;
|
2014-02-26 19:11:54 +01:00
|
|
|
|
2016-09-06 09:43:16 +03:00
|
|
|
/** Look up a tablespace.
|
|
|
|
|
The caller should hold an InnoDB table lock or a MDL that prevents
|
|
|
|
|
the tablespace from being dropped during the operation,
|
|
|
|
|
or the caller should be in single-threaded crash recovery mode
|
|
|
|
|
(no user connections that could drop tablespaces).
|
2020-10-26 16:04:12 +02:00
|
|
|
Normally, fil_space_t::get() should be used instead.
|
2016-09-06 09:43:16 +03:00
|
|
|
@param[in] id tablespace ID
|
|
|
|
|
@return tablespace, or NULL if not found */
|
|
|
|
|
fil_space_t*
|
|
|
|
|
fil_space_get(
|
|
|
|
|
ulint id)
|
|
|
|
|
MY_ATTRIBUTE((warn_unused_result));
|
|
|
|
|
|
2014-10-25 08:21:52 +03:00
|
|
|
/** The tablespace memory cache; also the totality of logs (the log
|
2020-01-12 02:05:28 +07:00
|
|
|
data space) is stored here; below we talk about tablespaces */
|
2014-10-25 08:21:52 +03:00
|
|
|
struct fil_system_t {
|
2018-03-28 09:00:06 +03:00
|
|
|
/**
|
|
|
|
|
Constructor.
|
|
|
|
|
|
|
|
|
|
Some members may require late initialisation, thus we just mark object as
|
|
|
|
|
uninitialised. Real initialisation happens in create().
|
|
|
|
|
*/
|
2021-03-20 01:31:01 +03:00
|
|
|
fil_system_t() : m_initialised(false) {}
|
2018-03-28 09:00:06 +03:00
|
|
|
|
|
|
|
|
bool is_initialised() const { return m_initialised; }
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
|
Create the file system interface at database start.
|
|
|
|
|
|
|
|
|
|
@param[in] hash_size hash table size
|
|
|
|
|
*/
|
|
|
|
|
void create(ulint hash_size);
|
|
|
|
|
|
|
|
|
|
/** Close the file system interface at shutdown */
|
|
|
|
|
void close();
|
|
|
|
|
|
|
|
|
|
private:
|
|
|
|
|
bool m_initialised;
|
MDEV-17380 innodb_flush_neighbors=ON should be ignored on SSD
For tablespaces that do not reside on spinning storage, it does
not make sense to attempt to write nearby pages when writing out
dirty pages from the InnoDB buffer pool. It is actually detrimental
to performance and to the life span of flash ROM storage.
With this change, MariaDB will detect whether an InnoDB file resides
on solid-state storage. The detection has been implemented for Linux
and Microsoft Windows. For other systems, we will err on the safe side
and assume that files reside on SSD.
As part of this change, we will reduce the number of fstat() calls
when opening data files on POSIX systems and slightly clean up some
file I/O code.
FIXME: os_is_sparse_file_supported() on POSIX works in a destructive
manner. Thus, we can only invoke it when creating files, not when
opening them.
For diagnostics, we introduce the column ON_SSD to the table
INFORMATION_SCHEMA.INNODB_TABLESPACES_SCRUBBING. The table
INNODB_SYS_TABLESPACES might seem more appropriate, but its purpose
is to reflect the contents of the InnoDB system table SYS_TABLESPACES,
which we would like to remove at some point.
On Microsoft Windows, querying StorageDeviceSeekPenaltyProperty
sometimes returns ERROR_GEN_FAILURE instead of ERROR_INVALID_FUNCTION
or ERROR_NOT_SUPPORTED. We will silently ignore also this error,
and assume that the file does not reside on SSD.
On Linux, the detection will be based on the files
/sys/block/*/queue/rotational and /sys/block/*/dev.
Especially for USB storage, it is possible that
/sys/block/*/queue/rotational will wrongly report 1 instead of 0.
fil_node_t::on_ssd: Whether the InnoDB data file resides on
solid-state storage.
fil_system_t::ssd: Collection of Linux block devices that reside on
non-rotational storage.
fil_system_t::create(): Detect ssd on Linux based on the contents
of /sys/block/*/queue/rotational and /sys/block/*/dev.
fil_system_t::is_ssd(dev_t): Determine if a Linux block device is
non-rotational. Partitions will be identified with the containing
block device by assuming that the least significant 4 bits of the
minor number identify a partition, and that the "partition number"
of the entire device is 0.
2019-04-01 11:57:06 +03:00
|
|
|
#ifdef UNIV_LINUX
|
|
|
|
|
/** available block devices that reside on non-rotational storage */
|
|
|
|
|
std::vector<dev_t> ssd;
|
|
|
|
|
public:
|
|
|
|
|
/** @return whether a file system device is on non-rotational storage */
|
|
|
|
|
bool is_ssd(dev_t dev) const
|
|
|
|
|
{
|
|
|
|
|
/* Linux seems to allow up to 15 partitions per block device.
|
|
|
|
|
If the detected ssd carries "partition number 0" (it is the whole device),
|
|
|
|
|
compare the candidate file system number without the partition number. */
|
|
|
|
|
for (const auto s : ssd)
|
|
|
|
|
if (dev == s || (dev & ~15U) == s)
|
|
|
|
|
return true;
|
|
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
#endif
|
2018-03-28 09:00:06 +03:00
|
|
|
public:
|
2020-05-29 01:24:50 +03:00
|
|
|
/** Detach a tablespace from the cache and close the files.
|
|
|
|
|
@param space tablespace
|
MDEV-25691: Simplify handlerton::drop_database for InnoDB
The implementation of handlerton::drop_database in InnoDB is
unnecessarily complex. The minimal implementation should check
that no conflicting locks or references exist on the tables,
delete all table metadata in a single transaction, and finally
delete the tablespaces.
Note: DROP DATABASE will delete each individual table that the
SQL layer knows about, one table per transaction.
The handlerton::drop_database is basically a final cleanup step
for removing any garbage that could have been left behind
in InnoDB due to some bug, or not having atomic DDL in the past.
hash_node_t: Remove. Use the proper data type name in pointers.
dict_drop_index_tree(): Do not take the table as a parameter.
Instead, return the tablespace ID if the tablespace should be dropped
(we are dropping a clustered index tree).
fil_delete_tablespace(), fil_system_t::detach(): Return a single
detached file handle. Multi-file tablespaces cannot be deleted
via this interface.
ha_innobase::delete_table(): Remove a work-around for non-atomic DDL
and do not try to drop tables with similar-looking name.
innodb_drop_database(): Complete rewrite.
innobase_drop_database(), dict_get_first_table_name_in_db(),
row_drop_database_for_mysql(), drop_all_foreign_keys_in_db(): Remove.
row_purge_remove_clust_if_poss_low(), row_undo_ins_remove_clust_rec():
If the tablespace is to be deleted, try to evict the table definition
from the cache. Failing that, set dict_table_t::space to nullptr.
lock_release_on_rollback(): On the rollback of CREATE TABLE, release all
locks that the transaction had on the table, to avoid heap-use-after-free.
2021-05-18 12:53:40 +03:00
|
|
|
@param detach_handle whether to detach the handle, instead of closing
|
|
|
|
|
@return detached handle
|
|
|
|
|
@retval OS_FILE_CLOSED if no handle was detached */
|
|
|
|
|
pfs_os_file_t detach(fil_space_t *space, bool detach_handle= false);
|
MDEV-15053 Reduce buf_pool_t::mutex contention
User-visible changes: The INFORMATION_SCHEMA views INNODB_BUFFER_PAGE
and INNODB_BUFFER_PAGE_LRU will report a dummy value FLUSH_TYPE=0
and will no longer report the PAGE_STATE value READY_FOR_USE.
We will remove some fields from buf_page_t and move much code to
member functions of buf_pool_t and buf_page_t, so that the access
rules of data members can be enforced consistently.
Evicting or adding pages in buf_pool.LRU will remain covered by
buf_pool.mutex.
Evicting or adding pages in buf_pool.page_hash will remain
covered by both buf_pool.mutex and the buf_pool.page_hash X-latch.
After this fix, buf_pool.page_hash lookups can entirely
avoid acquiring buf_pool.mutex, only relying on
buf_pool.hash_lock_get() S-latch.
Similarly, buf_flush_check_neighbors() can will rely solely on
buf_pool.mutex, no buf_pool.page_hash latch at all.
The buf_pool.mutex is rather contended in I/O heavy benchmarks,
especially when the workload does not fit in the buffer pool.
The first attempt to alleviate the contention was the
buf_pool_t::mutex split in
commit 4ed7082eefe56b3e97e0edefb3df76dd7ef5e858
which introduced buf_block_t::mutex, which we are now removing.
Later, multiple instances of buf_pool_t were introduced
in commit c18084f71b02ea707c6461353e6cfc15d7553bc6
and recently removed by us in
commit 1a6f708ec594ac0ae2dd30db926ab07b100fa24b (MDEV-15058).
UNIV_BUF_DEBUG: Remove. This option to enable some buffer pool
related debugging in otherwise non-debug builds has not been used
for years. Instead, we have been using UNIV_DEBUG, which is enabled
in CMAKE_BUILD_TYPE=Debug.
buf_block_t::mutex, buf_pool_t::zip_mutex: Remove. We can mainly rely on
std::atomic and the buf_pool.page_hash latches, and in some cases
depend on buf_pool.mutex or buf_pool.flush_list_mutex just like before.
We must always release buf_block_t::lock before invoking
unfix() or io_unfix(), to prevent a glitch where a block that was
added to the buf_pool.free list would apper X-latched. See
commit c5883debd6ef440a037011c11873b396923e93c5 how this glitch
was finally caught in a debug environment.
We move some buf_pool_t::page_hash specific code from the
ha and hash modules to buf_pool, for improved readability.
buf_pool_t::close(): Assert that all blocks are clean, except
on aborted startup or crash-like shutdown.
buf_pool_t::validate(): No longer attempt to validate
n_flush[] against the number of BUF_IO_WRITE fixed blocks,
because buf_page_t::flush_type no longer exists.
buf_pool_t::watch_set(): Replaces buf_pool_watch_set().
Reduce mutex contention by separating the buf_pool.watch[]
allocation and the insert into buf_pool.page_hash.
buf_pool_t::page_hash_lock<bool exclusive>(): Acquire a
buf_pool.page_hash latch.
Replaces and extends buf_page_hash_lock_s_confirm()
and buf_page_hash_lock_x_confirm().
buf_pool_t::READ_AHEAD_PAGES: Renamed from BUF_READ_AHEAD_PAGES.
buf_pool_t::curr_size, old_size, read_ahead_area, n_pend_reads:
Use Atomic_counter.
buf_pool_t::running_out(): Replaces buf_LRU_buf_pool_running_out().
buf_pool_t::LRU_remove(): Remove a block from the LRU list
and return its predecessor. Incorporates buf_LRU_adjust_hp(),
which was removed.
buf_page_get_gen(): Remove a redundant call of fsp_is_system_temporary(),
for mode == BUF_GET_IF_IN_POOL_OR_WATCH, which is only used by
BTR_DELETE_OP (purge), which is never invoked on temporary tables.
buf_free_from_unzip_LRU_list_batch(): Avoid redundant assignments.
buf_LRU_free_from_unzip_LRU_list(): Simplify the loop condition.
buf_LRU_free_page(): Clarify the function comment.
buf_flush_check_neighbor(), buf_flush_check_neighbors():
Rewrite the construction of the page hash range. We will hold
the buf_pool.mutex for up to buf_pool.read_ahead_area (at most 64)
consecutive lookups of buf_pool.page_hash.
buf_flush_page_and_try_neighbors(): Remove.
Merge to its only callers, and remove redundant operations in
buf_flush_LRU_list_batch().
buf_read_ahead_random(), buf_read_ahead_linear(): Rewrite.
Do not acquire buf_pool.mutex, and iterate directly with page_id_t.
ut_2_power_up(): Remove. my_round_up_to_next_power() is inlined
and avoids any loops.
fil_page_get_prev(), fil_page_get_next(), fil_addr_is_null(): Remove.
buf_flush_page(): Add a fil_space_t* parameter. Minimize the
buf_pool.mutex hold time. buf_pool.n_flush[] is no longer updated
atomically with the io_fix, and we will protect most buf_block_t
fields with buf_block_t::lock. The function
buf_flush_write_block_low() is removed and merged here.
buf_page_init_for_read(): Use static linkage. Initialize the newly
allocated block and acquire the exclusive buf_block_t::lock while not
holding any mutex.
IORequest::IORequest(): Remove the body. We only need to invoke
set_punch_hole() in buf_flush_page() and nowhere else.
buf_page_t::flush_type: Remove. Replaced by IORequest::flush_type.
This field is only used during a fil_io() call.
That function already takes IORequest as a parameter, so we had
better introduce for the rarely changing field.
buf_block_t::init(): Replaces buf_page_init().
buf_page_t::init(): Replaces buf_page_init_low().
buf_block_t::initialise(): Initialise many fields, but
keep the buf_page_t::state(). Both buf_pool_t::validate() and
buf_page_optimistic_get() requires that buf_page_t::in_file()
be protected atomically with buf_page_t::in_page_hash
and buf_page_t::in_LRU_list.
buf_page_optimistic_get(): Now that buf_block_t::mutex
no longer exists, we must check buf_page_t::io_fix()
after acquiring the buf_pool.page_hash lock, to detect
whether buf_page_init_for_read() has been initiated.
We will also check the io_fix() before acquiring hash_lock
in order to avoid unnecessary computation.
The field buf_block_t::modify_clock (protected by buf_block_t::lock)
allows buf_page_optimistic_get() to validate the block.
buf_page_t::real_size: Remove. It was only used while flushing
pages of page_compressed tables.
buf_page_encrypt(): Add an output parameter that allows us ot eliminate
buf_page_t::real_size. Replace a condition with debug assertion.
buf_page_should_punch_hole(): Remove.
buf_dblwr_t::add_to_batch(): Replaces buf_dblwr_add_to_batch().
Add the parameter size (to replace buf_page_t::real_size).
buf_dblwr_t::write_single_page(): Replaces buf_dblwr_write_single_page().
Add the parameter size (to replace buf_page_t::real_size).
fil_system_t::detach(): Replaces fil_space_detach().
Ensure that fil_validate() will not be violated even if
fil_system.mutex is released and reacquired.
fil_node_t::complete_io(): Renamed from fil_node_complete_io().
fil_node_t::close_to_free(): Replaces fil_node_close_to_free().
Avoid invoking fil_node_t::close() because fil_system.n_open
has already been decremented in fil_space_t::detach().
BUF_BLOCK_READY_FOR_USE: Remove. Directly use BUF_BLOCK_MEMORY.
BUF_BLOCK_ZIP_DIRTY: Remove. Directly use BUF_BLOCK_ZIP_PAGE,
and distinguish dirty pages by buf_page_t::oldest_modification().
BUF_BLOCK_POOL_WATCH: Remove. Use BUF_BLOCK_NOT_USED instead.
This state was only being used for buf_page_t that are in
buf_pool.watch.
buf_pool_t::watch[]: Remove pointer indirection.
buf_page_t::in_flush_list: Remove. It was set if and only if
buf_page_t::oldest_modification() is nonzero.
buf_page_decrypt_after_read(), buf_corrupt_page_release(),
buf_page_check_corrupt(): Change the const fil_space_t* parameter
to const fil_node_t& so that we can report the correct file name.
buf_page_monitor(): Declare as an ATTRIBUTE_COLD global function.
buf_page_io_complete(): Split to buf_page_read_complete() and
buf_page_write_complete().
buf_dblwr_t::in_use: Remove.
buf_dblwr_t::buf_block_array: Add IORequest::flush_t.
buf_dblwr_sync_datafiles(): Remove. It was a useless wrapper of
os_aio_wait_until_no_pending_writes().
buf_flush_write_complete(): Declare static, not global.
Add the parameter IORequest::flush_t.
buf_flush_freed_page(): Simplify the code.
recv_sys_t::flush_lru: Renamed from flush_type and changed to bool.
fil_read(), fil_write(): Replaced with direct use of fil_io().
fil_buffering_disabled(): Remove. Check srv_file_flush_method directly.
fil_mutex_enter_and_prepare_for_io(): Return the resolved
fil_space_t* to avoid a duplicated lookup in the caller.
fil_report_invalid_page_access(): Clean up the parameters.
fil_io(): Return fil_io_t, which comprises fil_node_t and error code.
Always invoke fil_space_t::acquire_for_io() and let either the
sync=true caller or fil_aio_callback() invoke
fil_space_t::release_for_io().
fil_aio_callback(): Rewrite to replace buf_page_io_complete().
fil_check_pending_operations(): Remove a parameter, and remove some
redundant lookups.
fil_node_close_to_free(): Wait for n_pending==0. Because we no longer
do an extra lookup of the tablespace between fil_io() and the
completion of the operation, we must give fil_node_t::complete_io() a
chance to decrement the counter.
fil_close_tablespace(): Remove unused parameter trx, and document
that this is only invoked during the error handling of IMPORT TABLESPACE.
row_import_discard_changes(): Merged with the only caller,
row_import_cleanup(). Do not lock up the data dictionary while
invoking fil_close_tablespace().
logs_empty_and_mark_files_at_shutdown(): Do not invoke
fil_close_all_files(), to avoid a !needs_flush assertion failure
on fil_node_t::close().
innodb_shutdown(): Invoke os_aio_free() before fil_close_all_files().
fil_close_all_files(): Invoke fil_flush_file_spaces()
to ensure proper durability.
thread_pool::unbind(): Fix a crash that would occur on Windows
after srv_thread_pool->disable_aio() and os_file_close().
This fix was submitted by Vladislav Vaintroub.
Thanks to Matthias Leich and Axel Schwenke for extensive testing,
Vladislav Vaintroub for helpful comments, and Eugene Kosov for a review.
2020-06-05 12:35:46 +03:00
|
|
|
|
2020-12-04 19:02:58 +02:00
|
|
|
/** the mutex protecting most data fields, and some fields of fil_space_t */
|
|
|
|
|
mysql_mutex_t mutex;
|
2018-03-28 09:29:14 +03:00
|
|
|
fil_space_t* sys_space; /*!< The innodb_system tablespace */
|
|
|
|
|
fil_space_t* temp_space; /*!< The innodb_temporary tablespace */
|
2020-06-18 12:26:28 +03:00
|
|
|
/** Map of fil_space_t::id to fil_space_t* */
|
|
|
|
|
hash_table_t spaces;
|
2020-10-26 16:04:12 +02:00
|
|
|
/** tablespaces for which fil_space_t::needs_flush() holds */
|
|
|
|
|
sized_ilist<fil_space_t, unflushed_spaces_tag_t> unflushed_spaces;
|
2020-10-26 15:59:30 +02:00
|
|
|
/** number of currently open files; protected by mutex */
|
|
|
|
|
ulint n_open;
|
2021-09-30 10:01:10 +03:00
|
|
|
/** last time we noted n_open exceeding the limit; protected by mutex */
|
|
|
|
|
time_t n_open_exceeded_time;
|
2014-10-25 08:21:52 +03:00
|
|
|
ulint max_assigned_id;/*!< maximum space id in the existing
|
|
|
|
|
tables, or assigned during the time
|
|
|
|
|
mysqld has been up; at an InnoDB
|
|
|
|
|
startup we scan the data dictionary
|
|
|
|
|
and set here the maximum of the
|
|
|
|
|
space id's of the tables there */
|
2020-10-26 15:59:30 +02:00
|
|
|
/** nonzero if fil_node_open_file_low() should avoid moving the tablespace
|
|
|
|
|
to the end of space_list, for FIFO policy of try_to_close() */
|
|
|
|
|
ulint freeze_space_list;
|
2021-03-20 01:31:01 +03:00
|
|
|
ilist<fil_space_t, space_list_tag_t> space_list;
|
2014-10-25 08:21:52 +03:00
|
|
|
/*!< list of all file spaces */
|
2021-03-20 01:31:01 +03:00
|
|
|
ilist<fil_space_t, named_spaces_tag_t> named_spaces;
|
2016-08-12 11:17:45 +03:00
|
|
|
/*!< list of all file spaces
|
MDEV-12353: Change the redo log encoding
log_t::FORMAT_10_5: physical redo log format tag
log_phys_t: Buffered records in the physical format.
The log record bytes will follow the last data field,
making use of alignment padding that would otherwise be wasted.
If there are multiple records for the same page, also those
may be appended to an existing log_phys_t object if the memory
is available.
In the physical format, the first byte of a record identifies the
record and its length (up to 15 bytes). For longer records, the
immediately following bytes will encode the remaining length
in a variable-length encoding. Usually, a variable-length-encoded
page identifier will follow, followed by optional payload, whose
length is included in the initially encoded total record length.
When a mini-transaction is updating multiple fields in a page,
it can avoid repeating the tablespace identifier and page number
by setting the same_page flag (most significant bit) in the first
byte of the log record. The byte offset of the record will be
relative to where the previous record for that page ended.
Until MDEV-14425 introduces a separate file-level log for
redo log checkpoints and file operations, we will write the
file-level records in the page-level redo log file.
The record FILE_CHECKPOINT (which replaces MLOG_CHECKPOINT)
will be removed in MDEV-14425, and one sequential scan of the
page recovery log will suffice.
Compared to MLOG_FILE_CREATE2, FILE_CREATE will not include any flags.
If the information is needed, it can be parsed from WRITE records that
modify FSP_SPACE_FLAGS.
MLOG_ZIP_WRITE_STRING: Remove. The record was only introduced temporarily
as part of this work, before being replaced with WRITE (along with
MLOG_WRITE_STRING, MLOG_1BYTE, MLOG_nBYTES).
mtr_buf_t::empty(): Check if the buffer is empty.
mtr_t::m_n_log_recs: Remove. It suffices to check if m_log is empty.
mtr_t::m_last, mtr_t::m_last_offset: End of the latest m_log record,
for the same_page encoding.
page_recv_t::last_offset: Reflects mtr_t::m_last_offset.
Valid values for last_offset during recovery should be 0 or above 8.
(The first 8 bytes of a page are the checksum and the page number,
and neither are ever updated directly by log records.)
Internally, the special value 1 indicates that the same_page form
will not be allowed for the subsequent record.
mtr_t::page_create(): Take the block descriptor as parameter,
so that it can be compared to mtr_t::m_last. The INIT_INDEX_PAGE
record will always followed by a subtype byte, because same_page
records must be longer than 1 byte.
trx_undo_page_init(): Combine the writes in WRITE record.
trx_undo_header_create(): Write 4 bytes using a special MEMSET
record that includes 1 bytes of length and 2 bytes of payload.
flst_write_addr(): Define as a static function. Combine the writes.
flst_zero_both(): Replaces two flst_zero_addr() calls.
flst_init(): Do not inline the function.
fsp_free_seg_inode(): Zerofill the whole inode.
fsp_apply_init_file_page(): Initialize FIL_PAGE_PREV,FIL_PAGE_NEXT
to FIL_NULL when using the physical format.
btr_create(): Assert !page_has_siblings() because fsp_apply_init_file_page()
must have been invoked.
fil_ibd_create(): Do not write FILE_MODIFY after FILE_CREATE.
fil_names_dirty_and_write(): Remove the parameter mtr.
Write the records using a separate mini-transaction object,
because any FILE_ records must be at the start of a mini-transaction log.
recv_recover_page(): Add a fil_space_t* parameter.
After applying log to the a ROW_FORMAT=COMPRESSED page,
invoke buf_zip_decompress() to restore the uncompressed page.
buf_page_io_complete(): Remove the temporary hack to discard the
uncompressed page of a ROW_FORMAT=COMPRESSED page.
page_zip_write_header(): Remove. Use mtr_t::write() or
mtr_t::memset() instead, and update the compressed page frame
separately.
trx_undo_header_add_space_for_xid(): Remove.
trx_undo_seg_create(): Perform the changes that were previously
made by trx_undo_header_add_space_for_xid().
btr_reset_instant(): New function: Reset the table to MariaDB 10.2
or 10.3 format when rolling back an instant ALTER TABLE operation.
page_rec_find_owner_rec(): Merge with the only callers.
page_cur_insert_rec_low(): Combine writes by using a local buffer.
MEMMOVE data from the preceding record whenever feasible
(copying at least 3 bytes).
page_cur_insert_rec_zip(): Combine writes to page header fields.
PageBulk::insertPage(): Issue MEMMOVE records to copy a matching
part from the preceding record.
PageBulk::finishPage(): Combine the writes to the page header
and to the sparse page directory slots.
mtr_t::write(): Only log the least significant (last) bytes
of multi-byte fields that actually differ.
For updating FSP_SIZE, we must always write all 4 bytes to the
redo log, so that the fil_space_set_recv_size() logic in
recv_sys_t::parse() will work.
mtr_t::memcpy(), mtr_t::zmemcpy(): Take a pointer argument
instead of a numeric offset to the page frame. Only log the
last bytes of multi-byte fields that actually differ.
In fil_space_crypt_t::write_page0(), we must log also any
unchanged bytes, so that recovery will recognize the record
and invoke fil_crypt_parse().
Future work:
MDEV-21724 Optimize page_cur_insert_rec_low() redo logging
MDEV-21725 Optimize btr_page_reorganize_low() redo logging
MDEV-21727 Optimize redo logging for ROW_FORMAT=COMPRESSED
2020-02-13 19:12:17 +02:00
|
|
|
for which a FILE_MODIFY
|
2016-08-12 11:17:45 +03:00
|
|
|
record has been written since
|
|
|
|
|
the latest redo log checkpoint.
|
2018-04-27 10:06:14 +03:00
|
|
|
Protected only by log_sys.mutex. */
|
2021-06-14 19:37:06 +05:30
|
|
|
|
|
|
|
|
/** List of all file spaces need key rotation */
|
|
|
|
|
ilist<fil_space_t, rotation_list_tag_t> default_encrypt_tables;
|
2017-03-14 12:56:01 +02:00
|
|
|
|
2018-03-28 09:00:06 +03:00
|
|
|
bool space_id_reuse_warned;
|
2020-10-26 15:59:30 +02:00
|
|
|
/*!< whether fil_space_t::create()
|
2014-10-25 08:21:52 +03:00
|
|
|
has issued a warning about
|
|
|
|
|
potential space_id reuse */
|
2019-05-01 17:24:58 +05:30
|
|
|
|
2021-06-14 19:37:06 +05:30
|
|
|
/** Return the next tablespace from default_encrypt_tables list.
|
2021-06-21 14:22:22 +03:00
|
|
|
@param space previous tablespace (nullptr to start from the start)
|
2020-08-10 17:57:14 +03:00
|
|
|
@param recheck whether the removal condition needs to be rechecked after
|
|
|
|
|
the encryption parameters were changed
|
|
|
|
|
@param encrypt expected state of innodb_encrypt_tables
|
|
|
|
|
@return the next tablespace to process (n_pending_ops incremented)
|
2021-06-21 14:22:22 +03:00
|
|
|
@retval fil_system.temp_space if there is no work to do
|
|
|
|
|
@retval nullptr upon reaching the end of the iteration */
|
|
|
|
|
inline fil_space_t* default_encrypt_next(fil_space_t *space, bool recheck,
|
|
|
|
|
bool encrypt);
|
2021-03-10 19:33:09 +02:00
|
|
|
|
2021-05-17 18:12:33 +03:00
|
|
|
/** Extend all open data files to the recovered size */
|
|
|
|
|
ATTRIBUTE_COLD void extend_to_recv_size();
|
|
|
|
|
|
|
|
|
|
/** Determine if a tablespace associated with a file name exists.
|
|
|
|
|
@param path tablespace file name to look for
|
|
|
|
|
@return a matching tablespace */
|
|
|
|
|
inline fil_space_t *find(const char *path) const;
|
2014-10-25 08:21:52 +03:00
|
|
|
};
|
|
|
|
|
|
2018-03-28 09:00:06 +03:00
|
|
|
/** The tablespace memory cache. */
|
|
|
|
|
extern fil_system_t fil_system;
|
2016-08-12 11:17:45 +03:00
|
|
|
|
2020-10-26 16:04:12 +02:00
|
|
|
inline void fil_space_t::reacquire()
|
|
|
|
|
{
|
|
|
|
|
ut_d(uint32_t n=) n_pending.fetch_add(1, std::memory_order_relaxed);
|
2020-12-04 19:02:58 +02:00
|
|
|
#ifdef SAFE_MUTEX
|
|
|
|
|
if (mysql_mutex_is_owner(&fil_system.mutex)) return;
|
2020-10-26 16:04:12 +02:00
|
|
|
ut_ad(n & PENDING);
|
|
|
|
|
ut_ad(UT_LIST_GET_FIRST(chain)->is_open());
|
2020-12-04 19:02:58 +02:00
|
|
|
#endif /* SAFE_MUTEX */
|
2020-10-26 16:04:12 +02:00
|
|
|
}
|
|
|
|
|
|
MDEV-25506 (3 of 3): Do not delete .ibd files before commit
This is a complete rewrite of DROP TABLE, also as part of other DDL,
such as ALTER TABLE, CREATE TABLE...SELECT, TRUNCATE TABLE.
The background DROP TABLE queue hack is removed.
If a transaction needs to drop and create a table by the same name
(like TRUNCATE TABLE does), it must first rename the table to an
internal #sql-ib name. No committed version of the data dictionary
will include any #sql-ib tables, because whenever a transaction
renames a table to a #sql-ib name, it will also drop that table.
Either the rename will be rolled back, or the drop will be committed.
Data files will be unlinked after the transaction has been committed
and a FILE_RENAME record has been durably written. The file will
actually be deleted when the detached file handle returned by
fil_delete_tablespace() will be closed, after the latches have been
released. It is possible that a purge of the delete of the SYS_INDEXES
record for the clustered index will execute fil_delete_tablespace()
concurrently with the DDL transaction. In that case, the thread that
arrives later will wait for the other thread to finish.
HTON_TRUNCATE_REQUIRES_EXCLUSIVE_USE: A new handler flag.
ha_innobase::truncate() now requires that all other references to
the table be released in advance. This was implemented by Monty.
ha_innobase::delete_table(): If CREATE TABLE..SELECT is detected,
we will "hijack" the current transaction, drop the table in
the current transaction and commit the current transaction.
This essentially fixes MDEV-21602. There is a FIXME comment about
making the check less failure-prone.
ha_innobase::truncate(), ha_innobase::delete_table():
Implement a fast path for temporary tables. We will no longer allow
temporary tables to use the adaptive hash index.
dict_table_t::mdl_name: The original table name for the purpose of
acquiring MDL in purge, to prevent a race condition between a
DDL transaction that is dropping a table, and purge processing
undo log records of DML that had executed before the DDL operation.
For #sql-backup- tables during ALTER TABLE...ALGORITHM=COPY, the
dict_table_t::mdl_name will differ from dict_table_t::name.
dict_table_t::parse_name(): Use mdl_name instead of name.
dict_table_rename_in_cache(): Update mdl_name.
For the internal FTS_ tables of FULLTEXT INDEX, purge would
acquire MDL on the FTS_ table name, but not on the main table,
and therefore it would be able to run concurrently with a
DDL transaction that is dropping the table. Previously, the
DROP TABLE queue hack prevented a race between purge and DDL.
For now, we introduce purge_sys.stop_FTS() to prevent purge from
opening any table, while a DDL transaction that may drop FTS_
tables is in progress. The function fts_lock_table(), which will
be invoked before the dictionary is locked, will wait for
purge to release any table handles.
trx_t::drop_table_statistics(): Drop statistics for the table.
This replaces dict_stats_drop_index(). We will drop or rename
persistent statistics atomically as part of DDL transactions.
On lock conflict for dropping statistics, we will fail instantly
with DB_LOCK_WAIT_TIMEOUT, because we will be holding the
exclusive data dictionary latch.
trx_t::commit_cleanup(): Separated from trx_t::commit_in_memory().
Relax an assertion around fts_commit() and allow DB_LOCK_WAIT_TIMEOUT
in addition to DB_DUPLICATE_KEY. The call to fts_commit() is
entirely misplaced here and may obviously break the consistency
of transactions that affect FULLTEXT INDEX. It needs to be fixed
separately.
dict_table_t::n_foreign_key_checks_running: Remove (MDEV-21175).
The counter was a work-around for missing meta-data locking (MDL)
on the SQL layer, and not really needed in MariaDB.
ER_TABLE_IN_FK_CHECK: Replaced with ER_UNUSED_28.
HA_ERR_TABLE_IN_FK_CHECK: Remove.
row_ins_check_foreign_constraints(): Do not acquire
dict_sys.latch either. The SQL-layer MDL will protect us.
This was reviewed by Thirunarayanan Balathandayuthapani
and tested by Matthias Leich.
2021-06-09 17:02:55 +03:00
|
|
|
/** Note that operations on the tablespace must stop.
|
|
|
|
|
@return whether the operations were already stopped */
|
2021-10-03 13:49:40 +03:00
|
|
|
inline bool fil_space_t::set_stopping_check()
|
MDEV-25506 (3 of 3): Do not delete .ibd files before commit
This is a complete rewrite of DROP TABLE, also as part of other DDL,
such as ALTER TABLE, CREATE TABLE...SELECT, TRUNCATE TABLE.
The background DROP TABLE queue hack is removed.
If a transaction needs to drop and create a table by the same name
(like TRUNCATE TABLE does), it must first rename the table to an
internal #sql-ib name. No committed version of the data dictionary
will include any #sql-ib tables, because whenever a transaction
renames a table to a #sql-ib name, it will also drop that table.
Either the rename will be rolled back, or the drop will be committed.
Data files will be unlinked after the transaction has been committed
and a FILE_RENAME record has been durably written. The file will
actually be deleted when the detached file handle returned by
fil_delete_tablespace() will be closed, after the latches have been
released. It is possible that a purge of the delete of the SYS_INDEXES
record for the clustered index will execute fil_delete_tablespace()
concurrently with the DDL transaction. In that case, the thread that
arrives later will wait for the other thread to finish.
HTON_TRUNCATE_REQUIRES_EXCLUSIVE_USE: A new handler flag.
ha_innobase::truncate() now requires that all other references to
the table be released in advance. This was implemented by Monty.
ha_innobase::delete_table(): If CREATE TABLE..SELECT is detected,
we will "hijack" the current transaction, drop the table in
the current transaction and commit the current transaction.
This essentially fixes MDEV-21602. There is a FIXME comment about
making the check less failure-prone.
ha_innobase::truncate(), ha_innobase::delete_table():
Implement a fast path for temporary tables. We will no longer allow
temporary tables to use the adaptive hash index.
dict_table_t::mdl_name: The original table name for the purpose of
acquiring MDL in purge, to prevent a race condition between a
DDL transaction that is dropping a table, and purge processing
undo log records of DML that had executed before the DDL operation.
For #sql-backup- tables during ALTER TABLE...ALGORITHM=COPY, the
dict_table_t::mdl_name will differ from dict_table_t::name.
dict_table_t::parse_name(): Use mdl_name instead of name.
dict_table_rename_in_cache(): Update mdl_name.
For the internal FTS_ tables of FULLTEXT INDEX, purge would
acquire MDL on the FTS_ table name, but not on the main table,
and therefore it would be able to run concurrently with a
DDL transaction that is dropping the table. Previously, the
DROP TABLE queue hack prevented a race between purge and DDL.
For now, we introduce purge_sys.stop_FTS() to prevent purge from
opening any table, while a DDL transaction that may drop FTS_
tables is in progress. The function fts_lock_table(), which will
be invoked before the dictionary is locked, will wait for
purge to release any table handles.
trx_t::drop_table_statistics(): Drop statistics for the table.
This replaces dict_stats_drop_index(). We will drop or rename
persistent statistics atomically as part of DDL transactions.
On lock conflict for dropping statistics, we will fail instantly
with DB_LOCK_WAIT_TIMEOUT, because we will be holding the
exclusive data dictionary latch.
trx_t::commit_cleanup(): Separated from trx_t::commit_in_memory().
Relax an assertion around fts_commit() and allow DB_LOCK_WAIT_TIMEOUT
in addition to DB_DUPLICATE_KEY. The call to fts_commit() is
entirely misplaced here and may obviously break the consistency
of transactions that affect FULLTEXT INDEX. It needs to be fixed
separately.
dict_table_t::n_foreign_key_checks_running: Remove (MDEV-21175).
The counter was a work-around for missing meta-data locking (MDL)
on the SQL layer, and not really needed in MariaDB.
ER_TABLE_IN_FK_CHECK: Replaced with ER_UNUSED_28.
HA_ERR_TABLE_IN_FK_CHECK: Remove.
row_ins_check_foreign_constraints(): Do not acquire
dict_sys.latch either. The SQL-layer MDL will protect us.
This was reviewed by Thirunarayanan Balathandayuthapani
and tested by Matthias Leich.
2021-06-09 17:02:55 +03:00
|
|
|
{
|
|
|
|
|
mysql_mutex_assert_owner(&fil_system.mutex);
|
2022-01-28 08:31:57 +02:00
|
|
|
#if (defined __clang_major__ && __clang_major__ < 10) || defined __APPLE_CC__
|
2021-10-12 07:47:10 +03:00
|
|
|
/* Only clang-10 introduced support for asm goto */
|
|
|
|
|
return n_pending.fetch_or(STOPPING, std::memory_order_relaxed) & STOPPING;
|
|
|
|
|
#elif defined __GNUC__ && (defined __i386__ || defined __x86_64__)
|
2021-10-03 13:49:40 +03:00
|
|
|
static_assert(STOPPING == 1U << 31, "compatibility");
|
|
|
|
|
__asm__ goto("lock btsl $31, %0\t\njnc %l1" : : "m" (n_pending)
|
|
|
|
|
: "cc", "memory" : not_stopped);
|
|
|
|
|
return true;
|
|
|
|
|
not_stopped:
|
|
|
|
|
return false;
|
2022-01-28 16:42:37 +02:00
|
|
|
#elif defined _MSC_VER && (defined _M_IX86 || defined _M_X64)
|
2021-10-03 13:49:40 +03:00
|
|
|
static_assert(STOPPING == 1U << 31, "compatibility");
|
|
|
|
|
return _interlockedbittestandset(reinterpret_cast<volatile long*>
|
|
|
|
|
(&n_pending), 31);
|
|
|
|
|
#else
|
MDEV-25506 (3 of 3): Do not delete .ibd files before commit
This is a complete rewrite of DROP TABLE, also as part of other DDL,
such as ALTER TABLE, CREATE TABLE...SELECT, TRUNCATE TABLE.
The background DROP TABLE queue hack is removed.
If a transaction needs to drop and create a table by the same name
(like TRUNCATE TABLE does), it must first rename the table to an
internal #sql-ib name. No committed version of the data dictionary
will include any #sql-ib tables, because whenever a transaction
renames a table to a #sql-ib name, it will also drop that table.
Either the rename will be rolled back, or the drop will be committed.
Data files will be unlinked after the transaction has been committed
and a FILE_RENAME record has been durably written. The file will
actually be deleted when the detached file handle returned by
fil_delete_tablespace() will be closed, after the latches have been
released. It is possible that a purge of the delete of the SYS_INDEXES
record for the clustered index will execute fil_delete_tablespace()
concurrently with the DDL transaction. In that case, the thread that
arrives later will wait for the other thread to finish.
HTON_TRUNCATE_REQUIRES_EXCLUSIVE_USE: A new handler flag.
ha_innobase::truncate() now requires that all other references to
the table be released in advance. This was implemented by Monty.
ha_innobase::delete_table(): If CREATE TABLE..SELECT is detected,
we will "hijack" the current transaction, drop the table in
the current transaction and commit the current transaction.
This essentially fixes MDEV-21602. There is a FIXME comment about
making the check less failure-prone.
ha_innobase::truncate(), ha_innobase::delete_table():
Implement a fast path for temporary tables. We will no longer allow
temporary tables to use the adaptive hash index.
dict_table_t::mdl_name: The original table name for the purpose of
acquiring MDL in purge, to prevent a race condition between a
DDL transaction that is dropping a table, and purge processing
undo log records of DML that had executed before the DDL operation.
For #sql-backup- tables during ALTER TABLE...ALGORITHM=COPY, the
dict_table_t::mdl_name will differ from dict_table_t::name.
dict_table_t::parse_name(): Use mdl_name instead of name.
dict_table_rename_in_cache(): Update mdl_name.
For the internal FTS_ tables of FULLTEXT INDEX, purge would
acquire MDL on the FTS_ table name, but not on the main table,
and therefore it would be able to run concurrently with a
DDL transaction that is dropping the table. Previously, the
DROP TABLE queue hack prevented a race between purge and DDL.
For now, we introduce purge_sys.stop_FTS() to prevent purge from
opening any table, while a DDL transaction that may drop FTS_
tables is in progress. The function fts_lock_table(), which will
be invoked before the dictionary is locked, will wait for
purge to release any table handles.
trx_t::drop_table_statistics(): Drop statistics for the table.
This replaces dict_stats_drop_index(). We will drop or rename
persistent statistics atomically as part of DDL transactions.
On lock conflict for dropping statistics, we will fail instantly
with DB_LOCK_WAIT_TIMEOUT, because we will be holding the
exclusive data dictionary latch.
trx_t::commit_cleanup(): Separated from trx_t::commit_in_memory().
Relax an assertion around fts_commit() and allow DB_LOCK_WAIT_TIMEOUT
in addition to DB_DUPLICATE_KEY. The call to fts_commit() is
entirely misplaced here and may obviously break the consistency
of transactions that affect FULLTEXT INDEX. It needs to be fixed
separately.
dict_table_t::n_foreign_key_checks_running: Remove (MDEV-21175).
The counter was a work-around for missing meta-data locking (MDL)
on the SQL layer, and not really needed in MariaDB.
ER_TABLE_IN_FK_CHECK: Replaced with ER_UNUSED_28.
HA_ERR_TABLE_IN_FK_CHECK: Remove.
row_ins_check_foreign_constraints(): Do not acquire
dict_sys.latch either. The SQL-layer MDL will protect us.
This was reviewed by Thirunarayanan Balathandayuthapani
and tested by Matthias Leich.
2021-06-09 17:02:55 +03:00
|
|
|
return n_pending.fetch_or(STOPPING, std::memory_order_relaxed) & STOPPING;
|
2021-10-03 13:49:40 +03:00
|
|
|
#endif
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/** Note that operations on the tablespace must stop.
|
|
|
|
|
@return whether the operations were already stopped */
|
|
|
|
|
inline void fil_space_t::set_stopping()
|
|
|
|
|
{
|
|
|
|
|
mysql_mutex_assert_owner(&fil_system.mutex);
|
|
|
|
|
#if defined __GNUC__ && (defined __i386__ || defined __x86_64__)
|
|
|
|
|
static_assert(STOPPING == 1U << 31, "compatibility");
|
|
|
|
|
__asm__ __volatile__("lock btsl $31, %0" : "+m" (n_pending));
|
2022-01-28 16:42:37 +02:00
|
|
|
#elif defined _MSC_VER && (defined _M_IX86 || defined _M_X64)
|
2021-10-03 13:49:40 +03:00
|
|
|
static_assert(STOPPING == 1U << 31, "compatibility");
|
|
|
|
|
_interlockedbittestandset(reinterpret_cast<volatile long*>(&n_pending), 31);
|
|
|
|
|
#else
|
|
|
|
|
n_pending.fetch_or(STOPPING, std::memory_order_relaxed);
|
|
|
|
|
#endif
|
MDEV-25506 (3 of 3): Do not delete .ibd files before commit
This is a complete rewrite of DROP TABLE, also as part of other DDL,
such as ALTER TABLE, CREATE TABLE...SELECT, TRUNCATE TABLE.
The background DROP TABLE queue hack is removed.
If a transaction needs to drop and create a table by the same name
(like TRUNCATE TABLE does), it must first rename the table to an
internal #sql-ib name. No committed version of the data dictionary
will include any #sql-ib tables, because whenever a transaction
renames a table to a #sql-ib name, it will also drop that table.
Either the rename will be rolled back, or the drop will be committed.
Data files will be unlinked after the transaction has been committed
and a FILE_RENAME record has been durably written. The file will
actually be deleted when the detached file handle returned by
fil_delete_tablespace() will be closed, after the latches have been
released. It is possible that a purge of the delete of the SYS_INDEXES
record for the clustered index will execute fil_delete_tablespace()
concurrently with the DDL transaction. In that case, the thread that
arrives later will wait for the other thread to finish.
HTON_TRUNCATE_REQUIRES_EXCLUSIVE_USE: A new handler flag.
ha_innobase::truncate() now requires that all other references to
the table be released in advance. This was implemented by Monty.
ha_innobase::delete_table(): If CREATE TABLE..SELECT is detected,
we will "hijack" the current transaction, drop the table in
the current transaction and commit the current transaction.
This essentially fixes MDEV-21602. There is a FIXME comment about
making the check less failure-prone.
ha_innobase::truncate(), ha_innobase::delete_table():
Implement a fast path for temporary tables. We will no longer allow
temporary tables to use the adaptive hash index.
dict_table_t::mdl_name: The original table name for the purpose of
acquiring MDL in purge, to prevent a race condition between a
DDL transaction that is dropping a table, and purge processing
undo log records of DML that had executed before the DDL operation.
For #sql-backup- tables during ALTER TABLE...ALGORITHM=COPY, the
dict_table_t::mdl_name will differ from dict_table_t::name.
dict_table_t::parse_name(): Use mdl_name instead of name.
dict_table_rename_in_cache(): Update mdl_name.
For the internal FTS_ tables of FULLTEXT INDEX, purge would
acquire MDL on the FTS_ table name, but not on the main table,
and therefore it would be able to run concurrently with a
DDL transaction that is dropping the table. Previously, the
DROP TABLE queue hack prevented a race between purge and DDL.
For now, we introduce purge_sys.stop_FTS() to prevent purge from
opening any table, while a DDL transaction that may drop FTS_
tables is in progress. The function fts_lock_table(), which will
be invoked before the dictionary is locked, will wait for
purge to release any table handles.
trx_t::drop_table_statistics(): Drop statistics for the table.
This replaces dict_stats_drop_index(). We will drop or rename
persistent statistics atomically as part of DDL transactions.
On lock conflict for dropping statistics, we will fail instantly
with DB_LOCK_WAIT_TIMEOUT, because we will be holding the
exclusive data dictionary latch.
trx_t::commit_cleanup(): Separated from trx_t::commit_in_memory().
Relax an assertion around fts_commit() and allow DB_LOCK_WAIT_TIMEOUT
in addition to DB_DUPLICATE_KEY. The call to fts_commit() is
entirely misplaced here and may obviously break the consistency
of transactions that affect FULLTEXT INDEX. It needs to be fixed
separately.
dict_table_t::n_foreign_key_checks_running: Remove (MDEV-21175).
The counter was a work-around for missing meta-data locking (MDL)
on the SQL layer, and not really needed in MariaDB.
ER_TABLE_IN_FK_CHECK: Replaced with ER_UNUSED_28.
HA_ERR_TABLE_IN_FK_CHECK: Remove.
row_ins_check_foreign_constraints(): Do not acquire
dict_sys.latch either. The SQL-layer MDL will protect us.
This was reviewed by Thirunarayanan Balathandayuthapani
and tested by Matthias Leich.
2021-06-09 17:02:55 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
inline void fil_space_t::clear_stopping()
|
MDEV-15053 Reduce buf_pool_t::mutex contention
User-visible changes: The INFORMATION_SCHEMA views INNODB_BUFFER_PAGE
and INNODB_BUFFER_PAGE_LRU will report a dummy value FLUSH_TYPE=0
and will no longer report the PAGE_STATE value READY_FOR_USE.
We will remove some fields from buf_page_t and move much code to
member functions of buf_pool_t and buf_page_t, so that the access
rules of data members can be enforced consistently.
Evicting or adding pages in buf_pool.LRU will remain covered by
buf_pool.mutex.
Evicting or adding pages in buf_pool.page_hash will remain
covered by both buf_pool.mutex and the buf_pool.page_hash X-latch.
After this fix, buf_pool.page_hash lookups can entirely
avoid acquiring buf_pool.mutex, only relying on
buf_pool.hash_lock_get() S-latch.
Similarly, buf_flush_check_neighbors() can will rely solely on
buf_pool.mutex, no buf_pool.page_hash latch at all.
The buf_pool.mutex is rather contended in I/O heavy benchmarks,
especially when the workload does not fit in the buffer pool.
The first attempt to alleviate the contention was the
buf_pool_t::mutex split in
commit 4ed7082eefe56b3e97e0edefb3df76dd7ef5e858
which introduced buf_block_t::mutex, which we are now removing.
Later, multiple instances of buf_pool_t were introduced
in commit c18084f71b02ea707c6461353e6cfc15d7553bc6
and recently removed by us in
commit 1a6f708ec594ac0ae2dd30db926ab07b100fa24b (MDEV-15058).
UNIV_BUF_DEBUG: Remove. This option to enable some buffer pool
related debugging in otherwise non-debug builds has not been used
for years. Instead, we have been using UNIV_DEBUG, which is enabled
in CMAKE_BUILD_TYPE=Debug.
buf_block_t::mutex, buf_pool_t::zip_mutex: Remove. We can mainly rely on
std::atomic and the buf_pool.page_hash latches, and in some cases
depend on buf_pool.mutex or buf_pool.flush_list_mutex just like before.
We must always release buf_block_t::lock before invoking
unfix() or io_unfix(), to prevent a glitch where a block that was
added to the buf_pool.free list would apper X-latched. See
commit c5883debd6ef440a037011c11873b396923e93c5 how this glitch
was finally caught in a debug environment.
We move some buf_pool_t::page_hash specific code from the
ha and hash modules to buf_pool, for improved readability.
buf_pool_t::close(): Assert that all blocks are clean, except
on aborted startup or crash-like shutdown.
buf_pool_t::validate(): No longer attempt to validate
n_flush[] against the number of BUF_IO_WRITE fixed blocks,
because buf_page_t::flush_type no longer exists.
buf_pool_t::watch_set(): Replaces buf_pool_watch_set().
Reduce mutex contention by separating the buf_pool.watch[]
allocation and the insert into buf_pool.page_hash.
buf_pool_t::page_hash_lock<bool exclusive>(): Acquire a
buf_pool.page_hash latch.
Replaces and extends buf_page_hash_lock_s_confirm()
and buf_page_hash_lock_x_confirm().
buf_pool_t::READ_AHEAD_PAGES: Renamed from BUF_READ_AHEAD_PAGES.
buf_pool_t::curr_size, old_size, read_ahead_area, n_pend_reads:
Use Atomic_counter.
buf_pool_t::running_out(): Replaces buf_LRU_buf_pool_running_out().
buf_pool_t::LRU_remove(): Remove a block from the LRU list
and return its predecessor. Incorporates buf_LRU_adjust_hp(),
which was removed.
buf_page_get_gen(): Remove a redundant call of fsp_is_system_temporary(),
for mode == BUF_GET_IF_IN_POOL_OR_WATCH, which is only used by
BTR_DELETE_OP (purge), which is never invoked on temporary tables.
buf_free_from_unzip_LRU_list_batch(): Avoid redundant assignments.
buf_LRU_free_from_unzip_LRU_list(): Simplify the loop condition.
buf_LRU_free_page(): Clarify the function comment.
buf_flush_check_neighbor(), buf_flush_check_neighbors():
Rewrite the construction of the page hash range. We will hold
the buf_pool.mutex for up to buf_pool.read_ahead_area (at most 64)
consecutive lookups of buf_pool.page_hash.
buf_flush_page_and_try_neighbors(): Remove.
Merge to its only callers, and remove redundant operations in
buf_flush_LRU_list_batch().
buf_read_ahead_random(), buf_read_ahead_linear(): Rewrite.
Do not acquire buf_pool.mutex, and iterate directly with page_id_t.
ut_2_power_up(): Remove. my_round_up_to_next_power() is inlined
and avoids any loops.
fil_page_get_prev(), fil_page_get_next(), fil_addr_is_null(): Remove.
buf_flush_page(): Add a fil_space_t* parameter. Minimize the
buf_pool.mutex hold time. buf_pool.n_flush[] is no longer updated
atomically with the io_fix, and we will protect most buf_block_t
fields with buf_block_t::lock. The function
buf_flush_write_block_low() is removed and merged here.
buf_page_init_for_read(): Use static linkage. Initialize the newly
allocated block and acquire the exclusive buf_block_t::lock while not
holding any mutex.
IORequest::IORequest(): Remove the body. We only need to invoke
set_punch_hole() in buf_flush_page() and nowhere else.
buf_page_t::flush_type: Remove. Replaced by IORequest::flush_type.
This field is only used during a fil_io() call.
That function already takes IORequest as a parameter, so we had
better introduce for the rarely changing field.
buf_block_t::init(): Replaces buf_page_init().
buf_page_t::init(): Replaces buf_page_init_low().
buf_block_t::initialise(): Initialise many fields, but
keep the buf_page_t::state(). Both buf_pool_t::validate() and
buf_page_optimistic_get() requires that buf_page_t::in_file()
be protected atomically with buf_page_t::in_page_hash
and buf_page_t::in_LRU_list.
buf_page_optimistic_get(): Now that buf_block_t::mutex
no longer exists, we must check buf_page_t::io_fix()
after acquiring the buf_pool.page_hash lock, to detect
whether buf_page_init_for_read() has been initiated.
We will also check the io_fix() before acquiring hash_lock
in order to avoid unnecessary computation.
The field buf_block_t::modify_clock (protected by buf_block_t::lock)
allows buf_page_optimistic_get() to validate the block.
buf_page_t::real_size: Remove. It was only used while flushing
pages of page_compressed tables.
buf_page_encrypt(): Add an output parameter that allows us ot eliminate
buf_page_t::real_size. Replace a condition with debug assertion.
buf_page_should_punch_hole(): Remove.
buf_dblwr_t::add_to_batch(): Replaces buf_dblwr_add_to_batch().
Add the parameter size (to replace buf_page_t::real_size).
buf_dblwr_t::write_single_page(): Replaces buf_dblwr_write_single_page().
Add the parameter size (to replace buf_page_t::real_size).
fil_system_t::detach(): Replaces fil_space_detach().
Ensure that fil_validate() will not be violated even if
fil_system.mutex is released and reacquired.
fil_node_t::complete_io(): Renamed from fil_node_complete_io().
fil_node_t::close_to_free(): Replaces fil_node_close_to_free().
Avoid invoking fil_node_t::close() because fil_system.n_open
has already been decremented in fil_space_t::detach().
BUF_BLOCK_READY_FOR_USE: Remove. Directly use BUF_BLOCK_MEMORY.
BUF_BLOCK_ZIP_DIRTY: Remove. Directly use BUF_BLOCK_ZIP_PAGE,
and distinguish dirty pages by buf_page_t::oldest_modification().
BUF_BLOCK_POOL_WATCH: Remove. Use BUF_BLOCK_NOT_USED instead.
This state was only being used for buf_page_t that are in
buf_pool.watch.
buf_pool_t::watch[]: Remove pointer indirection.
buf_page_t::in_flush_list: Remove. It was set if and only if
buf_page_t::oldest_modification() is nonzero.
buf_page_decrypt_after_read(), buf_corrupt_page_release(),
buf_page_check_corrupt(): Change the const fil_space_t* parameter
to const fil_node_t& so that we can report the correct file name.
buf_page_monitor(): Declare as an ATTRIBUTE_COLD global function.
buf_page_io_complete(): Split to buf_page_read_complete() and
buf_page_write_complete().
buf_dblwr_t::in_use: Remove.
buf_dblwr_t::buf_block_array: Add IORequest::flush_t.
buf_dblwr_sync_datafiles(): Remove. It was a useless wrapper of
os_aio_wait_until_no_pending_writes().
buf_flush_write_complete(): Declare static, not global.
Add the parameter IORequest::flush_t.
buf_flush_freed_page(): Simplify the code.
recv_sys_t::flush_lru: Renamed from flush_type and changed to bool.
fil_read(), fil_write(): Replaced with direct use of fil_io().
fil_buffering_disabled(): Remove. Check srv_file_flush_method directly.
fil_mutex_enter_and_prepare_for_io(): Return the resolved
fil_space_t* to avoid a duplicated lookup in the caller.
fil_report_invalid_page_access(): Clean up the parameters.
fil_io(): Return fil_io_t, which comprises fil_node_t and error code.
Always invoke fil_space_t::acquire_for_io() and let either the
sync=true caller or fil_aio_callback() invoke
fil_space_t::release_for_io().
fil_aio_callback(): Rewrite to replace buf_page_io_complete().
fil_check_pending_operations(): Remove a parameter, and remove some
redundant lookups.
fil_node_close_to_free(): Wait for n_pending==0. Because we no longer
do an extra lookup of the tablespace between fil_io() and the
completion of the operation, we must give fil_node_t::complete_io() a
chance to decrement the counter.
fil_close_tablespace(): Remove unused parameter trx, and document
that this is only invoked during the error handling of IMPORT TABLESPACE.
row_import_discard_changes(): Merged with the only caller,
row_import_cleanup(). Do not lock up the data dictionary while
invoking fil_close_tablespace().
logs_empty_and_mark_files_at_shutdown(): Do not invoke
fil_close_all_files(), to avoid a !needs_flush assertion failure
on fil_node_t::close().
innodb_shutdown(): Invoke os_aio_free() before fil_close_all_files().
fil_close_all_files(): Invoke fil_flush_file_spaces()
to ensure proper durability.
thread_pool::unbind(): Fix a crash that would occur on Windows
after srv_thread_pool->disable_aio() and os_file_close().
This fix was submitted by Vladislav Vaintroub.
Thanks to Matthias Leich and Axel Schwenke for extensive testing,
Vladislav Vaintroub for helpful comments, and Eugene Kosov for a review.
2020-06-05 12:35:46 +03:00
|
|
|
{
|
2020-12-04 19:02:58 +02:00
|
|
|
mysql_mutex_assert_owner(&fil_system.mutex);
|
2021-10-03 13:49:40 +03:00
|
|
|
static_assert(STOPPING == 1U << 31, "compatibility");
|
|
|
|
|
ut_d(auto n=) n_pending.fetch_sub(STOPPING, std::memory_order_relaxed);
|
MDEV-25506 (3 of 3): Do not delete .ibd files before commit
This is a complete rewrite of DROP TABLE, also as part of other DDL,
such as ALTER TABLE, CREATE TABLE...SELECT, TRUNCATE TABLE.
The background DROP TABLE queue hack is removed.
If a transaction needs to drop and create a table by the same name
(like TRUNCATE TABLE does), it must first rename the table to an
internal #sql-ib name. No committed version of the data dictionary
will include any #sql-ib tables, because whenever a transaction
renames a table to a #sql-ib name, it will also drop that table.
Either the rename will be rolled back, or the drop will be committed.
Data files will be unlinked after the transaction has been committed
and a FILE_RENAME record has been durably written. The file will
actually be deleted when the detached file handle returned by
fil_delete_tablespace() will be closed, after the latches have been
released. It is possible that a purge of the delete of the SYS_INDEXES
record for the clustered index will execute fil_delete_tablespace()
concurrently with the DDL transaction. In that case, the thread that
arrives later will wait for the other thread to finish.
HTON_TRUNCATE_REQUIRES_EXCLUSIVE_USE: A new handler flag.
ha_innobase::truncate() now requires that all other references to
the table be released in advance. This was implemented by Monty.
ha_innobase::delete_table(): If CREATE TABLE..SELECT is detected,
we will "hijack" the current transaction, drop the table in
the current transaction and commit the current transaction.
This essentially fixes MDEV-21602. There is a FIXME comment about
making the check less failure-prone.
ha_innobase::truncate(), ha_innobase::delete_table():
Implement a fast path for temporary tables. We will no longer allow
temporary tables to use the adaptive hash index.
dict_table_t::mdl_name: The original table name for the purpose of
acquiring MDL in purge, to prevent a race condition between a
DDL transaction that is dropping a table, and purge processing
undo log records of DML that had executed before the DDL operation.
For #sql-backup- tables during ALTER TABLE...ALGORITHM=COPY, the
dict_table_t::mdl_name will differ from dict_table_t::name.
dict_table_t::parse_name(): Use mdl_name instead of name.
dict_table_rename_in_cache(): Update mdl_name.
For the internal FTS_ tables of FULLTEXT INDEX, purge would
acquire MDL on the FTS_ table name, but not on the main table,
and therefore it would be able to run concurrently with a
DDL transaction that is dropping the table. Previously, the
DROP TABLE queue hack prevented a race between purge and DDL.
For now, we introduce purge_sys.stop_FTS() to prevent purge from
opening any table, while a DDL transaction that may drop FTS_
tables is in progress. The function fts_lock_table(), which will
be invoked before the dictionary is locked, will wait for
purge to release any table handles.
trx_t::drop_table_statistics(): Drop statistics for the table.
This replaces dict_stats_drop_index(). We will drop or rename
persistent statistics atomically as part of DDL transactions.
On lock conflict for dropping statistics, we will fail instantly
with DB_LOCK_WAIT_TIMEOUT, because we will be holding the
exclusive data dictionary latch.
trx_t::commit_cleanup(): Separated from trx_t::commit_in_memory().
Relax an assertion around fts_commit() and allow DB_LOCK_WAIT_TIMEOUT
in addition to DB_DUPLICATE_KEY. The call to fts_commit() is
entirely misplaced here and may obviously break the consistency
of transactions that affect FULLTEXT INDEX. It needs to be fixed
separately.
dict_table_t::n_foreign_key_checks_running: Remove (MDEV-21175).
The counter was a work-around for missing meta-data locking (MDL)
on the SQL layer, and not really needed in MariaDB.
ER_TABLE_IN_FK_CHECK: Replaced with ER_UNUSED_28.
HA_ERR_TABLE_IN_FK_CHECK: Remove.
row_ins_check_foreign_constraints(): Do not acquire
dict_sys.latch either. The SQL-layer MDL will protect us.
This was reviewed by Thirunarayanan Balathandayuthapani
and tested by Matthias Leich.
2021-06-09 17:02:55 +03:00
|
|
|
ut_ad(n & STOPPING);
|
2020-10-26 16:04:12 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/** Flush pending writes from the file system cache to the file. */
|
2020-12-11 09:05:26 +02:00
|
|
|
template<bool have_reference> inline void fil_space_t::flush()
|
2020-10-26 16:04:12 +02:00
|
|
|
{
|
2020-12-04 19:02:58 +02:00
|
|
|
mysql_mutex_assert_not_owner(&fil_system.mutex);
|
2020-12-11 09:05:26 +02:00
|
|
|
ut_ad(!have_reference || (pending() & PENDING));
|
2020-10-26 16:04:12 +02:00
|
|
|
ut_ad(purpose == FIL_TYPE_TABLESPACE || purpose == FIL_TYPE_IMPORT);
|
|
|
|
|
if (srv_file_flush_method == SRV_O_DIRECT_NO_FSYNC)
|
|
|
|
|
{
|
|
|
|
|
ut_ad(!is_in_unflushed_spaces);
|
|
|
|
|
ut_ad(!needs_flush());
|
|
|
|
|
}
|
2020-12-11 09:05:26 +02:00
|
|
|
else if (have_reference)
|
|
|
|
|
flush_low();
|
|
|
|
|
else if (!(acquire_low() & STOPPING))
|
|
|
|
|
{
|
2020-10-26 16:04:12 +02:00
|
|
|
flush_low();
|
2020-12-11 09:05:26 +02:00
|
|
|
release();
|
|
|
|
|
}
|
2020-10-26 15:59:30 +02:00
|
|
|
}
|
MDEV-15053 Reduce buf_pool_t::mutex contention
User-visible changes: The INFORMATION_SCHEMA views INNODB_BUFFER_PAGE
and INNODB_BUFFER_PAGE_LRU will report a dummy value FLUSH_TYPE=0
and will no longer report the PAGE_STATE value READY_FOR_USE.
We will remove some fields from buf_page_t and move much code to
member functions of buf_pool_t and buf_page_t, so that the access
rules of data members can be enforced consistently.
Evicting or adding pages in buf_pool.LRU will remain covered by
buf_pool.mutex.
Evicting or adding pages in buf_pool.page_hash will remain
covered by both buf_pool.mutex and the buf_pool.page_hash X-latch.
After this fix, buf_pool.page_hash lookups can entirely
avoid acquiring buf_pool.mutex, only relying on
buf_pool.hash_lock_get() S-latch.
Similarly, buf_flush_check_neighbors() can will rely solely on
buf_pool.mutex, no buf_pool.page_hash latch at all.
The buf_pool.mutex is rather contended in I/O heavy benchmarks,
especially when the workload does not fit in the buffer pool.
The first attempt to alleviate the contention was the
buf_pool_t::mutex split in
commit 4ed7082eefe56b3e97e0edefb3df76dd7ef5e858
which introduced buf_block_t::mutex, which we are now removing.
Later, multiple instances of buf_pool_t were introduced
in commit c18084f71b02ea707c6461353e6cfc15d7553bc6
and recently removed by us in
commit 1a6f708ec594ac0ae2dd30db926ab07b100fa24b (MDEV-15058).
UNIV_BUF_DEBUG: Remove. This option to enable some buffer pool
related debugging in otherwise non-debug builds has not been used
for years. Instead, we have been using UNIV_DEBUG, which is enabled
in CMAKE_BUILD_TYPE=Debug.
buf_block_t::mutex, buf_pool_t::zip_mutex: Remove. We can mainly rely on
std::atomic and the buf_pool.page_hash latches, and in some cases
depend on buf_pool.mutex or buf_pool.flush_list_mutex just like before.
We must always release buf_block_t::lock before invoking
unfix() or io_unfix(), to prevent a glitch where a block that was
added to the buf_pool.free list would apper X-latched. See
commit c5883debd6ef440a037011c11873b396923e93c5 how this glitch
was finally caught in a debug environment.
We move some buf_pool_t::page_hash specific code from the
ha and hash modules to buf_pool, for improved readability.
buf_pool_t::close(): Assert that all blocks are clean, except
on aborted startup or crash-like shutdown.
buf_pool_t::validate(): No longer attempt to validate
n_flush[] against the number of BUF_IO_WRITE fixed blocks,
because buf_page_t::flush_type no longer exists.
buf_pool_t::watch_set(): Replaces buf_pool_watch_set().
Reduce mutex contention by separating the buf_pool.watch[]
allocation and the insert into buf_pool.page_hash.
buf_pool_t::page_hash_lock<bool exclusive>(): Acquire a
buf_pool.page_hash latch.
Replaces and extends buf_page_hash_lock_s_confirm()
and buf_page_hash_lock_x_confirm().
buf_pool_t::READ_AHEAD_PAGES: Renamed from BUF_READ_AHEAD_PAGES.
buf_pool_t::curr_size, old_size, read_ahead_area, n_pend_reads:
Use Atomic_counter.
buf_pool_t::running_out(): Replaces buf_LRU_buf_pool_running_out().
buf_pool_t::LRU_remove(): Remove a block from the LRU list
and return its predecessor. Incorporates buf_LRU_adjust_hp(),
which was removed.
buf_page_get_gen(): Remove a redundant call of fsp_is_system_temporary(),
for mode == BUF_GET_IF_IN_POOL_OR_WATCH, which is only used by
BTR_DELETE_OP (purge), which is never invoked on temporary tables.
buf_free_from_unzip_LRU_list_batch(): Avoid redundant assignments.
buf_LRU_free_from_unzip_LRU_list(): Simplify the loop condition.
buf_LRU_free_page(): Clarify the function comment.
buf_flush_check_neighbor(), buf_flush_check_neighbors():
Rewrite the construction of the page hash range. We will hold
the buf_pool.mutex for up to buf_pool.read_ahead_area (at most 64)
consecutive lookups of buf_pool.page_hash.
buf_flush_page_and_try_neighbors(): Remove.
Merge to its only callers, and remove redundant operations in
buf_flush_LRU_list_batch().
buf_read_ahead_random(), buf_read_ahead_linear(): Rewrite.
Do not acquire buf_pool.mutex, and iterate directly with page_id_t.
ut_2_power_up(): Remove. my_round_up_to_next_power() is inlined
and avoids any loops.
fil_page_get_prev(), fil_page_get_next(), fil_addr_is_null(): Remove.
buf_flush_page(): Add a fil_space_t* parameter. Minimize the
buf_pool.mutex hold time. buf_pool.n_flush[] is no longer updated
atomically with the io_fix, and we will protect most buf_block_t
fields with buf_block_t::lock. The function
buf_flush_write_block_low() is removed and merged here.
buf_page_init_for_read(): Use static linkage. Initialize the newly
allocated block and acquire the exclusive buf_block_t::lock while not
holding any mutex.
IORequest::IORequest(): Remove the body. We only need to invoke
set_punch_hole() in buf_flush_page() and nowhere else.
buf_page_t::flush_type: Remove. Replaced by IORequest::flush_type.
This field is only used during a fil_io() call.
That function already takes IORequest as a parameter, so we had
better introduce for the rarely changing field.
buf_block_t::init(): Replaces buf_page_init().
buf_page_t::init(): Replaces buf_page_init_low().
buf_block_t::initialise(): Initialise many fields, but
keep the buf_page_t::state(). Both buf_pool_t::validate() and
buf_page_optimistic_get() requires that buf_page_t::in_file()
be protected atomically with buf_page_t::in_page_hash
and buf_page_t::in_LRU_list.
buf_page_optimistic_get(): Now that buf_block_t::mutex
no longer exists, we must check buf_page_t::io_fix()
after acquiring the buf_pool.page_hash lock, to detect
whether buf_page_init_for_read() has been initiated.
We will also check the io_fix() before acquiring hash_lock
in order to avoid unnecessary computation.
The field buf_block_t::modify_clock (protected by buf_block_t::lock)
allows buf_page_optimistic_get() to validate the block.
buf_page_t::real_size: Remove. It was only used while flushing
pages of page_compressed tables.
buf_page_encrypt(): Add an output parameter that allows us ot eliminate
buf_page_t::real_size. Replace a condition with debug assertion.
buf_page_should_punch_hole(): Remove.
buf_dblwr_t::add_to_batch(): Replaces buf_dblwr_add_to_batch().
Add the parameter size (to replace buf_page_t::real_size).
buf_dblwr_t::write_single_page(): Replaces buf_dblwr_write_single_page().
Add the parameter size (to replace buf_page_t::real_size).
fil_system_t::detach(): Replaces fil_space_detach().
Ensure that fil_validate() will not be violated even if
fil_system.mutex is released and reacquired.
fil_node_t::complete_io(): Renamed from fil_node_complete_io().
fil_node_t::close_to_free(): Replaces fil_node_close_to_free().
Avoid invoking fil_node_t::close() because fil_system.n_open
has already been decremented in fil_space_t::detach().
BUF_BLOCK_READY_FOR_USE: Remove. Directly use BUF_BLOCK_MEMORY.
BUF_BLOCK_ZIP_DIRTY: Remove. Directly use BUF_BLOCK_ZIP_PAGE,
and distinguish dirty pages by buf_page_t::oldest_modification().
BUF_BLOCK_POOL_WATCH: Remove. Use BUF_BLOCK_NOT_USED instead.
This state was only being used for buf_page_t that are in
buf_pool.watch.
buf_pool_t::watch[]: Remove pointer indirection.
buf_page_t::in_flush_list: Remove. It was set if and only if
buf_page_t::oldest_modification() is nonzero.
buf_page_decrypt_after_read(), buf_corrupt_page_release(),
buf_page_check_corrupt(): Change the const fil_space_t* parameter
to const fil_node_t& so that we can report the correct file name.
buf_page_monitor(): Declare as an ATTRIBUTE_COLD global function.
buf_page_io_complete(): Split to buf_page_read_complete() and
buf_page_write_complete().
buf_dblwr_t::in_use: Remove.
buf_dblwr_t::buf_block_array: Add IORequest::flush_t.
buf_dblwr_sync_datafiles(): Remove. It was a useless wrapper of
os_aio_wait_until_no_pending_writes().
buf_flush_write_complete(): Declare static, not global.
Add the parameter IORequest::flush_t.
buf_flush_freed_page(): Simplify the code.
recv_sys_t::flush_lru: Renamed from flush_type and changed to bool.
fil_read(), fil_write(): Replaced with direct use of fil_io().
fil_buffering_disabled(): Remove. Check srv_file_flush_method directly.
fil_mutex_enter_and_prepare_for_io(): Return the resolved
fil_space_t* to avoid a duplicated lookup in the caller.
fil_report_invalid_page_access(): Clean up the parameters.
fil_io(): Return fil_io_t, which comprises fil_node_t and error code.
Always invoke fil_space_t::acquire_for_io() and let either the
sync=true caller or fil_aio_callback() invoke
fil_space_t::release_for_io().
fil_aio_callback(): Rewrite to replace buf_page_io_complete().
fil_check_pending_operations(): Remove a parameter, and remove some
redundant lookups.
fil_node_close_to_free(): Wait for n_pending==0. Because we no longer
do an extra lookup of the tablespace between fil_io() and the
completion of the operation, we must give fil_node_t::complete_io() a
chance to decrement the counter.
fil_close_tablespace(): Remove unused parameter trx, and document
that this is only invoked during the error handling of IMPORT TABLESPACE.
row_import_discard_changes(): Merged with the only caller,
row_import_cleanup(). Do not lock up the data dictionary while
invoking fil_close_tablespace().
logs_empty_and_mark_files_at_shutdown(): Do not invoke
fil_close_all_files(), to avoid a !needs_flush assertion failure
on fil_node_t::close().
innodb_shutdown(): Invoke os_aio_free() before fil_close_all_files().
fil_close_all_files(): Invoke fil_flush_file_spaces()
to ensure proper durability.
thread_pool::unbind(): Fix a crash that would occur on Windows
after srv_thread_pool->disable_aio() and os_file_close().
This fix was submitted by Vladislav Vaintroub.
Thanks to Matthias Leich and Axel Schwenke for extensive testing,
Vladislav Vaintroub for helpful comments, and Eugene Kosov for a review.
2020-06-05 12:35:46 +03:00
|
|
|
|
2020-10-26 15:59:30 +02:00
|
|
|
/** @return the size in pages (0 if unreadable) */
|
|
|
|
|
inline uint32_t fil_space_t::get_size()
|
|
|
|
|
{
|
|
|
|
|
if (!size)
|
MDEV-15053 Reduce buf_pool_t::mutex contention
User-visible changes: The INFORMATION_SCHEMA views INNODB_BUFFER_PAGE
and INNODB_BUFFER_PAGE_LRU will report a dummy value FLUSH_TYPE=0
and will no longer report the PAGE_STATE value READY_FOR_USE.
We will remove some fields from buf_page_t and move much code to
member functions of buf_pool_t and buf_page_t, so that the access
rules of data members can be enforced consistently.
Evicting or adding pages in buf_pool.LRU will remain covered by
buf_pool.mutex.
Evicting or adding pages in buf_pool.page_hash will remain
covered by both buf_pool.mutex and the buf_pool.page_hash X-latch.
After this fix, buf_pool.page_hash lookups can entirely
avoid acquiring buf_pool.mutex, only relying on
buf_pool.hash_lock_get() S-latch.
Similarly, buf_flush_check_neighbors() can will rely solely on
buf_pool.mutex, no buf_pool.page_hash latch at all.
The buf_pool.mutex is rather contended in I/O heavy benchmarks,
especially when the workload does not fit in the buffer pool.
The first attempt to alleviate the contention was the
buf_pool_t::mutex split in
commit 4ed7082eefe56b3e97e0edefb3df76dd7ef5e858
which introduced buf_block_t::mutex, which we are now removing.
Later, multiple instances of buf_pool_t were introduced
in commit c18084f71b02ea707c6461353e6cfc15d7553bc6
and recently removed by us in
commit 1a6f708ec594ac0ae2dd30db926ab07b100fa24b (MDEV-15058).
UNIV_BUF_DEBUG: Remove. This option to enable some buffer pool
related debugging in otherwise non-debug builds has not been used
for years. Instead, we have been using UNIV_DEBUG, which is enabled
in CMAKE_BUILD_TYPE=Debug.
buf_block_t::mutex, buf_pool_t::zip_mutex: Remove. We can mainly rely on
std::atomic and the buf_pool.page_hash latches, and in some cases
depend on buf_pool.mutex or buf_pool.flush_list_mutex just like before.
We must always release buf_block_t::lock before invoking
unfix() or io_unfix(), to prevent a glitch where a block that was
added to the buf_pool.free list would apper X-latched. See
commit c5883debd6ef440a037011c11873b396923e93c5 how this glitch
was finally caught in a debug environment.
We move some buf_pool_t::page_hash specific code from the
ha and hash modules to buf_pool, for improved readability.
buf_pool_t::close(): Assert that all blocks are clean, except
on aborted startup or crash-like shutdown.
buf_pool_t::validate(): No longer attempt to validate
n_flush[] against the number of BUF_IO_WRITE fixed blocks,
because buf_page_t::flush_type no longer exists.
buf_pool_t::watch_set(): Replaces buf_pool_watch_set().
Reduce mutex contention by separating the buf_pool.watch[]
allocation and the insert into buf_pool.page_hash.
buf_pool_t::page_hash_lock<bool exclusive>(): Acquire a
buf_pool.page_hash latch.
Replaces and extends buf_page_hash_lock_s_confirm()
and buf_page_hash_lock_x_confirm().
buf_pool_t::READ_AHEAD_PAGES: Renamed from BUF_READ_AHEAD_PAGES.
buf_pool_t::curr_size, old_size, read_ahead_area, n_pend_reads:
Use Atomic_counter.
buf_pool_t::running_out(): Replaces buf_LRU_buf_pool_running_out().
buf_pool_t::LRU_remove(): Remove a block from the LRU list
and return its predecessor. Incorporates buf_LRU_adjust_hp(),
which was removed.
buf_page_get_gen(): Remove a redundant call of fsp_is_system_temporary(),
for mode == BUF_GET_IF_IN_POOL_OR_WATCH, which is only used by
BTR_DELETE_OP (purge), which is never invoked on temporary tables.
buf_free_from_unzip_LRU_list_batch(): Avoid redundant assignments.
buf_LRU_free_from_unzip_LRU_list(): Simplify the loop condition.
buf_LRU_free_page(): Clarify the function comment.
buf_flush_check_neighbor(), buf_flush_check_neighbors():
Rewrite the construction of the page hash range. We will hold
the buf_pool.mutex for up to buf_pool.read_ahead_area (at most 64)
consecutive lookups of buf_pool.page_hash.
buf_flush_page_and_try_neighbors(): Remove.
Merge to its only callers, and remove redundant operations in
buf_flush_LRU_list_batch().
buf_read_ahead_random(), buf_read_ahead_linear(): Rewrite.
Do not acquire buf_pool.mutex, and iterate directly with page_id_t.
ut_2_power_up(): Remove. my_round_up_to_next_power() is inlined
and avoids any loops.
fil_page_get_prev(), fil_page_get_next(), fil_addr_is_null(): Remove.
buf_flush_page(): Add a fil_space_t* parameter. Minimize the
buf_pool.mutex hold time. buf_pool.n_flush[] is no longer updated
atomically with the io_fix, and we will protect most buf_block_t
fields with buf_block_t::lock. The function
buf_flush_write_block_low() is removed and merged here.
buf_page_init_for_read(): Use static linkage. Initialize the newly
allocated block and acquire the exclusive buf_block_t::lock while not
holding any mutex.
IORequest::IORequest(): Remove the body. We only need to invoke
set_punch_hole() in buf_flush_page() and nowhere else.
buf_page_t::flush_type: Remove. Replaced by IORequest::flush_type.
This field is only used during a fil_io() call.
That function already takes IORequest as a parameter, so we had
better introduce for the rarely changing field.
buf_block_t::init(): Replaces buf_page_init().
buf_page_t::init(): Replaces buf_page_init_low().
buf_block_t::initialise(): Initialise many fields, but
keep the buf_page_t::state(). Both buf_pool_t::validate() and
buf_page_optimistic_get() requires that buf_page_t::in_file()
be protected atomically with buf_page_t::in_page_hash
and buf_page_t::in_LRU_list.
buf_page_optimistic_get(): Now that buf_block_t::mutex
no longer exists, we must check buf_page_t::io_fix()
after acquiring the buf_pool.page_hash lock, to detect
whether buf_page_init_for_read() has been initiated.
We will also check the io_fix() before acquiring hash_lock
in order to avoid unnecessary computation.
The field buf_block_t::modify_clock (protected by buf_block_t::lock)
allows buf_page_optimistic_get() to validate the block.
buf_page_t::real_size: Remove. It was only used while flushing
pages of page_compressed tables.
buf_page_encrypt(): Add an output parameter that allows us ot eliminate
buf_page_t::real_size. Replace a condition with debug assertion.
buf_page_should_punch_hole(): Remove.
buf_dblwr_t::add_to_batch(): Replaces buf_dblwr_add_to_batch().
Add the parameter size (to replace buf_page_t::real_size).
buf_dblwr_t::write_single_page(): Replaces buf_dblwr_write_single_page().
Add the parameter size (to replace buf_page_t::real_size).
fil_system_t::detach(): Replaces fil_space_detach().
Ensure that fil_validate() will not be violated even if
fil_system.mutex is released and reacquired.
fil_node_t::complete_io(): Renamed from fil_node_complete_io().
fil_node_t::close_to_free(): Replaces fil_node_close_to_free().
Avoid invoking fil_node_t::close() because fil_system.n_open
has already been decremented in fil_space_t::detach().
BUF_BLOCK_READY_FOR_USE: Remove. Directly use BUF_BLOCK_MEMORY.
BUF_BLOCK_ZIP_DIRTY: Remove. Directly use BUF_BLOCK_ZIP_PAGE,
and distinguish dirty pages by buf_page_t::oldest_modification().
BUF_BLOCK_POOL_WATCH: Remove. Use BUF_BLOCK_NOT_USED instead.
This state was only being used for buf_page_t that are in
buf_pool.watch.
buf_pool_t::watch[]: Remove pointer indirection.
buf_page_t::in_flush_list: Remove. It was set if and only if
buf_page_t::oldest_modification() is nonzero.
buf_page_decrypt_after_read(), buf_corrupt_page_release(),
buf_page_check_corrupt(): Change the const fil_space_t* parameter
to const fil_node_t& so that we can report the correct file name.
buf_page_monitor(): Declare as an ATTRIBUTE_COLD global function.
buf_page_io_complete(): Split to buf_page_read_complete() and
buf_page_write_complete().
buf_dblwr_t::in_use: Remove.
buf_dblwr_t::buf_block_array: Add IORequest::flush_t.
buf_dblwr_sync_datafiles(): Remove. It was a useless wrapper of
os_aio_wait_until_no_pending_writes().
buf_flush_write_complete(): Declare static, not global.
Add the parameter IORequest::flush_t.
buf_flush_freed_page(): Simplify the code.
recv_sys_t::flush_lru: Renamed from flush_type and changed to bool.
fil_read(), fil_write(): Replaced with direct use of fil_io().
fil_buffering_disabled(): Remove. Check srv_file_flush_method directly.
fil_mutex_enter_and_prepare_for_io(): Return the resolved
fil_space_t* to avoid a duplicated lookup in the caller.
fil_report_invalid_page_access(): Clean up the parameters.
fil_io(): Return fil_io_t, which comprises fil_node_t and error code.
Always invoke fil_space_t::acquire_for_io() and let either the
sync=true caller or fil_aio_callback() invoke
fil_space_t::release_for_io().
fil_aio_callback(): Rewrite to replace buf_page_io_complete().
fil_check_pending_operations(): Remove a parameter, and remove some
redundant lookups.
fil_node_close_to_free(): Wait for n_pending==0. Because we no longer
do an extra lookup of the tablespace between fil_io() and the
completion of the operation, we must give fil_node_t::complete_io() a
chance to decrement the counter.
fil_close_tablespace(): Remove unused parameter trx, and document
that this is only invoked during the error handling of IMPORT TABLESPACE.
row_import_discard_changes(): Merged with the only caller,
row_import_cleanup(). Do not lock up the data dictionary while
invoking fil_close_tablespace().
logs_empty_and_mark_files_at_shutdown(): Do not invoke
fil_close_all_files(), to avoid a !needs_flush assertion failure
on fil_node_t::close().
innodb_shutdown(): Invoke os_aio_free() before fil_close_all_files().
fil_close_all_files(): Invoke fil_flush_file_spaces()
to ensure proper durability.
thread_pool::unbind(): Fix a crash that would occur on Windows
after srv_thread_pool->disable_aio() and os_file_close().
This fix was submitted by Vladislav Vaintroub.
Thanks to Matthias Leich and Axel Schwenke for extensive testing,
Vladislav Vaintroub for helpful comments, and Eugene Kosov for a review.
2020-06-05 12:35:46 +03:00
|
|
|
{
|
2020-12-04 19:02:58 +02:00
|
|
|
mysql_mutex_lock(&fil_system.mutex);
|
2020-10-26 15:59:30 +02:00
|
|
|
read_page0();
|
2020-12-04 19:02:58 +02:00
|
|
|
mysql_mutex_unlock(&fil_system.mutex);
|
MDEV-15053 Reduce buf_pool_t::mutex contention
User-visible changes: The INFORMATION_SCHEMA views INNODB_BUFFER_PAGE
and INNODB_BUFFER_PAGE_LRU will report a dummy value FLUSH_TYPE=0
and will no longer report the PAGE_STATE value READY_FOR_USE.
We will remove some fields from buf_page_t and move much code to
member functions of buf_pool_t and buf_page_t, so that the access
rules of data members can be enforced consistently.
Evicting or adding pages in buf_pool.LRU will remain covered by
buf_pool.mutex.
Evicting or adding pages in buf_pool.page_hash will remain
covered by both buf_pool.mutex and the buf_pool.page_hash X-latch.
After this fix, buf_pool.page_hash lookups can entirely
avoid acquiring buf_pool.mutex, only relying on
buf_pool.hash_lock_get() S-latch.
Similarly, buf_flush_check_neighbors() can will rely solely on
buf_pool.mutex, no buf_pool.page_hash latch at all.
The buf_pool.mutex is rather contended in I/O heavy benchmarks,
especially when the workload does not fit in the buffer pool.
The first attempt to alleviate the contention was the
buf_pool_t::mutex split in
commit 4ed7082eefe56b3e97e0edefb3df76dd7ef5e858
which introduced buf_block_t::mutex, which we are now removing.
Later, multiple instances of buf_pool_t were introduced
in commit c18084f71b02ea707c6461353e6cfc15d7553bc6
and recently removed by us in
commit 1a6f708ec594ac0ae2dd30db926ab07b100fa24b (MDEV-15058).
UNIV_BUF_DEBUG: Remove. This option to enable some buffer pool
related debugging in otherwise non-debug builds has not been used
for years. Instead, we have been using UNIV_DEBUG, which is enabled
in CMAKE_BUILD_TYPE=Debug.
buf_block_t::mutex, buf_pool_t::zip_mutex: Remove. We can mainly rely on
std::atomic and the buf_pool.page_hash latches, and in some cases
depend on buf_pool.mutex or buf_pool.flush_list_mutex just like before.
We must always release buf_block_t::lock before invoking
unfix() or io_unfix(), to prevent a glitch where a block that was
added to the buf_pool.free list would apper X-latched. See
commit c5883debd6ef440a037011c11873b396923e93c5 how this glitch
was finally caught in a debug environment.
We move some buf_pool_t::page_hash specific code from the
ha and hash modules to buf_pool, for improved readability.
buf_pool_t::close(): Assert that all blocks are clean, except
on aborted startup or crash-like shutdown.
buf_pool_t::validate(): No longer attempt to validate
n_flush[] against the number of BUF_IO_WRITE fixed blocks,
because buf_page_t::flush_type no longer exists.
buf_pool_t::watch_set(): Replaces buf_pool_watch_set().
Reduce mutex contention by separating the buf_pool.watch[]
allocation and the insert into buf_pool.page_hash.
buf_pool_t::page_hash_lock<bool exclusive>(): Acquire a
buf_pool.page_hash latch.
Replaces and extends buf_page_hash_lock_s_confirm()
and buf_page_hash_lock_x_confirm().
buf_pool_t::READ_AHEAD_PAGES: Renamed from BUF_READ_AHEAD_PAGES.
buf_pool_t::curr_size, old_size, read_ahead_area, n_pend_reads:
Use Atomic_counter.
buf_pool_t::running_out(): Replaces buf_LRU_buf_pool_running_out().
buf_pool_t::LRU_remove(): Remove a block from the LRU list
and return its predecessor. Incorporates buf_LRU_adjust_hp(),
which was removed.
buf_page_get_gen(): Remove a redundant call of fsp_is_system_temporary(),
for mode == BUF_GET_IF_IN_POOL_OR_WATCH, which is only used by
BTR_DELETE_OP (purge), which is never invoked on temporary tables.
buf_free_from_unzip_LRU_list_batch(): Avoid redundant assignments.
buf_LRU_free_from_unzip_LRU_list(): Simplify the loop condition.
buf_LRU_free_page(): Clarify the function comment.
buf_flush_check_neighbor(), buf_flush_check_neighbors():
Rewrite the construction of the page hash range. We will hold
the buf_pool.mutex for up to buf_pool.read_ahead_area (at most 64)
consecutive lookups of buf_pool.page_hash.
buf_flush_page_and_try_neighbors(): Remove.
Merge to its only callers, and remove redundant operations in
buf_flush_LRU_list_batch().
buf_read_ahead_random(), buf_read_ahead_linear(): Rewrite.
Do not acquire buf_pool.mutex, and iterate directly with page_id_t.
ut_2_power_up(): Remove. my_round_up_to_next_power() is inlined
and avoids any loops.
fil_page_get_prev(), fil_page_get_next(), fil_addr_is_null(): Remove.
buf_flush_page(): Add a fil_space_t* parameter. Minimize the
buf_pool.mutex hold time. buf_pool.n_flush[] is no longer updated
atomically with the io_fix, and we will protect most buf_block_t
fields with buf_block_t::lock. The function
buf_flush_write_block_low() is removed and merged here.
buf_page_init_for_read(): Use static linkage. Initialize the newly
allocated block and acquire the exclusive buf_block_t::lock while not
holding any mutex.
IORequest::IORequest(): Remove the body. We only need to invoke
set_punch_hole() in buf_flush_page() and nowhere else.
buf_page_t::flush_type: Remove. Replaced by IORequest::flush_type.
This field is only used during a fil_io() call.
That function already takes IORequest as a parameter, so we had
better introduce for the rarely changing field.
buf_block_t::init(): Replaces buf_page_init().
buf_page_t::init(): Replaces buf_page_init_low().
buf_block_t::initialise(): Initialise many fields, but
keep the buf_page_t::state(). Both buf_pool_t::validate() and
buf_page_optimistic_get() requires that buf_page_t::in_file()
be protected atomically with buf_page_t::in_page_hash
and buf_page_t::in_LRU_list.
buf_page_optimistic_get(): Now that buf_block_t::mutex
no longer exists, we must check buf_page_t::io_fix()
after acquiring the buf_pool.page_hash lock, to detect
whether buf_page_init_for_read() has been initiated.
We will also check the io_fix() before acquiring hash_lock
in order to avoid unnecessary computation.
The field buf_block_t::modify_clock (protected by buf_block_t::lock)
allows buf_page_optimistic_get() to validate the block.
buf_page_t::real_size: Remove. It was only used while flushing
pages of page_compressed tables.
buf_page_encrypt(): Add an output parameter that allows us ot eliminate
buf_page_t::real_size. Replace a condition with debug assertion.
buf_page_should_punch_hole(): Remove.
buf_dblwr_t::add_to_batch(): Replaces buf_dblwr_add_to_batch().
Add the parameter size (to replace buf_page_t::real_size).
buf_dblwr_t::write_single_page(): Replaces buf_dblwr_write_single_page().
Add the parameter size (to replace buf_page_t::real_size).
fil_system_t::detach(): Replaces fil_space_detach().
Ensure that fil_validate() will not be violated even if
fil_system.mutex is released and reacquired.
fil_node_t::complete_io(): Renamed from fil_node_complete_io().
fil_node_t::close_to_free(): Replaces fil_node_close_to_free().
Avoid invoking fil_node_t::close() because fil_system.n_open
has already been decremented in fil_space_t::detach().
BUF_BLOCK_READY_FOR_USE: Remove. Directly use BUF_BLOCK_MEMORY.
BUF_BLOCK_ZIP_DIRTY: Remove. Directly use BUF_BLOCK_ZIP_PAGE,
and distinguish dirty pages by buf_page_t::oldest_modification().
BUF_BLOCK_POOL_WATCH: Remove. Use BUF_BLOCK_NOT_USED instead.
This state was only being used for buf_page_t that are in
buf_pool.watch.
buf_pool_t::watch[]: Remove pointer indirection.
buf_page_t::in_flush_list: Remove. It was set if and only if
buf_page_t::oldest_modification() is nonzero.
buf_page_decrypt_after_read(), buf_corrupt_page_release(),
buf_page_check_corrupt(): Change the const fil_space_t* parameter
to const fil_node_t& so that we can report the correct file name.
buf_page_monitor(): Declare as an ATTRIBUTE_COLD global function.
buf_page_io_complete(): Split to buf_page_read_complete() and
buf_page_write_complete().
buf_dblwr_t::in_use: Remove.
buf_dblwr_t::buf_block_array: Add IORequest::flush_t.
buf_dblwr_sync_datafiles(): Remove. It was a useless wrapper of
os_aio_wait_until_no_pending_writes().
buf_flush_write_complete(): Declare static, not global.
Add the parameter IORequest::flush_t.
buf_flush_freed_page(): Simplify the code.
recv_sys_t::flush_lru: Renamed from flush_type and changed to bool.
fil_read(), fil_write(): Replaced with direct use of fil_io().
fil_buffering_disabled(): Remove. Check srv_file_flush_method directly.
fil_mutex_enter_and_prepare_for_io(): Return the resolved
fil_space_t* to avoid a duplicated lookup in the caller.
fil_report_invalid_page_access(): Clean up the parameters.
fil_io(): Return fil_io_t, which comprises fil_node_t and error code.
Always invoke fil_space_t::acquire_for_io() and let either the
sync=true caller or fil_aio_callback() invoke
fil_space_t::release_for_io().
fil_aio_callback(): Rewrite to replace buf_page_io_complete().
fil_check_pending_operations(): Remove a parameter, and remove some
redundant lookups.
fil_node_close_to_free(): Wait for n_pending==0. Because we no longer
do an extra lookup of the tablespace between fil_io() and the
completion of the operation, we must give fil_node_t::complete_io() a
chance to decrement the counter.
fil_close_tablespace(): Remove unused parameter trx, and document
that this is only invoked during the error handling of IMPORT TABLESPACE.
row_import_discard_changes(): Merged with the only caller,
row_import_cleanup(). Do not lock up the data dictionary while
invoking fil_close_tablespace().
logs_empty_and_mark_files_at_shutdown(): Do not invoke
fil_close_all_files(), to avoid a !needs_flush assertion failure
on fil_node_t::close().
innodb_shutdown(): Invoke os_aio_free() before fil_close_all_files().
fil_close_all_files(): Invoke fil_flush_file_spaces()
to ensure proper durability.
thread_pool::unbind(): Fix a crash that would occur on Windows
after srv_thread_pool->disable_aio() and os_file_close().
This fix was submitted by Vladislav Vaintroub.
Thanks to Matthias Leich and Axel Schwenke for extensive testing,
Vladislav Vaintroub for helpful comments, and Eugene Kosov for a review.
2020-06-05 12:35:46 +03:00
|
|
|
}
|
2020-10-26 15:59:30 +02:00
|
|
|
return size;
|
MDEV-15053 Reduce buf_pool_t::mutex contention
User-visible changes: The INFORMATION_SCHEMA views INNODB_BUFFER_PAGE
and INNODB_BUFFER_PAGE_LRU will report a dummy value FLUSH_TYPE=0
and will no longer report the PAGE_STATE value READY_FOR_USE.
We will remove some fields from buf_page_t and move much code to
member functions of buf_pool_t and buf_page_t, so that the access
rules of data members can be enforced consistently.
Evicting or adding pages in buf_pool.LRU will remain covered by
buf_pool.mutex.
Evicting or adding pages in buf_pool.page_hash will remain
covered by both buf_pool.mutex and the buf_pool.page_hash X-latch.
After this fix, buf_pool.page_hash lookups can entirely
avoid acquiring buf_pool.mutex, only relying on
buf_pool.hash_lock_get() S-latch.
Similarly, buf_flush_check_neighbors() can will rely solely on
buf_pool.mutex, no buf_pool.page_hash latch at all.
The buf_pool.mutex is rather contended in I/O heavy benchmarks,
especially when the workload does not fit in the buffer pool.
The first attempt to alleviate the contention was the
buf_pool_t::mutex split in
commit 4ed7082eefe56b3e97e0edefb3df76dd7ef5e858
which introduced buf_block_t::mutex, which we are now removing.
Later, multiple instances of buf_pool_t were introduced
in commit c18084f71b02ea707c6461353e6cfc15d7553bc6
and recently removed by us in
commit 1a6f708ec594ac0ae2dd30db926ab07b100fa24b (MDEV-15058).
UNIV_BUF_DEBUG: Remove. This option to enable some buffer pool
related debugging in otherwise non-debug builds has not been used
for years. Instead, we have been using UNIV_DEBUG, which is enabled
in CMAKE_BUILD_TYPE=Debug.
buf_block_t::mutex, buf_pool_t::zip_mutex: Remove. We can mainly rely on
std::atomic and the buf_pool.page_hash latches, and in some cases
depend on buf_pool.mutex or buf_pool.flush_list_mutex just like before.
We must always release buf_block_t::lock before invoking
unfix() or io_unfix(), to prevent a glitch where a block that was
added to the buf_pool.free list would apper X-latched. See
commit c5883debd6ef440a037011c11873b396923e93c5 how this glitch
was finally caught in a debug environment.
We move some buf_pool_t::page_hash specific code from the
ha and hash modules to buf_pool, for improved readability.
buf_pool_t::close(): Assert that all blocks are clean, except
on aborted startup or crash-like shutdown.
buf_pool_t::validate(): No longer attempt to validate
n_flush[] against the number of BUF_IO_WRITE fixed blocks,
because buf_page_t::flush_type no longer exists.
buf_pool_t::watch_set(): Replaces buf_pool_watch_set().
Reduce mutex contention by separating the buf_pool.watch[]
allocation and the insert into buf_pool.page_hash.
buf_pool_t::page_hash_lock<bool exclusive>(): Acquire a
buf_pool.page_hash latch.
Replaces and extends buf_page_hash_lock_s_confirm()
and buf_page_hash_lock_x_confirm().
buf_pool_t::READ_AHEAD_PAGES: Renamed from BUF_READ_AHEAD_PAGES.
buf_pool_t::curr_size, old_size, read_ahead_area, n_pend_reads:
Use Atomic_counter.
buf_pool_t::running_out(): Replaces buf_LRU_buf_pool_running_out().
buf_pool_t::LRU_remove(): Remove a block from the LRU list
and return its predecessor. Incorporates buf_LRU_adjust_hp(),
which was removed.
buf_page_get_gen(): Remove a redundant call of fsp_is_system_temporary(),
for mode == BUF_GET_IF_IN_POOL_OR_WATCH, which is only used by
BTR_DELETE_OP (purge), which is never invoked on temporary tables.
buf_free_from_unzip_LRU_list_batch(): Avoid redundant assignments.
buf_LRU_free_from_unzip_LRU_list(): Simplify the loop condition.
buf_LRU_free_page(): Clarify the function comment.
buf_flush_check_neighbor(), buf_flush_check_neighbors():
Rewrite the construction of the page hash range. We will hold
the buf_pool.mutex for up to buf_pool.read_ahead_area (at most 64)
consecutive lookups of buf_pool.page_hash.
buf_flush_page_and_try_neighbors(): Remove.
Merge to its only callers, and remove redundant operations in
buf_flush_LRU_list_batch().
buf_read_ahead_random(), buf_read_ahead_linear(): Rewrite.
Do not acquire buf_pool.mutex, and iterate directly with page_id_t.
ut_2_power_up(): Remove. my_round_up_to_next_power() is inlined
and avoids any loops.
fil_page_get_prev(), fil_page_get_next(), fil_addr_is_null(): Remove.
buf_flush_page(): Add a fil_space_t* parameter. Minimize the
buf_pool.mutex hold time. buf_pool.n_flush[] is no longer updated
atomically with the io_fix, and we will protect most buf_block_t
fields with buf_block_t::lock. The function
buf_flush_write_block_low() is removed and merged here.
buf_page_init_for_read(): Use static linkage. Initialize the newly
allocated block and acquire the exclusive buf_block_t::lock while not
holding any mutex.
IORequest::IORequest(): Remove the body. We only need to invoke
set_punch_hole() in buf_flush_page() and nowhere else.
buf_page_t::flush_type: Remove. Replaced by IORequest::flush_type.
This field is only used during a fil_io() call.
That function already takes IORequest as a parameter, so we had
better introduce for the rarely changing field.
buf_block_t::init(): Replaces buf_page_init().
buf_page_t::init(): Replaces buf_page_init_low().
buf_block_t::initialise(): Initialise many fields, but
keep the buf_page_t::state(). Both buf_pool_t::validate() and
buf_page_optimistic_get() requires that buf_page_t::in_file()
be protected atomically with buf_page_t::in_page_hash
and buf_page_t::in_LRU_list.
buf_page_optimistic_get(): Now that buf_block_t::mutex
no longer exists, we must check buf_page_t::io_fix()
after acquiring the buf_pool.page_hash lock, to detect
whether buf_page_init_for_read() has been initiated.
We will also check the io_fix() before acquiring hash_lock
in order to avoid unnecessary computation.
The field buf_block_t::modify_clock (protected by buf_block_t::lock)
allows buf_page_optimistic_get() to validate the block.
buf_page_t::real_size: Remove. It was only used while flushing
pages of page_compressed tables.
buf_page_encrypt(): Add an output parameter that allows us ot eliminate
buf_page_t::real_size. Replace a condition with debug assertion.
buf_page_should_punch_hole(): Remove.
buf_dblwr_t::add_to_batch(): Replaces buf_dblwr_add_to_batch().
Add the parameter size (to replace buf_page_t::real_size).
buf_dblwr_t::write_single_page(): Replaces buf_dblwr_write_single_page().
Add the parameter size (to replace buf_page_t::real_size).
fil_system_t::detach(): Replaces fil_space_detach().
Ensure that fil_validate() will not be violated even if
fil_system.mutex is released and reacquired.
fil_node_t::complete_io(): Renamed from fil_node_complete_io().
fil_node_t::close_to_free(): Replaces fil_node_close_to_free().
Avoid invoking fil_node_t::close() because fil_system.n_open
has already been decremented in fil_space_t::detach().
BUF_BLOCK_READY_FOR_USE: Remove. Directly use BUF_BLOCK_MEMORY.
BUF_BLOCK_ZIP_DIRTY: Remove. Directly use BUF_BLOCK_ZIP_PAGE,
and distinguish dirty pages by buf_page_t::oldest_modification().
BUF_BLOCK_POOL_WATCH: Remove. Use BUF_BLOCK_NOT_USED instead.
This state was only being used for buf_page_t that are in
buf_pool.watch.
buf_pool_t::watch[]: Remove pointer indirection.
buf_page_t::in_flush_list: Remove. It was set if and only if
buf_page_t::oldest_modification() is nonzero.
buf_page_decrypt_after_read(), buf_corrupt_page_release(),
buf_page_check_corrupt(): Change the const fil_space_t* parameter
to const fil_node_t& so that we can report the correct file name.
buf_page_monitor(): Declare as an ATTRIBUTE_COLD global function.
buf_page_io_complete(): Split to buf_page_read_complete() and
buf_page_write_complete().
buf_dblwr_t::in_use: Remove.
buf_dblwr_t::buf_block_array: Add IORequest::flush_t.
buf_dblwr_sync_datafiles(): Remove. It was a useless wrapper of
os_aio_wait_until_no_pending_writes().
buf_flush_write_complete(): Declare static, not global.
Add the parameter IORequest::flush_t.
buf_flush_freed_page(): Simplify the code.
recv_sys_t::flush_lru: Renamed from flush_type and changed to bool.
fil_read(), fil_write(): Replaced with direct use of fil_io().
fil_buffering_disabled(): Remove. Check srv_file_flush_method directly.
fil_mutex_enter_and_prepare_for_io(): Return the resolved
fil_space_t* to avoid a duplicated lookup in the caller.
fil_report_invalid_page_access(): Clean up the parameters.
fil_io(): Return fil_io_t, which comprises fil_node_t and error code.
Always invoke fil_space_t::acquire_for_io() and let either the
sync=true caller or fil_aio_callback() invoke
fil_space_t::release_for_io().
fil_aio_callback(): Rewrite to replace buf_page_io_complete().
fil_check_pending_operations(): Remove a parameter, and remove some
redundant lookups.
fil_node_close_to_free(): Wait for n_pending==0. Because we no longer
do an extra lookup of the tablespace between fil_io() and the
completion of the operation, we must give fil_node_t::complete_io() a
chance to decrement the counter.
fil_close_tablespace(): Remove unused parameter trx, and document
that this is only invoked during the error handling of IMPORT TABLESPACE.
row_import_discard_changes(): Merged with the only caller,
row_import_cleanup(). Do not lock up the data dictionary while
invoking fil_close_tablespace().
logs_empty_and_mark_files_at_shutdown(): Do not invoke
fil_close_all_files(), to avoid a !needs_flush assertion failure
on fil_node_t::close().
innodb_shutdown(): Invoke os_aio_free() before fil_close_all_files().
fil_close_all_files(): Invoke fil_flush_file_spaces()
to ensure proper durability.
thread_pool::unbind(): Fix a crash that would occur on Windows
after srv_thread_pool->disable_aio() and os_file_close().
This fix was submitted by Vladislav Vaintroub.
Thanks to Matthias Leich and Axel Schwenke for extensive testing,
Vladislav Vaintroub for helpful comments, and Eugene Kosov for a review.
2020-06-05 12:35:46 +03:00
|
|
|
}
|
|
|
|
|
|
2016-09-06 09:43:16 +03:00
|
|
|
#include "fil0crypt.h"
|
|
|
|
|
|
2014-02-26 19:11:54 +01:00
|
|
|
/*******************************************************************//**
|
|
|
|
|
Assigns a new space id for a new single-table tablespace. This works simply by
|
|
|
|
|
incrementing the global counter. If 4 billion id's is not enough, we may need
|
|
|
|
|
to recycle id's.
|
2016-08-12 11:17:45 +03:00
|
|
|
@return true if assigned, false if not */
|
|
|
|
|
bool
|
2014-02-26 19:11:54 +01:00
|
|
|
fil_assign_new_space_id(
|
|
|
|
|
/*====================*/
|
|
|
|
|
ulint* space_id); /*!< in/out: space id */
|
2016-08-12 11:17:45 +03:00
|
|
|
|
|
|
|
|
/** Frees a space object from the tablespace memory cache.
|
|
|
|
|
Closes the files in the chain but does not delete them.
|
|
|
|
|
There must not be any pending i/o's or flushes on the files.
|
|
|
|
|
@param[in] id tablespace identifier
|
|
|
|
|
@param[in] x_latched whether the caller holds X-mode space->latch
|
|
|
|
|
@return true if success */
|
|
|
|
|
bool
|
|
|
|
|
fil_space_free(
|
|
|
|
|
ulint id,
|
|
|
|
|
bool x_latched);
|
|
|
|
|
|
MDEV-11556 InnoDB redo log apply fails to adjust data file sizes
fil_space_t::recv_size: New member: recovered tablespace size in pages;
0 if no size change was read from the redo log,
or if the size change was implemented.
fil_space_set_recv_size(): New function for setting space->recv_size.
innodb_data_file_size_debug: A debug parameter for setting the system
tablespace size in recovery even when the redo log does not contain
any size changes. It is hard to write a small test case that would
cause the system tablespace to be extended at the critical moment.
recv_parse_log_rec(): Note those tablespaces whose size is being changed
by the redo log, by invoking fil_space_set_recv_size().
innobase_init(): Correct an error message, and do not require a larger
innodb_buffer_pool_size when starting up with a smaller innodb_page_size.
innobase_start_or_create_for_mysql(): Allow startup with any initial
size of the ibdata1 file if the autoextend attribute is set. Require
the minimum size of fixed-size system tablespaces to be 640 pages,
not 10 megabytes. Implement innodb_data_file_size_debug.
open_or_create_data_files(): Round the system tablespace size down
to pages, not to full megabytes, (Our test truncates the system
tablespace to more than 800 pages with innodb_page_size=4k.
InnoDB should not imagine that it was truncated to 768 pages
and then overwrite good pages in the tablespace.)
fil_flush_low(): Refactored from fil_flush().
fil_space_extend_must_retry(): Refactored from
fil_extend_space_to_desired_size().
fil_mutex_enter_and_prepare_for_io(): Extend the tablespace if
fil_space_set_recv_size() was called.
The test case has been successfully run with all the
innodb_page_size values 4k, 8k, 16k, 32k, 64k.
2016-12-28 12:05:43 +02:00
|
|
|
/** Set the recovered size of a tablespace in pages.
|
2020-09-07 14:23:05 +05:30
|
|
|
@param id tablespace ID
|
|
|
|
|
@param size recovered size in pages
|
|
|
|
|
@param flags tablespace flags */
|
2020-10-15 16:28:19 +03:00
|
|
|
void fil_space_set_recv_size_and_flags(ulint id, uint32_t size,
|
|
|
|
|
uint32_t flags);
|
2020-09-07 14:23:05 +05:30
|
|
|
|
2014-02-26 19:11:54 +01:00
|
|
|
/*******************************************************************//**
|
|
|
|
|
Sets the max tablespace id counter if the given number is bigger than the
|
|
|
|
|
previous value. */
|
|
|
|
|
void
|
|
|
|
|
fil_set_max_space_id_if_bigger(
|
|
|
|
|
/*===========================*/
|
|
|
|
|
ulint max_id);/*!< in: maximum known id */
|
2016-09-06 09:43:16 +03:00
|
|
|
|
|
|
|
|
/** Write the flushed LSN to the page header of the first page in the
|
|
|
|
|
system tablespace.
|
|
|
|
|
@param[in] lsn flushed LSN
|
|
|
|
|
@return DB_SUCCESS or error number */
|
|
|
|
|
dberr_t
|
|
|
|
|
fil_write_flushed_lsn(
|
|
|
|
|
lsn_t lsn)
|
|
|
|
|
MY_ATTRIBUTE((warn_unused_result));
|
2016-08-12 11:17:45 +03:00
|
|
|
|
MDEV-25506 (3 of 3): Do not delete .ibd files before commit
This is a complete rewrite of DROP TABLE, also as part of other DDL,
such as ALTER TABLE, CREATE TABLE...SELECT, TRUNCATE TABLE.
The background DROP TABLE queue hack is removed.
If a transaction needs to drop and create a table by the same name
(like TRUNCATE TABLE does), it must first rename the table to an
internal #sql-ib name. No committed version of the data dictionary
will include any #sql-ib tables, because whenever a transaction
renames a table to a #sql-ib name, it will also drop that table.
Either the rename will be rolled back, or the drop will be committed.
Data files will be unlinked after the transaction has been committed
and a FILE_RENAME record has been durably written. The file will
actually be deleted when the detached file handle returned by
fil_delete_tablespace() will be closed, after the latches have been
released. It is possible that a purge of the delete of the SYS_INDEXES
record for the clustered index will execute fil_delete_tablespace()
concurrently with the DDL transaction. In that case, the thread that
arrives later will wait for the other thread to finish.
HTON_TRUNCATE_REQUIRES_EXCLUSIVE_USE: A new handler flag.
ha_innobase::truncate() now requires that all other references to
the table be released in advance. This was implemented by Monty.
ha_innobase::delete_table(): If CREATE TABLE..SELECT is detected,
we will "hijack" the current transaction, drop the table in
the current transaction and commit the current transaction.
This essentially fixes MDEV-21602. There is a FIXME comment about
making the check less failure-prone.
ha_innobase::truncate(), ha_innobase::delete_table():
Implement a fast path for temporary tables. We will no longer allow
temporary tables to use the adaptive hash index.
dict_table_t::mdl_name: The original table name for the purpose of
acquiring MDL in purge, to prevent a race condition between a
DDL transaction that is dropping a table, and purge processing
undo log records of DML that had executed before the DDL operation.
For #sql-backup- tables during ALTER TABLE...ALGORITHM=COPY, the
dict_table_t::mdl_name will differ from dict_table_t::name.
dict_table_t::parse_name(): Use mdl_name instead of name.
dict_table_rename_in_cache(): Update mdl_name.
For the internal FTS_ tables of FULLTEXT INDEX, purge would
acquire MDL on the FTS_ table name, but not on the main table,
and therefore it would be able to run concurrently with a
DDL transaction that is dropping the table. Previously, the
DROP TABLE queue hack prevented a race between purge and DDL.
For now, we introduce purge_sys.stop_FTS() to prevent purge from
opening any table, while a DDL transaction that may drop FTS_
tables is in progress. The function fts_lock_table(), which will
be invoked before the dictionary is locked, will wait for
purge to release any table handles.
trx_t::drop_table_statistics(): Drop statistics for the table.
This replaces dict_stats_drop_index(). We will drop or rename
persistent statistics atomically as part of DDL transactions.
On lock conflict for dropping statistics, we will fail instantly
with DB_LOCK_WAIT_TIMEOUT, because we will be holding the
exclusive data dictionary latch.
trx_t::commit_cleanup(): Separated from trx_t::commit_in_memory().
Relax an assertion around fts_commit() and allow DB_LOCK_WAIT_TIMEOUT
in addition to DB_DUPLICATE_KEY. The call to fts_commit() is
entirely misplaced here and may obviously break the consistency
of transactions that affect FULLTEXT INDEX. It needs to be fixed
separately.
dict_table_t::n_foreign_key_checks_running: Remove (MDEV-21175).
The counter was a work-around for missing meta-data locking (MDL)
on the SQL layer, and not really needed in MariaDB.
ER_TABLE_IN_FK_CHECK: Replaced with ER_UNUSED_28.
HA_ERR_TABLE_IN_FK_CHECK: Remove.
row_ins_check_foreign_constraints(): Do not acquire
dict_sys.latch either. The SQL-layer MDL will protect us.
This was reviewed by Thirunarayanan Balathandayuthapani
and tested by Matthias Leich.
2021-06-09 17:02:55 +03:00
|
|
|
MY_ATTRIBUTE((warn_unused_result))
|
MDEV-13328 ALTER TABLE…DISCARD TABLESPACE takes a lot of time
With a big buffer pool that contains many data pages,
DISCARD TABLESPACE took a long time, because it would scan the
entire buffer pool to remove any pages that belong to the tablespace.
With a large buffer pool, this would take a lot of time, especially
when the table-to-discard is empty.
The minimum amount of work that DISCARD TABLESPACE must do is to
remove the pages of the to-be-discarded table from the
buf_pool->flush_list because any writes to the data file must be
prevented before the file is deleted.
If DISCARD TABLESPACE does not evict the pages from the buffer pool,
then IMPORT TABLESPACE must do it, because we must prevent pre-DISCARD,
not-yet-evicted pages from being mistaken for pages of the imported
tablespace.
It would not be a useful fix to simply move the buffer pool scan to
the IMPORT TABLESPACE step. What we can do is to actively evict those
pages that could be mistaken for imported pages. In this way, when
importing a small table into a big buffer pool, the import should
still run relatively fast.
Import is bypassing the buffer pool when reading pages for the
adjustment phase. In the adjustment phase, if a page exists in
the buffer pool, we could replace it with the page from the imported
file. Unfortunately I did not get this to work properly, so instead
we will simply evict any matching page from the buffer pool.
buf_page_get_gen(): Implement BUF_EVICT_IF_IN_POOL, a new mode
where the requested page will be evicted if it is found. There
must be no unwritten changes for the page.
buf_remove_t: Remove. Instead, use trx!=NULL to signify that a write
to file is desired, and use a separate parameter bool drop_ahi.
buf_LRU_flush_or_remove_pages(), fil_delete_tablespace():
Replace buf_remove_t.
buf_LRU_remove_pages(), buf_LRU_remove_all_pages(): Remove.
PageConverter::m_mtr: A dummy mini-transaction buffer
PageConverter::PageConverter(): Complete the member initialization list.
PageConverter::operator()(): Evict any 'shadow' pages from the
buffer pool so that pre-existing (garbage) pages cannot be mistaken
for pages that exist in the being-imported file.
row_discard_tablespace(): Remove a bogus comment that seems to
refer to IMPORT TABLESPACE, not DISCARD TABLESPACE.
2017-11-02 22:38:37 +02:00
|
|
|
/** Delete a tablespace and associated .ibd file.
|
MDEV-25506 (3 of 3): Do not delete .ibd files before commit
This is a complete rewrite of DROP TABLE, also as part of other DDL,
such as ALTER TABLE, CREATE TABLE...SELECT, TRUNCATE TABLE.
The background DROP TABLE queue hack is removed.
If a transaction needs to drop and create a table by the same name
(like TRUNCATE TABLE does), it must first rename the table to an
internal #sql-ib name. No committed version of the data dictionary
will include any #sql-ib tables, because whenever a transaction
renames a table to a #sql-ib name, it will also drop that table.
Either the rename will be rolled back, or the drop will be committed.
Data files will be unlinked after the transaction has been committed
and a FILE_RENAME record has been durably written. The file will
actually be deleted when the detached file handle returned by
fil_delete_tablespace() will be closed, after the latches have been
released. It is possible that a purge of the delete of the SYS_INDEXES
record for the clustered index will execute fil_delete_tablespace()
concurrently with the DDL transaction. In that case, the thread that
arrives later will wait for the other thread to finish.
HTON_TRUNCATE_REQUIRES_EXCLUSIVE_USE: A new handler flag.
ha_innobase::truncate() now requires that all other references to
the table be released in advance. This was implemented by Monty.
ha_innobase::delete_table(): If CREATE TABLE..SELECT is detected,
we will "hijack" the current transaction, drop the table in
the current transaction and commit the current transaction.
This essentially fixes MDEV-21602. There is a FIXME comment about
making the check less failure-prone.
ha_innobase::truncate(), ha_innobase::delete_table():
Implement a fast path for temporary tables. We will no longer allow
temporary tables to use the adaptive hash index.
dict_table_t::mdl_name: The original table name for the purpose of
acquiring MDL in purge, to prevent a race condition between a
DDL transaction that is dropping a table, and purge processing
undo log records of DML that had executed before the DDL operation.
For #sql-backup- tables during ALTER TABLE...ALGORITHM=COPY, the
dict_table_t::mdl_name will differ from dict_table_t::name.
dict_table_t::parse_name(): Use mdl_name instead of name.
dict_table_rename_in_cache(): Update mdl_name.
For the internal FTS_ tables of FULLTEXT INDEX, purge would
acquire MDL on the FTS_ table name, but not on the main table,
and therefore it would be able to run concurrently with a
DDL transaction that is dropping the table. Previously, the
DROP TABLE queue hack prevented a race between purge and DDL.
For now, we introduce purge_sys.stop_FTS() to prevent purge from
opening any table, while a DDL transaction that may drop FTS_
tables is in progress. The function fts_lock_table(), which will
be invoked before the dictionary is locked, will wait for
purge to release any table handles.
trx_t::drop_table_statistics(): Drop statistics for the table.
This replaces dict_stats_drop_index(). We will drop or rename
persistent statistics atomically as part of DDL transactions.
On lock conflict for dropping statistics, we will fail instantly
with DB_LOCK_WAIT_TIMEOUT, because we will be holding the
exclusive data dictionary latch.
trx_t::commit_cleanup(): Separated from trx_t::commit_in_memory().
Relax an assertion around fts_commit() and allow DB_LOCK_WAIT_TIMEOUT
in addition to DB_DUPLICATE_KEY. The call to fts_commit() is
entirely misplaced here and may obviously break the consistency
of transactions that affect FULLTEXT INDEX. It needs to be fixed
separately.
dict_table_t::n_foreign_key_checks_running: Remove (MDEV-21175).
The counter was a work-around for missing meta-data locking (MDL)
on the SQL layer, and not really needed in MariaDB.
ER_TABLE_IN_FK_CHECK: Replaced with ER_UNUSED_28.
HA_ERR_TABLE_IN_FK_CHECK: Remove.
row_ins_check_foreign_constraints(): Do not acquire
dict_sys.latch either. The SQL-layer MDL will protect us.
This was reviewed by Thirunarayanan Balathandayuthapani
and tested by Matthias Leich.
2021-06-09 17:02:55 +03:00
|
|
|
@param id tablespace identifier
|
|
|
|
|
@return detached file handle (to be closed by the caller)
|
|
|
|
|
@return OS_FILE_CLOSED if no file existed */
|
|
|
|
|
pfs_os_file_t fil_delete_tablespace(ulint id);
|
2016-08-12 11:17:45 +03:00
|
|
|
|
MDEV-15053 Reduce buf_pool_t::mutex contention
User-visible changes: The INFORMATION_SCHEMA views INNODB_BUFFER_PAGE
and INNODB_BUFFER_PAGE_LRU will report a dummy value FLUSH_TYPE=0
and will no longer report the PAGE_STATE value READY_FOR_USE.
We will remove some fields from buf_page_t and move much code to
member functions of buf_pool_t and buf_page_t, so that the access
rules of data members can be enforced consistently.
Evicting or adding pages in buf_pool.LRU will remain covered by
buf_pool.mutex.
Evicting or adding pages in buf_pool.page_hash will remain
covered by both buf_pool.mutex and the buf_pool.page_hash X-latch.
After this fix, buf_pool.page_hash lookups can entirely
avoid acquiring buf_pool.mutex, only relying on
buf_pool.hash_lock_get() S-latch.
Similarly, buf_flush_check_neighbors() can will rely solely on
buf_pool.mutex, no buf_pool.page_hash latch at all.
The buf_pool.mutex is rather contended in I/O heavy benchmarks,
especially when the workload does not fit in the buffer pool.
The first attempt to alleviate the contention was the
buf_pool_t::mutex split in
commit 4ed7082eefe56b3e97e0edefb3df76dd7ef5e858
which introduced buf_block_t::mutex, which we are now removing.
Later, multiple instances of buf_pool_t were introduced
in commit c18084f71b02ea707c6461353e6cfc15d7553bc6
and recently removed by us in
commit 1a6f708ec594ac0ae2dd30db926ab07b100fa24b (MDEV-15058).
UNIV_BUF_DEBUG: Remove. This option to enable some buffer pool
related debugging in otherwise non-debug builds has not been used
for years. Instead, we have been using UNIV_DEBUG, which is enabled
in CMAKE_BUILD_TYPE=Debug.
buf_block_t::mutex, buf_pool_t::zip_mutex: Remove. We can mainly rely on
std::atomic and the buf_pool.page_hash latches, and in some cases
depend on buf_pool.mutex or buf_pool.flush_list_mutex just like before.
We must always release buf_block_t::lock before invoking
unfix() or io_unfix(), to prevent a glitch where a block that was
added to the buf_pool.free list would apper X-latched. See
commit c5883debd6ef440a037011c11873b396923e93c5 how this glitch
was finally caught in a debug environment.
We move some buf_pool_t::page_hash specific code from the
ha and hash modules to buf_pool, for improved readability.
buf_pool_t::close(): Assert that all blocks are clean, except
on aborted startup or crash-like shutdown.
buf_pool_t::validate(): No longer attempt to validate
n_flush[] against the number of BUF_IO_WRITE fixed blocks,
because buf_page_t::flush_type no longer exists.
buf_pool_t::watch_set(): Replaces buf_pool_watch_set().
Reduce mutex contention by separating the buf_pool.watch[]
allocation and the insert into buf_pool.page_hash.
buf_pool_t::page_hash_lock<bool exclusive>(): Acquire a
buf_pool.page_hash latch.
Replaces and extends buf_page_hash_lock_s_confirm()
and buf_page_hash_lock_x_confirm().
buf_pool_t::READ_AHEAD_PAGES: Renamed from BUF_READ_AHEAD_PAGES.
buf_pool_t::curr_size, old_size, read_ahead_area, n_pend_reads:
Use Atomic_counter.
buf_pool_t::running_out(): Replaces buf_LRU_buf_pool_running_out().
buf_pool_t::LRU_remove(): Remove a block from the LRU list
and return its predecessor. Incorporates buf_LRU_adjust_hp(),
which was removed.
buf_page_get_gen(): Remove a redundant call of fsp_is_system_temporary(),
for mode == BUF_GET_IF_IN_POOL_OR_WATCH, which is only used by
BTR_DELETE_OP (purge), which is never invoked on temporary tables.
buf_free_from_unzip_LRU_list_batch(): Avoid redundant assignments.
buf_LRU_free_from_unzip_LRU_list(): Simplify the loop condition.
buf_LRU_free_page(): Clarify the function comment.
buf_flush_check_neighbor(), buf_flush_check_neighbors():
Rewrite the construction of the page hash range. We will hold
the buf_pool.mutex for up to buf_pool.read_ahead_area (at most 64)
consecutive lookups of buf_pool.page_hash.
buf_flush_page_and_try_neighbors(): Remove.
Merge to its only callers, and remove redundant operations in
buf_flush_LRU_list_batch().
buf_read_ahead_random(), buf_read_ahead_linear(): Rewrite.
Do not acquire buf_pool.mutex, and iterate directly with page_id_t.
ut_2_power_up(): Remove. my_round_up_to_next_power() is inlined
and avoids any loops.
fil_page_get_prev(), fil_page_get_next(), fil_addr_is_null(): Remove.
buf_flush_page(): Add a fil_space_t* parameter. Minimize the
buf_pool.mutex hold time. buf_pool.n_flush[] is no longer updated
atomically with the io_fix, and we will protect most buf_block_t
fields with buf_block_t::lock. The function
buf_flush_write_block_low() is removed and merged here.
buf_page_init_for_read(): Use static linkage. Initialize the newly
allocated block and acquire the exclusive buf_block_t::lock while not
holding any mutex.
IORequest::IORequest(): Remove the body. We only need to invoke
set_punch_hole() in buf_flush_page() and nowhere else.
buf_page_t::flush_type: Remove. Replaced by IORequest::flush_type.
This field is only used during a fil_io() call.
That function already takes IORequest as a parameter, so we had
better introduce for the rarely changing field.
buf_block_t::init(): Replaces buf_page_init().
buf_page_t::init(): Replaces buf_page_init_low().
buf_block_t::initialise(): Initialise many fields, but
keep the buf_page_t::state(). Both buf_pool_t::validate() and
buf_page_optimistic_get() requires that buf_page_t::in_file()
be protected atomically with buf_page_t::in_page_hash
and buf_page_t::in_LRU_list.
buf_page_optimistic_get(): Now that buf_block_t::mutex
no longer exists, we must check buf_page_t::io_fix()
after acquiring the buf_pool.page_hash lock, to detect
whether buf_page_init_for_read() has been initiated.
We will also check the io_fix() before acquiring hash_lock
in order to avoid unnecessary computation.
The field buf_block_t::modify_clock (protected by buf_block_t::lock)
allows buf_page_optimistic_get() to validate the block.
buf_page_t::real_size: Remove. It was only used while flushing
pages of page_compressed tables.
buf_page_encrypt(): Add an output parameter that allows us ot eliminate
buf_page_t::real_size. Replace a condition with debug assertion.
buf_page_should_punch_hole(): Remove.
buf_dblwr_t::add_to_batch(): Replaces buf_dblwr_add_to_batch().
Add the parameter size (to replace buf_page_t::real_size).
buf_dblwr_t::write_single_page(): Replaces buf_dblwr_write_single_page().
Add the parameter size (to replace buf_page_t::real_size).
fil_system_t::detach(): Replaces fil_space_detach().
Ensure that fil_validate() will not be violated even if
fil_system.mutex is released and reacquired.
fil_node_t::complete_io(): Renamed from fil_node_complete_io().
fil_node_t::close_to_free(): Replaces fil_node_close_to_free().
Avoid invoking fil_node_t::close() because fil_system.n_open
has already been decremented in fil_space_t::detach().
BUF_BLOCK_READY_FOR_USE: Remove. Directly use BUF_BLOCK_MEMORY.
BUF_BLOCK_ZIP_DIRTY: Remove. Directly use BUF_BLOCK_ZIP_PAGE,
and distinguish dirty pages by buf_page_t::oldest_modification().
BUF_BLOCK_POOL_WATCH: Remove. Use BUF_BLOCK_NOT_USED instead.
This state was only being used for buf_page_t that are in
buf_pool.watch.
buf_pool_t::watch[]: Remove pointer indirection.
buf_page_t::in_flush_list: Remove. It was set if and only if
buf_page_t::oldest_modification() is nonzero.
buf_page_decrypt_after_read(), buf_corrupt_page_release(),
buf_page_check_corrupt(): Change the const fil_space_t* parameter
to const fil_node_t& so that we can report the correct file name.
buf_page_monitor(): Declare as an ATTRIBUTE_COLD global function.
buf_page_io_complete(): Split to buf_page_read_complete() and
buf_page_write_complete().
buf_dblwr_t::in_use: Remove.
buf_dblwr_t::buf_block_array: Add IORequest::flush_t.
buf_dblwr_sync_datafiles(): Remove. It was a useless wrapper of
os_aio_wait_until_no_pending_writes().
buf_flush_write_complete(): Declare static, not global.
Add the parameter IORequest::flush_t.
buf_flush_freed_page(): Simplify the code.
recv_sys_t::flush_lru: Renamed from flush_type and changed to bool.
fil_read(), fil_write(): Replaced with direct use of fil_io().
fil_buffering_disabled(): Remove. Check srv_file_flush_method directly.
fil_mutex_enter_and_prepare_for_io(): Return the resolved
fil_space_t* to avoid a duplicated lookup in the caller.
fil_report_invalid_page_access(): Clean up the parameters.
fil_io(): Return fil_io_t, which comprises fil_node_t and error code.
Always invoke fil_space_t::acquire_for_io() and let either the
sync=true caller or fil_aio_callback() invoke
fil_space_t::release_for_io().
fil_aio_callback(): Rewrite to replace buf_page_io_complete().
fil_check_pending_operations(): Remove a parameter, and remove some
redundant lookups.
fil_node_close_to_free(): Wait for n_pending==0. Because we no longer
do an extra lookup of the tablespace between fil_io() and the
completion of the operation, we must give fil_node_t::complete_io() a
chance to decrement the counter.
fil_close_tablespace(): Remove unused parameter trx, and document
that this is only invoked during the error handling of IMPORT TABLESPACE.
row_import_discard_changes(): Merged with the only caller,
row_import_cleanup(). Do not lock up the data dictionary while
invoking fil_close_tablespace().
logs_empty_and_mark_files_at_shutdown(): Do not invoke
fil_close_all_files(), to avoid a !needs_flush assertion failure
on fil_node_t::close().
innodb_shutdown(): Invoke os_aio_free() before fil_close_all_files().
fil_close_all_files(): Invoke fil_flush_file_spaces()
to ensure proper durability.
thread_pool::unbind(): Fix a crash that would occur on Windows
after srv_thread_pool->disable_aio() and os_file_close().
This fix was submitted by Vladislav Vaintroub.
Thanks to Matthias Leich and Axel Schwenke for extensive testing,
Vladislav Vaintroub for helpful comments, and Eugene Kosov for a review.
2020-06-05 12:35:46 +03:00
|
|
|
/** Close a single-table tablespace on failed IMPORT TABLESPACE.
|
|
|
|
|
The tablespace must be cached in the memory cache.
|
|
|
|
|
Free all pages used by the tablespace. */
|
|
|
|
|
void fil_close_tablespace(ulint id);
|
2016-12-30 15:04:10 +02:00
|
|
|
|
2014-02-26 19:11:54 +01:00
|
|
|
/*******************************************************************//**
|
2016-08-12 11:17:45 +03:00
|
|
|
Allocates and builds a file name from a path, a table or tablespace name
|
|
|
|
|
and a suffix. The string must be freed by caller with ut_free().
|
2020-12-20 21:07:38 +02:00
|
|
|
@param[in] path NULL or the directory path or the full path and filename.
|
MDEV-25312 Replace fil_space_t::name with fil_space_t::name()
A consistency check for fil_space_t::name is causing recovery failures
in MDEV-25180 (Atomic ALTER TABLE). So, we'd better remove that field
altogether.
fil_space_t::name was more or less a copy of dict_table_t::name
(except for some special cases), and it was not being used for
anything useful.
There used to be a name_hash, but it had been removed already in
commit a75dbfd7183cc96680f3e3e684fd36500dac8158 (MDEV-12266).
We will also remove os_normalize_path(), OS_PATH_SEPARATOR,
OS_PATH_SEPATOR_ALT. On Microsoft Windows, we will treat \ and /
roughly in the same way. The intention is that for per-table
tablespaces, the filenames will always follow the pattern
prefix/databasename/tablename.ibd. (Any \ in the prefix must not
be converted.)
ut_basename_noext(): Remove (unused function).
read_link_file(): Replaces RemoteDatafile::read_link_file().
We will ensure that the last two path component separators are
forward slashes (converting up to 2 trailing backslashes on
Microsoft Windows), so that everywhere else we can
assume that data file names end in "/databasename/tablename.ibd".
Note: On Microsoft Windows, path names that start with \\?\ must
not contain / as path component separators. Previously, such paths
did work in the DATA DIRECTORY argument of InnoDB tables.
Reviewed by: Vladislav Vaintroub
2021-04-07 18:01:13 +03:00
|
|
|
@param[in] name {} if path is full, or Table/Tablespace name
|
|
|
|
|
@param[in] ext the file extension to use
|
|
|
|
|
@param[in] trim_name true if the last name on the path should be trimmed.
|
2016-08-12 11:17:45 +03:00
|
|
|
@return own: file name */
|
MDEV-25312 Replace fil_space_t::name with fil_space_t::name()
A consistency check for fil_space_t::name is causing recovery failures
in MDEV-25180 (Atomic ALTER TABLE). So, we'd better remove that field
altogether.
fil_space_t::name was more or less a copy of dict_table_t::name
(except for some special cases), and it was not being used for
anything useful.
There used to be a name_hash, but it had been removed already in
commit a75dbfd7183cc96680f3e3e684fd36500dac8158 (MDEV-12266).
We will also remove os_normalize_path(), OS_PATH_SEPARATOR,
OS_PATH_SEPATOR_ALT. On Microsoft Windows, we will treat \ and /
roughly in the same way. The intention is that for per-table
tablespaces, the filenames will always follow the pattern
prefix/databasename/tablename.ibd. (Any \ in the prefix must not
be converted.)
ut_basename_noext(): Remove (unused function).
read_link_file(): Replaces RemoteDatafile::read_link_file().
We will ensure that the last two path component separators are
forward slashes (converting up to 2 trailing backslashes on
Microsoft Windows), so that everywhere else we can
assume that data file names end in "/databasename/tablename.ibd".
Note: On Microsoft Windows, path names that start with \\?\ must
not contain / as path component separators. Previously, such paths
did work in the DATA DIRECTORY argument of InnoDB tables.
Reviewed by: Vladislav Vaintroub
2021-04-07 18:01:13 +03:00
|
|
|
char* fil_make_filepath(const char *path, const fil_space_t::name_type &name,
|
|
|
|
|
ib_extention ext, bool trim_name);
|
|
|
|
|
|
|
|
|
|
char *fil_make_filepath(const char* path, const table_name_t name,
|
2021-05-20 14:58:25 +03:00
|
|
|
ib_extention suffix, bool strip_name);
|
2015-08-14 11:09:06 +03:00
|
|
|
|
2017-01-17 11:37:49 +02:00
|
|
|
/** Create a tablespace file.
|
2016-08-12 11:17:45 +03:00
|
|
|
@param[in] space_id Tablespace ID
|
|
|
|
|
@param[in] name Tablespace name in dbname/tablename format.
|
|
|
|
|
@param[in] path Path and filename of the datafile to create.
|
|
|
|
|
@param[in] flags Tablespace flags
|
|
|
|
|
@param[in] size Initial size of the tablespace file in pages,
|
|
|
|
|
must be >= FIL_IBD_FILE_INITIAL_SIZE
|
2017-01-17 11:37:49 +02:00
|
|
|
@param[in] mode MariaDB encryption mode
|
|
|
|
|
@param[in] key_id MariaDB encryption key_id
|
2018-03-22 14:17:43 +02:00
|
|
|
@param[out] err DB_SUCCESS or error code
|
|
|
|
|
@return the created tablespace
|
|
|
|
|
@retval NULL on error */
|
|
|
|
|
fil_space_t*
|
2016-08-12 11:17:45 +03:00
|
|
|
fil_ibd_create(
|
|
|
|
|
ulint space_id,
|
MDEV-25312 Replace fil_space_t::name with fil_space_t::name()
A consistency check for fil_space_t::name is causing recovery failures
in MDEV-25180 (Atomic ALTER TABLE). So, we'd better remove that field
altogether.
fil_space_t::name was more or less a copy of dict_table_t::name
(except for some special cases), and it was not being used for
anything useful.
There used to be a name_hash, but it had been removed already in
commit a75dbfd7183cc96680f3e3e684fd36500dac8158 (MDEV-12266).
We will also remove os_normalize_path(), OS_PATH_SEPARATOR,
OS_PATH_SEPATOR_ALT. On Microsoft Windows, we will treat \ and /
roughly in the same way. The intention is that for per-table
tablespaces, the filenames will always follow the pattern
prefix/databasename/tablename.ibd. (Any \ in the prefix must not
be converted.)
ut_basename_noext(): Remove (unused function).
read_link_file(): Replaces RemoteDatafile::read_link_file().
We will ensure that the last two path component separators are
forward slashes (converting up to 2 trailing backslashes on
Microsoft Windows), so that everywhere else we can
assume that data file names end in "/databasename/tablename.ibd".
Note: On Microsoft Windows, path names that start with \\?\ must
not contain / as path component separators. Previously, such paths
did work in the DATA DIRECTORY argument of InnoDB tables.
Reviewed by: Vladislav Vaintroub
2021-04-07 18:01:13 +03:00
|
|
|
const table_name_t name,
|
2016-08-12 11:17:45 +03:00
|
|
|
const char* path,
|
|
|
|
|
ulint flags,
|
2020-10-15 16:28:19 +03:00
|
|
|
uint32_t size,
|
2017-01-17 11:37:49 +02:00
|
|
|
fil_encryption_t mode,
|
2018-03-22 14:17:43 +02:00
|
|
|
uint32_t key_id,
|
|
|
|
|
dberr_t* err)
|
MDEV-25312 Replace fil_space_t::name with fil_space_t::name()
A consistency check for fil_space_t::name is causing recovery failures
in MDEV-25180 (Atomic ALTER TABLE). So, we'd better remove that field
altogether.
fil_space_t::name was more or less a copy of dict_table_t::name
(except for some special cases), and it was not being used for
anything useful.
There used to be a name_hash, but it had been removed already in
commit a75dbfd7183cc96680f3e3e684fd36500dac8158 (MDEV-12266).
We will also remove os_normalize_path(), OS_PATH_SEPARATOR,
OS_PATH_SEPATOR_ALT. On Microsoft Windows, we will treat \ and /
roughly in the same way. The intention is that for per-table
tablespaces, the filenames will always follow the pattern
prefix/databasename/tablename.ibd. (Any \ in the prefix must not
be converted.)
ut_basename_noext(): Remove (unused function).
read_link_file(): Replaces RemoteDatafile::read_link_file().
We will ensure that the last two path component separators are
forward slashes (converting up to 2 trailing backslashes on
Microsoft Windows), so that everywhere else we can
assume that data file names end in "/databasename/tablename.ibd".
Note: On Microsoft Windows, path names that start with \\?\ must
not contain / as path component separators. Previously, such paths
did work in the DATA DIRECTORY argument of InnoDB tables.
Reviewed by: Vladislav Vaintroub
2021-04-07 18:01:13 +03:00
|
|
|
MY_ATTRIBUTE((nonnull, warn_unused_result));
|
2016-08-12 11:17:45 +03:00
|
|
|
|
MDEV-11623 MariaDB 10.1 fails to start datadir created with
MariaDB 10.0/MySQL 5.6 using innodb-page-size!=16K
The storage format of FSP_SPACE_FLAGS was accidentally broken
already in MariaDB 10.1.0. This fix is bringing the format in
line with other MySQL and MariaDB release series.
Please refer to the comments that were added to fsp0fsp.h
for details.
This is an INCOMPATIBLE CHANGE that affects users of
page_compression and non-default innodb_page_size. Upgrading
to this release will correct the flags in the data files.
If you want to downgrade to earlier MariaDB 10.1.x, please refer
to the test innodb.101_compatibility how to reset the
FSP_SPACE_FLAGS in the files.
NOTE: MariaDB 10.1.0 to 10.1.20 can misinterpret
uncompressed data files with innodb_page_size=4k or 64k as
compressed innodb_page_size=16k files, and then probably fail
when trying to access the pages. See the comments in the
function fsp_flags_convert_from_101() for detailed analysis.
Move PAGE_COMPRESSION to FSP_SPACE_FLAGS bit position 16.
In this way, compressed innodb_page_size=16k tablespaces will not
be mistaken for uncompressed ones by MariaDB 10.1.0 to 10.1.20.
Derive PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR from the
dict_table_t::flags when the table is available, in
fil_space_for_table_exists_in_mem() or fil_open_single_table_tablespace().
During crash recovery, fil_load_single_table_tablespace() will use
innodb_compression_level for the PAGE_COMPRESSION_LEVEL.
FSP_FLAGS_MEM_MASK: A bitmap of the memory-only fil_space_t::flags
that are not to be written to FSP_SPACE_FLAGS. Currently, these will
include PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR.
Introduce the macro FSP_FLAGS_PAGE_SSIZE(). We only support
one innodb_page_size for the whole instance.
When creating a dummy tablespace for the redo log, use
fil_space_t::flags=0. The flags are never written to the redo log files.
Remove many FSP_FLAGS_SET_ macros.
dict_tf_verify_flags(): Remove. This is basically only duplicating
the logic of dict_tf_to_fsp_flags(), used in a debug assertion.
fil_space_t::mark: Remove. This flag was not used for anything.
fil_space_for_table_exists_in_mem(): Remove the unnecessary parameter
mark_space, and add a parameter for table flags. Check that
fil_space_t::flags match the table flags, and adjust the (memory-only)
flags based on the table flags.
fil_node_open_file(): Remove some redundant or unreachable conditions,
do not use stderr for output, and avoid unnecessary server aborts.
fil_user_tablespace_restore_page(): Convert the flags, so that the
correct page_size will be used when restoring a page from the
doublewrite buffer.
fil_space_get_page_compressed(), fsp_flags_is_page_compressed(): Remove.
It suffices to have fil_space_is_page_compressed().
FSP_FLAGS_WIDTH_DATA_DIR, FSP_FLAGS_WIDTH_PAGE_COMPRESSION_LEVEL,
FSP_FLAGS_WIDTH_ATOMIC_WRITES: Remove, because these flags do not
exist in the FSP_SPACE_FLAGS but only in memory.
fsp_flags_try_adjust(): New function, to adjust the FSP_SPACE_FLAGS
in page 0. Called by fil_open_single_table_tablespace(),
fil_space_for_table_exists_in_mem(), innobase_start_or_create_for_mysql()
except if --innodb-read-only is active.
fsp_flags_is_valid(ulint): Reimplement from the scratch, with
accurate comments. Do not display any details of detected
inconsistencies, because the output could be confusing when
dealing with MariaDB 10.1.x data files.
fsp_flags_convert_from_101(ulint): Convert flags from buggy
MariaDB 10.1.x format, or return ULINT_UNDEFINED if the flags
cannot be in MariaDB 10.1.x format.
fsp_flags_match(): Check the flags when probing files.
Implemented based on fsp_flags_is_valid()
and fsp_flags_convert_from_101().
dict_check_tablespaces_and_store_max_id(): Do not access the
page after committing the mini-transaction.
IMPORT TABLESPACE fixes:
AbstractCallback::init(): Convert the flags.
FetchIndexRootPages::operator(): Check that the tablespace flags match the
table flags. Do not attempt to convert tablespace flags to table flags,
because the conversion would necessarily be lossy.
PageConverter::update_header(): Write back the correct flags.
This takes care of the flags in IMPORT TABLESPACE.
2017-01-14 00:13:16 +02:00
|
|
|
/** Try to adjust FSP_SPACE_FLAGS if they differ from the expectations.
|
|
|
|
|
(Typically when upgrading from MariaDB 10.1.0..10.1.20.)
|
2018-03-27 11:49:57 +03:00
|
|
|
@param[in,out] space tablespace
|
MDEV-11623 MariaDB 10.1 fails to start datadir created with
MariaDB 10.0/MySQL 5.6 using innodb-page-size!=16K
The storage format of FSP_SPACE_FLAGS was accidentally broken
already in MariaDB 10.1.0. This fix is bringing the format in
line with other MySQL and MariaDB release series.
Please refer to the comments that were added to fsp0fsp.h
for details.
This is an INCOMPATIBLE CHANGE that affects users of
page_compression and non-default innodb_page_size. Upgrading
to this release will correct the flags in the data files.
If you want to downgrade to earlier MariaDB 10.1.x, please refer
to the test innodb.101_compatibility how to reset the
FSP_SPACE_FLAGS in the files.
NOTE: MariaDB 10.1.0 to 10.1.20 can misinterpret
uncompressed data files with innodb_page_size=4k or 64k as
compressed innodb_page_size=16k files, and then probably fail
when trying to access the pages. See the comments in the
function fsp_flags_convert_from_101() for detailed analysis.
Move PAGE_COMPRESSION to FSP_SPACE_FLAGS bit position 16.
In this way, compressed innodb_page_size=16k tablespaces will not
be mistaken for uncompressed ones by MariaDB 10.1.0 to 10.1.20.
Derive PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR from the
dict_table_t::flags when the table is available, in
fil_space_for_table_exists_in_mem() or fil_open_single_table_tablespace().
During crash recovery, fil_load_single_table_tablespace() will use
innodb_compression_level for the PAGE_COMPRESSION_LEVEL.
FSP_FLAGS_MEM_MASK: A bitmap of the memory-only fil_space_t::flags
that are not to be written to FSP_SPACE_FLAGS. Currently, these will
include PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR.
Introduce the macro FSP_FLAGS_PAGE_SSIZE(). We only support
one innodb_page_size for the whole instance.
When creating a dummy tablespace for the redo log, use
fil_space_t::flags=0. The flags are never written to the redo log files.
Remove many FSP_FLAGS_SET_ macros.
dict_tf_verify_flags(): Remove. This is basically only duplicating
the logic of dict_tf_to_fsp_flags(), used in a debug assertion.
fil_space_t::mark: Remove. This flag was not used for anything.
fil_space_for_table_exists_in_mem(): Remove the unnecessary parameter
mark_space, and add a parameter for table flags. Check that
fil_space_t::flags match the table flags, and adjust the (memory-only)
flags based on the table flags.
fil_node_open_file(): Remove some redundant or unreachable conditions,
do not use stderr for output, and avoid unnecessary server aborts.
fil_user_tablespace_restore_page(): Convert the flags, so that the
correct page_size will be used when restoring a page from the
doublewrite buffer.
fil_space_get_page_compressed(), fsp_flags_is_page_compressed(): Remove.
It suffices to have fil_space_is_page_compressed().
FSP_FLAGS_WIDTH_DATA_DIR, FSP_FLAGS_WIDTH_PAGE_COMPRESSION_LEVEL,
FSP_FLAGS_WIDTH_ATOMIC_WRITES: Remove, because these flags do not
exist in the FSP_SPACE_FLAGS but only in memory.
fsp_flags_try_adjust(): New function, to adjust the FSP_SPACE_FLAGS
in page 0. Called by fil_open_single_table_tablespace(),
fil_space_for_table_exists_in_mem(), innobase_start_or_create_for_mysql()
except if --innodb-read-only is active.
fsp_flags_is_valid(ulint): Reimplement from the scratch, with
accurate comments. Do not display any details of detected
inconsistencies, because the output could be confusing when
dealing with MariaDB 10.1.x data files.
fsp_flags_convert_from_101(ulint): Convert flags from buggy
MariaDB 10.1.x format, or return ULINT_UNDEFINED if the flags
cannot be in MariaDB 10.1.x format.
fsp_flags_match(): Check the flags when probing files.
Implemented based on fsp_flags_is_valid()
and fsp_flags_convert_from_101().
dict_check_tablespaces_and_store_max_id(): Do not access the
page after committing the mini-transaction.
IMPORT TABLESPACE fixes:
AbstractCallback::init(): Convert the flags.
FetchIndexRootPages::operator(): Check that the tablespace flags match the
table flags. Do not attempt to convert tablespace flags to table flags,
because the conversion would necessarily be lossy.
PageConverter::update_header(): Write back the correct flags.
This takes care of the flags in IMPORT TABLESPACE.
2017-01-14 00:13:16 +02:00
|
|
|
@param[in] flags desired tablespace flags */
|
2018-03-27 11:49:57 +03:00
|
|
|
void fsp_flags_try_adjust(fil_space_t* space, ulint flags);
|
MDEV-11623 MariaDB 10.1 fails to start datadir created with
MariaDB 10.0/MySQL 5.6 using innodb-page-size!=16K
The storage format of FSP_SPACE_FLAGS was accidentally broken
already in MariaDB 10.1.0. This fix is bringing the format in
line with other MySQL and MariaDB release series.
Please refer to the comments that were added to fsp0fsp.h
for details.
This is an INCOMPATIBLE CHANGE that affects users of
page_compression and non-default innodb_page_size. Upgrading
to this release will correct the flags in the data files.
If you want to downgrade to earlier MariaDB 10.1.x, please refer
to the test innodb.101_compatibility how to reset the
FSP_SPACE_FLAGS in the files.
NOTE: MariaDB 10.1.0 to 10.1.20 can misinterpret
uncompressed data files with innodb_page_size=4k or 64k as
compressed innodb_page_size=16k files, and then probably fail
when trying to access the pages. See the comments in the
function fsp_flags_convert_from_101() for detailed analysis.
Move PAGE_COMPRESSION to FSP_SPACE_FLAGS bit position 16.
In this way, compressed innodb_page_size=16k tablespaces will not
be mistaken for uncompressed ones by MariaDB 10.1.0 to 10.1.20.
Derive PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR from the
dict_table_t::flags when the table is available, in
fil_space_for_table_exists_in_mem() or fil_open_single_table_tablespace().
During crash recovery, fil_load_single_table_tablespace() will use
innodb_compression_level for the PAGE_COMPRESSION_LEVEL.
FSP_FLAGS_MEM_MASK: A bitmap of the memory-only fil_space_t::flags
that are not to be written to FSP_SPACE_FLAGS. Currently, these will
include PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR.
Introduce the macro FSP_FLAGS_PAGE_SSIZE(). We only support
one innodb_page_size for the whole instance.
When creating a dummy tablespace for the redo log, use
fil_space_t::flags=0. The flags are never written to the redo log files.
Remove many FSP_FLAGS_SET_ macros.
dict_tf_verify_flags(): Remove. This is basically only duplicating
the logic of dict_tf_to_fsp_flags(), used in a debug assertion.
fil_space_t::mark: Remove. This flag was not used for anything.
fil_space_for_table_exists_in_mem(): Remove the unnecessary parameter
mark_space, and add a parameter for table flags. Check that
fil_space_t::flags match the table flags, and adjust the (memory-only)
flags based on the table flags.
fil_node_open_file(): Remove some redundant or unreachable conditions,
do not use stderr for output, and avoid unnecessary server aborts.
fil_user_tablespace_restore_page(): Convert the flags, so that the
correct page_size will be used when restoring a page from the
doublewrite buffer.
fil_space_get_page_compressed(), fsp_flags_is_page_compressed(): Remove.
It suffices to have fil_space_is_page_compressed().
FSP_FLAGS_WIDTH_DATA_DIR, FSP_FLAGS_WIDTH_PAGE_COMPRESSION_LEVEL,
FSP_FLAGS_WIDTH_ATOMIC_WRITES: Remove, because these flags do not
exist in the FSP_SPACE_FLAGS but only in memory.
fsp_flags_try_adjust(): New function, to adjust the FSP_SPACE_FLAGS
in page 0. Called by fil_open_single_table_tablespace(),
fil_space_for_table_exists_in_mem(), innobase_start_or_create_for_mysql()
except if --innodb-read-only is active.
fsp_flags_is_valid(ulint): Reimplement from the scratch, with
accurate comments. Do not display any details of detected
inconsistencies, because the output could be confusing when
dealing with MariaDB 10.1.x data files.
fsp_flags_convert_from_101(ulint): Convert flags from buggy
MariaDB 10.1.x format, or return ULINT_UNDEFINED if the flags
cannot be in MariaDB 10.1.x format.
fsp_flags_match(): Check the flags when probing files.
Implemented based on fsp_flags_is_valid()
and fsp_flags_convert_from_101().
dict_check_tablespaces_and_store_max_id(): Do not access the
page after committing the mini-transaction.
IMPORT TABLESPACE fixes:
AbstractCallback::init(): Convert the flags.
FetchIndexRootPages::operator(): Check that the tablespace flags match the
table flags. Do not attempt to convert tablespace flags to table flags,
because the conversion would necessarily be lossy.
PageConverter::update_header(): Write back the correct flags.
This takes care of the flags in IMPORT TABLESPACE.
2017-01-14 00:13:16 +02:00
|
|
|
|
2014-02-26 19:11:54 +01:00
|
|
|
/********************************************************************//**
|
|
|
|
|
Tries to open a single-table tablespace and optionally checks the space id is
|
|
|
|
|
right in it. If does not succeed, prints an error message to the .err log. This
|
|
|
|
|
function is used to open a tablespace when we start up mysqld, and also in
|
|
|
|
|
IMPORT TABLESPACE.
|
|
|
|
|
NOTE that we assume this operation is used either at the database startup
|
MDEV-24258 Merge dict_sys.mutex into dict_sys.latch
In the parent commit, dict_sys.latch could theoretically have been
replaced with a mutex. But, we can do better and merge dict_sys.mutex
into dict_sys.latch. Generally, every occurrence of dict_sys.mutex_lock()
will be replaced with dict_sys.lock().
The PERFORMANCE_SCHEMA instrumentation for dict_sys_mutex
will be removed along with dict_sys.mutex. The dict_sys.latch
will remain instrumented as dict_operation_lock.
Some use of dict_sys.lock() will be replaced with dict_sys.freeze(),
which we will reintroduce for the new shared mode. Most notably,
concurrent table lookups are possible as long as the tables are present
in the dict_sys cache. In particular, this will allow more concurrency
among InnoDB purge workers.
Because dict_sys.mutex will no longer 'throttle' the threads that purge
InnoDB transaction history, a performance degradation may be observed
unless innodb_purge_threads=1.
The table cache eviction policy will become FIFO-like,
similar to what happened to fil_system.LRU
in commit 45ed9dd957eebc7fc84feb2509f4aa6baa908a95.
The name of the list dict_sys.table_LRU will become somewhat misleading;
that list contains tables that may be evicted, even though the
eviction policy no longer is least-recently-used but first-in-first-out.
(Note: Tables can never be evicted as long as locks exist on them or
the tables are in use by some thread.)
As demonstrated by the test perfschema.sxlock_func, there
will be less contention on dict_sys.latch, because some previous
use of exclusive latches will be replaced with shared latches.
fts_parse_sql_no_dict_lock(): Replaced with pars_sql().
fts_get_table_name_prefix(): Merged to fts_optimize_create().
dict_stats_update_transient_for_index(): Deduplicated some code.
ha_innobase::info_low(), dict_stats_stop_bg(): Use a combination
of dict_sys.latch and table->stats_mutex_lock() to cover the
changes of BG_STAT_SHOULD_QUIT, because the flag is being read
in dict_stats_update_persistent() while not holding dict_sys.latch.
row_discard_tablespace_for_mysql(): Protect stats_bg_flag by
exclusive dict_sys.latch, like most other code does.
row_quiesce_table_has_fts_index(): Remove unnecessary mutex
acquisition. FLUSH TABLES...FOR EXPORT is protected by MDL.
row_import::set_root_by_heuristic(): Remove unnecessary mutex
acquisition. ALTER TABLE...IMPORT TABLESPACE is protected by MDL.
row_ins_sec_index_entry_low(): Replace a call
to dict_set_corrupted_index_cache_only(). Reads of index->type
were not really protected by dict_sys.mutex, and writes
(flagging an index corrupted) should be extremely rare.
dict_stats_process_entry_from_defrag_pool(): Only freeze the dictionary,
do not lock it exclusively.
dict_stats_wait_bg_to_stop_using_table(), DICT_BG_YIELD: Remove trx.
We can simply invoke dict_sys.unlock() and dict_sys.lock() directly.
dict_acquire_mdl_shared()<trylock=false>: Assert that dict_sys.latch is
only held in shared more, not exclusive mode. Only acquire it in
exclusive mode if the table needs to be loaded to the cache.
dict_sys_t::acquire(): Remove. Relocating elements in dict_sys.table_LRU
would require holding an exclusive latch, which we want to avoid
for performance reasons.
dict_sys_t::allow_eviction(): Add the table first to dict_sys.table_LRU,
to compensate for the removal of dict_sys_t::acquire(). This function
is only invoked by INFORMATION_SCHEMA.INNODB_SYS_TABLESTATS.
dict_table_open_on_id(), dict_table_open_on_name(): If dict_locked=false,
try to acquire dict_sys.latch in shared mode. Only acquire the latch in
exclusive mode if the table is not found in the cache.
Reviewed by: Thirunarayanan Balathandayuthapani
2021-08-31 13:51:35 +03:00
|
|
|
or under the protection of dict_sys.latch, so that two users cannot
|
2014-02-26 19:11:54 +01:00
|
|
|
race here. This operation does not leave the file associated with the
|
|
|
|
|
tablespace open, but closes it after we have looked at the space id in it.
|
|
|
|
|
|
|
|
|
|
If the validate boolean is set, we read the first page of the file and
|
|
|
|
|
check that the space id in the file is what we expect. We assume that
|
|
|
|
|
this function runs much faster if no check is made, since accessing the
|
|
|
|
|
file inode probably is much faster (the OS caches them) than accessing
|
2016-08-12 11:17:45 +03:00
|
|
|
the first page of the file. This boolean may be initially false, but if
|
2014-02-26 19:11:54 +01:00
|
|
|
a remote tablespace is found it will be changed to true.
|
|
|
|
|
|
MDEV-27234: Data dictionary recovery was not READ COMMITTED
This also fixes MDEV-20198: Instant ALTER TABLE is not crash safe
InnoDB dictionary recovery wrongly used the READ UNCOMMITTED isolation
level, causing some mismatch. For example, if a table was renamed or
replaced in a transaction, according to READ UNCOMMITTED the table might
not exist at all.
We implement READ COMMITTED isolation level for accessing the dictionary
tables SYS_TABLES, SYS_COLUMNS, SYS_INDEXES, SYS_FIELDS, SYS_VIRTUAL,
SYS_FOREIGN, SYS_FOREIGN_COLS. For most of these tables, no secondary
index exists. For the secondary indexes (on SYS_TABLES.ID,
SYS_FOREIGN.FOR_NAME, SYS_FOREIGN.REF_NAME), we will always look up
the primary key in the clustered index and check if the record actually
is a committed version.
dict_check_sys_tables(): Recover tablespaces also from delete-marked
committed records, so that if a matching .ibd file exists, it will
be removed by fil_delete_tablespace() when the committed delete-marked
SYS_INDEXES record of the clustered index is purged
in row_purge_remove_clust_if_poss_low().
fil_ibd_open(): Change the Boolean parameter "validate" to a ternary
one, to suppress error messages when the file might not exist.
It is possible that a .ibd file was deleted and the server shut down
before the SYS_INDEXES and SYS_TABLES records were purged. Hence, if
dict_check_sys_tables() finds a committed delete-marked record,
we must not complain if the tablespace file is not found.
On Windows, we msut treat ERROR_PATH_NOT_FOUND (directory not found)
in the same way as ERROR_FILE_NOT_FOUND. This fixes a few failures where
a previous test successfully executed DROP DATABASE (and deleted all
files and the directory), but a committed delete-marked SYS_TABLES
record had not been purged before server restart.
dict_getnext_system_low(): Do not filter out delete-marked records.
dict_startscan_system(), dict_getnext_system(): Do filter out
delete-marked records, for accessing the INFORMATION_SCHEMA tables.
dict_sys_tables_rec_read(): Return the DB_TRX_ID of the committed
version of the record. This is needed in dict_load_table_low().
dict_load_foreign_cols(), dict_load_foreign(): Add a parameter for
the current transaction identifier. In some DDL operations, the
FOREIGN KEY constraints are being loaded from the data dictionary
before the DDL transaction has been committed. For SYS_FOREIGN
and SYS_FOREIGN_COLS, we must implement the special case of
READ COMMITTED that the changes of the uncommitted current transaction
are visible.
dict_load_foreign(): Validate the table name. We could find a
SYS_FOREIGN.ID via a committed delete-marked secondary index record
that does not match the REF_NAME or FOR_NAME of the secondary index record.
dict_load_index_low(): Optionally take the table as a parameter,
so that table->def_trx_id can be updated in case of a
committed delete-marked SYS_INDEXES record corresponding
to DROP INDEX, but not corresponding to an index stub of ADD INDEX.
dict_load_indexes(): Do not update table->def_trx_id
in case of delete-marked records.
rec_is_metadata(), rec_offs_make_valid(), rec_get_offsets_func(),
row_build_low(): Relax some assertions. We may now have
!index->is_instant() even if a metadata record is present in the index.
Previously, the recovery of instant ADD/DROP COLUMN assumed
that READ UNCOMMITTED of the data dictionary will be performed.
Now, we will have a READ COMMITTED copy of the data dictionary
cache, and a READ UNCOMMITTED copy of the metadata record.
btr_page_reorganize_low(): Correctly update the FIL_PAGE_TYPE
when rolling back an instant ADD/DROP COLUMN operation.
row_rec_to_index_entry_impl(): Relax some assertions,
and disallow accessing "extra" fields. This fixes the recovery
of a crash during an instant ADD COLUMN after a successful
instant DROP COLUMN, in the test innodb.instant_alter_crash.
Tested by: Matthias Leich
2022-03-21 17:15:06 +02:00
|
|
|
@param[in] validate 0=maybe missing, 1=do not validate, 2=validate
|
2016-08-12 11:17:45 +03:00
|
|
|
@param[in] purpose FIL_TYPE_TABLESPACE or FIL_TYPE_TEMPORARY
|
|
|
|
|
@param[in] id tablespace ID
|
2017-01-19 12:06:13 +02:00
|
|
|
@param[in] flags expected FSP_SPACE_FLAGS
|
2021-05-20 14:58:25 +03:00
|
|
|
@param[in] name table name
|
2016-08-12 11:17:45 +03:00
|
|
|
If file-per-table, it is the table name in the databasename/tablename format
|
|
|
|
|
@param[in] path_in expected filepath, usually read from dictionary
|
MDEV-12266: Change dict_table_t::space to fil_space_t*
InnoDB always keeps all tablespaces in the fil_system cache.
The fil_system.LRU is only for closing file handles; the
fil_space_t and fil_node_t for all data files will remain
in main memory. Between startup to shutdown, they can only be
created and removed by DDL statements. Therefore, we can
let dict_table_t::space point directly to the fil_space_t.
dict_table_t::space_id: A numeric tablespace ID for the corner cases
where we do not have a tablespace. The most prominent examples are
ALTER TABLE...DISCARD TABLESPACE or a missing or corrupted file.
There are a few functional differences; most notably:
(1) DROP TABLE will delete matching .ibd and .cfg files,
even if they were not attached to the data dictionary.
(2) Some error messages will report file names instead of numeric IDs.
There still are many functions that use numeric tablespace IDs instead
of fil_space_t*, and many functions could be converted to fil_space_t
member functions. Also, Tablespace and Datafile should be merged with
fil_space_t and fil_node_t. page_id_t and buf_page_get_gen() could use
fil_space_t& instead of a numeric ID, and after moving to a single
buffer pool (MDEV-15058), buf_pool_t::page_hash could be moved to
fil_space_t::page_hash.
FilSpace: Remove. Only few calls to fil_space_acquire() will remain,
and gradually they should be removed.
mtr_t::set_named_space_id(ulint): Renamed from set_named_space(),
to prevent accidental calls to this slower function. Very few
callers remain.
fseg_create(), fsp_reserve_free_extents(): Take fil_space_t*
as a parameter instead of a space_id.
fil_space_t::rename(): Wrapper for fil_rename_tablespace_check(),
fil_name_write_rename(), fil_rename_tablespace(). Mariabackup
passes the parameter log=false; InnoDB passes log=true.
dict_mem_table_create(): Take fil_space_t* instead of space_id
as parameter.
dict_process_sys_tables_rec_and_mtr_commit(): Replace the parameter
'status' with 'bool cached'.
dict_get_and_save_data_dir_path(): Avoid copying the fil_node_t::name.
fil_ibd_open(): Return the tablespace.
fil_space_t::set_imported(): Replaces fil_space_set_imported().
truncate_t: Change many member function parameters to fil_space_t*,
and remove page_size parameters.
row_truncate_prepare(): Merge to its only caller.
row_drop_table_from_cache(): Assert that the table is persistent.
dict_create_sys_indexes_tuple(): Write SYS_INDEXES.SPACE=FIL_NULL
if the tablespace has been discarded.
row_import_update_discarded_flag(): Remove a constant parameter.
2018-03-27 16:31:10 +03:00
|
|
|
@param[out] err DB_SUCCESS or error code
|
|
|
|
|
@return tablespace
|
|
|
|
|
@retval NULL if the tablespace could not be opened */
|
|
|
|
|
fil_space_t*
|
2016-08-12 11:17:45 +03:00
|
|
|
fil_ibd_open(
|
MDEV-27234: Data dictionary recovery was not READ COMMITTED
This also fixes MDEV-20198: Instant ALTER TABLE is not crash safe
InnoDB dictionary recovery wrongly used the READ UNCOMMITTED isolation
level, causing some mismatch. For example, if a table was renamed or
replaced in a transaction, according to READ UNCOMMITTED the table might
not exist at all.
We implement READ COMMITTED isolation level for accessing the dictionary
tables SYS_TABLES, SYS_COLUMNS, SYS_INDEXES, SYS_FIELDS, SYS_VIRTUAL,
SYS_FOREIGN, SYS_FOREIGN_COLS. For most of these tables, no secondary
index exists. For the secondary indexes (on SYS_TABLES.ID,
SYS_FOREIGN.FOR_NAME, SYS_FOREIGN.REF_NAME), we will always look up
the primary key in the clustered index and check if the record actually
is a committed version.
dict_check_sys_tables(): Recover tablespaces also from delete-marked
committed records, so that if a matching .ibd file exists, it will
be removed by fil_delete_tablespace() when the committed delete-marked
SYS_INDEXES record of the clustered index is purged
in row_purge_remove_clust_if_poss_low().
fil_ibd_open(): Change the Boolean parameter "validate" to a ternary
one, to suppress error messages when the file might not exist.
It is possible that a .ibd file was deleted and the server shut down
before the SYS_INDEXES and SYS_TABLES records were purged. Hence, if
dict_check_sys_tables() finds a committed delete-marked record,
we must not complain if the tablespace file is not found.
On Windows, we msut treat ERROR_PATH_NOT_FOUND (directory not found)
in the same way as ERROR_FILE_NOT_FOUND. This fixes a few failures where
a previous test successfully executed DROP DATABASE (and deleted all
files and the directory), but a committed delete-marked SYS_TABLES
record had not been purged before server restart.
dict_getnext_system_low(): Do not filter out delete-marked records.
dict_startscan_system(), dict_getnext_system(): Do filter out
delete-marked records, for accessing the INFORMATION_SCHEMA tables.
dict_sys_tables_rec_read(): Return the DB_TRX_ID of the committed
version of the record. This is needed in dict_load_table_low().
dict_load_foreign_cols(), dict_load_foreign(): Add a parameter for
the current transaction identifier. In some DDL operations, the
FOREIGN KEY constraints are being loaded from the data dictionary
before the DDL transaction has been committed. For SYS_FOREIGN
and SYS_FOREIGN_COLS, we must implement the special case of
READ COMMITTED that the changes of the uncommitted current transaction
are visible.
dict_load_foreign(): Validate the table name. We could find a
SYS_FOREIGN.ID via a committed delete-marked secondary index record
that does not match the REF_NAME or FOR_NAME of the secondary index record.
dict_load_index_low(): Optionally take the table as a parameter,
so that table->def_trx_id can be updated in case of a
committed delete-marked SYS_INDEXES record corresponding
to DROP INDEX, but not corresponding to an index stub of ADD INDEX.
dict_load_indexes(): Do not update table->def_trx_id
in case of delete-marked records.
rec_is_metadata(), rec_offs_make_valid(), rec_get_offsets_func(),
row_build_low(): Relax some assertions. We may now have
!index->is_instant() even if a metadata record is present in the index.
Previously, the recovery of instant ADD/DROP COLUMN assumed
that READ UNCOMMITTED of the data dictionary will be performed.
Now, we will have a READ COMMITTED copy of the data dictionary
cache, and a READ UNCOMMITTED copy of the metadata record.
btr_page_reorganize_low(): Correctly update the FIL_PAGE_TYPE
when rolling back an instant ADD/DROP COLUMN operation.
row_rec_to_index_entry_impl(): Relax some assertions,
and disallow accessing "extra" fields. This fixes the recovery
of a crash during an instant ADD COLUMN after a successful
instant DROP COLUMN, in the test innodb.instant_alter_crash.
Tested by: Matthias Leich
2022-03-21 17:15:06 +02:00
|
|
|
unsigned validate,
|
MDEV-12266: Change dict_table_t::space to fil_space_t*
InnoDB always keeps all tablespaces in the fil_system cache.
The fil_system.LRU is only for closing file handles; the
fil_space_t and fil_node_t for all data files will remain
in main memory. Between startup to shutdown, they can only be
created and removed by DDL statements. Therefore, we can
let dict_table_t::space point directly to the fil_space_t.
dict_table_t::space_id: A numeric tablespace ID for the corner cases
where we do not have a tablespace. The most prominent examples are
ALTER TABLE...DISCARD TABLESPACE or a missing or corrupted file.
There are a few functional differences; most notably:
(1) DROP TABLE will delete matching .ibd and .cfg files,
even if they were not attached to the data dictionary.
(2) Some error messages will report file names instead of numeric IDs.
There still are many functions that use numeric tablespace IDs instead
of fil_space_t*, and many functions could be converted to fil_space_t
member functions. Also, Tablespace and Datafile should be merged with
fil_space_t and fil_node_t. page_id_t and buf_page_get_gen() could use
fil_space_t& instead of a numeric ID, and after moving to a single
buffer pool (MDEV-15058), buf_pool_t::page_hash could be moved to
fil_space_t::page_hash.
FilSpace: Remove. Only few calls to fil_space_acquire() will remain,
and gradually they should be removed.
mtr_t::set_named_space_id(ulint): Renamed from set_named_space(),
to prevent accidental calls to this slower function. Very few
callers remain.
fseg_create(), fsp_reserve_free_extents(): Take fil_space_t*
as a parameter instead of a space_id.
fil_space_t::rename(): Wrapper for fil_rename_tablespace_check(),
fil_name_write_rename(), fil_rename_tablespace(). Mariabackup
passes the parameter log=false; InnoDB passes log=true.
dict_mem_table_create(): Take fil_space_t* instead of space_id
as parameter.
dict_process_sys_tables_rec_and_mtr_commit(): Replace the parameter
'status' with 'bool cached'.
dict_get_and_save_data_dir_path(): Avoid copying the fil_node_t::name.
fil_ibd_open(): Return the tablespace.
fil_space_t::set_imported(): Replaces fil_space_set_imported().
truncate_t: Change many member function parameters to fil_space_t*,
and remove page_size parameters.
row_truncate_prepare(): Merge to its only caller.
row_drop_table_from_cache(): Assert that the table is persistent.
dict_create_sys_indexes_tuple(): Write SYS_INDEXES.SPACE=FIL_NULL
if the tablespace has been discarded.
row_import_update_discarded_flag(): Remove a constant parameter.
2018-03-27 16:31:10 +03:00
|
|
|
fil_type_t purpose,
|
|
|
|
|
ulint id,
|
|
|
|
|
ulint flags,
|
2021-05-20 14:58:25 +03:00
|
|
|
fil_space_t::name_type name,
|
MDEV-12266: Change dict_table_t::space to fil_space_t*
InnoDB always keeps all tablespaces in the fil_system cache.
The fil_system.LRU is only for closing file handles; the
fil_space_t and fil_node_t for all data files will remain
in main memory. Between startup to shutdown, they can only be
created and removed by DDL statements. Therefore, we can
let dict_table_t::space point directly to the fil_space_t.
dict_table_t::space_id: A numeric tablespace ID for the corner cases
where we do not have a tablespace. The most prominent examples are
ALTER TABLE...DISCARD TABLESPACE or a missing or corrupted file.
There are a few functional differences; most notably:
(1) DROP TABLE will delete matching .ibd and .cfg files,
even if they were not attached to the data dictionary.
(2) Some error messages will report file names instead of numeric IDs.
There still are many functions that use numeric tablespace IDs instead
of fil_space_t*, and many functions could be converted to fil_space_t
member functions. Also, Tablespace and Datafile should be merged with
fil_space_t and fil_node_t. page_id_t and buf_page_get_gen() could use
fil_space_t& instead of a numeric ID, and after moving to a single
buffer pool (MDEV-15058), buf_pool_t::page_hash could be moved to
fil_space_t::page_hash.
FilSpace: Remove. Only few calls to fil_space_acquire() will remain,
and gradually they should be removed.
mtr_t::set_named_space_id(ulint): Renamed from set_named_space(),
to prevent accidental calls to this slower function. Very few
callers remain.
fseg_create(), fsp_reserve_free_extents(): Take fil_space_t*
as a parameter instead of a space_id.
fil_space_t::rename(): Wrapper for fil_rename_tablespace_check(),
fil_name_write_rename(), fil_rename_tablespace(). Mariabackup
passes the parameter log=false; InnoDB passes log=true.
dict_mem_table_create(): Take fil_space_t* instead of space_id
as parameter.
dict_process_sys_tables_rec_and_mtr_commit(): Replace the parameter
'status' with 'bool cached'.
dict_get_and_save_data_dir_path(): Avoid copying the fil_node_t::name.
fil_ibd_open(): Return the tablespace.
fil_space_t::set_imported(): Replaces fil_space_set_imported().
truncate_t: Change many member function parameters to fil_space_t*,
and remove page_size parameters.
row_truncate_prepare(): Merge to its only caller.
row_drop_table_from_cache(): Assert that the table is persistent.
dict_create_sys_indexes_tuple(): Write SYS_INDEXES.SPACE=FIL_NULL
if the tablespace has been discarded.
row_import_update_discarded_flag(): Remove a constant parameter.
2018-03-27 16:31:10 +03:00
|
|
|
const char* path_in,
|
|
|
|
|
dberr_t* err = NULL)
|
2016-09-06 09:43:16 +03:00
|
|
|
MY_ATTRIBUTE((warn_unused_result));
|
2016-08-12 11:17:45 +03:00
|
|
|
|
|
|
|
|
enum fil_load_status {
|
|
|
|
|
/** The tablespace file(s) were found and valid. */
|
|
|
|
|
FIL_LOAD_OK,
|
|
|
|
|
/** The name no longer matches space_id */
|
|
|
|
|
FIL_LOAD_ID_CHANGED,
|
|
|
|
|
/** The file(s) were not found */
|
|
|
|
|
FIL_LOAD_NOT_FOUND,
|
|
|
|
|
/** The file(s) were not valid */
|
2021-05-17 18:12:33 +03:00
|
|
|
FIL_LOAD_INVALID,
|
|
|
|
|
/** The tablespace file was deferred to open */
|
|
|
|
|
FIL_LOAD_DEFER
|
2016-08-12 11:17:45 +03:00
|
|
|
};
|
|
|
|
|
|
|
|
|
|
/** Open a single-file tablespace and add it to the InnoDB data structures.
|
|
|
|
|
@param[in] space_id tablespace ID
|
|
|
|
|
@param[in] filename path/to/databasename/tablename.ibd
|
|
|
|
|
@param[out] space the tablespace, or NULL on error
|
|
|
|
|
@return status of the operation */
|
|
|
|
|
enum fil_load_status
|
|
|
|
|
fil_ibd_load(
|
|
|
|
|
ulint space_id,
|
|
|
|
|
const char* filename,
|
|
|
|
|
fil_space_t*& space)
|
2016-09-06 09:43:16 +03:00
|
|
|
MY_ATTRIBUTE((warn_unused_result));
|
2014-02-26 19:11:54 +01:00
|
|
|
|
2016-08-12 11:17:45 +03:00
|
|
|
|
2017-12-20 22:19:47 +02:00
|
|
|
/** Determine if a matching tablespace exists in the InnoDB tablespace
|
|
|
|
|
memory cache. Note that if we have not done a crash recovery at the database
|
|
|
|
|
startup, there may be many tablespaces which are not yet in the memory cache.
|
|
|
|
|
@param[in] id Tablespace ID
|
|
|
|
|
@param[in] table_flags table flags
|
MDEV-12266: Change dict_table_t::space to fil_space_t*
InnoDB always keeps all tablespaces in the fil_system cache.
The fil_system.LRU is only for closing file handles; the
fil_space_t and fil_node_t for all data files will remain
in main memory. Between startup to shutdown, they can only be
created and removed by DDL statements. Therefore, we can
let dict_table_t::space point directly to the fil_space_t.
dict_table_t::space_id: A numeric tablespace ID for the corner cases
where we do not have a tablespace. The most prominent examples are
ALTER TABLE...DISCARD TABLESPACE or a missing or corrupted file.
There are a few functional differences; most notably:
(1) DROP TABLE will delete matching .ibd and .cfg files,
even if they were not attached to the data dictionary.
(2) Some error messages will report file names instead of numeric IDs.
There still are many functions that use numeric tablespace IDs instead
of fil_space_t*, and many functions could be converted to fil_space_t
member functions. Also, Tablespace and Datafile should be merged with
fil_space_t and fil_node_t. page_id_t and buf_page_get_gen() could use
fil_space_t& instead of a numeric ID, and after moving to a single
buffer pool (MDEV-15058), buf_pool_t::page_hash could be moved to
fil_space_t::page_hash.
FilSpace: Remove. Only few calls to fil_space_acquire() will remain,
and gradually they should be removed.
mtr_t::set_named_space_id(ulint): Renamed from set_named_space(),
to prevent accidental calls to this slower function. Very few
callers remain.
fseg_create(), fsp_reserve_free_extents(): Take fil_space_t*
as a parameter instead of a space_id.
fil_space_t::rename(): Wrapper for fil_rename_tablespace_check(),
fil_name_write_rename(), fil_rename_tablespace(). Mariabackup
passes the parameter log=false; InnoDB passes log=true.
dict_mem_table_create(): Take fil_space_t* instead of space_id
as parameter.
dict_process_sys_tables_rec_and_mtr_commit(): Replace the parameter
'status' with 'bool cached'.
dict_get_and_save_data_dir_path(): Avoid copying the fil_node_t::name.
fil_ibd_open(): Return the tablespace.
fil_space_t::set_imported(): Replaces fil_space_set_imported().
truncate_t: Change many member function parameters to fil_space_t*,
and remove page_size parameters.
row_truncate_prepare(): Merge to its only caller.
row_drop_table_from_cache(): Assert that the table is persistent.
dict_create_sys_indexes_tuple(): Write SYS_INDEXES.SPACE=FIL_NULL
if the tablespace has been discarded.
row_import_update_discarded_flag(): Remove a constant parameter.
2018-03-27 16:31:10 +03:00
|
|
|
@return the tablespace
|
|
|
|
|
@retval NULL if no matching tablespace exists in the memory cache */
|
MDEV-25312 Replace fil_space_t::name with fil_space_t::name()
A consistency check for fil_space_t::name is causing recovery failures
in MDEV-25180 (Atomic ALTER TABLE). So, we'd better remove that field
altogether.
fil_space_t::name was more or less a copy of dict_table_t::name
(except for some special cases), and it was not being used for
anything useful.
There used to be a name_hash, but it had been removed already in
commit a75dbfd7183cc96680f3e3e684fd36500dac8158 (MDEV-12266).
We will also remove os_normalize_path(), OS_PATH_SEPARATOR,
OS_PATH_SEPATOR_ALT. On Microsoft Windows, we will treat \ and /
roughly in the same way. The intention is that for per-table
tablespaces, the filenames will always follow the pattern
prefix/databasename/tablename.ibd. (Any \ in the prefix must not
be converted.)
ut_basename_noext(): Remove (unused function).
read_link_file(): Replaces RemoteDatafile::read_link_file().
We will ensure that the last two path component separators are
forward slashes (converting up to 2 trailing backslashes on
Microsoft Windows), so that everywhere else we can
assume that data file names end in "/databasename/tablename.ibd".
Note: On Microsoft Windows, path names that start with \\?\ must
not contain / as path component separators. Previously, such paths
did work in the DATA DIRECTORY argument of InnoDB tables.
Reviewed by: Vladislav Vaintroub
2021-04-07 18:01:13 +03:00
|
|
|
fil_space_t *fil_space_for_table_exists_in_mem(ulint id, ulint table_flags);
|
2016-12-30 15:04:10 +02:00
|
|
|
|
2016-08-12 11:17:45 +03:00
|
|
|
/** Try to extend a tablespace if it is smaller than the specified size.
|
|
|
|
|
@param[in,out] space tablespace
|
|
|
|
|
@param[in] size desired size in pages
|
|
|
|
|
@return whether the tablespace is at least as big as requested */
|
2020-10-15 16:28:19 +03:00
|
|
|
bool fil_space_extend(fil_space_t *space, uint32_t size);
|
2016-08-12 11:17:45 +03:00
|
|
|
|
|
|
|
|
/** Flush to disk the writes in file spaces of the given type
|
2020-05-05 07:06:37 +03:00
|
|
|
possibly cached by the OS. */
|
|
|
|
|
void fil_flush_file_spaces();
|
2014-02-26 19:11:54 +01:00
|
|
|
/******************************************************************//**
|
|
|
|
|
Checks the consistency of the tablespace cache.
|
2016-08-12 11:17:45 +03:00
|
|
|
@return true if ok */
|
2020-05-05 07:06:37 +03:00
|
|
|
bool fil_validate();
|
2014-02-26 19:11:54 +01:00
|
|
|
/*********************************************************************//**
|
|
|
|
|
Sets the file page type. */
|
|
|
|
|
void
|
|
|
|
|
fil_page_set_type(
|
|
|
|
|
/*==============*/
|
|
|
|
|
byte* page, /*!< in/out: file page */
|
|
|
|
|
ulint type); /*!< in: type */
|
2016-08-12 11:17:45 +03:00
|
|
|
|
2014-02-26 19:11:54 +01:00
|
|
|
/********************************************************************//**
|
|
|
|
|
Delete the tablespace file and any related files like .cfg.
|
|
|
|
|
This should not be called for temporary tables. */
|
|
|
|
|
void
|
|
|
|
|
fil_delete_file(
|
|
|
|
|
/*============*/
|
|
|
|
|
const char* path); /*!< in: filepath of the ibd tablespace */
|
|
|
|
|
|
2016-09-06 09:43:16 +03:00
|
|
|
/*******************************************************************//**
|
|
|
|
|
Returns the table space by a given id, NULL if not found. */
|
|
|
|
|
fil_space_t*
|
|
|
|
|
fil_space_get_by_id(
|
|
|
|
|
/*================*/
|
|
|
|
|
ulint id); /*!< in: space id */
|
|
|
|
|
|
2018-03-27 11:34:43 +03:00
|
|
|
/** Note that a non-predefined persistent tablespace has been modified
|
2016-08-12 11:17:45 +03:00
|
|
|
by redo log.
|
|
|
|
|
@param[in,out] space tablespace */
|
|
|
|
|
void
|
|
|
|
|
fil_names_dirty(
|
|
|
|
|
fil_space_t* space);
|
|
|
|
|
|
MDEV-12353: Change the redo log encoding
log_t::FORMAT_10_5: physical redo log format tag
log_phys_t: Buffered records in the physical format.
The log record bytes will follow the last data field,
making use of alignment padding that would otherwise be wasted.
If there are multiple records for the same page, also those
may be appended to an existing log_phys_t object if the memory
is available.
In the physical format, the first byte of a record identifies the
record and its length (up to 15 bytes). For longer records, the
immediately following bytes will encode the remaining length
in a variable-length encoding. Usually, a variable-length-encoded
page identifier will follow, followed by optional payload, whose
length is included in the initially encoded total record length.
When a mini-transaction is updating multiple fields in a page,
it can avoid repeating the tablespace identifier and page number
by setting the same_page flag (most significant bit) in the first
byte of the log record. The byte offset of the record will be
relative to where the previous record for that page ended.
Until MDEV-14425 introduces a separate file-level log for
redo log checkpoints and file operations, we will write the
file-level records in the page-level redo log file.
The record FILE_CHECKPOINT (which replaces MLOG_CHECKPOINT)
will be removed in MDEV-14425, and one sequential scan of the
page recovery log will suffice.
Compared to MLOG_FILE_CREATE2, FILE_CREATE will not include any flags.
If the information is needed, it can be parsed from WRITE records that
modify FSP_SPACE_FLAGS.
MLOG_ZIP_WRITE_STRING: Remove. The record was only introduced temporarily
as part of this work, before being replaced with WRITE (along with
MLOG_WRITE_STRING, MLOG_1BYTE, MLOG_nBYTES).
mtr_buf_t::empty(): Check if the buffer is empty.
mtr_t::m_n_log_recs: Remove. It suffices to check if m_log is empty.
mtr_t::m_last, mtr_t::m_last_offset: End of the latest m_log record,
for the same_page encoding.
page_recv_t::last_offset: Reflects mtr_t::m_last_offset.
Valid values for last_offset during recovery should be 0 or above 8.
(The first 8 bytes of a page are the checksum and the page number,
and neither are ever updated directly by log records.)
Internally, the special value 1 indicates that the same_page form
will not be allowed for the subsequent record.
mtr_t::page_create(): Take the block descriptor as parameter,
so that it can be compared to mtr_t::m_last. The INIT_INDEX_PAGE
record will always followed by a subtype byte, because same_page
records must be longer than 1 byte.
trx_undo_page_init(): Combine the writes in WRITE record.
trx_undo_header_create(): Write 4 bytes using a special MEMSET
record that includes 1 bytes of length and 2 bytes of payload.
flst_write_addr(): Define as a static function. Combine the writes.
flst_zero_both(): Replaces two flst_zero_addr() calls.
flst_init(): Do not inline the function.
fsp_free_seg_inode(): Zerofill the whole inode.
fsp_apply_init_file_page(): Initialize FIL_PAGE_PREV,FIL_PAGE_NEXT
to FIL_NULL when using the physical format.
btr_create(): Assert !page_has_siblings() because fsp_apply_init_file_page()
must have been invoked.
fil_ibd_create(): Do not write FILE_MODIFY after FILE_CREATE.
fil_names_dirty_and_write(): Remove the parameter mtr.
Write the records using a separate mini-transaction object,
because any FILE_ records must be at the start of a mini-transaction log.
recv_recover_page(): Add a fil_space_t* parameter.
After applying log to the a ROW_FORMAT=COMPRESSED page,
invoke buf_zip_decompress() to restore the uncompressed page.
buf_page_io_complete(): Remove the temporary hack to discard the
uncompressed page of a ROW_FORMAT=COMPRESSED page.
page_zip_write_header(): Remove. Use mtr_t::write() or
mtr_t::memset() instead, and update the compressed page frame
separately.
trx_undo_header_add_space_for_xid(): Remove.
trx_undo_seg_create(): Perform the changes that were previously
made by trx_undo_header_add_space_for_xid().
btr_reset_instant(): New function: Reset the table to MariaDB 10.2
or 10.3 format when rolling back an instant ALTER TABLE operation.
page_rec_find_owner_rec(): Merge with the only callers.
page_cur_insert_rec_low(): Combine writes by using a local buffer.
MEMMOVE data from the preceding record whenever feasible
(copying at least 3 bytes).
page_cur_insert_rec_zip(): Combine writes to page header fields.
PageBulk::insertPage(): Issue MEMMOVE records to copy a matching
part from the preceding record.
PageBulk::finishPage(): Combine the writes to the page header
and to the sparse page directory slots.
mtr_t::write(): Only log the least significant (last) bytes
of multi-byte fields that actually differ.
For updating FSP_SIZE, we must always write all 4 bytes to the
redo log, so that the fil_space_set_recv_size() logic in
recv_sys_t::parse() will work.
mtr_t::memcpy(), mtr_t::zmemcpy(): Take a pointer argument
instead of a numeric offset to the page frame. Only log the
last bytes of multi-byte fields that actually differ.
In fil_space_crypt_t::write_page0(), we must log also any
unchanged bytes, so that recovery will recognize the record
and invoke fil_crypt_parse().
Future work:
MDEV-21724 Optimize page_cur_insert_rec_low() redo logging
MDEV-21725 Optimize btr_page_reorganize_low() redo logging
MDEV-21727 Optimize redo logging for ROW_FORMAT=COMPRESSED
2020-02-13 19:12:17 +02:00
|
|
|
/** Write FILE_MODIFY records when a non-predefined persistent
|
2016-08-12 11:17:45 +03:00
|
|
|
tablespace was modified for the first time since the latest
|
|
|
|
|
fil_names_clear().
|
MDEV-12353: Change the redo log encoding
log_t::FORMAT_10_5: physical redo log format tag
log_phys_t: Buffered records in the physical format.
The log record bytes will follow the last data field,
making use of alignment padding that would otherwise be wasted.
If there are multiple records for the same page, also those
may be appended to an existing log_phys_t object if the memory
is available.
In the physical format, the first byte of a record identifies the
record and its length (up to 15 bytes). For longer records, the
immediately following bytes will encode the remaining length
in a variable-length encoding. Usually, a variable-length-encoded
page identifier will follow, followed by optional payload, whose
length is included in the initially encoded total record length.
When a mini-transaction is updating multiple fields in a page,
it can avoid repeating the tablespace identifier and page number
by setting the same_page flag (most significant bit) in the first
byte of the log record. The byte offset of the record will be
relative to where the previous record for that page ended.
Until MDEV-14425 introduces a separate file-level log for
redo log checkpoints and file operations, we will write the
file-level records in the page-level redo log file.
The record FILE_CHECKPOINT (which replaces MLOG_CHECKPOINT)
will be removed in MDEV-14425, and one sequential scan of the
page recovery log will suffice.
Compared to MLOG_FILE_CREATE2, FILE_CREATE will not include any flags.
If the information is needed, it can be parsed from WRITE records that
modify FSP_SPACE_FLAGS.
MLOG_ZIP_WRITE_STRING: Remove. The record was only introduced temporarily
as part of this work, before being replaced with WRITE (along with
MLOG_WRITE_STRING, MLOG_1BYTE, MLOG_nBYTES).
mtr_buf_t::empty(): Check if the buffer is empty.
mtr_t::m_n_log_recs: Remove. It suffices to check if m_log is empty.
mtr_t::m_last, mtr_t::m_last_offset: End of the latest m_log record,
for the same_page encoding.
page_recv_t::last_offset: Reflects mtr_t::m_last_offset.
Valid values for last_offset during recovery should be 0 or above 8.
(The first 8 bytes of a page are the checksum and the page number,
and neither are ever updated directly by log records.)
Internally, the special value 1 indicates that the same_page form
will not be allowed for the subsequent record.
mtr_t::page_create(): Take the block descriptor as parameter,
so that it can be compared to mtr_t::m_last. The INIT_INDEX_PAGE
record will always followed by a subtype byte, because same_page
records must be longer than 1 byte.
trx_undo_page_init(): Combine the writes in WRITE record.
trx_undo_header_create(): Write 4 bytes using a special MEMSET
record that includes 1 bytes of length and 2 bytes of payload.
flst_write_addr(): Define as a static function. Combine the writes.
flst_zero_both(): Replaces two flst_zero_addr() calls.
flst_init(): Do not inline the function.
fsp_free_seg_inode(): Zerofill the whole inode.
fsp_apply_init_file_page(): Initialize FIL_PAGE_PREV,FIL_PAGE_NEXT
to FIL_NULL when using the physical format.
btr_create(): Assert !page_has_siblings() because fsp_apply_init_file_page()
must have been invoked.
fil_ibd_create(): Do not write FILE_MODIFY after FILE_CREATE.
fil_names_dirty_and_write(): Remove the parameter mtr.
Write the records using a separate mini-transaction object,
because any FILE_ records must be at the start of a mini-transaction log.
recv_recover_page(): Add a fil_space_t* parameter.
After applying log to the a ROW_FORMAT=COMPRESSED page,
invoke buf_zip_decompress() to restore the uncompressed page.
buf_page_io_complete(): Remove the temporary hack to discard the
uncompressed page of a ROW_FORMAT=COMPRESSED page.
page_zip_write_header(): Remove. Use mtr_t::write() or
mtr_t::memset() instead, and update the compressed page frame
separately.
trx_undo_header_add_space_for_xid(): Remove.
trx_undo_seg_create(): Perform the changes that were previously
made by trx_undo_header_add_space_for_xid().
btr_reset_instant(): New function: Reset the table to MariaDB 10.2
or 10.3 format when rolling back an instant ALTER TABLE operation.
page_rec_find_owner_rec(): Merge with the only callers.
page_cur_insert_rec_low(): Combine writes by using a local buffer.
MEMMOVE data from the preceding record whenever feasible
(copying at least 3 bytes).
page_cur_insert_rec_zip(): Combine writes to page header fields.
PageBulk::insertPage(): Issue MEMMOVE records to copy a matching
part from the preceding record.
PageBulk::finishPage(): Combine the writes to the page header
and to the sparse page directory slots.
mtr_t::write(): Only log the least significant (last) bytes
of multi-byte fields that actually differ.
For updating FSP_SIZE, we must always write all 4 bytes to the
redo log, so that the fil_space_set_recv_size() logic in
recv_sys_t::parse() will work.
mtr_t::memcpy(), mtr_t::zmemcpy(): Take a pointer argument
instead of a numeric offset to the page frame. Only log the
last bytes of multi-byte fields that actually differ.
In fil_space_crypt_t::write_page0(), we must log also any
unchanged bytes, so that recovery will recognize the record
and invoke fil_crypt_parse().
Future work:
MDEV-21724 Optimize page_cur_insert_rec_low() redo logging
MDEV-21725 Optimize btr_page_reorganize_low() redo logging
MDEV-21727 Optimize redo logging for ROW_FORMAT=COMPRESSED
2020-02-13 19:12:17 +02:00
|
|
|
@param[in,out] space tablespace */
|
|
|
|
|
void fil_names_dirty_and_write(fil_space_t* space);
|
2016-08-12 11:17:45 +03:00
|
|
|
|
MDEV-12353: Change the redo log encoding
log_t::FORMAT_10_5: physical redo log format tag
log_phys_t: Buffered records in the physical format.
The log record bytes will follow the last data field,
making use of alignment padding that would otherwise be wasted.
If there are multiple records for the same page, also those
may be appended to an existing log_phys_t object if the memory
is available.
In the physical format, the first byte of a record identifies the
record and its length (up to 15 bytes). For longer records, the
immediately following bytes will encode the remaining length
in a variable-length encoding. Usually, a variable-length-encoded
page identifier will follow, followed by optional payload, whose
length is included in the initially encoded total record length.
When a mini-transaction is updating multiple fields in a page,
it can avoid repeating the tablespace identifier and page number
by setting the same_page flag (most significant bit) in the first
byte of the log record. The byte offset of the record will be
relative to where the previous record for that page ended.
Until MDEV-14425 introduces a separate file-level log for
redo log checkpoints and file operations, we will write the
file-level records in the page-level redo log file.
The record FILE_CHECKPOINT (which replaces MLOG_CHECKPOINT)
will be removed in MDEV-14425, and one sequential scan of the
page recovery log will suffice.
Compared to MLOG_FILE_CREATE2, FILE_CREATE will not include any flags.
If the information is needed, it can be parsed from WRITE records that
modify FSP_SPACE_FLAGS.
MLOG_ZIP_WRITE_STRING: Remove. The record was only introduced temporarily
as part of this work, before being replaced with WRITE (along with
MLOG_WRITE_STRING, MLOG_1BYTE, MLOG_nBYTES).
mtr_buf_t::empty(): Check if the buffer is empty.
mtr_t::m_n_log_recs: Remove. It suffices to check if m_log is empty.
mtr_t::m_last, mtr_t::m_last_offset: End of the latest m_log record,
for the same_page encoding.
page_recv_t::last_offset: Reflects mtr_t::m_last_offset.
Valid values for last_offset during recovery should be 0 or above 8.
(The first 8 bytes of a page are the checksum and the page number,
and neither are ever updated directly by log records.)
Internally, the special value 1 indicates that the same_page form
will not be allowed for the subsequent record.
mtr_t::page_create(): Take the block descriptor as parameter,
so that it can be compared to mtr_t::m_last. The INIT_INDEX_PAGE
record will always followed by a subtype byte, because same_page
records must be longer than 1 byte.
trx_undo_page_init(): Combine the writes in WRITE record.
trx_undo_header_create(): Write 4 bytes using a special MEMSET
record that includes 1 bytes of length and 2 bytes of payload.
flst_write_addr(): Define as a static function. Combine the writes.
flst_zero_both(): Replaces two flst_zero_addr() calls.
flst_init(): Do not inline the function.
fsp_free_seg_inode(): Zerofill the whole inode.
fsp_apply_init_file_page(): Initialize FIL_PAGE_PREV,FIL_PAGE_NEXT
to FIL_NULL when using the physical format.
btr_create(): Assert !page_has_siblings() because fsp_apply_init_file_page()
must have been invoked.
fil_ibd_create(): Do not write FILE_MODIFY after FILE_CREATE.
fil_names_dirty_and_write(): Remove the parameter mtr.
Write the records using a separate mini-transaction object,
because any FILE_ records must be at the start of a mini-transaction log.
recv_recover_page(): Add a fil_space_t* parameter.
After applying log to the a ROW_FORMAT=COMPRESSED page,
invoke buf_zip_decompress() to restore the uncompressed page.
buf_page_io_complete(): Remove the temporary hack to discard the
uncompressed page of a ROW_FORMAT=COMPRESSED page.
page_zip_write_header(): Remove. Use mtr_t::write() or
mtr_t::memset() instead, and update the compressed page frame
separately.
trx_undo_header_add_space_for_xid(): Remove.
trx_undo_seg_create(): Perform the changes that were previously
made by trx_undo_header_add_space_for_xid().
btr_reset_instant(): New function: Reset the table to MariaDB 10.2
or 10.3 format when rolling back an instant ALTER TABLE operation.
page_rec_find_owner_rec(): Merge with the only callers.
page_cur_insert_rec_low(): Combine writes by using a local buffer.
MEMMOVE data from the preceding record whenever feasible
(copying at least 3 bytes).
page_cur_insert_rec_zip(): Combine writes to page header fields.
PageBulk::insertPage(): Issue MEMMOVE records to copy a matching
part from the preceding record.
PageBulk::finishPage(): Combine the writes to the page header
and to the sparse page directory slots.
mtr_t::write(): Only log the least significant (last) bytes
of multi-byte fields that actually differ.
For updating FSP_SIZE, we must always write all 4 bytes to the
redo log, so that the fil_space_set_recv_size() logic in
recv_sys_t::parse() will work.
mtr_t::memcpy(), mtr_t::zmemcpy(): Take a pointer argument
instead of a numeric offset to the page frame. Only log the
last bytes of multi-byte fields that actually differ.
In fil_space_crypt_t::write_page0(), we must log also any
unchanged bytes, so that recovery will recognize the record
and invoke fil_crypt_parse().
Future work:
MDEV-21724 Optimize page_cur_insert_rec_low() redo logging
MDEV-21725 Optimize btr_page_reorganize_low() redo logging
MDEV-21727 Optimize redo logging for ROW_FORMAT=COMPRESSED
2020-02-13 19:12:17 +02:00
|
|
|
/** Write FILE_MODIFY records if a persistent tablespace was modified
|
2016-08-12 11:17:45 +03:00
|
|
|
for the first time since the latest fil_names_clear().
|
|
|
|
|
@param[in,out] space tablespace
|
|
|
|
|
@param[in,out] mtr mini-transaction
|
MDEV-12353: Change the redo log encoding
log_t::FORMAT_10_5: physical redo log format tag
log_phys_t: Buffered records in the physical format.
The log record bytes will follow the last data field,
making use of alignment padding that would otherwise be wasted.
If there are multiple records for the same page, also those
may be appended to an existing log_phys_t object if the memory
is available.
In the physical format, the first byte of a record identifies the
record and its length (up to 15 bytes). For longer records, the
immediately following bytes will encode the remaining length
in a variable-length encoding. Usually, a variable-length-encoded
page identifier will follow, followed by optional payload, whose
length is included in the initially encoded total record length.
When a mini-transaction is updating multiple fields in a page,
it can avoid repeating the tablespace identifier and page number
by setting the same_page flag (most significant bit) in the first
byte of the log record. The byte offset of the record will be
relative to where the previous record for that page ended.
Until MDEV-14425 introduces a separate file-level log for
redo log checkpoints and file operations, we will write the
file-level records in the page-level redo log file.
The record FILE_CHECKPOINT (which replaces MLOG_CHECKPOINT)
will be removed in MDEV-14425, and one sequential scan of the
page recovery log will suffice.
Compared to MLOG_FILE_CREATE2, FILE_CREATE will not include any flags.
If the information is needed, it can be parsed from WRITE records that
modify FSP_SPACE_FLAGS.
MLOG_ZIP_WRITE_STRING: Remove. The record was only introduced temporarily
as part of this work, before being replaced with WRITE (along with
MLOG_WRITE_STRING, MLOG_1BYTE, MLOG_nBYTES).
mtr_buf_t::empty(): Check if the buffer is empty.
mtr_t::m_n_log_recs: Remove. It suffices to check if m_log is empty.
mtr_t::m_last, mtr_t::m_last_offset: End of the latest m_log record,
for the same_page encoding.
page_recv_t::last_offset: Reflects mtr_t::m_last_offset.
Valid values for last_offset during recovery should be 0 or above 8.
(The first 8 bytes of a page are the checksum and the page number,
and neither are ever updated directly by log records.)
Internally, the special value 1 indicates that the same_page form
will not be allowed for the subsequent record.
mtr_t::page_create(): Take the block descriptor as parameter,
so that it can be compared to mtr_t::m_last. The INIT_INDEX_PAGE
record will always followed by a subtype byte, because same_page
records must be longer than 1 byte.
trx_undo_page_init(): Combine the writes in WRITE record.
trx_undo_header_create(): Write 4 bytes using a special MEMSET
record that includes 1 bytes of length and 2 bytes of payload.
flst_write_addr(): Define as a static function. Combine the writes.
flst_zero_both(): Replaces two flst_zero_addr() calls.
flst_init(): Do not inline the function.
fsp_free_seg_inode(): Zerofill the whole inode.
fsp_apply_init_file_page(): Initialize FIL_PAGE_PREV,FIL_PAGE_NEXT
to FIL_NULL when using the physical format.
btr_create(): Assert !page_has_siblings() because fsp_apply_init_file_page()
must have been invoked.
fil_ibd_create(): Do not write FILE_MODIFY after FILE_CREATE.
fil_names_dirty_and_write(): Remove the parameter mtr.
Write the records using a separate mini-transaction object,
because any FILE_ records must be at the start of a mini-transaction log.
recv_recover_page(): Add a fil_space_t* parameter.
After applying log to the a ROW_FORMAT=COMPRESSED page,
invoke buf_zip_decompress() to restore the uncompressed page.
buf_page_io_complete(): Remove the temporary hack to discard the
uncompressed page of a ROW_FORMAT=COMPRESSED page.
page_zip_write_header(): Remove. Use mtr_t::write() or
mtr_t::memset() instead, and update the compressed page frame
separately.
trx_undo_header_add_space_for_xid(): Remove.
trx_undo_seg_create(): Perform the changes that were previously
made by trx_undo_header_add_space_for_xid().
btr_reset_instant(): New function: Reset the table to MariaDB 10.2
or 10.3 format when rolling back an instant ALTER TABLE operation.
page_rec_find_owner_rec(): Merge with the only callers.
page_cur_insert_rec_low(): Combine writes by using a local buffer.
MEMMOVE data from the preceding record whenever feasible
(copying at least 3 bytes).
page_cur_insert_rec_zip(): Combine writes to page header fields.
PageBulk::insertPage(): Issue MEMMOVE records to copy a matching
part from the preceding record.
PageBulk::finishPage(): Combine the writes to the page header
and to the sparse page directory slots.
mtr_t::write(): Only log the least significant (last) bytes
of multi-byte fields that actually differ.
For updating FSP_SIZE, we must always write all 4 bytes to the
redo log, so that the fil_space_set_recv_size() logic in
recv_sys_t::parse() will work.
mtr_t::memcpy(), mtr_t::zmemcpy(): Take a pointer argument
instead of a numeric offset to the page frame. Only log the
last bytes of multi-byte fields that actually differ.
In fil_space_crypt_t::write_page0(), we must log also any
unchanged bytes, so that recovery will recognize the record
and invoke fil_crypt_parse().
Future work:
MDEV-21724 Optimize page_cur_insert_rec_low() redo logging
MDEV-21725 Optimize btr_page_reorganize_low() redo logging
MDEV-21727 Optimize redo logging for ROW_FORMAT=COMPRESSED
2020-02-13 19:12:17 +02:00
|
|
|
@return whether any FILE_MODIFY record was written */
|
|
|
|
|
inline bool fil_names_write_if_was_clean(fil_space_t* space)
|
2016-08-12 11:17:45 +03:00
|
|
|
{
|
2020-10-26 16:43:52 +02:00
|
|
|
mysql_mutex_assert_owner(&log_sys.mutex);
|
2016-08-12 11:17:45 +03:00
|
|
|
|
|
|
|
|
if (space == NULL) {
|
|
|
|
|
return(false);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
const bool was_clean = space->max_lsn == 0;
|
2020-03-05 11:45:28 +02:00
|
|
|
ut_ad(space->max_lsn <= log_sys.get_lsn());
|
|
|
|
|
space->max_lsn = log_sys.get_lsn();
|
2016-08-12 11:17:45 +03:00
|
|
|
|
|
|
|
|
if (was_clean) {
|
MDEV-12353: Change the redo log encoding
log_t::FORMAT_10_5: physical redo log format tag
log_phys_t: Buffered records in the physical format.
The log record bytes will follow the last data field,
making use of alignment padding that would otherwise be wasted.
If there are multiple records for the same page, also those
may be appended to an existing log_phys_t object if the memory
is available.
In the physical format, the first byte of a record identifies the
record and its length (up to 15 bytes). For longer records, the
immediately following bytes will encode the remaining length
in a variable-length encoding. Usually, a variable-length-encoded
page identifier will follow, followed by optional payload, whose
length is included in the initially encoded total record length.
When a mini-transaction is updating multiple fields in a page,
it can avoid repeating the tablespace identifier and page number
by setting the same_page flag (most significant bit) in the first
byte of the log record. The byte offset of the record will be
relative to where the previous record for that page ended.
Until MDEV-14425 introduces a separate file-level log for
redo log checkpoints and file operations, we will write the
file-level records in the page-level redo log file.
The record FILE_CHECKPOINT (which replaces MLOG_CHECKPOINT)
will be removed in MDEV-14425, and one sequential scan of the
page recovery log will suffice.
Compared to MLOG_FILE_CREATE2, FILE_CREATE will not include any flags.
If the information is needed, it can be parsed from WRITE records that
modify FSP_SPACE_FLAGS.
MLOG_ZIP_WRITE_STRING: Remove. The record was only introduced temporarily
as part of this work, before being replaced with WRITE (along with
MLOG_WRITE_STRING, MLOG_1BYTE, MLOG_nBYTES).
mtr_buf_t::empty(): Check if the buffer is empty.
mtr_t::m_n_log_recs: Remove. It suffices to check if m_log is empty.
mtr_t::m_last, mtr_t::m_last_offset: End of the latest m_log record,
for the same_page encoding.
page_recv_t::last_offset: Reflects mtr_t::m_last_offset.
Valid values for last_offset during recovery should be 0 or above 8.
(The first 8 bytes of a page are the checksum and the page number,
and neither are ever updated directly by log records.)
Internally, the special value 1 indicates that the same_page form
will not be allowed for the subsequent record.
mtr_t::page_create(): Take the block descriptor as parameter,
so that it can be compared to mtr_t::m_last. The INIT_INDEX_PAGE
record will always followed by a subtype byte, because same_page
records must be longer than 1 byte.
trx_undo_page_init(): Combine the writes in WRITE record.
trx_undo_header_create(): Write 4 bytes using a special MEMSET
record that includes 1 bytes of length and 2 bytes of payload.
flst_write_addr(): Define as a static function. Combine the writes.
flst_zero_both(): Replaces two flst_zero_addr() calls.
flst_init(): Do not inline the function.
fsp_free_seg_inode(): Zerofill the whole inode.
fsp_apply_init_file_page(): Initialize FIL_PAGE_PREV,FIL_PAGE_NEXT
to FIL_NULL when using the physical format.
btr_create(): Assert !page_has_siblings() because fsp_apply_init_file_page()
must have been invoked.
fil_ibd_create(): Do not write FILE_MODIFY after FILE_CREATE.
fil_names_dirty_and_write(): Remove the parameter mtr.
Write the records using a separate mini-transaction object,
because any FILE_ records must be at the start of a mini-transaction log.
recv_recover_page(): Add a fil_space_t* parameter.
After applying log to the a ROW_FORMAT=COMPRESSED page,
invoke buf_zip_decompress() to restore the uncompressed page.
buf_page_io_complete(): Remove the temporary hack to discard the
uncompressed page of a ROW_FORMAT=COMPRESSED page.
page_zip_write_header(): Remove. Use mtr_t::write() or
mtr_t::memset() instead, and update the compressed page frame
separately.
trx_undo_header_add_space_for_xid(): Remove.
trx_undo_seg_create(): Perform the changes that were previously
made by trx_undo_header_add_space_for_xid().
btr_reset_instant(): New function: Reset the table to MariaDB 10.2
or 10.3 format when rolling back an instant ALTER TABLE operation.
page_rec_find_owner_rec(): Merge with the only callers.
page_cur_insert_rec_low(): Combine writes by using a local buffer.
MEMMOVE data from the preceding record whenever feasible
(copying at least 3 bytes).
page_cur_insert_rec_zip(): Combine writes to page header fields.
PageBulk::insertPage(): Issue MEMMOVE records to copy a matching
part from the preceding record.
PageBulk::finishPage(): Combine the writes to the page header
and to the sparse page directory slots.
mtr_t::write(): Only log the least significant (last) bytes
of multi-byte fields that actually differ.
For updating FSP_SIZE, we must always write all 4 bytes to the
redo log, so that the fil_space_set_recv_size() logic in
recv_sys_t::parse() will work.
mtr_t::memcpy(), mtr_t::zmemcpy(): Take a pointer argument
instead of a numeric offset to the page frame. Only log the
last bytes of multi-byte fields that actually differ.
In fil_space_crypt_t::write_page0(), we must log also any
unchanged bytes, so that recovery will recognize the record
and invoke fil_crypt_parse().
Future work:
MDEV-21724 Optimize page_cur_insert_rec_low() redo logging
MDEV-21725 Optimize btr_page_reorganize_low() redo logging
MDEV-21727 Optimize redo logging for ROW_FORMAT=COMPRESSED
2020-02-13 19:12:17 +02:00
|
|
|
fil_names_dirty_and_write(space);
|
2016-08-12 11:17:45 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return(was_clean);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/** On a log checkpoint, reset fil_names_dirty_and_write() flags
|
MDEV-12353: Change the redo log encoding
log_t::FORMAT_10_5: physical redo log format tag
log_phys_t: Buffered records in the physical format.
The log record bytes will follow the last data field,
making use of alignment padding that would otherwise be wasted.
If there are multiple records for the same page, also those
may be appended to an existing log_phys_t object if the memory
is available.
In the physical format, the first byte of a record identifies the
record and its length (up to 15 bytes). For longer records, the
immediately following bytes will encode the remaining length
in a variable-length encoding. Usually, a variable-length-encoded
page identifier will follow, followed by optional payload, whose
length is included in the initially encoded total record length.
When a mini-transaction is updating multiple fields in a page,
it can avoid repeating the tablespace identifier and page number
by setting the same_page flag (most significant bit) in the first
byte of the log record. The byte offset of the record will be
relative to where the previous record for that page ended.
Until MDEV-14425 introduces a separate file-level log for
redo log checkpoints and file operations, we will write the
file-level records in the page-level redo log file.
The record FILE_CHECKPOINT (which replaces MLOG_CHECKPOINT)
will be removed in MDEV-14425, and one sequential scan of the
page recovery log will suffice.
Compared to MLOG_FILE_CREATE2, FILE_CREATE will not include any flags.
If the information is needed, it can be parsed from WRITE records that
modify FSP_SPACE_FLAGS.
MLOG_ZIP_WRITE_STRING: Remove. The record was only introduced temporarily
as part of this work, before being replaced with WRITE (along with
MLOG_WRITE_STRING, MLOG_1BYTE, MLOG_nBYTES).
mtr_buf_t::empty(): Check if the buffer is empty.
mtr_t::m_n_log_recs: Remove. It suffices to check if m_log is empty.
mtr_t::m_last, mtr_t::m_last_offset: End of the latest m_log record,
for the same_page encoding.
page_recv_t::last_offset: Reflects mtr_t::m_last_offset.
Valid values for last_offset during recovery should be 0 or above 8.
(The first 8 bytes of a page are the checksum and the page number,
and neither are ever updated directly by log records.)
Internally, the special value 1 indicates that the same_page form
will not be allowed for the subsequent record.
mtr_t::page_create(): Take the block descriptor as parameter,
so that it can be compared to mtr_t::m_last. The INIT_INDEX_PAGE
record will always followed by a subtype byte, because same_page
records must be longer than 1 byte.
trx_undo_page_init(): Combine the writes in WRITE record.
trx_undo_header_create(): Write 4 bytes using a special MEMSET
record that includes 1 bytes of length and 2 bytes of payload.
flst_write_addr(): Define as a static function. Combine the writes.
flst_zero_both(): Replaces two flst_zero_addr() calls.
flst_init(): Do not inline the function.
fsp_free_seg_inode(): Zerofill the whole inode.
fsp_apply_init_file_page(): Initialize FIL_PAGE_PREV,FIL_PAGE_NEXT
to FIL_NULL when using the physical format.
btr_create(): Assert !page_has_siblings() because fsp_apply_init_file_page()
must have been invoked.
fil_ibd_create(): Do not write FILE_MODIFY after FILE_CREATE.
fil_names_dirty_and_write(): Remove the parameter mtr.
Write the records using a separate mini-transaction object,
because any FILE_ records must be at the start of a mini-transaction log.
recv_recover_page(): Add a fil_space_t* parameter.
After applying log to the a ROW_FORMAT=COMPRESSED page,
invoke buf_zip_decompress() to restore the uncompressed page.
buf_page_io_complete(): Remove the temporary hack to discard the
uncompressed page of a ROW_FORMAT=COMPRESSED page.
page_zip_write_header(): Remove. Use mtr_t::write() or
mtr_t::memset() instead, and update the compressed page frame
separately.
trx_undo_header_add_space_for_xid(): Remove.
trx_undo_seg_create(): Perform the changes that were previously
made by trx_undo_header_add_space_for_xid().
btr_reset_instant(): New function: Reset the table to MariaDB 10.2
or 10.3 format when rolling back an instant ALTER TABLE operation.
page_rec_find_owner_rec(): Merge with the only callers.
page_cur_insert_rec_low(): Combine writes by using a local buffer.
MEMMOVE data from the preceding record whenever feasible
(copying at least 3 bytes).
page_cur_insert_rec_zip(): Combine writes to page header fields.
PageBulk::insertPage(): Issue MEMMOVE records to copy a matching
part from the preceding record.
PageBulk::finishPage(): Combine the writes to the page header
and to the sparse page directory slots.
mtr_t::write(): Only log the least significant (last) bytes
of multi-byte fields that actually differ.
For updating FSP_SIZE, we must always write all 4 bytes to the
redo log, so that the fil_space_set_recv_size() logic in
recv_sys_t::parse() will work.
mtr_t::memcpy(), mtr_t::zmemcpy(): Take a pointer argument
instead of a numeric offset to the page frame. Only log the
last bytes of multi-byte fields that actually differ.
In fil_space_crypt_t::write_page0(), we must log also any
unchanged bytes, so that recovery will recognize the record
and invoke fil_crypt_parse().
Future work:
MDEV-21724 Optimize page_cur_insert_rec_low() redo logging
MDEV-21725 Optimize btr_page_reorganize_low() redo logging
MDEV-21727 Optimize redo logging for ROW_FORMAT=COMPRESSED
2020-02-13 19:12:17 +02:00
|
|
|
and write out FILE_MODIFY and FILE_CHECKPOINT if needed.
|
2016-08-12 11:17:45 +03:00
|
|
|
@param[in] lsn checkpoint LSN
|
MDEV-12353: Change the redo log encoding
log_t::FORMAT_10_5: physical redo log format tag
log_phys_t: Buffered records in the physical format.
The log record bytes will follow the last data field,
making use of alignment padding that would otherwise be wasted.
If there are multiple records for the same page, also those
may be appended to an existing log_phys_t object if the memory
is available.
In the physical format, the first byte of a record identifies the
record and its length (up to 15 bytes). For longer records, the
immediately following bytes will encode the remaining length
in a variable-length encoding. Usually, a variable-length-encoded
page identifier will follow, followed by optional payload, whose
length is included in the initially encoded total record length.
When a mini-transaction is updating multiple fields in a page,
it can avoid repeating the tablespace identifier and page number
by setting the same_page flag (most significant bit) in the first
byte of the log record. The byte offset of the record will be
relative to where the previous record for that page ended.
Until MDEV-14425 introduces a separate file-level log for
redo log checkpoints and file operations, we will write the
file-level records in the page-level redo log file.
The record FILE_CHECKPOINT (which replaces MLOG_CHECKPOINT)
will be removed in MDEV-14425, and one sequential scan of the
page recovery log will suffice.
Compared to MLOG_FILE_CREATE2, FILE_CREATE will not include any flags.
If the information is needed, it can be parsed from WRITE records that
modify FSP_SPACE_FLAGS.
MLOG_ZIP_WRITE_STRING: Remove. The record was only introduced temporarily
as part of this work, before being replaced with WRITE (along with
MLOG_WRITE_STRING, MLOG_1BYTE, MLOG_nBYTES).
mtr_buf_t::empty(): Check if the buffer is empty.
mtr_t::m_n_log_recs: Remove. It suffices to check if m_log is empty.
mtr_t::m_last, mtr_t::m_last_offset: End of the latest m_log record,
for the same_page encoding.
page_recv_t::last_offset: Reflects mtr_t::m_last_offset.
Valid values for last_offset during recovery should be 0 or above 8.
(The first 8 bytes of a page are the checksum and the page number,
and neither are ever updated directly by log records.)
Internally, the special value 1 indicates that the same_page form
will not be allowed for the subsequent record.
mtr_t::page_create(): Take the block descriptor as parameter,
so that it can be compared to mtr_t::m_last. The INIT_INDEX_PAGE
record will always followed by a subtype byte, because same_page
records must be longer than 1 byte.
trx_undo_page_init(): Combine the writes in WRITE record.
trx_undo_header_create(): Write 4 bytes using a special MEMSET
record that includes 1 bytes of length and 2 bytes of payload.
flst_write_addr(): Define as a static function. Combine the writes.
flst_zero_both(): Replaces two flst_zero_addr() calls.
flst_init(): Do not inline the function.
fsp_free_seg_inode(): Zerofill the whole inode.
fsp_apply_init_file_page(): Initialize FIL_PAGE_PREV,FIL_PAGE_NEXT
to FIL_NULL when using the physical format.
btr_create(): Assert !page_has_siblings() because fsp_apply_init_file_page()
must have been invoked.
fil_ibd_create(): Do not write FILE_MODIFY after FILE_CREATE.
fil_names_dirty_and_write(): Remove the parameter mtr.
Write the records using a separate mini-transaction object,
because any FILE_ records must be at the start of a mini-transaction log.
recv_recover_page(): Add a fil_space_t* parameter.
After applying log to the a ROW_FORMAT=COMPRESSED page,
invoke buf_zip_decompress() to restore the uncompressed page.
buf_page_io_complete(): Remove the temporary hack to discard the
uncompressed page of a ROW_FORMAT=COMPRESSED page.
page_zip_write_header(): Remove. Use mtr_t::write() or
mtr_t::memset() instead, and update the compressed page frame
separately.
trx_undo_header_add_space_for_xid(): Remove.
trx_undo_seg_create(): Perform the changes that were previously
made by trx_undo_header_add_space_for_xid().
btr_reset_instant(): New function: Reset the table to MariaDB 10.2
or 10.3 format when rolling back an instant ALTER TABLE operation.
page_rec_find_owner_rec(): Merge with the only callers.
page_cur_insert_rec_low(): Combine writes by using a local buffer.
MEMMOVE data from the preceding record whenever feasible
(copying at least 3 bytes).
page_cur_insert_rec_zip(): Combine writes to page header fields.
PageBulk::insertPage(): Issue MEMMOVE records to copy a matching
part from the preceding record.
PageBulk::finishPage(): Combine the writes to the page header
and to the sparse page directory slots.
mtr_t::write(): Only log the least significant (last) bytes
of multi-byte fields that actually differ.
For updating FSP_SIZE, we must always write all 4 bytes to the
redo log, so that the fil_space_set_recv_size() logic in
recv_sys_t::parse() will work.
mtr_t::memcpy(), mtr_t::zmemcpy(): Take a pointer argument
instead of a numeric offset to the page frame. Only log the
last bytes of multi-byte fields that actually differ.
In fil_space_crypt_t::write_page0(), we must log also any
unchanged bytes, so that recovery will recognize the record
and invoke fil_crypt_parse().
Future work:
MDEV-21724 Optimize page_cur_insert_rec_low() redo logging
MDEV-21725 Optimize btr_page_reorganize_low() redo logging
MDEV-21727 Optimize redo logging for ROW_FORMAT=COMPRESSED
2020-02-13 19:12:17 +02:00
|
|
|
@param[in] do_write whether to always write FILE_CHECKPOINT
|
2016-08-12 11:17:45 +03:00
|
|
|
@return whether anything was written to the redo log
|
|
|
|
|
@retval false if no flags were set and nothing written
|
|
|
|
|
@retval true if anything was written to the redo log */
|
|
|
|
|
bool
|
|
|
|
|
fil_names_clear(
|
|
|
|
|
lsn_t lsn,
|
|
|
|
|
bool do_write);
|
|
|
|
|
|
|
|
|
|
#ifdef UNIV_ENABLE_UNIT_TEST_MAKE_FILEPATH
|
|
|
|
|
void test_make_filepath();
|
|
|
|
|
#endif /* UNIV_ENABLE_UNIT_TEST_MAKE_FILEPATH */
|
2014-05-06 21:13:16 +02:00
|
|
|
|
2017-04-21 11:28:18 +03:00
|
|
|
/** Determine the block size of the data file.
|
|
|
|
|
@param[in] space tablespace
|
|
|
|
|
@param[in] offset page number
|
2015-04-01 19:37:00 +03:00
|
|
|
@return block size */
|
MDEV-25791: Remove UNIV_INTERN
Back in 2006 or 2007, when MySQL AB and Innobase Oy existed as
separately controlled entities (Innobase had been acquired by
Oracle Corporation), MySQL 5.1 introduced a storage engine plugin
interface and Oracle made use of it by distributing a separate
InnoDB Plugin, which would contain some more bug fixes and
improvements, compared to the version of InnoDB that was statically
linked with the mysqld server that was distributed by MySQL AB.
The built-in InnoDB would export global symbols, which would clash
with the symbols of the dynamic InnoDB Plugin (which was supposed
to override the built-in one when present).
The solution to this problem was to declare all global symbols with
UNIV_INTERN, so that they would get the GCC function attribute that
specifies hidden visibility.
Later, in MariaDB Server, something based on Percona XtraDB (a fork of
MySQL InnoDB) became the statically linked implementation, and something
closer to MySQL InnoDB was available as a dynamic plugin. Starting with
version 10.2, MariaDB Server includes only one InnoDB implementation,
and hence any reason to have the UNIV_INTERN definition was lost.
btr_get_size_and_reserved(): Move to the same compilation unit with
the only caller.
innodb_set_buf_pool_size(): Remove. Modify innobase_buffer_pool_size
directly.
fil_crypt_calculate_checksum(): Merge to the only caller.
ha_innobase::innobase_reset_autoinc(): Merge to the only caller.
thd_query_start_micro(): Remove. Call thd_start_utime() directly.
2021-05-27 10:13:14 +03:00
|
|
|
ulint fil_space_get_block_size(const fil_space_t* space, unsigned offset);
|
2014-07-31 11:31:39 +03:00
|
|
|
|
2022-01-13 17:27:28 +01:00
|
|
|
#include "fil0fil.inl"
|
2016-04-22 10:50:45 +02:00
|
|
|
#endif /* UNIV_INNOCHECKSUM */
|
2014-07-31 11:31:39 +03:00
|
|
|
|
2014-02-26 19:11:54 +01:00
|
|
|
#endif /* fil0fil_h */
|