mirror of
https://github.com/MariaDB/server.git
synced 2026-10-23 18:49:29 +03:00
Implement an improved binlog implementation that is integrated into the storage engine. The new implementation is enabled with the --binlog-storage-engine option. Initially the InnoDB storage engine implements the binlog. Integrating the binlog in the storage engine improves performance, since it makes the InnoDB redo log the single source of truth and avoids the need for expensive two-phase commit between binlog and engine. It also makes it possible to disable durability (set --innodb-flush-log-at-trx-commit=0) to further improve performance, while still preserving the ability to recover the binlog and database into a consistent state after a crash. The new binlog implementation also greatly improves the internal design and implementation of the binlog, and enables future enhancements for replication. This is a squash of the original 11.4-based patch series. Signed-off-by: Kristian Nielsen <knielsen@knielsen-hq.org>
460 lines
12 KiB
C++
460 lines
12 KiB
C++
/*
|
|
Copyright (c) 2023, MariaDB plc
|
|
|
|
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., 51 Franklin St, Fifth Floor, Boston, MA 02110-1335 USA
|
|
*/
|
|
|
|
#include "my_global.h"
|
|
#include "handler.h"
|
|
#include "sql_class.h"
|
|
#include "log_cache.h"
|
|
|
|
typedef void *sv_id_t;
|
|
|
|
static int online_alter_close_connection(THD *thd);
|
|
static int online_alter_savepoint_set(THD *thd, sv_id_t sv_id);
|
|
static int online_alter_savepoint_rollback(THD *thd, sv_id_t sv_id);
|
|
static int online_alter_commit(THD *thd, bool all);
|
|
static int online_alter_rollback(THD *thd, bool all);
|
|
static int online_alter_prepare(THD *thd, bool all);
|
|
static int online_alter_commit_by_xid(XID *x);
|
|
static int online_alter_rollback_by_xid(XID *x);
|
|
|
|
static transaction_participant online_alter_tp=
|
|
{
|
|
0, 0, HTON_NO_ROLLBACK,
|
|
online_alter_close_connection,
|
|
online_alter_savepoint_set, online_alter_savepoint_rollback,
|
|
[](THD *thd){ return true; }, /*savepoint_rollback_can_release_mdl*/
|
|
NULL, /*savepoint_release*/
|
|
online_alter_commit, online_alter_rollback, online_alter_prepare,
|
|
[](XID*, uint){ return 0; }, /*recover*/
|
|
online_alter_commit_by_xid,
|
|
online_alter_rollback_by_xid,
|
|
NULL, NULL,
|
|
NULL, NULL, NULL, NULL, NULL /* snapshot, *_ordered, checkpoint, versioned*/
|
|
};
|
|
|
|
struct Online_alter_cache_list: ilist<online_alter_cache_data>
|
|
{
|
|
sv_id_t savepoint_id= 0;
|
|
};
|
|
|
|
struct table_savepoint: ilist_node<>
|
|
{
|
|
sv_id_t id;
|
|
my_off_t log_prev_pos;
|
|
table_savepoint(sv_id_t id, my_off_t pos): id(id), log_prev_pos(pos){}
|
|
};
|
|
|
|
class online_alter_cache_data: public ilist_node<>,
|
|
public binlog_cache_data
|
|
{
|
|
public:
|
|
online_alter_cache_data() : binlog_cache_data(false, false),
|
|
hton(nullptr), sink_log(nullptr) { }
|
|
void store_prev_position()
|
|
{
|
|
before_stmt_pos= my_b_write_tell(&cache_log);
|
|
}
|
|
|
|
handlerton *hton;
|
|
Cache_flip_event_log *sink_log;
|
|
ilist<table_savepoint> sv_list;
|
|
/**
|
|
Finds savepoint with specified id and returns its associated data.
|
|
Cleans up all the savepoints up to the found one.
|
|
*/
|
|
my_off_t pop_sv_until(sv_id_t id)
|
|
{
|
|
my_off_t pos= 0;
|
|
auto it= sv_list.begin();
|
|
auto sentinel= it->prev;
|
|
while (pos == 0 && it != sv_list.end())
|
|
{
|
|
table_savepoint &sv= *it;
|
|
++it;
|
|
if (sv.id == id)
|
|
pos= sv.log_prev_pos;
|
|
delete &sv;
|
|
}
|
|
sentinel->next= &*it; // drop the range from the list
|
|
it->prev= sentinel;
|
|
return pos;
|
|
}
|
|
void cleanup_sv()
|
|
{
|
|
pop_sv_until((sv_id_t)1); // Erase the whole list
|
|
}
|
|
};
|
|
|
|
|
|
static
|
|
online_alter_cache_data *setup_cache_data(MEM_ROOT *root, TABLE_SHARE *share)
|
|
{
|
|
static ulong online_alter_cache_use= 0, online_alter_cache_disk_use= 0;
|
|
|
|
auto cache= new online_alter_cache_data();
|
|
if (!cache || open_cached_file(&cache->cache_log, mysql_tmpdir, LOG_PREFIX,
|
|
(size_t)binlog_cache_size, MYF(MY_WME)))
|
|
{
|
|
delete cache;
|
|
return NULL;
|
|
}
|
|
|
|
share->online_alter_binlog->acquire();
|
|
cache->hton= share->db_type();
|
|
cache->sink_log= share->online_alter_binlog;
|
|
|
|
my_off_t binlog_max_size= SIZE_T_MAX; // maximum possible cache size
|
|
DBUG_EXECUTE_IF("online_alter_small_cache", binlog_max_size= 4096;);
|
|
|
|
cache->set_binlog_cache_info(binlog_max_size,
|
|
&online_alter_cache_use,
|
|
&online_alter_cache_disk_use);
|
|
cache->store_prev_position();
|
|
return cache;
|
|
}
|
|
|
|
|
|
static Online_alter_cache_list &get_cache_list(transaction_participant *ht,
|
|
THD *thd)
|
|
{
|
|
void *data= thd_get_ha_data(thd, ht);
|
|
DBUG_ASSERT(data);
|
|
return *(Online_alter_cache_list*)data;
|
|
}
|
|
|
|
|
|
static Online_alter_cache_list &get_or_create_cache_list(THD *thd)
|
|
{
|
|
void *data= thd_get_ha_data(thd, &online_alter_tp);
|
|
if (!data)
|
|
{
|
|
data= new Online_alter_cache_list();
|
|
thd_set_ha_data(thd, &online_alter_tp, data);
|
|
}
|
|
return *(Online_alter_cache_list*)data;
|
|
}
|
|
|
|
|
|
static online_alter_cache_data* get_cache_data(THD *thd, TABLE *table)
|
|
{
|
|
auto &cache_list= get_or_create_cache_list(thd);
|
|
/* we assume it's very rare to have more than one online ALTER running */
|
|
for (auto &cache: cache_list)
|
|
{
|
|
if (cache.sink_log == table->s->online_alter_binlog)
|
|
return &cache;
|
|
}
|
|
|
|
MEM_ROOT *root= &thd->transaction->mem_root;
|
|
auto *new_cache_data= setup_cache_data(root, table->s);
|
|
cache_list.push_back(*new_cache_data);
|
|
|
|
return new_cache_data;
|
|
}
|
|
|
|
|
|
int online_alter_log_row(TABLE* table, const uchar *before_record,
|
|
const uchar *after_record, Log_func *log_func)
|
|
{
|
|
THD *thd= table->in_use;
|
|
|
|
if (!table->online_alter_cache)
|
|
{
|
|
table->online_alter_cache= get_cache_data(thd, table);
|
|
DBUG_ASSERT(table->online_alter_cache->cache_log.type == WRITE_CACHE);
|
|
trans_register_ha(thd, false, &online_alter_tp, 0);
|
|
if (thd->in_multi_stmt_transaction_mode())
|
|
trans_register_ha(thd, true, &online_alter_tp, 0);
|
|
}
|
|
|
|
// We need to log all columns for the case if alter table changes primary key
|
|
DBUG_ASSERT(!before_record || bitmap_is_set_all(table->read_set));
|
|
MY_BITMAP *old_rpl_write_set= table->rpl_write_set;
|
|
table->rpl_write_set= &table->s->all_set;
|
|
|
|
table->online_alter_cache->store_prev_position();
|
|
int error= (*log_func)(thd, table, table->s->online_alter_binlog,
|
|
table->online_alter_cache,
|
|
table->file->has_transactions_and_rollback(),
|
|
BINLOG_ROW_IMAGE_FULL,
|
|
before_record, after_record);
|
|
|
|
table->rpl_write_set= old_rpl_write_set;
|
|
|
|
if (unlikely(error))
|
|
{
|
|
table->online_alter_cache->restore_prev_position();
|
|
return HA_ERR_RBR_LOGGING_FAILED;
|
|
}
|
|
|
|
return 0;
|
|
}
|
|
|
|
|
|
static void cleanup_cache_list(ilist<online_alter_cache_data> &list)
|
|
{
|
|
auto it= list.begin();
|
|
while (it != list.end())
|
|
{
|
|
auto &cache= *it++;
|
|
cache.sink_log->release();
|
|
cache.reset();
|
|
cache.cleanup_sv();
|
|
delete &cache;
|
|
}
|
|
list.clear();
|
|
DBUG_ASSERT(list.empty());
|
|
}
|
|
|
|
|
|
static
|
|
int online_alter_end_trans(Online_alter_cache_list &cache_list, THD *thd,
|
|
bool is_ending_transaction, bool commit)
|
|
{
|
|
DBUG_ENTER("online_alter_end_trans");
|
|
int error= 0;
|
|
|
|
if (cache_list.empty())
|
|
DBUG_RETURN(0);
|
|
|
|
for (auto &cache: cache_list)
|
|
{
|
|
auto *binlog= cache.sink_log;
|
|
DBUG_ASSERT(binlog);
|
|
bool non_trans= cache.hton->flags & HTON_NO_ROLLBACK // Aria
|
|
|| !cache.hton->rollback;
|
|
bool do_commit= (commit && is_ending_transaction) || non_trans;
|
|
|
|
if (commit || non_trans)
|
|
{
|
|
// Do not set STMT_END for last event to leave table open in altering thd
|
|
error= binlog_flush_pending_rows_event(thd, false, true, binlog, &cache);
|
|
if (error)
|
|
do_commit= commit= 0;
|
|
}
|
|
|
|
if (do_commit)
|
|
{
|
|
/*
|
|
If the cache wasn't reinited to write, then it remains empty after
|
|
the last write.
|
|
*/
|
|
if (my_b_bytes_in_cache(&cache.cache_log) && likely(!error))
|
|
{
|
|
DBUG_ASSERT(cache.cache_log.type != READ_CACHE);
|
|
mysql_mutex_lock(binlog->get_log_lock());
|
|
error= binlog->write_cache_raw(thd, &cache.cache_log);
|
|
mysql_mutex_unlock(binlog->get_log_lock());
|
|
if (!is_ending_transaction)
|
|
cache.reset();
|
|
}
|
|
}
|
|
else if (!commit) // rollback
|
|
{
|
|
DBUG_ASSERT(!non_trans);
|
|
cache.restore_prev_position();
|
|
}
|
|
else
|
|
{
|
|
DBUG_ASSERT(!is_ending_transaction);
|
|
cache.store_prev_position();
|
|
}
|
|
|
|
|
|
if (error)
|
|
{
|
|
my_error(ER_ERROR_ON_WRITE, MYF(ME_ERROR_LOG),
|
|
binlog->get_name(), errno);
|
|
break;
|
|
}
|
|
}
|
|
|
|
if (is_ending_transaction)
|
|
cleanup_cache_list(cache_list);
|
|
|
|
DBUG_RETURN(error);
|
|
}
|
|
|
|
void cleanup_tables(THD *thd)
|
|
{
|
|
for (TABLE *table= thd->open_tables; table; table= table->next)
|
|
table->online_alter_cache= NULL;
|
|
}
|
|
|
|
static
|
|
int online_alter_savepoint_set(THD *thd, sv_id_t sv_id)
|
|
{
|
|
DBUG_ENTER("binlog_online_alter_savepoint");
|
|
auto &cache_list= get_cache_list(&online_alter_tp, thd);
|
|
if (cache_list.empty())
|
|
DBUG_RETURN(0);
|
|
|
|
for (auto &cache: cache_list)
|
|
{
|
|
if (cache.hton->savepoint_set == NULL)
|
|
continue;
|
|
|
|
auto *sv= new table_savepoint(sv_id, cache.get_byte_position());
|
|
if(unlikely(sv == NULL))
|
|
DBUG_RETURN(1);
|
|
cache.sv_list.push_front(*sv);
|
|
}
|
|
DBUG_RETURN(0);
|
|
}
|
|
|
|
static
|
|
int online_alter_savepoint_rollback(THD *thd, sv_id_t sv_id)
|
|
{
|
|
DBUG_ENTER("online_alter_savepoint_rollback");
|
|
|
|
auto &cache_list= get_cache_list(&online_alter_tp, thd);
|
|
for (auto &cache: cache_list)
|
|
{
|
|
if (cache.hton->savepoint_set == NULL)
|
|
continue;
|
|
|
|
// There's no savepoint if it was set up before online table was modified.
|
|
// In that case, restore to 0.
|
|
my_off_t pos= cache.pop_sv_until(sv_id);
|
|
|
|
cache.restore_savepoint(pos);
|
|
}
|
|
|
|
DBUG_RETURN(0);
|
|
}
|
|
|
|
static int online_alter_commit_by_xid(XID *x)
|
|
{
|
|
auto *xid= static_cast<XA_data*>(x);
|
|
if (likely(xid->online_alter_cache == NULL))
|
|
return 1;
|
|
int res= online_alter_end_trans(*xid->online_alter_cache, current_thd,
|
|
true, true);
|
|
delete xid->online_alter_cache;
|
|
xid->online_alter_cache= NULL;
|
|
return res;
|
|
};
|
|
|
|
static int online_alter_rollback_by_xid(XID *x)
|
|
{
|
|
auto *xid= static_cast<XA_data*>(x);
|
|
if (likely(xid->online_alter_cache == NULL))
|
|
return 1;
|
|
int res= online_alter_end_trans(*xid->online_alter_cache, current_thd,
|
|
true, false);
|
|
delete xid->online_alter_cache;
|
|
xid->online_alter_cache= NULL;
|
|
return res;
|
|
};
|
|
|
|
static int online_alter_commit(THD *thd, bool all)
|
|
{
|
|
int res;
|
|
bool is_ending_transaction= ending_trans(thd, all);
|
|
if (is_ending_transaction
|
|
&& thd->transaction->xid_state.get_state_code() == XA_PREPARED)
|
|
{
|
|
res= online_alter_commit_by_xid(thd->transaction->xid_state.get_xid());
|
|
// cleanup was already done by prepare()
|
|
}
|
|
else
|
|
{
|
|
res= online_alter_end_trans(get_cache_list(&online_alter_tp, thd), thd,
|
|
is_ending_transaction, true);
|
|
cleanup_tables(thd);
|
|
}
|
|
return res;
|
|
};
|
|
|
|
static int online_alter_rollback(THD *thd, bool all)
|
|
{
|
|
int res;
|
|
bool is_ending_transaction= ending_trans(thd, all);
|
|
if (is_ending_transaction &&
|
|
(thd->transaction->xid_state.get_state_code() == XA_PREPARED ||
|
|
thd->transaction->xid_state.get_state_code() == XA_ROLLBACK_ONLY))
|
|
{
|
|
res= online_alter_rollback_by_xid(thd->transaction->xid_state.get_xid());
|
|
// cleanup was already done by prepare()
|
|
}
|
|
else
|
|
{
|
|
res= online_alter_end_trans(get_cache_list(&online_alter_tp, thd), thd,
|
|
is_ending_transaction, false);
|
|
cleanup_tables(thd);
|
|
}
|
|
return res;
|
|
};
|
|
|
|
static int online_alter_prepare(THD *thd, bool all)
|
|
{
|
|
auto &cache_list= get_cache_list(&online_alter_tp, thd);
|
|
int res= 0;
|
|
if (ending_trans(thd, all))
|
|
{
|
|
thd->transaction->xid_state.set_online_alter_cache(&cache_list);
|
|
thd_set_ha_data(thd, &online_alter_tp, NULL);
|
|
}
|
|
else
|
|
{
|
|
res= online_alter_end_trans(cache_list, thd, false, true);
|
|
}
|
|
|
|
cleanup_tables(thd);
|
|
return res;
|
|
};
|
|
|
|
static int online_alter_close_connection(THD *thd)
|
|
{
|
|
auto *cache_list= (Online_alter_cache_list*)thd_get_ha_data(thd, &online_alter_tp);
|
|
|
|
DBUG_ASSERT(!cache_list || cache_list->empty());
|
|
delete cache_list;
|
|
thd_set_ha_data(thd, &online_alter_tp, NULL);
|
|
return 0;
|
|
}
|
|
|
|
|
|
static int online_alter_log_init(void *p)
|
|
{
|
|
auto plugin= (st_plugin_int*)p;
|
|
plugin->data= &online_alter_tp;
|
|
return setup_transaction_participant(plugin);
|
|
}
|
|
|
|
struct st_mysql_daemon online_alter_plugin=
|
|
{ MYSQL_DAEMON_INTERFACE_VERSION };
|
|
|
|
maria_declare_plugin(online_alter_log)
|
|
{
|
|
MYSQL_DAEMON_PLUGIN,
|
|
&online_alter_plugin,
|
|
"online_alter_log",
|
|
"MariaDB PLC",
|
|
"This is a plugin to represent the online alter log in a transaction",
|
|
PLUGIN_LICENSE_GPL,
|
|
online_alter_log_init,
|
|
NULL,
|
|
0x0200, // 2.0
|
|
NULL, // no status vars
|
|
NULL, // no sysvars
|
|
"2.0",
|
|
MariaDB_PLUGIN_MATURITY_STABLE
|
|
}
|
|
maria_declare_plugin_end;
|