mariadb/mysql-test/main/lock_multi.test

1064 lines
26 KiB
Text
Raw Normal View History

# Save the initial number of concurrent sessions
--source include/count_sessions.inc
Changed mysql-test to print warnings for not existing table to DROP TABLE Cleaned up test; Removed wrong DROP TABLE commands and use standard table and database names. changed store_warning() -> push_warning_print() BitKeeper/deleted/.del-rpl000016-slave.opt~ef76f85ddcc13b87: Delete: mysql-test/t/rpl000016-slave.opt BitKeeper/deleted/.del-sel000001.test~9567c1646058cc: Delete: mysql-test/t/sel000001.test BitKeeper/deleted/.del-sel000002.test~9f500639572e18e1: Delete: mysql-test/t/sel000002.test BitKeeper/deleted/.del-sel000003.test~63a5512d18cd20a2: Delete: mysql-test/t/sel000003.test BitKeeper/deleted/.del-sel000001.result~383913ae4505ec86: Delete: mysql-test/r/sel000001.result BitKeeper/deleted/.del-sel000002.result~d1787e6fd5dbc1cc: Delete: mysql-test/r/sel000002.result BitKeeper/deleted/.del-sel000003.result~d7b657b1e3a286a7: Delete: mysql-test/r/sel000003.result BitKeeper/deleted/.del-sel000031.result~d49aeac63ad7db4d: Delete: mysql-test/r/sel000031.result BitKeeper/deleted/.del-sel000031.test~50a19a8e204e99bc: Delete: mysql-test/t/sel000031.test BitKeeper/deleted/.del-sel000032.result~6cb30e23cbca9fb0: Delete: mysql-test/r/sel000032.result BitKeeper/deleted/.del-sel000032.test~e32da7c3fc4b7ace: Delete: mysql-test/t/sel000032.test BitKeeper/deleted/.del-rpl000003.result~68d6ee00beaa011: Delete: mysql-test/r/rpl000003.result BitKeeper/deleted/.del-rpl000003.test~b7cfc4c5576fbafd: Delete: mysql-test/t/rpl000003.test client/mysql.cc: Don't yet print information about SQL help client/mysqltest.c: Added test options: --enable_warnings --disable_warnings --enable_info --disable_info configure.in: changed version number of shared libraries mysql-test/include/master-slave.inc: Don't write warnings on init mysql-test/r/backup.result: Updated results mysql-test/r/bdb.result: Updated results mysql-test/r/bigint.result: Updated results mysql-test/r/bool.result: Updated results mysql-test/r/create.result: Updated results mysql-test/r/delete.result: Updated results mysql-test/r/derived.result: Updated results mysql-test/r/distinct.result: Updated results mysql-test/r/drop.result: Updated results mysql-test/r/flush.result: Updated results mysql-test/r/fulltext.result: Updated results mysql-test/r/fulltext_multi.result: Updated results mysql-test/r/fulltext_order_by.result: Updated results mysql-test/r/func_equal.result: Updated results mysql-test/r/func_in.result: Updated results mysql-test/r/func_set.result: Updated results mysql-test/r/gcc296.result: Updated results mysql-test/r/group_by.result: Updated results mysql-test/r/innodb-deadlock.result: Updated results mysql-test/r/innodb.result: Updated results mysql-test/r/innodb_cache.result: Updated results mysql-test/r/innodb_handler.result: Updated results mysql-test/r/insert.result: Updated results mysql-test/r/insert_select.result: Updated results mysql-test/r/isam.result: Updated results mysql-test/r/join_outer.result: Updated results mysql-test/r/key.result: Updated results mysql-test/r/merge.result: Updated results mysql-test/r/multi_update.result: Updated results mysql-test/r/myisam.result: Updated results mysql-test/r/null.result: Updated results mysql-test/r/null_key.result: Updated results mysql-test/r/odbc.result: Updated results mysql-test/r/olap.result: Updated results mysql-test/r/order_by.result: Updated results mysql-test/r/query_cache.result: Updated results mysql-test/r/rename.result: Updated results mysql-test/r/row.result: Updated results mysql-test/r/rpl000001.result: Updated results mysql-test/r/rpl000002.result: Updated results mysql-test/r/rpl000004.result: Updated results mysql-test/r/rpl000005.result: Updated results mysql-test/r/rpl000006.result: Updated results mysql-test/r/rpl000008.result: Updated results mysql-test/r/rpl000009.result: Updated results mysql-test/r/rpl000010.result: Updated results mysql-test/r/rpl000011.result: Updated results mysql-test/r/rpl000012.result: Updated results mysql-test/r/rpl000013.result: Updated results mysql-test/r/rpl_alter.result: Updated results mysql-test/r/rpl_empty_master_crash.result: Updated results mysql-test/r/rpl_redirect.result: Updated results mysql-test/r/rpl_replicate_do.result: Updated results mysql-test/r/rpl_rotate_logs.result: Updated results mysql-test/r/rpl_skip_error.result: Updated results mysql-test/r/rpl_temporary.result: Updated results mysql-test/r/select.result: Updated results mysql-test/r/subselect.result: Updated results mysql-test/r/temp_table.result: Updated results mysql-test/r/type_date.result: Updated results mysql-test/r/type_float.result: Updated results mysql-test/r/union.result: Updated results mysql-test/r/update.result: Updated results mysql-test/r/user_var.result: Updated results mysql-test/r/varbinary.result: Updated results mysql-test/r/variables.result: Updated results mysql-test/r/warnings.result: Updated results mysql-test/t/alias.test: Don't write warnings when initializing test mysql-test/t/alter_table.test: Don't write warnings when initializing test mysql-test/t/analyse.test: Don't write warnings when initializing test mysql-test/t/auto_increment.test: Don't write warnings when initializing test mysql-test/t/backup.test: Don't write warnings when initializing test mysql-test/t/bdb-alter-table-1.test: Don't write warnings when initializing test mysql-test/t/bdb-crash.test: Don't write warnings when initializing test mysql-test/t/bdb-deadlock.test: Don't write warnings when initializing test mysql-test/t/bdb.test: Don't write warnings when initializing test cleaned up test mysql-test/t/bdb_cache.test: Don't write warnings when initializing test mysql-test/t/bench_count_distinct.test: Don't write warnings when initializing test mysql-test/t/bigint.test: Don't write warnings when initializing test mysql-test/t/binary.test: Don't write warnings when initializing test mysql-test/t/bool.test: Don't write warnings when initializing test Changed to use standard table names mysql-test/t/bulk_replace.test: Don't write warnings when initializing test mysql-test/t/case.test: Don't write warnings when initializing test mysql-test/t/check.test: Don't write warnings when initializing test mysql-test/t/compare.test: Don't write warnings when initializing test mysql-test/t/connect.test: Removed empty line mysql-test/t/constraints.test: Don't write warnings when initializing test mysql-test/t/count_distinct.test: Don't write warnings when initializing test mysql-test/t/count_distinct2.test: Don't write warnings when initializing test mysql-test/t/create.test: Don't write warnings when initializing test Cleaned up test mysql-test/t/ctype_latin1_de.test: Don't write warnings when initializing test mysql-test/t/ctype_many.test: Don't write warnings when initializing test mysql-test/t/delayed.test: Don't write warnings when initializing test mysql-test/t/delete.test: Don't write warnings when initializing test Cleaned up test mysql-test/t/derived.test: Don't write warnings when initializing test Cleaned up test mysql-test/t/dirty_close.test: Don't write warnings when initializing test mysql-test/t/distinct.test: Don't write warnings when initializing test Cleaned up test mysql-test/t/drop.test: Don't write warnings when initializing test mysql-test/t/empty_table.test: Don't write warnings when initializing test mysql-test/t/err000001.test: Don't write warnings when initializing test Cleaned up test mysql-test/t/explain.test: Don't write warnings when initializing test mysql-test/t/flush.test: Don't write warnings when initializing test mysql-test/t/foreign_key.test: Don't write warnings when initializing test mysql-test/t/fulltext.test: Don't write warnings when initializing test mysql-test/t/fulltext_cache.test: Don't write warnings when initializing test mysql-test/t/fulltext_distinct.test: Don't write warnings when initializing test mysql-test/t/fulltext_left_join.test: Don't write warnings when initializing test mysql-test/t/fulltext_multi.test: Don't write warnings when initializing test mysql-test/t/fulltext_order_by.test: Don't write warnings when initializing test mysql-test/t/fulltext_update.test: Don't write warnings when initializing test mysql-test/t/func_concat.test: Don't write warnings when initializing test mysql-test/t/func_date_add.test: Don't write warnings when initializing test mysql-test/t/func_encrypt.test: Don't write warnings when initializing test mysql-test/t/func_equal.test: Don't write warnings when initializing test mysql-test/t/func_group.test: Don't write warnings when initializing test mysql-test/t/func_if.test: Don't write warnings when initializing test mysql-test/t/func_in.test: Don't write warnings when initializing test mysql-test/t/func_isnull.test: Don't write warnings when initializing test mysql-test/t/func_like.test: Don't write warnings when initializing test mysql-test/t/func_regexp.test: Don't write warnings when initializing test mysql-test/t/func_set.test: Don't write warnings when initializing test Merged test with other tests mysql-test/t/func_str.test: Don't write warnings when initializing test mysql-test/t/func_time.test: Don't write warnings when initializing test mysql-test/t/func_timestamp.test: Don't write warnings when initializing test mysql-test/t/gcc296.test: Don't write warnings when initializing test Cleaned up test mysql-test/t/grant_cache.test: Don't write warnings when initializing test mysql-test/t/group_by.test: Don't write warnings when initializing test Cleaned up test mysql-test/t/handler.test: Don't write warnings when initializing test mysql-test/t/having.test: Don't write warnings when initializing test mysql-test/t/heap.test: Don't write warnings when initializing test mysql-test/t/heap_auto_increment.test: Don't write warnings when initializing test mysql-test/t/heap_btree.test: Don't write warnings when initializing test mysql-test/t/heap_hash.test: Don't write warnings when initializing test mysql-test/t/innodb-deadlock.test: Don't write warnings when initializing test mysql-test/t/innodb.test: Don't write warnings when initializing test mysql-test/t/innodb_cache.test: Don't write warnings when initializing test mysql-test/t/innodb_handler.test: Don't write warnings when initializing test mysql-test/t/ins000001.test: Don't write warnings when initializing test mysql-test/t/insert.test: Don't write warnings when initializing test cleaned up test. Changed to use standard database and table names mysql-test/t/insert_select.test: Don't write warnings when initializing test Changed to use standard table names mysql-test/t/insert_update.test: Don't write warnings when initializing test mysql-test/t/isam.test: Don't write warnings when initializing test cleaned up test mysql-test/t/join.test: Don't write warnings when initializing test mysql-test/t/join_crash.test: Don't write warnings when initializing test mysql-test/t/join_outer.test: Don't write warnings when initializing test mysql-test/t/key.test: Don't write warnings when initializing test mysql-test/t/key_diff.test: Don't write warnings when initializing test mysql-test/t/key_primary.test: Don't write warnings when initializing test mysql-test/t/keywords.test: Don't write warnings when initializing test mysql-test/t/kill.test: Don't write warnings when initializing test mysql-test/t/limit.test: Don't write warnings when initializing test mysql-test/t/lock.test: Don't write warnings when initializing test mysql-test/t/lock_multi.test: Don't write warnings when initializing test mysql-test/t/lowercase_table.test: Don't write warnings when initializing test mysql-test/t/merge.test: Don't write warnings when initializing test cleaned up test mysql-test/t/multi_update.test: Don't write warnings when initializing test mysql-test/t/myisam.test: Don't write warnings when initializing test mysql-test/t/null.test: Don't write warnings when initializing test mysql-test/t/null_key.test: Don't write warnings when initializing test mysql-test/t/odbc.test: Don't write warnings when initializing test mysql-test/t/olap.test: Don't write warnings when initializing test mysql-test/t/order_by.test: Don't write warnings when initializing test mysql-test/t/order_fill_sortbuf.test: Don't write warnings when initializing test mysql-test/t/query_cache.test: Don't write warnings when initializing test mysql-test/t/raid.test: Don't write warnings when initializing test mysql-test/t/range.test: Don't write warnings when initializing test mysql-test/t/rename.test: Don't write warnings when initializing test mysql-test/t/repair.test: Don't write warnings when initializing test mysql-test/t/replace.test: Don't write warnings when initializing test mysql-test/t/rollback.test: Don't write warnings when initializing test mysql-test/t/row.test: Don't write warnings when initializing test mysql-test/t/rpl000001.test: Don't write warnings when initializing test mysql-test/t/rpl000002.test: Don't write warnings when initializing test mysql-test/t/rpl000004.test: Don't write warnings when initializing test mysql-test/t/rpl000005.test: Don't write warnings when initializing test mysql-test/t/rpl000006.test: Don't write warnings when initializing test mysql-test/t/rpl000008-slave.opt: Don't write warnings when initializing test mysql-test/t/rpl000008.test: Don't write warnings when initializing test mysql-test/t/rpl000009-slave.opt: Don't write warnings when initializing test mysql-test/t/rpl000009.test: Don't write warnings when initializing test mysql-test/t/rpl000010.test: Don't write warnings when initializing test mysql-test/t/rpl000011.test: Don't write warnings when initializing test mysql-test/t/rpl000012.test: Don't write warnings when initializing test mysql-test/t/rpl000013.test: Don't write warnings when initializing test mysql-test/t/rpl000015.test: Don't write warnings when initializing test mysql-test/t/rpl000017.test: Don't write warnings when initializing test mysql-test/t/rpl000018.test: Don't write warnings when initializing test mysql-test/t/rpl_alter.test: Don't write warnings when initializing test mysql-test/t/rpl_empty_master_crash.test: Don't write warnings when initializing test mysql-test/t/rpl_redirect.test: Don't write warnings when initializing test mysql-test/t/rpl_replicate_do.test: Don't write warnings when initializing test mysql-test/t/rpl_rotate_logs.test: Don't write warnings when initializing test mysql-test/t/rpl_skip_error.test: Don't write warnings when initializing test mysql-test/t/rpl_temporary.test: Don't write warnings when initializing test mysql-test/t/sel000033.test: Don't write warnings when initializing test mysql-test/t/sel000100.test: Don't write warnings when initializing test mysql-test/t/select.test: Don't write warnings when initializing test mysql-test/t/select_found.test: Don't write warnings when initializing test mysql-test/t/select_safe.test: Don't write warnings when initializing test mysql-test/t/show_check.test: Don't write warnings when initializing test mysql-test/t/status.test: Don't write warnings when initializing test mysql-test/t/subselect.test: Don't write warnings when initializing test mysql-test/t/symlink.test: Don't write warnings when initializing test mysql-test/t/tablelock.test: Don't write warnings when initializing test mysql-test/t/temp_table.test: Don't write warnings when initializing test mysql-test/t/truncate.test: Don't write warnings when initializing test mysql-test/t/type_blob.test: Don't write warnings when initializing test mysql-test/t/type_date.test: Don't write warnings when initializing test mysql-test/t/type_datetime.test: Don't write warnings when initializing test mysql-test/t/type_decimal.test: Don't write warnings when initializing test mysql-test/t/type_enum.test: Don't write warnings when initializing test mysql-test/t/type_float.test: Don't write warnings when initializing test mysql-test/t/type_ranges.test: Don't write warnings when initializing test mysql-test/t/type_set.test: Don't write warnings when initializing test mysql-test/t/type_time.test: Don't write warnings when initializing test mysql-test/t/type_timestamp.test: Don't write warnings when initializing test mysql-test/t/type_uint.test: Don't write warnings when initializing test mysql-test/t/type_year.test: Don't write warnings when initializing test mysql-test/t/union.test: Don't write warnings when initializing test mysql-test/t/update.test: Don't write warnings when initializing test mysql-test/t/user_var.test: Don't write warnings when initializing test mysql-test/t/varbinary.test: Don't write warnings when initializing test mysql-test/t/variables.test: Don't write warnings when initializing test mysql-test/t/warnings.test: Don't write warnings when initializing test mysys/my_vsnprintf.c: Safety fix readline/terminal.c: Removed compiler warnings sql/ha_berkeley.cc: Indentation changes sql/mysql_priv.h: Change store_warning -> push_warning_printf sql/sql_db.cc: Change store_warning -> push_warning_printf sql/sql_error.cc: Change store_warning -> push_warning_printf sql/sql_table.cc: Change store_warning -> push_warning_printf
2003-01-06 00:48:59 +01:00
--disable_warnings
drop table if exists t1,t2;
drop DATABASE if exists mysqltest_1;
Changed mysql-test to print warnings for not existing table to DROP TABLE Cleaned up test; Removed wrong DROP TABLE commands and use standard table and database names. changed store_warning() -> push_warning_print() BitKeeper/deleted/.del-rpl000016-slave.opt~ef76f85ddcc13b87: Delete: mysql-test/t/rpl000016-slave.opt BitKeeper/deleted/.del-sel000001.test~9567c1646058cc: Delete: mysql-test/t/sel000001.test BitKeeper/deleted/.del-sel000002.test~9f500639572e18e1: Delete: mysql-test/t/sel000002.test BitKeeper/deleted/.del-sel000003.test~63a5512d18cd20a2: Delete: mysql-test/t/sel000003.test BitKeeper/deleted/.del-sel000001.result~383913ae4505ec86: Delete: mysql-test/r/sel000001.result BitKeeper/deleted/.del-sel000002.result~d1787e6fd5dbc1cc: Delete: mysql-test/r/sel000002.result BitKeeper/deleted/.del-sel000003.result~d7b657b1e3a286a7: Delete: mysql-test/r/sel000003.result BitKeeper/deleted/.del-sel000031.result~d49aeac63ad7db4d: Delete: mysql-test/r/sel000031.result BitKeeper/deleted/.del-sel000031.test~50a19a8e204e99bc: Delete: mysql-test/t/sel000031.test BitKeeper/deleted/.del-sel000032.result~6cb30e23cbca9fb0: Delete: mysql-test/r/sel000032.result BitKeeper/deleted/.del-sel000032.test~e32da7c3fc4b7ace: Delete: mysql-test/t/sel000032.test BitKeeper/deleted/.del-rpl000003.result~68d6ee00beaa011: Delete: mysql-test/r/rpl000003.result BitKeeper/deleted/.del-rpl000003.test~b7cfc4c5576fbafd: Delete: mysql-test/t/rpl000003.test client/mysql.cc: Don't yet print information about SQL help client/mysqltest.c: Added test options: --enable_warnings --disable_warnings --enable_info --disable_info configure.in: changed version number of shared libraries mysql-test/include/master-slave.inc: Don't write warnings on init mysql-test/r/backup.result: Updated results mysql-test/r/bdb.result: Updated results mysql-test/r/bigint.result: Updated results mysql-test/r/bool.result: Updated results mysql-test/r/create.result: Updated results mysql-test/r/delete.result: Updated results mysql-test/r/derived.result: Updated results mysql-test/r/distinct.result: Updated results mysql-test/r/drop.result: Updated results mysql-test/r/flush.result: Updated results mysql-test/r/fulltext.result: Updated results mysql-test/r/fulltext_multi.result: Updated results mysql-test/r/fulltext_order_by.result: Updated results mysql-test/r/func_equal.result: Updated results mysql-test/r/func_in.result: Updated results mysql-test/r/func_set.result: Updated results mysql-test/r/gcc296.result: Updated results mysql-test/r/group_by.result: Updated results mysql-test/r/innodb-deadlock.result: Updated results mysql-test/r/innodb.result: Updated results mysql-test/r/innodb_cache.result: Updated results mysql-test/r/innodb_handler.result: Updated results mysql-test/r/insert.result: Updated results mysql-test/r/insert_select.result: Updated results mysql-test/r/isam.result: Updated results mysql-test/r/join_outer.result: Updated results mysql-test/r/key.result: Updated results mysql-test/r/merge.result: Updated results mysql-test/r/multi_update.result: Updated results mysql-test/r/myisam.result: Updated results mysql-test/r/null.result: Updated results mysql-test/r/null_key.result: Updated results mysql-test/r/odbc.result: Updated results mysql-test/r/olap.result: Updated results mysql-test/r/order_by.result: Updated results mysql-test/r/query_cache.result: Updated results mysql-test/r/rename.result: Updated results mysql-test/r/row.result: Updated results mysql-test/r/rpl000001.result: Updated results mysql-test/r/rpl000002.result: Updated results mysql-test/r/rpl000004.result: Updated results mysql-test/r/rpl000005.result: Updated results mysql-test/r/rpl000006.result: Updated results mysql-test/r/rpl000008.result: Updated results mysql-test/r/rpl000009.result: Updated results mysql-test/r/rpl000010.result: Updated results mysql-test/r/rpl000011.result: Updated results mysql-test/r/rpl000012.result: Updated results mysql-test/r/rpl000013.result: Updated results mysql-test/r/rpl_alter.result: Updated results mysql-test/r/rpl_empty_master_crash.result: Updated results mysql-test/r/rpl_redirect.result: Updated results mysql-test/r/rpl_replicate_do.result: Updated results mysql-test/r/rpl_rotate_logs.result: Updated results mysql-test/r/rpl_skip_error.result: Updated results mysql-test/r/rpl_temporary.result: Updated results mysql-test/r/select.result: Updated results mysql-test/r/subselect.result: Updated results mysql-test/r/temp_table.result: Updated results mysql-test/r/type_date.result: Updated results mysql-test/r/type_float.result: Updated results mysql-test/r/union.result: Updated results mysql-test/r/update.result: Updated results mysql-test/r/user_var.result: Updated results mysql-test/r/varbinary.result: Updated results mysql-test/r/variables.result: Updated results mysql-test/r/warnings.result: Updated results mysql-test/t/alias.test: Don't write warnings when initializing test mysql-test/t/alter_table.test: Don't write warnings when initializing test mysql-test/t/analyse.test: Don't write warnings when initializing test mysql-test/t/auto_increment.test: Don't write warnings when initializing test mysql-test/t/backup.test: Don't write warnings when initializing test mysql-test/t/bdb-alter-table-1.test: Don't write warnings when initializing test mysql-test/t/bdb-crash.test: Don't write warnings when initializing test mysql-test/t/bdb-deadlock.test: Don't write warnings when initializing test mysql-test/t/bdb.test: Don't write warnings when initializing test cleaned up test mysql-test/t/bdb_cache.test: Don't write warnings when initializing test mysql-test/t/bench_count_distinct.test: Don't write warnings when initializing test mysql-test/t/bigint.test: Don't write warnings when initializing test mysql-test/t/binary.test: Don't write warnings when initializing test mysql-test/t/bool.test: Don't write warnings when initializing test Changed to use standard table names mysql-test/t/bulk_replace.test: Don't write warnings when initializing test mysql-test/t/case.test: Don't write warnings when initializing test mysql-test/t/check.test: Don't write warnings when initializing test mysql-test/t/compare.test: Don't write warnings when initializing test mysql-test/t/connect.test: Removed empty line mysql-test/t/constraints.test: Don't write warnings when initializing test mysql-test/t/count_distinct.test: Don't write warnings when initializing test mysql-test/t/count_distinct2.test: Don't write warnings when initializing test mysql-test/t/create.test: Don't write warnings when initializing test Cleaned up test mysql-test/t/ctype_latin1_de.test: Don't write warnings when initializing test mysql-test/t/ctype_many.test: Don't write warnings when initializing test mysql-test/t/delayed.test: Don't write warnings when initializing test mysql-test/t/delete.test: Don't write warnings when initializing test Cleaned up test mysql-test/t/derived.test: Don't write warnings when initializing test Cleaned up test mysql-test/t/dirty_close.test: Don't write warnings when initializing test mysql-test/t/distinct.test: Don't write warnings when initializing test Cleaned up test mysql-test/t/drop.test: Don't write warnings when initializing test mysql-test/t/empty_table.test: Don't write warnings when initializing test mysql-test/t/err000001.test: Don't write warnings when initializing test Cleaned up test mysql-test/t/explain.test: Don't write warnings when initializing test mysql-test/t/flush.test: Don't write warnings when initializing test mysql-test/t/foreign_key.test: Don't write warnings when initializing test mysql-test/t/fulltext.test: Don't write warnings when initializing test mysql-test/t/fulltext_cache.test: Don't write warnings when initializing test mysql-test/t/fulltext_distinct.test: Don't write warnings when initializing test mysql-test/t/fulltext_left_join.test: Don't write warnings when initializing test mysql-test/t/fulltext_multi.test: Don't write warnings when initializing test mysql-test/t/fulltext_order_by.test: Don't write warnings when initializing test mysql-test/t/fulltext_update.test: Don't write warnings when initializing test mysql-test/t/func_concat.test: Don't write warnings when initializing test mysql-test/t/func_date_add.test: Don't write warnings when initializing test mysql-test/t/func_encrypt.test: Don't write warnings when initializing test mysql-test/t/func_equal.test: Don't write warnings when initializing test mysql-test/t/func_group.test: Don't write warnings when initializing test mysql-test/t/func_if.test: Don't write warnings when initializing test mysql-test/t/func_in.test: Don't write warnings when initializing test mysql-test/t/func_isnull.test: Don't write warnings when initializing test mysql-test/t/func_like.test: Don't write warnings when initializing test mysql-test/t/func_regexp.test: Don't write warnings when initializing test mysql-test/t/func_set.test: Don't write warnings when initializing test Merged test with other tests mysql-test/t/func_str.test: Don't write warnings when initializing test mysql-test/t/func_time.test: Don't write warnings when initializing test mysql-test/t/func_timestamp.test: Don't write warnings when initializing test mysql-test/t/gcc296.test: Don't write warnings when initializing test Cleaned up test mysql-test/t/grant_cache.test: Don't write warnings when initializing test mysql-test/t/group_by.test: Don't write warnings when initializing test Cleaned up test mysql-test/t/handler.test: Don't write warnings when initializing test mysql-test/t/having.test: Don't write warnings when initializing test mysql-test/t/heap.test: Don't write warnings when initializing test mysql-test/t/heap_auto_increment.test: Don't write warnings when initializing test mysql-test/t/heap_btree.test: Don't write warnings when initializing test mysql-test/t/heap_hash.test: Don't write warnings when initializing test mysql-test/t/innodb-deadlock.test: Don't write warnings when initializing test mysql-test/t/innodb.test: Don't write warnings when initializing test mysql-test/t/innodb_cache.test: Don't write warnings when initializing test mysql-test/t/innodb_handler.test: Don't write warnings when initializing test mysql-test/t/ins000001.test: Don't write warnings when initializing test mysql-test/t/insert.test: Don't write warnings when initializing test cleaned up test. Changed to use standard database and table names mysql-test/t/insert_select.test: Don't write warnings when initializing test Changed to use standard table names mysql-test/t/insert_update.test: Don't write warnings when initializing test mysql-test/t/isam.test: Don't write warnings when initializing test cleaned up test mysql-test/t/join.test: Don't write warnings when initializing test mysql-test/t/join_crash.test: Don't write warnings when initializing test mysql-test/t/join_outer.test: Don't write warnings when initializing test mysql-test/t/key.test: Don't write warnings when initializing test mysql-test/t/key_diff.test: Don't write warnings when initializing test mysql-test/t/key_primary.test: Don't write warnings when initializing test mysql-test/t/keywords.test: Don't write warnings when initializing test mysql-test/t/kill.test: Don't write warnings when initializing test mysql-test/t/limit.test: Don't write warnings when initializing test mysql-test/t/lock.test: Don't write warnings when initializing test mysql-test/t/lock_multi.test: Don't write warnings when initializing test mysql-test/t/lowercase_table.test: Don't write warnings when initializing test mysql-test/t/merge.test: Don't write warnings when initializing test cleaned up test mysql-test/t/multi_update.test: Don't write warnings when initializing test mysql-test/t/myisam.test: Don't write warnings when initializing test mysql-test/t/null.test: Don't write warnings when initializing test mysql-test/t/null_key.test: Don't write warnings when initializing test mysql-test/t/odbc.test: Don't write warnings when initializing test mysql-test/t/olap.test: Don't write warnings when initializing test mysql-test/t/order_by.test: Don't write warnings when initializing test mysql-test/t/order_fill_sortbuf.test: Don't write warnings when initializing test mysql-test/t/query_cache.test: Don't write warnings when initializing test mysql-test/t/raid.test: Don't write warnings when initializing test mysql-test/t/range.test: Don't write warnings when initializing test mysql-test/t/rename.test: Don't write warnings when initializing test mysql-test/t/repair.test: Don't write warnings when initializing test mysql-test/t/replace.test: Don't write warnings when initializing test mysql-test/t/rollback.test: Don't write warnings when initializing test mysql-test/t/row.test: Don't write warnings when initializing test mysql-test/t/rpl000001.test: Don't write warnings when initializing test mysql-test/t/rpl000002.test: Don't write warnings when initializing test mysql-test/t/rpl000004.test: Don't write warnings when initializing test mysql-test/t/rpl000005.test: Don't write warnings when initializing test mysql-test/t/rpl000006.test: Don't write warnings when initializing test mysql-test/t/rpl000008-slave.opt: Don't write warnings when initializing test mysql-test/t/rpl000008.test: Don't write warnings when initializing test mysql-test/t/rpl000009-slave.opt: Don't write warnings when initializing test mysql-test/t/rpl000009.test: Don't write warnings when initializing test mysql-test/t/rpl000010.test: Don't write warnings when initializing test mysql-test/t/rpl000011.test: Don't write warnings when initializing test mysql-test/t/rpl000012.test: Don't write warnings when initializing test mysql-test/t/rpl000013.test: Don't write warnings when initializing test mysql-test/t/rpl000015.test: Don't write warnings when initializing test mysql-test/t/rpl000017.test: Don't write warnings when initializing test mysql-test/t/rpl000018.test: Don't write warnings when initializing test mysql-test/t/rpl_alter.test: Don't write warnings when initializing test mysql-test/t/rpl_empty_master_crash.test: Don't write warnings when initializing test mysql-test/t/rpl_redirect.test: Don't write warnings when initializing test mysql-test/t/rpl_replicate_do.test: Don't write warnings when initializing test mysql-test/t/rpl_rotate_logs.test: Don't write warnings when initializing test mysql-test/t/rpl_skip_error.test: Don't write warnings when initializing test mysql-test/t/rpl_temporary.test: Don't write warnings when initializing test mysql-test/t/sel000033.test: Don't write warnings when initializing test mysql-test/t/sel000100.test: Don't write warnings when initializing test mysql-test/t/select.test: Don't write warnings when initializing test mysql-test/t/select_found.test: Don't write warnings when initializing test mysql-test/t/select_safe.test: Don't write warnings when initializing test mysql-test/t/show_check.test: Don't write warnings when initializing test mysql-test/t/status.test: Don't write warnings when initializing test mysql-test/t/subselect.test: Don't write warnings when initializing test mysql-test/t/symlink.test: Don't write warnings when initializing test mysql-test/t/tablelock.test: Don't write warnings when initializing test mysql-test/t/temp_table.test: Don't write warnings when initializing test mysql-test/t/truncate.test: Don't write warnings when initializing test mysql-test/t/type_blob.test: Don't write warnings when initializing test mysql-test/t/type_date.test: Don't write warnings when initializing test mysql-test/t/type_datetime.test: Don't write warnings when initializing test mysql-test/t/type_decimal.test: Don't write warnings when initializing test mysql-test/t/type_enum.test: Don't write warnings when initializing test mysql-test/t/type_float.test: Don't write warnings when initializing test mysql-test/t/type_ranges.test: Don't write warnings when initializing test mysql-test/t/type_set.test: Don't write warnings when initializing test mysql-test/t/type_time.test: Don't write warnings when initializing test mysql-test/t/type_timestamp.test: Don't write warnings when initializing test mysql-test/t/type_uint.test: Don't write warnings when initializing test mysql-test/t/type_year.test: Don't write warnings when initializing test mysql-test/t/union.test: Don't write warnings when initializing test mysql-test/t/update.test: Don't write warnings when initializing test mysql-test/t/user_var.test: Don't write warnings when initializing test mysql-test/t/varbinary.test: Don't write warnings when initializing test mysql-test/t/variables.test: Don't write warnings when initializing test mysql-test/t/warnings.test: Don't write warnings when initializing test mysys/my_vsnprintf.c: Safety fix readline/terminal.c: Removed compiler warnings sql/ha_berkeley.cc: Indentation changes sql/mysql_priv.h: Change store_warning -> push_warning_printf sql/sql_db.cc: Change store_warning -> push_warning_printf sql/sql_error.cc: Change store_warning -> push_warning_printf sql/sql_table.cc: Change store_warning -> push_warning_printf
2003-01-06 00:48:59 +01:00
--enable_warnings
# Test to see if select will get the lock ahead of low priority update
Updated manual about embedded version. Speed up column-completion in 'mysql' Don't use ISAM if HAVE_ISAM is not defined A lot of fixes for the embedded version. All libraries are now included in libmysqld.a Changed arguments to convert_dirname() to make it more general. Renamed files in the 'merge' directory to all use a common prefix. Don't compile both assembler and C functions on x86 BitKeeper/deleted/.del-mf_pack2.c~f07795bbcf57be7: Delete: mysys/mf_pack2.c Docs/manual.texi: Updated chapter about embedded version acinclude.m4: Fix for using BDB and InnoDB with embedded client/completion_hash.cc: Speed up memory allocation client/completion_hash.h: Speed up memory allocation client/mysql.cc: Speed up memory allocation client/mysqldump.c: Fix to use now convert_dirname client/mysqltest.c: Fixed memory allocation bugs. Added --basedir=#, --compress=#, --server-arg, --server-file. Fixes for embedded version Changed silent mode to -s instead of -q include/my_global.h: Update to use HAVE_ISAM include/my_sys.h: Cleanup of fn_format() include/mysql.h: Prepare FIELD struct for 4.1 include/mysql_embed.h: Don't use ISAM in embedded version innobase/include/srv0srv.h: Make InnoDB startup/shutdown silent in embedded version innobase/log/log0log.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0srv.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0start.c: Make InnoDB startup/shutdown silent in embedded version isam/isamlog.c: new convert_dirname libmysql/libmysql.c: Prepare for 4.1 libmysqld/Makefile.am: Changed to create one libmysqld.a file that includes all other libraries. libmysqld/examples/Makefile.am: Changed to use new libmysqld.a libmysqld/examples/test-run: Lot's of fixes to get new mysqltest with embedded MySQL to work. libmysqld/lib_sql.cc: Changed type of arguments for mysql_server_init() to make code more portable. libmysqld/libmysqld.c: Cleanup merge/Makefile.am: Rename to use common prefix. merge/mrg_close.c: Rename to use common prefix. merge/mrg_create.c: Rename to use common prefix. merge/mrg_def.h: Rename to use common prefix. merge/mrg_delete.c: Rename to use common prefix. merge/mrg_extra.c: Rename to use common prefix. merge/mrg_info.c: Rename to use common prefix. merge/mrg_locking.c: Rename to use common prefix. merge/mrg_open.c: Rename to use common prefix. merge/mrg_panic.c: Rename to use common prefix. merge/mrg_rrnd.c: Rename to use common prefix. merge/mrg_rsame.c: Rename to use common prefix. merge/mrg_static.c: Rename to use common prefix. merge/mrg_update.c: Rename to use common prefix. myisam/myisamlog.c: Use new convert_dirname myisammrg/Makefile.am: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_close.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_create.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_def.h: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_delete.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_extra.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_info.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_locking.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_open.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_panic.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_queue.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rfirst.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rkey.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rlast.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rnext.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rprev.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rrnd.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rsame.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_static.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_update.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_write.c: Renamed mymrgdef.h -> myrg_def.h mysql-test/include/master-slave.inc: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/mysql-test-run.sh: cleanup mysql-test/r/auto_increment.result: Cleanup of tests for embedded version mysql-test/r/func_system.result: Cleanup of tests for embedded version mysql-test/r/isam.result: Cleanup of tests for embedded version mysql-test/r/lock.result: Cleanup of tests for embedded version mysql-test/r/show_check.result: Cleanup of tests for embedded version mysql-test/t/auto_increment.test: Cleanup of tests for embedded version mysql-test/t/backup.test: Cleanup of tests for embedded version mysql-test/t/count_distinct2-master.opt: Cleanup of tests for embedded version mysql-test/t/count_distinct2.test: Cleanup of tests for embedded version mysql-test/t/create.test: Cleanup of tests for embedded version mysql-test/t/flush.test: Cleanup of tests for embedded version mysql-test/t/func_system.test: Cleanup of tests for embedded version mysql-test/t/isam.test: Cleanup of tests for embedded version mysql-test/t/kill.test: Cleanup of tests for embedded version mysql-test/t/lock.test: Cleanup of tests for embedded version mysql-test/t/order_fill_sortbuf-master.opt: Cleanup of tests for embedded version mysql-test/t/rpl000015.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000016.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000017.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000018.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/show_check.test: Move ISAM specific tests to isam.test mysql-test/t/status.test: Cleanup of tests for embedded version mysql-test/t/tablelock.test: Cleanup mysys/Makefile.am: Removed not used mf_pack2.c mysys/charset.c: new convert_dirname mysys/default.c: new convert_dirname mysys/mf_dirname.c: Changed convert_dirname() to be more general mysys/mf_format.c: Changed bit flags to fn_format() to defines. Added handling of relative filenames BitKeeper/etc/ignore: Added libmysqld/examples/test-gdbinit scripts/mysql_explain_log to the ignore list mysys/mf_pack.c: new convert_dirname mysys/mf_tempfile.c: new convert_dirname scripts/Makefile.am: Adde mysql_explain_log scripts/mysql_config.sh: Added support of --libmysqld-libs sql/Makefile.am: Fix to use 'innodb_system_libs' sql/ha_innobase.cc: Make InnoDB startup/shutdown silent in embedded version sql/ha_isam.cc: Added handling of HAVE_ISAM sql/ha_isammrg.cc: Added handling of HAVE_ISAM sql/ha_myisam.cc: Handle relative paths; Needed to support BACKUP TABLE in embedded version sql/ha_myisammrg.cc: Rename of filenames sql/handler.cc: Added handling of HAVE_ISAM sql/item_func.cc: Fix for ecc (Intel Compiler) sql/mysql_priv.h: Added global variable 'mysql_embedded' sql/mysqld.cc: Use HAVE_ISAM sql/sql_parse.cc: Use new convert_dirname sql/sql_select.cc: Fix for ecc (Intel Compiler) sql/sql_table.cc: Added handling of relative filenames strings/Makefile.am: Don't compile both assembler and C functions on x86 strings/bchange.c: cleanup
2001-10-08 03:58:07 +02:00
connect (locker,localhost,root,,);
Implement new type-of-operation-aware metadata locks. Add a wait-for graph based deadlock detector to the MDL subsystem. Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and bug #37346 "innodb does not detect deadlock between update and alter table". The first bug manifested itself as an unwarranted abort of a transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER statement, when this transaction tried to repeat use of a table, which it has already used in a similar fashion before ALTER started. The second bug showed up as a deadlock between table-level locks and InnoDB row locks, which was "detected" only after innodb_lock_wait_timeout timeout. A transaction would start using the table and modify a few rows. Then ALTER TABLE would come in, and start copying rows into a temporary table. Eventually it would stumble on the modified records and get blocked on a row lock. The first transaction would try to do more updates, and get blocked on thr_lock.c lock. This situation of circular wait would only get resolved by a timeout. Both these bugs stemmed from inadequate solutions to the problem of deadlocks occurring between different locking subsystems. In the first case we tried to avoid deadlocks between metadata locking and table-level locking subsystems, when upgrading shared metadata lock to exclusive one. Transactions holding the shared lock on the table and waiting for some table-level lock used to be aborted too aggressively. We also allowed ALTER TABLE to start in presence of transactions that modify the subject table. ALTER TABLE acquires TL_WRITE_ALLOW_READ lock at start, and that block all writes against the table (naturally, we don't want any writes to be lost when switching the old and the new table). TL_WRITE_ALLOW_READ lock, in turn, would block the started transaction on thr_lock.c lock, should they do more updates. This, again, lead to the need to abort such transactions. The second bug occurred simply because we didn't have any mechanism to detect deadlocks between the table-level locks in thr_lock.c and row-level locks in InnoDB, other than innodb_lock_wait_timeout. This patch solves both these problems by moving lock conflicts which are causing these deadlocks into the metadata locking subsystem, thus making it possible to avoid or detect such deadlocks inside MDL. To do this we introduce new type-of-operation-aware metadata locks, which allow MDL subsystem to know not only the fact that transaction has used or is going to use some object but also what kind of operation it has carried out or going to carry out on the object. This, along with the addition of a special kind of upgradable metadata lock, allows ALTER TABLE to wait until all transactions which has updated the table to go away. This solves the second issue. Another special type of upgradable metadata lock is acquired by LOCK TABLE WRITE. This second lock type allows to solve the first issue, since abortion of table-level locks in event of DDL under LOCK TABLES becomes also unnecessary. Below follows the list of incompatible changes introduced by this patch: - From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those statements that acquire TL_WRITE_ALLOW_READ lock) wait for all transactions which has *updated* the table to complete. - From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE (i.e. all statements which acquire TL_WRITE table-level lock) wait for all transaction which *updated or read* from the table to complete. As a consequence, innodb_table_locks=0 option no longer applies to LOCK TABLES ... WRITE. - DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort statements or transactions which use tables being dropped or renamed, and instead wait for these transactions to complete. - Since LOCK TABLES WRITE now takes a special metadata lock, not compatible with with reads or writes against the subject table and transaction-wide, thr_lock.c deadlock avoidance algorithm that used to ensure absence of deadlocks between LOCK TABLES WRITE and other statements is no longer sufficient, even for MyISAM. The wait-for graph based deadlock detector of MDL subsystem may sometimes be necessary and is involved. This may lead to ER_LOCK_DEADLOCK error produced for multi-statement transactions even if these only use MyISAM: session 1: session 2: begin; update t1 ... lock table t2 write, t1 write; -- gets a lock on t2, blocks on t1 update t2 ... (ER_LOCK_DEADLOCK) - Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE was abandoned. LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same priority as the usual LOCK TABLE ... WRITE. SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in the wait queue. - We do not take upgradable metadata locks on implicitly locked tables. So if one has, say, a view v1 that uses table t1, and issues: LOCK TABLE v1 WRITE; FLUSH TABLE t1; -- (or just 'FLUSH TABLES'), an error is produced. In order to be able to perform DDL on a table under LOCK TABLES, the table must be locked explicitly in the LOCK TABLES list. mysql-test/include/handler.inc: Adjusted test case to trigger an execution path on which bug 41110 "crash with handler command when used concurrently with alter table" and bug 41112 "crash in mysql_ha_close_table/get_lock_data with alter table" were originally discovered. Left old test case which no longer triggers this execution path for the sake of coverage. Added test coverage for HANDLER SQL statements and type-aware metadata locks. Added a test for the global shared lock and HANDLER SQL. Updated tests to take into account that the old simple deadlock detection heuristics was replaced with a graph-based deadlock detector. mysql-test/r/debug_sync.result: Updated results (see debug_sync.test). mysql-test/r/handler_innodb.result: Updated results (see handler.inc test). mysql-test/r/handler_myisam.result: Updated results (see handler.inc test). mysql-test/r/innodb-lock.result: Updated results (see innodb-lock.test). mysql-test/r/innodb_mysql_lock.result: Updated results (see innodb_mysql_lock.test). mysql-test/r/lock.result: Updated results (see lock.test). mysql-test/r/lock_multi.result: Updated results (see lock_multi.test). mysql-test/r/lock_sync.result: Updated results (see lock_sync.test). mysql-test/r/mdl_sync.result: Updated results (see mdl_sync.test). mysql-test/r/sp-threads.result: SHOW PROCESSLIST output has changed due to the fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/r/truncate_coverage.result: Updated results (see truncate_coverage.test). mysql-test/suite/funcs_1/datadict/processlist_val.inc: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/funcs_1/r/processlist_val_no_prot.result: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/rpl/t/rpl_sp.test: Updated to a new SHOW PROCESSLIST state name. mysql-test/t/debug_sync.test: Use LOCK TABLES READ instead of LOCK TABLES WRITE as the latter no longer allows to trigger execution path involving waiting on thr_lock.c lock and therefore reaching debug sync-point covered by this test. mysql-test/t/innodb-lock.test: Adjusted test case to the fact that innodb_table_locks=0 option is no longer supported, since LOCK TABLES WRITE handles all its conflicts within MDL subsystem. mysql-test/t/innodb_mysql_lock.test: Added test for bug #37346 "innodb does not detect deadlock between update and alter table". mysql-test/t/lock.test: Added test coverage which checks the fact that we no longer support DDL under LOCK TABLES on tables which were locked implicitly. Adjusted existing test cases accordingly. mysql-test/t/lock_multi.test: Added test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock". Adjusted other test cases to take into account the fact that waiting for LOCK TABLES ... WRITE now happens within MDL subsystem. mysql-test/t/lock_sync.test: Since LOCK TABLES ... WRITE now takes SNRW metadata lock for tables locked explicitly we have to implicitly lock InnoDB tables (through view) to trigger the table-level lock conflict between TL_WRITE and TL_WRITE_ALLOW_WRITE. mysql-test/t/mdl_sync.test: Added basic test coverage for type-of-operation-aware metadata locks. Also covered with tests some use cases involving HANDLER statements in which a deadlock could arise. Adjusted existing tests to take type-of-operation-aware MDL into account. mysql-test/t/multi_update.test: Update to a new SHOW PROCESSLIST state name. mysql-test/t/truncate_coverage.test: Adjusted test case after making LOCK TABLES WRITE to wait until transactions that use the table to be locked are completed. Updated to the changed name of DEBUG_SYNC point. sql/handler.cc: Global read lock functionality has been moved into a class. sql/lock.cc: Global read lock functionality has been moved into a class. Updated code to use the new MDL API. sql/mdl.cc: Introduced new type-of-operation aware metadata locks. To do this: - Changed MDL_lock to use one list for waiting requests and one list for granted requests. For each list, added a bitmap that holds information what lock types a list contains. Added a helper class MDL_lock::List to manipulate with granted and waited lists while keeping the bitmaps in sync with list contents. - Changed lock-compatibility functions to use bitmaps that define compatibility. - Introduced a graph based deadlock detector inspired by waiting_threads.c from Maria implementation. - Now that we have a deadlock detector, and no longer have a global lock to protect individual lock objects, but rather use an rw lock per object, removed redundant code for upgrade, and the global read lock. Changed the MDL API to no longer require the caller to acquire the global intention exclusive lock by means of a separate method. Removed a few more methods that became redundant. - Removed deadlock detection heuristic, it has been made obsolete by the deadlock detector. - With operation-type-aware metadata locks, MDL subsystem has become aware of potential conflicts between DDL and open transactions. This made it possible to remove calls to mysql_abort_transactions_with_shared_lock() from acquisition paths for exclusive lock and lock upgrade. Now we can simply wait for these transactions to complete without fear of deadlock. Function mysql_lock_abort() has also become unnecessary for all conflicting cases except when a DDL conflicts with a connection that has an open HANDLER. sql/mdl.h: Introduced new type-of-operation aware metadata locks. Introduced a graph based deadlock detector and supporting methods. Added comments. God rid of redundant API calls. Renamed m_lt_or_ha_sentinel to m_trans_sentinel, since now it guards the global read lock as well as LOCK TABLES and HANDLER locks. sql/mysql_priv.h: Moved the global read lock functionality into a class. Added MYSQL_OPEN_FORCE_SHARED_MDL flag which forces open_tables() to take MDL_SHARED on tables instead of metadata locks specified in the parser. We use this to allow PREPARE run concurrently in presence of LOCK TABLES ... WRITE. Added signature for find_table_for_mdl_ugprade(). sql/set_var.cc: Global read lock functionality has been moved into a class. sql/sp_head.cc: When creating TABLE_LIST elements for prelocking or system tables set the type of request for metadata lock according to the operation that will be performed on the table. sql/sql_base.cc: - Updated code to use the new MDL API. - In order to avoid locks starvation we take upgradable locks all at once. As result implicitly locked tables no longer get an upgradable lock. Consequently DDL and FLUSH TABLES for such tables is prohibited. find_write_locked_table() was replaced by find_table_for_mdl_upgrade() function. open_table() was adjusted to return TABLE instance with upgradable ticket when necessary. - We no longer wait for all locks on OT_WAIT back off action -- only on the lock that caused the wait conflict. Moreover, now we distinguish cases when we have to wait due to conflict in MDL and old version of table in TDC. - Upate mysql_notify_threads_having_share_locks() to only abort thr_lock.c waits of threads that have open HANDLERs, since lock conflicts with only these threads now can lead to deadlocks not detectable by the MDL deadlock detector. - Remove mysql_abort_transactions_with_shared_locks() which is no longer needed. sql/sql_class.cc: Global read lock functionality has been moved into a class. Re-arranged code in THD::cleanup() to simplify assert. sql/sql_class.h: Introduced class to incapsulate global read lock functionality. Now sentinel in MDL subsystem guards the global read lock as well as LOCK TABLES and HANDLER locks. Adjusted code accordingly. sql/sql_db.cc: Global read lock functionality has been moved into a class. sql/sql_delete.cc: We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result TRUNCATE TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_handler.cc: Inform MDL_context about presence of open HANDLERs. Since HANLDERs break MDL protocol by acquiring table-level lock while holding only S metadata lock on a table MDL subsystem should take special care about such contexts (Now this is the only case when mysql_lock_abort() is used). sql/sql_parse.cc: Global read lock functionality has been moved into a class. Do not take upgradable metadata locks when opening tables for CREATE TABLE SELECT as it is not necessary and limits concurrency. When initializing TABLE_LIST objects before adding them to the table list set the type of request for metadata lock according to the operation that will be performed on the table. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result FLUSH TABLES is no longer allowed for such tables. sql/sql_prepare.cc: Use MYSQL_OPEN_FORCE_SHARED_MDL flag when opening tables during PREPARE. This allows PREPARE to run concurrently in presence of LOCK TABLES ... WRITE. sql/sql_rename.cc: Global read lock functionality has been moved into a class. sql/sql_show.cc: Updated code to use the new MDL API. sql/sql_table.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result DROP TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_trigger.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result CREATE/DROP TRIGGER is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_view.cc: Global read lock functionality has been moved into a class. Fixed results of wrong merge that led to misuse of GLR API. CREATE VIEW statement is not a commit statement. sql/table.cc: When resetting TABLE_LIST objects for PS or SP re-execution set the type of request for metadata lock according to the operation that will be performed on the table. Do the same in auxiliary function initializing metadata lock requests in a table list. sql/table.h: When initializing TABLE_LIST objects set the type of request for metadata lock according to the operation that will be performed on the table. sql/transaction.cc: Global read lock functionality has been moved into a class.
2010-02-01 12:43:06 +01:00
connect (locker2,localhost,root,,);
Updated manual about embedded version. Speed up column-completion in 'mysql' Don't use ISAM if HAVE_ISAM is not defined A lot of fixes for the embedded version. All libraries are now included in libmysqld.a Changed arguments to convert_dirname() to make it more general. Renamed files in the 'merge' directory to all use a common prefix. Don't compile both assembler and C functions on x86 BitKeeper/deleted/.del-mf_pack2.c~f07795bbcf57be7: Delete: mysys/mf_pack2.c Docs/manual.texi: Updated chapter about embedded version acinclude.m4: Fix for using BDB and InnoDB with embedded client/completion_hash.cc: Speed up memory allocation client/completion_hash.h: Speed up memory allocation client/mysql.cc: Speed up memory allocation client/mysqldump.c: Fix to use now convert_dirname client/mysqltest.c: Fixed memory allocation bugs. Added --basedir=#, --compress=#, --server-arg, --server-file. Fixes for embedded version Changed silent mode to -s instead of -q include/my_global.h: Update to use HAVE_ISAM include/my_sys.h: Cleanup of fn_format() include/mysql.h: Prepare FIELD struct for 4.1 include/mysql_embed.h: Don't use ISAM in embedded version innobase/include/srv0srv.h: Make InnoDB startup/shutdown silent in embedded version innobase/log/log0log.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0srv.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0start.c: Make InnoDB startup/shutdown silent in embedded version isam/isamlog.c: new convert_dirname libmysql/libmysql.c: Prepare for 4.1 libmysqld/Makefile.am: Changed to create one libmysqld.a file that includes all other libraries. libmysqld/examples/Makefile.am: Changed to use new libmysqld.a libmysqld/examples/test-run: Lot's of fixes to get new mysqltest with embedded MySQL to work. libmysqld/lib_sql.cc: Changed type of arguments for mysql_server_init() to make code more portable. libmysqld/libmysqld.c: Cleanup merge/Makefile.am: Rename to use common prefix. merge/mrg_close.c: Rename to use common prefix. merge/mrg_create.c: Rename to use common prefix. merge/mrg_def.h: Rename to use common prefix. merge/mrg_delete.c: Rename to use common prefix. merge/mrg_extra.c: Rename to use common prefix. merge/mrg_info.c: Rename to use common prefix. merge/mrg_locking.c: Rename to use common prefix. merge/mrg_open.c: Rename to use common prefix. merge/mrg_panic.c: Rename to use common prefix. merge/mrg_rrnd.c: Rename to use common prefix. merge/mrg_rsame.c: Rename to use common prefix. merge/mrg_static.c: Rename to use common prefix. merge/mrg_update.c: Rename to use common prefix. myisam/myisamlog.c: Use new convert_dirname myisammrg/Makefile.am: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_close.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_create.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_def.h: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_delete.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_extra.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_info.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_locking.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_open.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_panic.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_queue.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rfirst.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rkey.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rlast.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rnext.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rprev.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rrnd.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rsame.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_static.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_update.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_write.c: Renamed mymrgdef.h -> myrg_def.h mysql-test/include/master-slave.inc: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/mysql-test-run.sh: cleanup mysql-test/r/auto_increment.result: Cleanup of tests for embedded version mysql-test/r/func_system.result: Cleanup of tests for embedded version mysql-test/r/isam.result: Cleanup of tests for embedded version mysql-test/r/lock.result: Cleanup of tests for embedded version mysql-test/r/show_check.result: Cleanup of tests for embedded version mysql-test/t/auto_increment.test: Cleanup of tests for embedded version mysql-test/t/backup.test: Cleanup of tests for embedded version mysql-test/t/count_distinct2-master.opt: Cleanup of tests for embedded version mysql-test/t/count_distinct2.test: Cleanup of tests for embedded version mysql-test/t/create.test: Cleanup of tests for embedded version mysql-test/t/flush.test: Cleanup of tests for embedded version mysql-test/t/func_system.test: Cleanup of tests for embedded version mysql-test/t/isam.test: Cleanup of tests for embedded version mysql-test/t/kill.test: Cleanup of tests for embedded version mysql-test/t/lock.test: Cleanup of tests for embedded version mysql-test/t/order_fill_sortbuf-master.opt: Cleanup of tests for embedded version mysql-test/t/rpl000015.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000016.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000017.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000018.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/show_check.test: Move ISAM specific tests to isam.test mysql-test/t/status.test: Cleanup of tests for embedded version mysql-test/t/tablelock.test: Cleanup mysys/Makefile.am: Removed not used mf_pack2.c mysys/charset.c: new convert_dirname mysys/default.c: new convert_dirname mysys/mf_dirname.c: Changed convert_dirname() to be more general mysys/mf_format.c: Changed bit flags to fn_format() to defines. Added handling of relative filenames BitKeeper/etc/ignore: Added libmysqld/examples/test-gdbinit scripts/mysql_explain_log to the ignore list mysys/mf_pack.c: new convert_dirname mysys/mf_tempfile.c: new convert_dirname scripts/Makefile.am: Adde mysql_explain_log scripts/mysql_config.sh: Added support of --libmysqld-libs sql/Makefile.am: Fix to use 'innodb_system_libs' sql/ha_innobase.cc: Make InnoDB startup/shutdown silent in embedded version sql/ha_isam.cc: Added handling of HAVE_ISAM sql/ha_isammrg.cc: Added handling of HAVE_ISAM sql/ha_myisam.cc: Handle relative paths; Needed to support BACKUP TABLE in embedded version sql/ha_myisammrg.cc: Rename of filenames sql/handler.cc: Added handling of HAVE_ISAM sql/item_func.cc: Fix for ecc (Intel Compiler) sql/mysql_priv.h: Added global variable 'mysql_embedded' sql/mysqld.cc: Use HAVE_ISAM sql/sql_parse.cc: Use new convert_dirname sql/sql_select.cc: Fix for ecc (Intel Compiler) sql/sql_table.cc: Added handling of relative filenames strings/Makefile.am: Don't compile both assembler and C functions on x86 strings/bchange.c: cleanup
2001-10-08 03:58:07 +02:00
connect (reader,localhost,root,,);
connect (writer,localhost,root,,);
connection locker;
create table t1(n int);
insert into t1 values (1);
Implement new type-of-operation-aware metadata locks. Add a wait-for graph based deadlock detector to the MDL subsystem. Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and bug #37346 "innodb does not detect deadlock between update and alter table". The first bug manifested itself as an unwarranted abort of a transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER statement, when this transaction tried to repeat use of a table, which it has already used in a similar fashion before ALTER started. The second bug showed up as a deadlock between table-level locks and InnoDB row locks, which was "detected" only after innodb_lock_wait_timeout timeout. A transaction would start using the table and modify a few rows. Then ALTER TABLE would come in, and start copying rows into a temporary table. Eventually it would stumble on the modified records and get blocked on a row lock. The first transaction would try to do more updates, and get blocked on thr_lock.c lock. This situation of circular wait would only get resolved by a timeout. Both these bugs stemmed from inadequate solutions to the problem of deadlocks occurring between different locking subsystems. In the first case we tried to avoid deadlocks between metadata locking and table-level locking subsystems, when upgrading shared metadata lock to exclusive one. Transactions holding the shared lock on the table and waiting for some table-level lock used to be aborted too aggressively. We also allowed ALTER TABLE to start in presence of transactions that modify the subject table. ALTER TABLE acquires TL_WRITE_ALLOW_READ lock at start, and that block all writes against the table (naturally, we don't want any writes to be lost when switching the old and the new table). TL_WRITE_ALLOW_READ lock, in turn, would block the started transaction on thr_lock.c lock, should they do more updates. This, again, lead to the need to abort such transactions. The second bug occurred simply because we didn't have any mechanism to detect deadlocks between the table-level locks in thr_lock.c and row-level locks in InnoDB, other than innodb_lock_wait_timeout. This patch solves both these problems by moving lock conflicts which are causing these deadlocks into the metadata locking subsystem, thus making it possible to avoid or detect such deadlocks inside MDL. To do this we introduce new type-of-operation-aware metadata locks, which allow MDL subsystem to know not only the fact that transaction has used or is going to use some object but also what kind of operation it has carried out or going to carry out on the object. This, along with the addition of a special kind of upgradable metadata lock, allows ALTER TABLE to wait until all transactions which has updated the table to go away. This solves the second issue. Another special type of upgradable metadata lock is acquired by LOCK TABLE WRITE. This second lock type allows to solve the first issue, since abortion of table-level locks in event of DDL under LOCK TABLES becomes also unnecessary. Below follows the list of incompatible changes introduced by this patch: - From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those statements that acquire TL_WRITE_ALLOW_READ lock) wait for all transactions which has *updated* the table to complete. - From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE (i.e. all statements which acquire TL_WRITE table-level lock) wait for all transaction which *updated or read* from the table to complete. As a consequence, innodb_table_locks=0 option no longer applies to LOCK TABLES ... WRITE. - DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort statements or transactions which use tables being dropped or renamed, and instead wait for these transactions to complete. - Since LOCK TABLES WRITE now takes a special metadata lock, not compatible with with reads or writes against the subject table and transaction-wide, thr_lock.c deadlock avoidance algorithm that used to ensure absence of deadlocks between LOCK TABLES WRITE and other statements is no longer sufficient, even for MyISAM. The wait-for graph based deadlock detector of MDL subsystem may sometimes be necessary and is involved. This may lead to ER_LOCK_DEADLOCK error produced for multi-statement transactions even if these only use MyISAM: session 1: session 2: begin; update t1 ... lock table t2 write, t1 write; -- gets a lock on t2, blocks on t1 update t2 ... (ER_LOCK_DEADLOCK) - Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE was abandoned. LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same priority as the usual LOCK TABLE ... WRITE. SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in the wait queue. - We do not take upgradable metadata locks on implicitly locked tables. So if one has, say, a view v1 that uses table t1, and issues: LOCK TABLE v1 WRITE; FLUSH TABLE t1; -- (or just 'FLUSH TABLES'), an error is produced. In order to be able to perform DDL on a table under LOCK TABLES, the table must be locked explicitly in the LOCK TABLES list. mysql-test/include/handler.inc: Adjusted test case to trigger an execution path on which bug 41110 "crash with handler command when used concurrently with alter table" and bug 41112 "crash in mysql_ha_close_table/get_lock_data with alter table" were originally discovered. Left old test case which no longer triggers this execution path for the sake of coverage. Added test coverage for HANDLER SQL statements and type-aware metadata locks. Added a test for the global shared lock and HANDLER SQL. Updated tests to take into account that the old simple deadlock detection heuristics was replaced with a graph-based deadlock detector. mysql-test/r/debug_sync.result: Updated results (see debug_sync.test). mysql-test/r/handler_innodb.result: Updated results (see handler.inc test). mysql-test/r/handler_myisam.result: Updated results (see handler.inc test). mysql-test/r/innodb-lock.result: Updated results (see innodb-lock.test). mysql-test/r/innodb_mysql_lock.result: Updated results (see innodb_mysql_lock.test). mysql-test/r/lock.result: Updated results (see lock.test). mysql-test/r/lock_multi.result: Updated results (see lock_multi.test). mysql-test/r/lock_sync.result: Updated results (see lock_sync.test). mysql-test/r/mdl_sync.result: Updated results (see mdl_sync.test). mysql-test/r/sp-threads.result: SHOW PROCESSLIST output has changed due to the fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/r/truncate_coverage.result: Updated results (see truncate_coverage.test). mysql-test/suite/funcs_1/datadict/processlist_val.inc: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/funcs_1/r/processlist_val_no_prot.result: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/rpl/t/rpl_sp.test: Updated to a new SHOW PROCESSLIST state name. mysql-test/t/debug_sync.test: Use LOCK TABLES READ instead of LOCK TABLES WRITE as the latter no longer allows to trigger execution path involving waiting on thr_lock.c lock and therefore reaching debug sync-point covered by this test. mysql-test/t/innodb-lock.test: Adjusted test case to the fact that innodb_table_locks=0 option is no longer supported, since LOCK TABLES WRITE handles all its conflicts within MDL subsystem. mysql-test/t/innodb_mysql_lock.test: Added test for bug #37346 "innodb does not detect deadlock between update and alter table". mysql-test/t/lock.test: Added test coverage which checks the fact that we no longer support DDL under LOCK TABLES on tables which were locked implicitly. Adjusted existing test cases accordingly. mysql-test/t/lock_multi.test: Added test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock". Adjusted other test cases to take into account the fact that waiting for LOCK TABLES ... WRITE now happens within MDL subsystem. mysql-test/t/lock_sync.test: Since LOCK TABLES ... WRITE now takes SNRW metadata lock for tables locked explicitly we have to implicitly lock InnoDB tables (through view) to trigger the table-level lock conflict between TL_WRITE and TL_WRITE_ALLOW_WRITE. mysql-test/t/mdl_sync.test: Added basic test coverage for type-of-operation-aware metadata locks. Also covered with tests some use cases involving HANDLER statements in which a deadlock could arise. Adjusted existing tests to take type-of-operation-aware MDL into account. mysql-test/t/multi_update.test: Update to a new SHOW PROCESSLIST state name. mysql-test/t/truncate_coverage.test: Adjusted test case after making LOCK TABLES WRITE to wait until transactions that use the table to be locked are completed. Updated to the changed name of DEBUG_SYNC point. sql/handler.cc: Global read lock functionality has been moved into a class. sql/lock.cc: Global read lock functionality has been moved into a class. Updated code to use the new MDL API. sql/mdl.cc: Introduced new type-of-operation aware metadata locks. To do this: - Changed MDL_lock to use one list for waiting requests and one list for granted requests. For each list, added a bitmap that holds information what lock types a list contains. Added a helper class MDL_lock::List to manipulate with granted and waited lists while keeping the bitmaps in sync with list contents. - Changed lock-compatibility functions to use bitmaps that define compatibility. - Introduced a graph based deadlock detector inspired by waiting_threads.c from Maria implementation. - Now that we have a deadlock detector, and no longer have a global lock to protect individual lock objects, but rather use an rw lock per object, removed redundant code for upgrade, and the global read lock. Changed the MDL API to no longer require the caller to acquire the global intention exclusive lock by means of a separate method. Removed a few more methods that became redundant. - Removed deadlock detection heuristic, it has been made obsolete by the deadlock detector. - With operation-type-aware metadata locks, MDL subsystem has become aware of potential conflicts between DDL and open transactions. This made it possible to remove calls to mysql_abort_transactions_with_shared_lock() from acquisition paths for exclusive lock and lock upgrade. Now we can simply wait for these transactions to complete without fear of deadlock. Function mysql_lock_abort() has also become unnecessary for all conflicting cases except when a DDL conflicts with a connection that has an open HANDLER. sql/mdl.h: Introduced new type-of-operation aware metadata locks. Introduced a graph based deadlock detector and supporting methods. Added comments. God rid of redundant API calls. Renamed m_lt_or_ha_sentinel to m_trans_sentinel, since now it guards the global read lock as well as LOCK TABLES and HANDLER locks. sql/mysql_priv.h: Moved the global read lock functionality into a class. Added MYSQL_OPEN_FORCE_SHARED_MDL flag which forces open_tables() to take MDL_SHARED on tables instead of metadata locks specified in the parser. We use this to allow PREPARE run concurrently in presence of LOCK TABLES ... WRITE. Added signature for find_table_for_mdl_ugprade(). sql/set_var.cc: Global read lock functionality has been moved into a class. sql/sp_head.cc: When creating TABLE_LIST elements for prelocking or system tables set the type of request for metadata lock according to the operation that will be performed on the table. sql/sql_base.cc: - Updated code to use the new MDL API. - In order to avoid locks starvation we take upgradable locks all at once. As result implicitly locked tables no longer get an upgradable lock. Consequently DDL and FLUSH TABLES for such tables is prohibited. find_write_locked_table() was replaced by find_table_for_mdl_upgrade() function. open_table() was adjusted to return TABLE instance with upgradable ticket when necessary. - We no longer wait for all locks on OT_WAIT back off action -- only on the lock that caused the wait conflict. Moreover, now we distinguish cases when we have to wait due to conflict in MDL and old version of table in TDC. - Upate mysql_notify_threads_having_share_locks() to only abort thr_lock.c waits of threads that have open HANDLERs, since lock conflicts with only these threads now can lead to deadlocks not detectable by the MDL deadlock detector. - Remove mysql_abort_transactions_with_shared_locks() which is no longer needed. sql/sql_class.cc: Global read lock functionality has been moved into a class. Re-arranged code in THD::cleanup() to simplify assert. sql/sql_class.h: Introduced class to incapsulate global read lock functionality. Now sentinel in MDL subsystem guards the global read lock as well as LOCK TABLES and HANDLER locks. Adjusted code accordingly. sql/sql_db.cc: Global read lock functionality has been moved into a class. sql/sql_delete.cc: We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result TRUNCATE TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_handler.cc: Inform MDL_context about presence of open HANDLERs. Since HANLDERs break MDL protocol by acquiring table-level lock while holding only S metadata lock on a table MDL subsystem should take special care about such contexts (Now this is the only case when mysql_lock_abort() is used). sql/sql_parse.cc: Global read lock functionality has been moved into a class. Do not take upgradable metadata locks when opening tables for CREATE TABLE SELECT as it is not necessary and limits concurrency. When initializing TABLE_LIST objects before adding them to the table list set the type of request for metadata lock according to the operation that will be performed on the table. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result FLUSH TABLES is no longer allowed for such tables. sql/sql_prepare.cc: Use MYSQL_OPEN_FORCE_SHARED_MDL flag when opening tables during PREPARE. This allows PREPARE to run concurrently in presence of LOCK TABLES ... WRITE. sql/sql_rename.cc: Global read lock functionality has been moved into a class. sql/sql_show.cc: Updated code to use the new MDL API. sql/sql_table.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result DROP TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_trigger.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result CREATE/DROP TRIGGER is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_view.cc: Global read lock functionality has been moved into a class. Fixed results of wrong merge that led to misuse of GLR API. CREATE VIEW statement is not a commit statement. sql/table.cc: When resetting TABLE_LIST objects for PS or SP re-execution set the type of request for metadata lock according to the operation that will be performed on the table. Do the same in auxiliary function initializing metadata lock requests in a table list. sql/table.h: When initializing TABLE_LIST objects set the type of request for metadata lock according to the operation that will be performed on the table. sql/transaction.cc: Global read lock functionality has been moved into a class.
2010-02-01 12:43:06 +01:00
connection locker2;
select get_lock("mysqltest_lock", 100);
connection locker;
send
update t1 set n = 2 and get_lock('mysqltest_lock', 100);
Updated manual about embedded version. Speed up column-completion in 'mysql' Don't use ISAM if HAVE_ISAM is not defined A lot of fixes for the embedded version. All libraries are now included in libmysqld.a Changed arguments to convert_dirname() to make it more general. Renamed files in the 'merge' directory to all use a common prefix. Don't compile both assembler and C functions on x86 BitKeeper/deleted/.del-mf_pack2.c~f07795bbcf57be7: Delete: mysys/mf_pack2.c Docs/manual.texi: Updated chapter about embedded version acinclude.m4: Fix for using BDB and InnoDB with embedded client/completion_hash.cc: Speed up memory allocation client/completion_hash.h: Speed up memory allocation client/mysql.cc: Speed up memory allocation client/mysqldump.c: Fix to use now convert_dirname client/mysqltest.c: Fixed memory allocation bugs. Added --basedir=#, --compress=#, --server-arg, --server-file. Fixes for embedded version Changed silent mode to -s instead of -q include/my_global.h: Update to use HAVE_ISAM include/my_sys.h: Cleanup of fn_format() include/mysql.h: Prepare FIELD struct for 4.1 include/mysql_embed.h: Don't use ISAM in embedded version innobase/include/srv0srv.h: Make InnoDB startup/shutdown silent in embedded version innobase/log/log0log.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0srv.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0start.c: Make InnoDB startup/shutdown silent in embedded version isam/isamlog.c: new convert_dirname libmysql/libmysql.c: Prepare for 4.1 libmysqld/Makefile.am: Changed to create one libmysqld.a file that includes all other libraries. libmysqld/examples/Makefile.am: Changed to use new libmysqld.a libmysqld/examples/test-run: Lot's of fixes to get new mysqltest with embedded MySQL to work. libmysqld/lib_sql.cc: Changed type of arguments for mysql_server_init() to make code more portable. libmysqld/libmysqld.c: Cleanup merge/Makefile.am: Rename to use common prefix. merge/mrg_close.c: Rename to use common prefix. merge/mrg_create.c: Rename to use common prefix. merge/mrg_def.h: Rename to use common prefix. merge/mrg_delete.c: Rename to use common prefix. merge/mrg_extra.c: Rename to use common prefix. merge/mrg_info.c: Rename to use common prefix. merge/mrg_locking.c: Rename to use common prefix. merge/mrg_open.c: Rename to use common prefix. merge/mrg_panic.c: Rename to use common prefix. merge/mrg_rrnd.c: Rename to use common prefix. merge/mrg_rsame.c: Rename to use common prefix. merge/mrg_static.c: Rename to use common prefix. merge/mrg_update.c: Rename to use common prefix. myisam/myisamlog.c: Use new convert_dirname myisammrg/Makefile.am: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_close.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_create.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_def.h: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_delete.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_extra.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_info.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_locking.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_open.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_panic.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_queue.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rfirst.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rkey.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rlast.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rnext.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rprev.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rrnd.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rsame.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_static.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_update.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_write.c: Renamed mymrgdef.h -> myrg_def.h mysql-test/include/master-slave.inc: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/mysql-test-run.sh: cleanup mysql-test/r/auto_increment.result: Cleanup of tests for embedded version mysql-test/r/func_system.result: Cleanup of tests for embedded version mysql-test/r/isam.result: Cleanup of tests for embedded version mysql-test/r/lock.result: Cleanup of tests for embedded version mysql-test/r/show_check.result: Cleanup of tests for embedded version mysql-test/t/auto_increment.test: Cleanup of tests for embedded version mysql-test/t/backup.test: Cleanup of tests for embedded version mysql-test/t/count_distinct2-master.opt: Cleanup of tests for embedded version mysql-test/t/count_distinct2.test: Cleanup of tests for embedded version mysql-test/t/create.test: Cleanup of tests for embedded version mysql-test/t/flush.test: Cleanup of tests for embedded version mysql-test/t/func_system.test: Cleanup of tests for embedded version mysql-test/t/isam.test: Cleanup of tests for embedded version mysql-test/t/kill.test: Cleanup of tests for embedded version mysql-test/t/lock.test: Cleanup of tests for embedded version mysql-test/t/order_fill_sortbuf-master.opt: Cleanup of tests for embedded version mysql-test/t/rpl000015.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000016.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000017.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000018.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/show_check.test: Move ISAM specific tests to isam.test mysql-test/t/status.test: Cleanup of tests for embedded version mysql-test/t/tablelock.test: Cleanup mysys/Makefile.am: Removed not used mf_pack2.c mysys/charset.c: new convert_dirname mysys/default.c: new convert_dirname mysys/mf_dirname.c: Changed convert_dirname() to be more general mysys/mf_format.c: Changed bit flags to fn_format() to defines. Added handling of relative filenames BitKeeper/etc/ignore: Added libmysqld/examples/test-gdbinit scripts/mysql_explain_log to the ignore list mysys/mf_pack.c: new convert_dirname mysys/mf_tempfile.c: new convert_dirname scripts/Makefile.am: Adde mysql_explain_log scripts/mysql_config.sh: Added support of --libmysqld-libs sql/Makefile.am: Fix to use 'innodb_system_libs' sql/ha_innobase.cc: Make InnoDB startup/shutdown silent in embedded version sql/ha_isam.cc: Added handling of HAVE_ISAM sql/ha_isammrg.cc: Added handling of HAVE_ISAM sql/ha_myisam.cc: Handle relative paths; Needed to support BACKUP TABLE in embedded version sql/ha_myisammrg.cc: Rename of filenames sql/handler.cc: Added handling of HAVE_ISAM sql/item_func.cc: Fix for ecc (Intel Compiler) sql/mysql_priv.h: Added global variable 'mysql_embedded' sql/mysqld.cc: Use HAVE_ISAM sql/sql_parse.cc: Use new convert_dirname sql/sql_select.cc: Fix for ecc (Intel Compiler) sql/sql_table.cc: Added handling of relative filenames strings/Makefile.am: Don't compile both assembler and C functions on x86 strings/bchange.c: cleanup
2001-10-08 03:58:07 +02:00
connection writer;
Implement new type-of-operation-aware metadata locks. Add a wait-for graph based deadlock detector to the MDL subsystem. Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and bug #37346 "innodb does not detect deadlock between update and alter table". The first bug manifested itself as an unwarranted abort of a transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER statement, when this transaction tried to repeat use of a table, which it has already used in a similar fashion before ALTER started. The second bug showed up as a deadlock between table-level locks and InnoDB row locks, which was "detected" only after innodb_lock_wait_timeout timeout. A transaction would start using the table and modify a few rows. Then ALTER TABLE would come in, and start copying rows into a temporary table. Eventually it would stumble on the modified records and get blocked on a row lock. The first transaction would try to do more updates, and get blocked on thr_lock.c lock. This situation of circular wait would only get resolved by a timeout. Both these bugs stemmed from inadequate solutions to the problem of deadlocks occurring between different locking subsystems. In the first case we tried to avoid deadlocks between metadata locking and table-level locking subsystems, when upgrading shared metadata lock to exclusive one. Transactions holding the shared lock on the table and waiting for some table-level lock used to be aborted too aggressively. We also allowed ALTER TABLE to start in presence of transactions that modify the subject table. ALTER TABLE acquires TL_WRITE_ALLOW_READ lock at start, and that block all writes against the table (naturally, we don't want any writes to be lost when switching the old and the new table). TL_WRITE_ALLOW_READ lock, in turn, would block the started transaction on thr_lock.c lock, should they do more updates. This, again, lead to the need to abort such transactions. The second bug occurred simply because we didn't have any mechanism to detect deadlocks between the table-level locks in thr_lock.c and row-level locks in InnoDB, other than innodb_lock_wait_timeout. This patch solves both these problems by moving lock conflicts which are causing these deadlocks into the metadata locking subsystem, thus making it possible to avoid or detect such deadlocks inside MDL. To do this we introduce new type-of-operation-aware metadata locks, which allow MDL subsystem to know not only the fact that transaction has used or is going to use some object but also what kind of operation it has carried out or going to carry out on the object. This, along with the addition of a special kind of upgradable metadata lock, allows ALTER TABLE to wait until all transactions which has updated the table to go away. This solves the second issue. Another special type of upgradable metadata lock is acquired by LOCK TABLE WRITE. This second lock type allows to solve the first issue, since abortion of table-level locks in event of DDL under LOCK TABLES becomes also unnecessary. Below follows the list of incompatible changes introduced by this patch: - From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those statements that acquire TL_WRITE_ALLOW_READ lock) wait for all transactions which has *updated* the table to complete. - From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE (i.e. all statements which acquire TL_WRITE table-level lock) wait for all transaction which *updated or read* from the table to complete. As a consequence, innodb_table_locks=0 option no longer applies to LOCK TABLES ... WRITE. - DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort statements or transactions which use tables being dropped or renamed, and instead wait for these transactions to complete. - Since LOCK TABLES WRITE now takes a special metadata lock, not compatible with with reads or writes against the subject table and transaction-wide, thr_lock.c deadlock avoidance algorithm that used to ensure absence of deadlocks between LOCK TABLES WRITE and other statements is no longer sufficient, even for MyISAM. The wait-for graph based deadlock detector of MDL subsystem may sometimes be necessary and is involved. This may lead to ER_LOCK_DEADLOCK error produced for multi-statement transactions even if these only use MyISAM: session 1: session 2: begin; update t1 ... lock table t2 write, t1 write; -- gets a lock on t2, blocks on t1 update t2 ... (ER_LOCK_DEADLOCK) - Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE was abandoned. LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same priority as the usual LOCK TABLE ... WRITE. SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in the wait queue. - We do not take upgradable metadata locks on implicitly locked tables. So if one has, say, a view v1 that uses table t1, and issues: LOCK TABLE v1 WRITE; FLUSH TABLE t1; -- (or just 'FLUSH TABLES'), an error is produced. In order to be able to perform DDL on a table under LOCK TABLES, the table must be locked explicitly in the LOCK TABLES list. mysql-test/include/handler.inc: Adjusted test case to trigger an execution path on which bug 41110 "crash with handler command when used concurrently with alter table" and bug 41112 "crash in mysql_ha_close_table/get_lock_data with alter table" were originally discovered. Left old test case which no longer triggers this execution path for the sake of coverage. Added test coverage for HANDLER SQL statements and type-aware metadata locks. Added a test for the global shared lock and HANDLER SQL. Updated tests to take into account that the old simple deadlock detection heuristics was replaced with a graph-based deadlock detector. mysql-test/r/debug_sync.result: Updated results (see debug_sync.test). mysql-test/r/handler_innodb.result: Updated results (see handler.inc test). mysql-test/r/handler_myisam.result: Updated results (see handler.inc test). mysql-test/r/innodb-lock.result: Updated results (see innodb-lock.test). mysql-test/r/innodb_mysql_lock.result: Updated results (see innodb_mysql_lock.test). mysql-test/r/lock.result: Updated results (see lock.test). mysql-test/r/lock_multi.result: Updated results (see lock_multi.test). mysql-test/r/lock_sync.result: Updated results (see lock_sync.test). mysql-test/r/mdl_sync.result: Updated results (see mdl_sync.test). mysql-test/r/sp-threads.result: SHOW PROCESSLIST output has changed due to the fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/r/truncate_coverage.result: Updated results (see truncate_coverage.test). mysql-test/suite/funcs_1/datadict/processlist_val.inc: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/funcs_1/r/processlist_val_no_prot.result: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/rpl/t/rpl_sp.test: Updated to a new SHOW PROCESSLIST state name. mysql-test/t/debug_sync.test: Use LOCK TABLES READ instead of LOCK TABLES WRITE as the latter no longer allows to trigger execution path involving waiting on thr_lock.c lock and therefore reaching debug sync-point covered by this test. mysql-test/t/innodb-lock.test: Adjusted test case to the fact that innodb_table_locks=0 option is no longer supported, since LOCK TABLES WRITE handles all its conflicts within MDL subsystem. mysql-test/t/innodb_mysql_lock.test: Added test for bug #37346 "innodb does not detect deadlock between update and alter table". mysql-test/t/lock.test: Added test coverage which checks the fact that we no longer support DDL under LOCK TABLES on tables which were locked implicitly. Adjusted existing test cases accordingly. mysql-test/t/lock_multi.test: Added test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock". Adjusted other test cases to take into account the fact that waiting for LOCK TABLES ... WRITE now happens within MDL subsystem. mysql-test/t/lock_sync.test: Since LOCK TABLES ... WRITE now takes SNRW metadata lock for tables locked explicitly we have to implicitly lock InnoDB tables (through view) to trigger the table-level lock conflict between TL_WRITE and TL_WRITE_ALLOW_WRITE. mysql-test/t/mdl_sync.test: Added basic test coverage for type-of-operation-aware metadata locks. Also covered with tests some use cases involving HANDLER statements in which a deadlock could arise. Adjusted existing tests to take type-of-operation-aware MDL into account. mysql-test/t/multi_update.test: Update to a new SHOW PROCESSLIST state name. mysql-test/t/truncate_coverage.test: Adjusted test case after making LOCK TABLES WRITE to wait until transactions that use the table to be locked are completed. Updated to the changed name of DEBUG_SYNC point. sql/handler.cc: Global read lock functionality has been moved into a class. sql/lock.cc: Global read lock functionality has been moved into a class. Updated code to use the new MDL API. sql/mdl.cc: Introduced new type-of-operation aware metadata locks. To do this: - Changed MDL_lock to use one list for waiting requests and one list for granted requests. For each list, added a bitmap that holds information what lock types a list contains. Added a helper class MDL_lock::List to manipulate with granted and waited lists while keeping the bitmaps in sync with list contents. - Changed lock-compatibility functions to use bitmaps that define compatibility. - Introduced a graph based deadlock detector inspired by waiting_threads.c from Maria implementation. - Now that we have a deadlock detector, and no longer have a global lock to protect individual lock objects, but rather use an rw lock per object, removed redundant code for upgrade, and the global read lock. Changed the MDL API to no longer require the caller to acquire the global intention exclusive lock by means of a separate method. Removed a few more methods that became redundant. - Removed deadlock detection heuristic, it has been made obsolete by the deadlock detector. - With operation-type-aware metadata locks, MDL subsystem has become aware of potential conflicts between DDL and open transactions. This made it possible to remove calls to mysql_abort_transactions_with_shared_lock() from acquisition paths for exclusive lock and lock upgrade. Now we can simply wait for these transactions to complete without fear of deadlock. Function mysql_lock_abort() has also become unnecessary for all conflicting cases except when a DDL conflicts with a connection that has an open HANDLER. sql/mdl.h: Introduced new type-of-operation aware metadata locks. Introduced a graph based deadlock detector and supporting methods. Added comments. God rid of redundant API calls. Renamed m_lt_or_ha_sentinel to m_trans_sentinel, since now it guards the global read lock as well as LOCK TABLES and HANDLER locks. sql/mysql_priv.h: Moved the global read lock functionality into a class. Added MYSQL_OPEN_FORCE_SHARED_MDL flag which forces open_tables() to take MDL_SHARED on tables instead of metadata locks specified in the parser. We use this to allow PREPARE run concurrently in presence of LOCK TABLES ... WRITE. Added signature for find_table_for_mdl_ugprade(). sql/set_var.cc: Global read lock functionality has been moved into a class. sql/sp_head.cc: When creating TABLE_LIST elements for prelocking or system tables set the type of request for metadata lock according to the operation that will be performed on the table. sql/sql_base.cc: - Updated code to use the new MDL API. - In order to avoid locks starvation we take upgradable locks all at once. As result implicitly locked tables no longer get an upgradable lock. Consequently DDL and FLUSH TABLES for such tables is prohibited. find_write_locked_table() was replaced by find_table_for_mdl_upgrade() function. open_table() was adjusted to return TABLE instance with upgradable ticket when necessary. - We no longer wait for all locks on OT_WAIT back off action -- only on the lock that caused the wait conflict. Moreover, now we distinguish cases when we have to wait due to conflict in MDL and old version of table in TDC. - Upate mysql_notify_threads_having_share_locks() to only abort thr_lock.c waits of threads that have open HANDLERs, since lock conflicts with only these threads now can lead to deadlocks not detectable by the MDL deadlock detector. - Remove mysql_abort_transactions_with_shared_locks() which is no longer needed. sql/sql_class.cc: Global read lock functionality has been moved into a class. Re-arranged code in THD::cleanup() to simplify assert. sql/sql_class.h: Introduced class to incapsulate global read lock functionality. Now sentinel in MDL subsystem guards the global read lock as well as LOCK TABLES and HANDLER locks. Adjusted code accordingly. sql/sql_db.cc: Global read lock functionality has been moved into a class. sql/sql_delete.cc: We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result TRUNCATE TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_handler.cc: Inform MDL_context about presence of open HANDLERs. Since HANLDERs break MDL protocol by acquiring table-level lock while holding only S metadata lock on a table MDL subsystem should take special care about such contexts (Now this is the only case when mysql_lock_abort() is used). sql/sql_parse.cc: Global read lock functionality has been moved into a class. Do not take upgradable metadata locks when opening tables for CREATE TABLE SELECT as it is not necessary and limits concurrency. When initializing TABLE_LIST objects before adding them to the table list set the type of request for metadata lock according to the operation that will be performed on the table. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result FLUSH TABLES is no longer allowed for such tables. sql/sql_prepare.cc: Use MYSQL_OPEN_FORCE_SHARED_MDL flag when opening tables during PREPARE. This allows PREPARE to run concurrently in presence of LOCK TABLES ... WRITE. sql/sql_rename.cc: Global read lock functionality has been moved into a class. sql/sql_show.cc: Updated code to use the new MDL API. sql/sql_table.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result DROP TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_trigger.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result CREATE/DROP TRIGGER is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_view.cc: Global read lock functionality has been moved into a class. Fixed results of wrong merge that led to misuse of GLR API. CREATE VIEW statement is not a commit statement. sql/table.cc: When resetting TABLE_LIST objects for PS or SP re-execution set the type of request for metadata lock according to the operation that will be performed on the table. Do the same in auxiliary function initializing metadata lock requests in a table list. sql/table.h: When initializing TABLE_LIST objects set the type of request for metadata lock according to the operation that will be performed on the table. sql/transaction.cc: Global read lock functionality has been moved into a class.
2010-02-01 12:43:06 +01:00
# Wait till above update gets blocked on a user lock.
let $wait_condition=
select count(*) = 1 from information_schema.processlist
where state = "User lock" and info = "update t1 set n = 2 and get_lock('mysqltest_lock', 100)";
--source include/wait_condition.inc
send
update low_priority t1 set n = 4;
Updated manual about embedded version. Speed up column-completion in 'mysql' Don't use ISAM if HAVE_ISAM is not defined A lot of fixes for the embedded version. All libraries are now included in libmysqld.a Changed arguments to convert_dirname() to make it more general. Renamed files in the 'merge' directory to all use a common prefix. Don't compile both assembler and C functions on x86 BitKeeper/deleted/.del-mf_pack2.c~f07795bbcf57be7: Delete: mysys/mf_pack2.c Docs/manual.texi: Updated chapter about embedded version acinclude.m4: Fix for using BDB and InnoDB with embedded client/completion_hash.cc: Speed up memory allocation client/completion_hash.h: Speed up memory allocation client/mysql.cc: Speed up memory allocation client/mysqldump.c: Fix to use now convert_dirname client/mysqltest.c: Fixed memory allocation bugs. Added --basedir=#, --compress=#, --server-arg, --server-file. Fixes for embedded version Changed silent mode to -s instead of -q include/my_global.h: Update to use HAVE_ISAM include/my_sys.h: Cleanup of fn_format() include/mysql.h: Prepare FIELD struct for 4.1 include/mysql_embed.h: Don't use ISAM in embedded version innobase/include/srv0srv.h: Make InnoDB startup/shutdown silent in embedded version innobase/log/log0log.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0srv.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0start.c: Make InnoDB startup/shutdown silent in embedded version isam/isamlog.c: new convert_dirname libmysql/libmysql.c: Prepare for 4.1 libmysqld/Makefile.am: Changed to create one libmysqld.a file that includes all other libraries. libmysqld/examples/Makefile.am: Changed to use new libmysqld.a libmysqld/examples/test-run: Lot's of fixes to get new mysqltest with embedded MySQL to work. libmysqld/lib_sql.cc: Changed type of arguments for mysql_server_init() to make code more portable. libmysqld/libmysqld.c: Cleanup merge/Makefile.am: Rename to use common prefix. merge/mrg_close.c: Rename to use common prefix. merge/mrg_create.c: Rename to use common prefix. merge/mrg_def.h: Rename to use common prefix. merge/mrg_delete.c: Rename to use common prefix. merge/mrg_extra.c: Rename to use common prefix. merge/mrg_info.c: Rename to use common prefix. merge/mrg_locking.c: Rename to use common prefix. merge/mrg_open.c: Rename to use common prefix. merge/mrg_panic.c: Rename to use common prefix. merge/mrg_rrnd.c: Rename to use common prefix. merge/mrg_rsame.c: Rename to use common prefix. merge/mrg_static.c: Rename to use common prefix. merge/mrg_update.c: Rename to use common prefix. myisam/myisamlog.c: Use new convert_dirname myisammrg/Makefile.am: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_close.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_create.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_def.h: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_delete.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_extra.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_info.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_locking.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_open.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_panic.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_queue.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rfirst.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rkey.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rlast.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rnext.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rprev.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rrnd.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rsame.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_static.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_update.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_write.c: Renamed mymrgdef.h -> myrg_def.h mysql-test/include/master-slave.inc: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/mysql-test-run.sh: cleanup mysql-test/r/auto_increment.result: Cleanup of tests for embedded version mysql-test/r/func_system.result: Cleanup of tests for embedded version mysql-test/r/isam.result: Cleanup of tests for embedded version mysql-test/r/lock.result: Cleanup of tests for embedded version mysql-test/r/show_check.result: Cleanup of tests for embedded version mysql-test/t/auto_increment.test: Cleanup of tests for embedded version mysql-test/t/backup.test: Cleanup of tests for embedded version mysql-test/t/count_distinct2-master.opt: Cleanup of tests for embedded version mysql-test/t/count_distinct2.test: Cleanup of tests for embedded version mysql-test/t/create.test: Cleanup of tests for embedded version mysql-test/t/flush.test: Cleanup of tests for embedded version mysql-test/t/func_system.test: Cleanup of tests for embedded version mysql-test/t/isam.test: Cleanup of tests for embedded version mysql-test/t/kill.test: Cleanup of tests for embedded version mysql-test/t/lock.test: Cleanup of tests for embedded version mysql-test/t/order_fill_sortbuf-master.opt: Cleanup of tests for embedded version mysql-test/t/rpl000015.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000016.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000017.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000018.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/show_check.test: Move ISAM specific tests to isam.test mysql-test/t/status.test: Cleanup of tests for embedded version mysql-test/t/tablelock.test: Cleanup mysys/Makefile.am: Removed not used mf_pack2.c mysys/charset.c: new convert_dirname mysys/default.c: new convert_dirname mysys/mf_dirname.c: Changed convert_dirname() to be more general mysys/mf_format.c: Changed bit flags to fn_format() to defines. Added handling of relative filenames BitKeeper/etc/ignore: Added libmysqld/examples/test-gdbinit scripts/mysql_explain_log to the ignore list mysys/mf_pack.c: new convert_dirname mysys/mf_tempfile.c: new convert_dirname scripts/Makefile.am: Adde mysql_explain_log scripts/mysql_config.sh: Added support of --libmysqld-libs sql/Makefile.am: Fix to use 'innodb_system_libs' sql/ha_innobase.cc: Make InnoDB startup/shutdown silent in embedded version sql/ha_isam.cc: Added handling of HAVE_ISAM sql/ha_isammrg.cc: Added handling of HAVE_ISAM sql/ha_myisam.cc: Handle relative paths; Needed to support BACKUP TABLE in embedded version sql/ha_myisammrg.cc: Rename of filenames sql/handler.cc: Added handling of HAVE_ISAM sql/item_func.cc: Fix for ecc (Intel Compiler) sql/mysql_priv.h: Added global variable 'mysql_embedded' sql/mysqld.cc: Use HAVE_ISAM sql/sql_parse.cc: Use new convert_dirname sql/sql_select.cc: Fix for ecc (Intel Compiler) sql/sql_table.cc: Added handling of relative filenames strings/Makefile.am: Don't compile both assembler and C functions on x86 strings/bchange.c: cleanup
2001-10-08 03:58:07 +02:00
connection reader;
# Sleep a bit till the update of connection writer is in work and hangs
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
where state = "Waiting for table level lock" and
info = "update low_priority t1 set n = 4";
--source include/wait_condition.inc
send
select n from t1;
Implement new type-of-operation-aware metadata locks. Add a wait-for graph based deadlock detector to the MDL subsystem. Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and bug #37346 "innodb does not detect deadlock between update and alter table". The first bug manifested itself as an unwarranted abort of a transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER statement, when this transaction tried to repeat use of a table, which it has already used in a similar fashion before ALTER started. The second bug showed up as a deadlock between table-level locks and InnoDB row locks, which was "detected" only after innodb_lock_wait_timeout timeout. A transaction would start using the table and modify a few rows. Then ALTER TABLE would come in, and start copying rows into a temporary table. Eventually it would stumble on the modified records and get blocked on a row lock. The first transaction would try to do more updates, and get blocked on thr_lock.c lock. This situation of circular wait would only get resolved by a timeout. Both these bugs stemmed from inadequate solutions to the problem of deadlocks occurring between different locking subsystems. In the first case we tried to avoid deadlocks between metadata locking and table-level locking subsystems, when upgrading shared metadata lock to exclusive one. Transactions holding the shared lock on the table and waiting for some table-level lock used to be aborted too aggressively. We also allowed ALTER TABLE to start in presence of transactions that modify the subject table. ALTER TABLE acquires TL_WRITE_ALLOW_READ lock at start, and that block all writes against the table (naturally, we don't want any writes to be lost when switching the old and the new table). TL_WRITE_ALLOW_READ lock, in turn, would block the started transaction on thr_lock.c lock, should they do more updates. This, again, lead to the need to abort such transactions. The second bug occurred simply because we didn't have any mechanism to detect deadlocks between the table-level locks in thr_lock.c and row-level locks in InnoDB, other than innodb_lock_wait_timeout. This patch solves both these problems by moving lock conflicts which are causing these deadlocks into the metadata locking subsystem, thus making it possible to avoid or detect such deadlocks inside MDL. To do this we introduce new type-of-operation-aware metadata locks, which allow MDL subsystem to know not only the fact that transaction has used or is going to use some object but also what kind of operation it has carried out or going to carry out on the object. This, along with the addition of a special kind of upgradable metadata lock, allows ALTER TABLE to wait until all transactions which has updated the table to go away. This solves the second issue. Another special type of upgradable metadata lock is acquired by LOCK TABLE WRITE. This second lock type allows to solve the first issue, since abortion of table-level locks in event of DDL under LOCK TABLES becomes also unnecessary. Below follows the list of incompatible changes introduced by this patch: - From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those statements that acquire TL_WRITE_ALLOW_READ lock) wait for all transactions which has *updated* the table to complete. - From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE (i.e. all statements which acquire TL_WRITE table-level lock) wait for all transaction which *updated or read* from the table to complete. As a consequence, innodb_table_locks=0 option no longer applies to LOCK TABLES ... WRITE. - DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort statements or transactions which use tables being dropped or renamed, and instead wait for these transactions to complete. - Since LOCK TABLES WRITE now takes a special metadata lock, not compatible with with reads or writes against the subject table and transaction-wide, thr_lock.c deadlock avoidance algorithm that used to ensure absence of deadlocks between LOCK TABLES WRITE and other statements is no longer sufficient, even for MyISAM. The wait-for graph based deadlock detector of MDL subsystem may sometimes be necessary and is involved. This may lead to ER_LOCK_DEADLOCK error produced for multi-statement transactions even if these only use MyISAM: session 1: session 2: begin; update t1 ... lock table t2 write, t1 write; -- gets a lock on t2, blocks on t1 update t2 ... (ER_LOCK_DEADLOCK) - Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE was abandoned. LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same priority as the usual LOCK TABLE ... WRITE. SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in the wait queue. - We do not take upgradable metadata locks on implicitly locked tables. So if one has, say, a view v1 that uses table t1, and issues: LOCK TABLE v1 WRITE; FLUSH TABLE t1; -- (or just 'FLUSH TABLES'), an error is produced. In order to be able to perform DDL on a table under LOCK TABLES, the table must be locked explicitly in the LOCK TABLES list. mysql-test/include/handler.inc: Adjusted test case to trigger an execution path on which bug 41110 "crash with handler command when used concurrently with alter table" and bug 41112 "crash in mysql_ha_close_table/get_lock_data with alter table" were originally discovered. Left old test case which no longer triggers this execution path for the sake of coverage. Added test coverage for HANDLER SQL statements and type-aware metadata locks. Added a test for the global shared lock and HANDLER SQL. Updated tests to take into account that the old simple deadlock detection heuristics was replaced with a graph-based deadlock detector. mysql-test/r/debug_sync.result: Updated results (see debug_sync.test). mysql-test/r/handler_innodb.result: Updated results (see handler.inc test). mysql-test/r/handler_myisam.result: Updated results (see handler.inc test). mysql-test/r/innodb-lock.result: Updated results (see innodb-lock.test). mysql-test/r/innodb_mysql_lock.result: Updated results (see innodb_mysql_lock.test). mysql-test/r/lock.result: Updated results (see lock.test). mysql-test/r/lock_multi.result: Updated results (see lock_multi.test). mysql-test/r/lock_sync.result: Updated results (see lock_sync.test). mysql-test/r/mdl_sync.result: Updated results (see mdl_sync.test). mysql-test/r/sp-threads.result: SHOW PROCESSLIST output has changed due to the fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/r/truncate_coverage.result: Updated results (see truncate_coverage.test). mysql-test/suite/funcs_1/datadict/processlist_val.inc: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/funcs_1/r/processlist_val_no_prot.result: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/rpl/t/rpl_sp.test: Updated to a new SHOW PROCESSLIST state name. mysql-test/t/debug_sync.test: Use LOCK TABLES READ instead of LOCK TABLES WRITE as the latter no longer allows to trigger execution path involving waiting on thr_lock.c lock and therefore reaching debug sync-point covered by this test. mysql-test/t/innodb-lock.test: Adjusted test case to the fact that innodb_table_locks=0 option is no longer supported, since LOCK TABLES WRITE handles all its conflicts within MDL subsystem. mysql-test/t/innodb_mysql_lock.test: Added test for bug #37346 "innodb does not detect deadlock between update and alter table". mysql-test/t/lock.test: Added test coverage which checks the fact that we no longer support DDL under LOCK TABLES on tables which were locked implicitly. Adjusted existing test cases accordingly. mysql-test/t/lock_multi.test: Added test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock". Adjusted other test cases to take into account the fact that waiting for LOCK TABLES ... WRITE now happens within MDL subsystem. mysql-test/t/lock_sync.test: Since LOCK TABLES ... WRITE now takes SNRW metadata lock for tables locked explicitly we have to implicitly lock InnoDB tables (through view) to trigger the table-level lock conflict between TL_WRITE and TL_WRITE_ALLOW_WRITE. mysql-test/t/mdl_sync.test: Added basic test coverage for type-of-operation-aware metadata locks. Also covered with tests some use cases involving HANDLER statements in which a deadlock could arise. Adjusted existing tests to take type-of-operation-aware MDL into account. mysql-test/t/multi_update.test: Update to a new SHOW PROCESSLIST state name. mysql-test/t/truncate_coverage.test: Adjusted test case after making LOCK TABLES WRITE to wait until transactions that use the table to be locked are completed. Updated to the changed name of DEBUG_SYNC point. sql/handler.cc: Global read lock functionality has been moved into a class. sql/lock.cc: Global read lock functionality has been moved into a class. Updated code to use the new MDL API. sql/mdl.cc: Introduced new type-of-operation aware metadata locks. To do this: - Changed MDL_lock to use one list for waiting requests and one list for granted requests. For each list, added a bitmap that holds information what lock types a list contains. Added a helper class MDL_lock::List to manipulate with granted and waited lists while keeping the bitmaps in sync with list contents. - Changed lock-compatibility functions to use bitmaps that define compatibility. - Introduced a graph based deadlock detector inspired by waiting_threads.c from Maria implementation. - Now that we have a deadlock detector, and no longer have a global lock to protect individual lock objects, but rather use an rw lock per object, removed redundant code for upgrade, and the global read lock. Changed the MDL API to no longer require the caller to acquire the global intention exclusive lock by means of a separate method. Removed a few more methods that became redundant. - Removed deadlock detection heuristic, it has been made obsolete by the deadlock detector. - With operation-type-aware metadata locks, MDL subsystem has become aware of potential conflicts between DDL and open transactions. This made it possible to remove calls to mysql_abort_transactions_with_shared_lock() from acquisition paths for exclusive lock and lock upgrade. Now we can simply wait for these transactions to complete without fear of deadlock. Function mysql_lock_abort() has also become unnecessary for all conflicting cases except when a DDL conflicts with a connection that has an open HANDLER. sql/mdl.h: Introduced new type-of-operation aware metadata locks. Introduced a graph based deadlock detector and supporting methods. Added comments. God rid of redundant API calls. Renamed m_lt_or_ha_sentinel to m_trans_sentinel, since now it guards the global read lock as well as LOCK TABLES and HANDLER locks. sql/mysql_priv.h: Moved the global read lock functionality into a class. Added MYSQL_OPEN_FORCE_SHARED_MDL flag which forces open_tables() to take MDL_SHARED on tables instead of metadata locks specified in the parser. We use this to allow PREPARE run concurrently in presence of LOCK TABLES ... WRITE. Added signature for find_table_for_mdl_ugprade(). sql/set_var.cc: Global read lock functionality has been moved into a class. sql/sp_head.cc: When creating TABLE_LIST elements for prelocking or system tables set the type of request for metadata lock according to the operation that will be performed on the table. sql/sql_base.cc: - Updated code to use the new MDL API. - In order to avoid locks starvation we take upgradable locks all at once. As result implicitly locked tables no longer get an upgradable lock. Consequently DDL and FLUSH TABLES for such tables is prohibited. find_write_locked_table() was replaced by find_table_for_mdl_upgrade() function. open_table() was adjusted to return TABLE instance with upgradable ticket when necessary. - We no longer wait for all locks on OT_WAIT back off action -- only on the lock that caused the wait conflict. Moreover, now we distinguish cases when we have to wait due to conflict in MDL and old version of table in TDC. - Upate mysql_notify_threads_having_share_locks() to only abort thr_lock.c waits of threads that have open HANDLERs, since lock conflicts with only these threads now can lead to deadlocks not detectable by the MDL deadlock detector. - Remove mysql_abort_transactions_with_shared_locks() which is no longer needed. sql/sql_class.cc: Global read lock functionality has been moved into a class. Re-arranged code in THD::cleanup() to simplify assert. sql/sql_class.h: Introduced class to incapsulate global read lock functionality. Now sentinel in MDL subsystem guards the global read lock as well as LOCK TABLES and HANDLER locks. Adjusted code accordingly. sql/sql_db.cc: Global read lock functionality has been moved into a class. sql/sql_delete.cc: We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result TRUNCATE TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_handler.cc: Inform MDL_context about presence of open HANDLERs. Since HANLDERs break MDL protocol by acquiring table-level lock while holding only S metadata lock on a table MDL subsystem should take special care about such contexts (Now this is the only case when mysql_lock_abort() is used). sql/sql_parse.cc: Global read lock functionality has been moved into a class. Do not take upgradable metadata locks when opening tables for CREATE TABLE SELECT as it is not necessary and limits concurrency. When initializing TABLE_LIST objects before adding them to the table list set the type of request for metadata lock according to the operation that will be performed on the table. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result FLUSH TABLES is no longer allowed for such tables. sql/sql_prepare.cc: Use MYSQL_OPEN_FORCE_SHARED_MDL flag when opening tables during PREPARE. This allows PREPARE to run concurrently in presence of LOCK TABLES ... WRITE. sql/sql_rename.cc: Global read lock functionality has been moved into a class. sql/sql_show.cc: Updated code to use the new MDL API. sql/sql_table.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result DROP TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_trigger.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result CREATE/DROP TRIGGER is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_view.cc: Global read lock functionality has been moved into a class. Fixed results of wrong merge that led to misuse of GLR API. CREATE VIEW statement is not a commit statement. sql/table.cc: When resetting TABLE_LIST objects for PS or SP re-execution set the type of request for metadata lock according to the operation that will be performed on the table. Do the same in auxiliary function initializing metadata lock requests in a table list. sql/table.h: When initializing TABLE_LIST objects set the type of request for metadata lock according to the operation that will be performed on the table. sql/transaction.cc: Global read lock functionality has been moved into a class.
2010-02-01 12:43:06 +01:00
connection locker2;
# Sleep a bit till the select of connection reader is in work and hangs
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
where state = "Waiting for table level lock" and
info = "select n from t1";
--source include/wait_condition.inc
Implement new type-of-operation-aware metadata locks. Add a wait-for graph based deadlock detector to the MDL subsystem. Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and bug #37346 "innodb does not detect deadlock between update and alter table". The first bug manifested itself as an unwarranted abort of a transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER statement, when this transaction tried to repeat use of a table, which it has already used in a similar fashion before ALTER started. The second bug showed up as a deadlock between table-level locks and InnoDB row locks, which was "detected" only after innodb_lock_wait_timeout timeout. A transaction would start using the table and modify a few rows. Then ALTER TABLE would come in, and start copying rows into a temporary table. Eventually it would stumble on the modified records and get blocked on a row lock. The first transaction would try to do more updates, and get blocked on thr_lock.c lock. This situation of circular wait would only get resolved by a timeout. Both these bugs stemmed from inadequate solutions to the problem of deadlocks occurring between different locking subsystems. In the first case we tried to avoid deadlocks between metadata locking and table-level locking subsystems, when upgrading shared metadata lock to exclusive one. Transactions holding the shared lock on the table and waiting for some table-level lock used to be aborted too aggressively. We also allowed ALTER TABLE to start in presence of transactions that modify the subject table. ALTER TABLE acquires TL_WRITE_ALLOW_READ lock at start, and that block all writes against the table (naturally, we don't want any writes to be lost when switching the old and the new table). TL_WRITE_ALLOW_READ lock, in turn, would block the started transaction on thr_lock.c lock, should they do more updates. This, again, lead to the need to abort such transactions. The second bug occurred simply because we didn't have any mechanism to detect deadlocks between the table-level locks in thr_lock.c and row-level locks in InnoDB, other than innodb_lock_wait_timeout. This patch solves both these problems by moving lock conflicts which are causing these deadlocks into the metadata locking subsystem, thus making it possible to avoid or detect such deadlocks inside MDL. To do this we introduce new type-of-operation-aware metadata locks, which allow MDL subsystem to know not only the fact that transaction has used or is going to use some object but also what kind of operation it has carried out or going to carry out on the object. This, along with the addition of a special kind of upgradable metadata lock, allows ALTER TABLE to wait until all transactions which has updated the table to go away. This solves the second issue. Another special type of upgradable metadata lock is acquired by LOCK TABLE WRITE. This second lock type allows to solve the first issue, since abortion of table-level locks in event of DDL under LOCK TABLES becomes also unnecessary. Below follows the list of incompatible changes introduced by this patch: - From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those statements that acquire TL_WRITE_ALLOW_READ lock) wait for all transactions which has *updated* the table to complete. - From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE (i.e. all statements which acquire TL_WRITE table-level lock) wait for all transaction which *updated or read* from the table to complete. As a consequence, innodb_table_locks=0 option no longer applies to LOCK TABLES ... WRITE. - DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort statements or transactions which use tables being dropped or renamed, and instead wait for these transactions to complete. - Since LOCK TABLES WRITE now takes a special metadata lock, not compatible with with reads or writes against the subject table and transaction-wide, thr_lock.c deadlock avoidance algorithm that used to ensure absence of deadlocks between LOCK TABLES WRITE and other statements is no longer sufficient, even for MyISAM. The wait-for graph based deadlock detector of MDL subsystem may sometimes be necessary and is involved. This may lead to ER_LOCK_DEADLOCK error produced for multi-statement transactions even if these only use MyISAM: session 1: session 2: begin; update t1 ... lock table t2 write, t1 write; -- gets a lock on t2, blocks on t1 update t2 ... (ER_LOCK_DEADLOCK) - Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE was abandoned. LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same priority as the usual LOCK TABLE ... WRITE. SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in the wait queue. - We do not take upgradable metadata locks on implicitly locked tables. So if one has, say, a view v1 that uses table t1, and issues: LOCK TABLE v1 WRITE; FLUSH TABLE t1; -- (or just 'FLUSH TABLES'), an error is produced. In order to be able to perform DDL on a table under LOCK TABLES, the table must be locked explicitly in the LOCK TABLES list. mysql-test/include/handler.inc: Adjusted test case to trigger an execution path on which bug 41110 "crash with handler command when used concurrently with alter table" and bug 41112 "crash in mysql_ha_close_table/get_lock_data with alter table" were originally discovered. Left old test case which no longer triggers this execution path for the sake of coverage. Added test coverage for HANDLER SQL statements and type-aware metadata locks. Added a test for the global shared lock and HANDLER SQL. Updated tests to take into account that the old simple deadlock detection heuristics was replaced with a graph-based deadlock detector. mysql-test/r/debug_sync.result: Updated results (see debug_sync.test). mysql-test/r/handler_innodb.result: Updated results (see handler.inc test). mysql-test/r/handler_myisam.result: Updated results (see handler.inc test). mysql-test/r/innodb-lock.result: Updated results (see innodb-lock.test). mysql-test/r/innodb_mysql_lock.result: Updated results (see innodb_mysql_lock.test). mysql-test/r/lock.result: Updated results (see lock.test). mysql-test/r/lock_multi.result: Updated results (see lock_multi.test). mysql-test/r/lock_sync.result: Updated results (see lock_sync.test). mysql-test/r/mdl_sync.result: Updated results (see mdl_sync.test). mysql-test/r/sp-threads.result: SHOW PROCESSLIST output has changed due to the fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/r/truncate_coverage.result: Updated results (see truncate_coverage.test). mysql-test/suite/funcs_1/datadict/processlist_val.inc: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/funcs_1/r/processlist_val_no_prot.result: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/rpl/t/rpl_sp.test: Updated to a new SHOW PROCESSLIST state name. mysql-test/t/debug_sync.test: Use LOCK TABLES READ instead of LOCK TABLES WRITE as the latter no longer allows to trigger execution path involving waiting on thr_lock.c lock and therefore reaching debug sync-point covered by this test. mysql-test/t/innodb-lock.test: Adjusted test case to the fact that innodb_table_locks=0 option is no longer supported, since LOCK TABLES WRITE handles all its conflicts within MDL subsystem. mysql-test/t/innodb_mysql_lock.test: Added test for bug #37346 "innodb does not detect deadlock between update and alter table". mysql-test/t/lock.test: Added test coverage which checks the fact that we no longer support DDL under LOCK TABLES on tables which were locked implicitly. Adjusted existing test cases accordingly. mysql-test/t/lock_multi.test: Added test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock". Adjusted other test cases to take into account the fact that waiting for LOCK TABLES ... WRITE now happens within MDL subsystem. mysql-test/t/lock_sync.test: Since LOCK TABLES ... WRITE now takes SNRW metadata lock for tables locked explicitly we have to implicitly lock InnoDB tables (through view) to trigger the table-level lock conflict between TL_WRITE and TL_WRITE_ALLOW_WRITE. mysql-test/t/mdl_sync.test: Added basic test coverage for type-of-operation-aware metadata locks. Also covered with tests some use cases involving HANDLER statements in which a deadlock could arise. Adjusted existing tests to take type-of-operation-aware MDL into account. mysql-test/t/multi_update.test: Update to a new SHOW PROCESSLIST state name. mysql-test/t/truncate_coverage.test: Adjusted test case after making LOCK TABLES WRITE to wait until transactions that use the table to be locked are completed. Updated to the changed name of DEBUG_SYNC point. sql/handler.cc: Global read lock functionality has been moved into a class. sql/lock.cc: Global read lock functionality has been moved into a class. Updated code to use the new MDL API. sql/mdl.cc: Introduced new type-of-operation aware metadata locks. To do this: - Changed MDL_lock to use one list for waiting requests and one list for granted requests. For each list, added a bitmap that holds information what lock types a list contains. Added a helper class MDL_lock::List to manipulate with granted and waited lists while keeping the bitmaps in sync with list contents. - Changed lock-compatibility functions to use bitmaps that define compatibility. - Introduced a graph based deadlock detector inspired by waiting_threads.c from Maria implementation. - Now that we have a deadlock detector, and no longer have a global lock to protect individual lock objects, but rather use an rw lock per object, removed redundant code for upgrade, and the global read lock. Changed the MDL API to no longer require the caller to acquire the global intention exclusive lock by means of a separate method. Removed a few more methods that became redundant. - Removed deadlock detection heuristic, it has been made obsolete by the deadlock detector. - With operation-type-aware metadata locks, MDL subsystem has become aware of potential conflicts between DDL and open transactions. This made it possible to remove calls to mysql_abort_transactions_with_shared_lock() from acquisition paths for exclusive lock and lock upgrade. Now we can simply wait for these transactions to complete without fear of deadlock. Function mysql_lock_abort() has also become unnecessary for all conflicting cases except when a DDL conflicts with a connection that has an open HANDLER. sql/mdl.h: Introduced new type-of-operation aware metadata locks. Introduced a graph based deadlock detector and supporting methods. Added comments. God rid of redundant API calls. Renamed m_lt_or_ha_sentinel to m_trans_sentinel, since now it guards the global read lock as well as LOCK TABLES and HANDLER locks. sql/mysql_priv.h: Moved the global read lock functionality into a class. Added MYSQL_OPEN_FORCE_SHARED_MDL flag which forces open_tables() to take MDL_SHARED on tables instead of metadata locks specified in the parser. We use this to allow PREPARE run concurrently in presence of LOCK TABLES ... WRITE. Added signature for find_table_for_mdl_ugprade(). sql/set_var.cc: Global read lock functionality has been moved into a class. sql/sp_head.cc: When creating TABLE_LIST elements for prelocking or system tables set the type of request for metadata lock according to the operation that will be performed on the table. sql/sql_base.cc: - Updated code to use the new MDL API. - In order to avoid locks starvation we take upgradable locks all at once. As result implicitly locked tables no longer get an upgradable lock. Consequently DDL and FLUSH TABLES for such tables is prohibited. find_write_locked_table() was replaced by find_table_for_mdl_upgrade() function. open_table() was adjusted to return TABLE instance with upgradable ticket when necessary. - We no longer wait for all locks on OT_WAIT back off action -- only on the lock that caused the wait conflict. Moreover, now we distinguish cases when we have to wait due to conflict in MDL and old version of table in TDC. - Upate mysql_notify_threads_having_share_locks() to only abort thr_lock.c waits of threads that have open HANDLERs, since lock conflicts with only these threads now can lead to deadlocks not detectable by the MDL deadlock detector. - Remove mysql_abort_transactions_with_shared_locks() which is no longer needed. sql/sql_class.cc: Global read lock functionality has been moved into a class. Re-arranged code in THD::cleanup() to simplify assert. sql/sql_class.h: Introduced class to incapsulate global read lock functionality. Now sentinel in MDL subsystem guards the global read lock as well as LOCK TABLES and HANDLER locks. Adjusted code accordingly. sql/sql_db.cc: Global read lock functionality has been moved into a class. sql/sql_delete.cc: We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result TRUNCATE TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_handler.cc: Inform MDL_context about presence of open HANDLERs. Since HANLDERs break MDL protocol by acquiring table-level lock while holding only S metadata lock on a table MDL subsystem should take special care about such contexts (Now this is the only case when mysql_lock_abort() is used). sql/sql_parse.cc: Global read lock functionality has been moved into a class. Do not take upgradable metadata locks when opening tables for CREATE TABLE SELECT as it is not necessary and limits concurrency. When initializing TABLE_LIST objects before adding them to the table list set the type of request for metadata lock according to the operation that will be performed on the table. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result FLUSH TABLES is no longer allowed for such tables. sql/sql_prepare.cc: Use MYSQL_OPEN_FORCE_SHARED_MDL flag when opening tables during PREPARE. This allows PREPARE to run concurrently in presence of LOCK TABLES ... WRITE. sql/sql_rename.cc: Global read lock functionality has been moved into a class. sql/sql_show.cc: Updated code to use the new MDL API. sql/sql_table.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result DROP TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_trigger.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result CREATE/DROP TRIGGER is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_view.cc: Global read lock functionality has been moved into a class. Fixed results of wrong merge that led to misuse of GLR API. CREATE VIEW statement is not a commit statement. sql/table.cc: When resetting TABLE_LIST objects for PS or SP re-execution set the type of request for metadata lock according to the operation that will be performed on the table. Do the same in auxiliary function initializing metadata lock requests in a table list. sql/table.h: When initializing TABLE_LIST objects set the type of request for metadata lock according to the operation that will be performed on the table. sql/transaction.cc: Global read lock functionality has been moved into a class.
2010-02-01 12:43:06 +01:00
select release_lock("mysqltest_lock");
connection locker;
reap;
select release_lock("mysqltest_lock");
Updated manual about embedded version. Speed up column-completion in 'mysql' Don't use ISAM if HAVE_ISAM is not defined A lot of fixes for the embedded version. All libraries are now included in libmysqld.a Changed arguments to convert_dirname() to make it more general. Renamed files in the 'merge' directory to all use a common prefix. Don't compile both assembler and C functions on x86 BitKeeper/deleted/.del-mf_pack2.c~f07795bbcf57be7: Delete: mysys/mf_pack2.c Docs/manual.texi: Updated chapter about embedded version acinclude.m4: Fix for using BDB and InnoDB with embedded client/completion_hash.cc: Speed up memory allocation client/completion_hash.h: Speed up memory allocation client/mysql.cc: Speed up memory allocation client/mysqldump.c: Fix to use now convert_dirname client/mysqltest.c: Fixed memory allocation bugs. Added --basedir=#, --compress=#, --server-arg, --server-file. Fixes for embedded version Changed silent mode to -s instead of -q include/my_global.h: Update to use HAVE_ISAM include/my_sys.h: Cleanup of fn_format() include/mysql.h: Prepare FIELD struct for 4.1 include/mysql_embed.h: Don't use ISAM in embedded version innobase/include/srv0srv.h: Make InnoDB startup/shutdown silent in embedded version innobase/log/log0log.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0srv.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0start.c: Make InnoDB startup/shutdown silent in embedded version isam/isamlog.c: new convert_dirname libmysql/libmysql.c: Prepare for 4.1 libmysqld/Makefile.am: Changed to create one libmysqld.a file that includes all other libraries. libmysqld/examples/Makefile.am: Changed to use new libmysqld.a libmysqld/examples/test-run: Lot's of fixes to get new mysqltest with embedded MySQL to work. libmysqld/lib_sql.cc: Changed type of arguments for mysql_server_init() to make code more portable. libmysqld/libmysqld.c: Cleanup merge/Makefile.am: Rename to use common prefix. merge/mrg_close.c: Rename to use common prefix. merge/mrg_create.c: Rename to use common prefix. merge/mrg_def.h: Rename to use common prefix. merge/mrg_delete.c: Rename to use common prefix. merge/mrg_extra.c: Rename to use common prefix. merge/mrg_info.c: Rename to use common prefix. merge/mrg_locking.c: Rename to use common prefix. merge/mrg_open.c: Rename to use common prefix. merge/mrg_panic.c: Rename to use common prefix. merge/mrg_rrnd.c: Rename to use common prefix. merge/mrg_rsame.c: Rename to use common prefix. merge/mrg_static.c: Rename to use common prefix. merge/mrg_update.c: Rename to use common prefix. myisam/myisamlog.c: Use new convert_dirname myisammrg/Makefile.am: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_close.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_create.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_def.h: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_delete.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_extra.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_info.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_locking.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_open.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_panic.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_queue.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rfirst.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rkey.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rlast.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rnext.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rprev.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rrnd.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rsame.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_static.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_update.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_write.c: Renamed mymrgdef.h -> myrg_def.h mysql-test/include/master-slave.inc: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/mysql-test-run.sh: cleanup mysql-test/r/auto_increment.result: Cleanup of tests for embedded version mysql-test/r/func_system.result: Cleanup of tests for embedded version mysql-test/r/isam.result: Cleanup of tests for embedded version mysql-test/r/lock.result: Cleanup of tests for embedded version mysql-test/r/show_check.result: Cleanup of tests for embedded version mysql-test/t/auto_increment.test: Cleanup of tests for embedded version mysql-test/t/backup.test: Cleanup of tests for embedded version mysql-test/t/count_distinct2-master.opt: Cleanup of tests for embedded version mysql-test/t/count_distinct2.test: Cleanup of tests for embedded version mysql-test/t/create.test: Cleanup of tests for embedded version mysql-test/t/flush.test: Cleanup of tests for embedded version mysql-test/t/func_system.test: Cleanup of tests for embedded version mysql-test/t/isam.test: Cleanup of tests for embedded version mysql-test/t/kill.test: Cleanup of tests for embedded version mysql-test/t/lock.test: Cleanup of tests for embedded version mysql-test/t/order_fill_sortbuf-master.opt: Cleanup of tests for embedded version mysql-test/t/rpl000015.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000016.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000017.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000018.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/show_check.test: Move ISAM specific tests to isam.test mysql-test/t/status.test: Cleanup of tests for embedded version mysql-test/t/tablelock.test: Cleanup mysys/Makefile.am: Removed not used mf_pack2.c mysys/charset.c: new convert_dirname mysys/default.c: new convert_dirname mysys/mf_dirname.c: Changed convert_dirname() to be more general mysys/mf_format.c: Changed bit flags to fn_format() to defines. Added handling of relative filenames BitKeeper/etc/ignore: Added libmysqld/examples/test-gdbinit scripts/mysql_explain_log to the ignore list mysys/mf_pack.c: new convert_dirname mysys/mf_tempfile.c: new convert_dirname scripts/Makefile.am: Adde mysql_explain_log scripts/mysql_config.sh: Added support of --libmysqld-libs sql/Makefile.am: Fix to use 'innodb_system_libs' sql/ha_innobase.cc: Make InnoDB startup/shutdown silent in embedded version sql/ha_isam.cc: Added handling of HAVE_ISAM sql/ha_isammrg.cc: Added handling of HAVE_ISAM sql/ha_myisam.cc: Handle relative paths; Needed to support BACKUP TABLE in embedded version sql/ha_myisammrg.cc: Rename of filenames sql/handler.cc: Added handling of HAVE_ISAM sql/item_func.cc: Fix for ecc (Intel Compiler) sql/mysql_priv.h: Added global variable 'mysql_embedded' sql/mysqld.cc: Use HAVE_ISAM sql/sql_parse.cc: Use new convert_dirname sql/sql_select.cc: Fix for ecc (Intel Compiler) sql/sql_table.cc: Added handling of relative filenames strings/Makefile.am: Don't compile both assembler and C functions on x86 strings/bchange.c: cleanup
2001-10-08 03:58:07 +02:00
connection writer;
reap;
connection reader;
reap;
drop table t1;
connection locker;
create table t1(n int);
2013-08-20 13:48:29 +02:00
insert into t1 values (1),(2);
Implement new type-of-operation-aware metadata locks. Add a wait-for graph based deadlock detector to the MDL subsystem. Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and bug #37346 "innodb does not detect deadlock between update and alter table". The first bug manifested itself as an unwarranted abort of a transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER statement, when this transaction tried to repeat use of a table, which it has already used in a similar fashion before ALTER started. The second bug showed up as a deadlock between table-level locks and InnoDB row locks, which was "detected" only after innodb_lock_wait_timeout timeout. A transaction would start using the table and modify a few rows. Then ALTER TABLE would come in, and start copying rows into a temporary table. Eventually it would stumble on the modified records and get blocked on a row lock. The first transaction would try to do more updates, and get blocked on thr_lock.c lock. This situation of circular wait would only get resolved by a timeout. Both these bugs stemmed from inadequate solutions to the problem of deadlocks occurring between different locking subsystems. In the first case we tried to avoid deadlocks between metadata locking and table-level locking subsystems, when upgrading shared metadata lock to exclusive one. Transactions holding the shared lock on the table and waiting for some table-level lock used to be aborted too aggressively. We also allowed ALTER TABLE to start in presence of transactions that modify the subject table. ALTER TABLE acquires TL_WRITE_ALLOW_READ lock at start, and that block all writes against the table (naturally, we don't want any writes to be lost when switching the old and the new table). TL_WRITE_ALLOW_READ lock, in turn, would block the started transaction on thr_lock.c lock, should they do more updates. This, again, lead to the need to abort such transactions. The second bug occurred simply because we didn't have any mechanism to detect deadlocks between the table-level locks in thr_lock.c and row-level locks in InnoDB, other than innodb_lock_wait_timeout. This patch solves both these problems by moving lock conflicts which are causing these deadlocks into the metadata locking subsystem, thus making it possible to avoid or detect such deadlocks inside MDL. To do this we introduce new type-of-operation-aware metadata locks, which allow MDL subsystem to know not only the fact that transaction has used or is going to use some object but also what kind of operation it has carried out or going to carry out on the object. This, along with the addition of a special kind of upgradable metadata lock, allows ALTER TABLE to wait until all transactions which has updated the table to go away. This solves the second issue. Another special type of upgradable metadata lock is acquired by LOCK TABLE WRITE. This second lock type allows to solve the first issue, since abortion of table-level locks in event of DDL under LOCK TABLES becomes also unnecessary. Below follows the list of incompatible changes introduced by this patch: - From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those statements that acquire TL_WRITE_ALLOW_READ lock) wait for all transactions which has *updated* the table to complete. - From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE (i.e. all statements which acquire TL_WRITE table-level lock) wait for all transaction which *updated or read* from the table to complete. As a consequence, innodb_table_locks=0 option no longer applies to LOCK TABLES ... WRITE. - DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort statements or transactions which use tables being dropped or renamed, and instead wait for these transactions to complete. - Since LOCK TABLES WRITE now takes a special metadata lock, not compatible with with reads or writes against the subject table and transaction-wide, thr_lock.c deadlock avoidance algorithm that used to ensure absence of deadlocks between LOCK TABLES WRITE and other statements is no longer sufficient, even for MyISAM. The wait-for graph based deadlock detector of MDL subsystem may sometimes be necessary and is involved. This may lead to ER_LOCK_DEADLOCK error produced for multi-statement transactions even if these only use MyISAM: session 1: session 2: begin; update t1 ... lock table t2 write, t1 write; -- gets a lock on t2, blocks on t1 update t2 ... (ER_LOCK_DEADLOCK) - Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE was abandoned. LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same priority as the usual LOCK TABLE ... WRITE. SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in the wait queue. - We do not take upgradable metadata locks on implicitly locked tables. So if one has, say, a view v1 that uses table t1, and issues: LOCK TABLE v1 WRITE; FLUSH TABLE t1; -- (or just 'FLUSH TABLES'), an error is produced. In order to be able to perform DDL on a table under LOCK TABLES, the table must be locked explicitly in the LOCK TABLES list. mysql-test/include/handler.inc: Adjusted test case to trigger an execution path on which bug 41110 "crash with handler command when used concurrently with alter table" and bug 41112 "crash in mysql_ha_close_table/get_lock_data with alter table" were originally discovered. Left old test case which no longer triggers this execution path for the sake of coverage. Added test coverage for HANDLER SQL statements and type-aware metadata locks. Added a test for the global shared lock and HANDLER SQL. Updated tests to take into account that the old simple deadlock detection heuristics was replaced with a graph-based deadlock detector. mysql-test/r/debug_sync.result: Updated results (see debug_sync.test). mysql-test/r/handler_innodb.result: Updated results (see handler.inc test). mysql-test/r/handler_myisam.result: Updated results (see handler.inc test). mysql-test/r/innodb-lock.result: Updated results (see innodb-lock.test). mysql-test/r/innodb_mysql_lock.result: Updated results (see innodb_mysql_lock.test). mysql-test/r/lock.result: Updated results (see lock.test). mysql-test/r/lock_multi.result: Updated results (see lock_multi.test). mysql-test/r/lock_sync.result: Updated results (see lock_sync.test). mysql-test/r/mdl_sync.result: Updated results (see mdl_sync.test). mysql-test/r/sp-threads.result: SHOW PROCESSLIST output has changed due to the fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/r/truncate_coverage.result: Updated results (see truncate_coverage.test). mysql-test/suite/funcs_1/datadict/processlist_val.inc: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/funcs_1/r/processlist_val_no_prot.result: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/rpl/t/rpl_sp.test: Updated to a new SHOW PROCESSLIST state name. mysql-test/t/debug_sync.test: Use LOCK TABLES READ instead of LOCK TABLES WRITE as the latter no longer allows to trigger execution path involving waiting on thr_lock.c lock and therefore reaching debug sync-point covered by this test. mysql-test/t/innodb-lock.test: Adjusted test case to the fact that innodb_table_locks=0 option is no longer supported, since LOCK TABLES WRITE handles all its conflicts within MDL subsystem. mysql-test/t/innodb_mysql_lock.test: Added test for bug #37346 "innodb does not detect deadlock between update and alter table". mysql-test/t/lock.test: Added test coverage which checks the fact that we no longer support DDL under LOCK TABLES on tables which were locked implicitly. Adjusted existing test cases accordingly. mysql-test/t/lock_multi.test: Added test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock". Adjusted other test cases to take into account the fact that waiting for LOCK TABLES ... WRITE now happens within MDL subsystem. mysql-test/t/lock_sync.test: Since LOCK TABLES ... WRITE now takes SNRW metadata lock for tables locked explicitly we have to implicitly lock InnoDB tables (through view) to trigger the table-level lock conflict between TL_WRITE and TL_WRITE_ALLOW_WRITE. mysql-test/t/mdl_sync.test: Added basic test coverage for type-of-operation-aware metadata locks. Also covered with tests some use cases involving HANDLER statements in which a deadlock could arise. Adjusted existing tests to take type-of-operation-aware MDL into account. mysql-test/t/multi_update.test: Update to a new SHOW PROCESSLIST state name. mysql-test/t/truncate_coverage.test: Adjusted test case after making LOCK TABLES WRITE to wait until transactions that use the table to be locked are completed. Updated to the changed name of DEBUG_SYNC point. sql/handler.cc: Global read lock functionality has been moved into a class. sql/lock.cc: Global read lock functionality has been moved into a class. Updated code to use the new MDL API. sql/mdl.cc: Introduced new type-of-operation aware metadata locks. To do this: - Changed MDL_lock to use one list for waiting requests and one list for granted requests. For each list, added a bitmap that holds information what lock types a list contains. Added a helper class MDL_lock::List to manipulate with granted and waited lists while keeping the bitmaps in sync with list contents. - Changed lock-compatibility functions to use bitmaps that define compatibility. - Introduced a graph based deadlock detector inspired by waiting_threads.c from Maria implementation. - Now that we have a deadlock detector, and no longer have a global lock to protect individual lock objects, but rather use an rw lock per object, removed redundant code for upgrade, and the global read lock. Changed the MDL API to no longer require the caller to acquire the global intention exclusive lock by means of a separate method. Removed a few more methods that became redundant. - Removed deadlock detection heuristic, it has been made obsolete by the deadlock detector. - With operation-type-aware metadata locks, MDL subsystem has become aware of potential conflicts between DDL and open transactions. This made it possible to remove calls to mysql_abort_transactions_with_shared_lock() from acquisition paths for exclusive lock and lock upgrade. Now we can simply wait for these transactions to complete without fear of deadlock. Function mysql_lock_abort() has also become unnecessary for all conflicting cases except when a DDL conflicts with a connection that has an open HANDLER. sql/mdl.h: Introduced new type-of-operation aware metadata locks. Introduced a graph based deadlock detector and supporting methods. Added comments. God rid of redundant API calls. Renamed m_lt_or_ha_sentinel to m_trans_sentinel, since now it guards the global read lock as well as LOCK TABLES and HANDLER locks. sql/mysql_priv.h: Moved the global read lock functionality into a class. Added MYSQL_OPEN_FORCE_SHARED_MDL flag which forces open_tables() to take MDL_SHARED on tables instead of metadata locks specified in the parser. We use this to allow PREPARE run concurrently in presence of LOCK TABLES ... WRITE. Added signature for find_table_for_mdl_ugprade(). sql/set_var.cc: Global read lock functionality has been moved into a class. sql/sp_head.cc: When creating TABLE_LIST elements for prelocking or system tables set the type of request for metadata lock according to the operation that will be performed on the table. sql/sql_base.cc: - Updated code to use the new MDL API. - In order to avoid locks starvation we take upgradable locks all at once. As result implicitly locked tables no longer get an upgradable lock. Consequently DDL and FLUSH TABLES for such tables is prohibited. find_write_locked_table() was replaced by find_table_for_mdl_upgrade() function. open_table() was adjusted to return TABLE instance with upgradable ticket when necessary. - We no longer wait for all locks on OT_WAIT back off action -- only on the lock that caused the wait conflict. Moreover, now we distinguish cases when we have to wait due to conflict in MDL and old version of table in TDC. - Upate mysql_notify_threads_having_share_locks() to only abort thr_lock.c waits of threads that have open HANDLERs, since lock conflicts with only these threads now can lead to deadlocks not detectable by the MDL deadlock detector. - Remove mysql_abort_transactions_with_shared_locks() which is no longer needed. sql/sql_class.cc: Global read lock functionality has been moved into a class. Re-arranged code in THD::cleanup() to simplify assert. sql/sql_class.h: Introduced class to incapsulate global read lock functionality. Now sentinel in MDL subsystem guards the global read lock as well as LOCK TABLES and HANDLER locks. Adjusted code accordingly. sql/sql_db.cc: Global read lock functionality has been moved into a class. sql/sql_delete.cc: We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result TRUNCATE TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_handler.cc: Inform MDL_context about presence of open HANDLERs. Since HANLDERs break MDL protocol by acquiring table-level lock while holding only S metadata lock on a table MDL subsystem should take special care about such contexts (Now this is the only case when mysql_lock_abort() is used). sql/sql_parse.cc: Global read lock functionality has been moved into a class. Do not take upgradable metadata locks when opening tables for CREATE TABLE SELECT as it is not necessary and limits concurrency. When initializing TABLE_LIST objects before adding them to the table list set the type of request for metadata lock according to the operation that will be performed on the table. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result FLUSH TABLES is no longer allowed for such tables. sql/sql_prepare.cc: Use MYSQL_OPEN_FORCE_SHARED_MDL flag when opening tables during PREPARE. This allows PREPARE to run concurrently in presence of LOCK TABLES ... WRITE. sql/sql_rename.cc: Global read lock functionality has been moved into a class. sql/sql_show.cc: Updated code to use the new MDL API. sql/sql_table.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result DROP TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_trigger.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result CREATE/DROP TRIGGER is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_view.cc: Global read lock functionality has been moved into a class. Fixed results of wrong merge that led to misuse of GLR API. CREATE VIEW statement is not a commit statement. sql/table.cc: When resetting TABLE_LIST objects for PS or SP re-execution set the type of request for metadata lock according to the operation that will be performed on the table. Do the same in auxiliary function initializing metadata lock requests in a table list. sql/table.h: When initializing TABLE_LIST objects set the type of request for metadata lock according to the operation that will be performed on the table. sql/transaction.cc: Global read lock functionality has been moved into a class.
2010-02-01 12:43:06 +01:00
connection locker2;
select get_lock("mysqltest_lock", 100);
connection locker;
send
select n from t1 where get_lock('mysqltest_lock', 100);
Updated manual about embedded version. Speed up column-completion in 'mysql' Don't use ISAM if HAVE_ISAM is not defined A lot of fixes for the embedded version. All libraries are now included in libmysqld.a Changed arguments to convert_dirname() to make it more general. Renamed files in the 'merge' directory to all use a common prefix. Don't compile both assembler and C functions on x86 BitKeeper/deleted/.del-mf_pack2.c~f07795bbcf57be7: Delete: mysys/mf_pack2.c Docs/manual.texi: Updated chapter about embedded version acinclude.m4: Fix for using BDB and InnoDB with embedded client/completion_hash.cc: Speed up memory allocation client/completion_hash.h: Speed up memory allocation client/mysql.cc: Speed up memory allocation client/mysqldump.c: Fix to use now convert_dirname client/mysqltest.c: Fixed memory allocation bugs. Added --basedir=#, --compress=#, --server-arg, --server-file. Fixes for embedded version Changed silent mode to -s instead of -q include/my_global.h: Update to use HAVE_ISAM include/my_sys.h: Cleanup of fn_format() include/mysql.h: Prepare FIELD struct for 4.1 include/mysql_embed.h: Don't use ISAM in embedded version innobase/include/srv0srv.h: Make InnoDB startup/shutdown silent in embedded version innobase/log/log0log.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0srv.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0start.c: Make InnoDB startup/shutdown silent in embedded version isam/isamlog.c: new convert_dirname libmysql/libmysql.c: Prepare for 4.1 libmysqld/Makefile.am: Changed to create one libmysqld.a file that includes all other libraries. libmysqld/examples/Makefile.am: Changed to use new libmysqld.a libmysqld/examples/test-run: Lot's of fixes to get new mysqltest with embedded MySQL to work. libmysqld/lib_sql.cc: Changed type of arguments for mysql_server_init() to make code more portable. libmysqld/libmysqld.c: Cleanup merge/Makefile.am: Rename to use common prefix. merge/mrg_close.c: Rename to use common prefix. merge/mrg_create.c: Rename to use common prefix. merge/mrg_def.h: Rename to use common prefix. merge/mrg_delete.c: Rename to use common prefix. merge/mrg_extra.c: Rename to use common prefix. merge/mrg_info.c: Rename to use common prefix. merge/mrg_locking.c: Rename to use common prefix. merge/mrg_open.c: Rename to use common prefix. merge/mrg_panic.c: Rename to use common prefix. merge/mrg_rrnd.c: Rename to use common prefix. merge/mrg_rsame.c: Rename to use common prefix. merge/mrg_static.c: Rename to use common prefix. merge/mrg_update.c: Rename to use common prefix. myisam/myisamlog.c: Use new convert_dirname myisammrg/Makefile.am: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_close.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_create.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_def.h: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_delete.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_extra.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_info.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_locking.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_open.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_panic.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_queue.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rfirst.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rkey.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rlast.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rnext.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rprev.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rrnd.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rsame.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_static.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_update.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_write.c: Renamed mymrgdef.h -> myrg_def.h mysql-test/include/master-slave.inc: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/mysql-test-run.sh: cleanup mysql-test/r/auto_increment.result: Cleanup of tests for embedded version mysql-test/r/func_system.result: Cleanup of tests for embedded version mysql-test/r/isam.result: Cleanup of tests for embedded version mysql-test/r/lock.result: Cleanup of tests for embedded version mysql-test/r/show_check.result: Cleanup of tests for embedded version mysql-test/t/auto_increment.test: Cleanup of tests for embedded version mysql-test/t/backup.test: Cleanup of tests for embedded version mysql-test/t/count_distinct2-master.opt: Cleanup of tests for embedded version mysql-test/t/count_distinct2.test: Cleanup of tests for embedded version mysql-test/t/create.test: Cleanup of tests for embedded version mysql-test/t/flush.test: Cleanup of tests for embedded version mysql-test/t/func_system.test: Cleanup of tests for embedded version mysql-test/t/isam.test: Cleanup of tests for embedded version mysql-test/t/kill.test: Cleanup of tests for embedded version mysql-test/t/lock.test: Cleanup of tests for embedded version mysql-test/t/order_fill_sortbuf-master.opt: Cleanup of tests for embedded version mysql-test/t/rpl000015.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000016.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000017.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000018.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/show_check.test: Move ISAM specific tests to isam.test mysql-test/t/status.test: Cleanup of tests for embedded version mysql-test/t/tablelock.test: Cleanup mysys/Makefile.am: Removed not used mf_pack2.c mysys/charset.c: new convert_dirname mysys/default.c: new convert_dirname mysys/mf_dirname.c: Changed convert_dirname() to be more general mysys/mf_format.c: Changed bit flags to fn_format() to defines. Added handling of relative filenames BitKeeper/etc/ignore: Added libmysqld/examples/test-gdbinit scripts/mysql_explain_log to the ignore list mysys/mf_pack.c: new convert_dirname mysys/mf_tempfile.c: new convert_dirname scripts/Makefile.am: Adde mysql_explain_log scripts/mysql_config.sh: Added support of --libmysqld-libs sql/Makefile.am: Fix to use 'innodb_system_libs' sql/ha_innobase.cc: Make InnoDB startup/shutdown silent in embedded version sql/ha_isam.cc: Added handling of HAVE_ISAM sql/ha_isammrg.cc: Added handling of HAVE_ISAM sql/ha_myisam.cc: Handle relative paths; Needed to support BACKUP TABLE in embedded version sql/ha_myisammrg.cc: Rename of filenames sql/handler.cc: Added handling of HAVE_ISAM sql/item_func.cc: Fix for ecc (Intel Compiler) sql/mysql_priv.h: Added global variable 'mysql_embedded' sql/mysqld.cc: Use HAVE_ISAM sql/sql_parse.cc: Use new convert_dirname sql/sql_select.cc: Fix for ecc (Intel Compiler) sql/sql_table.cc: Added handling of relative filenames strings/Makefile.am: Don't compile both assembler and C functions on x86 strings/bchange.c: cleanup
2001-10-08 03:58:07 +02:00
connection writer;
Implement new type-of-operation-aware metadata locks. Add a wait-for graph based deadlock detector to the MDL subsystem. Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and bug #37346 "innodb does not detect deadlock between update and alter table". The first bug manifested itself as an unwarranted abort of a transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER statement, when this transaction tried to repeat use of a table, which it has already used in a similar fashion before ALTER started. The second bug showed up as a deadlock between table-level locks and InnoDB row locks, which was "detected" only after innodb_lock_wait_timeout timeout. A transaction would start using the table and modify a few rows. Then ALTER TABLE would come in, and start copying rows into a temporary table. Eventually it would stumble on the modified records and get blocked on a row lock. The first transaction would try to do more updates, and get blocked on thr_lock.c lock. This situation of circular wait would only get resolved by a timeout. Both these bugs stemmed from inadequate solutions to the problem of deadlocks occurring between different locking subsystems. In the first case we tried to avoid deadlocks between metadata locking and table-level locking subsystems, when upgrading shared metadata lock to exclusive one. Transactions holding the shared lock on the table and waiting for some table-level lock used to be aborted too aggressively. We also allowed ALTER TABLE to start in presence of transactions that modify the subject table. ALTER TABLE acquires TL_WRITE_ALLOW_READ lock at start, and that block all writes against the table (naturally, we don't want any writes to be lost when switching the old and the new table). TL_WRITE_ALLOW_READ lock, in turn, would block the started transaction on thr_lock.c lock, should they do more updates. This, again, lead to the need to abort such transactions. The second bug occurred simply because we didn't have any mechanism to detect deadlocks between the table-level locks in thr_lock.c and row-level locks in InnoDB, other than innodb_lock_wait_timeout. This patch solves both these problems by moving lock conflicts which are causing these deadlocks into the metadata locking subsystem, thus making it possible to avoid or detect such deadlocks inside MDL. To do this we introduce new type-of-operation-aware metadata locks, which allow MDL subsystem to know not only the fact that transaction has used or is going to use some object but also what kind of operation it has carried out or going to carry out on the object. This, along with the addition of a special kind of upgradable metadata lock, allows ALTER TABLE to wait until all transactions which has updated the table to go away. This solves the second issue. Another special type of upgradable metadata lock is acquired by LOCK TABLE WRITE. This second lock type allows to solve the first issue, since abortion of table-level locks in event of DDL under LOCK TABLES becomes also unnecessary. Below follows the list of incompatible changes introduced by this patch: - From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those statements that acquire TL_WRITE_ALLOW_READ lock) wait for all transactions which has *updated* the table to complete. - From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE (i.e. all statements which acquire TL_WRITE table-level lock) wait for all transaction which *updated or read* from the table to complete. As a consequence, innodb_table_locks=0 option no longer applies to LOCK TABLES ... WRITE. - DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort statements or transactions which use tables being dropped or renamed, and instead wait for these transactions to complete. - Since LOCK TABLES WRITE now takes a special metadata lock, not compatible with with reads or writes against the subject table and transaction-wide, thr_lock.c deadlock avoidance algorithm that used to ensure absence of deadlocks between LOCK TABLES WRITE and other statements is no longer sufficient, even for MyISAM. The wait-for graph based deadlock detector of MDL subsystem may sometimes be necessary and is involved. This may lead to ER_LOCK_DEADLOCK error produced for multi-statement transactions even if these only use MyISAM: session 1: session 2: begin; update t1 ... lock table t2 write, t1 write; -- gets a lock on t2, blocks on t1 update t2 ... (ER_LOCK_DEADLOCK) - Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE was abandoned. LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same priority as the usual LOCK TABLE ... WRITE. SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in the wait queue. - We do not take upgradable metadata locks on implicitly locked tables. So if one has, say, a view v1 that uses table t1, and issues: LOCK TABLE v1 WRITE; FLUSH TABLE t1; -- (or just 'FLUSH TABLES'), an error is produced. In order to be able to perform DDL on a table under LOCK TABLES, the table must be locked explicitly in the LOCK TABLES list. mysql-test/include/handler.inc: Adjusted test case to trigger an execution path on which bug 41110 "crash with handler command when used concurrently with alter table" and bug 41112 "crash in mysql_ha_close_table/get_lock_data with alter table" were originally discovered. Left old test case which no longer triggers this execution path for the sake of coverage. Added test coverage for HANDLER SQL statements and type-aware metadata locks. Added a test for the global shared lock and HANDLER SQL. Updated tests to take into account that the old simple deadlock detection heuristics was replaced with a graph-based deadlock detector. mysql-test/r/debug_sync.result: Updated results (see debug_sync.test). mysql-test/r/handler_innodb.result: Updated results (see handler.inc test). mysql-test/r/handler_myisam.result: Updated results (see handler.inc test). mysql-test/r/innodb-lock.result: Updated results (see innodb-lock.test). mysql-test/r/innodb_mysql_lock.result: Updated results (see innodb_mysql_lock.test). mysql-test/r/lock.result: Updated results (see lock.test). mysql-test/r/lock_multi.result: Updated results (see lock_multi.test). mysql-test/r/lock_sync.result: Updated results (see lock_sync.test). mysql-test/r/mdl_sync.result: Updated results (see mdl_sync.test). mysql-test/r/sp-threads.result: SHOW PROCESSLIST output has changed due to the fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/r/truncate_coverage.result: Updated results (see truncate_coverage.test). mysql-test/suite/funcs_1/datadict/processlist_val.inc: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/funcs_1/r/processlist_val_no_prot.result: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/rpl/t/rpl_sp.test: Updated to a new SHOW PROCESSLIST state name. mysql-test/t/debug_sync.test: Use LOCK TABLES READ instead of LOCK TABLES WRITE as the latter no longer allows to trigger execution path involving waiting on thr_lock.c lock and therefore reaching debug sync-point covered by this test. mysql-test/t/innodb-lock.test: Adjusted test case to the fact that innodb_table_locks=0 option is no longer supported, since LOCK TABLES WRITE handles all its conflicts within MDL subsystem. mysql-test/t/innodb_mysql_lock.test: Added test for bug #37346 "innodb does not detect deadlock between update and alter table". mysql-test/t/lock.test: Added test coverage which checks the fact that we no longer support DDL under LOCK TABLES on tables which were locked implicitly. Adjusted existing test cases accordingly. mysql-test/t/lock_multi.test: Added test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock". Adjusted other test cases to take into account the fact that waiting for LOCK TABLES ... WRITE now happens within MDL subsystem. mysql-test/t/lock_sync.test: Since LOCK TABLES ... WRITE now takes SNRW metadata lock for tables locked explicitly we have to implicitly lock InnoDB tables (through view) to trigger the table-level lock conflict between TL_WRITE and TL_WRITE_ALLOW_WRITE. mysql-test/t/mdl_sync.test: Added basic test coverage for type-of-operation-aware metadata locks. Also covered with tests some use cases involving HANDLER statements in which a deadlock could arise. Adjusted existing tests to take type-of-operation-aware MDL into account. mysql-test/t/multi_update.test: Update to a new SHOW PROCESSLIST state name. mysql-test/t/truncate_coverage.test: Adjusted test case after making LOCK TABLES WRITE to wait until transactions that use the table to be locked are completed. Updated to the changed name of DEBUG_SYNC point. sql/handler.cc: Global read lock functionality has been moved into a class. sql/lock.cc: Global read lock functionality has been moved into a class. Updated code to use the new MDL API. sql/mdl.cc: Introduced new type-of-operation aware metadata locks. To do this: - Changed MDL_lock to use one list for waiting requests and one list for granted requests. For each list, added a bitmap that holds information what lock types a list contains. Added a helper class MDL_lock::List to manipulate with granted and waited lists while keeping the bitmaps in sync with list contents. - Changed lock-compatibility functions to use bitmaps that define compatibility. - Introduced a graph based deadlock detector inspired by waiting_threads.c from Maria implementation. - Now that we have a deadlock detector, and no longer have a global lock to protect individual lock objects, but rather use an rw lock per object, removed redundant code for upgrade, and the global read lock. Changed the MDL API to no longer require the caller to acquire the global intention exclusive lock by means of a separate method. Removed a few more methods that became redundant. - Removed deadlock detection heuristic, it has been made obsolete by the deadlock detector. - With operation-type-aware metadata locks, MDL subsystem has become aware of potential conflicts between DDL and open transactions. This made it possible to remove calls to mysql_abort_transactions_with_shared_lock() from acquisition paths for exclusive lock and lock upgrade. Now we can simply wait for these transactions to complete without fear of deadlock. Function mysql_lock_abort() has also become unnecessary for all conflicting cases except when a DDL conflicts with a connection that has an open HANDLER. sql/mdl.h: Introduced new type-of-operation aware metadata locks. Introduced a graph based deadlock detector and supporting methods. Added comments. God rid of redundant API calls. Renamed m_lt_or_ha_sentinel to m_trans_sentinel, since now it guards the global read lock as well as LOCK TABLES and HANDLER locks. sql/mysql_priv.h: Moved the global read lock functionality into a class. Added MYSQL_OPEN_FORCE_SHARED_MDL flag which forces open_tables() to take MDL_SHARED on tables instead of metadata locks specified in the parser. We use this to allow PREPARE run concurrently in presence of LOCK TABLES ... WRITE. Added signature for find_table_for_mdl_ugprade(). sql/set_var.cc: Global read lock functionality has been moved into a class. sql/sp_head.cc: When creating TABLE_LIST elements for prelocking or system tables set the type of request for metadata lock according to the operation that will be performed on the table. sql/sql_base.cc: - Updated code to use the new MDL API. - In order to avoid locks starvation we take upgradable locks all at once. As result implicitly locked tables no longer get an upgradable lock. Consequently DDL and FLUSH TABLES for such tables is prohibited. find_write_locked_table() was replaced by find_table_for_mdl_upgrade() function. open_table() was adjusted to return TABLE instance with upgradable ticket when necessary. - We no longer wait for all locks on OT_WAIT back off action -- only on the lock that caused the wait conflict. Moreover, now we distinguish cases when we have to wait due to conflict in MDL and old version of table in TDC. - Upate mysql_notify_threads_having_share_locks() to only abort thr_lock.c waits of threads that have open HANDLERs, since lock conflicts with only these threads now can lead to deadlocks not detectable by the MDL deadlock detector. - Remove mysql_abort_transactions_with_shared_locks() which is no longer needed. sql/sql_class.cc: Global read lock functionality has been moved into a class. Re-arranged code in THD::cleanup() to simplify assert. sql/sql_class.h: Introduced class to incapsulate global read lock functionality. Now sentinel in MDL subsystem guards the global read lock as well as LOCK TABLES and HANDLER locks. Adjusted code accordingly. sql/sql_db.cc: Global read lock functionality has been moved into a class. sql/sql_delete.cc: We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result TRUNCATE TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_handler.cc: Inform MDL_context about presence of open HANDLERs. Since HANLDERs break MDL protocol by acquiring table-level lock while holding only S metadata lock on a table MDL subsystem should take special care about such contexts (Now this is the only case when mysql_lock_abort() is used). sql/sql_parse.cc: Global read lock functionality has been moved into a class. Do not take upgradable metadata locks when opening tables for CREATE TABLE SELECT as it is not necessary and limits concurrency. When initializing TABLE_LIST objects before adding them to the table list set the type of request for metadata lock according to the operation that will be performed on the table. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result FLUSH TABLES is no longer allowed for such tables. sql/sql_prepare.cc: Use MYSQL_OPEN_FORCE_SHARED_MDL flag when opening tables during PREPARE. This allows PREPARE to run concurrently in presence of LOCK TABLES ... WRITE. sql/sql_rename.cc: Global read lock functionality has been moved into a class. sql/sql_show.cc: Updated code to use the new MDL API. sql/sql_table.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result DROP TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_trigger.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result CREATE/DROP TRIGGER is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_view.cc: Global read lock functionality has been moved into a class. Fixed results of wrong merge that led to misuse of GLR API. CREATE VIEW statement is not a commit statement. sql/table.cc: When resetting TABLE_LIST objects for PS or SP re-execution set the type of request for metadata lock according to the operation that will be performed on the table. Do the same in auxiliary function initializing metadata lock requests in a table list. sql/table.h: When initializing TABLE_LIST objects set the type of request for metadata lock according to the operation that will be performed on the table. sql/transaction.cc: Global read lock functionality has been moved into a class.
2010-02-01 12:43:06 +01:00
# Wait till above select gets blocked on a user lock.
let $wait_condition=
select count(*) = 1 from information_schema.processlist
where state = "User lock" and info = "select n from t1 where get_lock('mysqltest_lock', 100)";
--source include/wait_condition.inc
send
update low_priority t1 set n = 4;
Updated manual about embedded version. Speed up column-completion in 'mysql' Don't use ISAM if HAVE_ISAM is not defined A lot of fixes for the embedded version. All libraries are now included in libmysqld.a Changed arguments to convert_dirname() to make it more general. Renamed files in the 'merge' directory to all use a common prefix. Don't compile both assembler and C functions on x86 BitKeeper/deleted/.del-mf_pack2.c~f07795bbcf57be7: Delete: mysys/mf_pack2.c Docs/manual.texi: Updated chapter about embedded version acinclude.m4: Fix for using BDB and InnoDB with embedded client/completion_hash.cc: Speed up memory allocation client/completion_hash.h: Speed up memory allocation client/mysql.cc: Speed up memory allocation client/mysqldump.c: Fix to use now convert_dirname client/mysqltest.c: Fixed memory allocation bugs. Added --basedir=#, --compress=#, --server-arg, --server-file. Fixes for embedded version Changed silent mode to -s instead of -q include/my_global.h: Update to use HAVE_ISAM include/my_sys.h: Cleanup of fn_format() include/mysql.h: Prepare FIELD struct for 4.1 include/mysql_embed.h: Don't use ISAM in embedded version innobase/include/srv0srv.h: Make InnoDB startup/shutdown silent in embedded version innobase/log/log0log.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0srv.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0start.c: Make InnoDB startup/shutdown silent in embedded version isam/isamlog.c: new convert_dirname libmysql/libmysql.c: Prepare for 4.1 libmysqld/Makefile.am: Changed to create one libmysqld.a file that includes all other libraries. libmysqld/examples/Makefile.am: Changed to use new libmysqld.a libmysqld/examples/test-run: Lot's of fixes to get new mysqltest with embedded MySQL to work. libmysqld/lib_sql.cc: Changed type of arguments for mysql_server_init() to make code more portable. libmysqld/libmysqld.c: Cleanup merge/Makefile.am: Rename to use common prefix. merge/mrg_close.c: Rename to use common prefix. merge/mrg_create.c: Rename to use common prefix. merge/mrg_def.h: Rename to use common prefix. merge/mrg_delete.c: Rename to use common prefix. merge/mrg_extra.c: Rename to use common prefix. merge/mrg_info.c: Rename to use common prefix. merge/mrg_locking.c: Rename to use common prefix. merge/mrg_open.c: Rename to use common prefix. merge/mrg_panic.c: Rename to use common prefix. merge/mrg_rrnd.c: Rename to use common prefix. merge/mrg_rsame.c: Rename to use common prefix. merge/mrg_static.c: Rename to use common prefix. merge/mrg_update.c: Rename to use common prefix. myisam/myisamlog.c: Use new convert_dirname myisammrg/Makefile.am: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_close.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_create.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_def.h: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_delete.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_extra.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_info.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_locking.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_open.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_panic.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_queue.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rfirst.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rkey.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rlast.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rnext.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rprev.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rrnd.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rsame.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_static.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_update.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_write.c: Renamed mymrgdef.h -> myrg_def.h mysql-test/include/master-slave.inc: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/mysql-test-run.sh: cleanup mysql-test/r/auto_increment.result: Cleanup of tests for embedded version mysql-test/r/func_system.result: Cleanup of tests for embedded version mysql-test/r/isam.result: Cleanup of tests for embedded version mysql-test/r/lock.result: Cleanup of tests for embedded version mysql-test/r/show_check.result: Cleanup of tests for embedded version mysql-test/t/auto_increment.test: Cleanup of tests for embedded version mysql-test/t/backup.test: Cleanup of tests for embedded version mysql-test/t/count_distinct2-master.opt: Cleanup of tests for embedded version mysql-test/t/count_distinct2.test: Cleanup of tests for embedded version mysql-test/t/create.test: Cleanup of tests for embedded version mysql-test/t/flush.test: Cleanup of tests for embedded version mysql-test/t/func_system.test: Cleanup of tests for embedded version mysql-test/t/isam.test: Cleanup of tests for embedded version mysql-test/t/kill.test: Cleanup of tests for embedded version mysql-test/t/lock.test: Cleanup of tests for embedded version mysql-test/t/order_fill_sortbuf-master.opt: Cleanup of tests for embedded version mysql-test/t/rpl000015.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000016.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000017.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000018.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/show_check.test: Move ISAM specific tests to isam.test mysql-test/t/status.test: Cleanup of tests for embedded version mysql-test/t/tablelock.test: Cleanup mysys/Makefile.am: Removed not used mf_pack2.c mysys/charset.c: new convert_dirname mysys/default.c: new convert_dirname mysys/mf_dirname.c: Changed convert_dirname() to be more general mysys/mf_format.c: Changed bit flags to fn_format() to defines. Added handling of relative filenames BitKeeper/etc/ignore: Added libmysqld/examples/test-gdbinit scripts/mysql_explain_log to the ignore list mysys/mf_pack.c: new convert_dirname mysys/mf_tempfile.c: new convert_dirname scripts/Makefile.am: Adde mysql_explain_log scripts/mysql_config.sh: Added support of --libmysqld-libs sql/Makefile.am: Fix to use 'innodb_system_libs' sql/ha_innobase.cc: Make InnoDB startup/shutdown silent in embedded version sql/ha_isam.cc: Added handling of HAVE_ISAM sql/ha_isammrg.cc: Added handling of HAVE_ISAM sql/ha_myisam.cc: Handle relative paths; Needed to support BACKUP TABLE in embedded version sql/ha_myisammrg.cc: Rename of filenames sql/handler.cc: Added handling of HAVE_ISAM sql/item_func.cc: Fix for ecc (Intel Compiler) sql/mysql_priv.h: Added global variable 'mysql_embedded' sql/mysqld.cc: Use HAVE_ISAM sql/sql_parse.cc: Use new convert_dirname sql/sql_select.cc: Fix for ecc (Intel Compiler) sql/sql_table.cc: Added handling of relative filenames strings/Makefile.am: Don't compile both assembler and C functions on x86 strings/bchange.c: cleanup
2001-10-08 03:58:07 +02:00
connection reader;
# Sleep a bit till the update of connection writer is in work and hangs
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
where state = "Waiting for table level lock" and
info = "update low_priority t1 set n = 4";
--source include/wait_condition.inc
select n from t1;
Implement new type-of-operation-aware metadata locks. Add a wait-for graph based deadlock detector to the MDL subsystem. Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and bug #37346 "innodb does not detect deadlock between update and alter table". The first bug manifested itself as an unwarranted abort of a transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER statement, when this transaction tried to repeat use of a table, which it has already used in a similar fashion before ALTER started. The second bug showed up as a deadlock between table-level locks and InnoDB row locks, which was "detected" only after innodb_lock_wait_timeout timeout. A transaction would start using the table and modify a few rows. Then ALTER TABLE would come in, and start copying rows into a temporary table. Eventually it would stumble on the modified records and get blocked on a row lock. The first transaction would try to do more updates, and get blocked on thr_lock.c lock. This situation of circular wait would only get resolved by a timeout. Both these bugs stemmed from inadequate solutions to the problem of deadlocks occurring between different locking subsystems. In the first case we tried to avoid deadlocks between metadata locking and table-level locking subsystems, when upgrading shared metadata lock to exclusive one. Transactions holding the shared lock on the table and waiting for some table-level lock used to be aborted too aggressively. We also allowed ALTER TABLE to start in presence of transactions that modify the subject table. ALTER TABLE acquires TL_WRITE_ALLOW_READ lock at start, and that block all writes against the table (naturally, we don't want any writes to be lost when switching the old and the new table). TL_WRITE_ALLOW_READ lock, in turn, would block the started transaction on thr_lock.c lock, should they do more updates. This, again, lead to the need to abort such transactions. The second bug occurred simply because we didn't have any mechanism to detect deadlocks between the table-level locks in thr_lock.c and row-level locks in InnoDB, other than innodb_lock_wait_timeout. This patch solves both these problems by moving lock conflicts which are causing these deadlocks into the metadata locking subsystem, thus making it possible to avoid or detect such deadlocks inside MDL. To do this we introduce new type-of-operation-aware metadata locks, which allow MDL subsystem to know not only the fact that transaction has used or is going to use some object but also what kind of operation it has carried out or going to carry out on the object. This, along with the addition of a special kind of upgradable metadata lock, allows ALTER TABLE to wait until all transactions which has updated the table to go away. This solves the second issue. Another special type of upgradable metadata lock is acquired by LOCK TABLE WRITE. This second lock type allows to solve the first issue, since abortion of table-level locks in event of DDL under LOCK TABLES becomes also unnecessary. Below follows the list of incompatible changes introduced by this patch: - From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those statements that acquire TL_WRITE_ALLOW_READ lock) wait for all transactions which has *updated* the table to complete. - From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE (i.e. all statements which acquire TL_WRITE table-level lock) wait for all transaction which *updated or read* from the table to complete. As a consequence, innodb_table_locks=0 option no longer applies to LOCK TABLES ... WRITE. - DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort statements or transactions which use tables being dropped or renamed, and instead wait for these transactions to complete. - Since LOCK TABLES WRITE now takes a special metadata lock, not compatible with with reads or writes against the subject table and transaction-wide, thr_lock.c deadlock avoidance algorithm that used to ensure absence of deadlocks between LOCK TABLES WRITE and other statements is no longer sufficient, even for MyISAM. The wait-for graph based deadlock detector of MDL subsystem may sometimes be necessary and is involved. This may lead to ER_LOCK_DEADLOCK error produced for multi-statement transactions even if these only use MyISAM: session 1: session 2: begin; update t1 ... lock table t2 write, t1 write; -- gets a lock on t2, blocks on t1 update t2 ... (ER_LOCK_DEADLOCK) - Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE was abandoned. LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same priority as the usual LOCK TABLE ... WRITE. SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in the wait queue. - We do not take upgradable metadata locks on implicitly locked tables. So if one has, say, a view v1 that uses table t1, and issues: LOCK TABLE v1 WRITE; FLUSH TABLE t1; -- (or just 'FLUSH TABLES'), an error is produced. In order to be able to perform DDL on a table under LOCK TABLES, the table must be locked explicitly in the LOCK TABLES list. mysql-test/include/handler.inc: Adjusted test case to trigger an execution path on which bug 41110 "crash with handler command when used concurrently with alter table" and bug 41112 "crash in mysql_ha_close_table/get_lock_data with alter table" were originally discovered. Left old test case which no longer triggers this execution path for the sake of coverage. Added test coverage for HANDLER SQL statements and type-aware metadata locks. Added a test for the global shared lock and HANDLER SQL. Updated tests to take into account that the old simple deadlock detection heuristics was replaced with a graph-based deadlock detector. mysql-test/r/debug_sync.result: Updated results (see debug_sync.test). mysql-test/r/handler_innodb.result: Updated results (see handler.inc test). mysql-test/r/handler_myisam.result: Updated results (see handler.inc test). mysql-test/r/innodb-lock.result: Updated results (see innodb-lock.test). mysql-test/r/innodb_mysql_lock.result: Updated results (see innodb_mysql_lock.test). mysql-test/r/lock.result: Updated results (see lock.test). mysql-test/r/lock_multi.result: Updated results (see lock_multi.test). mysql-test/r/lock_sync.result: Updated results (see lock_sync.test). mysql-test/r/mdl_sync.result: Updated results (see mdl_sync.test). mysql-test/r/sp-threads.result: SHOW PROCESSLIST output has changed due to the fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/r/truncate_coverage.result: Updated results (see truncate_coverage.test). mysql-test/suite/funcs_1/datadict/processlist_val.inc: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/funcs_1/r/processlist_val_no_prot.result: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/rpl/t/rpl_sp.test: Updated to a new SHOW PROCESSLIST state name. mysql-test/t/debug_sync.test: Use LOCK TABLES READ instead of LOCK TABLES WRITE as the latter no longer allows to trigger execution path involving waiting on thr_lock.c lock and therefore reaching debug sync-point covered by this test. mysql-test/t/innodb-lock.test: Adjusted test case to the fact that innodb_table_locks=0 option is no longer supported, since LOCK TABLES WRITE handles all its conflicts within MDL subsystem. mysql-test/t/innodb_mysql_lock.test: Added test for bug #37346 "innodb does not detect deadlock between update and alter table". mysql-test/t/lock.test: Added test coverage which checks the fact that we no longer support DDL under LOCK TABLES on tables which were locked implicitly. Adjusted existing test cases accordingly. mysql-test/t/lock_multi.test: Added test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock". Adjusted other test cases to take into account the fact that waiting for LOCK TABLES ... WRITE now happens within MDL subsystem. mysql-test/t/lock_sync.test: Since LOCK TABLES ... WRITE now takes SNRW metadata lock for tables locked explicitly we have to implicitly lock InnoDB tables (through view) to trigger the table-level lock conflict between TL_WRITE and TL_WRITE_ALLOW_WRITE. mysql-test/t/mdl_sync.test: Added basic test coverage for type-of-operation-aware metadata locks. Also covered with tests some use cases involving HANDLER statements in which a deadlock could arise. Adjusted existing tests to take type-of-operation-aware MDL into account. mysql-test/t/multi_update.test: Update to a new SHOW PROCESSLIST state name. mysql-test/t/truncate_coverage.test: Adjusted test case after making LOCK TABLES WRITE to wait until transactions that use the table to be locked are completed. Updated to the changed name of DEBUG_SYNC point. sql/handler.cc: Global read lock functionality has been moved into a class. sql/lock.cc: Global read lock functionality has been moved into a class. Updated code to use the new MDL API. sql/mdl.cc: Introduced new type-of-operation aware metadata locks. To do this: - Changed MDL_lock to use one list for waiting requests and one list for granted requests. For each list, added a bitmap that holds information what lock types a list contains. Added a helper class MDL_lock::List to manipulate with granted and waited lists while keeping the bitmaps in sync with list contents. - Changed lock-compatibility functions to use bitmaps that define compatibility. - Introduced a graph based deadlock detector inspired by waiting_threads.c from Maria implementation. - Now that we have a deadlock detector, and no longer have a global lock to protect individual lock objects, but rather use an rw lock per object, removed redundant code for upgrade, and the global read lock. Changed the MDL API to no longer require the caller to acquire the global intention exclusive lock by means of a separate method. Removed a few more methods that became redundant. - Removed deadlock detection heuristic, it has been made obsolete by the deadlock detector. - With operation-type-aware metadata locks, MDL subsystem has become aware of potential conflicts between DDL and open transactions. This made it possible to remove calls to mysql_abort_transactions_with_shared_lock() from acquisition paths for exclusive lock and lock upgrade. Now we can simply wait for these transactions to complete without fear of deadlock. Function mysql_lock_abort() has also become unnecessary for all conflicting cases except when a DDL conflicts with a connection that has an open HANDLER. sql/mdl.h: Introduced new type-of-operation aware metadata locks. Introduced a graph based deadlock detector and supporting methods. Added comments. God rid of redundant API calls. Renamed m_lt_or_ha_sentinel to m_trans_sentinel, since now it guards the global read lock as well as LOCK TABLES and HANDLER locks. sql/mysql_priv.h: Moved the global read lock functionality into a class. Added MYSQL_OPEN_FORCE_SHARED_MDL flag which forces open_tables() to take MDL_SHARED on tables instead of metadata locks specified in the parser. We use this to allow PREPARE run concurrently in presence of LOCK TABLES ... WRITE. Added signature for find_table_for_mdl_ugprade(). sql/set_var.cc: Global read lock functionality has been moved into a class. sql/sp_head.cc: When creating TABLE_LIST elements for prelocking or system tables set the type of request for metadata lock according to the operation that will be performed on the table. sql/sql_base.cc: - Updated code to use the new MDL API. - In order to avoid locks starvation we take upgradable locks all at once. As result implicitly locked tables no longer get an upgradable lock. Consequently DDL and FLUSH TABLES for such tables is prohibited. find_write_locked_table() was replaced by find_table_for_mdl_upgrade() function. open_table() was adjusted to return TABLE instance with upgradable ticket when necessary. - We no longer wait for all locks on OT_WAIT back off action -- only on the lock that caused the wait conflict. Moreover, now we distinguish cases when we have to wait due to conflict in MDL and old version of table in TDC. - Upate mysql_notify_threads_having_share_locks() to only abort thr_lock.c waits of threads that have open HANDLERs, since lock conflicts with only these threads now can lead to deadlocks not detectable by the MDL deadlock detector. - Remove mysql_abort_transactions_with_shared_locks() which is no longer needed. sql/sql_class.cc: Global read lock functionality has been moved into a class. Re-arranged code in THD::cleanup() to simplify assert. sql/sql_class.h: Introduced class to incapsulate global read lock functionality. Now sentinel in MDL subsystem guards the global read lock as well as LOCK TABLES and HANDLER locks. Adjusted code accordingly. sql/sql_db.cc: Global read lock functionality has been moved into a class. sql/sql_delete.cc: We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result TRUNCATE TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_handler.cc: Inform MDL_context about presence of open HANDLERs. Since HANLDERs break MDL protocol by acquiring table-level lock while holding only S metadata lock on a table MDL subsystem should take special care about such contexts (Now this is the only case when mysql_lock_abort() is used). sql/sql_parse.cc: Global read lock functionality has been moved into a class. Do not take upgradable metadata locks when opening tables for CREATE TABLE SELECT as it is not necessary and limits concurrency. When initializing TABLE_LIST objects before adding them to the table list set the type of request for metadata lock according to the operation that will be performed on the table. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result FLUSH TABLES is no longer allowed for such tables. sql/sql_prepare.cc: Use MYSQL_OPEN_FORCE_SHARED_MDL flag when opening tables during PREPARE. This allows PREPARE to run concurrently in presence of LOCK TABLES ... WRITE. sql/sql_rename.cc: Global read lock functionality has been moved into a class. sql/sql_show.cc: Updated code to use the new MDL API. sql/sql_table.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result DROP TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_trigger.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result CREATE/DROP TRIGGER is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_view.cc: Global read lock functionality has been moved into a class. Fixed results of wrong merge that led to misuse of GLR API. CREATE VIEW statement is not a commit statement. sql/table.cc: When resetting TABLE_LIST objects for PS or SP re-execution set the type of request for metadata lock according to the operation that will be performed on the table. Do the same in auxiliary function initializing metadata lock requests in a table list. sql/table.h: When initializing TABLE_LIST objects set the type of request for metadata lock according to the operation that will be performed on the table. sql/transaction.cc: Global read lock functionality has been moved into a class.
2010-02-01 12:43:06 +01:00
connection locker2;
select release_lock("mysqltest_lock");
Updated manual about embedded version. Speed up column-completion in 'mysql' Don't use ISAM if HAVE_ISAM is not defined A lot of fixes for the embedded version. All libraries are now included in libmysqld.a Changed arguments to convert_dirname() to make it more general. Renamed files in the 'merge' directory to all use a common prefix. Don't compile both assembler and C functions on x86 BitKeeper/deleted/.del-mf_pack2.c~f07795bbcf57be7: Delete: mysys/mf_pack2.c Docs/manual.texi: Updated chapter about embedded version acinclude.m4: Fix for using BDB and InnoDB with embedded client/completion_hash.cc: Speed up memory allocation client/completion_hash.h: Speed up memory allocation client/mysql.cc: Speed up memory allocation client/mysqldump.c: Fix to use now convert_dirname client/mysqltest.c: Fixed memory allocation bugs. Added --basedir=#, --compress=#, --server-arg, --server-file. Fixes for embedded version Changed silent mode to -s instead of -q include/my_global.h: Update to use HAVE_ISAM include/my_sys.h: Cleanup of fn_format() include/mysql.h: Prepare FIELD struct for 4.1 include/mysql_embed.h: Don't use ISAM in embedded version innobase/include/srv0srv.h: Make InnoDB startup/shutdown silent in embedded version innobase/log/log0log.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0srv.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0start.c: Make InnoDB startup/shutdown silent in embedded version isam/isamlog.c: new convert_dirname libmysql/libmysql.c: Prepare for 4.1 libmysqld/Makefile.am: Changed to create one libmysqld.a file that includes all other libraries. libmysqld/examples/Makefile.am: Changed to use new libmysqld.a libmysqld/examples/test-run: Lot's of fixes to get new mysqltest with embedded MySQL to work. libmysqld/lib_sql.cc: Changed type of arguments for mysql_server_init() to make code more portable. libmysqld/libmysqld.c: Cleanup merge/Makefile.am: Rename to use common prefix. merge/mrg_close.c: Rename to use common prefix. merge/mrg_create.c: Rename to use common prefix. merge/mrg_def.h: Rename to use common prefix. merge/mrg_delete.c: Rename to use common prefix. merge/mrg_extra.c: Rename to use common prefix. merge/mrg_info.c: Rename to use common prefix. merge/mrg_locking.c: Rename to use common prefix. merge/mrg_open.c: Rename to use common prefix. merge/mrg_panic.c: Rename to use common prefix. merge/mrg_rrnd.c: Rename to use common prefix. merge/mrg_rsame.c: Rename to use common prefix. merge/mrg_static.c: Rename to use common prefix. merge/mrg_update.c: Rename to use common prefix. myisam/myisamlog.c: Use new convert_dirname myisammrg/Makefile.am: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_close.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_create.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_def.h: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_delete.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_extra.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_info.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_locking.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_open.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_panic.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_queue.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rfirst.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rkey.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rlast.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rnext.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rprev.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rrnd.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rsame.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_static.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_update.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_write.c: Renamed mymrgdef.h -> myrg_def.h mysql-test/include/master-slave.inc: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/mysql-test-run.sh: cleanup mysql-test/r/auto_increment.result: Cleanup of tests for embedded version mysql-test/r/func_system.result: Cleanup of tests for embedded version mysql-test/r/isam.result: Cleanup of tests for embedded version mysql-test/r/lock.result: Cleanup of tests for embedded version mysql-test/r/show_check.result: Cleanup of tests for embedded version mysql-test/t/auto_increment.test: Cleanup of tests for embedded version mysql-test/t/backup.test: Cleanup of tests for embedded version mysql-test/t/count_distinct2-master.opt: Cleanup of tests for embedded version mysql-test/t/count_distinct2.test: Cleanup of tests for embedded version mysql-test/t/create.test: Cleanup of tests for embedded version mysql-test/t/flush.test: Cleanup of tests for embedded version mysql-test/t/func_system.test: Cleanup of tests for embedded version mysql-test/t/isam.test: Cleanup of tests for embedded version mysql-test/t/kill.test: Cleanup of tests for embedded version mysql-test/t/lock.test: Cleanup of tests for embedded version mysql-test/t/order_fill_sortbuf-master.opt: Cleanup of tests for embedded version mysql-test/t/rpl000015.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000016.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000017.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000018.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/show_check.test: Move ISAM specific tests to isam.test mysql-test/t/status.test: Cleanup of tests for embedded version mysql-test/t/tablelock.test: Cleanup mysys/Makefile.am: Removed not used mf_pack2.c mysys/charset.c: new convert_dirname mysys/default.c: new convert_dirname mysys/mf_dirname.c: Changed convert_dirname() to be more general mysys/mf_format.c: Changed bit flags to fn_format() to defines. Added handling of relative filenames BitKeeper/etc/ignore: Added libmysqld/examples/test-gdbinit scripts/mysql_explain_log to the ignore list mysys/mf_pack.c: new convert_dirname mysys/mf_tempfile.c: new convert_dirname scripts/Makefile.am: Adde mysql_explain_log scripts/mysql_config.sh: Added support of --libmysqld-libs sql/Makefile.am: Fix to use 'innodb_system_libs' sql/ha_innobase.cc: Make InnoDB startup/shutdown silent in embedded version sql/ha_isam.cc: Added handling of HAVE_ISAM sql/ha_isammrg.cc: Added handling of HAVE_ISAM sql/ha_myisam.cc: Handle relative paths; Needed to support BACKUP TABLE in embedded version sql/ha_myisammrg.cc: Rename of filenames sql/handler.cc: Added handling of HAVE_ISAM sql/item_func.cc: Fix for ecc (Intel Compiler) sql/mysql_priv.h: Added global variable 'mysql_embedded' sql/mysqld.cc: Use HAVE_ISAM sql/sql_parse.cc: Use new convert_dirname sql/sql_select.cc: Fix for ecc (Intel Compiler) sql/sql_table.cc: Added handling of relative filenames strings/Makefile.am: Don't compile both assembler and C functions on x86 strings/bchange.c: cleanup
2001-10-08 03:58:07 +02:00
connection locker;
Implement new type-of-operation-aware metadata locks. Add a wait-for graph based deadlock detector to the MDL subsystem. Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and bug #37346 "innodb does not detect deadlock between update and alter table". The first bug manifested itself as an unwarranted abort of a transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER statement, when this transaction tried to repeat use of a table, which it has already used in a similar fashion before ALTER started. The second bug showed up as a deadlock between table-level locks and InnoDB row locks, which was "detected" only after innodb_lock_wait_timeout timeout. A transaction would start using the table and modify a few rows. Then ALTER TABLE would come in, and start copying rows into a temporary table. Eventually it would stumble on the modified records and get blocked on a row lock. The first transaction would try to do more updates, and get blocked on thr_lock.c lock. This situation of circular wait would only get resolved by a timeout. Both these bugs stemmed from inadequate solutions to the problem of deadlocks occurring between different locking subsystems. In the first case we tried to avoid deadlocks between metadata locking and table-level locking subsystems, when upgrading shared metadata lock to exclusive one. Transactions holding the shared lock on the table and waiting for some table-level lock used to be aborted too aggressively. We also allowed ALTER TABLE to start in presence of transactions that modify the subject table. ALTER TABLE acquires TL_WRITE_ALLOW_READ lock at start, and that block all writes against the table (naturally, we don't want any writes to be lost when switching the old and the new table). TL_WRITE_ALLOW_READ lock, in turn, would block the started transaction on thr_lock.c lock, should they do more updates. This, again, lead to the need to abort such transactions. The second bug occurred simply because we didn't have any mechanism to detect deadlocks between the table-level locks in thr_lock.c and row-level locks in InnoDB, other than innodb_lock_wait_timeout. This patch solves both these problems by moving lock conflicts which are causing these deadlocks into the metadata locking subsystem, thus making it possible to avoid or detect such deadlocks inside MDL. To do this we introduce new type-of-operation-aware metadata locks, which allow MDL subsystem to know not only the fact that transaction has used or is going to use some object but also what kind of operation it has carried out or going to carry out on the object. This, along with the addition of a special kind of upgradable metadata lock, allows ALTER TABLE to wait until all transactions which has updated the table to go away. This solves the second issue. Another special type of upgradable metadata lock is acquired by LOCK TABLE WRITE. This second lock type allows to solve the first issue, since abortion of table-level locks in event of DDL under LOCK TABLES becomes also unnecessary. Below follows the list of incompatible changes introduced by this patch: - From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those statements that acquire TL_WRITE_ALLOW_READ lock) wait for all transactions which has *updated* the table to complete. - From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE (i.e. all statements which acquire TL_WRITE table-level lock) wait for all transaction which *updated or read* from the table to complete. As a consequence, innodb_table_locks=0 option no longer applies to LOCK TABLES ... WRITE. - DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort statements or transactions which use tables being dropped or renamed, and instead wait for these transactions to complete. - Since LOCK TABLES WRITE now takes a special metadata lock, not compatible with with reads or writes against the subject table and transaction-wide, thr_lock.c deadlock avoidance algorithm that used to ensure absence of deadlocks between LOCK TABLES WRITE and other statements is no longer sufficient, even for MyISAM. The wait-for graph based deadlock detector of MDL subsystem may sometimes be necessary and is involved. This may lead to ER_LOCK_DEADLOCK error produced for multi-statement transactions even if these only use MyISAM: session 1: session 2: begin; update t1 ... lock table t2 write, t1 write; -- gets a lock on t2, blocks on t1 update t2 ... (ER_LOCK_DEADLOCK) - Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE was abandoned. LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same priority as the usual LOCK TABLE ... WRITE. SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in the wait queue. - We do not take upgradable metadata locks on implicitly locked tables. So if one has, say, a view v1 that uses table t1, and issues: LOCK TABLE v1 WRITE; FLUSH TABLE t1; -- (or just 'FLUSH TABLES'), an error is produced. In order to be able to perform DDL on a table under LOCK TABLES, the table must be locked explicitly in the LOCK TABLES list. mysql-test/include/handler.inc: Adjusted test case to trigger an execution path on which bug 41110 "crash with handler command when used concurrently with alter table" and bug 41112 "crash in mysql_ha_close_table/get_lock_data with alter table" were originally discovered. Left old test case which no longer triggers this execution path for the sake of coverage. Added test coverage for HANDLER SQL statements and type-aware metadata locks. Added a test for the global shared lock and HANDLER SQL. Updated tests to take into account that the old simple deadlock detection heuristics was replaced with a graph-based deadlock detector. mysql-test/r/debug_sync.result: Updated results (see debug_sync.test). mysql-test/r/handler_innodb.result: Updated results (see handler.inc test). mysql-test/r/handler_myisam.result: Updated results (see handler.inc test). mysql-test/r/innodb-lock.result: Updated results (see innodb-lock.test). mysql-test/r/innodb_mysql_lock.result: Updated results (see innodb_mysql_lock.test). mysql-test/r/lock.result: Updated results (see lock.test). mysql-test/r/lock_multi.result: Updated results (see lock_multi.test). mysql-test/r/lock_sync.result: Updated results (see lock_sync.test). mysql-test/r/mdl_sync.result: Updated results (see mdl_sync.test). mysql-test/r/sp-threads.result: SHOW PROCESSLIST output has changed due to the fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/r/truncate_coverage.result: Updated results (see truncate_coverage.test). mysql-test/suite/funcs_1/datadict/processlist_val.inc: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/funcs_1/r/processlist_val_no_prot.result: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/rpl/t/rpl_sp.test: Updated to a new SHOW PROCESSLIST state name. mysql-test/t/debug_sync.test: Use LOCK TABLES READ instead of LOCK TABLES WRITE as the latter no longer allows to trigger execution path involving waiting on thr_lock.c lock and therefore reaching debug sync-point covered by this test. mysql-test/t/innodb-lock.test: Adjusted test case to the fact that innodb_table_locks=0 option is no longer supported, since LOCK TABLES WRITE handles all its conflicts within MDL subsystem. mysql-test/t/innodb_mysql_lock.test: Added test for bug #37346 "innodb does not detect deadlock between update and alter table". mysql-test/t/lock.test: Added test coverage which checks the fact that we no longer support DDL under LOCK TABLES on tables which were locked implicitly. Adjusted existing test cases accordingly. mysql-test/t/lock_multi.test: Added test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock". Adjusted other test cases to take into account the fact that waiting for LOCK TABLES ... WRITE now happens within MDL subsystem. mysql-test/t/lock_sync.test: Since LOCK TABLES ... WRITE now takes SNRW metadata lock for tables locked explicitly we have to implicitly lock InnoDB tables (through view) to trigger the table-level lock conflict between TL_WRITE and TL_WRITE_ALLOW_WRITE. mysql-test/t/mdl_sync.test: Added basic test coverage for type-of-operation-aware metadata locks. Also covered with tests some use cases involving HANDLER statements in which a deadlock could arise. Adjusted existing tests to take type-of-operation-aware MDL into account. mysql-test/t/multi_update.test: Update to a new SHOW PROCESSLIST state name. mysql-test/t/truncate_coverage.test: Adjusted test case after making LOCK TABLES WRITE to wait until transactions that use the table to be locked are completed. Updated to the changed name of DEBUG_SYNC point. sql/handler.cc: Global read lock functionality has been moved into a class. sql/lock.cc: Global read lock functionality has been moved into a class. Updated code to use the new MDL API. sql/mdl.cc: Introduced new type-of-operation aware metadata locks. To do this: - Changed MDL_lock to use one list for waiting requests and one list for granted requests. For each list, added a bitmap that holds information what lock types a list contains. Added a helper class MDL_lock::List to manipulate with granted and waited lists while keeping the bitmaps in sync with list contents. - Changed lock-compatibility functions to use bitmaps that define compatibility. - Introduced a graph based deadlock detector inspired by waiting_threads.c from Maria implementation. - Now that we have a deadlock detector, and no longer have a global lock to protect individual lock objects, but rather use an rw lock per object, removed redundant code for upgrade, and the global read lock. Changed the MDL API to no longer require the caller to acquire the global intention exclusive lock by means of a separate method. Removed a few more methods that became redundant. - Removed deadlock detection heuristic, it has been made obsolete by the deadlock detector. - With operation-type-aware metadata locks, MDL subsystem has become aware of potential conflicts between DDL and open transactions. This made it possible to remove calls to mysql_abort_transactions_with_shared_lock() from acquisition paths for exclusive lock and lock upgrade. Now we can simply wait for these transactions to complete without fear of deadlock. Function mysql_lock_abort() has also become unnecessary for all conflicting cases except when a DDL conflicts with a connection that has an open HANDLER. sql/mdl.h: Introduced new type-of-operation aware metadata locks. Introduced a graph based deadlock detector and supporting methods. Added comments. God rid of redundant API calls. Renamed m_lt_or_ha_sentinel to m_trans_sentinel, since now it guards the global read lock as well as LOCK TABLES and HANDLER locks. sql/mysql_priv.h: Moved the global read lock functionality into a class. Added MYSQL_OPEN_FORCE_SHARED_MDL flag which forces open_tables() to take MDL_SHARED on tables instead of metadata locks specified in the parser. We use this to allow PREPARE run concurrently in presence of LOCK TABLES ... WRITE. Added signature for find_table_for_mdl_ugprade(). sql/set_var.cc: Global read lock functionality has been moved into a class. sql/sp_head.cc: When creating TABLE_LIST elements for prelocking or system tables set the type of request for metadata lock according to the operation that will be performed on the table. sql/sql_base.cc: - Updated code to use the new MDL API. - In order to avoid locks starvation we take upgradable locks all at once. As result implicitly locked tables no longer get an upgradable lock. Consequently DDL and FLUSH TABLES for such tables is prohibited. find_write_locked_table() was replaced by find_table_for_mdl_upgrade() function. open_table() was adjusted to return TABLE instance with upgradable ticket when necessary. - We no longer wait for all locks on OT_WAIT back off action -- only on the lock that caused the wait conflict. Moreover, now we distinguish cases when we have to wait due to conflict in MDL and old version of table in TDC. - Upate mysql_notify_threads_having_share_locks() to only abort thr_lock.c waits of threads that have open HANDLERs, since lock conflicts with only these threads now can lead to deadlocks not detectable by the MDL deadlock detector. - Remove mysql_abort_transactions_with_shared_locks() which is no longer needed. sql/sql_class.cc: Global read lock functionality has been moved into a class. Re-arranged code in THD::cleanup() to simplify assert. sql/sql_class.h: Introduced class to incapsulate global read lock functionality. Now sentinel in MDL subsystem guards the global read lock as well as LOCK TABLES and HANDLER locks. Adjusted code accordingly. sql/sql_db.cc: Global read lock functionality has been moved into a class. sql/sql_delete.cc: We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result TRUNCATE TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_handler.cc: Inform MDL_context about presence of open HANDLERs. Since HANLDERs break MDL protocol by acquiring table-level lock while holding only S metadata lock on a table MDL subsystem should take special care about such contexts (Now this is the only case when mysql_lock_abort() is used). sql/sql_parse.cc: Global read lock functionality has been moved into a class. Do not take upgradable metadata locks when opening tables for CREATE TABLE SELECT as it is not necessary and limits concurrency. When initializing TABLE_LIST objects before adding them to the table list set the type of request for metadata lock according to the operation that will be performed on the table. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result FLUSH TABLES is no longer allowed for such tables. sql/sql_prepare.cc: Use MYSQL_OPEN_FORCE_SHARED_MDL flag when opening tables during PREPARE. This allows PREPARE to run concurrently in presence of LOCK TABLES ... WRITE. sql/sql_rename.cc: Global read lock functionality has been moved into a class. sql/sql_show.cc: Updated code to use the new MDL API. sql/sql_table.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result DROP TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_trigger.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result CREATE/DROP TRIGGER is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_view.cc: Global read lock functionality has been moved into a class. Fixed results of wrong merge that led to misuse of GLR API. CREATE VIEW statement is not a commit statement. sql/table.cc: When resetting TABLE_LIST objects for PS or SP re-execution set the type of request for metadata lock according to the operation that will be performed on the table. Do the same in auxiliary function initializing metadata lock requests in a table list. sql/table.h: When initializing TABLE_LIST objects set the type of request for metadata lock according to the operation that will be performed on the table. sql/transaction.cc: Global read lock functionality has been moved into a class.
2010-02-01 12:43:06 +01:00
reap;
select release_lock("mysqltest_lock");
Updated manual about embedded version. Speed up column-completion in 'mysql' Don't use ISAM if HAVE_ISAM is not defined A lot of fixes for the embedded version. All libraries are now included in libmysqld.a Changed arguments to convert_dirname() to make it more general. Renamed files in the 'merge' directory to all use a common prefix. Don't compile both assembler and C functions on x86 BitKeeper/deleted/.del-mf_pack2.c~f07795bbcf57be7: Delete: mysys/mf_pack2.c Docs/manual.texi: Updated chapter about embedded version acinclude.m4: Fix for using BDB and InnoDB with embedded client/completion_hash.cc: Speed up memory allocation client/completion_hash.h: Speed up memory allocation client/mysql.cc: Speed up memory allocation client/mysqldump.c: Fix to use now convert_dirname client/mysqltest.c: Fixed memory allocation bugs. Added --basedir=#, --compress=#, --server-arg, --server-file. Fixes for embedded version Changed silent mode to -s instead of -q include/my_global.h: Update to use HAVE_ISAM include/my_sys.h: Cleanup of fn_format() include/mysql.h: Prepare FIELD struct for 4.1 include/mysql_embed.h: Don't use ISAM in embedded version innobase/include/srv0srv.h: Make InnoDB startup/shutdown silent in embedded version innobase/log/log0log.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0srv.c: Make InnoDB startup/shutdown silent in embedded version innobase/srv/srv0start.c: Make InnoDB startup/shutdown silent in embedded version isam/isamlog.c: new convert_dirname libmysql/libmysql.c: Prepare for 4.1 libmysqld/Makefile.am: Changed to create one libmysqld.a file that includes all other libraries. libmysqld/examples/Makefile.am: Changed to use new libmysqld.a libmysqld/examples/test-run: Lot's of fixes to get new mysqltest with embedded MySQL to work. libmysqld/lib_sql.cc: Changed type of arguments for mysql_server_init() to make code more portable. libmysqld/libmysqld.c: Cleanup merge/Makefile.am: Rename to use common prefix. merge/mrg_close.c: Rename to use common prefix. merge/mrg_create.c: Rename to use common prefix. merge/mrg_def.h: Rename to use common prefix. merge/mrg_delete.c: Rename to use common prefix. merge/mrg_extra.c: Rename to use common prefix. merge/mrg_info.c: Rename to use common prefix. merge/mrg_locking.c: Rename to use common prefix. merge/mrg_open.c: Rename to use common prefix. merge/mrg_panic.c: Rename to use common prefix. merge/mrg_rrnd.c: Rename to use common prefix. merge/mrg_rsame.c: Rename to use common prefix. merge/mrg_static.c: Rename to use common prefix. merge/mrg_update.c: Rename to use common prefix. myisam/myisamlog.c: Use new convert_dirname myisammrg/Makefile.am: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_close.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_create.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_def.h: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_delete.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_extra.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_info.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_locking.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_open.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_panic.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_queue.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rfirst.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rkey.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rlast.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rnext.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rprev.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rrnd.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_rsame.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_static.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_update.c: Renamed mymrgdef.h -> myrg_def.h myisammrg/myrg_write.c: Renamed mymrgdef.h -> myrg_def.h mysql-test/include/master-slave.inc: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/mysql-test-run.sh: cleanup mysql-test/r/auto_increment.result: Cleanup of tests for embedded version mysql-test/r/func_system.result: Cleanup of tests for embedded version mysql-test/r/isam.result: Cleanup of tests for embedded version mysql-test/r/lock.result: Cleanup of tests for embedded version mysql-test/r/show_check.result: Cleanup of tests for embedded version mysql-test/t/auto_increment.test: Cleanup of tests for embedded version mysql-test/t/backup.test: Cleanup of tests for embedded version mysql-test/t/count_distinct2-master.opt: Cleanup of tests for embedded version mysql-test/t/count_distinct2.test: Cleanup of tests for embedded version mysql-test/t/create.test: Cleanup of tests for embedded version mysql-test/t/flush.test: Cleanup of tests for embedded version mysql-test/t/func_system.test: Cleanup of tests for embedded version mysql-test/t/isam.test: Cleanup of tests for embedded version mysql-test/t/kill.test: Cleanup of tests for embedded version mysql-test/t/lock.test: Cleanup of tests for embedded version mysql-test/t/order_fill_sortbuf-master.opt: Cleanup of tests for embedded version mysql-test/t/rpl000015.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000016.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000017.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/rpl000018.test: Use short filenames for sockets (portability problem on Mac OS X) mysql-test/t/show_check.test: Move ISAM specific tests to isam.test mysql-test/t/status.test: Cleanup of tests for embedded version mysql-test/t/tablelock.test: Cleanup mysys/Makefile.am: Removed not used mf_pack2.c mysys/charset.c: new convert_dirname mysys/default.c: new convert_dirname mysys/mf_dirname.c: Changed convert_dirname() to be more general mysys/mf_format.c: Changed bit flags to fn_format() to defines. Added handling of relative filenames BitKeeper/etc/ignore: Added libmysqld/examples/test-gdbinit scripts/mysql_explain_log to the ignore list mysys/mf_pack.c: new convert_dirname mysys/mf_tempfile.c: new convert_dirname scripts/Makefile.am: Adde mysql_explain_log scripts/mysql_config.sh: Added support of --libmysqld-libs sql/Makefile.am: Fix to use 'innodb_system_libs' sql/ha_innobase.cc: Make InnoDB startup/shutdown silent in embedded version sql/ha_isam.cc: Added handling of HAVE_ISAM sql/ha_isammrg.cc: Added handling of HAVE_ISAM sql/ha_myisam.cc: Handle relative paths; Needed to support BACKUP TABLE in embedded version sql/ha_myisammrg.cc: Rename of filenames sql/handler.cc: Added handling of HAVE_ISAM sql/item_func.cc: Fix for ecc (Intel Compiler) sql/mysql_priv.h: Added global variable 'mysql_embedded' sql/mysqld.cc: Use HAVE_ISAM sql/sql_parse.cc: Use new convert_dirname sql/sql_select.cc: Fix for ecc (Intel Compiler) sql/sql_table.cc: Added handling of relative filenames strings/Makefile.am: Don't compile both assembler and C functions on x86 strings/bchange.c: cleanup
2001-10-08 03:58:07 +02:00
connection writer;
reap;
drop table t1;
#
# Test problem when using locks with multi-updates
# It should not block when multi-update is reading on a read-locked table
#
connection locker;
create table t1 (a int, b int);
create table t2 (c int, d int);
insert into t1 values(1,1);
insert into t1 values(2,2);
insert into t2 values(1,2);
lock table t1 read;
connection writer;
update t1,t2 set c=a where b=d;
connection reader;
select c from t2;
connection locker;
Backport of: ------------------------------------------------------------ revno: 2476.784.3 committer: davi@moksha.local timestamp: Tue 2007-10-02 21:27:31 -0300 message: Bug#25858 Some DROP TABLE under LOCK TABLES can cause deadlocks When a client (connection) holds a lock on a table and attempts to drop (obtain a exclusive lock) on a second table that is already held by a second client and the second client then attempts to drop the table that is held by the first client, leads to a circular wait deadlock. This scenario is very similar to trying to drop (or rename) a table while holding read locks and are correctly forbidden. The solution is to allow a drop table operation to continue only if the table being dropped is write (exclusively) locked, or if the table is temporary, or if the client is not holding any locks. Using this scheme prevents the creation of a circular chain in which each client is waiting for one table that the next client in the chain is holding. This is incompatible change, as can be seen by number of tests cases that needed to be fixed, but is consistent with respect to behavior of the different scenarios in which the circular wait might happen. mysql-test/r/drop.result: Test case result for Bug#25858 mysql-test/r/group_by.result: Fix test case result wrt drop table under lock tables -- unlock tables before dropping table. mysql-test/r/insert_notembedded.result: Fix test case result wrt drop table under lock tables -- unlock tables before dropping table mysql-test/r/lock.result: Fix test case result wrt drop table under lock tables -- unlock tables before dropping table. mysql-test/r/lock_multi.result: Fix test case result wrt drop table under lock tables -- unlock tables before dropping table. mysql-test/r/myisam.result: Fix test case result wrt drop table under lock tables -- unlock tables before dropping table. mysql-test/r/view.result: Fix test case result wrt drop table under lock tables -- unlock tables before dropping table. mysql-test/t/drop.test: Add test case for Bug#25858 mysql-test/t/group_by.test: Fix test case: unlock tables in preparation for a drop table. In some circumstances, dropping tables while holding locks leads to a deadlock. mysql-test/t/insert_notembedded.test: Fix test case: unlock tables in preparation for a drop table. In some circumstances, dropping tables while holding locks leads to a deadlock. mysql-test/t/lock.test: Fix test case: unlock tables in preparation for a drop table. In some circumstances, dropping tables while holding locks leads to a deadlock. mysql-test/t/lock_multi.test: Fix test case: unlock tables in preparation for a drop table. In some circumstances, dropping tables while holding locks leads to a deadlock. mysql-test/t/myisam.test: Fix test case: unlock tables in preparation for a drop table. In some circumstances, dropping tables while holding locks leads to a deadlock. mysql-test/t/query_cache_notembedded.test: Fix test case: unlock tables in preparation for a drop table. In some circumstances, dropping tables while holding locks leads to a deadlock. mysql-test/t/view.test: Fix test case: unlock tables in preparation for a drop table. In some circumstances, dropping tables while holding locks leads to a deadlock. sql/lock.cc: When trying to obtain a name lock under lock tables, ensure that the table is properly exclusively locked and fail otherwise.
2009-11-20 20:51:12 +01:00
unlock tables;
drop table t1;
drop table t2;
#
# Test problem when using locks on many tables and dropping a table that
# is to-be-locked by another thread
#
#
connection locker;
create table t1 (a int);
create table t2 (a int);
lock table t1 write, t2 write;
connection reader;
send
insert t1 select * from t2;
connection locker;
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
where state = "Waiting for table metadata lock" and
info = "insert t1 select * from t2";
--source include/wait_condition.inc
drop table t2;
Implement new type-of-operation-aware metadata locks. Add a wait-for graph based deadlock detector to the MDL subsystem. Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and bug #37346 "innodb does not detect deadlock between update and alter table". The first bug manifested itself as an unwarranted abort of a transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER statement, when this transaction tried to repeat use of a table, which it has already used in a similar fashion before ALTER started. The second bug showed up as a deadlock between table-level locks and InnoDB row locks, which was "detected" only after innodb_lock_wait_timeout timeout. A transaction would start using the table and modify a few rows. Then ALTER TABLE would come in, and start copying rows into a temporary table. Eventually it would stumble on the modified records and get blocked on a row lock. The first transaction would try to do more updates, and get blocked on thr_lock.c lock. This situation of circular wait would only get resolved by a timeout. Both these bugs stemmed from inadequate solutions to the problem of deadlocks occurring between different locking subsystems. In the first case we tried to avoid deadlocks between metadata locking and table-level locking subsystems, when upgrading shared metadata lock to exclusive one. Transactions holding the shared lock on the table and waiting for some table-level lock used to be aborted too aggressively. We also allowed ALTER TABLE to start in presence of transactions that modify the subject table. ALTER TABLE acquires TL_WRITE_ALLOW_READ lock at start, and that block all writes against the table (naturally, we don't want any writes to be lost when switching the old and the new table). TL_WRITE_ALLOW_READ lock, in turn, would block the started transaction on thr_lock.c lock, should they do more updates. This, again, lead to the need to abort such transactions. The second bug occurred simply because we didn't have any mechanism to detect deadlocks between the table-level locks in thr_lock.c and row-level locks in InnoDB, other than innodb_lock_wait_timeout. This patch solves both these problems by moving lock conflicts which are causing these deadlocks into the metadata locking subsystem, thus making it possible to avoid or detect such deadlocks inside MDL. To do this we introduce new type-of-operation-aware metadata locks, which allow MDL subsystem to know not only the fact that transaction has used or is going to use some object but also what kind of operation it has carried out or going to carry out on the object. This, along with the addition of a special kind of upgradable metadata lock, allows ALTER TABLE to wait until all transactions which has updated the table to go away. This solves the second issue. Another special type of upgradable metadata lock is acquired by LOCK TABLE WRITE. This second lock type allows to solve the first issue, since abortion of table-level locks in event of DDL under LOCK TABLES becomes also unnecessary. Below follows the list of incompatible changes introduced by this patch: - From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those statements that acquire TL_WRITE_ALLOW_READ lock) wait for all transactions which has *updated* the table to complete. - From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE (i.e. all statements which acquire TL_WRITE table-level lock) wait for all transaction which *updated or read* from the table to complete. As a consequence, innodb_table_locks=0 option no longer applies to LOCK TABLES ... WRITE. - DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort statements or transactions which use tables being dropped or renamed, and instead wait for these transactions to complete. - Since LOCK TABLES WRITE now takes a special metadata lock, not compatible with with reads or writes against the subject table and transaction-wide, thr_lock.c deadlock avoidance algorithm that used to ensure absence of deadlocks between LOCK TABLES WRITE and other statements is no longer sufficient, even for MyISAM. The wait-for graph based deadlock detector of MDL subsystem may sometimes be necessary and is involved. This may lead to ER_LOCK_DEADLOCK error produced for multi-statement transactions even if these only use MyISAM: session 1: session 2: begin; update t1 ... lock table t2 write, t1 write; -- gets a lock on t2, blocks on t1 update t2 ... (ER_LOCK_DEADLOCK) - Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE was abandoned. LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same priority as the usual LOCK TABLE ... WRITE. SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in the wait queue. - We do not take upgradable metadata locks on implicitly locked tables. So if one has, say, a view v1 that uses table t1, and issues: LOCK TABLE v1 WRITE; FLUSH TABLE t1; -- (or just 'FLUSH TABLES'), an error is produced. In order to be able to perform DDL on a table under LOCK TABLES, the table must be locked explicitly in the LOCK TABLES list. mysql-test/include/handler.inc: Adjusted test case to trigger an execution path on which bug 41110 "crash with handler command when used concurrently with alter table" and bug 41112 "crash in mysql_ha_close_table/get_lock_data with alter table" were originally discovered. Left old test case which no longer triggers this execution path for the sake of coverage. Added test coverage for HANDLER SQL statements and type-aware metadata locks. Added a test for the global shared lock and HANDLER SQL. Updated tests to take into account that the old simple deadlock detection heuristics was replaced with a graph-based deadlock detector. mysql-test/r/debug_sync.result: Updated results (see debug_sync.test). mysql-test/r/handler_innodb.result: Updated results (see handler.inc test). mysql-test/r/handler_myisam.result: Updated results (see handler.inc test). mysql-test/r/innodb-lock.result: Updated results (see innodb-lock.test). mysql-test/r/innodb_mysql_lock.result: Updated results (see innodb_mysql_lock.test). mysql-test/r/lock.result: Updated results (see lock.test). mysql-test/r/lock_multi.result: Updated results (see lock_multi.test). mysql-test/r/lock_sync.result: Updated results (see lock_sync.test). mysql-test/r/mdl_sync.result: Updated results (see mdl_sync.test). mysql-test/r/sp-threads.result: SHOW PROCESSLIST output has changed due to the fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/r/truncate_coverage.result: Updated results (see truncate_coverage.test). mysql-test/suite/funcs_1/datadict/processlist_val.inc: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/funcs_1/r/processlist_val_no_prot.result: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/rpl/t/rpl_sp.test: Updated to a new SHOW PROCESSLIST state name. mysql-test/t/debug_sync.test: Use LOCK TABLES READ instead of LOCK TABLES WRITE as the latter no longer allows to trigger execution path involving waiting on thr_lock.c lock and therefore reaching debug sync-point covered by this test. mysql-test/t/innodb-lock.test: Adjusted test case to the fact that innodb_table_locks=0 option is no longer supported, since LOCK TABLES WRITE handles all its conflicts within MDL subsystem. mysql-test/t/innodb_mysql_lock.test: Added test for bug #37346 "innodb does not detect deadlock between update and alter table". mysql-test/t/lock.test: Added test coverage which checks the fact that we no longer support DDL under LOCK TABLES on tables which were locked implicitly. Adjusted existing test cases accordingly. mysql-test/t/lock_multi.test: Added test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock". Adjusted other test cases to take into account the fact that waiting for LOCK TABLES ... WRITE now happens within MDL subsystem. mysql-test/t/lock_sync.test: Since LOCK TABLES ... WRITE now takes SNRW metadata lock for tables locked explicitly we have to implicitly lock InnoDB tables (through view) to trigger the table-level lock conflict between TL_WRITE and TL_WRITE_ALLOW_WRITE. mysql-test/t/mdl_sync.test: Added basic test coverage for type-of-operation-aware metadata locks. Also covered with tests some use cases involving HANDLER statements in which a deadlock could arise. Adjusted existing tests to take type-of-operation-aware MDL into account. mysql-test/t/multi_update.test: Update to a new SHOW PROCESSLIST state name. mysql-test/t/truncate_coverage.test: Adjusted test case after making LOCK TABLES WRITE to wait until transactions that use the table to be locked are completed. Updated to the changed name of DEBUG_SYNC point. sql/handler.cc: Global read lock functionality has been moved into a class. sql/lock.cc: Global read lock functionality has been moved into a class. Updated code to use the new MDL API. sql/mdl.cc: Introduced new type-of-operation aware metadata locks. To do this: - Changed MDL_lock to use one list for waiting requests and one list for granted requests. For each list, added a bitmap that holds information what lock types a list contains. Added a helper class MDL_lock::List to manipulate with granted and waited lists while keeping the bitmaps in sync with list contents. - Changed lock-compatibility functions to use bitmaps that define compatibility. - Introduced a graph based deadlock detector inspired by waiting_threads.c from Maria implementation. - Now that we have a deadlock detector, and no longer have a global lock to protect individual lock objects, but rather use an rw lock per object, removed redundant code for upgrade, and the global read lock. Changed the MDL API to no longer require the caller to acquire the global intention exclusive lock by means of a separate method. Removed a few more methods that became redundant. - Removed deadlock detection heuristic, it has been made obsolete by the deadlock detector. - With operation-type-aware metadata locks, MDL subsystem has become aware of potential conflicts between DDL and open transactions. This made it possible to remove calls to mysql_abort_transactions_with_shared_lock() from acquisition paths for exclusive lock and lock upgrade. Now we can simply wait for these transactions to complete without fear of deadlock. Function mysql_lock_abort() has also become unnecessary for all conflicting cases except when a DDL conflicts with a connection that has an open HANDLER. sql/mdl.h: Introduced new type-of-operation aware metadata locks. Introduced a graph based deadlock detector and supporting methods. Added comments. God rid of redundant API calls. Renamed m_lt_or_ha_sentinel to m_trans_sentinel, since now it guards the global read lock as well as LOCK TABLES and HANDLER locks. sql/mysql_priv.h: Moved the global read lock functionality into a class. Added MYSQL_OPEN_FORCE_SHARED_MDL flag which forces open_tables() to take MDL_SHARED on tables instead of metadata locks specified in the parser. We use this to allow PREPARE run concurrently in presence of LOCK TABLES ... WRITE. Added signature for find_table_for_mdl_ugprade(). sql/set_var.cc: Global read lock functionality has been moved into a class. sql/sp_head.cc: When creating TABLE_LIST elements for prelocking or system tables set the type of request for metadata lock according to the operation that will be performed on the table. sql/sql_base.cc: - Updated code to use the new MDL API. - In order to avoid locks starvation we take upgradable locks all at once. As result implicitly locked tables no longer get an upgradable lock. Consequently DDL and FLUSH TABLES for such tables is prohibited. find_write_locked_table() was replaced by find_table_for_mdl_upgrade() function. open_table() was adjusted to return TABLE instance with upgradable ticket when necessary. - We no longer wait for all locks on OT_WAIT back off action -- only on the lock that caused the wait conflict. Moreover, now we distinguish cases when we have to wait due to conflict in MDL and old version of table in TDC. - Upate mysql_notify_threads_having_share_locks() to only abort thr_lock.c waits of threads that have open HANDLERs, since lock conflicts with only these threads now can lead to deadlocks not detectable by the MDL deadlock detector. - Remove mysql_abort_transactions_with_shared_locks() which is no longer needed. sql/sql_class.cc: Global read lock functionality has been moved into a class. Re-arranged code in THD::cleanup() to simplify assert. sql/sql_class.h: Introduced class to incapsulate global read lock functionality. Now sentinel in MDL subsystem guards the global read lock as well as LOCK TABLES and HANDLER locks. Adjusted code accordingly. sql/sql_db.cc: Global read lock functionality has been moved into a class. sql/sql_delete.cc: We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result TRUNCATE TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_handler.cc: Inform MDL_context about presence of open HANDLERs. Since HANLDERs break MDL protocol by acquiring table-level lock while holding only S metadata lock on a table MDL subsystem should take special care about such contexts (Now this is the only case when mysql_lock_abort() is used). sql/sql_parse.cc: Global read lock functionality has been moved into a class. Do not take upgradable metadata locks when opening tables for CREATE TABLE SELECT as it is not necessary and limits concurrency. When initializing TABLE_LIST objects before adding them to the table list set the type of request for metadata lock according to the operation that will be performed on the table. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result FLUSH TABLES is no longer allowed for such tables. sql/sql_prepare.cc: Use MYSQL_OPEN_FORCE_SHARED_MDL flag when opening tables during PREPARE. This allows PREPARE to run concurrently in presence of LOCK TABLES ... WRITE. sql/sql_rename.cc: Global read lock functionality has been moved into a class. sql/sql_show.cc: Updated code to use the new MDL API. sql/sql_table.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result DROP TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_trigger.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result CREATE/DROP TRIGGER is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_view.cc: Global read lock functionality has been moved into a class. Fixed results of wrong merge that led to misuse of GLR API. CREATE VIEW statement is not a commit statement. sql/table.cc: When resetting TABLE_LIST objects for PS or SP re-execution set the type of request for metadata lock according to the operation that will be performed on the table. Do the same in auxiliary function initializing metadata lock requests in a table list. sql/table.h: When initializing TABLE_LIST objects set the type of request for metadata lock according to the operation that will be performed on the table. sql/transaction.cc: Global read lock functionality has been moved into a class.
2010-02-01 12:43:06 +01:00
unlock tables;
connection reader;
--error ER_NO_SUCH_TABLE
reap;
connection locker;
drop table t1;
Fixed compiler and valgrind warnings Added missing DBUG_xxx_RETURN statements Fixed some usage of not initialized variables (as found by valgrind) Ensure that we don't remove locked tables used as name locks from open table cache until unlock_table_names() are called. This was fixed by having drop_locked_name() returning any table used as a name lock so that we can free it in unlock_table_names() This will allow Tomas to continue with his work to use namelocks to syncronize things. Note: valgrind still produces a lot of warnings about using not initialized code and shows memory loss errors when running the ndb tests BitKeeper/etc/ignore: added mysql-test/r/*.log client/mysqltest.c: Change type of variables to get rid of compiler warnings More debugging Fix memory leak mysql-test/mysql-test-run.sh: Collect warnings about missing DBUG_RETURN statements mysql-test/r/lock_multi.result: Add test of new code mysql-test/r/ndb_condition_pushdown.result: Drop used tables before test mysql-test/t/lock_multi.test: Add test of new code mysql-test/t/ndb_condition_pushdown.test: Drop used tables before test mysql-test/valgrind.supp: Ignore 'safe' warnings from libz (when used with archive) sql/event.cc: More comments Simplify code Fixed memory leak found by valgrind sql/ha_archive.cc: Remove compiler warnings (Wrong handlerton structure and signed/unsigned comparison) sql/ha_berkeley.cc: Fixed compiler warning sql/ha_blackhole.cc: Fixed compiler warning sql/ha_federated.cc: Fixed compiler warning sql/ha_heap.cc: Fixed compiler warning sql/ha_myisam.cc: Fixed compiler warning sql/ha_myisammrg.cc: Fixed compiler warning sql/ha_ndbcluster.cc: Fixed compiler warnings sql/ha_partition.cc: Fixed compiler warning Fixed error noticed by valgrind in ha_partition::rnd_init() sql/handler.cc: Fixed compiler warning sql/handler.h: Fixed compiler warning sql/item.cc: Fixed compiler warning sql/item_xmlfunc.cc: Fixed warning from valgrind when calling memcpy with wrong address sql/lock.cc: More debugging sql/log.cc: Fixed compiler warning Indentation fixes sql/log.h: Fixed compiler warning sql/mysql_priv.h: Changed prototype for 'drop_locked_tables' sql/opt_range.cc: Indentation fix sql/password.c: Removed compiler warnings sql/set_var.cc: Fixed compiler warning sql/slave.cc: Fixed compiler warning sql/sp_head.cc: Fixed compiler warning sql/sql_acl.cc: Fixed compiler warning sql/sql_analyse.cc: Added missing DBUG_RETURN statements sql/sql_base.cc: Removed de-reference of not initialized pointer More comments drop_locked_tables() changed to not delete tables used for name locking Fixed compiler warnings sql/sql_delete.cc: Fixed usage of not initialized variable (deleted could be referenced in some not common error conditions) sql/sql_parse.cc: Added missing DBUG_VOID_RETURN Simplify code sql/sql_partition.cc: Fixed usage of wrong variable (noticed by valgrind) sql/sql_plugin.cc: Removed compiler warning sql/sql_show.cc: Removed compiler warning sql/sql_table.cc: Ensure that we don't remove locked tables used as name locks from open table cache until unlock_table_names() are called. This was fixed by having drop_locked_name() returning any table used as a name lock so that we can free it in unlock_table_names() This will allow Tomas to continue with his work to use namelocks to syncronize things. Fixed wrong test of 'table_type' (path_length could otherwise be accessed uninitialized) Remove compile warnings about not initialized variables. sql/sql_yacc.yy: Ensure that no_write_to_binlog is properly initialized (Was accessed uninitialized by partition code) sql/table.cc: Removed valgrind warnings (not fatal) Removed compiler warnings sql/tztime.cc: Removed valgrind warning storage/ndb/include/ndbapi/NdbIndexStat.hpp: Removed compiler warning
2006-03-29 13:27:36 +02:00
#
# Same test as above, but with the dropped table locked twice
#
connection locker;
create table t1 (a int);
create table t2 (a int);
lock table t1 write, t2 write, t1 as t1_2 write, t2 as t2_2 write;
connection reader;
send
insert t1 select * from t2;
Fixed compiler and valgrind warnings Added missing DBUG_xxx_RETURN statements Fixed some usage of not initialized variables (as found by valgrind) Ensure that we don't remove locked tables used as name locks from open table cache until unlock_table_names() are called. This was fixed by having drop_locked_name() returning any table used as a name lock so that we can free it in unlock_table_names() This will allow Tomas to continue with his work to use namelocks to syncronize things. Note: valgrind still produces a lot of warnings about using not initialized code and shows memory loss errors when running the ndb tests BitKeeper/etc/ignore: added mysql-test/r/*.log client/mysqltest.c: Change type of variables to get rid of compiler warnings More debugging Fix memory leak mysql-test/mysql-test-run.sh: Collect warnings about missing DBUG_RETURN statements mysql-test/r/lock_multi.result: Add test of new code mysql-test/r/ndb_condition_pushdown.result: Drop used tables before test mysql-test/t/lock_multi.test: Add test of new code mysql-test/t/ndb_condition_pushdown.test: Drop used tables before test mysql-test/valgrind.supp: Ignore 'safe' warnings from libz (when used with archive) sql/event.cc: More comments Simplify code Fixed memory leak found by valgrind sql/ha_archive.cc: Remove compiler warnings (Wrong handlerton structure and signed/unsigned comparison) sql/ha_berkeley.cc: Fixed compiler warning sql/ha_blackhole.cc: Fixed compiler warning sql/ha_federated.cc: Fixed compiler warning sql/ha_heap.cc: Fixed compiler warning sql/ha_myisam.cc: Fixed compiler warning sql/ha_myisammrg.cc: Fixed compiler warning sql/ha_ndbcluster.cc: Fixed compiler warnings sql/ha_partition.cc: Fixed compiler warning Fixed error noticed by valgrind in ha_partition::rnd_init() sql/handler.cc: Fixed compiler warning sql/handler.h: Fixed compiler warning sql/item.cc: Fixed compiler warning sql/item_xmlfunc.cc: Fixed warning from valgrind when calling memcpy with wrong address sql/lock.cc: More debugging sql/log.cc: Fixed compiler warning Indentation fixes sql/log.h: Fixed compiler warning sql/mysql_priv.h: Changed prototype for 'drop_locked_tables' sql/opt_range.cc: Indentation fix sql/password.c: Removed compiler warnings sql/set_var.cc: Fixed compiler warning sql/slave.cc: Fixed compiler warning sql/sp_head.cc: Fixed compiler warning sql/sql_acl.cc: Fixed compiler warning sql/sql_analyse.cc: Added missing DBUG_RETURN statements sql/sql_base.cc: Removed de-reference of not initialized pointer More comments drop_locked_tables() changed to not delete tables used for name locking Fixed compiler warnings sql/sql_delete.cc: Fixed usage of not initialized variable (deleted could be referenced in some not common error conditions) sql/sql_parse.cc: Added missing DBUG_VOID_RETURN Simplify code sql/sql_partition.cc: Fixed usage of wrong variable (noticed by valgrind) sql/sql_plugin.cc: Removed compiler warning sql/sql_show.cc: Removed compiler warning sql/sql_table.cc: Ensure that we don't remove locked tables used as name locks from open table cache until unlock_table_names() are called. This was fixed by having drop_locked_name() returning any table used as a name lock so that we can free it in unlock_table_names() This will allow Tomas to continue with his work to use namelocks to syncronize things. Fixed wrong test of 'table_type' (path_length could otherwise be accessed uninitialized) Remove compile warnings about not initialized variables. sql/sql_yacc.yy: Ensure that no_write_to_binlog is properly initialized (Was accessed uninitialized by partition code) sql/table.cc: Removed valgrind warnings (not fatal) Removed compiler warnings sql/tztime.cc: Removed valgrind warning storage/ndb/include/ndbapi/NdbIndexStat.hpp: Removed compiler warning
2006-03-29 13:27:36 +02:00
connection locker;
# Sleep a bit till the insert of connection reader is in work and hangs
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
where state = "Waiting for table metadata lock" and
info = "insert t1 select * from t2";
--source include/wait_condition.inc
Fixed compiler and valgrind warnings Added missing DBUG_xxx_RETURN statements Fixed some usage of not initialized variables (as found by valgrind) Ensure that we don't remove locked tables used as name locks from open table cache until unlock_table_names() are called. This was fixed by having drop_locked_name() returning any table used as a name lock so that we can free it in unlock_table_names() This will allow Tomas to continue with his work to use namelocks to syncronize things. Note: valgrind still produces a lot of warnings about using not initialized code and shows memory loss errors when running the ndb tests BitKeeper/etc/ignore: added mysql-test/r/*.log client/mysqltest.c: Change type of variables to get rid of compiler warnings More debugging Fix memory leak mysql-test/mysql-test-run.sh: Collect warnings about missing DBUG_RETURN statements mysql-test/r/lock_multi.result: Add test of new code mysql-test/r/ndb_condition_pushdown.result: Drop used tables before test mysql-test/t/lock_multi.test: Add test of new code mysql-test/t/ndb_condition_pushdown.test: Drop used tables before test mysql-test/valgrind.supp: Ignore 'safe' warnings from libz (when used with archive) sql/event.cc: More comments Simplify code Fixed memory leak found by valgrind sql/ha_archive.cc: Remove compiler warnings (Wrong handlerton structure and signed/unsigned comparison) sql/ha_berkeley.cc: Fixed compiler warning sql/ha_blackhole.cc: Fixed compiler warning sql/ha_federated.cc: Fixed compiler warning sql/ha_heap.cc: Fixed compiler warning sql/ha_myisam.cc: Fixed compiler warning sql/ha_myisammrg.cc: Fixed compiler warning sql/ha_ndbcluster.cc: Fixed compiler warnings sql/ha_partition.cc: Fixed compiler warning Fixed error noticed by valgrind in ha_partition::rnd_init() sql/handler.cc: Fixed compiler warning sql/handler.h: Fixed compiler warning sql/item.cc: Fixed compiler warning sql/item_xmlfunc.cc: Fixed warning from valgrind when calling memcpy with wrong address sql/lock.cc: More debugging sql/log.cc: Fixed compiler warning Indentation fixes sql/log.h: Fixed compiler warning sql/mysql_priv.h: Changed prototype for 'drop_locked_tables' sql/opt_range.cc: Indentation fix sql/password.c: Removed compiler warnings sql/set_var.cc: Fixed compiler warning sql/slave.cc: Fixed compiler warning sql/sp_head.cc: Fixed compiler warning sql/sql_acl.cc: Fixed compiler warning sql/sql_analyse.cc: Added missing DBUG_RETURN statements sql/sql_base.cc: Removed de-reference of not initialized pointer More comments drop_locked_tables() changed to not delete tables used for name locking Fixed compiler warnings sql/sql_delete.cc: Fixed usage of not initialized variable (deleted could be referenced in some not common error conditions) sql/sql_parse.cc: Added missing DBUG_VOID_RETURN Simplify code sql/sql_partition.cc: Fixed usage of wrong variable (noticed by valgrind) sql/sql_plugin.cc: Removed compiler warning sql/sql_show.cc: Removed compiler warning sql/sql_table.cc: Ensure that we don't remove locked tables used as name locks from open table cache until unlock_table_names() are called. This was fixed by having drop_locked_name() returning any table used as a name lock so that we can free it in unlock_table_names() This will allow Tomas to continue with his work to use namelocks to syncronize things. Fixed wrong test of 'table_type' (path_length could otherwise be accessed uninitialized) Remove compile warnings about not initialized variables. sql/sql_yacc.yy: Ensure that no_write_to_binlog is properly initialized (Was accessed uninitialized by partition code) sql/table.cc: Removed valgrind warnings (not fatal) Removed compiler warnings sql/tztime.cc: Removed valgrind warning storage/ndb/include/ndbapi/NdbIndexStat.hpp: Removed compiler warning
2006-03-29 13:27:36 +02:00
drop table t2;
Implement new type-of-operation-aware metadata locks. Add a wait-for graph based deadlock detector to the MDL subsystem. Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and bug #37346 "innodb does not detect deadlock between update and alter table". The first bug manifested itself as an unwarranted abort of a transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER statement, when this transaction tried to repeat use of a table, which it has already used in a similar fashion before ALTER started. The second bug showed up as a deadlock between table-level locks and InnoDB row locks, which was "detected" only after innodb_lock_wait_timeout timeout. A transaction would start using the table and modify a few rows. Then ALTER TABLE would come in, and start copying rows into a temporary table. Eventually it would stumble on the modified records and get blocked on a row lock. The first transaction would try to do more updates, and get blocked on thr_lock.c lock. This situation of circular wait would only get resolved by a timeout. Both these bugs stemmed from inadequate solutions to the problem of deadlocks occurring between different locking subsystems. In the first case we tried to avoid deadlocks between metadata locking and table-level locking subsystems, when upgrading shared metadata lock to exclusive one. Transactions holding the shared lock on the table and waiting for some table-level lock used to be aborted too aggressively. We also allowed ALTER TABLE to start in presence of transactions that modify the subject table. ALTER TABLE acquires TL_WRITE_ALLOW_READ lock at start, and that block all writes against the table (naturally, we don't want any writes to be lost when switching the old and the new table). TL_WRITE_ALLOW_READ lock, in turn, would block the started transaction on thr_lock.c lock, should they do more updates. This, again, lead to the need to abort such transactions. The second bug occurred simply because we didn't have any mechanism to detect deadlocks between the table-level locks in thr_lock.c and row-level locks in InnoDB, other than innodb_lock_wait_timeout. This patch solves both these problems by moving lock conflicts which are causing these deadlocks into the metadata locking subsystem, thus making it possible to avoid or detect such deadlocks inside MDL. To do this we introduce new type-of-operation-aware metadata locks, which allow MDL subsystem to know not only the fact that transaction has used or is going to use some object but also what kind of operation it has carried out or going to carry out on the object. This, along with the addition of a special kind of upgradable metadata lock, allows ALTER TABLE to wait until all transactions which has updated the table to go away. This solves the second issue. Another special type of upgradable metadata lock is acquired by LOCK TABLE WRITE. This second lock type allows to solve the first issue, since abortion of table-level locks in event of DDL under LOCK TABLES becomes also unnecessary. Below follows the list of incompatible changes introduced by this patch: - From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those statements that acquire TL_WRITE_ALLOW_READ lock) wait for all transactions which has *updated* the table to complete. - From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE (i.e. all statements which acquire TL_WRITE table-level lock) wait for all transaction which *updated or read* from the table to complete. As a consequence, innodb_table_locks=0 option no longer applies to LOCK TABLES ... WRITE. - DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort statements or transactions which use tables being dropped or renamed, and instead wait for these transactions to complete. - Since LOCK TABLES WRITE now takes a special metadata lock, not compatible with with reads or writes against the subject table and transaction-wide, thr_lock.c deadlock avoidance algorithm that used to ensure absence of deadlocks between LOCK TABLES WRITE and other statements is no longer sufficient, even for MyISAM. The wait-for graph based deadlock detector of MDL subsystem may sometimes be necessary and is involved. This may lead to ER_LOCK_DEADLOCK error produced for multi-statement transactions even if these only use MyISAM: session 1: session 2: begin; update t1 ... lock table t2 write, t1 write; -- gets a lock on t2, blocks on t1 update t2 ... (ER_LOCK_DEADLOCK) - Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE was abandoned. LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same priority as the usual LOCK TABLE ... WRITE. SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in the wait queue. - We do not take upgradable metadata locks on implicitly locked tables. So if one has, say, a view v1 that uses table t1, and issues: LOCK TABLE v1 WRITE; FLUSH TABLE t1; -- (or just 'FLUSH TABLES'), an error is produced. In order to be able to perform DDL on a table under LOCK TABLES, the table must be locked explicitly in the LOCK TABLES list. mysql-test/include/handler.inc: Adjusted test case to trigger an execution path on which bug 41110 "crash with handler command when used concurrently with alter table" and bug 41112 "crash in mysql_ha_close_table/get_lock_data with alter table" were originally discovered. Left old test case which no longer triggers this execution path for the sake of coverage. Added test coverage for HANDLER SQL statements and type-aware metadata locks. Added a test for the global shared lock and HANDLER SQL. Updated tests to take into account that the old simple deadlock detection heuristics was replaced with a graph-based deadlock detector. mysql-test/r/debug_sync.result: Updated results (see debug_sync.test). mysql-test/r/handler_innodb.result: Updated results (see handler.inc test). mysql-test/r/handler_myisam.result: Updated results (see handler.inc test). mysql-test/r/innodb-lock.result: Updated results (see innodb-lock.test). mysql-test/r/innodb_mysql_lock.result: Updated results (see innodb_mysql_lock.test). mysql-test/r/lock.result: Updated results (see lock.test). mysql-test/r/lock_multi.result: Updated results (see lock_multi.test). mysql-test/r/lock_sync.result: Updated results (see lock_sync.test). mysql-test/r/mdl_sync.result: Updated results (see mdl_sync.test). mysql-test/r/sp-threads.result: SHOW PROCESSLIST output has changed due to the fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/r/truncate_coverage.result: Updated results (see truncate_coverage.test). mysql-test/suite/funcs_1/datadict/processlist_val.inc: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/funcs_1/r/processlist_val_no_prot.result: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/rpl/t/rpl_sp.test: Updated to a new SHOW PROCESSLIST state name. mysql-test/t/debug_sync.test: Use LOCK TABLES READ instead of LOCK TABLES WRITE as the latter no longer allows to trigger execution path involving waiting on thr_lock.c lock and therefore reaching debug sync-point covered by this test. mysql-test/t/innodb-lock.test: Adjusted test case to the fact that innodb_table_locks=0 option is no longer supported, since LOCK TABLES WRITE handles all its conflicts within MDL subsystem. mysql-test/t/innodb_mysql_lock.test: Added test for bug #37346 "innodb does not detect deadlock between update and alter table". mysql-test/t/lock.test: Added test coverage which checks the fact that we no longer support DDL under LOCK TABLES on tables which were locked implicitly. Adjusted existing test cases accordingly. mysql-test/t/lock_multi.test: Added test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock". Adjusted other test cases to take into account the fact that waiting for LOCK TABLES ... WRITE now happens within MDL subsystem. mysql-test/t/lock_sync.test: Since LOCK TABLES ... WRITE now takes SNRW metadata lock for tables locked explicitly we have to implicitly lock InnoDB tables (through view) to trigger the table-level lock conflict between TL_WRITE and TL_WRITE_ALLOW_WRITE. mysql-test/t/mdl_sync.test: Added basic test coverage for type-of-operation-aware metadata locks. Also covered with tests some use cases involving HANDLER statements in which a deadlock could arise. Adjusted existing tests to take type-of-operation-aware MDL into account. mysql-test/t/multi_update.test: Update to a new SHOW PROCESSLIST state name. mysql-test/t/truncate_coverage.test: Adjusted test case after making LOCK TABLES WRITE to wait until transactions that use the table to be locked are completed. Updated to the changed name of DEBUG_SYNC point. sql/handler.cc: Global read lock functionality has been moved into a class. sql/lock.cc: Global read lock functionality has been moved into a class. Updated code to use the new MDL API. sql/mdl.cc: Introduced new type-of-operation aware metadata locks. To do this: - Changed MDL_lock to use one list for waiting requests and one list for granted requests. For each list, added a bitmap that holds information what lock types a list contains. Added a helper class MDL_lock::List to manipulate with granted and waited lists while keeping the bitmaps in sync with list contents. - Changed lock-compatibility functions to use bitmaps that define compatibility. - Introduced a graph based deadlock detector inspired by waiting_threads.c from Maria implementation. - Now that we have a deadlock detector, and no longer have a global lock to protect individual lock objects, but rather use an rw lock per object, removed redundant code for upgrade, and the global read lock. Changed the MDL API to no longer require the caller to acquire the global intention exclusive lock by means of a separate method. Removed a few more methods that became redundant. - Removed deadlock detection heuristic, it has been made obsolete by the deadlock detector. - With operation-type-aware metadata locks, MDL subsystem has become aware of potential conflicts between DDL and open transactions. This made it possible to remove calls to mysql_abort_transactions_with_shared_lock() from acquisition paths for exclusive lock and lock upgrade. Now we can simply wait for these transactions to complete without fear of deadlock. Function mysql_lock_abort() has also become unnecessary for all conflicting cases except when a DDL conflicts with a connection that has an open HANDLER. sql/mdl.h: Introduced new type-of-operation aware metadata locks. Introduced a graph based deadlock detector and supporting methods. Added comments. God rid of redundant API calls. Renamed m_lt_or_ha_sentinel to m_trans_sentinel, since now it guards the global read lock as well as LOCK TABLES and HANDLER locks. sql/mysql_priv.h: Moved the global read lock functionality into a class. Added MYSQL_OPEN_FORCE_SHARED_MDL flag which forces open_tables() to take MDL_SHARED on tables instead of metadata locks specified in the parser. We use this to allow PREPARE run concurrently in presence of LOCK TABLES ... WRITE. Added signature for find_table_for_mdl_ugprade(). sql/set_var.cc: Global read lock functionality has been moved into a class. sql/sp_head.cc: When creating TABLE_LIST elements for prelocking or system tables set the type of request for metadata lock according to the operation that will be performed on the table. sql/sql_base.cc: - Updated code to use the new MDL API. - In order to avoid locks starvation we take upgradable locks all at once. As result implicitly locked tables no longer get an upgradable lock. Consequently DDL and FLUSH TABLES for such tables is prohibited. find_write_locked_table() was replaced by find_table_for_mdl_upgrade() function. open_table() was adjusted to return TABLE instance with upgradable ticket when necessary. - We no longer wait for all locks on OT_WAIT back off action -- only on the lock that caused the wait conflict. Moreover, now we distinguish cases when we have to wait due to conflict in MDL and old version of table in TDC. - Upate mysql_notify_threads_having_share_locks() to only abort thr_lock.c waits of threads that have open HANDLERs, since lock conflicts with only these threads now can lead to deadlocks not detectable by the MDL deadlock detector. - Remove mysql_abort_transactions_with_shared_locks() which is no longer needed. sql/sql_class.cc: Global read lock functionality has been moved into a class. Re-arranged code in THD::cleanup() to simplify assert. sql/sql_class.h: Introduced class to incapsulate global read lock functionality. Now sentinel in MDL subsystem guards the global read lock as well as LOCK TABLES and HANDLER locks. Adjusted code accordingly. sql/sql_db.cc: Global read lock functionality has been moved into a class. sql/sql_delete.cc: We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result TRUNCATE TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_handler.cc: Inform MDL_context about presence of open HANDLERs. Since HANLDERs break MDL protocol by acquiring table-level lock while holding only S metadata lock on a table MDL subsystem should take special care about such contexts (Now this is the only case when mysql_lock_abort() is used). sql/sql_parse.cc: Global read lock functionality has been moved into a class. Do not take upgradable metadata locks when opening tables for CREATE TABLE SELECT as it is not necessary and limits concurrency. When initializing TABLE_LIST objects before adding them to the table list set the type of request for metadata lock according to the operation that will be performed on the table. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result FLUSH TABLES is no longer allowed for such tables. sql/sql_prepare.cc: Use MYSQL_OPEN_FORCE_SHARED_MDL flag when opening tables during PREPARE. This allows PREPARE to run concurrently in presence of LOCK TABLES ... WRITE. sql/sql_rename.cc: Global read lock functionality has been moved into a class. sql/sql_show.cc: Updated code to use the new MDL API. sql/sql_table.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result DROP TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_trigger.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result CREATE/DROP TRIGGER is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_view.cc: Global read lock functionality has been moved into a class. Fixed results of wrong merge that led to misuse of GLR API. CREATE VIEW statement is not a commit statement. sql/table.cc: When resetting TABLE_LIST objects for PS or SP re-execution set the type of request for metadata lock according to the operation that will be performed on the table. Do the same in auxiliary function initializing metadata lock requests in a table list. sql/table.h: When initializing TABLE_LIST objects set the type of request for metadata lock according to the operation that will be performed on the table. sql/transaction.cc: Global read lock functionality has been moved into a class.
2010-02-01 12:43:06 +01:00
unlock tables;
Fixed compiler and valgrind warnings Added missing DBUG_xxx_RETURN statements Fixed some usage of not initialized variables (as found by valgrind) Ensure that we don't remove locked tables used as name locks from open table cache until unlock_table_names() are called. This was fixed by having drop_locked_name() returning any table used as a name lock so that we can free it in unlock_table_names() This will allow Tomas to continue with his work to use namelocks to syncronize things. Note: valgrind still produces a lot of warnings about using not initialized code and shows memory loss errors when running the ndb tests BitKeeper/etc/ignore: added mysql-test/r/*.log client/mysqltest.c: Change type of variables to get rid of compiler warnings More debugging Fix memory leak mysql-test/mysql-test-run.sh: Collect warnings about missing DBUG_RETURN statements mysql-test/r/lock_multi.result: Add test of new code mysql-test/r/ndb_condition_pushdown.result: Drop used tables before test mysql-test/t/lock_multi.test: Add test of new code mysql-test/t/ndb_condition_pushdown.test: Drop used tables before test mysql-test/valgrind.supp: Ignore 'safe' warnings from libz (when used with archive) sql/event.cc: More comments Simplify code Fixed memory leak found by valgrind sql/ha_archive.cc: Remove compiler warnings (Wrong handlerton structure and signed/unsigned comparison) sql/ha_berkeley.cc: Fixed compiler warning sql/ha_blackhole.cc: Fixed compiler warning sql/ha_federated.cc: Fixed compiler warning sql/ha_heap.cc: Fixed compiler warning sql/ha_myisam.cc: Fixed compiler warning sql/ha_myisammrg.cc: Fixed compiler warning sql/ha_ndbcluster.cc: Fixed compiler warnings sql/ha_partition.cc: Fixed compiler warning Fixed error noticed by valgrind in ha_partition::rnd_init() sql/handler.cc: Fixed compiler warning sql/handler.h: Fixed compiler warning sql/item.cc: Fixed compiler warning sql/item_xmlfunc.cc: Fixed warning from valgrind when calling memcpy with wrong address sql/lock.cc: More debugging sql/log.cc: Fixed compiler warning Indentation fixes sql/log.h: Fixed compiler warning sql/mysql_priv.h: Changed prototype for 'drop_locked_tables' sql/opt_range.cc: Indentation fix sql/password.c: Removed compiler warnings sql/set_var.cc: Fixed compiler warning sql/slave.cc: Fixed compiler warning sql/sp_head.cc: Fixed compiler warning sql/sql_acl.cc: Fixed compiler warning sql/sql_analyse.cc: Added missing DBUG_RETURN statements sql/sql_base.cc: Removed de-reference of not initialized pointer More comments drop_locked_tables() changed to not delete tables used for name locking Fixed compiler warnings sql/sql_delete.cc: Fixed usage of not initialized variable (deleted could be referenced in some not common error conditions) sql/sql_parse.cc: Added missing DBUG_VOID_RETURN Simplify code sql/sql_partition.cc: Fixed usage of wrong variable (noticed by valgrind) sql/sql_plugin.cc: Removed compiler warning sql/sql_show.cc: Removed compiler warning sql/sql_table.cc: Ensure that we don't remove locked tables used as name locks from open table cache until unlock_table_names() are called. This was fixed by having drop_locked_name() returning any table used as a name lock so that we can free it in unlock_table_names() This will allow Tomas to continue with his work to use namelocks to syncronize things. Fixed wrong test of 'table_type' (path_length could otherwise be accessed uninitialized) Remove compile warnings about not initialized variables. sql/sql_yacc.yy: Ensure that no_write_to_binlog is properly initialized (Was accessed uninitialized by partition code) sql/table.cc: Removed valgrind warnings (not fatal) Removed compiler warnings sql/tztime.cc: Removed valgrind warning storage/ndb/include/ndbapi/NdbIndexStat.hpp: Removed compiler warning
2006-03-29 13:27:36 +02:00
connection reader;
--error ER_NO_SUCH_TABLE
Fixed compiler and valgrind warnings Added missing DBUG_xxx_RETURN statements Fixed some usage of not initialized variables (as found by valgrind) Ensure that we don't remove locked tables used as name locks from open table cache until unlock_table_names() are called. This was fixed by having drop_locked_name() returning any table used as a name lock so that we can free it in unlock_table_names() This will allow Tomas to continue with his work to use namelocks to syncronize things. Note: valgrind still produces a lot of warnings about using not initialized code and shows memory loss errors when running the ndb tests BitKeeper/etc/ignore: added mysql-test/r/*.log client/mysqltest.c: Change type of variables to get rid of compiler warnings More debugging Fix memory leak mysql-test/mysql-test-run.sh: Collect warnings about missing DBUG_RETURN statements mysql-test/r/lock_multi.result: Add test of new code mysql-test/r/ndb_condition_pushdown.result: Drop used tables before test mysql-test/t/lock_multi.test: Add test of new code mysql-test/t/ndb_condition_pushdown.test: Drop used tables before test mysql-test/valgrind.supp: Ignore 'safe' warnings from libz (when used with archive) sql/event.cc: More comments Simplify code Fixed memory leak found by valgrind sql/ha_archive.cc: Remove compiler warnings (Wrong handlerton structure and signed/unsigned comparison) sql/ha_berkeley.cc: Fixed compiler warning sql/ha_blackhole.cc: Fixed compiler warning sql/ha_federated.cc: Fixed compiler warning sql/ha_heap.cc: Fixed compiler warning sql/ha_myisam.cc: Fixed compiler warning sql/ha_myisammrg.cc: Fixed compiler warning sql/ha_ndbcluster.cc: Fixed compiler warnings sql/ha_partition.cc: Fixed compiler warning Fixed error noticed by valgrind in ha_partition::rnd_init() sql/handler.cc: Fixed compiler warning sql/handler.h: Fixed compiler warning sql/item.cc: Fixed compiler warning sql/item_xmlfunc.cc: Fixed warning from valgrind when calling memcpy with wrong address sql/lock.cc: More debugging sql/log.cc: Fixed compiler warning Indentation fixes sql/log.h: Fixed compiler warning sql/mysql_priv.h: Changed prototype for 'drop_locked_tables' sql/opt_range.cc: Indentation fix sql/password.c: Removed compiler warnings sql/set_var.cc: Fixed compiler warning sql/slave.cc: Fixed compiler warning sql/sp_head.cc: Fixed compiler warning sql/sql_acl.cc: Fixed compiler warning sql/sql_analyse.cc: Added missing DBUG_RETURN statements sql/sql_base.cc: Removed de-reference of not initialized pointer More comments drop_locked_tables() changed to not delete tables used for name locking Fixed compiler warnings sql/sql_delete.cc: Fixed usage of not initialized variable (deleted could be referenced in some not common error conditions) sql/sql_parse.cc: Added missing DBUG_VOID_RETURN Simplify code sql/sql_partition.cc: Fixed usage of wrong variable (noticed by valgrind) sql/sql_plugin.cc: Removed compiler warning sql/sql_show.cc: Removed compiler warning sql/sql_table.cc: Ensure that we don't remove locked tables used as name locks from open table cache until unlock_table_names() are called. This was fixed by having drop_locked_name() returning any table used as a name lock so that we can free it in unlock_table_names() This will allow Tomas to continue with his work to use namelocks to syncronize things. Fixed wrong test of 'table_type' (path_length could otherwise be accessed uninitialized) Remove compile warnings about not initialized variables. sql/sql_yacc.yy: Ensure that no_write_to_binlog is properly initialized (Was accessed uninitialized by partition code) sql/table.cc: Removed valgrind warnings (not fatal) Removed compiler warnings sql/tztime.cc: Removed valgrind warning storage/ndb/include/ndbapi/NdbIndexStat.hpp: Removed compiler warning
2006-03-29 13:27:36 +02:00
reap;
connection locker;
drop table t1;
--echo End of 4.1 tests
#
# Bug#9998 MySQL client hangs on USE "database"
Merge mysql.com:/home/my/mysql-4.1 into mysql.com:/home/my/mysql-5.0 BitKeeper/etc/ignore: auto-union BitKeeper/deleted/.del-ctype_cp932.test: Auto merged BitKeeper/deleted/.del-isam.test~834fb0ee8196c445: Auto merged include/thr_lock.h: Auto merged mysql-test/t/alias.test: Auto merged mysql-test/t/alter_table.test: Auto merged mysql-test/t/archive.test: Auto merged mysql-test/t/backup.test: Auto merged mysql-test/t/bool.test: Auto merged mysql-test/t/connect.test: Auto merged mysql-test/t/count_distinct2.test: Auto merged mysql-test/t/create.test: Auto merged mysql-test/t/ctype_many.test: Auto merged mysql-test/t/ctype_ucs_binlog.test: Auto merged mysql-test/t/date_formats.test: Auto merged mysql-test/t/delayed.test: Auto merged mysql-test/t/derived.test: Auto merged mysql-test/t/distinct.test: Auto merged mysql-test/t/drop.test: Auto merged mysql-test/t/endspace.test: Auto merged mysql-test/t/flush.test: Auto merged mysql-test/t/fulltext.test: Auto merged mysql-test/t/fulltext_order_by.test: Auto merged mysql-test/t/func_compress.test: Auto merged mysql-test/t/func_concat.test: Auto merged mysql-test/t/func_date_add.test: Auto merged mysql-test/t/func_equal.test: Auto merged mysql-test/t/func_if.test: Auto merged mysql-test/t/func_sapdb.test: Auto merged mysql-test/t/func_set.test: Auto merged mysql-test/t/func_str.test: Auto merged mysql-test/t/gis-rtree.test: Auto merged mysql-test/t/gis.test: Auto merged mysql-test/t/grant.test: Auto merged mysql-test/t/grant2.test: Auto merged mysql-test/t/grant_cache.test: Auto merged mysql-test/t/heap.test: Auto merged mysql-test/t/heap_btree.test: Auto merged mysql-test/t/heap_hash.test: Auto merged mysql-test/t/init_connect.test: Auto merged mysql-test/t/insert_select.test: Auto merged mysql-test/t/insert_update.test: Auto merged mysql-test/t/key.test: Auto merged mysql-test/t/keywords.test: Auto merged mysql-test/t/limit.test: Auto merged mysql-test/t/lock.test: Auto merged mysql-test/t/lowercase_table.test: Auto merged mysql-test/t/lowercase_table3.test: Auto merged mysql-test/t/merge.test: Auto merged mysql-test/t/mix_innodb_myisam_binlog.test: Auto merged mysql-test/t/myisam.test: Auto merged mysql-test/t/mysqlbinlog2.test: Auto merged mysql-test/t/ndb_alter_table.test: Auto merged mysql-test/t/ndb_autodiscover.test: Auto merged mysql-test/t/ndb_charset.test: Auto merged mysql-test/t/ndb_grant.later: Auto merged mysql-test/t/ndb_index_ordered.test: Auto merged mysql-test/t/ndb_index_unique.test: Auto merged mysql-test/t/ndb_restore.test: Auto merged mysql-test/t/ndb_types.test: Auto merged mysql-test/t/ndb_update.test: Auto merged mysql-test/t/null.test: Auto merged mysql-test/t/null_key.test: Auto merged mysql-test/t/olap.test: Auto merged mysql-test/t/openssl_1.test: Auto merged mysql-test/t/order_by.test: Auto merged mysql-test/t/ps.test: Auto merged mysql-test/t/ps_1general.test: Auto merged mysql-test/t/ps_4heap.test: Auto merged mysql-test/t/ps_5merge.test: Auto merged mysql-test/t/query_cache.test: Auto merged mysql-test/t/replace.test: Auto merged mysql-test/t/row.test: Auto merged mysql-test/t/rpl000001.test: Auto merged mysql-test/t/rpl000015.test: Auto merged mysql-test/t/rpl000017.test: Auto merged mysql-test/t/rpl000018.test: Auto merged mysql-test/t/rpl_EE_error.test: Auto merged mysql-test/t/rpl_change_master.test: Auto merged mysql-test/t/rpl_charset.test: Auto merged mysql-test/t/rpl_create_database.test: Auto merged mysql-test/t/rpl_ddl.test: Auto merged mysql-test/t/rpl_deadlock.test: Auto merged mysql-test/t/rpl_empty_master_crash.test: Auto merged mysql-test/t/rpl_error_ignored_table.test: Auto merged mysql-test/t/rpl_flush_log_loop.test: Auto merged mysql-test/t/rpl_flush_tables.test: Auto merged mysql-test/t/rpl_get_lock.test: Auto merged mysql-test/t/rpl_heap.test: Auto merged mysql-test/t/rpl_loaddata.test: Auto merged mysql-test/t/rpl_loaddata_rule_m.test: Auto merged mysql-test/t/rpl_log.test: Auto merged mysql-test/t/rpl_log_pos.test: Auto merged mysql-test/t/rpl_max_relay_size.test: Auto merged mysql-test/t/rpl_multi_query.test: Auto merged mysql-test/t/rpl_openssl.test: Auto merged mysql-test/t/rpl_redirect.test: Auto merged mysql-test/t/rpl_relayrotate.test: Auto merged mysql-test/t/rpl_replicate_do.test: Auto merged mysql-test/t/rpl_reset_slave.test: Auto merged mysql-test/t/rpl_server_id2.test: Auto merged mysql-test/t/rpl_temporary.test: Auto merged mysql-test/t/rpl_timezone.test: Auto merged mysql-test/t/rpl_user_variables.test: Auto merged mysql-test/t/show_check.test: Auto merged mysql-test/t/subselect.test: Auto merged mysql-test/t/symlink.test: Auto merged mysql-test/t/synchronization.test: Auto merged mysql-test/t/system_mysql_db.test: Auto merged mysql-test/t/system_mysql_db_fix.test: Auto merged mysql-test/t/temp_table.test: Auto merged mysql-test/t/timezone2.test: Auto merged mysql-test/t/timezone_grant.test: Auto merged mysql-test/t/type_float.test: Auto merged mysql-test/t/type_ranges.test: Auto merged mysql-test/t/type_timestamp.test: Auto merged mysql-test/t/union.test: Auto merged mysql-test/t/update.test: Auto merged mysql-test/t/user_var-binlog.test: Auto merged mysql-test/t/warnings.test: Auto merged mysys/thr_lock.c: Auto merged sql/slave.cc: Auto merged sql/sql_base.cc: Auto merged sql/sql_lex.cc: Auto merged sql/sql_select.cc: Auto merged support-files/mysql.spec.sh: Auto merged BitKeeper/deleted/.del-rpl_trunc_binlog.test~961b1f6ac73d37c8: Simple merge mysql-test/r/ps_grant.result: Simple merge mysql-test/t/analyse.test: Simple merge mysql-test/t/auto_increment.test: Simple merge mysql-test/t/bdb.test: Simple merge mysql-test/t/bigint.test: Simple merge mysql-test/t/case.test: Simple merge mysql-test/t/cast.test: Simple merge mysql-test/t/check.test: Simple merge mysql-test/t/count_distinct.test: Simple merge mysql-test/t/ctype_latin1_de.test: Simple merge mysql-test/t/ctype_uca.test: Simple merge mysql-test/t/ctype_ucs.test: Simple merge mysql-test/t/ctype_utf8.test: Simple merge mysql-test/t/delete.test: Simple merge mysql-test/t/flush_block_commit.test: Simple merge mysql-test/t/func_default.test: Simple merge mysql-test/t/func_gconcat.test: Simple merge mysql-test/t/func_group.test: Aligned code with 4.1 mysql-test/t/func_in.test: Simple merge mysql-test/t/func_math.test: Simple merge mysql-test/t/func_misc.test: Simple merge mysql-test/t/func_test.test: Simple merge mysql-test/t/func_time.test: Simple merge mysql-test/t/group_by.test: Simple merge mysql-test/t/having.test: Simple merge mysql-test/t/innodb.test: Simple merge mysql-test/t/insert.test: Simple merge mysql-test/t/join_outer.test: Simple merge mysql-test/t/kill.test: Simple merge mysql-test/t/loaddata.test: Simple merge mysql-test/t/lock_multi.test: Simple merge mysql-test/t/multi_update.test: Simple merge mysql-test/t/mysqlbinlog.test: Simple merge mysql-test/t/mysqldump.test: Aligned code with 4.1 mysql-test/t/mysqltest.test: Simple merge mysql-test/t/ndb_basic.test: Simple merge mysql-test/t/ndb_cache.test: Simple merge mysql-test/t/ndb_subquery.test: Simple merge mysql-test/t/ps_grant.test: Simple merge mysql-test/t/range.test: Simple merge mysql-test/t/rpl_drop_temp.test: Simple merge mysql-test/t/rpl_loaddata_rule_s.test: Simple merge mysql-test/t/rpl_loaddatalocal.test: Simple merge mysql-test/t/rpl_rotate_logs.test: Simple merge mysql-test/t/rpl_until.test: Simple merge mysql-test/t/rpl_variables.test: Simple merge mysql-test/t/select.test: Simple merge mysql-test/t/sql_mode.test: Simple merge mysql-test/t/type_blob.test: Simple merge mysql-test/t/type_decimal.test: Simple merge mysql-test/t/user_var.test: Simple merge mysql-test/t/variables.test: Simple merge sql/lock.cc: Simple optimization sql/mysql_priv.h: Simple merge sql/sql_table.cc: Simple merge sql/table.cc: Simple merge sql/unireg.cc: Simple merge
2005-07-28 16:09:54 +02:00
#
create table t1(a int);
lock tables t1 write;
connection reader;
show columns from t1;
connection locker;
unlock tables;
drop table t1;
#
# Bug#16986 Deadlock condition with MyISAM tables
#
# Need a matching user in mysql.user for multi-table select
--source include/add_anonymous_users.inc
connection locker;
USE mysql;
LOCK TABLES columns_priv WRITE, db WRITE, host WRITE, user WRITE;
FLUSH TABLES;
#
connection reader;
USE mysql;
# Note: This must be a multi-table select, otherwise the deadlock will not occur
send
SELECT user.Select_priv FROM user, db WHERE user.user = db.user LIMIT 1;
#
connection locker;
# Sleep a bit till the select of connection reader is in work and hangs
let $wait_condition=
Backport of: ------------------------------------------------------------ revno: 3035.4.1 committer: Davi Arnaut <Davi.Arnaut@Sun.COM> branch nick: 39897-6.0 timestamp: Thu 2009-01-15 12:17:57 -0200 message: Bug#39897: lock_multi fails in pushbuild: timeout waiting for processlist The problem is that relying on the "Table lock" thread state in its current position to detect that a thread is waiting on a lock is race prone. The "Table lock" state change happens before the thread actually tries to grab a lock on a table. The solution is to move the "Table lock" state so that its set only when a thread is actually going to wait for a lock. The state change happens after the thread fails to grab the lock (because it is owned by other thread) and proceeds to wait on a condition. This is considered part of work related to WL#4284 "Transactional DDL locking" Warning: this patch contains an incompatible change. When waiting on a lock in thr_lock.c, the server used to display "Locked" processlist state. After this patch, the state is "Table lock". The new state was actually intended to be display since year 2002, when Monty added it. But up until removal of thd->locked boolean member, this state was ignored by SHOW PROCESSLIST code. mysql-test/r/lock_multi.result: A style fix. mysql-test/r/sp-threads.result: Changed output of SHOW PROCESSLIST (new wait state). mysql-test/t/lock_multi.test: Use a more accurate state description when waiting inside thr_lock.c. mysql-test/t/lock_sync.test: Use a more accurate state description when waiting inside thr_lock.c. mysql-test/t/multi_update.test: Use a more accurate state description when waiting inside thr_lock.c. mysql-test/t/query_cache_28249.test: Use a more accurate state description when waiting inside thr_lock.c. mysql-test/t/sp_notembedded.test: Use a more accurate state description when waiting inside thr_lock.c. mysql-test/t/status.test: Use a more accurate state description when waiting inside thr_lock.c. mysys/thr_lock.c: Update thread state while waiting for a table lock. sql/lock.cc: State change was moved inside thr_lock.c.
2009-12-03 21:08:27 +01:00
SELECT COUNT(*) = 1 FROM information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
WHERE state = "Waiting for table metadata lock" AND info =
"SELECT user.Select_priv FROM user, db WHERE user.user = db.user LIMIT 1";
--source include/wait_condition.inc
# Make test case independent from earlier grants.
--replace_result "Table is already up to date" "OK"
OPTIMIZE TABLES columns_priv, db, host, user;
UNLOCK TABLES;
#
connection reader;
reap;
USE test;
#
connection locker;
use test;
#
connection default;
Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. The stress test suite revealed some deadlocks. Some were related to the new condition variable (COND_global_read_lock) and some were general problems with the global read lock. It is now necessary to signal COND_global_read_lock whenever COND_refresh is signalled. We need to wait for the release of a global read lock if one is set before every operation that requires a write lock. But we must not wait if we have locked tables by LOCK TABLES. After setting a global read lock a thread waits until all write locks are released. mysql-test/r/lock_multi.result: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added test results. mysql-test/t/lock_multi.test: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added tests for possible deadlocks that did not occur with the stress test suite. mysys/thr_lock.c: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added a protection against an infinite loop that occurs with the test case for Bug #20662. sql/lock.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). Added the definition of a new function that signals COND_global_read_lock whenever COND_refresh is signalled. sql/mysql_priv.h: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added a declaration for a new function that signals COND_global_read_lock whenever COND_refresh is signalled. sql/sql_base.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). sql/sql_handler.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). sql/sql_insert.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Removed global read lock handling from inside of INSERT DELAYED. It is handled on a higher level now. sql/sql_parse.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Wait for the release of a global read lock if one is set before every operation that requires a write lock. But don't wait if locked tables exist already. sql/sql_table.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Removed global read lock handling from inside of CREATE TABLE. It is handled on a higher level now. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh().
2006-06-26 19:14:35 +02:00
#
# Test if CREATE TABLE with LOCK TABLE deadlocks.
#
connection writer;
CREATE TABLE t1 (c1 int);
LOCK TABLE t1 WRITE;
#
# This waits until t1 is unlocked.
connection locker;
send
FLUSH TABLES WITH READ LOCK;
Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. The stress test suite revealed some deadlocks. Some were related to the new condition variable (COND_global_read_lock) and some were general problems with the global read lock. It is now necessary to signal COND_global_read_lock whenever COND_refresh is signalled. We need to wait for the release of a global read lock if one is set before every operation that requires a write lock. But we must not wait if we have locked tables by LOCK TABLES. After setting a global read lock a thread waits until all write locks are released. mysql-test/r/lock_multi.result: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added test results. mysql-test/t/lock_multi.test: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added tests for possible deadlocks that did not occur with the stress test suite. mysys/thr_lock.c: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added a protection against an infinite loop that occurs with the test case for Bug #20662. sql/lock.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). Added the definition of a new function that signals COND_global_read_lock whenever COND_refresh is signalled. sql/mysql_priv.h: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added a declaration for a new function that signals COND_global_read_lock whenever COND_refresh is signalled. sql/sql_base.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). sql/sql_handler.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). sql/sql_insert.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Removed global read lock handling from inside of INSERT DELAYED. It is handled on a higher level now. sql/sql_parse.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Wait for the release of a global read lock if one is set before every operation that requires a write lock. But don't wait if locked tables exist already. sql/sql_table.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Removed global read lock handling from inside of CREATE TABLE. It is handled on a higher level now. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh().
2006-06-26 19:14:35 +02:00
#
connection writer;
# Sleep a bit till the flush of connection locker is in work and hangs
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
where state = "Waiting for global read lock" and
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
info = "FLUSH TABLES WITH READ LOCK";
--source include/wait_condition.inc
# This must not block.
--error ER_TABLE_NOT_LOCKED
Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. The stress test suite revealed some deadlocks. Some were related to the new condition variable (COND_global_read_lock) and some were general problems with the global read lock. It is now necessary to signal COND_global_read_lock whenever COND_refresh is signalled. We need to wait for the release of a global read lock if one is set before every operation that requires a write lock. But we must not wait if we have locked tables by LOCK TABLES. After setting a global read lock a thread waits until all write locks are released. mysql-test/r/lock_multi.result: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added test results. mysql-test/t/lock_multi.test: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added tests for possible deadlocks that did not occur with the stress test suite. mysys/thr_lock.c: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added a protection against an infinite loop that occurs with the test case for Bug #20662. sql/lock.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). Added the definition of a new function that signals COND_global_read_lock whenever COND_refresh is signalled. sql/mysql_priv.h: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added a declaration for a new function that signals COND_global_read_lock whenever COND_refresh is signalled. sql/sql_base.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). sql/sql_handler.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). sql/sql_insert.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Removed global read lock handling from inside of INSERT DELAYED. It is handled on a higher level now. sql/sql_parse.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Wait for the release of a global read lock if one is set before every operation that requires a write lock. But don't wait if locked tables exist already. sql/sql_table.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Removed global read lock handling from inside of CREATE TABLE. It is handled on a higher level now. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh().
2006-06-26 19:14:35 +02:00
CREATE TABLE t2 (c1 int);
UNLOCK TABLES;
#
# This awakes now.
connection locker;
reap;
UNLOCK TABLES;
#
connection default;
DROP TABLE t1;
Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. The stress test suite revealed some deadlocks. Some were related to the new condition variable (COND_global_read_lock) and some were general problems with the global read lock. It is now necessary to signal COND_global_read_lock whenever COND_refresh is signalled. We need to wait for the release of a global read lock if one is set before every operation that requires a write lock. But we must not wait if we have locked tables by LOCK TABLES. After setting a global read lock a thread waits until all write locks are released. mysql-test/r/lock_multi.result: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added test results. mysql-test/t/lock_multi.test: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added tests for possible deadlocks that did not occur with the stress test suite. mysys/thr_lock.c: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added a protection against an infinite loop that occurs with the test case for Bug #20662. sql/lock.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). Added the definition of a new function that signals COND_global_read_lock whenever COND_refresh is signalled. sql/mysql_priv.h: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added a declaration for a new function that signals COND_global_read_lock whenever COND_refresh is signalled. sql/sql_base.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). sql/sql_handler.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). sql/sql_insert.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Removed global read lock handling from inside of INSERT DELAYED. It is handled on a higher level now. sql/sql_parse.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Wait for the release of a global read lock if one is set before every operation that requires a write lock. But don't wait if locked tables exist already. sql/sql_table.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Removed global read lock handling from inside of CREATE TABLE. It is handled on a higher level now. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh().
2006-06-26 19:14:35 +02:00
#
# Test if CREATE TABLE SELECT with LOCK TABLE deadlocks.
#
connection writer;
CREATE TABLE t1 (c1 int);
LOCK TABLE t1 WRITE;
#
# This waits until t1 is unlocked.
connection locker;
send
FLUSH TABLES WITH READ LOCK;
Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. The stress test suite revealed some deadlocks. Some were related to the new condition variable (COND_global_read_lock) and some were general problems with the global read lock. It is now necessary to signal COND_global_read_lock whenever COND_refresh is signalled. We need to wait for the release of a global read lock if one is set before every operation that requires a write lock. But we must not wait if we have locked tables by LOCK TABLES. After setting a global read lock a thread waits until all write locks are released. mysql-test/r/lock_multi.result: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added test results. mysql-test/t/lock_multi.test: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added tests for possible deadlocks that did not occur with the stress test suite. mysys/thr_lock.c: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added a protection against an infinite loop that occurs with the test case for Bug #20662. sql/lock.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). Added the definition of a new function that signals COND_global_read_lock whenever COND_refresh is signalled. sql/mysql_priv.h: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added a declaration for a new function that signals COND_global_read_lock whenever COND_refresh is signalled. sql/sql_base.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). sql/sql_handler.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). sql/sql_insert.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Removed global read lock handling from inside of INSERT DELAYED. It is handled on a higher level now. sql/sql_parse.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Wait for the release of a global read lock if one is set before every operation that requires a write lock. But don't wait if locked tables exist already. sql/sql_table.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Removed global read lock handling from inside of CREATE TABLE. It is handled on a higher level now. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh().
2006-06-26 19:14:35 +02:00
#
# This must not block.
connection writer;
# Sleep a bit till the flush of connection locker is in work and hangs
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
where state = "Waiting for global read lock" and
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
info = "FLUSH TABLES WITH READ LOCK";
--source include/wait_condition.inc
--error ER_TABLE_NOT_LOCKED
Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. The stress test suite revealed some deadlocks. Some were related to the new condition variable (COND_global_read_lock) and some were general problems with the global read lock. It is now necessary to signal COND_global_read_lock whenever COND_refresh is signalled. We need to wait for the release of a global read lock if one is set before every operation that requires a write lock. But we must not wait if we have locked tables by LOCK TABLES. After setting a global read lock a thread waits until all write locks are released. mysql-test/r/lock_multi.result: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added test results. mysql-test/t/lock_multi.test: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added tests for possible deadlocks that did not occur with the stress test suite. mysys/thr_lock.c: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added a protection against an infinite loop that occurs with the test case for Bug #20662. sql/lock.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). Added the definition of a new function that signals COND_global_read_lock whenever COND_refresh is signalled. sql/mysql_priv.h: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Added a declaration for a new function that signals COND_global_read_lock whenever COND_refresh is signalled. sql/sql_base.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). sql/sql_handler.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh(). sql/sql_insert.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Removed global read lock handling from inside of INSERT DELAYED. It is handled on a higher level now. sql/sql_parse.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Wait for the release of a global read lock if one is set before every operation that requires a write lock. But don't wait if locked tables exist already. sql/sql_table.cc: Bug#16986 - Deadlock condition with MyISAM tables Addendum fixes after changing the condition variable for the global read lock. Removed global read lock handling from inside of CREATE TABLE. It is handled on a higher level now. Signal COND_global_read_lock whenever COND_refresh is signalled by using the new function broadcast_refresh().
2006-06-26 19:14:35 +02:00
CREATE TABLE t2 AS SELECT * FROM t1;
UNLOCK TABLES;
#
# This awakes now.
connection locker;
reap;
UNLOCK TABLES;
#
connection default;
DROP TABLE t1;
--source include/delete_anonymous_users.inc
2006-07-04 12:34:23 +02:00
#
# Bug#19815 CREATE/RENAME/DROP DATABASE can deadlock on a global read lock
2006-07-04 12:34:23 +02:00
#
connect (con1,localhost,root,,);
connect (con2,localhost,root,,);
#
connection con1;
CREATE DATABASE mysqltest_1;
FLUSH TABLES WITH READ LOCK;
#
# With bug in place: acquire LOCK_mysql_create_table and
# wait in wait_if_global_read_lock().
connection con2;
send
DROP DATABASE mysqltest_1;
2006-07-04 12:34:23 +02:00
#
# With bug in place: try to acquire LOCK_mysql_create_table...
# When fixed: Reject dropping db because of the read lock.
connection con1;
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
# Wait a bit so that the session con2 is in state
# "Waiting for global read lock"
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
where state = "Waiting for global read lock"
and info = "DROP DATABASE mysqltest_1";
--source include/wait_condition.inc
2006-07-04 12:34:23 +02:00
--error ER_CANT_UPDATE_WITH_READLOCK
DROP DATABASE mysqltest_1;
UNLOCK TABLES;
#
connection con2;
reap;
#
connection default;
disconnect con1;
disconnect con2;
# This must have been dropped by connection 2 already,
# which waited until the global read lock was released.
--error ER_DB_DROP_EXISTS
DROP DATABASE mysqltest_1;
#
# Bug#17264 MySQL Server freeze
#
connection locker;
# Disable warnings to allow test to run also without InnoDB
set sql_mode="";
--disable_warnings
create table t1 (f1 int(12) unsigned not null auto_increment, primary key(f1)) engine=innodb;
--enable_warnings
set sql_mode=default;
lock tables t1 write;
connection writer;
send
alter table t1 auto_increment=0;
connection reader;
# Wait till connection writer is blocked
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
where state = "Waiting for table metadata lock" and
info = "alter table t1 auto_increment=0";
--source include/wait_condition.inc
send
alter table t1 auto_increment=0;
connection locker;
# Wait till connection reader is blocked
let $wait_condition=
select count(*) = 2 from information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
where state = "Waiting for table metadata lock" and
info = "alter table t1 auto_increment=0";
--source include/wait_condition.inc
unlock tables;
connection writer;
reap;
connection reader;
reap;
connection locker;
drop table t1;
#
# Bug#43230: SELECT ... FOR UPDATE can hang with FLUSH TABLES WITH READ LOCK indefinitely
#
connect (con1,localhost,root,,);
connect (con2,localhost,root,,);
connect (con3,localhost,root,,);
connect (con4,localhost,root,,);
connect (con5,localhost,root,,);
create table t1 (a int);
create table t2 like t1;
connection con1;
lock tables t1 write;
connection con2;
send flush tables with read lock;
connection con5;
Initial import of WL#3726 "DDL locking for all metadata objects". Backport of: ------------------------------------------------------------ revno: 2630.4.1 committer: Dmitry Lenev <dlenev@mysql.com> branch nick: mysql-6.0-3726-w timestamp: Fri 2008-05-23 17:54:03 +0400 message: WL#3726 "DDL locking for all metadata objects". After review fixes in progress. ------------------------------------------------------------ This is the first patch in series. It transforms the metadata locking subsystem to use a dedicated module (mdl.h,cc). No significant changes in the locking protocol. The import passes the test suite with the exception of deprecated/removed 6.0 features, and MERGE tables. The latter are subject to a fix by WL#4144. Unfortunately, the original changeset comments got lost in a merge, thus this import has its own (largely insufficient) comments. This patch fixes Bug#25144 "replication / binlog with view breaks". Warning: this patch introduces an incompatible change: Under LOCK TABLES, it's no longer possible to FLUSH a table that was not locked for WRITE. Under LOCK TABLES, it's no longer possible to DROP a table or VIEW that was not locked for WRITE. ****** Backport of: ------------------------------------------------------------ revno: 2630.4.2 committer: Dmitry Lenev <dlenev@mysql.com> branch nick: mysql-6.0-3726-w timestamp: Sat 2008-05-24 14:03:45 +0400 message: WL#3726 "DDL locking for all metadata objects". After review fixes in progress. ****** Backport of: ------------------------------------------------------------ revno: 2630.4.3 committer: Dmitry Lenev <dlenev@mysql.com> branch nick: mysql-6.0-3726-w timestamp: Sat 2008-05-24 14:08:51 +0400 message: WL#3726 "DDL locking for all metadata objects" Fixed failing Windows builds by adding mdl.cc to the lists of files needed to build server/libmysqld on Windows. ****** Backport of: ------------------------------------------------------------ revno: 2630.4.4 committer: Dmitry Lenev <dlenev@mysql.com> branch nick: mysql-6.0-3726-w timestamp: Sat 2008-05-24 21:57:58 +0400 message: WL#3726 "DDL locking for all metadata objects". Fix for assert failures in kill.test which occured when one tried to kill ALTER TABLE statement on merge table while it was waiting in wait_while_table_is_used() for other connections to close this table. These assert failures stemmed from the fact that cleanup code in this case assumed that temporary table representing new version of table was open with adding to THD::temporary_tables list while code which were opening this temporary table wasn't always fulfilling this. This patch changes code that opens new version of table to always do this linking in. It also streamlines cleanup process for cases when error occurs while we have new version of table open. ****** WL#3726 "DDL locking for all metadata objects" Add libmysqld/mdl.cc to .bzrignore. ****** Backport of: ------------------------------------------------------------ revno: 2630.4.6 committer: Dmitry Lenev <dlenev@mysql.com> branch nick: mysql-6.0-3726-w timestamp: Sun 2008-05-25 00:33:22 +0400 message: WL#3726 "DDL locking for all metadata objects". Addition to the fix of assert failures in kill.test caused by changes for this worklog. Make sure we close the new table only once. .bzrignore: Add libmysqld/mdl.cc libmysqld/CMakeLists.txt: Added mdl.cc to the list of files needed for building of libmysqld. libmysqld/Makefile.am: Added files implementing new meta-data locking subsystem to the server. mysql-test/include/handler.inc: Use separate connection for waiting while threads performing DDL operations conflicting with open HANDLER tables reach blocked state. This is required because now we check and close tables open by HANDLER statements in this connection conflicting with DDL in another each time open_tables() is called and thus select from I_S which is used for waiting will unblock DDL operations if issued from connection with open HANDLERs. mysql-test/r/create.result: Adjusted test case after change in implementation of CREATE TABLE ... SELECT. We no longer have special check in open_table() which catches the case when we select from the table created. Instead we rely on unique_table() call which happens after opening and locking all tables. mysql-test/r/flush.result: FLUSH TABLES WITH READ LOCK can no longer happen under LOCK TABLES. Updated test accordingly. mysql-test/r/flush_table.result: Under LOCK TABLES we no longer allow to do FLUSH TABLES for tables locked for read. Updated test accordingly. mysql-test/r/handler_innodb.result: Use separate connection for waiting while threads performing DDL operations conflicting with open HANDLER tables reach blocked state. This is required because now we check and close tables open by HANDLER statements in this connection conflicting with DDL in another each time open_tables() is called and thus select from I_S which is used for waiting will unblock DDL operations if issued from connection with open HANDLERs. mysql-test/r/handler_myisam.result: Use separate connection for waiting while threads performing DDL operations conflicting with open HANDLER tables reach blocked state. This is required because now we check and close tables open by HANDLER statements in this connection conflicting with DDL in another each time open_tables() is called and thus select from I_S which is used for waiting will unblock DDL operations if issued from connection with open HANDLERs. mysql-test/r/information_schema.result: Additional test for WL#3726 "DDL locking for all metadata objects". Check that we use high-priority metadata lock requests when filling I_S tables. Rearrange tests to match 6.0 better (fewer merge conflicts). mysql-test/r/kill.result: Added tests checking that DDL and DML statements waiting for metadata locks can be interrupted by KILL command. mysql-test/r/lock.result: One no longer is allowed to do DROP VIEW under LOCK TABLES even if this view is locked by LOCK TABLES. The problem is that in such situation write locks on view are not mutually exclusive so upgrading metadata lock which is required for dropping of view will lead to deadlock. mysql-test/r/partition_column_prune.result: Update results (same results in 6.0), WL#3726 mysql-test/r/partition_pruning.result: Update results (same results in 6.0), WL#3726 mysql-test/r/ps_ddl.result: We no longer invalidate prepared CREATE TABLE ... SELECT statement if target table changes. This is OK since it is not strictly necessary. The first change is wrong, is caused by FLUSH TABLE now flushing all unused tables. This is a regression that Dmitri fixed in 6.0 in a follow up patch. mysql-test/r/sp.result: Under LOCK TABLES we no longer allow accessing views which were not explicitly locked. To access view we need to obtain metadata lock on it and doing this under LOCK TABLES may lead to deadlocks. mysql-test/r/view.result: One no longer is allowed to do DROP VIEW under LOCK TABLES even if this view is locked by LOCK TABLES. The problem is that in such situation even "write locks" on view are not mutually exclusive so upgrading metadata lock which is required for dropping of view will lead to deadlock mysql-test/r/view_grant.result: ALTER VIEW implementation was changed to open a view only after checking that user which does alter has appropriate privileges on it. This means that in case when user's privileges are insufficient for this we won't check that new view definer is the same as original one or user performing alter has SUPER privilege. Adjusted test case accordingly. mysql-test/r/view_multi.result: Added test case for bug#25144 "replication / binlog with view breaks". mysql-test/suite/rpl/t/disabled.def: Disable test for deprecated features (they don't work with new MDL). mysql-test/t/create.test: Adjusted test case after change in implementation of CREATE TABLE ... SELECT. We no longer have special check in open_table() which catches the case when we select from the table created. Instead we rely on unique_table() call which happens after opening and locking all tables. mysql-test/t/disabled.def: Disable merge.test, subject of WL#4144 mysql-test/t/flush.test: FLUSH TABLES WITH READ LOCK can no longer happen under LOCK TABLES. Updated test accordingly. mysql-test/t/flush_table.test: Under LOCK TABLES we no longer allow to do FLUSH TABLES for tables locked for read. Updated test accordingly. mysql-test/t/information_schema.test: Additional test for WL#3726 "DDL locking for all metadata objects". Check that we use high-priority metadata lock requests when filling I_S tables. Rearrange the results for easier merges with 6.0. mysql-test/t/kill.test: Added tests checking that DDL and DML statements waiting for metadata locks can be interrupted by KILL command. mysql-test/t/lock.test: One no longer is allowed to do DROP VIEW under LOCK TABLES even if this view is locked by LOCK TABLES. The problem is that in such situation write locks on view are not mutually exclusive so upgrading metadata lock which is required for dropping of view will lead to deadlock. mysql-test/t/lock_multi.test: Adjusted test case to the changes of status in various places caused by change in implementation FLUSH TABLES WITH READ LOCK, which is now takes global metadata lock before flushing tables and therefore waits on at these places. mysql-test/t/ps_ddl.test: We no longer invalidate prepared CREATE TABLE ... SELECT statement if target table changes. This is OK since it is not strictly necessary. The first change is wrong, is caused by FLUSH TABLE now flushing all unused tables. This is a regression that Dmitri fixed in 6.0 in a follow up patch. mysql-test/t/sp.test: Under LOCK TABLES we no longer allow accessing views which were not explicitly locked. To access view we need to obtain metadata lock on it and doing this under LOCK TABLES may lead to deadlocks. mysql-test/t/trigger_notembedded.test: Adjusted test case to the changes of status in various places caused by change in implementation FLUSH TABLES WITH READ LOCK, which is now takes global metadata lock before flushing tables and therefore waits on at these places. mysql-test/t/view.test: One no longer is allowed to do DROP VIEW under LOCK TABLES even if this view is locked by LOCK TABLES. The problem is that in such situation even "write locks" on view are not mutually exclusive so upgrading metadata lock which is required for dropping of view will lead to deadlock. mysql-test/t/view_grant.test: ALTER VIEW implementation was changed to open a view only after checking that user which does alter has appropriate privileges on it. This means that in case when user's privileges are insufficient for this we won't check that new view definer is the same as original one or user performing alter has SUPER privilege. Adjusted test case accordingly. mysql-test/t/view_multi.test: Added test case for bug#25144 "replication / binlog with view breaks". sql/CMakeLists.txt: Added mdl.cc to the list of files needed for building of server. sql/Makefile.am: Added files implementing new meta-data locking subsystem to the server. sql/event_db_repository.cc: Allocate metadata lock requests objects (MDL_LOCK) on execution memory root in cases when TABLE_LIST objects is also allocated there or on stack. sql/ha_ndbcluster.cc: Adjusted code to work nicely with new metadata locking subsystem. close_cached_tables() no longer has wait_for_placeholder argument. Instead of relying on this parameter and related behavior FLUSH TABLES WITH READ LOCK now takes global shared metadata lock. sql/ha_ndbcluster_binlog.cc: Adjusted code to work with new metadata locking subsystem. close_cached_tables() no longer has wait_for_placeholder argument. Instead of relying on this parameter and related behavior FLUSH TABLES WITH READ LOCK now takes global shared metadata lock. sql/handler.cc: update_frm_version(): Directly update TABLE_SHARE::mysql_version member instead of going through all TABLE instances for this table (old code was a legacy from pre-table-definition-cache days). sql/lock.cc: Use new metadata locking subsystem. Threw away most of functions related to name locking as now one is supposed to use metadata locking API instead. In lock_global_read_lock() and unlock_global_read_lock() in order to avoid problems with global read lock sneaking in at the moment when we perform FLUSH TABLES or ALTER TABLE under LOCK TABLES and when tables being reopened are protected only by metadata locks we also have to take global shared meta data lock. sql/log_event.cc: Adjusted code to work with new metadata locking subsystem. For tables open by slave thread for applying RBR events allocate memory for lock request object in the same chunk of memory as TABLE_LIST objects for them. In order to ensure that we keep these objects around until tables are open always close tables before calling Relay_log_info::clear_tables_to_lock(). Use new auxiliary Relay_log_info::slave_close_thread_tables() method to enforce this. sql/log_event_old.cc: Adjusted code to work with new metadata locking subsystem. Since for tables open by slave thread for applying RBR events memory for lock request object is allocated in the same chunk of memory as TABLE_LIST objects for them we have to ensure that we keep these objects around until tables are open. To ensure this we always close tables before calling Relay_log_info::clear_tables_to_lock(). To enfore this we use new auxiliary Relay_log_info::slave_close_thread_tables() method. sql/mdl.cc: Implemented new metadata locking subsystem and API described in WL3726 "DDL locking for all metadata objects". sql/mdl.h: Implemented new metadata locking subsystem and API described in WL3726 "DDL locking for all metadata objects". sql/mysql_priv.h: - close_thread_tables()/close_tables_for_reopen() now has one more argument which indicates that metadata locks should be released but not removed from the context in order to be used later in mdl_wait_for_locks() and tdc_wait_for_old_version(). - close_cached_table() routine is no longer public. - Thread waiting in wait_while_table_is_used() can be now killed so this function returns boolean to make caller aware of such situation. - We no longer have table cache as separate entity instead used and unused TABLE instances are linked to TABLE_SHARE objects in table definition cache. - Now third argument of open_table() is also used for requesting table repair or auto-discovery of table's new definition. So its type was changed from bool to enum. - Added tdc_open_view() function for opening view by getting its definition from disk (and table cache in future). - reopen_name_locked_table() no longer needs "link_in" argument as now we have exclusive metadata locks instead of dummy TABLE instances when this function is called. - find_locked_table() now takes head of list of TABLE instances instead of always scanning through THD::open_tables list. Also added find_write_locked_table() auxiliary. - reopen_tables(), close_cached_tables() no longer have mark_share_as_old and wait_for_placeholder arguments. Instead of relying on this parameters and related behavior FLUSH TABLES WITH READ LOCK now takes global shared metadata lock. - We no longer need drop_locked_tables() and abort_locked_tables(). - mysql_ha_rm_tables() now always assume that LOCK_open is not acquired by caller. - Added notify_thread_having_shared_lock() callback invoked by metadata locking subsystem when acquiring an exclusive lock, for each thread that has a conflicting shared metadata lock. - Introduced expel_table_from_cache() as replacement for remove_table_from_cache() (the main difference is that this new function assumes that caller follows metadata locking protocol and never waits). - Threw away most of functions related to name locking. One should use new metadata locking subsystem and API instead. sql/mysqld.cc: Got rid of call initializing/deinitializing table cache since now it is embedded into table definition cache. Added calls for initializing/ deinitializing metadata locking subsystem. sql/rpl_rli.cc: Introduced auxiliary Relay_log_info::slave_close_thread_tables() method which is used for enforcing that we always close tables open for RBR before deallocating TABLE_LIST elements and MDL_LOCK objects for them. sql/rpl_rli.h: Introduced auxiliary Relay_log_info::slave_close_thread_tables() method which is used for enforcing that we always close tables open for RBR before deallocating TABLE_LIST elements and MDL_LOCK objects for them. sql/set_var.cc: close_cached_tables() no longer has wait_for_placeholder argument. Instead of relying on this parameter and related behavior FLUSH TABLES WITH READ LOCK now takes global shared metadata lock. sql/sp_head.cc: For tables added to the statement's table list by prelocking algorithm we allocate these objects either on the same memory as corresponding table list elements or on THD::locked_tables_root (if we are building table list for LOCK TABLES). sql/sql_acl.cc: Allocate metadata lock requests objects (MDL_LOCK) on execution memory root in cases when we use stack TABLE_LIST objects to open tables. Got rid of redundant code by using unlock_locked_tables() function. sql/sql_base.cc: Changed code to use new MDL subsystem. Got rid of separate table cache. Now used and unused TABLE instances are linked to the TABLE_SHAREs in table definition cache. check_unused(): Adjusted code to the fact that we no longer have separate table cache. Removed dead code. table_def_free(): Free TABLE instances referenced from TABLE_SHARE objects before destroying table definition cache. get_table_share(): Added assert which ensures that noone will be able to access table (and its share) without acquiring some kind of metadata lock first. close_handle_and_leave_table_as_lock(): Adjusted code to the fact that TABLE instances now are linked to list in TABLE_SHARE. list_open_tables(): Changed this function to use table definition cache instead of table cache. free_cache_entry(): Unlink freed TABLE elements from the list of all TABLE instances for the table in TABLE_SHARE. kill_delayed_thread_for_table(): Added auxiliary for killing delayed insert threads for particular table. close_cached_tables(): Got rid of wait_for_refresh argument as we now rely on global shared metadata lock to prevent FLUSH WITH READ LOCK sneaking in when we are reopening tables. Heavily reworked this function to use new MDL code and not to rely on separate table cache entity. close_open_tables(): We no longer have separate table cache. close_thread_tables(): Release metadata locks after closing all tables. Added skip_mdl argument which allows us not to remove metadata lock requests from the context in case when we are going to use this requests later in mdl_wait_for_locks() and tdc_wait_for_old_versions(). close_thread_table()/close_table_for_reopen(): Since we no longer have separate table cache and all TABLE instances are linked to TABLE_SHARE objects in table definition cache we have to link/unlink TABLE object to/from appropriate lists in the share. name_lock_locked_table(): Moved redundant code to find_write_locked_table() function and adjusted code to the fact that wait_while_table_is_used() can now return with an error if our thread is killed. reopen_table_entry(): We no longer need "link_in" argument as with MDL we no longer call this function with dummy TABLE object pre-allocated and added to the THD::open_tables. Also now we add newly-open TABLE instance to the list of share's used TABLE instances. table_cache_insert_placeholder(): Got rid of name-locking legacy. lock_table_name_if_not_cached(): Moved to sql_table.cc the only place where it is used. It was also reimplemented using new MDL API. open_table(): - Reworked this function to use new MDL subsystem. - Changed code to deal with table definition cache directly instead of going through separate table cache. - Now third argument is also used for requesting table repair or auto-discovery of table's new definition. So its type was changed from bool to enum. find_locked_table()/find_write_locked_table(): Accept head of list of TABLE objects as first argument and use this list instead of always searching in THD::open_tables list. Also added auxiliary for finding write-locked locked tables. reopen_table(): Adjusted function to work with new MDL subsystem and to properly manuipulate with lists of used/unused TABLE instaces in TABLE_SHARE. reopen_tables(): Removed mark_share_as_old parameter. Instead of relying on it and related behavior FLUSH TABLES WITH READ LOCK now takes global shared metadata lock. Changed code after removing separate table cache. drop_locked_tables()/abort_locked_tables(): Got rid of functions which are no longer needed. unlock_locked_tables(): Moved this function from sql_parse.cc and changed it to release memory which was used for allocating metadata lock requests for tables open and locked by LOCK TABLES. tdc_open_view(): Intoduced function for opening a view by getting its definition from disk (and table cache in future). reopen_table_entry(): Introduced function for opening table definitions while holding exclusive metatadata lock on it. open_unireg_entry(): Got rid of this function. Most of its functionality is relocated to open_table() and open_table_fini() functions, and some of it to reopen_table_entry() and tdc_open_view(). Also code resposible for auto-repair and auto-discovery of tables was moved to separate function. open_table_entry_fini(): Introduced function which contains common actions which finalize process of TABLE object creation. auto_repair_table(): Moved code responsible for auto-repair of table being opened here. handle_failed_open_table_attempt() Moved code responsible for handling failing attempt to open table to one place (retry due to lock conflict/old version, auto-discovery and repair). open_tables(): - Flush open HANDLER tables if they have old version of if there is conflicting metadata lock against them (before this moment we had this code in open_table()). - When we open view which should be processed via derived table on the second execution of prepared statement or stored routine we still should call open_table() for it in order to obtain metadata lock on it and prepare its security context. - In cases when we discover that some special handling of failure to open table is needed call handle_failed_open_table_attempt() which handles all such scenarios. open_ltable(): Handling of various special scenarios of failure to open a table was moved to separate handle_failed_open_table_attempt() function. remove_db_from_cache(): Removed this function as it is no longer used. notify_thread_having_shared_lock(): Added callback which is invoked by MDL subsystem when acquiring an exclusive lock, for each thread that has a conflicting shared metadata lock. expel_table_from_cache(): Introduced function for removing unused TABLE instances. Unlike remove_table_from_cache() it relies on caller following MDL protocol and having appropriate locks when calling it and thus does not do any waiting if table is still in use. tdc_wait_for_old_version(): Added function which allows open_tables() to wait in cases when we discover that we should back-off due to presence of old version of table. abort_and_upgrade_lock(): Use new MDL calls. mysql_wait_completed_table(): Got rid of unused function. open_system_tables_for_read/for_update()/performance_schema_table(): Allocate MDL_LOCK objects on execution memory root in cases when TABLE_LIST objects for corresponding tables is allocated on stack. close_performance_schema_table(): Release metadata locks after closing tables. ****** Use I_P_List for free/used tables list in the table share. sql/sql_binlog.cc: Use Relay_log_info::slave_close_thread_tables() method to enforce that we always close tables open for RBR before deallocating TABLE_LIST elements and MDL_LOCK objects for them. sql/sql_class.cc: Added meta-data locking contexts as part of Open_tables_state context. Also introduced THD::locked_tables_root memory root which is to be used for allocating MDL_LOCK objects for tables in LOCK TABLES statement (end of lifetime for such objects is UNLOCK TABLES so we can't use statement or execution root for them). sql/sql_class.h: Added meta-data locking contexts as part of Open_tables_state context. Also introduced THD::locked_tables_root memory root which is to be used for allocating MDL_LOCK objects for tables in LOCK TABLES statement (end of lifetime for such objects is UNLOCK TABLES so we can't use statement or execution root for them). Note: handler_mdl_context and locked_tables_root and mdl_el_root will be removed by subsequent patches. sql/sql_db.cc: mysql_rm_db() does not really need to call remove_db_from_cache() as it drops each table in the database using mysql_rm_table_part2(), which performs all necessary operations on table (definition) cache. sql/sql_delete.cc: Use the new metadata locking API for TRUNCATE. sql/sql_handler.cc: Changed HANDLER implementation to use new metadata locking subsystem. Note that MDL_LOCK objects for HANDLER tables are allocated in the same chunk of heap memory as TABLE_LIST object for those tables. sql/sql_insert.cc: mysql_insert(): find_locked_table() now takes head of list of TABLE object as its argument instead of always scanning through THD::open_tables list. handle_delayed_insert(): Allocate metadata lock request object for table open by delayed insert thread on execution memroot. create_table_from_items(): We no longer allocate dummy TABLE objects for tables being created if they don't exist. As consequence reopen_name_locked_table() no longer has link_in argument. open_table() now has one more argument which is not relevant for temporary tables. sql/sql_parse.cc: - Moved unlock_locked_tables() routine to sql_base.cc and made available it in other files. Got rid of some redundant code by using this function. - Replaced boolean TABLE_LIST::create member with enum open_table_type member. - Use special memory root for allocating MDL_LOCK objects for tables open and locked by LOCK TABLES (these object should live till UNLOCK TABLES so we can't allocate them on statement nor execution memory root). Also properly set metadata lock upgradability attribure for those tables. - Under LOCK TABLES it is no longer allowed to flush tables which are not write-locked as this breaks metadata locking protocol and thus potentially might lead to deadlock. - Added auxiliary adjust_mdl_locks_upgradability() function. sql/sql_partition.cc: Adjusted code to the fact that reopen_tables() no longer has "mark_share_as_old" argument. Got rid of comments which are no longer true. sql/sql_plist.h: Added I_P_List template class for parametrized intrusive doubly linked lists and I_P_List_iterator for corresponding iterator. Unlike for I_List<> list elements of such list can participate in several lists. Unlike List<> such lists are doubly-linked and intrusive. sql/sql_plugin.cc: Allocate metadata lock requests objects (MDL_LOCK) on execution memory root in cases when we use stack TABLE_LIST objects to open tables. sql/sql_prepare.cc: Replaced boolean TABLE_LIST::create member with enum open_table_type member. This allows easily handle situation in which instead of opening the table we want only to take exclusive metadata lock on it. sql/sql_rename.cc: Use new metadata locking subsystem in implementation of RENAME TABLE. sql/sql_servers.cc: Allocate metadata lock requests objects (MDL_LOCK) on execution memory root in cases when we use stack TABLE_LIST objects to open tables. Got rid of redundant code by using unlock_locked_tables() function. sql/sql_show.cc: Acquire shared metadata lock when we are getting information for I_S table directly from TABLE_SHARE without doing full-blown table open. We use high priority lock request in this situation in order to avoid deadlocks. Also allocate metadata lock requests objects (MDL_LOCK) on execution memory root in cases when TABLE_LIST objects are also allocated there sql/sql_table.cc: mysql_rm_table(): Removed comment which is no longer relevant. mysql_rm_table_part2(): Now caller of mysql_ha_rm_tables() should not own LOCK_open. Adjusted code to use new metadata locking subsystem instead of name-locks. lock_table_name_if_not_cached(): Moved this function from sql_base.cc to this file and reimplemented it using metadata locking API. mysql_create_table(): Adjusted code to use new MDL API. wait_while_table_is_used(): Changed function to use new MDL subsystem. Made thread waiting in it killable (this also led to introduction of return value so caller can distinguish successful executions from situations when waiting was aborted). close_cached_tables(): Thread waiting in this function is killable now. As result it has return value for distinguishing between succes and failure. Got rid of redundant boradcast_refresh() call. prepare_for_repair(): Use MDL subsystem instead of name-locks. mysql_admin_table(): mysql_ha_rm_tables() now always assumes that caller doesn't own LOCK_open. mysql_repair_table(): We should mark all elements of table list as requiring upgradable metadata locks. mysql_create_table_like(): Use new MDL subsystem instead of name-locks. create_temporary_tables(): We don't need to obtain metadata locks when creating temporary table. mysql_fast_or_online_alter_table(): Thread waiting in wait_while_table_is_used() is now killable. mysql_alter_table(): Adjusted code to work with new MDL subsystem and to the fact that threads waiting in what_while_table_is_used() and close_cached_table() are now killable. sql/sql_test.cc: We no longer have separate table cache. TABLE instances are now associated with/linked to TABLE_SHARE objects in table definition cache. sql/sql_trigger.cc: Adjusted code to work with new metadata locking subsystem. Also reopen_tables() no longer has mark_share_as_old argument (Instead of relying on this parameter and related behavior FLUSH TABLES WITH READ LOCK now takes global shared metadata lock). sql/sql_udf.cc: Allocate metadata lock requests objects (MDL_LOCK) on execution memory root in cases when we use stack TABLE_LIST objects to open tables. sql/sql_update.cc: Adjusted code to work with new meta-data locking subsystem. sql/sql_view.cc: Added proper meta-data locking to implementations of CREATE/ALTER/DROP VIEW statements. Now we obtain exclusive meta-data lock on a view before creating/ changing/dropping it. This ensures that all concurrent statements that use this view will finish before our statement will proceed and therefore we will get correct order of statements in the binary log. Also ensure that TABLE_LIST::mdl_upgradable attribute is properly propagated for underlying tables of view. sql/table.cc: Added auxiliary alloc_mdl_locks() function for allocating metadata lock request objects for all elements of table list. sql/table.h: TABLE_SHARE: Got rid of unused members. Introduced members for storing lists of used and unused TABLE objects for this share. TABLE: Added members for linking TABLE objects into per-share lists of used and unused TABLE instances. Added member for holding pointer to metadata lock for this table. TABLE_LIST: Replaced boolean TABLE_LIST::create member with enum open_table_type member. This allows easily handle situation in which instead of opening the table we want only to take exclusive meta-data lock on it (we need this in order to handle ALTER VIEW and CREATE VIEW statements). Introduced new mdl_upgradable member for marking elements of table list for which we need to take upgradable shared metadata lock instead of plain shared metadata lock. Added pointer for holding pointer to MDL_LOCK for the table. Added auxiliary alloc_mdl_locks() function for allocating metadata lock requests objects for all elements of table list. Added auxiliary set_all_mdl_upgradable() function for marking all elements in table list as requiring upgradable metadata locks. storage/myisammrg/ha_myisammrg.cc: Allocate MDL_LOCK objects for underlying tables of MERGE table. To be reworked once Ingo pushes his patch for WL4144.
2009-11-30 16:55:03 +01:00
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
where state = "Waiting for global read lock" and
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
info = "flush tables with read lock";
Initial import of WL#3726 "DDL locking for all metadata objects". Backport of: ------------------------------------------------------------ revno: 2630.4.1 committer: Dmitry Lenev <dlenev@mysql.com> branch nick: mysql-6.0-3726-w timestamp: Fri 2008-05-23 17:54:03 +0400 message: WL#3726 "DDL locking for all metadata objects". After review fixes in progress. ------------------------------------------------------------ This is the first patch in series. It transforms the metadata locking subsystem to use a dedicated module (mdl.h,cc). No significant changes in the locking protocol. The import passes the test suite with the exception of deprecated/removed 6.0 features, and MERGE tables. The latter are subject to a fix by WL#4144. Unfortunately, the original changeset comments got lost in a merge, thus this import has its own (largely insufficient) comments. This patch fixes Bug#25144 "replication / binlog with view breaks". Warning: this patch introduces an incompatible change: Under LOCK TABLES, it's no longer possible to FLUSH a table that was not locked for WRITE. Under LOCK TABLES, it's no longer possible to DROP a table or VIEW that was not locked for WRITE. ****** Backport of: ------------------------------------------------------------ revno: 2630.4.2 committer: Dmitry Lenev <dlenev@mysql.com> branch nick: mysql-6.0-3726-w timestamp: Sat 2008-05-24 14:03:45 +0400 message: WL#3726 "DDL locking for all metadata objects". After review fixes in progress. ****** Backport of: ------------------------------------------------------------ revno: 2630.4.3 committer: Dmitry Lenev <dlenev@mysql.com> branch nick: mysql-6.0-3726-w timestamp: Sat 2008-05-24 14:08:51 +0400 message: WL#3726 "DDL locking for all metadata objects" Fixed failing Windows builds by adding mdl.cc to the lists of files needed to build server/libmysqld on Windows. ****** Backport of: ------------------------------------------------------------ revno: 2630.4.4 committer: Dmitry Lenev <dlenev@mysql.com> branch nick: mysql-6.0-3726-w timestamp: Sat 2008-05-24 21:57:58 +0400 message: WL#3726 "DDL locking for all metadata objects". Fix for assert failures in kill.test which occured when one tried to kill ALTER TABLE statement on merge table while it was waiting in wait_while_table_is_used() for other connections to close this table. These assert failures stemmed from the fact that cleanup code in this case assumed that temporary table representing new version of table was open with adding to THD::temporary_tables list while code which were opening this temporary table wasn't always fulfilling this. This patch changes code that opens new version of table to always do this linking in. It also streamlines cleanup process for cases when error occurs while we have new version of table open. ****** WL#3726 "DDL locking for all metadata objects" Add libmysqld/mdl.cc to .bzrignore. ****** Backport of: ------------------------------------------------------------ revno: 2630.4.6 committer: Dmitry Lenev <dlenev@mysql.com> branch nick: mysql-6.0-3726-w timestamp: Sun 2008-05-25 00:33:22 +0400 message: WL#3726 "DDL locking for all metadata objects". Addition to the fix of assert failures in kill.test caused by changes for this worklog. Make sure we close the new table only once. .bzrignore: Add libmysqld/mdl.cc libmysqld/CMakeLists.txt: Added mdl.cc to the list of files needed for building of libmysqld. libmysqld/Makefile.am: Added files implementing new meta-data locking subsystem to the server. mysql-test/include/handler.inc: Use separate connection for waiting while threads performing DDL operations conflicting with open HANDLER tables reach blocked state. This is required because now we check and close tables open by HANDLER statements in this connection conflicting with DDL in another each time open_tables() is called and thus select from I_S which is used for waiting will unblock DDL operations if issued from connection with open HANDLERs. mysql-test/r/create.result: Adjusted test case after change in implementation of CREATE TABLE ... SELECT. We no longer have special check in open_table() which catches the case when we select from the table created. Instead we rely on unique_table() call which happens after opening and locking all tables. mysql-test/r/flush.result: FLUSH TABLES WITH READ LOCK can no longer happen under LOCK TABLES. Updated test accordingly. mysql-test/r/flush_table.result: Under LOCK TABLES we no longer allow to do FLUSH TABLES for tables locked for read. Updated test accordingly. mysql-test/r/handler_innodb.result: Use separate connection for waiting while threads performing DDL operations conflicting with open HANDLER tables reach blocked state. This is required because now we check and close tables open by HANDLER statements in this connection conflicting with DDL in another each time open_tables() is called and thus select from I_S which is used for waiting will unblock DDL operations if issued from connection with open HANDLERs. mysql-test/r/handler_myisam.result: Use separate connection for waiting while threads performing DDL operations conflicting with open HANDLER tables reach blocked state. This is required because now we check and close tables open by HANDLER statements in this connection conflicting with DDL in another each time open_tables() is called and thus select from I_S which is used for waiting will unblock DDL operations if issued from connection with open HANDLERs. mysql-test/r/information_schema.result: Additional test for WL#3726 "DDL locking for all metadata objects". Check that we use high-priority metadata lock requests when filling I_S tables. Rearrange tests to match 6.0 better (fewer merge conflicts). mysql-test/r/kill.result: Added tests checking that DDL and DML statements waiting for metadata locks can be interrupted by KILL command. mysql-test/r/lock.result: One no longer is allowed to do DROP VIEW under LOCK TABLES even if this view is locked by LOCK TABLES. The problem is that in such situation write locks on view are not mutually exclusive so upgrading metadata lock which is required for dropping of view will lead to deadlock. mysql-test/r/partition_column_prune.result: Update results (same results in 6.0), WL#3726 mysql-test/r/partition_pruning.result: Update results (same results in 6.0), WL#3726 mysql-test/r/ps_ddl.result: We no longer invalidate prepared CREATE TABLE ... SELECT statement if target table changes. This is OK since it is not strictly necessary. The first change is wrong, is caused by FLUSH TABLE now flushing all unused tables. This is a regression that Dmitri fixed in 6.0 in a follow up patch. mysql-test/r/sp.result: Under LOCK TABLES we no longer allow accessing views which were not explicitly locked. To access view we need to obtain metadata lock on it and doing this under LOCK TABLES may lead to deadlocks. mysql-test/r/view.result: One no longer is allowed to do DROP VIEW under LOCK TABLES even if this view is locked by LOCK TABLES. The problem is that in such situation even "write locks" on view are not mutually exclusive so upgrading metadata lock which is required for dropping of view will lead to deadlock mysql-test/r/view_grant.result: ALTER VIEW implementation was changed to open a view only after checking that user which does alter has appropriate privileges on it. This means that in case when user's privileges are insufficient for this we won't check that new view definer is the same as original one or user performing alter has SUPER privilege. Adjusted test case accordingly. mysql-test/r/view_multi.result: Added test case for bug#25144 "replication / binlog with view breaks". mysql-test/suite/rpl/t/disabled.def: Disable test for deprecated features (they don't work with new MDL). mysql-test/t/create.test: Adjusted test case after change in implementation of CREATE TABLE ... SELECT. We no longer have special check in open_table() which catches the case when we select from the table created. Instead we rely on unique_table() call which happens after opening and locking all tables. mysql-test/t/disabled.def: Disable merge.test, subject of WL#4144 mysql-test/t/flush.test: FLUSH TABLES WITH READ LOCK can no longer happen under LOCK TABLES. Updated test accordingly. mysql-test/t/flush_table.test: Under LOCK TABLES we no longer allow to do FLUSH TABLES for tables locked for read. Updated test accordingly. mysql-test/t/information_schema.test: Additional test for WL#3726 "DDL locking for all metadata objects". Check that we use high-priority metadata lock requests when filling I_S tables. Rearrange the results for easier merges with 6.0. mysql-test/t/kill.test: Added tests checking that DDL and DML statements waiting for metadata locks can be interrupted by KILL command. mysql-test/t/lock.test: One no longer is allowed to do DROP VIEW under LOCK TABLES even if this view is locked by LOCK TABLES. The problem is that in such situation write locks on view are not mutually exclusive so upgrading metadata lock which is required for dropping of view will lead to deadlock. mysql-test/t/lock_multi.test: Adjusted test case to the changes of status in various places caused by change in implementation FLUSH TABLES WITH READ LOCK, which is now takes global metadata lock before flushing tables and therefore waits on at these places. mysql-test/t/ps_ddl.test: We no longer invalidate prepared CREATE TABLE ... SELECT statement if target table changes. This is OK since it is not strictly necessary. The first change is wrong, is caused by FLUSH TABLE now flushing all unused tables. This is a regression that Dmitri fixed in 6.0 in a follow up patch. mysql-test/t/sp.test: Under LOCK TABLES we no longer allow accessing views which were not explicitly locked. To access view we need to obtain metadata lock on it and doing this under LOCK TABLES may lead to deadlocks. mysql-test/t/trigger_notembedded.test: Adjusted test case to the changes of status in various places caused by change in implementation FLUSH TABLES WITH READ LOCK, which is now takes global metadata lock before flushing tables and therefore waits on at these places. mysql-test/t/view.test: One no longer is allowed to do DROP VIEW under LOCK TABLES even if this view is locked by LOCK TABLES. The problem is that in such situation even "write locks" on view are not mutually exclusive so upgrading metadata lock which is required for dropping of view will lead to deadlock. mysql-test/t/view_grant.test: ALTER VIEW implementation was changed to open a view only after checking that user which does alter has appropriate privileges on it. This means that in case when user's privileges are insufficient for this we won't check that new view definer is the same as original one or user performing alter has SUPER privilege. Adjusted test case accordingly. mysql-test/t/view_multi.test: Added test case for bug#25144 "replication / binlog with view breaks". sql/CMakeLists.txt: Added mdl.cc to the list of files needed for building of server. sql/Makefile.am: Added files implementing new meta-data locking subsystem to the server. sql/event_db_repository.cc: Allocate metadata lock requests objects (MDL_LOCK) on execution memory root in cases when TABLE_LIST objects is also allocated there or on stack. sql/ha_ndbcluster.cc: Adjusted code to work nicely with new metadata locking subsystem. close_cached_tables() no longer has wait_for_placeholder argument. Instead of relying on this parameter and related behavior FLUSH TABLES WITH READ LOCK now takes global shared metadata lock. sql/ha_ndbcluster_binlog.cc: Adjusted code to work with new metadata locking subsystem. close_cached_tables() no longer has wait_for_placeholder argument. Instead of relying on this parameter and related behavior FLUSH TABLES WITH READ LOCK now takes global shared metadata lock. sql/handler.cc: update_frm_version(): Directly update TABLE_SHARE::mysql_version member instead of going through all TABLE instances for this table (old code was a legacy from pre-table-definition-cache days). sql/lock.cc: Use new metadata locking subsystem. Threw away most of functions related to name locking as now one is supposed to use metadata locking API instead. In lock_global_read_lock() and unlock_global_read_lock() in order to avoid problems with global read lock sneaking in at the moment when we perform FLUSH TABLES or ALTER TABLE under LOCK TABLES and when tables being reopened are protected only by metadata locks we also have to take global shared meta data lock. sql/log_event.cc: Adjusted code to work with new metadata locking subsystem. For tables open by slave thread for applying RBR events allocate memory for lock request object in the same chunk of memory as TABLE_LIST objects for them. In order to ensure that we keep these objects around until tables are open always close tables before calling Relay_log_info::clear_tables_to_lock(). Use new auxiliary Relay_log_info::slave_close_thread_tables() method to enforce this. sql/log_event_old.cc: Adjusted code to work with new metadata locking subsystem. Since for tables open by slave thread for applying RBR events memory for lock request object is allocated in the same chunk of memory as TABLE_LIST objects for them we have to ensure that we keep these objects around until tables are open. To ensure this we always close tables before calling Relay_log_info::clear_tables_to_lock(). To enfore this we use new auxiliary Relay_log_info::slave_close_thread_tables() method. sql/mdl.cc: Implemented new metadata locking subsystem and API described in WL3726 "DDL locking for all metadata objects". sql/mdl.h: Implemented new metadata locking subsystem and API described in WL3726 "DDL locking for all metadata objects". sql/mysql_priv.h: - close_thread_tables()/close_tables_for_reopen() now has one more argument which indicates that metadata locks should be released but not removed from the context in order to be used later in mdl_wait_for_locks() and tdc_wait_for_old_version(). - close_cached_table() routine is no longer public. - Thread waiting in wait_while_table_is_used() can be now killed so this function returns boolean to make caller aware of such situation. - We no longer have table cache as separate entity instead used and unused TABLE instances are linked to TABLE_SHARE objects in table definition cache. - Now third argument of open_table() is also used for requesting table repair or auto-discovery of table's new definition. So its type was changed from bool to enum. - Added tdc_open_view() function for opening view by getting its definition from disk (and table cache in future). - reopen_name_locked_table() no longer needs "link_in" argument as now we have exclusive metadata locks instead of dummy TABLE instances when this function is called. - find_locked_table() now takes head of list of TABLE instances instead of always scanning through THD::open_tables list. Also added find_write_locked_table() auxiliary. - reopen_tables(), close_cached_tables() no longer have mark_share_as_old and wait_for_placeholder arguments. Instead of relying on this parameters and related behavior FLUSH TABLES WITH READ LOCK now takes global shared metadata lock. - We no longer need drop_locked_tables() and abort_locked_tables(). - mysql_ha_rm_tables() now always assume that LOCK_open is not acquired by caller. - Added notify_thread_having_shared_lock() callback invoked by metadata locking subsystem when acquiring an exclusive lock, for each thread that has a conflicting shared metadata lock. - Introduced expel_table_from_cache() as replacement for remove_table_from_cache() (the main difference is that this new function assumes that caller follows metadata locking protocol and never waits). - Threw away most of functions related to name locking. One should use new metadata locking subsystem and API instead. sql/mysqld.cc: Got rid of call initializing/deinitializing table cache since now it is embedded into table definition cache. Added calls for initializing/ deinitializing metadata locking subsystem. sql/rpl_rli.cc: Introduced auxiliary Relay_log_info::slave_close_thread_tables() method which is used for enforcing that we always close tables open for RBR before deallocating TABLE_LIST elements and MDL_LOCK objects for them. sql/rpl_rli.h: Introduced auxiliary Relay_log_info::slave_close_thread_tables() method which is used for enforcing that we always close tables open for RBR before deallocating TABLE_LIST elements and MDL_LOCK objects for them. sql/set_var.cc: close_cached_tables() no longer has wait_for_placeholder argument. Instead of relying on this parameter and related behavior FLUSH TABLES WITH READ LOCK now takes global shared metadata lock. sql/sp_head.cc: For tables added to the statement's table list by prelocking algorithm we allocate these objects either on the same memory as corresponding table list elements or on THD::locked_tables_root (if we are building table list for LOCK TABLES). sql/sql_acl.cc: Allocate metadata lock requests objects (MDL_LOCK) on execution memory root in cases when we use stack TABLE_LIST objects to open tables. Got rid of redundant code by using unlock_locked_tables() function. sql/sql_base.cc: Changed code to use new MDL subsystem. Got rid of separate table cache. Now used and unused TABLE instances are linked to the TABLE_SHAREs in table definition cache. check_unused(): Adjusted code to the fact that we no longer have separate table cache. Removed dead code. table_def_free(): Free TABLE instances referenced from TABLE_SHARE objects before destroying table definition cache. get_table_share(): Added assert which ensures that noone will be able to access table (and its share) without acquiring some kind of metadata lock first. close_handle_and_leave_table_as_lock(): Adjusted code to the fact that TABLE instances now are linked to list in TABLE_SHARE. list_open_tables(): Changed this function to use table definition cache instead of table cache. free_cache_entry(): Unlink freed TABLE elements from the list of all TABLE instances for the table in TABLE_SHARE. kill_delayed_thread_for_table(): Added auxiliary for killing delayed insert threads for particular table. close_cached_tables(): Got rid of wait_for_refresh argument as we now rely on global shared metadata lock to prevent FLUSH WITH READ LOCK sneaking in when we are reopening tables. Heavily reworked this function to use new MDL code and not to rely on separate table cache entity. close_open_tables(): We no longer have separate table cache. close_thread_tables(): Release metadata locks after closing all tables. Added skip_mdl argument which allows us not to remove metadata lock requests from the context in case when we are going to use this requests later in mdl_wait_for_locks() and tdc_wait_for_old_versions(). close_thread_table()/close_table_for_reopen(): Since we no longer have separate table cache and all TABLE instances are linked to TABLE_SHARE objects in table definition cache we have to link/unlink TABLE object to/from appropriate lists in the share. name_lock_locked_table(): Moved redundant code to find_write_locked_table() function and adjusted code to the fact that wait_while_table_is_used() can now return with an error if our thread is killed. reopen_table_entry(): We no longer need "link_in" argument as with MDL we no longer call this function with dummy TABLE object pre-allocated and added to the THD::open_tables. Also now we add newly-open TABLE instance to the list of share's used TABLE instances. table_cache_insert_placeholder(): Got rid of name-locking legacy. lock_table_name_if_not_cached(): Moved to sql_table.cc the only place where it is used. It was also reimplemented using new MDL API. open_table(): - Reworked this function to use new MDL subsystem. - Changed code to deal with table definition cache directly instead of going through separate table cache. - Now third argument is also used for requesting table repair or auto-discovery of table's new definition. So its type was changed from bool to enum. find_locked_table()/find_write_locked_table(): Accept head of list of TABLE objects as first argument and use this list instead of always searching in THD::open_tables list. Also added auxiliary for finding write-locked locked tables. reopen_table(): Adjusted function to work with new MDL subsystem and to properly manuipulate with lists of used/unused TABLE instaces in TABLE_SHARE. reopen_tables(): Removed mark_share_as_old parameter. Instead of relying on it and related behavior FLUSH TABLES WITH READ LOCK now takes global shared metadata lock. Changed code after removing separate table cache. drop_locked_tables()/abort_locked_tables(): Got rid of functions which are no longer needed. unlock_locked_tables(): Moved this function from sql_parse.cc and changed it to release memory which was used for allocating metadata lock requests for tables open and locked by LOCK TABLES. tdc_open_view(): Intoduced function for opening a view by getting its definition from disk (and table cache in future). reopen_table_entry(): Introduced function for opening table definitions while holding exclusive metatadata lock on it. open_unireg_entry(): Got rid of this function. Most of its functionality is relocated to open_table() and open_table_fini() functions, and some of it to reopen_table_entry() and tdc_open_view(). Also code resposible for auto-repair and auto-discovery of tables was moved to separate function. open_table_entry_fini(): Introduced function which contains common actions which finalize process of TABLE object creation. auto_repair_table(): Moved code responsible for auto-repair of table being opened here. handle_failed_open_table_attempt() Moved code responsible for handling failing attempt to open table to one place (retry due to lock conflict/old version, auto-discovery and repair). open_tables(): - Flush open HANDLER tables if they have old version of if there is conflicting metadata lock against them (before this moment we had this code in open_table()). - When we open view which should be processed via derived table on the second execution of prepared statement or stored routine we still should call open_table() for it in order to obtain metadata lock on it and prepare its security context. - In cases when we discover that some special handling of failure to open table is needed call handle_failed_open_table_attempt() which handles all such scenarios. open_ltable(): Handling of various special scenarios of failure to open a table was moved to separate handle_failed_open_table_attempt() function. remove_db_from_cache(): Removed this function as it is no longer used. notify_thread_having_shared_lock(): Added callback which is invoked by MDL subsystem when acquiring an exclusive lock, for each thread that has a conflicting shared metadata lock. expel_table_from_cache(): Introduced function for removing unused TABLE instances. Unlike remove_table_from_cache() it relies on caller following MDL protocol and having appropriate locks when calling it and thus does not do any waiting if table is still in use. tdc_wait_for_old_version(): Added function which allows open_tables() to wait in cases when we discover that we should back-off due to presence of old version of table. abort_and_upgrade_lock(): Use new MDL calls. mysql_wait_completed_table(): Got rid of unused function. open_system_tables_for_read/for_update()/performance_schema_table(): Allocate MDL_LOCK objects on execution memory root in cases when TABLE_LIST objects for corresponding tables is allocated on stack. close_performance_schema_table(): Release metadata locks after closing tables. ****** Use I_P_List for free/used tables list in the table share. sql/sql_binlog.cc: Use Relay_log_info::slave_close_thread_tables() method to enforce that we always close tables open for RBR before deallocating TABLE_LIST elements and MDL_LOCK objects for them. sql/sql_class.cc: Added meta-data locking contexts as part of Open_tables_state context. Also introduced THD::locked_tables_root memory root which is to be used for allocating MDL_LOCK objects for tables in LOCK TABLES statement (end of lifetime for such objects is UNLOCK TABLES so we can't use statement or execution root for them). sql/sql_class.h: Added meta-data locking contexts as part of Open_tables_state context. Also introduced THD::locked_tables_root memory root which is to be used for allocating MDL_LOCK objects for tables in LOCK TABLES statement (end of lifetime for such objects is UNLOCK TABLES so we can't use statement or execution root for them). Note: handler_mdl_context and locked_tables_root and mdl_el_root will be removed by subsequent patches. sql/sql_db.cc: mysql_rm_db() does not really need to call remove_db_from_cache() as it drops each table in the database using mysql_rm_table_part2(), which performs all necessary operations on table (definition) cache. sql/sql_delete.cc: Use the new metadata locking API for TRUNCATE. sql/sql_handler.cc: Changed HANDLER implementation to use new metadata locking subsystem. Note that MDL_LOCK objects for HANDLER tables are allocated in the same chunk of heap memory as TABLE_LIST object for those tables. sql/sql_insert.cc: mysql_insert(): find_locked_table() now takes head of list of TABLE object as its argument instead of always scanning through THD::open_tables list. handle_delayed_insert(): Allocate metadata lock request object for table open by delayed insert thread on execution memroot. create_table_from_items(): We no longer allocate dummy TABLE objects for tables being created if they don't exist. As consequence reopen_name_locked_table() no longer has link_in argument. open_table() now has one more argument which is not relevant for temporary tables. sql/sql_parse.cc: - Moved unlock_locked_tables() routine to sql_base.cc and made available it in other files. Got rid of some redundant code by using this function. - Replaced boolean TABLE_LIST::create member with enum open_table_type member. - Use special memory root for allocating MDL_LOCK objects for tables open and locked by LOCK TABLES (these object should live till UNLOCK TABLES so we can't allocate them on statement nor execution memory root). Also properly set metadata lock upgradability attribure for those tables. - Under LOCK TABLES it is no longer allowed to flush tables which are not write-locked as this breaks metadata locking protocol and thus potentially might lead to deadlock. - Added auxiliary adjust_mdl_locks_upgradability() function. sql/sql_partition.cc: Adjusted code to the fact that reopen_tables() no longer has "mark_share_as_old" argument. Got rid of comments which are no longer true. sql/sql_plist.h: Added I_P_List template class for parametrized intrusive doubly linked lists and I_P_List_iterator for corresponding iterator. Unlike for I_List<> list elements of such list can participate in several lists. Unlike List<> such lists are doubly-linked and intrusive. sql/sql_plugin.cc: Allocate metadata lock requests objects (MDL_LOCK) on execution memory root in cases when we use stack TABLE_LIST objects to open tables. sql/sql_prepare.cc: Replaced boolean TABLE_LIST::create member with enum open_table_type member. This allows easily handle situation in which instead of opening the table we want only to take exclusive metadata lock on it. sql/sql_rename.cc: Use new metadata locking subsystem in implementation of RENAME TABLE. sql/sql_servers.cc: Allocate metadata lock requests objects (MDL_LOCK) on execution memory root in cases when we use stack TABLE_LIST objects to open tables. Got rid of redundant code by using unlock_locked_tables() function. sql/sql_show.cc: Acquire shared metadata lock when we are getting information for I_S table directly from TABLE_SHARE without doing full-blown table open. We use high priority lock request in this situation in order to avoid deadlocks. Also allocate metadata lock requests objects (MDL_LOCK) on execution memory root in cases when TABLE_LIST objects are also allocated there sql/sql_table.cc: mysql_rm_table(): Removed comment which is no longer relevant. mysql_rm_table_part2(): Now caller of mysql_ha_rm_tables() should not own LOCK_open. Adjusted code to use new metadata locking subsystem instead of name-locks. lock_table_name_if_not_cached(): Moved this function from sql_base.cc to this file and reimplemented it using metadata locking API. mysql_create_table(): Adjusted code to use new MDL API. wait_while_table_is_used(): Changed function to use new MDL subsystem. Made thread waiting in it killable (this also led to introduction of return value so caller can distinguish successful executions from situations when waiting was aborted). close_cached_tables(): Thread waiting in this function is killable now. As result it has return value for distinguishing between succes and failure. Got rid of redundant boradcast_refresh() call. prepare_for_repair(): Use MDL subsystem instead of name-locks. mysql_admin_table(): mysql_ha_rm_tables() now always assumes that caller doesn't own LOCK_open. mysql_repair_table(): We should mark all elements of table list as requiring upgradable metadata locks. mysql_create_table_like(): Use new MDL subsystem instead of name-locks. create_temporary_tables(): We don't need to obtain metadata locks when creating temporary table. mysql_fast_or_online_alter_table(): Thread waiting in wait_while_table_is_used() is now killable. mysql_alter_table(): Adjusted code to work with new MDL subsystem and to the fact that threads waiting in what_while_table_is_used() and close_cached_table() are now killable. sql/sql_test.cc: We no longer have separate table cache. TABLE instances are now associated with/linked to TABLE_SHARE objects in table definition cache. sql/sql_trigger.cc: Adjusted code to work with new metadata locking subsystem. Also reopen_tables() no longer has mark_share_as_old argument (Instead of relying on this parameter and related behavior FLUSH TABLES WITH READ LOCK now takes global shared metadata lock). sql/sql_udf.cc: Allocate metadata lock requests objects (MDL_LOCK) on execution memory root in cases when we use stack TABLE_LIST objects to open tables. sql/sql_update.cc: Adjusted code to work with new meta-data locking subsystem. sql/sql_view.cc: Added proper meta-data locking to implementations of CREATE/ALTER/DROP VIEW statements. Now we obtain exclusive meta-data lock on a view before creating/ changing/dropping it. This ensures that all concurrent statements that use this view will finish before our statement will proceed and therefore we will get correct order of statements in the binary log. Also ensure that TABLE_LIST::mdl_upgradable attribute is properly propagated for underlying tables of view. sql/table.cc: Added auxiliary alloc_mdl_locks() function for allocating metadata lock request objects for all elements of table list. sql/table.h: TABLE_SHARE: Got rid of unused members. Introduced members for storing lists of used and unused TABLE objects for this share. TABLE: Added members for linking TABLE objects into per-share lists of used and unused TABLE instances. Added member for holding pointer to metadata lock for this table. TABLE_LIST: Replaced boolean TABLE_LIST::create member with enum open_table_type member. This allows easily handle situation in which instead of opening the table we want only to take exclusive meta-data lock on it (we need this in order to handle ALTER VIEW and CREATE VIEW statements). Introduced new mdl_upgradable member for marking elements of table list for which we need to take upgradable shared metadata lock instead of plain shared metadata lock. Added pointer for holding pointer to MDL_LOCK for the table. Added auxiliary alloc_mdl_locks() function for allocating metadata lock requests objects for all elements of table list. Added auxiliary set_all_mdl_upgradable() function for marking all elements in table list as requiring upgradable metadata locks. storage/myisammrg/ha_myisammrg.cc: Allocate MDL_LOCK objects for underlying tables of MERGE table. To be reworked once Ingo pushes his patch for WL4144.
2009-11-30 16:55:03 +01:00
--source include/wait_condition.inc
--echo # global read lock is taken
connection con3;
send select * from t2 for update;
connection con5;
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
let $wait_condition=
select count(*) = 1 from information_schema.processlist
where state = "Waiting for global read lock" and
info = "select * from t2 for update";
--source include/wait_condition.inc
--echo # waiting for release of read lock
connection con4;
--echo # would hang and later cause a deadlock
flush tables t2;
connection con1;
--echo # clean up
unlock tables;
connection con2;
--reap
unlock tables;
connection con3;
--reap
connection default;
disconnect con5;
disconnect con4;
disconnect con3;
disconnect con2;
disconnect con1;
drop table t1,t2;
--echo #
--echo # Lightweight version:
--echo # Ensure that the wait for a GRL is done before opening tables.
--echo #
connect (con1,localhost,root,,);
connect (con2,localhost,root,,);
create table t1 (a int);
create table t2 like t1;
--echo #
--echo # UPDATE
--echo #
connection default;
flush tables with read lock;
connection con1;
send update t2 set a = 1;
connection default;
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
let $wait_condition=
select count(*) = 1 from information_schema.processlist
where state = "Waiting for global read lock" and
info = "update t2 set a = 1";
--source include/wait_condition.inc
--echo # statement is waiting for release of read lock
connection con2;
flush table t2;
connection default;
unlock tables;
connection con1;
--reap
--echo #
--echo # LOCK TABLES .. WRITE
--echo #
connection default;
flush tables with read lock;
connection con1;
send lock tables t2 write;
connection default;
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
let $wait_condition=
select count(*) = 1 from information_schema.processlist
where state = "Waiting for global read lock" and
info = "lock tables t2 write";
--source include/wait_condition.inc
--echo # statement is waiting for release of read lock
connection con2;
flush table t2;
connection default;
unlock tables;
connection con1;
--reap
unlock tables;
connection default;
disconnect con2;
disconnect con1;
drop table t1,t2;
--echo End of 5.0 tests
#
# Bug#21281 Pending write lock is incorrectly removed when its
# statement being KILLed
#
create table t1 (i int);
connection locker;
lock table t1 read;
connection writer;
send
update t1 set i= 10;
connection reader;
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
where state = "Waiting for table level lock" and
info = "update t1 set i= 10";
--source include/wait_condition.inc
send
select * from t1;
connection default;
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
where state = "Waiting for table level lock" and
info = "select * from t1";
--source include/wait_condition.inc
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
let $ID= `select id from information_schema.processlist
where state = "Waiting for table level lock" and
info = "update t1 set i= 10"`;
--replace_result $ID ID
eval kill query $ID;
connection reader;
--reap
connection writer;
--error ER_QUERY_INTERRUPTED
--reap
connection locker;
unlock tables;
connection default;
drop table t1;
# Disconnect sessions used in many subtests above
disconnect locker;
Implement new type-of-operation-aware metadata locks. Add a wait-for graph based deadlock detector to the MDL subsystem. Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and bug #37346 "innodb does not detect deadlock between update and alter table". The first bug manifested itself as an unwarranted abort of a transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER statement, when this transaction tried to repeat use of a table, which it has already used in a similar fashion before ALTER started. The second bug showed up as a deadlock between table-level locks and InnoDB row locks, which was "detected" only after innodb_lock_wait_timeout timeout. A transaction would start using the table and modify a few rows. Then ALTER TABLE would come in, and start copying rows into a temporary table. Eventually it would stumble on the modified records and get blocked on a row lock. The first transaction would try to do more updates, and get blocked on thr_lock.c lock. This situation of circular wait would only get resolved by a timeout. Both these bugs stemmed from inadequate solutions to the problem of deadlocks occurring between different locking subsystems. In the first case we tried to avoid deadlocks between metadata locking and table-level locking subsystems, when upgrading shared metadata lock to exclusive one. Transactions holding the shared lock on the table and waiting for some table-level lock used to be aborted too aggressively. We also allowed ALTER TABLE to start in presence of transactions that modify the subject table. ALTER TABLE acquires TL_WRITE_ALLOW_READ lock at start, and that block all writes against the table (naturally, we don't want any writes to be lost when switching the old and the new table). TL_WRITE_ALLOW_READ lock, in turn, would block the started transaction on thr_lock.c lock, should they do more updates. This, again, lead to the need to abort such transactions. The second bug occurred simply because we didn't have any mechanism to detect deadlocks between the table-level locks in thr_lock.c and row-level locks in InnoDB, other than innodb_lock_wait_timeout. This patch solves both these problems by moving lock conflicts which are causing these deadlocks into the metadata locking subsystem, thus making it possible to avoid or detect such deadlocks inside MDL. To do this we introduce new type-of-operation-aware metadata locks, which allow MDL subsystem to know not only the fact that transaction has used or is going to use some object but also what kind of operation it has carried out or going to carry out on the object. This, along with the addition of a special kind of upgradable metadata lock, allows ALTER TABLE to wait until all transactions which has updated the table to go away. This solves the second issue. Another special type of upgradable metadata lock is acquired by LOCK TABLE WRITE. This second lock type allows to solve the first issue, since abortion of table-level locks in event of DDL under LOCK TABLES becomes also unnecessary. Below follows the list of incompatible changes introduced by this patch: - From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those statements that acquire TL_WRITE_ALLOW_READ lock) wait for all transactions which has *updated* the table to complete. - From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE (i.e. all statements which acquire TL_WRITE table-level lock) wait for all transaction which *updated or read* from the table to complete. As a consequence, innodb_table_locks=0 option no longer applies to LOCK TABLES ... WRITE. - DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort statements or transactions which use tables being dropped or renamed, and instead wait for these transactions to complete. - Since LOCK TABLES WRITE now takes a special metadata lock, not compatible with with reads or writes against the subject table and transaction-wide, thr_lock.c deadlock avoidance algorithm that used to ensure absence of deadlocks between LOCK TABLES WRITE and other statements is no longer sufficient, even for MyISAM. The wait-for graph based deadlock detector of MDL subsystem may sometimes be necessary and is involved. This may lead to ER_LOCK_DEADLOCK error produced for multi-statement transactions even if these only use MyISAM: session 1: session 2: begin; update t1 ... lock table t2 write, t1 write; -- gets a lock on t2, blocks on t1 update t2 ... (ER_LOCK_DEADLOCK) - Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE was abandoned. LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same priority as the usual LOCK TABLE ... WRITE. SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in the wait queue. - We do not take upgradable metadata locks on implicitly locked tables. So if one has, say, a view v1 that uses table t1, and issues: LOCK TABLE v1 WRITE; FLUSH TABLE t1; -- (or just 'FLUSH TABLES'), an error is produced. In order to be able to perform DDL on a table under LOCK TABLES, the table must be locked explicitly in the LOCK TABLES list. mysql-test/include/handler.inc: Adjusted test case to trigger an execution path on which bug 41110 "crash with handler command when used concurrently with alter table" and bug 41112 "crash in mysql_ha_close_table/get_lock_data with alter table" were originally discovered. Left old test case which no longer triggers this execution path for the sake of coverage. Added test coverage for HANDLER SQL statements and type-aware metadata locks. Added a test for the global shared lock and HANDLER SQL. Updated tests to take into account that the old simple deadlock detection heuristics was replaced with a graph-based deadlock detector. mysql-test/r/debug_sync.result: Updated results (see debug_sync.test). mysql-test/r/handler_innodb.result: Updated results (see handler.inc test). mysql-test/r/handler_myisam.result: Updated results (see handler.inc test). mysql-test/r/innodb-lock.result: Updated results (see innodb-lock.test). mysql-test/r/innodb_mysql_lock.result: Updated results (see innodb_mysql_lock.test). mysql-test/r/lock.result: Updated results (see lock.test). mysql-test/r/lock_multi.result: Updated results (see lock_multi.test). mysql-test/r/lock_sync.result: Updated results (see lock_sync.test). mysql-test/r/mdl_sync.result: Updated results (see mdl_sync.test). mysql-test/r/sp-threads.result: SHOW PROCESSLIST output has changed due to the fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/r/truncate_coverage.result: Updated results (see truncate_coverage.test). mysql-test/suite/funcs_1/datadict/processlist_val.inc: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/funcs_1/r/processlist_val_no_prot.result: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/rpl/t/rpl_sp.test: Updated to a new SHOW PROCESSLIST state name. mysql-test/t/debug_sync.test: Use LOCK TABLES READ instead of LOCK TABLES WRITE as the latter no longer allows to trigger execution path involving waiting on thr_lock.c lock and therefore reaching debug sync-point covered by this test. mysql-test/t/innodb-lock.test: Adjusted test case to the fact that innodb_table_locks=0 option is no longer supported, since LOCK TABLES WRITE handles all its conflicts within MDL subsystem. mysql-test/t/innodb_mysql_lock.test: Added test for bug #37346 "innodb does not detect deadlock between update and alter table". mysql-test/t/lock.test: Added test coverage which checks the fact that we no longer support DDL under LOCK TABLES on tables which were locked implicitly. Adjusted existing test cases accordingly. mysql-test/t/lock_multi.test: Added test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock". Adjusted other test cases to take into account the fact that waiting for LOCK TABLES ... WRITE now happens within MDL subsystem. mysql-test/t/lock_sync.test: Since LOCK TABLES ... WRITE now takes SNRW metadata lock for tables locked explicitly we have to implicitly lock InnoDB tables (through view) to trigger the table-level lock conflict between TL_WRITE and TL_WRITE_ALLOW_WRITE. mysql-test/t/mdl_sync.test: Added basic test coverage for type-of-operation-aware metadata locks. Also covered with tests some use cases involving HANDLER statements in which a deadlock could arise. Adjusted existing tests to take type-of-operation-aware MDL into account. mysql-test/t/multi_update.test: Update to a new SHOW PROCESSLIST state name. mysql-test/t/truncate_coverage.test: Adjusted test case after making LOCK TABLES WRITE to wait until transactions that use the table to be locked are completed. Updated to the changed name of DEBUG_SYNC point. sql/handler.cc: Global read lock functionality has been moved into a class. sql/lock.cc: Global read lock functionality has been moved into a class. Updated code to use the new MDL API. sql/mdl.cc: Introduced new type-of-operation aware metadata locks. To do this: - Changed MDL_lock to use one list for waiting requests and one list for granted requests. For each list, added a bitmap that holds information what lock types a list contains. Added a helper class MDL_lock::List to manipulate with granted and waited lists while keeping the bitmaps in sync with list contents. - Changed lock-compatibility functions to use bitmaps that define compatibility. - Introduced a graph based deadlock detector inspired by waiting_threads.c from Maria implementation. - Now that we have a deadlock detector, and no longer have a global lock to protect individual lock objects, but rather use an rw lock per object, removed redundant code for upgrade, and the global read lock. Changed the MDL API to no longer require the caller to acquire the global intention exclusive lock by means of a separate method. Removed a few more methods that became redundant. - Removed deadlock detection heuristic, it has been made obsolete by the deadlock detector. - With operation-type-aware metadata locks, MDL subsystem has become aware of potential conflicts between DDL and open transactions. This made it possible to remove calls to mysql_abort_transactions_with_shared_lock() from acquisition paths for exclusive lock and lock upgrade. Now we can simply wait for these transactions to complete without fear of deadlock. Function mysql_lock_abort() has also become unnecessary for all conflicting cases except when a DDL conflicts with a connection that has an open HANDLER. sql/mdl.h: Introduced new type-of-operation aware metadata locks. Introduced a graph based deadlock detector and supporting methods. Added comments. God rid of redundant API calls. Renamed m_lt_or_ha_sentinel to m_trans_sentinel, since now it guards the global read lock as well as LOCK TABLES and HANDLER locks. sql/mysql_priv.h: Moved the global read lock functionality into a class. Added MYSQL_OPEN_FORCE_SHARED_MDL flag which forces open_tables() to take MDL_SHARED on tables instead of metadata locks specified in the parser. We use this to allow PREPARE run concurrently in presence of LOCK TABLES ... WRITE. Added signature for find_table_for_mdl_ugprade(). sql/set_var.cc: Global read lock functionality has been moved into a class. sql/sp_head.cc: When creating TABLE_LIST elements for prelocking or system tables set the type of request for metadata lock according to the operation that will be performed on the table. sql/sql_base.cc: - Updated code to use the new MDL API. - In order to avoid locks starvation we take upgradable locks all at once. As result implicitly locked tables no longer get an upgradable lock. Consequently DDL and FLUSH TABLES for such tables is prohibited. find_write_locked_table() was replaced by find_table_for_mdl_upgrade() function. open_table() was adjusted to return TABLE instance with upgradable ticket when necessary. - We no longer wait for all locks on OT_WAIT back off action -- only on the lock that caused the wait conflict. Moreover, now we distinguish cases when we have to wait due to conflict in MDL and old version of table in TDC. - Upate mysql_notify_threads_having_share_locks() to only abort thr_lock.c waits of threads that have open HANDLERs, since lock conflicts with only these threads now can lead to deadlocks not detectable by the MDL deadlock detector. - Remove mysql_abort_transactions_with_shared_locks() which is no longer needed. sql/sql_class.cc: Global read lock functionality has been moved into a class. Re-arranged code in THD::cleanup() to simplify assert. sql/sql_class.h: Introduced class to incapsulate global read lock functionality. Now sentinel in MDL subsystem guards the global read lock as well as LOCK TABLES and HANDLER locks. Adjusted code accordingly. sql/sql_db.cc: Global read lock functionality has been moved into a class. sql/sql_delete.cc: We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result TRUNCATE TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_handler.cc: Inform MDL_context about presence of open HANDLERs. Since HANLDERs break MDL protocol by acquiring table-level lock while holding only S metadata lock on a table MDL subsystem should take special care about such contexts (Now this is the only case when mysql_lock_abort() is used). sql/sql_parse.cc: Global read lock functionality has been moved into a class. Do not take upgradable metadata locks when opening tables for CREATE TABLE SELECT as it is not necessary and limits concurrency. When initializing TABLE_LIST objects before adding them to the table list set the type of request for metadata lock according to the operation that will be performed on the table. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result FLUSH TABLES is no longer allowed for such tables. sql/sql_prepare.cc: Use MYSQL_OPEN_FORCE_SHARED_MDL flag when opening tables during PREPARE. This allows PREPARE to run concurrently in presence of LOCK TABLES ... WRITE. sql/sql_rename.cc: Global read lock functionality has been moved into a class. sql/sql_show.cc: Updated code to use the new MDL API. sql/sql_table.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result DROP TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_trigger.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result CREATE/DROP TRIGGER is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_view.cc: Global read lock functionality has been moved into a class. Fixed results of wrong merge that led to misuse of GLR API. CREATE VIEW statement is not a commit statement. sql/table.cc: When resetting TABLE_LIST objects for PS or SP re-execution set the type of request for metadata lock according to the operation that will be performed on the table. Do the same in auxiliary function initializing metadata lock requests in a table list. sql/table.h: When initializing TABLE_LIST objects set the type of request for metadata lock according to the operation that will be performed on the table. sql/transaction.cc: Global read lock functionality has been moved into a class.
2010-02-01 12:43:06 +01:00
disconnect locker2;
disconnect reader;
disconnect writer;
Bug#32395 Alter table under a impending global read lock causes a server crash The problem is that some DDL statements (ALTER TABLE, CREATE TRIGGER, FLUSH TABLES, ...) when under LOCK TABLES need to momentarily drop the lock, reopen the table and grab the write lock again (using reopen_tables). When grabbing the lock again, reopen_tables doesn't pass a flag to mysql_lock_tables in order to ignore the impending global read lock, which causes a assertion because LOCK_open is being hold. Also dropping the lock must not signal to any threads that the table has been relinquished (related to the locking/flushing protocol). The solution is to correct the way the table is reopenned and the locks grabbed. When reopening the table and under LOCK TABLES, the table version should be set to 0 so other threads have to wait for the table. When grabbing the lock, any other flush should be ignored because it's theoretically a atomic operation. The chosen solution also fixes a potential discrepancy between binlog and GRL (global read lock) because table placeholders were being ignored, now a FLUSH TABLES WITH READ LOCK will properly for table with open placeholders. It's also important to mention that this patch doesn't fix a potential deadlock if one uses two GRLs under LOCK TABLES concurrently. mysql-test/r/lock_multi.result: Add test case result for Bug#32395 mysql-test/r/trigger_notembedded.result: Add test case result for Bug#32395 mysql-test/t/lock_multi.test: Add test case for Bug#32395 mysql-test/t/trigger_notembedded.test: Enable test case for Bug#32395 sql/ha_ndbcluster.cc: Update close_cached_tables usage. sql/ha_ndbcluster_binlog.cc: Update close_cached_tables usage. sql/mysql_priv.h: Update close_cache_tables prototype. sql/set_var.cc: Update close_cached_tables usage and set flag to wait for tables with placeholders. This is one of the places where a GRL can be obtained. sql/sql_base.cc: Preserve old version for write locked tables and ignore pending flushes and update close_cache_tables to take into account name locked tables. sql/sql_parse.cc: Update close_cached_tables usage and pass flag so that name locked tables are waited for. sql/sql_table.cc: Protect the table against a impending GRL if under LOCK TABLES.
2007-12-12 22:44:14 +01:00
#
# Bug#32395 Alter table under a impending global read lock causes a server crash
#
#
# Test ALTER TABLE under LOCK TABLES and FLUSH TABLES WITH READ LOCK
#
--disable_warnings
drop table if exists t1;
--enable_warnings
create table t1 (i int);
connect (flush,localhost,root,,test,,);
connection default;
lock tables t1 write;
connection flush;
--send flush tables with read lock;
connection default;
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
where state = "Waiting for global read lock" and
info = "flush tables with read lock";
Bug#32395 Alter table under a impending global read lock causes a server crash The problem is that some DDL statements (ALTER TABLE, CREATE TRIGGER, FLUSH TABLES, ...) when under LOCK TABLES need to momentarily drop the lock, reopen the table and grab the write lock again (using reopen_tables). When grabbing the lock again, reopen_tables doesn't pass a flag to mysql_lock_tables in order to ignore the impending global read lock, which causes a assertion because LOCK_open is being hold. Also dropping the lock must not signal to any threads that the table has been relinquished (related to the locking/flushing protocol). The solution is to correct the way the table is reopenned and the locks grabbed. When reopening the table and under LOCK TABLES, the table version should be set to 0 so other threads have to wait for the table. When grabbing the lock, any other flush should be ignored because it's theoretically a atomic operation. The chosen solution also fixes a potential discrepancy between binlog and GRL (global read lock) because table placeholders were being ignored, now a FLUSH TABLES WITH READ LOCK will properly for table with open placeholders. It's also important to mention that this patch doesn't fix a potential deadlock if one uses two GRLs under LOCK TABLES concurrently. mysql-test/r/lock_multi.result: Add test case result for Bug#32395 mysql-test/r/trigger_notembedded.result: Add test case result for Bug#32395 mysql-test/t/lock_multi.test: Add test case for Bug#32395 mysql-test/t/trigger_notembedded.test: Enable test case for Bug#32395 sql/ha_ndbcluster.cc: Update close_cached_tables usage. sql/ha_ndbcluster_binlog.cc: Update close_cached_tables usage. sql/mysql_priv.h: Update close_cache_tables prototype. sql/set_var.cc: Update close_cached_tables usage and set flag to wait for tables with placeholders. This is one of the places where a GRL can be obtained. sql/sql_base.cc: Preserve old version for write locked tables and ignore pending flushes and update close_cache_tables to take into account name locked tables. sql/sql_parse.cc: Update close_cached_tables usage and pass flag so that name locked tables are waited for. sql/sql_table.cc: Protect the table against a impending GRL if under LOCK TABLES.
2007-12-12 22:44:14 +01:00
--source include/wait_condition.inc
alter table t1 add column j int;
connect (insert,localhost,root,,test,,);
connection insert;
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
where state = "Waiting for global read lock" and
info = "flush tables with read lock";
Bug#32395 Alter table under a impending global read lock causes a server crash The problem is that some DDL statements (ALTER TABLE, CREATE TRIGGER, FLUSH TABLES, ...) when under LOCK TABLES need to momentarily drop the lock, reopen the table and grab the write lock again (using reopen_tables). When grabbing the lock again, reopen_tables doesn't pass a flag to mysql_lock_tables in order to ignore the impending global read lock, which causes a assertion because LOCK_open is being hold. Also dropping the lock must not signal to any threads that the table has been relinquished (related to the locking/flushing protocol). The solution is to correct the way the table is reopenned and the locks grabbed. When reopening the table and under LOCK TABLES, the table version should be set to 0 so other threads have to wait for the table. When grabbing the lock, any other flush should be ignored because it's theoretically a atomic operation. The chosen solution also fixes a potential discrepancy between binlog and GRL (global read lock) because table placeholders were being ignored, now a FLUSH TABLES WITH READ LOCK will properly for table with open placeholders. It's also important to mention that this patch doesn't fix a potential deadlock if one uses two GRLs under LOCK TABLES concurrently. mysql-test/r/lock_multi.result: Add test case result for Bug#32395 mysql-test/r/trigger_notembedded.result: Add test case result for Bug#32395 mysql-test/t/lock_multi.test: Add test case for Bug#32395 mysql-test/t/trigger_notembedded.test: Enable test case for Bug#32395 sql/ha_ndbcluster.cc: Update close_cached_tables usage. sql/ha_ndbcluster_binlog.cc: Update close_cached_tables usage. sql/mysql_priv.h: Update close_cache_tables prototype. sql/set_var.cc: Update close_cached_tables usage and set flag to wait for tables with placeholders. This is one of the places where a GRL can be obtained. sql/sql_base.cc: Preserve old version for write locked tables and ignore pending flushes and update close_cache_tables to take into account name locked tables. sql/sql_parse.cc: Update close_cached_tables usage and pass flag so that name locked tables are waited for. sql/sql_table.cc: Protect the table against a impending GRL if under LOCK TABLES.
2007-12-12 22:44:14 +01:00
--source include/wait_condition.inc
--send insert into t1 values (1,2);
connection default;
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
where state = "Waiting for global read lock" and
info = "insert into t1 values (1,2)";
Bug#32395 Alter table under a impending global read lock causes a server crash The problem is that some DDL statements (ALTER TABLE, CREATE TRIGGER, FLUSH TABLES, ...) when under LOCK TABLES need to momentarily drop the lock, reopen the table and grab the write lock again (using reopen_tables). When grabbing the lock again, reopen_tables doesn't pass a flag to mysql_lock_tables in order to ignore the impending global read lock, which causes a assertion because LOCK_open is being hold. Also dropping the lock must not signal to any threads that the table has been relinquished (related to the locking/flushing protocol). The solution is to correct the way the table is reopenned and the locks grabbed. When reopening the table and under LOCK TABLES, the table version should be set to 0 so other threads have to wait for the table. When grabbing the lock, any other flush should be ignored because it's theoretically a atomic operation. The chosen solution also fixes a potential discrepancy between binlog and GRL (global read lock) because table placeholders were being ignored, now a FLUSH TABLES WITH READ LOCK will properly for table with open placeholders. It's also important to mention that this patch doesn't fix a potential deadlock if one uses two GRLs under LOCK TABLES concurrently. mysql-test/r/lock_multi.result: Add test case result for Bug#32395 mysql-test/r/trigger_notembedded.result: Add test case result for Bug#32395 mysql-test/t/lock_multi.test: Add test case for Bug#32395 mysql-test/t/trigger_notembedded.test: Enable test case for Bug#32395 sql/ha_ndbcluster.cc: Update close_cached_tables usage. sql/ha_ndbcluster_binlog.cc: Update close_cached_tables usage. sql/mysql_priv.h: Update close_cache_tables prototype. sql/set_var.cc: Update close_cached_tables usage and set flag to wait for tables with placeholders. This is one of the places where a GRL can be obtained. sql/sql_base.cc: Preserve old version for write locked tables and ignore pending flushes and update close_cache_tables to take into account name locked tables. sql/sql_parse.cc: Update close_cached_tables usage and pass flag so that name locked tables are waited for. sql/sql_table.cc: Protect the table against a impending GRL if under LOCK TABLES.
2007-12-12 22:44:14 +01:00
--source include/wait_condition.inc
unlock tables;
connection flush;
--reap
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
where state = "Waiting for global read lock" and
info = "insert into t1 values (1,2)";
Bug#32395 Alter table under a impending global read lock causes a server crash The problem is that some DDL statements (ALTER TABLE, CREATE TRIGGER, FLUSH TABLES, ...) when under LOCK TABLES need to momentarily drop the lock, reopen the table and grab the write lock again (using reopen_tables). When grabbing the lock again, reopen_tables doesn't pass a flag to mysql_lock_tables in order to ignore the impending global read lock, which causes a assertion because LOCK_open is being hold. Also dropping the lock must not signal to any threads that the table has been relinquished (related to the locking/flushing protocol). The solution is to correct the way the table is reopenned and the locks grabbed. When reopening the table and under LOCK TABLES, the table version should be set to 0 so other threads have to wait for the table. When grabbing the lock, any other flush should be ignored because it's theoretically a atomic operation. The chosen solution also fixes a potential discrepancy between binlog and GRL (global read lock) because table placeholders were being ignored, now a FLUSH TABLES WITH READ LOCK will properly for table with open placeholders. It's also important to mention that this patch doesn't fix a potential deadlock if one uses two GRLs under LOCK TABLES concurrently. mysql-test/r/lock_multi.result: Add test case result for Bug#32395 mysql-test/r/trigger_notembedded.result: Add test case result for Bug#32395 mysql-test/t/lock_multi.test: Add test case for Bug#32395 mysql-test/t/trigger_notembedded.test: Enable test case for Bug#32395 sql/ha_ndbcluster.cc: Update close_cached_tables usage. sql/ha_ndbcluster_binlog.cc: Update close_cached_tables usage. sql/mysql_priv.h: Update close_cache_tables prototype. sql/set_var.cc: Update close_cached_tables usage and set flag to wait for tables with placeholders. This is one of the places where a GRL can be obtained. sql/sql_base.cc: Preserve old version for write locked tables and ignore pending flushes and update close_cache_tables to take into account name locked tables. sql/sql_parse.cc: Update close_cached_tables usage and pass flag so that name locked tables are waited for. sql/sql_table.cc: Protect the table against a impending GRL if under LOCK TABLES.
2007-12-12 22:44:14 +01:00
--source include/wait_condition.inc
select * from t1;
unlock tables;
connection insert;
--reap
connection default;
let $wait_condition=
select count(*) = 1 from t1;
--source include/wait_condition.inc
Bug#32395 Alter table under a impending global read lock causes a server crash The problem is that some DDL statements (ALTER TABLE, CREATE TRIGGER, FLUSH TABLES, ...) when under LOCK TABLES need to momentarily drop the lock, reopen the table and grab the write lock again (using reopen_tables). When grabbing the lock again, reopen_tables doesn't pass a flag to mysql_lock_tables in order to ignore the impending global read lock, which causes a assertion because LOCK_open is being hold. Also dropping the lock must not signal to any threads that the table has been relinquished (related to the locking/flushing protocol). The solution is to correct the way the table is reopenned and the locks grabbed. When reopening the table and under LOCK TABLES, the table version should be set to 0 so other threads have to wait for the table. When grabbing the lock, any other flush should be ignored because it's theoretically a atomic operation. The chosen solution also fixes a potential discrepancy between binlog and GRL (global read lock) because table placeholders were being ignored, now a FLUSH TABLES WITH READ LOCK will properly for table with open placeholders. It's also important to mention that this patch doesn't fix a potential deadlock if one uses two GRLs under LOCK TABLES concurrently. mysql-test/r/lock_multi.result: Add test case result for Bug#32395 mysql-test/r/trigger_notembedded.result: Add test case result for Bug#32395 mysql-test/t/lock_multi.test: Add test case for Bug#32395 mysql-test/t/trigger_notembedded.test: Enable test case for Bug#32395 sql/ha_ndbcluster.cc: Update close_cached_tables usage. sql/ha_ndbcluster_binlog.cc: Update close_cached_tables usage. sql/mysql_priv.h: Update close_cache_tables prototype. sql/set_var.cc: Update close_cached_tables usage and set flag to wait for tables with placeholders. This is one of the places where a GRL can be obtained. sql/sql_base.cc: Preserve old version for write locked tables and ignore pending flushes and update close_cache_tables to take into account name locked tables. sql/sql_parse.cc: Update close_cached_tables usage and pass flag so that name locked tables are waited for. sql/sql_table.cc: Protect the table against a impending GRL if under LOCK TABLES.
2007-12-12 22:44:14 +01:00
select * from t1;
drop table t1;
disconnect flush;
disconnect insert;
#
# Test that FLUSH TABLES under LOCK TABLES protects write locked tables
# from a impending FLUSH TABLES WITH READ LOCK
#
--disable_warnings
drop table if exists t1;
--enable_warnings
create table t1 (i int);
connect (flush,localhost,root,,test,,);
connection default;
lock tables t1 write;
connection flush;
--send flush tables with read lock;
connection default;
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
where state = "Waiting for global read lock";
Bug#32395 Alter table under a impending global read lock causes a server crash The problem is that some DDL statements (ALTER TABLE, CREATE TRIGGER, FLUSH TABLES, ...) when under LOCK TABLES need to momentarily drop the lock, reopen the table and grab the write lock again (using reopen_tables). When grabbing the lock again, reopen_tables doesn't pass a flag to mysql_lock_tables in order to ignore the impending global read lock, which causes a assertion because LOCK_open is being hold. Also dropping the lock must not signal to any threads that the table has been relinquished (related to the locking/flushing protocol). The solution is to correct the way the table is reopenned and the locks grabbed. When reopening the table and under LOCK TABLES, the table version should be set to 0 so other threads have to wait for the table. When grabbing the lock, any other flush should be ignored because it's theoretically a atomic operation. The chosen solution also fixes a potential discrepancy between binlog and GRL (global read lock) because table placeholders were being ignored, now a FLUSH TABLES WITH READ LOCK will properly for table with open placeholders. It's also important to mention that this patch doesn't fix a potential deadlock if one uses two GRLs under LOCK TABLES concurrently. mysql-test/r/lock_multi.result: Add test case result for Bug#32395 mysql-test/r/trigger_notembedded.result: Add test case result for Bug#32395 mysql-test/t/lock_multi.test: Add test case for Bug#32395 mysql-test/t/trigger_notembedded.test: Enable test case for Bug#32395 sql/ha_ndbcluster.cc: Update close_cached_tables usage. sql/ha_ndbcluster_binlog.cc: Update close_cached_tables usage. sql/mysql_priv.h: Update close_cache_tables prototype. sql/set_var.cc: Update close_cached_tables usage and set flag to wait for tables with placeholders. This is one of the places where a GRL can be obtained. sql/sql_base.cc: Preserve old version for write locked tables and ignore pending flushes and update close_cache_tables to take into account name locked tables. sql/sql_parse.cc: Update close_cached_tables usage and pass flag so that name locked tables are waited for. sql/sql_table.cc: Protect the table against a impending GRL if under LOCK TABLES.
2007-12-12 22:44:14 +01:00
--source include/wait_condition.inc
flush tables;
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
where state = "Waiting for global read lock";
Bug#32395 Alter table under a impending global read lock causes a server crash The problem is that some DDL statements (ALTER TABLE, CREATE TRIGGER, FLUSH TABLES, ...) when under LOCK TABLES need to momentarily drop the lock, reopen the table and grab the write lock again (using reopen_tables). When grabbing the lock again, reopen_tables doesn't pass a flag to mysql_lock_tables in order to ignore the impending global read lock, which causes a assertion because LOCK_open is being hold. Also dropping the lock must not signal to any threads that the table has been relinquished (related to the locking/flushing protocol). The solution is to correct the way the table is reopenned and the locks grabbed. When reopening the table and under LOCK TABLES, the table version should be set to 0 so other threads have to wait for the table. When grabbing the lock, any other flush should be ignored because it's theoretically a atomic operation. The chosen solution also fixes a potential discrepancy between binlog and GRL (global read lock) because table placeholders were being ignored, now a FLUSH TABLES WITH READ LOCK will properly for table with open placeholders. It's also important to mention that this patch doesn't fix a potential deadlock if one uses two GRLs under LOCK TABLES concurrently. mysql-test/r/lock_multi.result: Add test case result for Bug#32395 mysql-test/r/trigger_notembedded.result: Add test case result for Bug#32395 mysql-test/t/lock_multi.test: Add test case for Bug#32395 mysql-test/t/trigger_notembedded.test: Enable test case for Bug#32395 sql/ha_ndbcluster.cc: Update close_cached_tables usage. sql/ha_ndbcluster_binlog.cc: Update close_cached_tables usage. sql/mysql_priv.h: Update close_cache_tables prototype. sql/set_var.cc: Update close_cached_tables usage and set flag to wait for tables with placeholders. This is one of the places where a GRL can be obtained. sql/sql_base.cc: Preserve old version for write locked tables and ignore pending flushes and update close_cache_tables to take into account name locked tables. sql/sql_parse.cc: Update close_cached_tables usage and pass flag so that name locked tables are waited for. sql/sql_table.cc: Protect the table against a impending GRL if under LOCK TABLES.
2007-12-12 22:44:14 +01:00
--source include/wait_condition.inc
unlock tables;
connection flush;
--reap
connection default;
disconnect flush;
drop table t1;
#
# Bug#30331 Table_locks_waited shows inaccurate values
#
--disable_warnings
drop table if exists t1,t2;
--enable_warnings
create table t1 (a int);
flush status;
lock tables t1 read;
let $tlwa= `show status like 'Table_locks_waited'`;
connect (waiter,localhost,root,,);
connection waiter;
Backport of: ------------------------------------------------------------ revno: 3035.4.1 committer: Davi Arnaut <Davi.Arnaut@Sun.COM> branch nick: 39897-6.0 timestamp: Thu 2009-01-15 12:17:57 -0200 message: Bug#39897: lock_multi fails in pushbuild: timeout waiting for processlist The problem is that relying on the "Table lock" thread state in its current position to detect that a thread is waiting on a lock is race prone. The "Table lock" state change happens before the thread actually tries to grab a lock on a table. The solution is to move the "Table lock" state so that its set only when a thread is actually going to wait for a lock. The state change happens after the thread fails to grab the lock (because it is owned by other thread) and proceeds to wait on a condition. This is considered part of work related to WL#4284 "Transactional DDL locking" Warning: this patch contains an incompatible change. When waiting on a lock in thr_lock.c, the server used to display "Locked" processlist state. After this patch, the state is "Table lock". The new state was actually intended to be display since year 2002, when Monty added it. But up until removal of thd->locked boolean member, this state was ignored by SHOW PROCESSLIST code. mysql-test/r/lock_multi.result: A style fix. mysql-test/r/sp-threads.result: Changed output of SHOW PROCESSLIST (new wait state). mysql-test/t/lock_multi.test: Use a more accurate state description when waiting inside thr_lock.c. mysql-test/t/lock_sync.test: Use a more accurate state description when waiting inside thr_lock.c. mysql-test/t/multi_update.test: Use a more accurate state description when waiting inside thr_lock.c. mysql-test/t/query_cache_28249.test: Use a more accurate state description when waiting inside thr_lock.c. mysql-test/t/sp_notembedded.test: Use a more accurate state description when waiting inside thr_lock.c. mysql-test/t/status.test: Use a more accurate state description when waiting inside thr_lock.c. mysys/thr_lock.c: Update thread state while waiting for a table lock. sql/lock.cc: State change was moved inside thr_lock.c.
2009-12-03 21:08:27 +01:00
send insert into t1 values(1);
connection default;
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
where state = "Waiting for table level lock" and
info = "insert into t1 values(1)";
--source include/wait_condition.inc
let $tlwb= `show status like 'Table_locks_waited'`;
unlock tables;
connection waiter;
--reap
connection default;
drop table t1;
disconnect waiter;
--disable_query_log
eval SET @tlwa= SUBSTRING_INDEX('$tlwa', ' ', -1);
eval SET @tlwb= SUBSTRING_INDEX('$tlwb', ' ', -1);
--enable_query_log
select @tlwa < @tlwb;
--echo End of 5.1 tests
#
# Test that DROP TABLES does not wait for a impending FLUSH TABLES
# WITH READ LOCK
#
--disable_warnings
drop table if exists t1;
--enable_warnings
create table t1 (i int);
connect (flush,localhost,root,,test,,);
connection default;
lock tables t1 write;
connection flush;
--send flush tables with read lock;
connection default;
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
where state = "Waiting for global read lock";
--source include/wait_condition.inc
flush tables;
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Patch that refactors global read lock implementation and fixes bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK" and bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'". The first bug manifested itself as a deadlock which occurred when a connection, which had some table open through HANDLER statement, tried to update some data through DML statement while another connection tried to execute FLUSH TABLES WITH READ LOCK concurrently. What happened was that FTWRL in the second connection managed to perform first step of GRL acquisition and thus blocked all upcoming DML. After that it started to wait for table open through HANDLER statement to be flushed. When the first connection tried to execute DML it has started to wait for GRL/the second connection creating deadlock. The second bug manifested itself as starvation of FLUSH TABLES WITH READ LOCK statements in cases when there was a constant stream of concurrent DML statements (in two or more connections). This has happened because requests for protection against GRL which were acquired by DML statements were ignoring presence of pending GRL and thus the latter was starved. This patch solves both these problems by re-implementing GRL using metadata locks. Similar to the old implementation acquisition of GRL in new implementation is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. The first problem is solved because waits for GRL become visible to deadlock detector in metadata locking subsystem and thus deadlocks like one in the first bug become impossible. The second problem is solved because global S locks which are used for GRL implementation are given preference over IX locks which are acquired by concurrent DML (and we can switch to fair scheduling in future if needed). Important change: FTWRL/GRL no longer blocks DML and DDL on temporary tables. Before this patch behavior was not consistent in this respect: in some cases DML/DDL statements on temporary tables were blocked while in others they were not. Since the main use cases for FTWRL are various forms of backups and temporary tables are not preserved during backups we have opted for consistently allowing DML/DDL on temporary tables during FTWRL/GRL. Important change: This patch changes thread state names which are used when DML/DDL of FTWRL is waiting for global read lock. It is now either "Waiting for global read lock" or "Waiting for commit lock" depending on the stage on which FTWRL is. Incompatible change: To solve deadlock in events code which was exposed by this patch we have to replace LOCK_event_metadata mutex with metadata locks on events. As result we have to prohibit DDL on events under LOCK TABLES. This patch also adds extensive test coverage for interaction of DML/DDL and FTWRL. Performance of new and old global read lock implementations in sysbench tests were compared. There were no significant difference between new and old implementations. mysql-test/include/check_ftwrl_compatible.inc: Added helper script which allows to check that a statement is compatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/check_ftwrl_incompatible.inc: Added helper script which allows to check that a statement is incompatible with FLUSH TABLES WITH READ LOCK. mysql-test/include/handler.inc: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/include/wait_show_condition.inc: Fixed small error in the timeout message. The correct name of variable used as parameter for this script is "$condition" and not "$wait_condition". mysql-test/r/delayed.result: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/r/events_2.result: Updated test results after prohibiting event DDL operations under LOCK TABLES. mysql-test/r/flush.result: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/r/flush_read_lock.result: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/r/flush_read_lock_kill.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/r/handler_innodb.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/handler_myisam.result: Adjusted test case to the fact that now DROP TABLE closes open HANDLERs for the table to be dropped before checking if there active FTWRL in this connection. mysql-test/r/mdl_sync.result: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. mysql-test/suite/perfschema/r/dml_setup_instruments.result: Updated test results after removing global COND_global_read_lock condition variable. mysql-test/suite/perfschema/r/func_file_io.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/func_mutex.result: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/r/global_read_lock.result: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/r/server_init.result: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/perfschema/t/func_file_io.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/func_mutex.test: Ensure that this test doesn't affect subsequent tests. At the end of its execution enable back P_S instrumentation which this test disables at some point. mysql-test/suite/perfschema/t/global_read_lock.test: Adjusted test case to take into account that new GRL implementation is based on MDL. mysql-test/suite/perfschema/t/server_init.test: Adjusted test case after replacing custom global read lock implementation with one based on MDL and replacing LOCK_event_metadata mutex with metadata lock. mysql-test/suite/rpl/r/rpl_tmp_table_and_DDL.result: Updated test results after prohibiting event DDL under LOCK TABLES. mysql-test/t/delayed.test: Added test coverage for scenario which triggered assert in metadata locking subsystem. mysql-test/t/events_2.test: Updated test case after prohibiting event DDL operations under LOCK TABLES. mysql-test/t/flush.test: Added test coverage for bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". mysql-test/t/flush_block_commit.test: Adjusted test case after changing thread state name which is used when COMMIT waits for FLUSH TABLES WITH READ LOCK from "Waiting for release of readlock" to "Waiting for commit lock". mysql-test/t/flush_block_commit_notembedded.test: Adjusted test case after changing thread state name which is used when DML waits for FLUSH TABLES WITH READ LOCK. Now we use "Waiting for global read lock" in this case. mysql-test/t/flush_read_lock.test: Added test coverage for various aspects of FLUSH TABLES WITH READ LOCK functionality. mysql-test/t/flush_read_lock_kill-master.opt: We no longer need to use make_global_read_lock_block_commit_loop debug tag in this test. Instead we rely on an appropriate debug_sync point in MDL code. mysql-test/t/flush_read_lock_kill.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Use new debug_sync point. Do not disable concurrent inserts as now InnoDB we always use InnoDB table. mysql-test/t/lock_multi.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". mysql-test/t/mdl_sync.test: Adjusted test case after replacing custom global read lock implementation with one based on metadata locks. Replaced usage of GRL-specific debug_sync's with appropriate sync points in MDL subsystem. Updated thread state names which are used when DDL waits for FTWRL. mysql-test/t/trigger_notembedded.test: Adjusted test case after changing thread state names which are used when DML or DDL waits for FLUSH TABLES WITH READ LOCK to "Waiting for global read lock". sql/event_data_objects.cc: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_data_objects.h: Removed Event_queue_element::status/last_executed_changed members and Event_queue_element::update_timing_fields() method. We no longer use this class for updating mysql.events once event is chosen for execution. Accesses to instances of this class in scheduler thread require protection by Event_queue::LOCK_event_queue mutex and we try to avoid updating table while holding this lock. sql/event_db_repository.cc: - Changed Event_db_repository methods to not release all metadata locks once they are done updating mysql.events table. This allows to keep metadata lock protecting against GRL and lock protecting particular event around until corresponding DDL statement is written to the binary log. - Removed logic for conditional update of "status" and "last_executed" fields from update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" is too much hassle. sql/event_db_repository.h: Removed logic for conditional update of "status" and "last_executed" fields from Event_db_repository:: update_timing_fields_for_event() method. In the only case when this method is called now "last_executed" is always modified and tracking change of "status" field is too much hassle. sql/event_queue.cc: Changed event scheduler code not to update mysql.events table while holding Event_queue::LOCK_event_queue mutex. Doing so led to a deadlock with a new GRL implementation. This deadlock didn't occur with old implementation due to fact that code acquiring protection against GRL ignored pending GRL requests (which lead to GRL starvation). One of goals of new implementation is to disallow GRL starvation and so we have to solve problem with this deadlock in a different way. sql/events.cc: Changed methods of Events class to acquire protection against GRL while perfoming DDL statement and keep it until statement is written to the binary log. Unfortunately this step together with new GRL implementation exposed deadlock involving Events::LOCK_event_metadata and GRL. To solve it Events::LOCK_event_metadata mutex was replaced with a metadata lock on event. As a side-effect events DDL has to be prohibited under LOCK TABLES even in cases when mysql.events table was explicitly locked for write. sql/events.h: Replaced Events::LOCK_event_metadata mutex with a metadata lock on event. sql/ha_ndbcluster.cc: Updated code after replacing custom global read lock implementation with one based on MDL. Since MDL subsystem should now be able to detect deadlocks involving metadata locks and GRL there is no need for special handling of active GRL. sql/handler.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. sql/lock.cc: Replaced custom implementation of global read lock with one based on metadata locks. This step allows to expose wait for GRL to deadlock detector of MDL subsystem and thus succesfully resolve deadlocks similar to one behind bug #57006 "Deadlock between HANDLER and FLUSH TABLES WITH READ LOCK". It also solves problem with GRL starvation described in bug #54673 "It takes too long to get readlock for 'FLUSH TABLES WITH READ LOCK'" since metadata locks used by GRL give preference to FTWRL statement instead of DML statements (if needed in future this can be changed to fair scheduling). Similar to old implementation of acquisition of GRL is two-step. During the first step we block all concurrent DML and DDL statements by acquiring global S metadata lock (each DML and DDL statement acquires global IX lock for its duration). During the second step we block commits by acquiring global S lock in COMMIT namespace (commit code acquires global IX lock in this namespace). Note that unlike in old implementation acquisition of protection against GRL in DML and DDL is semi-automatic. We assume that any statement which should be blocked by GRL will either open and acquires write-lock on tables or acquires metadata locks on objects it is going to modify. For any such statement global IX metadata lock is automatically acquired for its duration. To support this change: - Global_read_lock::lock/unlock_global_read_lock and make_global_read_lock_block_commit methods were changed accordingly. - Global_read_lock::wait_if_global_read_lock() and start_waiting_global_read_lock() methods were dropped. It is now responsibility of code acquiring metadata locks opening tables to acquire protection against GRL by explicitly taking global IX lock with statement duration. - Global variables, mutex and condition variable used by old implementation was removed. - lock_routine_name() was changed to use statement duration for its global IX lock. It was also renamed to lock_object_name() as it now also used to take metadata locks on events. - Global_read_lock::set_explicit_lock_duration() was added which allows not to release locks used for GRL when leaving prelocked mode. sql/lock.h: - Renamed lock_routine_name() to lock_object_name() and changed its signature to allow its usage for events. - Removed broadcast_refresh() function. It is no longer needed with new GRL implementation. sql/log_event.cc: Release metadata locks with statement duration at the end of processing legacy event for LOAD DATA. This ensures that replication thread processing such event properly releases its protection against global read lock. sql/mdl.cc: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Changed thread state name for GLOBAL namespace to "Waiting for global read lock". Optimized MDL_map::find_or_insert() method to avoid taking m_mutex mutex when looking up MDL_lock objects for GLOBAL or COMMIT namespaces. We keep pre-created MDL_lock objects for these namespaces around and simply return pointers to these global objects when needed. Changed MDL_lock/MDL_scoped_lock to properly handle notification of insert delayed handler threads when FTWRL takes global S lock. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mdl.h: Changed MDL subsystem to support new MDL-based implementation of global read lock. Added COMMIT and EVENTS namespaces for metadata locks. Introduced concept of lock duration. In addition to locks with transaction duration which work in the way which is similar to how locks worked before (i.e. they are released at the end of transaction), locks with statement and explicit duration were introduced. Locks with statement duration are automatically released at the end of statement. Locks with explicit duration require explicit release and obsolete concept of transactional sentinel. * Changed MDL_request and MDL_ticket classes to support notion of duration. * Changed MDL_context to keep locks with different duration in different lists. Changed code handling ticket list to take this into account. * Changed methods responsible for releasing locks to take into account duration of tickets. Particularly public MDL_context::release_lock() method now only can release tickets with explicit duration (there is still internal method which allows to specify duration). To release locks with statement or transaction duration one have to use release_statement/transactional_locks() methods. * Concept of savepoint for MDL subsystem now has to take into account locks with statement duration. Consequently MDL_savepoint class was introduced and methods working with savepoints were updated accordingly. * Added methods which allow to set duration for one or all locks in the context. sql/mysqld.cc: Removed global mutex and condition variables which were used by old implementation of GRL. Also we no longer need to initialize Events::LOCK_event_metadata mutex as it was replaced with metadata locks on events. sql/mysqld.h: Removed global variable, mutex and condition variables which were used by old implementation of GRL. sql/rpl_rli.cc: When slave thread closes tables which were open for handling of RBR events ensure that it releases global IX lock which was acquired as protection against GRL. sql/sp.cc: Adjusted code to the new signature of lock_object/routine_name(), to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sp_head.cc: Ensure that statements in stored procedures release statement metadata locks and thus release their protectiong against GRL in proper moment in time. Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_admin.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_base.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. Code doing this also responsible for checking that current connection has no active GRL by calling an Global_read_lock::can_acquire_protection() method. Changed code in open_table() and lock_table_names() accordingly. Note that as result of this change DDL and DML on temporary tables is always compatible with GRL (before it was incompatible in some cases and compatible in other cases). - To speed-up code acquiring protection against GRL introduced m_has_protection_against_grl member in Open_table_context class. It indicates that protection was already acquired sometime during open_tables() execution and new attempts can be skipped. - Thanks to new GRL implementation calls to broadcast_refresh() became unnecessary and were removed. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_base.h: Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. Also introduced Open_table_context::m_has_protection_against_grl member which allows to avoid acquiring protection against GRL while opening tables if such protection was already acquired. sql/sql_class.cc: Changed THD::leave_locked_tables_mode() after transactional sentinel for metadata locks was obsoleted by introduction of locks with explicit duration. sql/sql_class.h: - Adjusted code to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. - Changed Global_read_lock class according to changes in global read lock implementation: * wait_if_global_read_lock and start_waiting_global_read_lock are now gone. Instead code needing protection against GRL has to acquire global IX metadata lock with statement duration itself. To help it new can_acquire_protection() was introduced. Also as result of the above change m_protection_count member is gone too. * Added m_mdl_blocks_commits_lock member to store metadata lock blocking commits. * Adjusted code to the fact that concept of transactional sentinel was obsoleted by concept of lock duration. - Removed CF_PROTECT_AGAINST_GRL flag as it is no longer necessary. New GRL implementation acquires protection against global read lock automagically when statement acquires metadata locks on tables or other objects it is going to change. sql/sql_db.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/sql_handler.cc: Removed call to broadcast_refresh() function. It is no longer needed with new GRL implementation. Adjusted code after introducing duration concept for metadata locks. Particularly to the fact transactional sentinel was replaced with explicit duration. sql/sql_handler.h: Renamed mysql_ha_move_tickets_after_trans_sentinel() to mysql_ha_set_explicit_lock_duration() after transactional sentinel was obsoleted by locks with explicit duration. sql/sql_insert.cc: Adjusted code handling delaying inserts after switching to new GRL implementation. Now connection thread initiating delayed insert has to acquire global IX lock in addition to metadata lock on table being inserted into. This IX lock protects against GRL and similarly to SW lock on table being inserted into has to be passed to handler thread in order to avoid deadlocks. sql/sql_lex.cc: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_lex.h: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/sql_parse.cc: - Implemented support for new approach to acquiring protection against global read lock. We no longer acquire such protection explicitly on the basis of statement flags. Instead we always rely on code which is responsible for acquiring metadata locks on object to be changed acquiring this protection. This is achieved by acquiring global IX metadata lock with statement duration. This lock is automatically released at the end of statement execution. - Changed implementation of CREATE/DROP PROCEDURE/FUNCTION not to release metadata locks and thus protection against of GRL in the middle of statement execution. - Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_prepare.cc: Adjusted code to the to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_rename.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before renaming tables. This happens automatically in code which acquires metadata locks on tables being renamed. sql/sql_show.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request and to the fact that savepoints for MDL subsystem are now represented by MDL_savepoint class. sql/sql_table.cc: - With new GRL implementation there is no need to explicitly acquire protection against GRL before dropping tables. This happens automatically in code which acquires metadata locks on tables being dropped. - Changed mysql_alter_table() not to release lock on new table name explicitly and to rely on automatic release of locks at the end of statement instead. This was necessary since now MDL_context::release_lock() is supported only for locks for explicit duration. sql/sql_trigger.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before changing table triggers. This happens automatically in code which acquires metadata locks on tables which triggers are to be changed. sql/sql_update.cc: Fix bug exposed by GRL testing. During prepare phase acquire only S metadata locks instead of SW locks to keep prepare of multi-UPDATE compatible with concurrent LOCK TABLES WRITE and global read lock. sql/sql_view.cc: With new GRL implementation there is no need to explicitly acquire protection against GRL before creating view. This happens automatically in code which acquires metadata lock on view to be created. sql/sql_yacc.yy: LEX::protect_against_global_read_lock member is no longer necessary since protection against GRL is automatically taken by code acquiring metadata locks/opening tables. sql/table.cc: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/table.h: Adjusted code to the fact that one now needs specify duration of lock when initializing MDL_request. sql/transaction.cc: Replaced custom implementation of global read lock with one based on metadata locks. Consequently when doing commit instead of calling method of Global_read_lock class to acquire protection against GRL we simply acquire IX in COMMIT namespace. Also adjusted code to the fact that MDL savepoint is now represented by MDL_savepoint class.
2010-11-11 18:11:05 +01:00
where state = "Waiting for global read lock";
--source include/wait_condition.inc
drop table t1;
connection flush;
--reap
connection default;
disconnect flush;
2009-12-09 10:44:01 +01:00
Implement new type-of-operation-aware metadata locks. Add a wait-for graph based deadlock detector to the MDL subsystem. Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and bug #37346 "innodb does not detect deadlock between update and alter table". The first bug manifested itself as an unwarranted abort of a transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER statement, when this transaction tried to repeat use of a table, which it has already used in a similar fashion before ALTER started. The second bug showed up as a deadlock between table-level locks and InnoDB row locks, which was "detected" only after innodb_lock_wait_timeout timeout. A transaction would start using the table and modify a few rows. Then ALTER TABLE would come in, and start copying rows into a temporary table. Eventually it would stumble on the modified records and get blocked on a row lock. The first transaction would try to do more updates, and get blocked on thr_lock.c lock. This situation of circular wait would only get resolved by a timeout. Both these bugs stemmed from inadequate solutions to the problem of deadlocks occurring between different locking subsystems. In the first case we tried to avoid deadlocks between metadata locking and table-level locking subsystems, when upgrading shared metadata lock to exclusive one. Transactions holding the shared lock on the table and waiting for some table-level lock used to be aborted too aggressively. We also allowed ALTER TABLE to start in presence of transactions that modify the subject table. ALTER TABLE acquires TL_WRITE_ALLOW_READ lock at start, and that block all writes against the table (naturally, we don't want any writes to be lost when switching the old and the new table). TL_WRITE_ALLOW_READ lock, in turn, would block the started transaction on thr_lock.c lock, should they do more updates. This, again, lead to the need to abort such transactions. The second bug occurred simply because we didn't have any mechanism to detect deadlocks between the table-level locks in thr_lock.c and row-level locks in InnoDB, other than innodb_lock_wait_timeout. This patch solves both these problems by moving lock conflicts which are causing these deadlocks into the metadata locking subsystem, thus making it possible to avoid or detect such deadlocks inside MDL. To do this we introduce new type-of-operation-aware metadata locks, which allow MDL subsystem to know not only the fact that transaction has used or is going to use some object but also what kind of operation it has carried out or going to carry out on the object. This, along with the addition of a special kind of upgradable metadata lock, allows ALTER TABLE to wait until all transactions which has updated the table to go away. This solves the second issue. Another special type of upgradable metadata lock is acquired by LOCK TABLE WRITE. This second lock type allows to solve the first issue, since abortion of table-level locks in event of DDL under LOCK TABLES becomes also unnecessary. Below follows the list of incompatible changes introduced by this patch: - From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those statements that acquire TL_WRITE_ALLOW_READ lock) wait for all transactions which has *updated* the table to complete. - From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE (i.e. all statements which acquire TL_WRITE table-level lock) wait for all transaction which *updated or read* from the table to complete. As a consequence, innodb_table_locks=0 option no longer applies to LOCK TABLES ... WRITE. - DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort statements or transactions which use tables being dropped or renamed, and instead wait for these transactions to complete. - Since LOCK TABLES WRITE now takes a special metadata lock, not compatible with with reads or writes against the subject table and transaction-wide, thr_lock.c deadlock avoidance algorithm that used to ensure absence of deadlocks between LOCK TABLES WRITE and other statements is no longer sufficient, even for MyISAM. The wait-for graph based deadlock detector of MDL subsystem may sometimes be necessary and is involved. This may lead to ER_LOCK_DEADLOCK error produced for multi-statement transactions even if these only use MyISAM: session 1: session 2: begin; update t1 ... lock table t2 write, t1 write; -- gets a lock on t2, blocks on t1 update t2 ... (ER_LOCK_DEADLOCK) - Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE was abandoned. LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same priority as the usual LOCK TABLE ... WRITE. SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in the wait queue. - We do not take upgradable metadata locks on implicitly locked tables. So if one has, say, a view v1 that uses table t1, and issues: LOCK TABLE v1 WRITE; FLUSH TABLE t1; -- (or just 'FLUSH TABLES'), an error is produced. In order to be able to perform DDL on a table under LOCK TABLES, the table must be locked explicitly in the LOCK TABLES list. mysql-test/include/handler.inc: Adjusted test case to trigger an execution path on which bug 41110 "crash with handler command when used concurrently with alter table" and bug 41112 "crash in mysql_ha_close_table/get_lock_data with alter table" were originally discovered. Left old test case which no longer triggers this execution path for the sake of coverage. Added test coverage for HANDLER SQL statements and type-aware metadata locks. Added a test for the global shared lock and HANDLER SQL. Updated tests to take into account that the old simple deadlock detection heuristics was replaced with a graph-based deadlock detector. mysql-test/r/debug_sync.result: Updated results (see debug_sync.test). mysql-test/r/handler_innodb.result: Updated results (see handler.inc test). mysql-test/r/handler_myisam.result: Updated results (see handler.inc test). mysql-test/r/innodb-lock.result: Updated results (see innodb-lock.test). mysql-test/r/innodb_mysql_lock.result: Updated results (see innodb_mysql_lock.test). mysql-test/r/lock.result: Updated results (see lock.test). mysql-test/r/lock_multi.result: Updated results (see lock_multi.test). mysql-test/r/lock_sync.result: Updated results (see lock_sync.test). mysql-test/r/mdl_sync.result: Updated results (see mdl_sync.test). mysql-test/r/sp-threads.result: SHOW PROCESSLIST output has changed due to the fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/r/truncate_coverage.result: Updated results (see truncate_coverage.test). mysql-test/suite/funcs_1/datadict/processlist_val.inc: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/funcs_1/r/processlist_val_no_prot.result: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/rpl/t/rpl_sp.test: Updated to a new SHOW PROCESSLIST state name. mysql-test/t/debug_sync.test: Use LOCK TABLES READ instead of LOCK TABLES WRITE as the latter no longer allows to trigger execution path involving waiting on thr_lock.c lock and therefore reaching debug sync-point covered by this test. mysql-test/t/innodb-lock.test: Adjusted test case to the fact that innodb_table_locks=0 option is no longer supported, since LOCK TABLES WRITE handles all its conflicts within MDL subsystem. mysql-test/t/innodb_mysql_lock.test: Added test for bug #37346 "innodb does not detect deadlock between update and alter table". mysql-test/t/lock.test: Added test coverage which checks the fact that we no longer support DDL under LOCK TABLES on tables which were locked implicitly. Adjusted existing test cases accordingly. mysql-test/t/lock_multi.test: Added test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock". Adjusted other test cases to take into account the fact that waiting for LOCK TABLES ... WRITE now happens within MDL subsystem. mysql-test/t/lock_sync.test: Since LOCK TABLES ... WRITE now takes SNRW metadata lock for tables locked explicitly we have to implicitly lock InnoDB tables (through view) to trigger the table-level lock conflict between TL_WRITE and TL_WRITE_ALLOW_WRITE. mysql-test/t/mdl_sync.test: Added basic test coverage for type-of-operation-aware metadata locks. Also covered with tests some use cases involving HANDLER statements in which a deadlock could arise. Adjusted existing tests to take type-of-operation-aware MDL into account. mysql-test/t/multi_update.test: Update to a new SHOW PROCESSLIST state name. mysql-test/t/truncate_coverage.test: Adjusted test case after making LOCK TABLES WRITE to wait until transactions that use the table to be locked are completed. Updated to the changed name of DEBUG_SYNC point. sql/handler.cc: Global read lock functionality has been moved into a class. sql/lock.cc: Global read lock functionality has been moved into a class. Updated code to use the new MDL API. sql/mdl.cc: Introduced new type-of-operation aware metadata locks. To do this: - Changed MDL_lock to use one list for waiting requests and one list for granted requests. For each list, added a bitmap that holds information what lock types a list contains. Added a helper class MDL_lock::List to manipulate with granted and waited lists while keeping the bitmaps in sync with list contents. - Changed lock-compatibility functions to use bitmaps that define compatibility. - Introduced a graph based deadlock detector inspired by waiting_threads.c from Maria implementation. - Now that we have a deadlock detector, and no longer have a global lock to protect individual lock objects, but rather use an rw lock per object, removed redundant code for upgrade, and the global read lock. Changed the MDL API to no longer require the caller to acquire the global intention exclusive lock by means of a separate method. Removed a few more methods that became redundant. - Removed deadlock detection heuristic, it has been made obsolete by the deadlock detector. - With operation-type-aware metadata locks, MDL subsystem has become aware of potential conflicts between DDL and open transactions. This made it possible to remove calls to mysql_abort_transactions_with_shared_lock() from acquisition paths for exclusive lock and lock upgrade. Now we can simply wait for these transactions to complete without fear of deadlock. Function mysql_lock_abort() has also become unnecessary for all conflicting cases except when a DDL conflicts with a connection that has an open HANDLER. sql/mdl.h: Introduced new type-of-operation aware metadata locks. Introduced a graph based deadlock detector and supporting methods. Added comments. God rid of redundant API calls. Renamed m_lt_or_ha_sentinel to m_trans_sentinel, since now it guards the global read lock as well as LOCK TABLES and HANDLER locks. sql/mysql_priv.h: Moved the global read lock functionality into a class. Added MYSQL_OPEN_FORCE_SHARED_MDL flag which forces open_tables() to take MDL_SHARED on tables instead of metadata locks specified in the parser. We use this to allow PREPARE run concurrently in presence of LOCK TABLES ... WRITE. Added signature for find_table_for_mdl_ugprade(). sql/set_var.cc: Global read lock functionality has been moved into a class. sql/sp_head.cc: When creating TABLE_LIST elements for prelocking or system tables set the type of request for metadata lock according to the operation that will be performed on the table. sql/sql_base.cc: - Updated code to use the new MDL API. - In order to avoid locks starvation we take upgradable locks all at once. As result implicitly locked tables no longer get an upgradable lock. Consequently DDL and FLUSH TABLES for such tables is prohibited. find_write_locked_table() was replaced by find_table_for_mdl_upgrade() function. open_table() was adjusted to return TABLE instance with upgradable ticket when necessary. - We no longer wait for all locks on OT_WAIT back off action -- only on the lock that caused the wait conflict. Moreover, now we distinguish cases when we have to wait due to conflict in MDL and old version of table in TDC. - Upate mysql_notify_threads_having_share_locks() to only abort thr_lock.c waits of threads that have open HANDLERs, since lock conflicts with only these threads now can lead to deadlocks not detectable by the MDL deadlock detector. - Remove mysql_abort_transactions_with_shared_locks() which is no longer needed. sql/sql_class.cc: Global read lock functionality has been moved into a class. Re-arranged code in THD::cleanup() to simplify assert. sql/sql_class.h: Introduced class to incapsulate global read lock functionality. Now sentinel in MDL subsystem guards the global read lock as well as LOCK TABLES and HANDLER locks. Adjusted code accordingly. sql/sql_db.cc: Global read lock functionality has been moved into a class. sql/sql_delete.cc: We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result TRUNCATE TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_handler.cc: Inform MDL_context about presence of open HANDLERs. Since HANLDERs break MDL protocol by acquiring table-level lock while holding only S metadata lock on a table MDL subsystem should take special care about such contexts (Now this is the only case when mysql_lock_abort() is used). sql/sql_parse.cc: Global read lock functionality has been moved into a class. Do not take upgradable metadata locks when opening tables for CREATE TABLE SELECT as it is not necessary and limits concurrency. When initializing TABLE_LIST objects before adding them to the table list set the type of request for metadata lock according to the operation that will be performed on the table. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result FLUSH TABLES is no longer allowed for such tables. sql/sql_prepare.cc: Use MYSQL_OPEN_FORCE_SHARED_MDL flag when opening tables during PREPARE. This allows PREPARE to run concurrently in presence of LOCK TABLES ... WRITE. sql/sql_rename.cc: Global read lock functionality has been moved into a class. sql/sql_show.cc: Updated code to use the new MDL API. sql/sql_table.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result DROP TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_trigger.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result CREATE/DROP TRIGGER is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_view.cc: Global read lock functionality has been moved into a class. Fixed results of wrong merge that led to misuse of GLR API. CREATE VIEW statement is not a commit statement. sql/table.cc: When resetting TABLE_LIST objects for PS or SP re-execution set the type of request for metadata lock according to the operation that will be performed on the table. Do the same in auxiliary function initializing metadata lock requests in a table list. sql/table.h: When initializing TABLE_LIST objects set the type of request for metadata lock according to the operation that will be performed on the table. sql/transaction.cc: Global read lock functionality has been moved into a class.
2010-02-01 12:43:06 +01:00
--echo #
--echo # Test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock".
--echo #
--disable_warnings
drop table if exists t1;
--enable_warnings
create table t1 (c1 int primary key, c2 int, c3 int);
insert into t1 values (1,1,0),(2,2,0),(3,3,0),(4,4,0),(5,5,0);
begin;
update t1 set c3=c3+1 where c2=3;
connect (con46272,localhost,root,,test,,);
connection con46272;
--echo # The below ALTER TABLE statement should wait till transaction
--echo # in connection 'default' is complete and then succeed.
--echo # It should not deadlock or fail with ER_LOCK_DEADLOCK error.
--echo # Sending:
--send alter table t1 add column c4 int;
connection default;
--echo # Wait until the above ALTER TABLE gets blocked because this
--echo # connection holds SW metadata lock on table to be altered.
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
where state = "Waiting for table metadata lock" and
info = "alter table t1 add column c4 int";
Implement new type-of-operation-aware metadata locks. Add a wait-for graph based deadlock detector to the MDL subsystem. Fixes bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock" and bug #37346 "innodb does not detect deadlock between update and alter table". The first bug manifested itself as an unwarranted abort of a transaction with ER_LOCK_DEADLOCK error by a concurrent ALTER statement, when this transaction tried to repeat use of a table, which it has already used in a similar fashion before ALTER started. The second bug showed up as a deadlock between table-level locks and InnoDB row locks, which was "detected" only after innodb_lock_wait_timeout timeout. A transaction would start using the table and modify a few rows. Then ALTER TABLE would come in, and start copying rows into a temporary table. Eventually it would stumble on the modified records and get blocked on a row lock. The first transaction would try to do more updates, and get blocked on thr_lock.c lock. This situation of circular wait would only get resolved by a timeout. Both these bugs stemmed from inadequate solutions to the problem of deadlocks occurring between different locking subsystems. In the first case we tried to avoid deadlocks between metadata locking and table-level locking subsystems, when upgrading shared metadata lock to exclusive one. Transactions holding the shared lock on the table and waiting for some table-level lock used to be aborted too aggressively. We also allowed ALTER TABLE to start in presence of transactions that modify the subject table. ALTER TABLE acquires TL_WRITE_ALLOW_READ lock at start, and that block all writes against the table (naturally, we don't want any writes to be lost when switching the old and the new table). TL_WRITE_ALLOW_READ lock, in turn, would block the started transaction on thr_lock.c lock, should they do more updates. This, again, lead to the need to abort such transactions. The second bug occurred simply because we didn't have any mechanism to detect deadlocks between the table-level locks in thr_lock.c and row-level locks in InnoDB, other than innodb_lock_wait_timeout. This patch solves both these problems by moving lock conflicts which are causing these deadlocks into the metadata locking subsystem, thus making it possible to avoid or detect such deadlocks inside MDL. To do this we introduce new type-of-operation-aware metadata locks, which allow MDL subsystem to know not only the fact that transaction has used or is going to use some object but also what kind of operation it has carried out or going to carry out on the object. This, along with the addition of a special kind of upgradable metadata lock, allows ALTER TABLE to wait until all transactions which has updated the table to go away. This solves the second issue. Another special type of upgradable metadata lock is acquired by LOCK TABLE WRITE. This second lock type allows to solve the first issue, since abortion of table-level locks in event of DDL under LOCK TABLES becomes also unnecessary. Below follows the list of incompatible changes introduced by this patch: - From now on, ALTER TABLE and CREATE/DROP TRIGGER SQL (i.e. those statements that acquire TL_WRITE_ALLOW_READ lock) wait for all transactions which has *updated* the table to complete. - From now on, LOCK TABLES ... WRITE, REPAIR/OPTIMIZE TABLE (i.e. all statements which acquire TL_WRITE table-level lock) wait for all transaction which *updated or read* from the table to complete. As a consequence, innodb_table_locks=0 option no longer applies to LOCK TABLES ... WRITE. - DROP DATABASE, DROP TABLE, RENAME TABLE no longer abort statements or transactions which use tables being dropped or renamed, and instead wait for these transactions to complete. - Since LOCK TABLES WRITE now takes a special metadata lock, not compatible with with reads or writes against the subject table and transaction-wide, thr_lock.c deadlock avoidance algorithm that used to ensure absence of deadlocks between LOCK TABLES WRITE and other statements is no longer sufficient, even for MyISAM. The wait-for graph based deadlock detector of MDL subsystem may sometimes be necessary and is involved. This may lead to ER_LOCK_DEADLOCK error produced for multi-statement transactions even if these only use MyISAM: session 1: session 2: begin; update t1 ... lock table t2 write, t1 write; -- gets a lock on t2, blocks on t1 update t2 ... (ER_LOCK_DEADLOCK) - Finally, support of LOW_PRIORITY option for LOCK TABLES ... WRITE was abandoned. LOCK TABLE ... LOW_PRIORITY WRITE from now on has the same priority as the usual LOCK TABLE ... WRITE. SELECT HIGH PRIORITY no longer trumps LOCK TABLE ... WRITE in the wait queue. - We do not take upgradable metadata locks on implicitly locked tables. So if one has, say, a view v1 that uses table t1, and issues: LOCK TABLE v1 WRITE; FLUSH TABLE t1; -- (or just 'FLUSH TABLES'), an error is produced. In order to be able to perform DDL on a table under LOCK TABLES, the table must be locked explicitly in the LOCK TABLES list. mysql-test/include/handler.inc: Adjusted test case to trigger an execution path on which bug 41110 "crash with handler command when used concurrently with alter table" and bug 41112 "crash in mysql_ha_close_table/get_lock_data with alter table" were originally discovered. Left old test case which no longer triggers this execution path for the sake of coverage. Added test coverage for HANDLER SQL statements and type-aware metadata locks. Added a test for the global shared lock and HANDLER SQL. Updated tests to take into account that the old simple deadlock detection heuristics was replaced with a graph-based deadlock detector. mysql-test/r/debug_sync.result: Updated results (see debug_sync.test). mysql-test/r/handler_innodb.result: Updated results (see handler.inc test). mysql-test/r/handler_myisam.result: Updated results (see handler.inc test). mysql-test/r/innodb-lock.result: Updated results (see innodb-lock.test). mysql-test/r/innodb_mysql_lock.result: Updated results (see innodb_mysql_lock.test). mysql-test/r/lock.result: Updated results (see lock.test). mysql-test/r/lock_multi.result: Updated results (see lock_multi.test). mysql-test/r/lock_sync.result: Updated results (see lock_sync.test). mysql-test/r/mdl_sync.result: Updated results (see mdl_sync.test). mysql-test/r/sp-threads.result: SHOW PROCESSLIST output has changed due to the fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/r/truncate_coverage.result: Updated results (see truncate_coverage.test). mysql-test/suite/funcs_1/datadict/processlist_val.inc: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/funcs_1/r/processlist_val_no_prot.result: SELECT FROM I_S.PROCESSLIST output has changed due to fact that waiting for LOCK TABLES WRITE now happens within metadata locking subsystem. mysql-test/suite/rpl/t/rpl_sp.test: Updated to a new SHOW PROCESSLIST state name. mysql-test/t/debug_sync.test: Use LOCK TABLES READ instead of LOCK TABLES WRITE as the latter no longer allows to trigger execution path involving waiting on thr_lock.c lock and therefore reaching debug sync-point covered by this test. mysql-test/t/innodb-lock.test: Adjusted test case to the fact that innodb_table_locks=0 option is no longer supported, since LOCK TABLES WRITE handles all its conflicts within MDL subsystem. mysql-test/t/innodb_mysql_lock.test: Added test for bug #37346 "innodb does not detect deadlock between update and alter table". mysql-test/t/lock.test: Added test coverage which checks the fact that we no longer support DDL under LOCK TABLES on tables which were locked implicitly. Adjusted existing test cases accordingly. mysql-test/t/lock_multi.test: Added test for bug #46272 "MySQL 5.4.4, new MDL: unnecessary deadlock". Adjusted other test cases to take into account the fact that waiting for LOCK TABLES ... WRITE now happens within MDL subsystem. mysql-test/t/lock_sync.test: Since LOCK TABLES ... WRITE now takes SNRW metadata lock for tables locked explicitly we have to implicitly lock InnoDB tables (through view) to trigger the table-level lock conflict between TL_WRITE and TL_WRITE_ALLOW_WRITE. mysql-test/t/mdl_sync.test: Added basic test coverage for type-of-operation-aware metadata locks. Also covered with tests some use cases involving HANDLER statements in which a deadlock could arise. Adjusted existing tests to take type-of-operation-aware MDL into account. mysql-test/t/multi_update.test: Update to a new SHOW PROCESSLIST state name. mysql-test/t/truncate_coverage.test: Adjusted test case after making LOCK TABLES WRITE to wait until transactions that use the table to be locked are completed. Updated to the changed name of DEBUG_SYNC point. sql/handler.cc: Global read lock functionality has been moved into a class. sql/lock.cc: Global read lock functionality has been moved into a class. Updated code to use the new MDL API. sql/mdl.cc: Introduced new type-of-operation aware metadata locks. To do this: - Changed MDL_lock to use one list for waiting requests and one list for granted requests. For each list, added a bitmap that holds information what lock types a list contains. Added a helper class MDL_lock::List to manipulate with granted and waited lists while keeping the bitmaps in sync with list contents. - Changed lock-compatibility functions to use bitmaps that define compatibility. - Introduced a graph based deadlock detector inspired by waiting_threads.c from Maria implementation. - Now that we have a deadlock detector, and no longer have a global lock to protect individual lock objects, but rather use an rw lock per object, removed redundant code for upgrade, and the global read lock. Changed the MDL API to no longer require the caller to acquire the global intention exclusive lock by means of a separate method. Removed a few more methods that became redundant. - Removed deadlock detection heuristic, it has been made obsolete by the deadlock detector. - With operation-type-aware metadata locks, MDL subsystem has become aware of potential conflicts between DDL and open transactions. This made it possible to remove calls to mysql_abort_transactions_with_shared_lock() from acquisition paths for exclusive lock and lock upgrade. Now we can simply wait for these transactions to complete without fear of deadlock. Function mysql_lock_abort() has also become unnecessary for all conflicting cases except when a DDL conflicts with a connection that has an open HANDLER. sql/mdl.h: Introduced new type-of-operation aware metadata locks. Introduced a graph based deadlock detector and supporting methods. Added comments. God rid of redundant API calls. Renamed m_lt_or_ha_sentinel to m_trans_sentinel, since now it guards the global read lock as well as LOCK TABLES and HANDLER locks. sql/mysql_priv.h: Moved the global read lock functionality into a class. Added MYSQL_OPEN_FORCE_SHARED_MDL flag which forces open_tables() to take MDL_SHARED on tables instead of metadata locks specified in the parser. We use this to allow PREPARE run concurrently in presence of LOCK TABLES ... WRITE. Added signature for find_table_for_mdl_ugprade(). sql/set_var.cc: Global read lock functionality has been moved into a class. sql/sp_head.cc: When creating TABLE_LIST elements for prelocking or system tables set the type of request for metadata lock according to the operation that will be performed on the table. sql/sql_base.cc: - Updated code to use the new MDL API. - In order to avoid locks starvation we take upgradable locks all at once. As result implicitly locked tables no longer get an upgradable lock. Consequently DDL and FLUSH TABLES for such tables is prohibited. find_write_locked_table() was replaced by find_table_for_mdl_upgrade() function. open_table() was adjusted to return TABLE instance with upgradable ticket when necessary. - We no longer wait for all locks on OT_WAIT back off action -- only on the lock that caused the wait conflict. Moreover, now we distinguish cases when we have to wait due to conflict in MDL and old version of table in TDC. - Upate mysql_notify_threads_having_share_locks() to only abort thr_lock.c waits of threads that have open HANDLERs, since lock conflicts with only these threads now can lead to deadlocks not detectable by the MDL deadlock detector. - Remove mysql_abort_transactions_with_shared_locks() which is no longer needed. sql/sql_class.cc: Global read lock functionality has been moved into a class. Re-arranged code in THD::cleanup() to simplify assert. sql/sql_class.h: Introduced class to incapsulate global read lock functionality. Now sentinel in MDL subsystem guards the global read lock as well as LOCK TABLES and HANDLER locks. Adjusted code accordingly. sql/sql_db.cc: Global read lock functionality has been moved into a class. sql/sql_delete.cc: We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result TRUNCATE TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_handler.cc: Inform MDL_context about presence of open HANDLERs. Since HANLDERs break MDL protocol by acquiring table-level lock while holding only S metadata lock on a table MDL subsystem should take special care about such contexts (Now this is the only case when mysql_lock_abort() is used). sql/sql_parse.cc: Global read lock functionality has been moved into a class. Do not take upgradable metadata locks when opening tables for CREATE TABLE SELECT as it is not necessary and limits concurrency. When initializing TABLE_LIST objects before adding them to the table list set the type of request for metadata lock according to the operation that will be performed on the table. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result FLUSH TABLES is no longer allowed for such tables. sql/sql_prepare.cc: Use MYSQL_OPEN_FORCE_SHARED_MDL flag when opening tables during PREPARE. This allows PREPARE to run concurrently in presence of LOCK TABLES ... WRITE. sql/sql_rename.cc: Global read lock functionality has been moved into a class. sql/sql_show.cc: Updated code to use the new MDL API. sql/sql_table.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result DROP TABLE is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_trigger.cc: Global read lock functionality has been moved into a class. We no longer acquire upgradable metadata locks on tables which are locked by LOCK TABLES implicitly. As result CREATE/DROP TRIGGER is no longer allowed for such tables. Updated code to use the new MDL API. sql/sql_view.cc: Global read lock functionality has been moved into a class. Fixed results of wrong merge that led to misuse of GLR API. CREATE VIEW statement is not a commit statement. sql/table.cc: When resetting TABLE_LIST objects for PS or SP re-execution set the type of request for metadata lock according to the operation that will be performed on the table. Do the same in auxiliary function initializing metadata lock requests in a table list. sql/table.h: When initializing TABLE_LIST objects set the type of request for metadata lock according to the operation that will be performed on the table. sql/transaction.cc: Global read lock functionality has been moved into a class.
2010-02-01 12:43:06 +01:00
--source include/wait_condition.inc
--echo # The below statement should succeed. It should not
--echo # deadlock or end with ER_LOCK_DEADLOCK error.
update t1 set c3=c3+1 where c2=4;
--echo # Unblock ALTER TABLE by committing transaction.
commit;
connection con46272;
--echo # Reaping ALTER TABLE.
--reap
connection default;
disconnect con46272;
drop table t1;
2009-12-09 10:44:01 +01:00
--echo #
--echo # Bug#47249 assert in MDL_global_lock::is_lock_type_compatible
--echo #
--disable_warnings
DROP TABLE IF EXISTS t1;
DROP VIEW IF EXISTS v1;
--enable_warnings
--echo #
--echo # Test 1: LOCK TABLES v1 WRITE, t1 READ;
--echo #
CREATE TABLE t1 ( f1 integer );
CREATE VIEW v1 AS SELECT f1 FROM t1 ;
--echo # Connection 2
connect (con2,localhost,root);
2009-12-09 10:44:01 +01:00
LOCK TABLES v1 WRITE, t1 READ;
FLUSH TABLE t1;
disconnect con2;
--source include/wait_until_disconnected.inc
--echo # Connection 1
connection default;
LOCK TABLES t1 WRITE;
FLUSH TABLE t1; # Assertion happened here
2009-12-09 10:44:01 +01:00
# Cleanup
DROP TABLE t1;
DROP VIEW v1;
--echo #
--echo # Test 2: LOCK TABLES t1 WRITE, v1 READ;
--echo #
CREATE TABLE t1 ( f1 integer );
CREATE VIEW v1 AS SELECT f1 FROM t1 ;
connect (con2,localhost,root);
LOCK TABLES t1 WRITE, v1 READ;
FLUSH TABLE t1;
disconnect con2;
--source include/wait_until_disconnected.inc
connection default;
LOCK TABLES t1 WRITE;
FLUSH TABLE t1; # Assertion happened here
# Cleanup
DROP TABLE t1;
DROP VIEW v1;
Fix for bug #50913 "Deadlock between open_and_lock_tables_derived and MDL". Concurrent execution of a multi-DELETE statement and ALTER TABLE statement which affected one of the tables used in the multi-DELETE sometimes led to deadlock. Similar deadlocks might have occured when one performed INSERT/UPDATE/DELETE on a view and concurrently executed ALTER TABLE for the view's underlying table, or when one concurrently executed TRUNCATE TABLE for InnoDB table and ALTER TABLE for the same table. These deadlocks were caused by a discrepancy between types of metadata and thr_lock.cc locks acquired by those statements. What happened was that multi-DELETE/TRUNCATE/DML-through-the- view statement in the first connection acquired SR lock on a table, then ALTER TABLE would come in in the second connection and acquire SNW metadata lock and TL_WRITE_ALLOW_READ thr_lock.c lock and then would start waiting for the first connection during lock upgrade. After that the statement in the first connection would try to acquire TL_WRITE lock on table and would start waiting for the second connection, creating a deadlock. This patch solves this problem by ensuring that we acquire SW metadata lock in all cases in which we acquiring write thr_lock.c lock. This guarantees that deadlocks like the one described above won't occur since all lock conflicts in such situation are resolved within MDL subsystem. This patch also adds assert which should guarantee that such situations won't arise in future. mysql-test/r/lock_multi.result: Added main test for bug #50913 "Deadlock between open_and_lock_tables_derived and MDL". mysql-test/r/mdl_sync.result: Added additional coverage for bug #50913 "Deadlock between open_and_lock_tables_derived and MDL". mysql-test/t/lock_multi.test: Added main test for bug #50913 "Deadlock between open_and_lock_tables_derived and MDL". mysql-test/t/mdl_sync.test: Added additional coverage for bug #50913 "Deadlock between open_and_lock_tables_derived and MDL". sql/lock.cc: Added assert that enforces that when we are locking a non-temporary table we have an appropriate type of metadata lock on this table. sql/mysql_priv.h: Added separate flag for open_tables() to be able specify that SH metadata locks on table to be open should be acquired. We can no longer use MYSQL_LOCK_IGNORE_FLUSH flag for this as in addition to use in I_S implementation it is also used for opening system tables. Since in the latter case we also acquire thr_lock.c locks using SH metadata lock in it instead of SR or SW locks may lead to deadlock. sql/sql_base.cc: When opening tables don't interpret MYSQL_LOCK_IGNORE_FLUSH flag as request to acquire SH metadata locks. This flag is also used for opening system tables for which we also take thr_lock.c locks and thus proper metadata lock to take in this case is SR or SW lock (otherwise deadlocks can occur). In cases when SH lock is really required (e.g. when tables are open by I_S implementation) we rely on that newly introduced MYSQL_OPEN_FORCE_SHARED_HIGH_PRIO_MDL flag is used. sql/sql_delete.cc: mysql_truncate_by_delete(): Adjust type of metadata lock to be requested after changing type of thr_lock.c lock for table list element from one which was set in parser to TL_WRITE. This removes discrepancy between types of these locks which allowed deadlocks to creep in. sql/sql_handler.cc: When closing table which was open by HANDLER statement clear TABLE::open_by_handler flag. This allows to use this flag as a reliable indication that TABLE instance was open by HANDLER statement in assert which was added to mysql_lock_tables(). sql/sql_parse.cc: multi_delete_set_locks_and_link_aux_tables(): Adjust type of metadata lock to be requested after changing type of thr_lock.c lock for table list element from one which was set in parser to TL_WRITE. This removes discrepancy between types of these locks which allowed deadlocks to creep in. sql/sql_show.cc: Use newly introduced MYSQL_OPEN_FORCE_SHARED_HIGH_PRIO_MDL flag in order to acquire SH metadata locks when opening tables in I_S implementation. sql/sql_update.cc: Added comment explaining why in multi-update after deciding that we need weaker thr_lock.c lock on a table we don't downgrade metadata lock on it. sql/sql_view.cc: When merging view into main statement adjust type of metadata lock to be requested after changing type of thr_lock.c lock for table. This removes discrepancy between types of these locks which allowed deadlocks to creep in.
2010-02-08 21:19:55 +01:00
--echo #
--echo # Test for bug #50913 "Deadlock between open_and_lock_tables_derived
--echo # and MDL". Also see additional coverage in mdl_sync.test.
--echo #
--disable_warnings
drop table if exists t1;
drop view if exists v1;
--enable_warnings
connect (con50913,localhost,root);
connection default;
create table t1 (i int);
create view v1 as select i from t1;
begin;
select * from t1;
connection con50913;
--echo # Sending:
--send alter table t1 add column j int
connection default;
--echo # Wait until ALTER TABLE gets blocked.
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
where state = "Waiting for table metadata lock" and
info = "alter table t1 add column j int";
Fix for bug #50913 "Deadlock between open_and_lock_tables_derived and MDL". Concurrent execution of a multi-DELETE statement and ALTER TABLE statement which affected one of the tables used in the multi-DELETE sometimes led to deadlock. Similar deadlocks might have occured when one performed INSERT/UPDATE/DELETE on a view and concurrently executed ALTER TABLE for the view's underlying table, or when one concurrently executed TRUNCATE TABLE for InnoDB table and ALTER TABLE for the same table. These deadlocks were caused by a discrepancy between types of metadata and thr_lock.cc locks acquired by those statements. What happened was that multi-DELETE/TRUNCATE/DML-through-the- view statement in the first connection acquired SR lock on a table, then ALTER TABLE would come in in the second connection and acquire SNW metadata lock and TL_WRITE_ALLOW_READ thr_lock.c lock and then would start waiting for the first connection during lock upgrade. After that the statement in the first connection would try to acquire TL_WRITE lock on table and would start waiting for the second connection, creating a deadlock. This patch solves this problem by ensuring that we acquire SW metadata lock in all cases in which we acquiring write thr_lock.c lock. This guarantees that deadlocks like the one described above won't occur since all lock conflicts in such situation are resolved within MDL subsystem. This patch also adds assert which should guarantee that such situations won't arise in future. mysql-test/r/lock_multi.result: Added main test for bug #50913 "Deadlock between open_and_lock_tables_derived and MDL". mysql-test/r/mdl_sync.result: Added additional coverage for bug #50913 "Deadlock between open_and_lock_tables_derived and MDL". mysql-test/t/lock_multi.test: Added main test for bug #50913 "Deadlock between open_and_lock_tables_derived and MDL". mysql-test/t/mdl_sync.test: Added additional coverage for bug #50913 "Deadlock between open_and_lock_tables_derived and MDL". sql/lock.cc: Added assert that enforces that when we are locking a non-temporary table we have an appropriate type of metadata lock on this table. sql/mysql_priv.h: Added separate flag for open_tables() to be able specify that SH metadata locks on table to be open should be acquired. We can no longer use MYSQL_LOCK_IGNORE_FLUSH flag for this as in addition to use in I_S implementation it is also used for opening system tables. Since in the latter case we also acquire thr_lock.c locks using SH metadata lock in it instead of SR or SW locks may lead to deadlock. sql/sql_base.cc: When opening tables don't interpret MYSQL_LOCK_IGNORE_FLUSH flag as request to acquire SH metadata locks. This flag is also used for opening system tables for which we also take thr_lock.c locks and thus proper metadata lock to take in this case is SR or SW lock (otherwise deadlocks can occur). In cases when SH lock is really required (e.g. when tables are open by I_S implementation) we rely on that newly introduced MYSQL_OPEN_FORCE_SHARED_HIGH_PRIO_MDL flag is used. sql/sql_delete.cc: mysql_truncate_by_delete(): Adjust type of metadata lock to be requested after changing type of thr_lock.c lock for table list element from one which was set in parser to TL_WRITE. This removes discrepancy between types of these locks which allowed deadlocks to creep in. sql/sql_handler.cc: When closing table which was open by HANDLER statement clear TABLE::open_by_handler flag. This allows to use this flag as a reliable indication that TABLE instance was open by HANDLER statement in assert which was added to mysql_lock_tables(). sql/sql_parse.cc: multi_delete_set_locks_and_link_aux_tables(): Adjust type of metadata lock to be requested after changing type of thr_lock.c lock for table list element from one which was set in parser to TL_WRITE. This removes discrepancy between types of these locks which allowed deadlocks to creep in. sql/sql_show.cc: Use newly introduced MYSQL_OPEN_FORCE_SHARED_HIGH_PRIO_MDL flag in order to acquire SH metadata locks when opening tables in I_S implementation. sql/sql_update.cc: Added comment explaining why in multi-update after deciding that we need weaker thr_lock.c lock on a table we don't downgrade metadata lock on it. sql/sql_view.cc: When merging view into main statement adjust type of metadata lock to be requested after changing type of thr_lock.c lock for table. This removes discrepancy between types of these locks which allowed deadlocks to creep in.
2010-02-08 21:19:55 +01:00
--source include/wait_condition.inc
--echo # The below statement should try to acquire SW lock on 't1'
--echo # and therefore should get ER_LOCK_DEADLOCK error. Before
--echo # bug fix it acquired SR lock and hung on thr_lock.c lock.
--error ER_LOCK_DEADLOCK
delete a from t1 as a where i = 1;
--echo # Unblock ALTER TABLE.
commit;
connection con50913;
--echo # Reaping ALTER TABLE;
--reap
connection default;
begin;
select * from v1;
connection con50913;
--echo # Sending:
--send alter table t1 drop column j
connection default;
--echo # Wait until ALTER TABLE gets blocked.
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
where state = "Waiting for table metadata lock" and
info = "alter table t1 drop column j";
Fix for bug #50913 "Deadlock between open_and_lock_tables_derived and MDL". Concurrent execution of a multi-DELETE statement and ALTER TABLE statement which affected one of the tables used in the multi-DELETE sometimes led to deadlock. Similar deadlocks might have occured when one performed INSERT/UPDATE/DELETE on a view and concurrently executed ALTER TABLE for the view's underlying table, or when one concurrently executed TRUNCATE TABLE for InnoDB table and ALTER TABLE for the same table. These deadlocks were caused by a discrepancy between types of metadata and thr_lock.cc locks acquired by those statements. What happened was that multi-DELETE/TRUNCATE/DML-through-the- view statement in the first connection acquired SR lock on a table, then ALTER TABLE would come in in the second connection and acquire SNW metadata lock and TL_WRITE_ALLOW_READ thr_lock.c lock and then would start waiting for the first connection during lock upgrade. After that the statement in the first connection would try to acquire TL_WRITE lock on table and would start waiting for the second connection, creating a deadlock. This patch solves this problem by ensuring that we acquire SW metadata lock in all cases in which we acquiring write thr_lock.c lock. This guarantees that deadlocks like the one described above won't occur since all lock conflicts in such situation are resolved within MDL subsystem. This patch also adds assert which should guarantee that such situations won't arise in future. mysql-test/r/lock_multi.result: Added main test for bug #50913 "Deadlock between open_and_lock_tables_derived and MDL". mysql-test/r/mdl_sync.result: Added additional coverage for bug #50913 "Deadlock between open_and_lock_tables_derived and MDL". mysql-test/t/lock_multi.test: Added main test for bug #50913 "Deadlock between open_and_lock_tables_derived and MDL". mysql-test/t/mdl_sync.test: Added additional coverage for bug #50913 "Deadlock between open_and_lock_tables_derived and MDL". sql/lock.cc: Added assert that enforces that when we are locking a non-temporary table we have an appropriate type of metadata lock on this table. sql/mysql_priv.h: Added separate flag for open_tables() to be able specify that SH metadata locks on table to be open should be acquired. We can no longer use MYSQL_LOCK_IGNORE_FLUSH flag for this as in addition to use in I_S implementation it is also used for opening system tables. Since in the latter case we also acquire thr_lock.c locks using SH metadata lock in it instead of SR or SW locks may lead to deadlock. sql/sql_base.cc: When opening tables don't interpret MYSQL_LOCK_IGNORE_FLUSH flag as request to acquire SH metadata locks. This flag is also used for opening system tables for which we also take thr_lock.c locks and thus proper metadata lock to take in this case is SR or SW lock (otherwise deadlocks can occur). In cases when SH lock is really required (e.g. when tables are open by I_S implementation) we rely on that newly introduced MYSQL_OPEN_FORCE_SHARED_HIGH_PRIO_MDL flag is used. sql/sql_delete.cc: mysql_truncate_by_delete(): Adjust type of metadata lock to be requested after changing type of thr_lock.c lock for table list element from one which was set in parser to TL_WRITE. This removes discrepancy between types of these locks which allowed deadlocks to creep in. sql/sql_handler.cc: When closing table which was open by HANDLER statement clear TABLE::open_by_handler flag. This allows to use this flag as a reliable indication that TABLE instance was open by HANDLER statement in assert which was added to mysql_lock_tables(). sql/sql_parse.cc: multi_delete_set_locks_and_link_aux_tables(): Adjust type of metadata lock to be requested after changing type of thr_lock.c lock for table list element from one which was set in parser to TL_WRITE. This removes discrepancy between types of these locks which allowed deadlocks to creep in. sql/sql_show.cc: Use newly introduced MYSQL_OPEN_FORCE_SHARED_HIGH_PRIO_MDL flag in order to acquire SH metadata locks when opening tables in I_S implementation. sql/sql_update.cc: Added comment explaining why in multi-update after deciding that we need weaker thr_lock.c lock on a table we don't downgrade metadata lock on it. sql/sql_view.cc: When merging view into main statement adjust type of metadata lock to be requested after changing type of thr_lock.c lock for table. This removes discrepancy between types of these locks which allowed deadlocks to creep in.
2010-02-08 21:19:55 +01:00
--source include/wait_condition.inc
--echo # The below statement should try to acquire SW lock on 't1'
--echo # and therefore should get ER_LOCK_DEADLOCK error. Before
--echo # bug fix it acquired SR lock and hung on thr_lock.c lock.
--error ER_LOCK_DEADLOCK
insert into v1 values (1);
--echo # Unblock ALTER TABLE.
commit;
connection con50913;
--echo # Reaping ALTER TABLE;
--reap
connection default;
disconnect con50913;
drop view v1;
drop table t1;
Bug #45225 Locking: hang if drop table with no timeout This patch introduces timeouts for metadata locks. The timeout is specified in seconds using the new dynamic system variable "lock_wait_timeout" which has both GLOBAL and SESSION scopes. Allowed values range from 1 to 31536000 seconds (= 1 year). The default value is 1 year. The new server parameter "lock-wait-timeout" can be used to set the default value parameter upon server startup. "lock_wait_timeout" applies to all statements that use metadata locks. These include DML and DDL operations on tables, views, stored procedures and stored functions. They also include LOCK TABLES, FLUSH TABLES WITH READ LOCK and HANDLER statements. The patch also changes thr_lock.c code (table data locks used by MyISAM and other simplistic engines) to use the same system variable. InnoDB row locks are unaffected. One exception to the handling of the "lock_wait_timeout" variable is delayed inserts. All delayed inserts are executed with a timeout of 1 year regardless of the setting for the global variable. As the connection issuing the delayed insert gets no notification of delayed insert timeouts, we want to avoid unnecessary timeouts. It's important to note that the timeout value is used for each lock acquired and that one statement can take more than one lock. A statement can therefore block for longer than the lock_wait_timeout value before reporting a timeout error. When lock timeout occurs, ER_LOCK_WAIT_TIMEOUT is reported. Test case added to lock_multi.test. include/my_pthread.h: Added macros for comparing two timespec structs. include/thr_lock.h: Introduced timeouts for thr_lock.c locks. mysql-test/r/mysqld--help-notwin.result: Updated result file with the new server variable. mysql-test/r/mysqld--help-win.result: Updated result file with the new server variable. mysql-test/suite/sys_vars/r/lock_wait_timeout_basic.result: Added basic test for the new server variable. mysql-test/suite/sys_vars/t/lock_wait_timeout_basic.test: Added basic test for the new server variable. mysys/thr_lock.c: Introduced timeouts for thr_lock.c locks. sql/mdl.cc: Introduced timeouts for metadata locks. sql/mdl.h: Introduced timeouts for metadata locks. sql/sql_base.cc: Introduced timeouts in tdc_wait_for_old_versions(). sql/sql_class.h: Added new server variable lock_wait_timeout. sql/sys_vars.cc: Added new server variable lock_wait_timeout.
2010-02-11 11:23:39 +01:00
--echo #
--echo # Bug#45225 Locking: hang if drop table with no timeout
--echo #
--echo # These tests also provide function coverage for the
--echo # lock_wait_timeout server variable.
--echo #
--disable_warnings
DROP TABLE IF EXISTS t1;
--enable_warnings
CREATE TABLE t1 (id int);
connect(con2, localhost, root,,);
SET SESSION lock_wait_timeout= 1;
--echo #
--echo # Test 1: acquire exclusive lock
--echo #
connection default;
START TRANSACTION;
INSERT INTO t1 VALUES (1);
connection con2;
--error ER_LOCK_WAIT_TIMEOUT
DROP TABLE t1;
connection default;
COMMIT;
--echo #
--echo # Test 2: upgrade shared lock
--echo #
connection default;
START TRANSACTION;
SELECT * FROM t1;
connection con2;
--error ER_LOCK_WAIT_TIMEOUT
ALTER TABLE t1 RENAME TO t2;
connection default;
COMMIT;
--echo #
--echo # Test 3: acquire shared lock
--echo #
connection default;
LOCK TABLE t1 WRITE;
connection con2;
--error ER_LOCK_WAIT_TIMEOUT
INSERT INTO t1(id) VALUES (2);
connection default;
UNLOCK TABLES;
--echo #
--echo # Test 4: table level locks
--echo #
connection default;
LOCK TABLE t1 READ;
connection con2;
--error ER_LOCK_WAIT_TIMEOUT
INSERT INTO t1(id) VALUES(4);
connection default;
UNLOCK TABLES;
--echo #
--echo # Test 5: Waiting on Table Definition Cache (TDC)
--echo #
connect(con3, localhost, root);
connection default;
LOCK TABLE t1 READ;
connection con3;
--echo # Sending:
--send FLUSH TABLES
connection con2;
let $wait_condition=
SELECT COUNT(*) = 1 FROM information_schema.processlist
WHERE state = "Waiting for table flush" AND info = "FLUSH TABLES";
Bug #45225 Locking: hang if drop table with no timeout This patch introduces timeouts for metadata locks. The timeout is specified in seconds using the new dynamic system variable "lock_wait_timeout" which has both GLOBAL and SESSION scopes. Allowed values range from 1 to 31536000 seconds (= 1 year). The default value is 1 year. The new server parameter "lock-wait-timeout" can be used to set the default value parameter upon server startup. "lock_wait_timeout" applies to all statements that use metadata locks. These include DML and DDL operations on tables, views, stored procedures and stored functions. They also include LOCK TABLES, FLUSH TABLES WITH READ LOCK and HANDLER statements. The patch also changes thr_lock.c code (table data locks used by MyISAM and other simplistic engines) to use the same system variable. InnoDB row locks are unaffected. One exception to the handling of the "lock_wait_timeout" variable is delayed inserts. All delayed inserts are executed with a timeout of 1 year regardless of the setting for the global variable. As the connection issuing the delayed insert gets no notification of delayed insert timeouts, we want to avoid unnecessary timeouts. It's important to note that the timeout value is used for each lock acquired and that one statement can take more than one lock. A statement can therefore block for longer than the lock_wait_timeout value before reporting a timeout error. When lock timeout occurs, ER_LOCK_WAIT_TIMEOUT is reported. Test case added to lock_multi.test. include/my_pthread.h: Added macros for comparing two timespec structs. include/thr_lock.h: Introduced timeouts for thr_lock.c locks. mysql-test/r/mysqld--help-notwin.result: Updated result file with the new server variable. mysql-test/r/mysqld--help-win.result: Updated result file with the new server variable. mysql-test/suite/sys_vars/r/lock_wait_timeout_basic.result: Added basic test for the new server variable. mysql-test/suite/sys_vars/t/lock_wait_timeout_basic.test: Added basic test for the new server variable. mysys/thr_lock.c: Introduced timeouts for thr_lock.c locks. sql/mdl.cc: Introduced timeouts for metadata locks. sql/mdl.h: Introduced timeouts for metadata locks. sql/sql_base.cc: Introduced timeouts in tdc_wait_for_old_versions(). sql/sql_class.h: Added new server variable lock_wait_timeout. sql/sys_vars.cc: Added new server variable lock_wait_timeout.
2010-02-11 11:23:39 +01:00
--source include/wait_condition.inc
--error ER_LOCK_WAIT_TIMEOUT
SELECT * FROM t1;
connection default;
UNLOCK TABLES;
connection con3;
--echo # Reaping: FLUSH TABLES
--reap
--echo #
--echo # Test 6: Timeouts in I_S queries
--echo #
connection default;
CREATE TABLE t2 (id INT);
LOCK TABLE t2 WRITE;
connection con3;
--echo # Sending:
--send DROP TABLE t1, t2
connection con2;
let $wait_condition=
SELECT COUNT(*) = 1 FROM information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
WHERE state = "Waiting for table metadata lock" AND
info = "DROP TABLE t1, t2";
Bug #45225 Locking: hang if drop table with no timeout This patch introduces timeouts for metadata locks. The timeout is specified in seconds using the new dynamic system variable "lock_wait_timeout" which has both GLOBAL and SESSION scopes. Allowed values range from 1 to 31536000 seconds (= 1 year). The default value is 1 year. The new server parameter "lock-wait-timeout" can be used to set the default value parameter upon server startup. "lock_wait_timeout" applies to all statements that use metadata locks. These include DML and DDL operations on tables, views, stored procedures and stored functions. They also include LOCK TABLES, FLUSH TABLES WITH READ LOCK and HANDLER statements. The patch also changes thr_lock.c code (table data locks used by MyISAM and other simplistic engines) to use the same system variable. InnoDB row locks are unaffected. One exception to the handling of the "lock_wait_timeout" variable is delayed inserts. All delayed inserts are executed with a timeout of 1 year regardless of the setting for the global variable. As the connection issuing the delayed insert gets no notification of delayed insert timeouts, we want to avoid unnecessary timeouts. It's important to note that the timeout value is used for each lock acquired and that one statement can take more than one lock. A statement can therefore block for longer than the lock_wait_timeout value before reporting a timeout error. When lock timeout occurs, ER_LOCK_WAIT_TIMEOUT is reported. Test case added to lock_multi.test. include/my_pthread.h: Added macros for comparing two timespec structs. include/thr_lock.h: Introduced timeouts for thr_lock.c locks. mysql-test/r/mysqld--help-notwin.result: Updated result file with the new server variable. mysql-test/r/mysqld--help-win.result: Updated result file with the new server variable. mysql-test/suite/sys_vars/r/lock_wait_timeout_basic.result: Added basic test for the new server variable. mysql-test/suite/sys_vars/t/lock_wait_timeout_basic.test: Added basic test for the new server variable. mysys/thr_lock.c: Introduced timeouts for thr_lock.c locks. sql/mdl.cc: Introduced timeouts for metadata locks. sql/mdl.h: Introduced timeouts for metadata locks. sql/sql_base.cc: Introduced timeouts in tdc_wait_for_old_versions(). sql/sql_class.h: Added new server variable lock_wait_timeout. sql/sys_vars.cc: Added new server variable lock_wait_timeout.
2010-02-11 11:23:39 +01:00
--source include/wait_condition.inc
# Note: This query causes two timeouts.
# 1: try_acquire_high_prio_shared_mdl_lock on t1
# 2: recover_from_failed_open on t1
SELECT table_name, table_comment FROM information_schema.tables
WHERE table_schema= 'test' AND table_name= 't1';
connection default;
UNLOCK TABLES;
connection con3;
--echo # Reaping: DROP TABLE t1, t2
--reap
connection default;
--echo # Cleanup
disconnect con2;
disconnect con3;
Fix for bug #51134 "Crash in MDL_lock::destroy on a concurrent DDL workload". When a RENAME TABLE or LOCK TABLE ... WRITE statement which mentioned the same table several times were aborted during the process of acquring metadata locks (due to deadlock which was discovered or because of KILL statement) server might have crashed. When attempt to acquire all locks requested had failed we went through the list of requests and released locks which we have managed to acquire by that moment one by one. Since in the scenario described above list of requests contained duplicates this led to releasing the same ticket twice and a crash as result. This patch solves the problem by employing different approach to releasing locks in case of failure to acquire all locks requested. Now we take a MDL savepoint before starting acquiring locks and simply rollback to it if things go bad. mysql-test/r/lock_multi.result: Updated test results (see lock_multi.test). mysql-test/t/lock_multi.test: Added test case for bug #51134 "Crash in MDL_lock::destroy on a concurrent DDL workload". sql/mdl.cc: MDL_context::acquire_locks(): When attempt to acquire all locks requested has failed do not go through the list of requests and release locks which we have managed to acquire one by one. Since list of requests can contain duplicates such approach may lead to releasing the same ticket twice and a crash as result. Instead use the following approach - take a MDL savepoint before starting acquiring locks and simply rollback to it if things go bad.
2010-02-15 11:23:34 +01:00
--echo #
--echo # Test for bug #51134 "Crash in MDL_lock::destroy on a concurrent
--echo # DDL workload".
--echo #
--disable_warnings
drop tables if exists t1, t2, t3;
--enable_warnings
connect (con1, localhost, root, , );
connect (con2, localhost, root, , );
connection default;
create table t3 (i int);
connection con1;
--echo # Lock 't3' so upcoming RENAME is blocked.
lock table t3 read;
connection con2;
--echo # Remember ID for this connection.
let $ID= `select connection_id()`;
--echo # Start statement which will try to acquire two instances
--echo # of X metadata lock on the same object.
--echo # Sending:
--send rename tables t1 to t2, t2 to t3;
connection default;
--echo # Wait until RENAME TABLE is blocked on table 't3'.
let $wait_condition=
select count(*) = 1 from information_schema.processlist
Part of fix for bug#52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES <list> WITH READ LOCK are incompatible" to be pushed as separate patch. Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. Also replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it more consistent with other thread state names. Updated test cases and their results according to these changes. Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. mysql-test/r/query_cache.result: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysql-test/suite/sys_vars/r/query_cache_wlock_invalidate_func.result: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/suite/sys_vars/t/query_cache_wlock_invalidate_func.test: Fixed sys_vars.query_cache_wlock_invalidate_func test to not to wait for timeout of wait_condition.inc script. Reverted changes to test which introduced timeout and replaced waiting condition with a more appropriate one. Test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables was added to query_cache.test. mysql-test/t/query_cache.test: Added test coverage for query_cache_wlock_invalidate behavior for implicitly locked tables. mysys/thr_lock.c: Replaced "Table lock" thread state name, which was used by threads waiting on thr_lock.c table level lock, with more elaborate "Waiting for table level lock", to make it consistent with thread state names which are used while waiting for metadata locks and table flush. sql/mdl.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for a metadata lock or table flush, with a set of names which better reflect types of resources being waited for. To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/mdl.h: To implement this: - Adjusted MDL_wait::timed_wait() to take thread state name as parameter. - Introduced method of MDL_key class which allows to get thread state name to be used while waiting for resource corresponding to the key and changed code to use it. Added array translating namespaces to thread state names as part of this change. sql/sql_base.cc: Replaced thread state name "Waiting for table", which was used by threads waiting for table flush, with a more elaborate "Waiting for table flush".
2010-08-06 13:29:37 +02:00
where state = "Waiting for table metadata lock" and
info = "rename tables t1 to t2, t2 to t3";
Fix for bug #51134 "Crash in MDL_lock::destroy on a concurrent DDL workload". When a RENAME TABLE or LOCK TABLE ... WRITE statement which mentioned the same table several times were aborted during the process of acquring metadata locks (due to deadlock which was discovered or because of KILL statement) server might have crashed. When attempt to acquire all locks requested had failed we went through the list of requests and released locks which we have managed to acquire by that moment one by one. Since in the scenario described above list of requests contained duplicates this led to releasing the same ticket twice and a crash as result. This patch solves the problem by employing different approach to releasing locks in case of failure to acquire all locks requested. Now we take a MDL savepoint before starting acquiring locks and simply rollback to it if things go bad. mysql-test/r/lock_multi.result: Updated test results (see lock_multi.test). mysql-test/t/lock_multi.test: Added test case for bug #51134 "Crash in MDL_lock::destroy on a concurrent DDL workload". sql/mdl.cc: MDL_context::acquire_locks(): When attempt to acquire all locks requested has failed do not go through the list of requests and release locks which we have managed to acquire one by one. Since list of requests can contain duplicates such approach may lead to releasing the same ticket twice and a crash as result. Instead use the following approach - take a MDL savepoint before starting acquiring locks and simply rollback to it if things go bad.
2010-02-15 11:23:34 +01:00
--source include/wait_condition.inc
--echo # Kill RENAME TABLE.
--replace_result $ID ID
eval kill query $ID;
connection con2;
--echo # RENAME TABLE should be aborted but should not crash.
--error ER_QUERY_INTERRUPTED
--reap
connection con1;
unlock tables;
connection default;
disconnect con1;
disconnect con2;
drop table t3;
--echo #
--echo # Test for the bug where upgradable metadata locks was acquired
--echo # even if the table to altered was temporary.
--echo # Bug found while working on the related bug #51240.
--echo #
--disable_warnings
DROP TABLE IF EXISTS t1;
--enable_warnings
CREATE TABLE t1 (id INT);
LOCK TABLE t1 WRITE;
connect (con1, localhost, root);
CREATE TEMPORARY TABLE t1 (id INT);
# This alter should not block and timeout.
ALTER TABLE t1 ADD COLUMN j INT;
connection default;
disconnect con1;
UNLOCK TABLES;
DROP TABLE t1;
# Wait till all disconnects are completed
--source include/wait_until_count_sessions.inc