innodb_lock_waits. See
https://svn.innodb.com/innobase/InformationSchema/TransactionsAndLocks
for design notes.
Things that need to be resolved before this goes live:
* MySQL must add thd_get_thread_id() function to their code
http://bugs.mysql.com/30930
* Allocate memory from mem_heap instead of using mem_alloc()
* Copy table name and index name into the cache because they may be
freed later which will result in referencing freed memory
Approved by: Marko
existed before the fast-index-creation code was merged, in r1414.
Do not call row_prebuilt_table_obsolete(innodb_table) until after
the table has been renamed to a temporary name. Otherwise, the
following could happen when creating a clustered index on table T:
(a1) copy T to T#1
(a2) mark T obsolete
(b1) add the name of T to the background drop list
(a3) rename T to T#2 and T#1 to T
(b2) drop the table on the background
These steps are executed by two threads, a and b.
If b1 is executed after a3, all is fine: the name T#2 will be added
to the background drop list. If b1 is executed before a3, the name
T will be added, and T#2 will survive.
fast index creation.
innobase_convert_tablename(): New function, to remove the filename
encoding of a database and table name.
ha_innobase::add_index(): Invoke innobase_convert_tablename() before
calling my_error().
innodb-index.result: Update the error messages.
trx_t: Remove dict_undo_list and dict_redo_list.
innobase_create_temporary_tablename(): Replace TEMP_TABLE_PREFIX with
a table name suffix "#1" or "#2". In this way, the user can restore
precious data, should anything go wrong. It is possible to reach an
inconsistent state, because the creation, deletion and renaming of
single-table tablespaces are not transactional.
ut_print_namel(), fil_make_ibd_name(), innobase_rename_table(): Remove
the special treatment of TEMP_TABLE_PREFIX.
Introduce TEMP_INDEX_PREFIX == 0xff for temporary indexes. This byte
cannot occur in index names since MySQL 4.1. However, it might have
been possible to use this byte in MySQL 4.0.
recv_recovery_from_checkpoint_finish(): Call the new function
row_merge_drop_temp_indexes(), to drop all indexes whose name starts
with the byte 0xff.
row_merge_rename_indexes(): Renamed from row_merge_rename_index().
Remove the parameter "index".
row_drop_table_for_mysql(): Unconditionally call trx_commit_for_mysql().
row_drop_table_for_mysql_no_commit(): Correct the function commit,
based on the corrected comment of row_drop_table_for_mysql(). Rely on
table->to_be_dropped instead of TEMP_TABLE_PREFIX.
ha_innobase::add_index(): Simplify the control flow.
#ifdef ROW_MERGE_IS_INDEX_USABLE, to avoid growing the memory
footprint until MySQL implements and calls the virtual method
handler::is_index_available() or something equivalent.
in fast index creation. The prefix was previously TEMP_TABLE_PREFIX == '/',
which may occur in index names.
ut_print_namel(): Do not assume that all '/' are separators between
database and table names.
ha_innobase::add_index(): Add assertions about !trx->sync_cb and
!trx->dict_redo_list. Remove a bogus comment. Replace
row_lock_table_for_merge() with row_merge_lock_table().
row_merge_lock_table(): Renamed from row_lock_table_for_merge()
and moved from row0mysql.c to row0merge.c. Always lock the table in
the mode LOCK_X. Call que_graph_free() in order not to leak memory.
innodb-index.test: Add some test cases for FOREIGN KEY constraints.
row_merge_rename_tables(): New function,
adapted from row_rename_table_for_mysql().
ha_innobase::add_index(): Use row_merge_rename_tables().
have been set.
ha_innobase::add_index(), ha_innobase::prepare_drop_index(),
ha_innobase::final_drop_index(): Return HA_ERR_WRONG_COMMAND if
newraw or innodb_force_recovery have been set.
(PRIMARY KEY or UNIQUE INDEX on NOT NULL columns).
innodb-index.test: Remove the --replace_regex directives. Replace correct
values to the "Duplicate entry" messages. This will cause the test to
fail until we somehow return the duplicate key values to MySQL.
Add test cases for creating or dropping UNIQUE INDEX or PRIMARY KEY.
innobase_copy_index_def(): Allow the function to copy a clustered index
definition to a secondary one (UNIQUE INDEX).
innobase_create_key_def(): Check the NOT NULL flag from the correct
place. Copy the clustered index if it is a UNIQUE INDEX.
ha_innobase::prepare_drop_index(): Refuse to drop the clustered index.
ha_innobase::add_index(), ha_innobase::prepare_drop_index():
Remove the redundant thd_test_options() calls.
The options are tested in check_trx_exists().
ha_innobase::prepare_drop_index(): key cannot be NULL; remove the tests.
Move the condition for checking foreign keys outside of the loop.
Remove the redundant error log printout about foreign key dependency.
It was only set by ha_innobase::prepare_drop_index(), which can return
the appropriate MySQL error code (HA_ERR_DROP_INDEX_FK) directly.
ha_innobase::add_index(): Correct the function comment.
ha_innobase::prepare_drop_index(): Correct the function comment.
Return MySQL error codes directly.
ha_innobase::final_drop_index(): Correct the function comment.
trx_t: Change the type of error_info from void* to const dict_index_t*.
trx_get_error_info(): Add const qualifier to trx_t*. Make this an
inline function.
key name "PRIMARY". If the table has a generated clustered index,
also check for a UNIQUE INDEX containing no NULL columns.
innobase_create_index_def(): Add the parameter key_primary.
innobase_create_key_def(): Check if the first index is a UNIQUE index
consisting of NOT NULL columns.
Fixing this bug revealed another one: FOREIGN KEY constraints
will be lost when a PRIMARY KEY is created. This is the reason why
innodb.test will fail.
innobase_create_key_def(), row_table_got_default_clust_index(),
row_get_mysql_key_number_for_index(): Add a const qualifier.
dict_table_get_first_index(), dict_table_get_next_index(): Define
as a const-preserving macro. Preserve the old function for UNIV_DEBUG,
but add a const qualifier to the parameter and cast away the constness.
row_merge_create_temporary_table(): Add the parameter index_defs.
DB_PRIMARY_KEY_IS_NULL: New error code, mapped to ER_PRIMARY_CANT_HAVE_NULL.
row_merge_read_clustered_index(): Replace the parameter "table" with the
two parameters "old_table" and "new_table". Detect NULL values of columns
that are supposed to be NOT NULL.
row_merge_col_prtype(): New auxiliary function, used by
row_merge_create_temporary_table() to flag NOT NULL any columns
contained in an added PRIMARY KEY.
innobase_copy_index_field_def(): Remove parameter heap. The field->name
will not be deallocated before the copied index has been added to the
data dictionary cache.
innobase_copy_index_def(): Add const qualifier to dict_index_t*.
Add an assertion !dict_index_is_clust(index).
ha_innobase::add_index(): Add the missing user_thd parameter to the
convert_error_code_to_mysql() call that was forgotten from r1686.
merge_index_field_struct: Add const qualifier to field_name.
merge_index_def_struct: Add const qualifier to name.
row_merge_build_indexes(): Improve the function comment.
row_merge_drop_table(): Remove bogus assertion ut_a(table->to_be_dropped).
goto err_exit. The error handling should be rewritten. Many of the
error messages contain parameters, but this code is only returning
the error number to MySQL.
row_mysql_read_true_varchar(), row_mysql_read_blob_ref(),
row_mysql_store_col_in_innobase_format(),
row_sel_convert_mysql_key_to_innobase(): Add const qualifiers.
row_sel_convert_mysql_key_to_innobase(): Document that the first parameter
is in/out.
the data dictionary locked across the operations.
dict_table_decrement_handle_count(), row_prebuilt_free(): Add the flag
dict_locked, to prevent the acquisition of dict_sys->mutex.
innobase_rename_table(): Rename commit_flag to lock_and_commit,
and do not acquire dict_sys->mutex or flush the log unless the flag is set.
Remove bogus comment about utility threads, because the threads will
be waken up by the upper-level function ha_innobase::rename_table().
ha_innobase::add_index(): After creating a primary key, hold dict_sys->mutex
across all dictionary operations.
Some things still fail in innodb-index.test, and there seems to be
a race condition (data dictionary lock wait) when running with --valgrind.
dfield_t: Add an "external storage" flag, dfield->ext.
dfield_is_null(), dfield_is_ext(), dfield_set_ext(), dfield_set_null():
New functions.
dfield_copy(), dfield_copy_data(): Add const qualifiers, fix in/out comments.
data_write_sql_null(): Use memset().
big_rec_field_t: Replace byte* data with const void* data.
ut_ulint_sort(): Remove.
upd_field_t: Remove extern_storage.
upd_node_t: Replace ext_vec, n_ext_vec with n_ext.
row_merge_copy_blobs(): New function.
row_ins_index_entry(): Add the parameter "ibool foreign" for suppressing
foreign key checks during fast index creation or when inserting into
secondary indexes.
btr_page_insert_fits(): Add const qualifiers.
btr_cur_add_ext(), upd_ext_vec_contains(): Remove.
dfield_print_also_hex(), dfield_print(): Replace if...else if with switch.
Observe dfield_is_ext().
ha_innobase::add_index(): Do commit the transaction, because the error
handler (label err1) in mysql_alter_table() won't commit it as of now.
This was detected in a failure of type_bit_innodb.
now that all of InnoDB code is built from a single Makefile and it should
not be possible to build the modules with mutually incompatible options.
#define INSIDE_HA_INNOBASE_CC: Remove.
srv_sizeof_trx_t_in_ha_innodb_cc: Remove.
dict_table_get_low_noninlined(): Remove. This function was unused.
Remove all _noninline functions. Remove the _noninline suffix from
all function calls in ha_innodb.cc.
Remove the variable parent_trx.
Call innobase_register_stmt() in order to work around a MySQL bug
in mysql_alter_table(), which, as of
ChangeSet@1.2482.61.2, 2007-06-07 16:37:15+02:00, joerg@trift2. +8 -0
commits the transaction before calling ha_innobase::add_index().
Without re-registering the statement, the ha_commit_stmt(thd)
in mysql_alter_table() would not invoke innobase_commit.
