2009-03-25 23:11:11 -07:00
|
|
|
/*****************************************************************************
|
|
|
|
|
2016-08-10 19:24:58 +02:00
|
|
|
Copyright (c) 1996, 2016, Oracle and/or its affiliates. All Rights Reserved.
|
MDEV-11623 MariaDB 10.1 fails to start datadir created with
MariaDB 10.0/MySQL 5.6 using innodb-page-size!=16K
The storage format of FSP_SPACE_FLAGS was accidentally broken
already in MariaDB 10.1.0. This fix is bringing the format in
line with other MySQL and MariaDB release series.
Please refer to the comments that were added to fsp0fsp.h
for details.
This is an INCOMPATIBLE CHANGE that affects users of
page_compression and non-default innodb_page_size. Upgrading
to this release will correct the flags in the data files.
If you want to downgrade to earlier MariaDB 10.1.x, please refer
to the test innodb.101_compatibility how to reset the
FSP_SPACE_FLAGS in the files.
NOTE: MariaDB 10.1.0 to 10.1.20 can misinterpret
uncompressed data files with innodb_page_size=4k or 64k as
compressed innodb_page_size=16k files, and then probably fail
when trying to access the pages. See the comments in the
function fsp_flags_convert_from_101() for detailed analysis.
Move PAGE_COMPRESSION to FSP_SPACE_FLAGS bit position 16.
In this way, compressed innodb_page_size=16k tablespaces will not
be mistaken for uncompressed ones by MariaDB 10.1.0 to 10.1.20.
Derive PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR from the
dict_table_t::flags when the table is available, in
fil_space_for_table_exists_in_mem() or fil_open_single_table_tablespace().
During crash recovery, fil_load_single_table_tablespace() will use
innodb_compression_level for the PAGE_COMPRESSION_LEVEL.
FSP_FLAGS_MEM_MASK: A bitmap of the memory-only fil_space_t::flags
that are not to be written to FSP_SPACE_FLAGS. Currently, these will
include PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR.
Introduce the macro FSP_FLAGS_PAGE_SSIZE(). We only support
one innodb_page_size for the whole instance.
When creating a dummy tablespace for the redo log, use
fil_space_t::flags=0. The flags are never written to the redo log files.
Remove many FSP_FLAGS_SET_ macros.
dict_tf_verify_flags(): Remove. This is basically only duplicating
the logic of dict_tf_to_fsp_flags(), used in a debug assertion.
fil_space_t::mark: Remove. This flag was not used for anything.
fil_space_for_table_exists_in_mem(): Remove the unnecessary parameter
mark_space, and add a parameter for table flags. Check that
fil_space_t::flags match the table flags, and adjust the (memory-only)
flags based on the table flags.
fil_node_open_file(): Remove some redundant or unreachable conditions,
do not use stderr for output, and avoid unnecessary server aborts.
fil_user_tablespace_restore_page(): Convert the flags, so that the
correct page_size will be used when restoring a page from the
doublewrite buffer.
fil_space_get_page_compressed(), fsp_flags_is_page_compressed(): Remove.
It suffices to have fil_space_is_page_compressed().
FSP_FLAGS_WIDTH_DATA_DIR, FSP_FLAGS_WIDTH_PAGE_COMPRESSION_LEVEL,
FSP_FLAGS_WIDTH_ATOMIC_WRITES: Remove, because these flags do not
exist in the FSP_SPACE_FLAGS but only in memory.
fsp_flags_try_adjust(): New function, to adjust the FSP_SPACE_FLAGS
in page 0. Called by fil_open_single_table_tablespace(),
fil_space_for_table_exists_in_mem(), innobase_start_or_create_for_mysql()
except if --innodb-read-only is active.
fsp_flags_is_valid(ulint): Reimplement from the scratch, with
accurate comments. Do not display any details of detected
inconsistencies, because the output could be confusing when
dealing with MariaDB 10.1.x data files.
fsp_flags_convert_from_101(ulint): Convert flags from buggy
MariaDB 10.1.x format, or return ULINT_UNDEFINED if the flags
cannot be in MariaDB 10.1.x format.
fsp_flags_match(): Check the flags when probing files.
Implemented based on fsp_flags_is_valid()
and fsp_flags_convert_from_101().
dict_check_tablespaces_and_store_max_id(): Do not access the
page after committing the mini-transaction.
IMPORT TABLESPACE fixes:
AbstractCallback::init(): Convert the flags.
FetchIndexRootPages::operator(): Check that the tablespace flags match the
table flags. Do not attempt to convert tablespace flags to table flags,
because the conversion would necessarily be lossy.
PageConverter::update_header(): Write back the correct flags.
This takes care of the flags in IMPORT TABLESPACE.
2017-01-14 00:13:16 +02:00
|
|
|
Copyright (c) 2013, 2017, MariaDB Corporation.
|
2009-03-25 23:11:11 -07:00
|
|
|
|
|
|
|
This program is free software; you can redistribute it and/or modify it under
|
|
|
|
the terms of the GNU General Public License as published by the Free Software
|
|
|
|
Foundation; version 2 of the License.
|
|
|
|
|
|
|
|
This program is distributed in the hope that it will be useful, but WITHOUT
|
|
|
|
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
|
|
|
|
FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
|
|
|
|
|
|
|
|
You should have received a copy of the GNU General Public License along with
|
2012-06-15 10:23:33 +02:00
|
|
|
this program; if not, write to the Free Software Foundation, Inc.,
|
|
|
|
51 Franklin Street, Suite 500, Boston, MA 02110-1335 USA
|
2009-03-25 23:11:11 -07:00
|
|
|
|
|
|
|
*****************************************************************************/
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/******************************************************************//**
|
|
|
|
@file include/dict0dict.ic
|
2008-11-30 22:10:29 -08:00
|
|
|
Data dictionary system
|
|
|
|
|
|
|
|
Created 1/8/1996 Heikki Tuuri
|
|
|
|
***********************************************************************/
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
#include "data0type.h"
|
|
|
|
#ifndef UNIV_HOTBACKUP
|
2008-11-30 22:10:29 -08:00
|
|
|
#include "dict0load.h"
|
|
|
|
#include "rem0types.h"
|
2013-12-16 15:38:05 +01:00
|
|
|
#include "fsp0fsp.h"
|
2011-12-14 14:58:22 +01:00
|
|
|
#include "srv0srv.h"
|
2013-12-16 15:38:05 +01:00
|
|
|
#include "sync0rw.h" /* RW_S_LATCH */
|
2008-11-30 22:10:29 -08:00
|
|
|
|
2011-07-14 21:22:41 +02:00
|
|
|
/*********************************************************************//**
|
|
|
|
Gets the minimum number of bytes per character.
|
|
|
|
@return minimum multi-byte char size, in bytes */
|
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_col_get_mbminlen(
|
|
|
|
/*==================*/
|
|
|
|
const dict_col_t* col) /*!< in: column */
|
|
|
|
{
|
|
|
|
return(DATA_MBMINLEN(col->mbminmaxlen));
|
|
|
|
}
|
|
|
|
/*********************************************************************//**
|
|
|
|
Gets the maximum number of bytes per character.
|
|
|
|
@return maximum multi-byte char size, in bytes */
|
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_col_get_mbmaxlen(
|
|
|
|
/*==================*/
|
|
|
|
const dict_col_t* col) /*!< in: column */
|
|
|
|
{
|
|
|
|
return(DATA_MBMAXLEN(col->mbminmaxlen));
|
|
|
|
}
|
|
|
|
/*********************************************************************//**
|
|
|
|
Sets the minimum and maximum number of bytes per character. */
|
|
|
|
UNIV_INLINE
|
|
|
|
void
|
|
|
|
dict_col_set_mbminmaxlen(
|
|
|
|
/*=====================*/
|
|
|
|
dict_col_t* col, /*!< in/out: column */
|
|
|
|
ulint mbminlen, /*!< in: minimum multi-byte
|
|
|
|
character size, in bytes */
|
|
|
|
ulint mbmaxlen) /*!< in: minimum multi-byte
|
|
|
|
character size, in bytes */
|
|
|
|
{
|
|
|
|
ut_ad(mbminlen < DATA_MBMAX);
|
|
|
|
ut_ad(mbmaxlen < DATA_MBMAX);
|
|
|
|
ut_ad(mbminlen <= mbmaxlen);
|
|
|
|
|
|
|
|
col->mbminmaxlen = DATA_MBMINMAXLEN(mbminlen, mbmaxlen);
|
|
|
|
}
|
2009-09-07 10:22:53 +00:00
|
|
|
/*********************************************************************//**
|
2008-11-30 22:10:29 -08:00
|
|
|
Gets the column data type. */
|
|
|
|
UNIV_INLINE
|
|
|
|
void
|
|
|
|
dict_col_copy_type(
|
|
|
|
/*===============*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_col_t* col, /*!< in: column */
|
|
|
|
dtype_t* type) /*!< out: data type */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
2016-08-10 19:24:58 +02:00
|
|
|
ut_ad(col != NULL);
|
|
|
|
ut_ad(type != NULL);
|
2008-11-30 22:10:29 -08:00
|
|
|
|
|
|
|
type->mtype = col->mtype;
|
|
|
|
type->prtype = col->prtype;
|
|
|
|
type->len = col->len;
|
2011-07-14 21:22:41 +02:00
|
|
|
type->mbminmaxlen = col->mbminmaxlen;
|
2008-11-30 22:10:29 -08:00
|
|
|
}
|
2009-09-07 10:22:53 +00:00
|
|
|
#endif /* !UNIV_HOTBACKUP */
|
2008-11-30 22:10:29 -08:00
|
|
|
|
|
|
|
#ifdef UNIV_DEBUG
|
2009-09-07 10:22:53 +00:00
|
|
|
/*********************************************************************//**
|
|
|
|
Assert that a column and a data type match.
|
|
|
|
@return TRUE */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ibool
|
|
|
|
dict_col_type_assert_equal(
|
|
|
|
/*=======================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_col_t* col, /*!< in: column */
|
|
|
|
const dtype_t* type) /*!< in: data type */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(col);
|
|
|
|
ut_ad(type);
|
|
|
|
|
|
|
|
ut_ad(col->mtype == type->mtype);
|
|
|
|
ut_ad(col->prtype == type->prtype);
|
2013-12-16 15:38:05 +01:00
|
|
|
//ut_ad(col->len == type->len);
|
2009-09-07 10:22:53 +00:00
|
|
|
# ifndef UNIV_HOTBACKUP
|
2011-07-14 21:22:41 +02:00
|
|
|
ut_ad(col->mbminmaxlen == type->mbminmaxlen);
|
2009-09-07 10:22:53 +00:00
|
|
|
# endif /* !UNIV_HOTBACKUP */
|
2008-11-30 22:10:29 -08:00
|
|
|
|
|
|
|
return(TRUE);
|
|
|
|
}
|
|
|
|
#endif /* UNIV_DEBUG */
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
#ifndef UNIV_HOTBACKUP
|
|
|
|
/***********************************************************************//**
|
|
|
|
Returns the minimum size of the column.
|
|
|
|
@return minimum size */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_col_get_min_size(
|
|
|
|
/*==================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_col_t* col) /*!< in: column */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
return(dtype_get_min_size_low(col->mtype, col->prtype, col->len,
|
2011-07-14 21:22:41 +02:00
|
|
|
col->mbminmaxlen));
|
2008-11-30 22:10:29 -08:00
|
|
|
}
|
2009-09-07 10:22:53 +00:00
|
|
|
/***********************************************************************//**
|
|
|
|
Returns the maximum size of the column.
|
|
|
|
@return maximum size */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_col_get_max_size(
|
|
|
|
/*==================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_col_t* col) /*!< in: column */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
return(dtype_get_max_size_low(col->mtype, col->len));
|
|
|
|
}
|
2009-09-07 10:22:53 +00:00
|
|
|
#endif /* !UNIV_HOTBACKUP */
|
|
|
|
/***********************************************************************//**
|
|
|
|
Returns the size of a fixed size column, 0 if not a fixed size column.
|
|
|
|
@return fixed size, or 0 */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_col_get_fixed_size(
|
|
|
|
/*====================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_col_t* col, /*!< in: column */
|
2013-12-16 15:38:05 +01:00
|
|
|
ulint comp) /*!< in: nonzero=ROW_FORMAT=COMPACT */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
return(dtype_get_fixed_size_low(col->mtype, col->prtype, col->len,
|
2011-07-14 21:22:41 +02:00
|
|
|
col->mbminmaxlen, comp));
|
2008-11-30 22:10:29 -08:00
|
|
|
}
|
2009-09-07 10:22:53 +00:00
|
|
|
/***********************************************************************//**
|
2008-11-30 22:10:29 -08:00
|
|
|
Returns the ROW_FORMAT=REDUNDANT stored SQL NULL size of a column.
|
2009-09-07 10:22:53 +00:00
|
|
|
For fixed length types it is the fixed length of the type, otherwise 0.
|
|
|
|
@return SQL null storage size in ROW_FORMAT=REDUNDANT */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_col_get_sql_null_size(
|
|
|
|
/*=======================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_col_t* col, /*!< in: column */
|
|
|
|
ulint comp) /*!< in: nonzero=ROW_FORMAT=COMPACT */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
2009-09-07 10:22:53 +00:00
|
|
|
return(dict_col_get_fixed_size(col, comp));
|
2008-11-30 22:10:29 -08:00
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/*********************************************************************//**
|
|
|
|
Gets the column number.
|
|
|
|
@return col->ind, table column position (starting from 0) */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_col_get_no(
|
|
|
|
/*============*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_col_t* col) /*!< in: column */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(col);
|
|
|
|
|
|
|
|
return(col->ind);
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/*********************************************************************//**
|
2008-11-30 22:10:29 -08:00
|
|
|
Gets the column position in the clustered index. */
|
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_col_get_clust_pos(
|
|
|
|
/*===================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_col_t* col, /*!< in: table column */
|
|
|
|
const dict_index_t* clust_index) /*!< in: clustered index */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ulint i;
|
|
|
|
|
|
|
|
ut_ad(col);
|
|
|
|
ut_ad(clust_index);
|
|
|
|
ut_ad(dict_index_is_clust(clust_index));
|
|
|
|
|
|
|
|
for (i = 0; i < clust_index->n_def; i++) {
|
|
|
|
const dict_field_t* field = &clust_index->fields[i];
|
|
|
|
|
|
|
|
if (!field->prefix_len && field->col == col) {
|
|
|
|
return(i);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return(ULINT_UNDEFINED);
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
#ifndef UNIV_HOTBACKUP
|
2008-11-30 22:10:29 -08:00
|
|
|
#ifdef UNIV_DEBUG
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Gets the first index on the table (the clustered index).
|
|
|
|
@return index, NULL if none exists */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
dict_index_t*
|
|
|
|
dict_table_get_first_index(
|
|
|
|
/*=======================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_table_t* table) /*!< in: table */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(table);
|
|
|
|
ut_ad(table->magic_n == DICT_TABLE_MAGIC_N);
|
|
|
|
|
|
|
|
return(UT_LIST_GET_FIRST(((dict_table_t*) table)->indexes));
|
|
|
|
}
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
/********************************************************************//**
|
|
|
|
Gets the last index on the table.
|
|
|
|
@return index, NULL if none exists */
|
|
|
|
UNIV_INLINE
|
|
|
|
dict_index_t*
|
|
|
|
dict_table_get_last_index(
|
|
|
|
/*=======================*/
|
|
|
|
const dict_table_t* table) /*!< in: table */
|
|
|
|
{
|
|
|
|
ut_ad(table);
|
|
|
|
ut_ad(table->magic_n == DICT_TABLE_MAGIC_N);
|
|
|
|
|
|
|
|
return(UT_LIST_GET_LAST((const_cast<dict_table_t*>(table))
|
|
|
|
->indexes));
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Gets the next index on the table.
|
|
|
|
@return index, NULL if none left */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
dict_index_t*
|
|
|
|
dict_table_get_next_index(
|
|
|
|
/*======================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_index_t* index) /*!< in: index */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(index);
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
|
|
|
|
|
|
|
return(UT_LIST_GET_NEXT(indexes, (dict_index_t*) index));
|
|
|
|
}
|
|
|
|
#endif /* UNIV_DEBUG */
|
2009-09-07 10:22:53 +00:00
|
|
|
#endif /* !UNIV_HOTBACKUP */
|
2008-11-30 22:10:29 -08:00
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Check whether the index is the clustered index.
|
|
|
|
@return nonzero for clustered index, zero for other indexes */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_is_clust(
|
|
|
|
/*================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_index_t* index) /*!< in: index */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
return(index->type & DICT_CLUSTERED);
|
2008-11-30 22:10:29 -08:00
|
|
|
}
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Check whether the index is unique.
|
|
|
|
@return nonzero for unique index, zero for other indexes */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_is_unique(
|
|
|
|
/*=================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_index_t* index) /*!< in: index */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
return(index->type & DICT_UNIQUE);
|
2008-11-30 22:10:29 -08:00
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Check whether the index is the insert buffer tree.
|
|
|
|
@return nonzero for insert buffer, zero for other indexes */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_is_ibuf(
|
|
|
|
/*===============*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_index_t* index) /*!< in: index */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
return(index->type & DICT_IBUF);
|
|
|
|
}
|
|
|
|
|
|
|
|
/********************************************************************//**
|
|
|
|
Check whether the index is an universal index tree.
|
|
|
|
@return nonzero for universal tree, zero for other indexes */
|
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_is_univ(
|
|
|
|
/*===============*/
|
|
|
|
const dict_index_t* index) /*!< in: index */
|
|
|
|
{
|
|
|
|
ut_ad(index);
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
|
|
|
|
|
|
|
return(index->type & DICT_UNIVERSAL);
|
2008-11-30 22:10:29 -08:00
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Check whether the index is a secondary index or the insert buffer tree.
|
|
|
|
@return nonzero for insert buffer, zero for other indexes */
|
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_is_sec_or_ibuf(
|
|
|
|
/*======================*/
|
|
|
|
const dict_index_t* index) /*!< in: index */
|
|
|
|
{
|
|
|
|
ulint type;
|
|
|
|
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
|
|
|
|
|
|
|
type = index->type;
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
return(!(type & DICT_CLUSTERED) || (type & DICT_IBUF));
|
2009-09-07 10:22:53 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/********************************************************************//**
|
2008-11-30 22:10:29 -08:00
|
|
|
Gets the number of user-defined columns in a table in the dictionary
|
2009-09-07 10:22:53 +00:00
|
|
|
cache.
|
|
|
|
@return number of user-defined (e.g., not ROW_ID) columns of a table */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_table_get_n_user_cols(
|
|
|
|
/*=======================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_table_t* table) /*!< in: table */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(table->magic_n == DICT_TABLE_MAGIC_N);
|
|
|
|
|
|
|
|
return(table->n_cols - DATA_N_SYS_COLS);
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Gets the number of system columns in a table in the dictionary cache.
|
|
|
|
@return number of system (e.g., ROW_ID) columns of a table */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_table_get_n_sys_cols(
|
|
|
|
/*======================*/
|
2016-08-10 19:24:58 +02:00
|
|
|
const dict_table_t* table MY_ATTRIBUTE((unused))) /*!< in: table */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(table);
|
|
|
|
ut_ad(table->magic_n == DICT_TABLE_MAGIC_N);
|
|
|
|
ut_ad(table->cached);
|
|
|
|
|
|
|
|
return(DATA_N_SYS_COLS);
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
2008-11-30 22:10:29 -08:00
|
|
|
Gets the number of all columns (also system) in a table in the dictionary
|
2009-09-07 10:22:53 +00:00
|
|
|
cache.
|
|
|
|
@return number of columns of a table */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_table_get_n_cols(
|
|
|
|
/*==================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_table_t* table) /*!< in: table */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(table->magic_n == DICT_TABLE_MAGIC_N);
|
|
|
|
|
|
|
|
return(table->n_cols);
|
|
|
|
}
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
/********************************************************************//**
|
|
|
|
Gets the approximately estimated number of rows in the table.
|
|
|
|
@return estimated number of rows */
|
|
|
|
UNIV_INLINE
|
|
|
|
ib_uint64_t
|
|
|
|
dict_table_get_n_rows(
|
|
|
|
/*==================*/
|
|
|
|
const dict_table_t* table) /*!< in: table */
|
|
|
|
{
|
|
|
|
ut_ad(table->stat_initialized);
|
|
|
|
|
|
|
|
return(table->stat_n_rows);
|
|
|
|
}
|
|
|
|
|
|
|
|
/********************************************************************//**
|
|
|
|
Increment the number of rows in the table by one.
|
|
|
|
Notice that this operation is not protected by any latch, the number is
|
|
|
|
approximate. */
|
|
|
|
UNIV_INLINE
|
|
|
|
void
|
|
|
|
dict_table_n_rows_inc(
|
|
|
|
/*==================*/
|
|
|
|
dict_table_t* table) /*!< in/out: table */
|
|
|
|
{
|
|
|
|
if (table->stat_initialized) {
|
|
|
|
ib_uint64_t n_rows = table->stat_n_rows;
|
|
|
|
if (n_rows < 0xFFFFFFFFFFFFFFFFULL) {
|
|
|
|
table->stat_n_rows = n_rows + 1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/********************************************************************//**
|
|
|
|
Decrement the number of rows in the table by one.
|
|
|
|
Notice that this operation is not protected by any latch, the number is
|
|
|
|
approximate. */
|
|
|
|
UNIV_INLINE
|
|
|
|
void
|
|
|
|
dict_table_n_rows_dec(
|
|
|
|
/*==================*/
|
|
|
|
dict_table_t* table) /*!< in/out: table */
|
|
|
|
{
|
|
|
|
if (table->stat_initialized) {
|
|
|
|
ib_uint64_t n_rows = table->stat_n_rows;
|
|
|
|
if (n_rows > 0) {
|
|
|
|
table->stat_n_rows = n_rows - 1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2008-11-30 22:10:29 -08:00
|
|
|
#ifdef UNIV_DEBUG
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Gets the nth column of a table.
|
|
|
|
@return pointer to column object */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
dict_col_t*
|
|
|
|
dict_table_get_nth_col(
|
|
|
|
/*===================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_table_t* table, /*!< in: table */
|
|
|
|
ulint pos) /*!< in: position of column */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(table);
|
|
|
|
ut_ad(pos < table->n_def);
|
|
|
|
ut_ad(table->magic_n == DICT_TABLE_MAGIC_N);
|
|
|
|
|
|
|
|
return((dict_col_t*) (table->cols) + pos);
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Gets the given system column of a table.
|
|
|
|
@return pointer to column object */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
dict_col_t*
|
|
|
|
dict_table_get_sys_col(
|
|
|
|
/*===================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_table_t* table, /*!< in: table */
|
|
|
|
ulint sys) /*!< in: DATA_ROW_ID, ... */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
dict_col_t* col;
|
|
|
|
|
|
|
|
ut_ad(table);
|
|
|
|
ut_ad(sys < DATA_N_SYS_COLS);
|
|
|
|
ut_ad(table->magic_n == DICT_TABLE_MAGIC_N);
|
|
|
|
|
|
|
|
col = dict_table_get_nth_col(table, table->n_cols
|
|
|
|
- DATA_N_SYS_COLS + sys);
|
|
|
|
ut_ad(col->mtype == DATA_SYS);
|
|
|
|
ut_ad(col->prtype == (sys | DATA_NOT_NULL));
|
|
|
|
|
|
|
|
return(col);
|
|
|
|
}
|
|
|
|
#endif /* UNIV_DEBUG */
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Gets the given system column number of a table.
|
|
|
|
@return column number */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_table_get_sys_col_no(
|
|
|
|
/*======================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_table_t* table, /*!< in: table */
|
|
|
|
ulint sys) /*!< in: DATA_ROW_ID, ... */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(table);
|
|
|
|
ut_ad(sys < DATA_N_SYS_COLS);
|
|
|
|
ut_ad(table->magic_n == DICT_TABLE_MAGIC_N);
|
|
|
|
|
|
|
|
return(table->n_cols - DATA_N_SYS_COLS + sys);
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Check whether the table uses the compact page format.
|
|
|
|
@return TRUE if table uses the compact page format */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ibool
|
|
|
|
dict_table_is_comp(
|
|
|
|
/*===============*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_table_t* table) /*!< in: table */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(table);
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
#if DICT_TF_COMPACT != 1
|
|
|
|
#error "DICT_TF_COMPACT must be 1"
|
2008-11-30 22:10:29 -08:00
|
|
|
#endif
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
return(table->flags & DICT_TF_COMPACT);
|
|
|
|
}
|
|
|
|
|
|
|
|
/************************************************************************
|
|
|
|
Check if the table has an FTS index. */
|
|
|
|
UNIV_INLINE
|
|
|
|
ibool
|
|
|
|
dict_table_has_fts_index(
|
|
|
|
/*=====================*/
|
|
|
|
/* out: TRUE if table has an FTS index */
|
|
|
|
dict_table_t* table) /* in: table */
|
|
|
|
{
|
|
|
|
ut_ad(table);
|
|
|
|
|
|
|
|
return(DICT_TF2_FLAG_IS_SET(table, DICT_TF2_FTS));
|
|
|
|
}
|
|
|
|
|
|
|
|
/********************************************************************//**
|
|
|
|
Validate the table flags.
|
|
|
|
@return true if valid. */
|
|
|
|
UNIV_INLINE
|
|
|
|
bool
|
|
|
|
dict_tf_is_valid(
|
|
|
|
/*=============*/
|
|
|
|
ulint flags) /*!< in: table flags */
|
|
|
|
{
|
|
|
|
ulint compact = DICT_TF_GET_COMPACT(flags);
|
|
|
|
ulint zip_ssize = DICT_TF_GET_ZIP_SSIZE(flags);
|
|
|
|
ulint atomic_blobs = DICT_TF_HAS_ATOMIC_BLOBS(flags);
|
|
|
|
ulint unused = DICT_TF_GET_UNUSED(flags);
|
2014-02-03 10:08:15 +02:00
|
|
|
ulint page_compression = DICT_TF_GET_PAGE_COMPRESSION(flags);
|
|
|
|
ulint page_compression_level = DICT_TF_GET_PAGE_COMPRESSION_LEVEL(flags);
|
|
|
|
ulint data_dir = DICT_TF_HAS_DATA_DIR(flags);
|
|
|
|
ulint atomic_writes = DICT_TF_GET_ATOMIC_WRITES(flags);
|
2013-12-16 15:38:05 +01:00
|
|
|
|
|
|
|
/* Make sure there are no bits that we do not know about. */
|
|
|
|
if (unused != 0) {
|
2014-02-03 10:08:15 +02:00
|
|
|
fprintf(stderr,
|
2017-04-21 16:06:05 +03:00
|
|
|
"InnoDB: Error: table unused flags are " ULINTPF
|
2014-02-03 10:08:15 +02:00
|
|
|
" in the data dictionary and are corrupted\n"
|
|
|
|
"InnoDB: Error: data dictionary flags are\n"
|
2017-04-21 16:06:05 +03:00
|
|
|
"InnoDB: compact " ULINTPF " atomic_blobs " ULINTPF
|
|
|
|
"\nInnoDB: unused " ULINTPF " data_dir " ULINTPF
|
|
|
|
" zip_ssize " ULINTPF
|
|
|
|
"\nInnoDB: page_compression " ULINTPF
|
|
|
|
" page_compression_level " ULINTPF
|
|
|
|
"\nInnoDB: atomic_writes " ULINTPF "\n",
|
2014-02-03 10:08:15 +02:00
|
|
|
unused,
|
|
|
|
compact, atomic_blobs, unused, data_dir, zip_ssize,
|
2015-04-01 11:50:21 +03:00
|
|
|
page_compression, page_compression_level, atomic_writes
|
2014-02-03 10:08:15 +02:00
|
|
|
);
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
return(false);
|
|
|
|
|
|
|
|
} else if (atomic_blobs) {
|
|
|
|
/* Barracuda row formats COMPRESSED and DYNAMIC build on
|
|
|
|
the page structure introduced for the COMPACT row format
|
|
|
|
by allowing keys in secondary indexes to be made from
|
|
|
|
data stored off-page in the clustered index. */
|
|
|
|
|
|
|
|
if (!compact) {
|
2014-02-03 10:08:15 +02:00
|
|
|
fprintf(stderr,
|
2017-04-21 16:06:05 +03:00
|
|
|
"InnoDB: Error: table compact flags are "
|
|
|
|
ULINTPF
|
2014-02-03 10:08:15 +02:00
|
|
|
" in the data dictionary and are corrupted\n"
|
|
|
|
"InnoDB: Error: data dictionary flags are\n"
|
2017-04-21 16:06:05 +03:00
|
|
|
"InnoDB: compact " ULINTPF
|
|
|
|
" atomic_blobs " ULINTPF "\n"
|
|
|
|
"InnoDB: unused " ULINTPF
|
|
|
|
" data_dir " ULINTPF " zip_ssize " ULINTPF
|
|
|
|
"\nInnoDB: page_compression " ULINTPF
|
|
|
|
" page_compression_level " ULINTPF
|
|
|
|
"\nInnoDB: atomic_writes " ULINTPF "\n",
|
2014-02-03 10:08:15 +02:00
|
|
|
compact, compact, atomic_blobs, unused, data_dir, zip_ssize,
|
|
|
|
page_compression, page_compression_level, atomic_writes
|
|
|
|
);
|
2013-12-16 15:38:05 +01:00
|
|
|
return(false);
|
|
|
|
}
|
|
|
|
|
|
|
|
} else if (zip_ssize) {
|
|
|
|
|
|
|
|
/* Antelope does not support COMPRESSED row format. */
|
2014-02-03 10:08:15 +02:00
|
|
|
fprintf(stderr,
|
2017-04-21 16:06:05 +03:00
|
|
|
"InnoDB: Error: table flags are " ULINTPF
|
2014-02-03 10:08:15 +02:00
|
|
|
" in the data dictionary and are corrupted\n"
|
|
|
|
"InnoDB: Error: data dictionary flags are\n"
|
2017-04-21 16:06:05 +03:00
|
|
|
"InnoDB: compact " ULINTPF " atomic_blobs " ULINTPF
|
|
|
|
"\nInnoDB: unused " ULINTPF " data_dir " ULINTPF
|
|
|
|
" zip_ssize " ULINTPF
|
|
|
|
"\nInnoDB: page_compression " ULINTPF
|
|
|
|
" page_compression_level " ULINTPF
|
|
|
|
"\nInnoDB: atomic_writes " ULINTPF "\n",
|
2014-02-03 10:08:15 +02:00
|
|
|
flags, compact, atomic_blobs, unused, data_dir, zip_ssize,
|
|
|
|
page_compression, page_compression_level, atomic_writes
|
|
|
|
);
|
2013-12-16 15:38:05 +01:00
|
|
|
return(false);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (zip_ssize) {
|
|
|
|
|
|
|
|
/* COMPRESSED row format must have compact and atomic_blobs
|
|
|
|
bits set and validate the number is within allowed range. */
|
|
|
|
|
|
|
|
if (!compact
|
|
|
|
|| !atomic_blobs
|
|
|
|
|| zip_ssize > PAGE_ZIP_SSIZE_MAX) {
|
|
|
|
|
2014-02-03 10:08:15 +02:00
|
|
|
fprintf(stderr,
|
2017-04-21 16:06:05 +03:00
|
|
|
"InnoDB: Error: table compact flags are "
|
|
|
|
ULINTPF
|
|
|
|
" in the data dictionary and are corrupted\n"
|
2014-02-03 10:08:15 +02:00
|
|
|
"InnoDB: Error: data dictionary flags are\n"
|
2017-04-21 16:06:05 +03:00
|
|
|
"InnoDB: compact " ULINTPF
|
|
|
|
" atomic_blobs " ULINTPF "\n"
|
|
|
|
"InnoDB: unused " ULINTPF
|
|
|
|
" data_dir " ULINTPF " zip_ssize " ULINTPF
|
|
|
|
"\nInnoDB: page_compression " ULINTPF
|
|
|
|
" page_compression_level " ULINTPF
|
|
|
|
"\nInnoDB: atomic_writes " ULINTPF "\n",
|
2014-02-03 10:08:15 +02:00
|
|
|
flags,
|
|
|
|
compact, atomic_blobs, unused, data_dir, zip_ssize,
|
|
|
|
page_compression, page_compression_level, atomic_writes
|
|
|
|
|
|
|
|
);
|
|
|
|
return(false);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (page_compression || page_compression_level) {
|
|
|
|
/* Page compression format must have compact and
|
|
|
|
atomic_blobs and page_compression_level requires
|
|
|
|
page_compression */
|
|
|
|
if (!compact
|
|
|
|
|| !page_compression
|
|
|
|
|| !atomic_blobs) {
|
|
|
|
|
|
|
|
fprintf(stderr,
|
2017-04-21 16:06:05 +03:00
|
|
|
"InnoDB: Error: table flags are " ULINTPF
|
|
|
|
" in the data dictionary and are corrupted\n"
|
2014-02-03 10:08:15 +02:00
|
|
|
"InnoDB: Error: data dictionary flags are\n"
|
2017-04-21 16:06:05 +03:00
|
|
|
"InnoDB: compact " ULINTPF
|
|
|
|
" atomic_blobs " ULINTPF "\n"
|
|
|
|
"InnoDB: unused " ULINTPF
|
|
|
|
" data_dir " ULINTPF " zip_ssize " ULINTPF
|
|
|
|
"\nInnoDB: page_compression " ULINTPF
|
|
|
|
" page_compression_level " ULINTPF
|
|
|
|
"\nInnoDB: atomic_writes " ULINTPF "\n",
|
2014-02-03 10:08:15 +02:00
|
|
|
flags, compact, atomic_blobs, unused, data_dir, zip_ssize,
|
|
|
|
page_compression, page_compression_level, atomic_writes
|
|
|
|
);
|
2013-12-16 15:38:05 +01:00
|
|
|
return(false);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2014-02-12 18:00:03 +02:00
|
|
|
if (atomic_writes) {
|
|
|
|
|
2014-08-03 15:12:10 +03:00
|
|
|
if(atomic_writes > ATOMIC_WRITES_OFF) {
|
2014-02-12 18:00:03 +02:00
|
|
|
|
|
|
|
fprintf(stderr,
|
2017-04-21 16:06:05 +03:00
|
|
|
"InnoDB: Error: table flags are " ULINTPF
|
|
|
|
" in the data dictionary and are corrupted\n"
|
2014-02-12 18:00:03 +02:00
|
|
|
"InnoDB: Error: data dictionary flags are\n"
|
2017-04-21 16:06:05 +03:00
|
|
|
"InnoDB: compact " ULINTPF
|
|
|
|
" atomic_blobs " ULINTPF "\n"
|
|
|
|
"InnoDB: unused " ULINTPF
|
|
|
|
" data_dir " ULINTPF " zip_ssize " ULINTPF
|
|
|
|
"\nInnoDB: page_compression " ULINTPF
|
|
|
|
" page_compression_level " ULINTPF
|
|
|
|
"\nInnoDB: atomic_writes " ULINTPF "\n",
|
2014-02-12 18:00:03 +02:00
|
|
|
flags, compact, atomic_blobs, unused, data_dir, zip_ssize,
|
|
|
|
page_compression, page_compression_level, atomic_writes
|
|
|
|
);
|
2013-12-16 15:38:05 +01:00
|
|
|
return(false);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* CREATE TABLE ... DATA DIRECTORY is supported for any row format,
|
|
|
|
so the DATA_DIR flag is compatible with all other table flags. */
|
|
|
|
|
|
|
|
return(true);
|
|
|
|
}
|
|
|
|
|
|
|
|
/********************************************************************//**
|
|
|
|
Validate a SYS_TABLES TYPE field and return it.
|
|
|
|
@return Same as input after validating it as a SYS_TABLES TYPE field.
|
|
|
|
If there is an error, return ULINT_UNDEFINED. */
|
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_sys_tables_type_validate(
|
|
|
|
/*==========================*/
|
|
|
|
ulint type, /*!< in: SYS_TABLES.TYPE */
|
|
|
|
ulint n_cols) /*!< in: SYS_TABLES.N_COLS */
|
|
|
|
{
|
|
|
|
ulint low_order_bit = DICT_TF_GET_COMPACT(type);
|
|
|
|
ulint redundant = !(n_cols & DICT_N_COLS_COMPACT);
|
|
|
|
ulint zip_ssize = DICT_TF_GET_ZIP_SSIZE(type);
|
|
|
|
ulint atomic_blobs = DICT_TF_HAS_ATOMIC_BLOBS(type);
|
|
|
|
ulint unused = DICT_TF_GET_UNUSED(type);
|
2014-02-03 10:08:15 +02:00
|
|
|
ulint page_compression = DICT_TF_GET_PAGE_COMPRESSION(type);
|
|
|
|
ulint page_compression_level = DICT_TF_GET_PAGE_COMPRESSION_LEVEL(type);
|
|
|
|
ulint atomic_writes = DICT_TF_GET_ATOMIC_WRITES(type);
|
2014-02-12 18:00:03 +02:00
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
/* The low order bit of SYS_TABLES.TYPE is always set to 1.
|
|
|
|
If the format is UNIV_FORMAT_B or higher, this field is the same
|
|
|
|
as dict_table_t::flags. Zero is not allowed here. */
|
|
|
|
if (!low_order_bit) {
|
|
|
|
return(ULINT_UNDEFINED);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (redundant) {
|
|
|
|
if (zip_ssize || atomic_blobs) {
|
|
|
|
return(ULINT_UNDEFINED);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Make sure there are no bits that we do not know about. */
|
|
|
|
if (unused) {
|
|
|
|
return(ULINT_UNDEFINED);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (atomic_blobs) {
|
|
|
|
/* Barracuda row formats COMPRESSED and DYNAMIC build on
|
|
|
|
the page structure introduced for the COMPACT row format
|
|
|
|
by allowing keys in secondary indexes to be made from
|
|
|
|
data stored off-page in the clustered index.
|
|
|
|
|
|
|
|
The DICT_N_COLS_COMPACT flag should be in N_COLS,
|
|
|
|
but we already know that. */
|
|
|
|
} else if (zip_ssize) {
|
|
|
|
/* Antelope does not support COMPRESSED format. */
|
|
|
|
return(ULINT_UNDEFINED);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (zip_ssize) {
|
|
|
|
/* COMPRESSED row format must have low_order_bit and
|
|
|
|
atomic_blobs bits set and the DICT_N_COLS_COMPACT flag
|
|
|
|
should be in N_COLS, but we already know about the
|
|
|
|
low_order_bit and DICT_N_COLS_COMPACT flags. */
|
|
|
|
if (!atomic_blobs) {
|
|
|
|
return(ULINT_UNDEFINED);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Validate that the number is within allowed range. */
|
|
|
|
if (zip_ssize > PAGE_ZIP_SSIZE_MAX) {
|
|
|
|
return(ULINT_UNDEFINED);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* There is nothing to validate for the data_dir field.
|
|
|
|
CREATE TABLE ... DATA DIRECTORY is supported for any row
|
|
|
|
format, so the DATA_DIR flag is compatible with any other
|
|
|
|
table flags. However, it is not used with TEMPORARY tables.*/
|
|
|
|
|
2014-02-03 10:08:15 +02:00
|
|
|
if (page_compression || page_compression_level) {
|
|
|
|
/* page compressed row format must have low_order_bit and
|
|
|
|
atomic_blobs bits set and the DICT_N_COLS_COMPACT flag
|
|
|
|
should be in N_COLS, but we already know about the
|
|
|
|
low_order_bit and DICT_N_COLS_COMPACT flags. */
|
|
|
|
|
|
|
|
if (!atomic_blobs || !page_compression) {
|
|
|
|
return(ULINT_UNDEFINED);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2014-02-13 09:13:56 +02:00
|
|
|
/* Validate that the atomic writes number is within allowed range. */
|
2014-08-03 15:12:10 +03:00
|
|
|
if (atomic_writes > ATOMIC_WRITES_OFF) {
|
2017-06-13 18:29:40 +03:00
|
|
|
return(ULINT_UNDEFINED);
|
2014-02-13 09:13:56 +02:00
|
|
|
}
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
/* Return the validated SYS_TABLES.TYPE. */
|
|
|
|
return(type);
|
|
|
|
}
|
|
|
|
|
|
|
|
/********************************************************************//**
|
|
|
|
Determine the file format from dict_table_t::flags
|
|
|
|
The low order bit will be zero for REDUNDANT and 1 for COMPACT. For any
|
|
|
|
other row_format, file_format is > 0 and DICT_TF_COMPACT will also be set.
|
|
|
|
@return file format version */
|
|
|
|
UNIV_INLINE
|
|
|
|
rec_format_t
|
|
|
|
dict_tf_get_rec_format(
|
|
|
|
/*===================*/
|
|
|
|
ulint flags) /*!< in: dict_table_t::flags */
|
|
|
|
{
|
|
|
|
ut_a(dict_tf_is_valid(flags));
|
|
|
|
|
|
|
|
if (!DICT_TF_GET_COMPACT(flags)) {
|
|
|
|
return(REC_FORMAT_REDUNDANT);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!DICT_TF_HAS_ATOMIC_BLOBS(flags)) {
|
|
|
|
return(REC_FORMAT_COMPACT);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (DICT_TF_GET_ZIP_SSIZE(flags)) {
|
|
|
|
return(REC_FORMAT_COMPRESSED);
|
|
|
|
}
|
|
|
|
|
|
|
|
return(REC_FORMAT_DYNAMIC);
|
|
|
|
}
|
|
|
|
|
|
|
|
/********************************************************************//**
|
|
|
|
Determine the file format from a dict_table_t::flags.
|
|
|
|
@return file format version */
|
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_tf_get_format(
|
|
|
|
/*===============*/
|
|
|
|
ulint flags) /*!< in: dict_table_t::flags */
|
|
|
|
{
|
|
|
|
if (DICT_TF_HAS_ATOMIC_BLOBS(flags)) {
|
|
|
|
return(UNIV_FORMAT_B);
|
|
|
|
}
|
|
|
|
|
|
|
|
return(UNIV_FORMAT_A);
|
2008-11-30 22:10:29 -08:00
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Determine the file format of a table.
|
|
|
|
@return file format version */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_table_get_format(
|
|
|
|
/*==================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_table_t* table) /*!< in: table */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(table);
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
return(dict_tf_get_format(table->flags));
|
2008-11-30 22:10:29 -08:00
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
2013-12-16 15:38:05 +01:00
|
|
|
Set the file format and zip size in a dict_table_t::flags. If zip size
|
|
|
|
is not needed, it should be 0. */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
void
|
2013-12-16 15:38:05 +01:00
|
|
|
dict_tf_set(
|
|
|
|
/*========*/
|
|
|
|
ulint* flags, /*!< in/out: table flags */
|
|
|
|
rec_format_t format, /*!< in: file format */
|
|
|
|
ulint zip_ssize, /*!< in: zip shift size */
|
2014-02-03 10:08:15 +02:00
|
|
|
bool use_data_dir, /*!< in: table uses DATA DIRECTORY
|
|
|
|
*/
|
|
|
|
bool page_compressed,/*!< in: table uses page compressed
|
|
|
|
pages */
|
|
|
|
ulint page_compression_level, /*!< in: table page compression
|
|
|
|
level */
|
2015-04-01 11:50:21 +03:00
|
|
|
ulint atomic_writes) /*!< in: table atomic writes setup */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
2014-02-03 10:08:15 +02:00
|
|
|
atomic_writes_t awrites = (atomic_writes_t)atomic_writes;
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
switch (format) {
|
|
|
|
case REC_FORMAT_REDUNDANT:
|
|
|
|
*flags = 0;
|
|
|
|
ut_ad(zip_ssize == 0);
|
|
|
|
break;
|
|
|
|
case REC_FORMAT_COMPACT:
|
|
|
|
*flags = DICT_TF_COMPACT;
|
|
|
|
ut_ad(zip_ssize == 0);
|
|
|
|
break;
|
|
|
|
case REC_FORMAT_COMPRESSED:
|
|
|
|
*flags = DICT_TF_COMPACT
|
|
|
|
| (1 << DICT_TF_POS_ATOMIC_BLOBS)
|
|
|
|
| (zip_ssize << DICT_TF_POS_ZIP_SSIZE);
|
|
|
|
break;
|
|
|
|
case REC_FORMAT_DYNAMIC:
|
|
|
|
*flags = DICT_TF_COMPACT
|
|
|
|
| (1 << DICT_TF_POS_ATOMIC_BLOBS);
|
|
|
|
ut_ad(zip_ssize == 0);
|
|
|
|
break;
|
|
|
|
}
|
2008-11-30 22:10:29 -08:00
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
if (use_data_dir) {
|
|
|
|
*flags |= (1 << DICT_TF_POS_DATA_DIR);
|
|
|
|
}
|
2014-02-03 10:08:15 +02:00
|
|
|
|
|
|
|
if (page_compressed) {
|
2014-02-12 12:52:34 +02:00
|
|
|
*flags |= (1 << DICT_TF_POS_ATOMIC_BLOBS)
|
|
|
|
| (1 << DICT_TF_POS_PAGE_COMPRESSION)
|
|
|
|
| (page_compression_level << DICT_TF_POS_PAGE_COMPRESSION_LEVEL);
|
2014-02-03 10:08:15 +02:00
|
|
|
|
|
|
|
ut_ad(zip_ssize == 0);
|
|
|
|
ut_ad(dict_tf_get_page_compression(*flags) == TRUE);
|
|
|
|
ut_ad(dict_tf_get_page_compression_level(*flags) == page_compression_level);
|
|
|
|
}
|
|
|
|
|
2014-02-12 18:00:03 +02:00
|
|
|
*flags |= (atomic_writes << DICT_TF_POS_ATOMIC_WRITES);
|
2014-02-13 12:35:37 +02:00
|
|
|
ut_a(dict_tf_get_atomic_writes(*flags) == awrites);
|
2008-11-30 22:10:29 -08:00
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
2013-12-16 15:38:05 +01:00
|
|
|
Convert a 32 bit integer table flags to the 32 bit integer that is
|
|
|
|
written into the tablespace header at the offset FSP_SPACE_FLAGS and is
|
|
|
|
also stored in the fil_space_t::flags field. The following chart shows
|
|
|
|
the translation of the low order bit. Other bits are the same.
|
|
|
|
========================= Low order bit ==========================
|
|
|
|
| REDUNDANT | COMPACT | COMPRESSED | DYNAMIC
|
|
|
|
dict_table_t::flags | 0 | 1 | 1 | 1
|
|
|
|
fil_space_t::flags | 0 | 0 | 1 | 1
|
|
|
|
==================================================================
|
|
|
|
@return tablespace flags (fil_space_t::flags) */
|
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_tf_to_fsp_flags(
|
|
|
|
/*=================*/
|
|
|
|
ulint table_flags) /*!< in: dict_table_t::flags */
|
|
|
|
{
|
|
|
|
ulint fsp_flags;
|
MDEV-11623 MariaDB 10.1 fails to start datadir created with
MariaDB 10.0/MySQL 5.6 using innodb-page-size!=16K
The storage format of FSP_SPACE_FLAGS was accidentally broken
already in MariaDB 10.1.0. This fix is bringing the format in
line with other MySQL and MariaDB release series.
Please refer to the comments that were added to fsp0fsp.h
for details.
This is an INCOMPATIBLE CHANGE that affects users of
page_compression and non-default innodb_page_size. Upgrading
to this release will correct the flags in the data files.
If you want to downgrade to earlier MariaDB 10.1.x, please refer
to the test innodb.101_compatibility how to reset the
FSP_SPACE_FLAGS in the files.
NOTE: MariaDB 10.1.0 to 10.1.20 can misinterpret
uncompressed data files with innodb_page_size=4k or 64k as
compressed innodb_page_size=16k files, and then probably fail
when trying to access the pages. See the comments in the
function fsp_flags_convert_from_101() for detailed analysis.
Move PAGE_COMPRESSION to FSP_SPACE_FLAGS bit position 16.
In this way, compressed innodb_page_size=16k tablespaces will not
be mistaken for uncompressed ones by MariaDB 10.1.0 to 10.1.20.
Derive PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR from the
dict_table_t::flags when the table is available, in
fil_space_for_table_exists_in_mem() or fil_open_single_table_tablespace().
During crash recovery, fil_load_single_table_tablespace() will use
innodb_compression_level for the PAGE_COMPRESSION_LEVEL.
FSP_FLAGS_MEM_MASK: A bitmap of the memory-only fil_space_t::flags
that are not to be written to FSP_SPACE_FLAGS. Currently, these will
include PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR.
Introduce the macro FSP_FLAGS_PAGE_SSIZE(). We only support
one innodb_page_size for the whole instance.
When creating a dummy tablespace for the redo log, use
fil_space_t::flags=0. The flags are never written to the redo log files.
Remove many FSP_FLAGS_SET_ macros.
dict_tf_verify_flags(): Remove. This is basically only duplicating
the logic of dict_tf_to_fsp_flags(), used in a debug assertion.
fil_space_t::mark: Remove. This flag was not used for anything.
fil_space_for_table_exists_in_mem(): Remove the unnecessary parameter
mark_space, and add a parameter for table flags. Check that
fil_space_t::flags match the table flags, and adjust the (memory-only)
flags based on the table flags.
fil_node_open_file(): Remove some redundant or unreachable conditions,
do not use stderr for output, and avoid unnecessary server aborts.
fil_user_tablespace_restore_page(): Convert the flags, so that the
correct page_size will be used when restoring a page from the
doublewrite buffer.
fil_space_get_page_compressed(), fsp_flags_is_page_compressed(): Remove.
It suffices to have fil_space_is_page_compressed().
FSP_FLAGS_WIDTH_DATA_DIR, FSP_FLAGS_WIDTH_PAGE_COMPRESSION_LEVEL,
FSP_FLAGS_WIDTH_ATOMIC_WRITES: Remove, because these flags do not
exist in the FSP_SPACE_FLAGS but only in memory.
fsp_flags_try_adjust(): New function, to adjust the FSP_SPACE_FLAGS
in page 0. Called by fil_open_single_table_tablespace(),
fil_space_for_table_exists_in_mem(), innobase_start_or_create_for_mysql()
except if --innodb-read-only is active.
fsp_flags_is_valid(ulint): Reimplement from the scratch, with
accurate comments. Do not display any details of detected
inconsistencies, because the output could be confusing when
dealing with MariaDB 10.1.x data files.
fsp_flags_convert_from_101(ulint): Convert flags from buggy
MariaDB 10.1.x format, or return ULINT_UNDEFINED if the flags
cannot be in MariaDB 10.1.x format.
fsp_flags_match(): Check the flags when probing files.
Implemented based on fsp_flags_is_valid()
and fsp_flags_convert_from_101().
dict_check_tablespaces_and_store_max_id(): Do not access the
page after committing the mini-transaction.
IMPORT TABLESPACE fixes:
AbstractCallback::init(): Convert the flags.
FetchIndexRootPages::operator(): Check that the tablespace flags match the
table flags. Do not attempt to convert tablespace flags to table flags,
because the conversion would necessarily be lossy.
PageConverter::update_header(): Write back the correct flags.
This takes care of the flags in IMPORT TABLESPACE.
2017-01-14 00:13:16 +02:00
|
|
|
ulint page_compression_level = DICT_TF_GET_PAGE_COMPRESSION_LEVEL(
|
|
|
|
table_flags);
|
2014-02-03 10:08:15 +02:00
|
|
|
ulint atomic_writes = DICT_TF_GET_ATOMIC_WRITES(table_flags);
|
2013-12-16 15:38:05 +01:00
|
|
|
|
MDEV-11623 MariaDB 10.1 fails to start datadir created with
MariaDB 10.0/MySQL 5.6 using innodb-page-size!=16K
The storage format of FSP_SPACE_FLAGS was accidentally broken
already in MariaDB 10.1.0. This fix is bringing the format in
line with other MySQL and MariaDB release series.
Please refer to the comments that were added to fsp0fsp.h
for details.
This is an INCOMPATIBLE CHANGE that affects users of
page_compression and non-default innodb_page_size. Upgrading
to this release will correct the flags in the data files.
If you want to downgrade to earlier MariaDB 10.1.x, please refer
to the test innodb.101_compatibility how to reset the
FSP_SPACE_FLAGS in the files.
NOTE: MariaDB 10.1.0 to 10.1.20 can misinterpret
uncompressed data files with innodb_page_size=4k or 64k as
compressed innodb_page_size=16k files, and then probably fail
when trying to access the pages. See the comments in the
function fsp_flags_convert_from_101() for detailed analysis.
Move PAGE_COMPRESSION to FSP_SPACE_FLAGS bit position 16.
In this way, compressed innodb_page_size=16k tablespaces will not
be mistaken for uncompressed ones by MariaDB 10.1.0 to 10.1.20.
Derive PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR from the
dict_table_t::flags when the table is available, in
fil_space_for_table_exists_in_mem() or fil_open_single_table_tablespace().
During crash recovery, fil_load_single_table_tablespace() will use
innodb_compression_level for the PAGE_COMPRESSION_LEVEL.
FSP_FLAGS_MEM_MASK: A bitmap of the memory-only fil_space_t::flags
that are not to be written to FSP_SPACE_FLAGS. Currently, these will
include PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR.
Introduce the macro FSP_FLAGS_PAGE_SSIZE(). We only support
one innodb_page_size for the whole instance.
When creating a dummy tablespace for the redo log, use
fil_space_t::flags=0. The flags are never written to the redo log files.
Remove many FSP_FLAGS_SET_ macros.
dict_tf_verify_flags(): Remove. This is basically only duplicating
the logic of dict_tf_to_fsp_flags(), used in a debug assertion.
fil_space_t::mark: Remove. This flag was not used for anything.
fil_space_for_table_exists_in_mem(): Remove the unnecessary parameter
mark_space, and add a parameter for table flags. Check that
fil_space_t::flags match the table flags, and adjust the (memory-only)
flags based on the table flags.
fil_node_open_file(): Remove some redundant or unreachable conditions,
do not use stderr for output, and avoid unnecessary server aborts.
fil_user_tablespace_restore_page(): Convert the flags, so that the
correct page_size will be used when restoring a page from the
doublewrite buffer.
fil_space_get_page_compressed(), fsp_flags_is_page_compressed(): Remove.
It suffices to have fil_space_is_page_compressed().
FSP_FLAGS_WIDTH_DATA_DIR, FSP_FLAGS_WIDTH_PAGE_COMPRESSION_LEVEL,
FSP_FLAGS_WIDTH_ATOMIC_WRITES: Remove, because these flags do not
exist in the FSP_SPACE_FLAGS but only in memory.
fsp_flags_try_adjust(): New function, to adjust the FSP_SPACE_FLAGS
in page 0. Called by fil_open_single_table_tablespace(),
fil_space_for_table_exists_in_mem(), innobase_start_or_create_for_mysql()
except if --innodb-read-only is active.
fsp_flags_is_valid(ulint): Reimplement from the scratch, with
accurate comments. Do not display any details of detected
inconsistencies, because the output could be confusing when
dealing with MariaDB 10.1.x data files.
fsp_flags_convert_from_101(ulint): Convert flags from buggy
MariaDB 10.1.x format, or return ULINT_UNDEFINED if the flags
cannot be in MariaDB 10.1.x format.
fsp_flags_match(): Check the flags when probing files.
Implemented based on fsp_flags_is_valid()
and fsp_flags_convert_from_101().
dict_check_tablespaces_and_store_max_id(): Do not access the
page after committing the mini-transaction.
IMPORT TABLESPACE fixes:
AbstractCallback::init(): Convert the flags.
FetchIndexRootPages::operator(): Check that the tablespace flags match the
table flags. Do not attempt to convert tablespace flags to table flags,
because the conversion would necessarily be lossy.
PageConverter::update_header(): Write back the correct flags.
This takes care of the flags in IMPORT TABLESPACE.
2017-01-14 00:13:16 +02:00
|
|
|
ut_ad((DICT_TF_GET_PAGE_COMPRESSION(table_flags) == 0)
|
|
|
|
== (page_compression_level == 0));
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
DBUG_EXECUTE_IF("dict_tf_to_fsp_flags_failure",
|
|
|
|
return(ULINT_UNDEFINED););
|
|
|
|
|
|
|
|
/* Adjust bit zero. */
|
|
|
|
fsp_flags = DICT_TF_HAS_ATOMIC_BLOBS(table_flags) ? 1 : 0;
|
|
|
|
|
|
|
|
/* ZIP_SSIZE and ATOMIC_BLOBS are at the same position. */
|
MDEV-11623 MariaDB 10.1 fails to start datadir created with
MariaDB 10.0/MySQL 5.6 using innodb-page-size!=16K
The storage format of FSP_SPACE_FLAGS was accidentally broken
already in MariaDB 10.1.0. This fix is bringing the format in
line with other MySQL and MariaDB release series.
Please refer to the comments that were added to fsp0fsp.h
for details.
This is an INCOMPATIBLE CHANGE that affects users of
page_compression and non-default innodb_page_size. Upgrading
to this release will correct the flags in the data files.
If you want to downgrade to earlier MariaDB 10.1.x, please refer
to the test innodb.101_compatibility how to reset the
FSP_SPACE_FLAGS in the files.
NOTE: MariaDB 10.1.0 to 10.1.20 can misinterpret
uncompressed data files with innodb_page_size=4k or 64k as
compressed innodb_page_size=16k files, and then probably fail
when trying to access the pages. See the comments in the
function fsp_flags_convert_from_101() for detailed analysis.
Move PAGE_COMPRESSION to FSP_SPACE_FLAGS bit position 16.
In this way, compressed innodb_page_size=16k tablespaces will not
be mistaken for uncompressed ones by MariaDB 10.1.0 to 10.1.20.
Derive PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR from the
dict_table_t::flags when the table is available, in
fil_space_for_table_exists_in_mem() or fil_open_single_table_tablespace().
During crash recovery, fil_load_single_table_tablespace() will use
innodb_compression_level for the PAGE_COMPRESSION_LEVEL.
FSP_FLAGS_MEM_MASK: A bitmap of the memory-only fil_space_t::flags
that are not to be written to FSP_SPACE_FLAGS. Currently, these will
include PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR.
Introduce the macro FSP_FLAGS_PAGE_SSIZE(). We only support
one innodb_page_size for the whole instance.
When creating a dummy tablespace for the redo log, use
fil_space_t::flags=0. The flags are never written to the redo log files.
Remove many FSP_FLAGS_SET_ macros.
dict_tf_verify_flags(): Remove. This is basically only duplicating
the logic of dict_tf_to_fsp_flags(), used in a debug assertion.
fil_space_t::mark: Remove. This flag was not used for anything.
fil_space_for_table_exists_in_mem(): Remove the unnecessary parameter
mark_space, and add a parameter for table flags. Check that
fil_space_t::flags match the table flags, and adjust the (memory-only)
flags based on the table flags.
fil_node_open_file(): Remove some redundant or unreachable conditions,
do not use stderr for output, and avoid unnecessary server aborts.
fil_user_tablespace_restore_page(): Convert the flags, so that the
correct page_size will be used when restoring a page from the
doublewrite buffer.
fil_space_get_page_compressed(), fsp_flags_is_page_compressed(): Remove.
It suffices to have fil_space_is_page_compressed().
FSP_FLAGS_WIDTH_DATA_DIR, FSP_FLAGS_WIDTH_PAGE_COMPRESSION_LEVEL,
FSP_FLAGS_WIDTH_ATOMIC_WRITES: Remove, because these flags do not
exist in the FSP_SPACE_FLAGS but only in memory.
fsp_flags_try_adjust(): New function, to adjust the FSP_SPACE_FLAGS
in page 0. Called by fil_open_single_table_tablespace(),
fil_space_for_table_exists_in_mem(), innobase_start_or_create_for_mysql()
except if --innodb-read-only is active.
fsp_flags_is_valid(ulint): Reimplement from the scratch, with
accurate comments. Do not display any details of detected
inconsistencies, because the output could be confusing when
dealing with MariaDB 10.1.x data files.
fsp_flags_convert_from_101(ulint): Convert flags from buggy
MariaDB 10.1.x format, or return ULINT_UNDEFINED if the flags
cannot be in MariaDB 10.1.x format.
fsp_flags_match(): Check the flags when probing files.
Implemented based on fsp_flags_is_valid()
and fsp_flags_convert_from_101().
dict_check_tablespaces_and_store_max_id(): Do not access the
page after committing the mini-transaction.
IMPORT TABLESPACE fixes:
AbstractCallback::init(): Convert the flags.
FetchIndexRootPages::operator(): Check that the tablespace flags match the
table flags. Do not attempt to convert tablespace flags to table flags,
because the conversion would necessarily be lossy.
PageConverter::update_header(): Write back the correct flags.
This takes care of the flags in IMPORT TABLESPACE.
2017-01-14 00:13:16 +02:00
|
|
|
fsp_flags |= table_flags
|
|
|
|
& (DICT_TF_MASK_ZIP_SSIZE | DICT_TF_MASK_ATOMIC_BLOBS);
|
2013-12-16 15:38:05 +01:00
|
|
|
|
MDEV-11623 MariaDB 10.1 fails to start datadir created with
MariaDB 10.0/MySQL 5.6 using innodb-page-size!=16K
The storage format of FSP_SPACE_FLAGS was accidentally broken
already in MariaDB 10.1.0. This fix is bringing the format in
line with other MySQL and MariaDB release series.
Please refer to the comments that were added to fsp0fsp.h
for details.
This is an INCOMPATIBLE CHANGE that affects users of
page_compression and non-default innodb_page_size. Upgrading
to this release will correct the flags in the data files.
If you want to downgrade to earlier MariaDB 10.1.x, please refer
to the test innodb.101_compatibility how to reset the
FSP_SPACE_FLAGS in the files.
NOTE: MariaDB 10.1.0 to 10.1.20 can misinterpret
uncompressed data files with innodb_page_size=4k or 64k as
compressed innodb_page_size=16k files, and then probably fail
when trying to access the pages. See the comments in the
function fsp_flags_convert_from_101() for detailed analysis.
Move PAGE_COMPRESSION to FSP_SPACE_FLAGS bit position 16.
In this way, compressed innodb_page_size=16k tablespaces will not
be mistaken for uncompressed ones by MariaDB 10.1.0 to 10.1.20.
Derive PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR from the
dict_table_t::flags when the table is available, in
fil_space_for_table_exists_in_mem() or fil_open_single_table_tablespace().
During crash recovery, fil_load_single_table_tablespace() will use
innodb_compression_level for the PAGE_COMPRESSION_LEVEL.
FSP_FLAGS_MEM_MASK: A bitmap of the memory-only fil_space_t::flags
that are not to be written to FSP_SPACE_FLAGS. Currently, these will
include PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR.
Introduce the macro FSP_FLAGS_PAGE_SSIZE(). We only support
one innodb_page_size for the whole instance.
When creating a dummy tablespace for the redo log, use
fil_space_t::flags=0. The flags are never written to the redo log files.
Remove many FSP_FLAGS_SET_ macros.
dict_tf_verify_flags(): Remove. This is basically only duplicating
the logic of dict_tf_to_fsp_flags(), used in a debug assertion.
fil_space_t::mark: Remove. This flag was not used for anything.
fil_space_for_table_exists_in_mem(): Remove the unnecessary parameter
mark_space, and add a parameter for table flags. Check that
fil_space_t::flags match the table flags, and adjust the (memory-only)
flags based on the table flags.
fil_node_open_file(): Remove some redundant or unreachable conditions,
do not use stderr for output, and avoid unnecessary server aborts.
fil_user_tablespace_restore_page(): Convert the flags, so that the
correct page_size will be used when restoring a page from the
doublewrite buffer.
fil_space_get_page_compressed(), fsp_flags_is_page_compressed(): Remove.
It suffices to have fil_space_is_page_compressed().
FSP_FLAGS_WIDTH_DATA_DIR, FSP_FLAGS_WIDTH_PAGE_COMPRESSION_LEVEL,
FSP_FLAGS_WIDTH_ATOMIC_WRITES: Remove, because these flags do not
exist in the FSP_SPACE_FLAGS but only in memory.
fsp_flags_try_adjust(): New function, to adjust the FSP_SPACE_FLAGS
in page 0. Called by fil_open_single_table_tablespace(),
fil_space_for_table_exists_in_mem(), innobase_start_or_create_for_mysql()
except if --innodb-read-only is active.
fsp_flags_is_valid(ulint): Reimplement from the scratch, with
accurate comments. Do not display any details of detected
inconsistencies, because the output could be confusing when
dealing with MariaDB 10.1.x data files.
fsp_flags_convert_from_101(ulint): Convert flags from buggy
MariaDB 10.1.x format, or return ULINT_UNDEFINED if the flags
cannot be in MariaDB 10.1.x format.
fsp_flags_match(): Check the flags when probing files.
Implemented based on fsp_flags_is_valid()
and fsp_flags_convert_from_101().
dict_check_tablespaces_and_store_max_id(): Do not access the
page after committing the mini-transaction.
IMPORT TABLESPACE fixes:
AbstractCallback::init(): Convert the flags.
FetchIndexRootPages::operator(): Check that the tablespace flags match the
table flags. Do not attempt to convert tablespace flags to table flags,
because the conversion would necessarily be lossy.
PageConverter::update_header(): Write back the correct flags.
This takes care of the flags in IMPORT TABLESPACE.
2017-01-14 00:13:16 +02:00
|
|
|
fsp_flags |= FSP_FLAGS_PAGE_SSIZE();
|
2013-12-16 15:38:05 +01:00
|
|
|
|
MDEV-11623 MariaDB 10.1 fails to start datadir created with
MariaDB 10.0/MySQL 5.6 using innodb-page-size!=16K
The storage format of FSP_SPACE_FLAGS was accidentally broken
already in MariaDB 10.1.0. This fix is bringing the format in
line with other MySQL and MariaDB release series.
Please refer to the comments that were added to fsp0fsp.h
for details.
This is an INCOMPATIBLE CHANGE that affects users of
page_compression and non-default innodb_page_size. Upgrading
to this release will correct the flags in the data files.
If you want to downgrade to earlier MariaDB 10.1.x, please refer
to the test innodb.101_compatibility how to reset the
FSP_SPACE_FLAGS in the files.
NOTE: MariaDB 10.1.0 to 10.1.20 can misinterpret
uncompressed data files with innodb_page_size=4k or 64k as
compressed innodb_page_size=16k files, and then probably fail
when trying to access the pages. See the comments in the
function fsp_flags_convert_from_101() for detailed analysis.
Move PAGE_COMPRESSION to FSP_SPACE_FLAGS bit position 16.
In this way, compressed innodb_page_size=16k tablespaces will not
be mistaken for uncompressed ones by MariaDB 10.1.0 to 10.1.20.
Derive PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR from the
dict_table_t::flags when the table is available, in
fil_space_for_table_exists_in_mem() or fil_open_single_table_tablespace().
During crash recovery, fil_load_single_table_tablespace() will use
innodb_compression_level for the PAGE_COMPRESSION_LEVEL.
FSP_FLAGS_MEM_MASK: A bitmap of the memory-only fil_space_t::flags
that are not to be written to FSP_SPACE_FLAGS. Currently, these will
include PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR.
Introduce the macro FSP_FLAGS_PAGE_SSIZE(). We only support
one innodb_page_size for the whole instance.
When creating a dummy tablespace for the redo log, use
fil_space_t::flags=0. The flags are never written to the redo log files.
Remove many FSP_FLAGS_SET_ macros.
dict_tf_verify_flags(): Remove. This is basically only duplicating
the logic of dict_tf_to_fsp_flags(), used in a debug assertion.
fil_space_t::mark: Remove. This flag was not used for anything.
fil_space_for_table_exists_in_mem(): Remove the unnecessary parameter
mark_space, and add a parameter for table flags. Check that
fil_space_t::flags match the table flags, and adjust the (memory-only)
flags based on the table flags.
fil_node_open_file(): Remove some redundant or unreachable conditions,
do not use stderr for output, and avoid unnecessary server aborts.
fil_user_tablespace_restore_page(): Convert the flags, so that the
correct page_size will be used when restoring a page from the
doublewrite buffer.
fil_space_get_page_compressed(), fsp_flags_is_page_compressed(): Remove.
It suffices to have fil_space_is_page_compressed().
FSP_FLAGS_WIDTH_DATA_DIR, FSP_FLAGS_WIDTH_PAGE_COMPRESSION_LEVEL,
FSP_FLAGS_WIDTH_ATOMIC_WRITES: Remove, because these flags do not
exist in the FSP_SPACE_FLAGS but only in memory.
fsp_flags_try_adjust(): New function, to adjust the FSP_SPACE_FLAGS
in page 0. Called by fil_open_single_table_tablespace(),
fil_space_for_table_exists_in_mem(), innobase_start_or_create_for_mysql()
except if --innodb-read-only is active.
fsp_flags_is_valid(ulint): Reimplement from the scratch, with
accurate comments. Do not display any details of detected
inconsistencies, because the output could be confusing when
dealing with MariaDB 10.1.x data files.
fsp_flags_convert_from_101(ulint): Convert flags from buggy
MariaDB 10.1.x format, or return ULINT_UNDEFINED if the flags
cannot be in MariaDB 10.1.x format.
fsp_flags_match(): Check the flags when probing files.
Implemented based on fsp_flags_is_valid()
and fsp_flags_convert_from_101().
dict_check_tablespaces_and_store_max_id(): Do not access the
page after committing the mini-transaction.
IMPORT TABLESPACE fixes:
AbstractCallback::init(): Convert the flags.
FetchIndexRootPages::operator(): Check that the tablespace flags match the
table flags. Do not attempt to convert tablespace flags to table flags,
because the conversion would necessarily be lossy.
PageConverter::update_header(): Write back the correct flags.
This takes care of the flags in IMPORT TABLESPACE.
2017-01-14 00:13:16 +02:00
|
|
|
if (page_compression_level) {
|
|
|
|
fsp_flags |= FSP_FLAGS_MASK_PAGE_COMPRESSION;
|
|
|
|
}
|
2014-02-03 10:08:15 +02:00
|
|
|
|
MDEV-11623 MariaDB 10.1 fails to start datadir created with
MariaDB 10.0/MySQL 5.6 using innodb-page-size!=16K
The storage format of FSP_SPACE_FLAGS was accidentally broken
already in MariaDB 10.1.0. This fix is bringing the format in
line with other MySQL and MariaDB release series.
Please refer to the comments that were added to fsp0fsp.h
for details.
This is an INCOMPATIBLE CHANGE that affects users of
page_compression and non-default innodb_page_size. Upgrading
to this release will correct the flags in the data files.
If you want to downgrade to earlier MariaDB 10.1.x, please refer
to the test innodb.101_compatibility how to reset the
FSP_SPACE_FLAGS in the files.
NOTE: MariaDB 10.1.0 to 10.1.20 can misinterpret
uncompressed data files with innodb_page_size=4k or 64k as
compressed innodb_page_size=16k files, and then probably fail
when trying to access the pages. See the comments in the
function fsp_flags_convert_from_101() for detailed analysis.
Move PAGE_COMPRESSION to FSP_SPACE_FLAGS bit position 16.
In this way, compressed innodb_page_size=16k tablespaces will not
be mistaken for uncompressed ones by MariaDB 10.1.0 to 10.1.20.
Derive PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR from the
dict_table_t::flags when the table is available, in
fil_space_for_table_exists_in_mem() or fil_open_single_table_tablespace().
During crash recovery, fil_load_single_table_tablespace() will use
innodb_compression_level for the PAGE_COMPRESSION_LEVEL.
FSP_FLAGS_MEM_MASK: A bitmap of the memory-only fil_space_t::flags
that are not to be written to FSP_SPACE_FLAGS. Currently, these will
include PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR.
Introduce the macro FSP_FLAGS_PAGE_SSIZE(). We only support
one innodb_page_size for the whole instance.
When creating a dummy tablespace for the redo log, use
fil_space_t::flags=0. The flags are never written to the redo log files.
Remove many FSP_FLAGS_SET_ macros.
dict_tf_verify_flags(): Remove. This is basically only duplicating
the logic of dict_tf_to_fsp_flags(), used in a debug assertion.
fil_space_t::mark: Remove. This flag was not used for anything.
fil_space_for_table_exists_in_mem(): Remove the unnecessary parameter
mark_space, and add a parameter for table flags. Check that
fil_space_t::flags match the table flags, and adjust the (memory-only)
flags based on the table flags.
fil_node_open_file(): Remove some redundant or unreachable conditions,
do not use stderr for output, and avoid unnecessary server aborts.
fil_user_tablespace_restore_page(): Convert the flags, so that the
correct page_size will be used when restoring a page from the
doublewrite buffer.
fil_space_get_page_compressed(), fsp_flags_is_page_compressed(): Remove.
It suffices to have fil_space_is_page_compressed().
FSP_FLAGS_WIDTH_DATA_DIR, FSP_FLAGS_WIDTH_PAGE_COMPRESSION_LEVEL,
FSP_FLAGS_WIDTH_ATOMIC_WRITES: Remove, because these flags do not
exist in the FSP_SPACE_FLAGS but only in memory.
fsp_flags_try_adjust(): New function, to adjust the FSP_SPACE_FLAGS
in page 0. Called by fil_open_single_table_tablespace(),
fil_space_for_table_exists_in_mem(), innobase_start_or_create_for_mysql()
except if --innodb-read-only is active.
fsp_flags_is_valid(ulint): Reimplement from the scratch, with
accurate comments. Do not display any details of detected
inconsistencies, because the output could be confusing when
dealing with MariaDB 10.1.x data files.
fsp_flags_convert_from_101(ulint): Convert flags from buggy
MariaDB 10.1.x format, or return ULINT_UNDEFINED if the flags
cannot be in MariaDB 10.1.x format.
fsp_flags_match(): Check the flags when probing files.
Implemented based on fsp_flags_is_valid()
and fsp_flags_convert_from_101().
dict_check_tablespaces_and_store_max_id(): Do not access the
page after committing the mini-transaction.
IMPORT TABLESPACE fixes:
AbstractCallback::init(): Convert the flags.
FetchIndexRootPages::operator(): Check that the tablespace flags match the
table flags. Do not attempt to convert tablespace flags to table flags,
because the conversion would necessarily be lossy.
PageConverter::update_header(): Write back the correct flags.
This takes care of the flags in IMPORT TABLESPACE.
2017-01-14 00:13:16 +02:00
|
|
|
ut_a(fsp_flags_is_valid(fsp_flags));
|
2014-02-03 10:08:15 +02:00
|
|
|
|
MDEV-11623 MariaDB 10.1 fails to start datadir created with
MariaDB 10.0/MySQL 5.6 using innodb-page-size!=16K
The storage format of FSP_SPACE_FLAGS was accidentally broken
already in MariaDB 10.1.0. This fix is bringing the format in
line with other MySQL and MariaDB release series.
Please refer to the comments that were added to fsp0fsp.h
for details.
This is an INCOMPATIBLE CHANGE that affects users of
page_compression and non-default innodb_page_size. Upgrading
to this release will correct the flags in the data files.
If you want to downgrade to earlier MariaDB 10.1.x, please refer
to the test innodb.101_compatibility how to reset the
FSP_SPACE_FLAGS in the files.
NOTE: MariaDB 10.1.0 to 10.1.20 can misinterpret
uncompressed data files with innodb_page_size=4k or 64k as
compressed innodb_page_size=16k files, and then probably fail
when trying to access the pages. See the comments in the
function fsp_flags_convert_from_101() for detailed analysis.
Move PAGE_COMPRESSION to FSP_SPACE_FLAGS bit position 16.
In this way, compressed innodb_page_size=16k tablespaces will not
be mistaken for uncompressed ones by MariaDB 10.1.0 to 10.1.20.
Derive PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR from the
dict_table_t::flags when the table is available, in
fil_space_for_table_exists_in_mem() or fil_open_single_table_tablespace().
During crash recovery, fil_load_single_table_tablespace() will use
innodb_compression_level for the PAGE_COMPRESSION_LEVEL.
FSP_FLAGS_MEM_MASK: A bitmap of the memory-only fil_space_t::flags
that are not to be written to FSP_SPACE_FLAGS. Currently, these will
include PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR.
Introduce the macro FSP_FLAGS_PAGE_SSIZE(). We only support
one innodb_page_size for the whole instance.
When creating a dummy tablespace for the redo log, use
fil_space_t::flags=0. The flags are never written to the redo log files.
Remove many FSP_FLAGS_SET_ macros.
dict_tf_verify_flags(): Remove. This is basically only duplicating
the logic of dict_tf_to_fsp_flags(), used in a debug assertion.
fil_space_t::mark: Remove. This flag was not used for anything.
fil_space_for_table_exists_in_mem(): Remove the unnecessary parameter
mark_space, and add a parameter for table flags. Check that
fil_space_t::flags match the table flags, and adjust the (memory-only)
flags based on the table flags.
fil_node_open_file(): Remove some redundant or unreachable conditions,
do not use stderr for output, and avoid unnecessary server aborts.
fil_user_tablespace_restore_page(): Convert the flags, so that the
correct page_size will be used when restoring a page from the
doublewrite buffer.
fil_space_get_page_compressed(), fsp_flags_is_page_compressed(): Remove.
It suffices to have fil_space_is_page_compressed().
FSP_FLAGS_WIDTH_DATA_DIR, FSP_FLAGS_WIDTH_PAGE_COMPRESSION_LEVEL,
FSP_FLAGS_WIDTH_ATOMIC_WRITES: Remove, because these flags do not
exist in the FSP_SPACE_FLAGS but only in memory.
fsp_flags_try_adjust(): New function, to adjust the FSP_SPACE_FLAGS
in page 0. Called by fil_open_single_table_tablespace(),
fil_space_for_table_exists_in_mem(), innobase_start_or_create_for_mysql()
except if --innodb-read-only is active.
fsp_flags_is_valid(ulint): Reimplement from the scratch, with
accurate comments. Do not display any details of detected
inconsistencies, because the output could be confusing when
dealing with MariaDB 10.1.x data files.
fsp_flags_convert_from_101(ulint): Convert flags from buggy
MariaDB 10.1.x format, or return ULINT_UNDEFINED if the flags
cannot be in MariaDB 10.1.x format.
fsp_flags_match(): Check the flags when probing files.
Implemented based on fsp_flags_is_valid()
and fsp_flags_convert_from_101().
dict_check_tablespaces_and_store_max_id(): Do not access the
page after committing the mini-transaction.
IMPORT TABLESPACE fixes:
AbstractCallback::init(): Convert the flags.
FetchIndexRootPages::operator(): Check that the tablespace flags match the
table flags. Do not attempt to convert tablespace flags to table flags,
because the conversion would necessarily be lossy.
PageConverter::update_header(): Write back the correct flags.
This takes care of the flags in IMPORT TABLESPACE.
2017-01-14 00:13:16 +02:00
|
|
|
if (DICT_TF_HAS_DATA_DIR(table_flags)) {
|
|
|
|
fsp_flags |= 1U << FSP_FLAGS_MEM_DATA_DIR;
|
|
|
|
}
|
2014-02-03 10:08:15 +02:00
|
|
|
|
MDEV-11623 MariaDB 10.1 fails to start datadir created with
MariaDB 10.0/MySQL 5.6 using innodb-page-size!=16K
The storage format of FSP_SPACE_FLAGS was accidentally broken
already in MariaDB 10.1.0. This fix is bringing the format in
line with other MySQL and MariaDB release series.
Please refer to the comments that were added to fsp0fsp.h
for details.
This is an INCOMPATIBLE CHANGE that affects users of
page_compression and non-default innodb_page_size. Upgrading
to this release will correct the flags in the data files.
If you want to downgrade to earlier MariaDB 10.1.x, please refer
to the test innodb.101_compatibility how to reset the
FSP_SPACE_FLAGS in the files.
NOTE: MariaDB 10.1.0 to 10.1.20 can misinterpret
uncompressed data files with innodb_page_size=4k or 64k as
compressed innodb_page_size=16k files, and then probably fail
when trying to access the pages. See the comments in the
function fsp_flags_convert_from_101() for detailed analysis.
Move PAGE_COMPRESSION to FSP_SPACE_FLAGS bit position 16.
In this way, compressed innodb_page_size=16k tablespaces will not
be mistaken for uncompressed ones by MariaDB 10.1.0 to 10.1.20.
Derive PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR from the
dict_table_t::flags when the table is available, in
fil_space_for_table_exists_in_mem() or fil_open_single_table_tablespace().
During crash recovery, fil_load_single_table_tablespace() will use
innodb_compression_level for the PAGE_COMPRESSION_LEVEL.
FSP_FLAGS_MEM_MASK: A bitmap of the memory-only fil_space_t::flags
that are not to be written to FSP_SPACE_FLAGS. Currently, these will
include PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES and DATA_DIR.
Introduce the macro FSP_FLAGS_PAGE_SSIZE(). We only support
one innodb_page_size for the whole instance.
When creating a dummy tablespace for the redo log, use
fil_space_t::flags=0. The flags are never written to the redo log files.
Remove many FSP_FLAGS_SET_ macros.
dict_tf_verify_flags(): Remove. This is basically only duplicating
the logic of dict_tf_to_fsp_flags(), used in a debug assertion.
fil_space_t::mark: Remove. This flag was not used for anything.
fil_space_for_table_exists_in_mem(): Remove the unnecessary parameter
mark_space, and add a parameter for table flags. Check that
fil_space_t::flags match the table flags, and adjust the (memory-only)
flags based on the table flags.
fil_node_open_file(): Remove some redundant or unreachable conditions,
do not use stderr for output, and avoid unnecessary server aborts.
fil_user_tablespace_restore_page(): Convert the flags, so that the
correct page_size will be used when restoring a page from the
doublewrite buffer.
fil_space_get_page_compressed(), fsp_flags_is_page_compressed(): Remove.
It suffices to have fil_space_is_page_compressed().
FSP_FLAGS_WIDTH_DATA_DIR, FSP_FLAGS_WIDTH_PAGE_COMPRESSION_LEVEL,
FSP_FLAGS_WIDTH_ATOMIC_WRITES: Remove, because these flags do not
exist in the FSP_SPACE_FLAGS but only in memory.
fsp_flags_try_adjust(): New function, to adjust the FSP_SPACE_FLAGS
in page 0. Called by fil_open_single_table_tablespace(),
fil_space_for_table_exists_in_mem(), innobase_start_or_create_for_mysql()
except if --innodb-read-only is active.
fsp_flags_is_valid(ulint): Reimplement from the scratch, with
accurate comments. Do not display any details of detected
inconsistencies, because the output could be confusing when
dealing with MariaDB 10.1.x data files.
fsp_flags_convert_from_101(ulint): Convert flags from buggy
MariaDB 10.1.x format, or return ULINT_UNDEFINED if the flags
cannot be in MariaDB 10.1.x format.
fsp_flags_match(): Check the flags when probing files.
Implemented based on fsp_flags_is_valid()
and fsp_flags_convert_from_101().
dict_check_tablespaces_and_store_max_id(): Do not access the
page after committing the mini-transaction.
IMPORT TABLESPACE fixes:
AbstractCallback::init(): Convert the flags.
FetchIndexRootPages::operator(): Check that the tablespace flags match the
table flags. Do not attempt to convert tablespace flags to table flags,
because the conversion would necessarily be lossy.
PageConverter::update_header(): Write back the correct flags.
This takes care of the flags in IMPORT TABLESPACE.
2017-01-14 00:13:16 +02:00
|
|
|
fsp_flags |= atomic_writes << FSP_FLAGS_MEM_ATOMIC_WRITES;
|
|
|
|
fsp_flags |= page_compression_level << FSP_FLAGS_MEM_COMPRESSION_LEVEL;
|
2013-12-16 15:38:05 +01:00
|
|
|
|
|
|
|
return(fsp_flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
/********************************************************************//**
|
|
|
|
Convert a 32 bit integer from SYS_TABLES.TYPE to dict_table_t::flags
|
|
|
|
The following chart shows the translation of the low order bit.
|
|
|
|
Other bits are the same.
|
|
|
|
========================= Low order bit ==========================
|
|
|
|
| REDUNDANT | COMPACT | COMPRESSED and DYNAMIC
|
|
|
|
SYS_TABLES.TYPE | 1 | 1 | 1
|
|
|
|
dict_table_t::flags | 0 | 1 | 1
|
|
|
|
==================================================================
|
|
|
|
@return ulint containing SYS_TABLES.TYPE */
|
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_sys_tables_type_to_tf(
|
|
|
|
/*=======================*/
|
|
|
|
ulint type, /*!< in: SYS_TABLES.TYPE field */
|
|
|
|
ulint n_cols) /*!< in: SYS_TABLES.N_COLS field */
|
|
|
|
{
|
|
|
|
ulint flags;
|
|
|
|
ulint redundant = !(n_cols & DICT_N_COLS_COMPACT);
|
|
|
|
|
|
|
|
/* Adjust bit zero. */
|
|
|
|
flags = redundant ? 0 : 1;
|
|
|
|
|
2014-02-03 10:08:15 +02:00
|
|
|
/* ZIP_SSIZE, ATOMIC_BLOBS, DATA_DIR, PAGE_COMPRESSION,
|
|
|
|
PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES are the same. */
|
2013-12-16 15:38:05 +01:00
|
|
|
flags |= type & (DICT_TF_MASK_ZIP_SSIZE
|
|
|
|
| DICT_TF_MASK_ATOMIC_BLOBS
|
2014-02-03 10:08:15 +02:00
|
|
|
| DICT_TF_MASK_DATA_DIR
|
|
|
|
| DICT_TF_MASK_PAGE_COMPRESSION
|
|
|
|
| DICT_TF_MASK_PAGE_COMPRESSION_LEVEL
|
|
|
|
| DICT_TF_MASK_ATOMIC_WRITES
|
2014-12-22 16:53:17 +02:00
|
|
|
|
2014-02-03 10:08:15 +02:00
|
|
|
);
|
2013-12-16 15:38:05 +01:00
|
|
|
|
|
|
|
return(flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
/********************************************************************//**
|
|
|
|
Convert a 32 bit integer table flags to the 32bit integer that is written
|
|
|
|
to a SYS_TABLES.TYPE field. The following chart shows the translation of
|
|
|
|
the low order bit. Other bits are the same.
|
|
|
|
========================= Low order bit ==========================
|
|
|
|
| REDUNDANT | COMPACT | COMPRESSED and DYNAMIC
|
|
|
|
dict_table_t::flags | 0 | 1 | 1
|
|
|
|
SYS_TABLES.TYPE | 1 | 1 | 1
|
|
|
|
==================================================================
|
|
|
|
@return ulint containing SYS_TABLES.TYPE */
|
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_tf_to_sys_tables_type(
|
|
|
|
/*=======================*/
|
|
|
|
ulint flags) /*!< in: dict_table_t::flags */
|
|
|
|
{
|
|
|
|
ulint type;
|
|
|
|
|
|
|
|
ut_a(dict_tf_is_valid(flags));
|
|
|
|
|
|
|
|
/* Adjust bit zero. It is always 1 in SYS_TABLES.TYPE */
|
|
|
|
type = 1;
|
|
|
|
|
2014-02-03 10:08:15 +02:00
|
|
|
/* ZIP_SSIZE, ATOMIC_BLOBS, DATA_DIR, PAGE_COMPRESSION,
|
|
|
|
PAGE_COMPRESSION_LEVEL, ATOMIC_WRITES are the same. */
|
2013-12-16 15:38:05 +01:00
|
|
|
type |= flags & (DICT_TF_MASK_ZIP_SSIZE
|
|
|
|
| DICT_TF_MASK_ATOMIC_BLOBS
|
2014-02-03 10:08:15 +02:00
|
|
|
| DICT_TF_MASK_DATA_DIR
|
|
|
|
| DICT_TF_MASK_PAGE_COMPRESSION
|
|
|
|
| DICT_TF_MASK_PAGE_COMPRESSION_LEVEL
|
2015-04-01 11:50:21 +03:00
|
|
|
| DICT_TF_MASK_ATOMIC_WRITES);
|
2013-12-16 15:38:05 +01:00
|
|
|
|
|
|
|
return(type);
|
|
|
|
}
|
|
|
|
|
|
|
|
/********************************************************************//**
|
|
|
|
Extract the compressed page size from dict_table_t::flags.
|
|
|
|
These flags are in memory, so assert that they are valid.
|
2009-09-07 10:22:53 +00:00
|
|
|
@return compressed page size, or 0 if not compressed */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
2013-12-16 15:38:05 +01:00
|
|
|
dict_tf_get_zip_size(
|
|
|
|
/*=================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
ulint flags) /*!< in: flags */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
2013-12-16 15:38:05 +01:00
|
|
|
ulint zip_ssize = DICT_TF_GET_ZIP_SSIZE(flags);
|
|
|
|
ulint zip_size = (zip_ssize
|
|
|
|
? (UNIV_ZIP_SIZE_MIN >> 1) << zip_ssize
|
|
|
|
: 0);
|
2008-11-30 22:10:29 -08:00
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
ut_ad(zip_size <= UNIV_ZIP_SIZE_MAX);
|
2008-11-30 22:10:29 -08:00
|
|
|
|
|
|
|
return(zip_size);
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Check whether the table uses the compressed compact page format.
|
|
|
|
@return compressed page size, or 0 if not compressed */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_table_zip_size(
|
|
|
|
/*================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_table_t* table) /*!< in: table */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(table);
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
return(dict_tf_get_zip_size(table->flags));
|
2008-11-30 22:10:29 -08:00
|
|
|
}
|
|
|
|
|
2012-09-05 10:44:23 +02:00
|
|
|
#ifndef UNIV_HOTBACKUP
|
2010-10-19 12:27:31 +02:00
|
|
|
/*********************************************************************//**
|
|
|
|
Obtain exclusive locks on all index trees of the table. This is to prevent
|
|
|
|
accessing index trees while InnoDB is updating internal metadata for
|
|
|
|
operations such as truncate tables. */
|
|
|
|
UNIV_INLINE
|
|
|
|
void
|
|
|
|
dict_table_x_lock_indexes(
|
|
|
|
/*======================*/
|
|
|
|
dict_table_t* table) /*!< in: table */
|
|
|
|
{
|
|
|
|
dict_index_t* index;
|
|
|
|
|
|
|
|
ut_a(table);
|
|
|
|
ut_ad(mutex_own(&(dict_sys->mutex)));
|
|
|
|
|
|
|
|
/* Loop through each index of the table and lock them */
|
|
|
|
for (index = dict_table_get_first_index(table);
|
|
|
|
index != NULL;
|
|
|
|
index = dict_table_get_next_index(index)) {
|
|
|
|
rw_lock_x_lock(dict_index_get_lock(index));
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*********************************************************************//**
|
|
|
|
Release the exclusive locks on all index tree. */
|
|
|
|
UNIV_INLINE
|
|
|
|
void
|
|
|
|
dict_table_x_unlock_indexes(
|
|
|
|
/*========================*/
|
|
|
|
dict_table_t* table) /*!< in: table */
|
|
|
|
{
|
|
|
|
dict_index_t* index;
|
|
|
|
|
|
|
|
ut_a(table);
|
|
|
|
ut_ad(mutex_own(&(dict_sys->mutex)));
|
|
|
|
|
|
|
|
for (index = dict_table_get_first_index(table);
|
|
|
|
index != NULL;
|
|
|
|
index = dict_table_get_next_index(index)) {
|
|
|
|
rw_lock_x_unlock(dict_index_get_lock(index));
|
|
|
|
}
|
|
|
|
}
|
2012-09-05 10:44:23 +02:00
|
|
|
#endif /* !UNIV_HOTBACKUP */
|
2013-12-16 15:38:05 +01:00
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
2008-11-30 22:10:29 -08:00
|
|
|
Gets the number of fields in the internal representation of an index,
|
2009-09-07 10:22:53 +00:00
|
|
|
including fields added by the dictionary system.
|
|
|
|
@return number of fields */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_get_n_fields(
|
|
|
|
/*====================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_index_t* index) /*!< in: an internal
|
2008-11-30 22:10:29 -08:00
|
|
|
representation of index (in
|
|
|
|
the dictionary cache) */
|
|
|
|
{
|
|
|
|
ut_ad(index);
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
|
|
|
|
|
|
|
return(index->n_fields);
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
2008-11-30 22:10:29 -08:00
|
|
|
Gets the number of fields in the internal representation of an index
|
|
|
|
that uniquely determine the position of an index entry in the index, if
|
|
|
|
we do not take multiversioning into account: in the B-tree use the value
|
2009-09-07 10:22:53 +00:00
|
|
|
returned by dict_index_get_n_unique_in_tree.
|
|
|
|
@return number of fields */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_get_n_unique(
|
|
|
|
/*====================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_index_t* index) /*!< in: an internal representation
|
2008-11-30 22:10:29 -08:00
|
|
|
of index (in the dictionary cache) */
|
|
|
|
{
|
|
|
|
ut_ad(index);
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
|
|
|
ut_ad(index->cached);
|
|
|
|
|
|
|
|
return(index->n_uniq);
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
2008-11-30 22:10:29 -08:00
|
|
|
Gets the number of fields in the internal representation of an index
|
|
|
|
which uniquely determine the position of an index entry in the index, if
|
2009-09-07 10:22:53 +00:00
|
|
|
we also take multiversioning into account.
|
|
|
|
@return number of fields */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_get_n_unique_in_tree(
|
|
|
|
/*============================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_index_t* index) /*!< in: an internal representation
|
2008-11-30 22:10:29 -08:00
|
|
|
of index (in the dictionary cache) */
|
|
|
|
{
|
|
|
|
ut_ad(index);
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
|
|
|
ut_ad(index->cached);
|
|
|
|
|
|
|
|
if (dict_index_is_clust(index)) {
|
|
|
|
|
|
|
|
return(dict_index_get_n_unique(index));
|
|
|
|
}
|
|
|
|
|
|
|
|
return(dict_index_get_n_fields(index));
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
2008-11-30 22:10:29 -08:00
|
|
|
Gets the number of user-defined ordering fields in the index. In the internal
|
|
|
|
representation of clustered indexes we add the row id to the ordering fields
|
|
|
|
to make a clustered index unique, but this function returns the number of
|
2009-09-07 10:22:53 +00:00
|
|
|
fields the user defined in the index as ordering fields.
|
|
|
|
@return number of fields */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_get_n_ordering_defined_by_user(
|
|
|
|
/*======================================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_index_t* index) /*!< in: an internal representation
|
2008-11-30 22:10:29 -08:00
|
|
|
of index (in the dictionary cache) */
|
|
|
|
{
|
|
|
|
return(index->n_user_defined_cols);
|
|
|
|
}
|
|
|
|
|
|
|
|
#ifdef UNIV_DEBUG
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Gets the nth field of an index.
|
|
|
|
@return pointer to field object */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
dict_field_t*
|
|
|
|
dict_index_get_nth_field(
|
|
|
|
/*=====================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_index_t* index, /*!< in: index */
|
|
|
|
ulint pos) /*!< in: position of field */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(index);
|
|
|
|
ut_ad(pos < index->n_def);
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
|
|
|
|
|
|
|
return((dict_field_t*) (index->fields) + pos);
|
|
|
|
}
|
|
|
|
#endif /* UNIV_DEBUG */
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Returns the position of a system column in an index.
|
|
|
|
@return position, ULINT_UNDEFINED if not contained */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_get_sys_col_pos(
|
|
|
|
/*=======================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_index_t* index, /*!< in: index */
|
|
|
|
ulint type) /*!< in: DATA_ROW_ID, ... */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(index);
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
2013-12-16 15:38:05 +01:00
|
|
|
ut_ad(!dict_index_is_univ(index));
|
2008-11-30 22:10:29 -08:00
|
|
|
|
|
|
|
if (dict_index_is_clust(index)) {
|
|
|
|
|
|
|
|
return(dict_col_get_clust_pos(
|
|
|
|
dict_table_get_sys_col(index->table, type),
|
|
|
|
index));
|
|
|
|
}
|
|
|
|
|
|
|
|
return(dict_index_get_nth_col_pos(
|
2014-11-03 11:18:52 +02:00
|
|
|
index, dict_table_get_sys_col_no(index->table, type),
|
|
|
|
NULL));
|
2008-11-30 22:10:29 -08:00
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/*********************************************************************//**
|
|
|
|
Gets the field column.
|
|
|
|
@return field->col, pointer to the table column */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
const dict_col_t*
|
|
|
|
dict_field_get_col(
|
|
|
|
/*===============*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_field_t* field) /*!< in: index field */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(field);
|
|
|
|
|
|
|
|
return(field->col);
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Gets pointer to the nth column in an index.
|
|
|
|
@return column */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
const dict_col_t*
|
|
|
|
dict_index_get_nth_col(
|
|
|
|
/*===================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_index_t* index, /*!< in: index */
|
|
|
|
ulint pos) /*!< in: position of the field */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
return(dict_field_get_col(dict_index_get_nth_field(index, pos)));
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
|
|
|
Gets the column number the nth field in an index.
|
|
|
|
@return column number */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_get_nth_col_no(
|
|
|
|
/*======================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_index_t* index, /*!< in: index */
|
|
|
|
ulint pos) /*!< in: position of the field */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
return(dict_col_get_no(dict_index_get_nth_col(index, pos)));
|
|
|
|
}
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
/********************************************************************//**
|
|
|
|
Looks for column n in an index.
|
|
|
|
@return position in internal representation of the index;
|
|
|
|
ULINT_UNDEFINED if not contained */
|
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_get_nth_col_pos(
|
|
|
|
/*=======================*/
|
|
|
|
const dict_index_t* index, /*!< in: index */
|
2014-11-03 11:18:52 +02:00
|
|
|
ulint n, /*!< in: column number */
|
|
|
|
ulint* prefix_col_pos) /*!< out: col num if prefix */
|
2013-12-16 15:38:05 +01:00
|
|
|
{
|
2014-11-03 11:18:52 +02:00
|
|
|
return(dict_index_get_nth_col_or_prefix_pos(index, n, FALSE,
|
|
|
|
prefix_col_pos));
|
2013-12-16 15:38:05 +01:00
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
#ifndef UNIV_HOTBACKUP
|
|
|
|
/********************************************************************//**
|
|
|
|
Returns the minimum data size of an index record.
|
|
|
|
@return minimum data size in bytes */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_get_min_size(
|
|
|
|
/*====================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_index_t* index) /*!< in: index */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ulint n = dict_index_get_n_fields(index);
|
|
|
|
ulint size = 0;
|
|
|
|
|
|
|
|
while (n--) {
|
|
|
|
size += dict_col_get_min_size(dict_index_get_nth_col(index,
|
|
|
|
n));
|
|
|
|
}
|
|
|
|
|
|
|
|
return(size);
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/*********************************************************************//**
|
|
|
|
Gets the space id of the root of the index tree.
|
|
|
|
@return space id */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_get_space(
|
|
|
|
/*=================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_index_t* index) /*!< in: index */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(index);
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
|
|
|
|
|
|
|
return(index->space);
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/*********************************************************************//**
|
2008-11-30 22:10:29 -08:00
|
|
|
Sets the space id of the root of the index tree. */
|
|
|
|
UNIV_INLINE
|
|
|
|
void
|
|
|
|
dict_index_set_space(
|
|
|
|
/*=================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
dict_index_t* index, /*!< in/out: index */
|
|
|
|
ulint space) /*!< in: space id */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(index);
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
|
|
|
|
|
|
|
index->space = space;
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/*********************************************************************//**
|
|
|
|
Gets the page number of the root of the index tree.
|
|
|
|
@return page number */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_get_page(
|
|
|
|
/*================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
const dict_index_t* index) /*!< in: index */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(index);
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
|
|
|
|
|
|
|
return(index->page);
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/*********************************************************************//**
|
|
|
|
Gets the read-write lock of the index tree.
|
|
|
|
@return read-write lock */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
2013-12-16 15:38:05 +01:00
|
|
|
prio_rw_lock_t*
|
2008-11-30 22:10:29 -08:00
|
|
|
dict_index_get_lock(
|
|
|
|
/*================*/
|
2009-09-07 10:22:53 +00:00
|
|
|
dict_index_t* index) /*!< in: index */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
|
|
|
ut_ad(index);
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
|
|
|
|
|
|
|
return(&(index->lock));
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/********************************************************************//**
|
2008-11-30 22:10:29 -08:00
|
|
|
Returns free space reserved for future updates of records. This is
|
|
|
|
relevant only in the case of many consecutive inserts, as updates
|
2009-09-07 10:22:53 +00:00
|
|
|
which make the records bigger might fragment the index.
|
|
|
|
@return number of free bytes on page, reserved for updates */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_get_space_reserve(void)
|
|
|
|
/*==============================*/
|
|
|
|
{
|
|
|
|
return(UNIV_PAGE_SIZE / 16);
|
|
|
|
}
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
/********************************************************************//**
|
|
|
|
Gets the status of online index creation.
|
|
|
|
@return the status */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
2013-12-16 15:38:05 +01:00
|
|
|
enum online_index_status
|
|
|
|
dict_index_get_online_status(
|
|
|
|
/*=========================*/
|
|
|
|
const dict_index_t* index) /*!< in: secondary index */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
2013-12-16 15:38:05 +01:00
|
|
|
enum online_index_status status;
|
2008-11-30 22:10:29 -08:00
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
status = (enum online_index_status) index->online_status;
|
2008-11-30 22:10:29 -08:00
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
/* Without the index->lock protection, the online
|
|
|
|
status can change from ONLINE_INDEX_CREATION to
|
|
|
|
ONLINE_INDEX_COMPLETE (or ONLINE_INDEX_ABORTED) in
|
|
|
|
row_log_apply() once log application is done. So to make
|
|
|
|
sure the status is ONLINE_INDEX_CREATION or ONLINE_INDEX_COMPLETE
|
|
|
|
you should always do the recheck after acquiring index->lock */
|
2009-06-24 18:43:25 -07:00
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
#ifdef UNIV_DEBUG
|
|
|
|
switch (status) {
|
|
|
|
case ONLINE_INDEX_COMPLETE:
|
|
|
|
case ONLINE_INDEX_CREATION:
|
|
|
|
case ONLINE_INDEX_ABORTED:
|
|
|
|
case ONLINE_INDEX_ABORTED_DROPPED:
|
|
|
|
return(status);
|
2009-06-24 18:43:25 -07:00
|
|
|
}
|
2013-12-16 15:38:05 +01:00
|
|
|
ut_error;
|
|
|
|
#endif /* UNIV_DEBUG */
|
|
|
|
return(status);
|
2008-11-30 22:10:29 -08:00
|
|
|
}
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
/********************************************************************//**
|
|
|
|
Sets the status of online index creation. */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
2013-12-16 15:38:05 +01:00
|
|
|
void
|
|
|
|
dict_index_set_online_status(
|
|
|
|
/*=========================*/
|
|
|
|
dict_index_t* index, /*!< in/out: index */
|
|
|
|
enum online_index_status status) /*!< in: status */
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
2013-12-16 15:38:05 +01:00
|
|
|
ut_ad(!(index->type & DICT_FTS));
|
|
|
|
#ifdef UNIV_SYNC_DEBUG
|
|
|
|
ut_ad(rw_lock_own(dict_index_get_lock(index), RW_LOCK_EX));
|
|
|
|
#endif /* UNIV_SYNC_DEBUG */
|
|
|
|
#ifdef UNIV_DEBUG
|
|
|
|
switch (dict_index_get_online_status(index)) {
|
|
|
|
case ONLINE_INDEX_COMPLETE:
|
|
|
|
case ONLINE_INDEX_CREATION:
|
|
|
|
break;
|
|
|
|
case ONLINE_INDEX_ABORTED:
|
|
|
|
ut_ad(status == ONLINE_INDEX_ABORTED_DROPPED);
|
|
|
|
break;
|
|
|
|
case ONLINE_INDEX_ABORTED_DROPPED:
|
|
|
|
ut_error;
|
|
|
|
}
|
|
|
|
#endif /* UNIV_DEBUG */
|
2008-11-30 22:10:29 -08:00
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
index->online_status = status;
|
|
|
|
ut_ad(dict_index_get_online_status(index) == status);
|
|
|
|
}
|
2008-11-30 22:10:29 -08:00
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
/********************************************************************//**
|
|
|
|
Determines if a secondary index is being or has been created online,
|
|
|
|
or if the table is being rebuilt online, allowing concurrent modifications
|
|
|
|
to the table.
|
|
|
|
@retval true if the index is being or has been built online, or
|
|
|
|
if this is a clustered index and the table is being or has been rebuilt online
|
|
|
|
@retval false if the index has been created or the table has been
|
|
|
|
rebuilt completely */
|
|
|
|
UNIV_INLINE
|
|
|
|
bool
|
|
|
|
dict_index_is_online_ddl(
|
|
|
|
/*=====================*/
|
|
|
|
const dict_index_t* index) /*!< in: index */
|
|
|
|
{
|
|
|
|
#ifdef UNIV_DEBUG
|
|
|
|
if (dict_index_is_clust(index)) {
|
|
|
|
switch (dict_index_get_online_status(index)) {
|
|
|
|
case ONLINE_INDEX_CREATION:
|
|
|
|
return(true);
|
|
|
|
case ONLINE_INDEX_COMPLETE:
|
|
|
|
return(false);
|
|
|
|
case ONLINE_INDEX_ABORTED:
|
|
|
|
case ONLINE_INDEX_ABORTED_DROPPED:
|
|
|
|
break;
|
2011-12-14 14:58:22 +01:00
|
|
|
}
|
2013-12-16 15:38:05 +01:00
|
|
|
ut_ad(0);
|
|
|
|
return(false);
|
2011-12-14 14:58:22 +01:00
|
|
|
}
|
2013-12-16 15:38:05 +01:00
|
|
|
#endif /* UNIV_DEBUG */
|
2011-12-14 14:58:22 +01:00
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
return(UNIV_UNLIKELY(dict_index_get_online_status(index)
|
|
|
|
!= ONLINE_INDEX_COMPLETE));
|
2008-11-30 22:10:29 -08:00
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
/**********************************************************************//**
|
2013-12-16 15:38:05 +01:00
|
|
|
Check whether a column exists in an FTS index.
|
|
|
|
@return ULINT_UNDEFINED if no match else the offset within the vector */
|
2008-11-30 22:10:29 -08:00
|
|
|
UNIV_INLINE
|
2013-12-16 15:38:05 +01:00
|
|
|
ulint
|
|
|
|
dict_table_is_fts_column(
|
2008-11-30 22:10:29 -08:00
|
|
|
/*=====================*/
|
2013-12-16 15:38:05 +01:00
|
|
|
ib_vector_t* indexes,/*!< in: vector containing only FTS indexes */
|
|
|
|
ulint col_no) /*!< in: col number to search for */
|
|
|
|
|
2008-11-30 22:10:29 -08:00
|
|
|
{
|
2013-12-16 15:38:05 +01:00
|
|
|
ulint i;
|
2008-11-30 22:10:29 -08:00
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
for (i = 0; i < ib_vector_size(indexes); ++i) {
|
|
|
|
dict_index_t* index;
|
2008-11-30 22:10:29 -08:00
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
index = (dict_index_t*) ib_vector_getp(indexes, i);
|
2008-11-30 22:10:29 -08:00
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
if (dict_index_contains_col_or_prefix(index, col_no)) {
|
2008-11-30 22:10:29 -08:00
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
return(i);
|
|
|
|
}
|
2009-06-24 18:43:25 -07:00
|
|
|
}
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
return(ULINT_UNDEFINED);
|
2008-11-30 22:10:29 -08:00
|
|
|
}
|
2011-07-17 11:28:48 +02:00
|
|
|
|
|
|
|
/**********************************************************************//**
|
|
|
|
Determine bytes of column prefix to be stored in the undo log. Please
|
2013-12-16 15:38:05 +01:00
|
|
|
note if the table format is UNIV_FORMAT_A (< UNIV_FORMAT_B), no prefix
|
2011-07-17 11:28:48 +02:00
|
|
|
needs to be stored in the undo log.
|
|
|
|
@return bytes of column prefix to be stored in the undo log */
|
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_max_field_len_store_undo(
|
|
|
|
/*==========================*/
|
|
|
|
dict_table_t* table, /*!< in: table */
|
|
|
|
const dict_col_t* col) /*!< in: column which index prefix
|
|
|
|
is based on */
|
|
|
|
{
|
2013-12-16 15:38:05 +01:00
|
|
|
ulint prefix_len = 0;
|
2011-07-17 11:28:48 +02:00
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
if (dict_table_get_format(table) >= UNIV_FORMAT_B)
|
2011-07-17 11:28:48 +02:00
|
|
|
{
|
|
|
|
prefix_len = col->max_prefix
|
|
|
|
? col->max_prefix
|
|
|
|
: DICT_MAX_FIELD_LEN_BY_FORMAT(table);
|
|
|
|
}
|
|
|
|
|
|
|
|
return(prefix_len);
|
|
|
|
}
|
|
|
|
|
2011-12-14 14:58:22 +01:00
|
|
|
/********************************************************************//**
|
|
|
|
Check whether the index is corrupted.
|
|
|
|
@return nonzero for corrupted index, zero for valid indexes */
|
|
|
|
UNIV_INLINE
|
|
|
|
ulint
|
|
|
|
dict_index_is_corrupted(
|
|
|
|
/*====================*/
|
|
|
|
const dict_index_t* index) /*!< in: index */
|
|
|
|
{
|
|
|
|
ut_ad(index->magic_n == DICT_INDEX_MAGIC_N);
|
|
|
|
|
2013-12-16 15:38:05 +01:00
|
|
|
return((index->type & DICT_CORRUPT)
|
MDEV-12253: Buffer pool blocks are accessed after they have been freed
Problem was that bpage was referenced after it was already freed
from LRU. Fixed by adding a new variable encrypted that is
passed down to buf_page_check_corrupt() and used in
buf_page_get_gen() to stop processing page read.
This patch should also address following test failures and
bugs:
MDEV-12419: IMPORT should not look up tablespace in
PageConverter::validate(). This is now removed.
MDEV-10099: encryption.innodb_onlinealter_encryption fails
sporadically in buildbot
MDEV-11420: encryption.innodb_encryption-page-compression
failed in buildbot
MDEV-11222: encryption.encrypt_and_grep failed in buildbot on P8
Removed dict_table_t::is_encrypted and dict_table_t::ibd_file_missing
and replaced these with dict_table_t::file_unreadable. Table
ibd file is missing if fil_get_space(space_id) returns NULL
and encrypted if not. Removed dict_table_t::is_corrupted field.
Ported FilSpace class from 10.2 and using that on buf_page_check_corrupt(),
buf_page_decrypt_after_read(), buf_page_encrypt_before_write(),
buf_dblwr_process(), buf_read_page(), dict_stats_save_defrag_stats().
Added test cases when enrypted page could be read while doing
redo log crash recovery. Also added test case for row compressed
blobs.
btr_cur_open_at_index_side_func(),
btr_cur_open_at_rnd_pos_func(): Avoid referencing block that is
NULL.
buf_page_get_zip(): Issue error if page read fails.
buf_page_get_gen(): Use dberr_t for error detection and
do not reference bpage after we hare freed it.
buf_mark_space_corrupt(): remove bpage from LRU also when
it is encrypted.
buf_page_check_corrupt(): @return DB_SUCCESS if page has
been read and is not corrupted,
DB_PAGE_CORRUPTED if page based on checksum check is corrupted,
DB_DECRYPTION_FAILED if page post encryption checksum matches but
after decryption normal page checksum does not match. In read
case only DB_SUCCESS is possible.
buf_page_io_complete(): use dberr_t for error handling.
buf_flush_write_block_low(),
buf_read_ahead_random(),
buf_read_page_async(),
buf_read_ahead_linear(),
buf_read_ibuf_merge_pages(),
buf_read_recv_pages(),
fil_aio_wait():
Issue error if page read fails.
btr_pcur_move_to_next_page(): Do not reference page if it is
NULL.
Introduced dict_table_t::is_readable() and dict_index_t::is_readable()
that will return true if tablespace exists and pages read from
tablespace are not corrupted or page decryption failed.
Removed buf_page_t::key_version. After page decryption the
key version is not removed from page frame. For unencrypted
pages, old key_version is removed at buf_page_encrypt_before_write()
dict_stats_update_transient_for_index(),
dict_stats_update_transient()
Do not continue if table decryption failed or table
is corrupted.
dict0stats.cc: Introduced a dict_stats_report_error function
to avoid code duplication.
fil_parse_write_crypt_data():
Check that key read from redo log entry is found from
encryption plugin and if it is not, refuse to start.
PageConverter::validate(): Removed access to fil_space_t as
tablespace is not available during import.
Fixed error code on innodb.innodb test.
Merged test cased innodb-bad-key-change5 and innodb-bad-key-shutdown
to innodb-bad-key-change2. Removed innodb-bad-key-change5 test.
Decreased unnecessary complexity on some long lasting tests.
Removed fil_inc_pending_ops(), fil_decr_pending_ops(),
fil_get_first_space(), fil_get_next_space(),
fil_get_first_space_safe(), fil_get_next_space_safe()
functions.
fil_space_verify_crypt_checksum(): Fixed bug found using ASAN
where FIL_PAGE_END_LSN_OLD_CHECKSUM field was incorrectly
accessed from row compressed tables. Fixed out of page frame
bug for row compressed tables in
fil_space_verify_crypt_checksum() found using ASAN. Incorrect
function was called for compressed table.
Added new tests for discard, rename table and drop (we should allow them
even when page decryption fails). Alter table rename is not allowed.
Added test for restart with innodb-force-recovery=1 when page read on
redo-recovery cant be decrypted. Added test for corrupted table where
both page data and FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION is corrupted.
Adjusted the test case innodb_bug14147491 so that it does not anymore
expect crash. Instead table is just mostly not usable.
fil0fil.h: fil_space_acquire_low is not visible function
and fil_space_acquire and fil_space_acquire_silent are
inline functions. FilSpace class uses fil_space_acquire_low
directly.
recv_apply_hashed_log_recs() does not return anything.
2017-04-26 15:19:16 +03:00
|
|
|
|| (index->table && index->table->corrupted));
|
2013-12-16 15:38:05 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
/********************************************************************//**
|
|
|
|
Check if the tablespace for the table has been discarded.
|
|
|
|
@return true if the tablespace has been discarded. */
|
|
|
|
UNIV_INLINE
|
|
|
|
bool
|
|
|
|
dict_table_is_discarded(
|
|
|
|
/*====================*/
|
|
|
|
const dict_table_t* table) /*!< in: table to check */
|
|
|
|
{
|
|
|
|
return(DICT_TF2_FLAG_IS_SET(table, DICT_TF2_DISCARDED));
|
|
|
|
}
|
|
|
|
|
|
|
|
/********************************************************************//**
|
|
|
|
Check if it is a temporary table.
|
|
|
|
@return true if temporary table flag is set. */
|
|
|
|
UNIV_INLINE
|
|
|
|
bool
|
|
|
|
dict_table_is_temporary(
|
|
|
|
/*====================*/
|
|
|
|
const dict_table_t* table) /*!< in: table to check */
|
|
|
|
{
|
|
|
|
return(DICT_TF2_FLAG_IS_SET(table, DICT_TF2_TEMPORARY));
|
2011-12-14 14:58:22 +01:00
|
|
|
}
|
|
|
|
|
2014-05-05 18:16:30 +02:00
|
|
|
/**********************************************************************//**
|
|
|
|
Get index by first field of the index
|
|
|
|
@return index which is having first field matches
|
|
|
|
with the field present in field_index position of table */
|
|
|
|
UNIV_INLINE
|
|
|
|
dict_index_t*
|
|
|
|
dict_table_get_index_on_first_col(
|
|
|
|
/*==============================*/
|
|
|
|
const dict_table_t* table, /*!< in: table */
|
|
|
|
ulint col_index) /*!< in: position of column
|
|
|
|
in table */
|
|
|
|
{
|
|
|
|
ut_ad(col_index < table->n_cols);
|
|
|
|
|
|
|
|
dict_col_t* column = dict_table_get_nth_col(table, col_index);
|
|
|
|
|
|
|
|
for (dict_index_t* index = dict_table_get_first_index(table);
|
|
|
|
index != NULL; index = dict_table_get_next_index(index)) {
|
|
|
|
|
|
|
|
if (index->fields[0].col == column) {
|
|
|
|
return(index);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
ut_error;
|
|
|
|
return(0);
|
|
|
|
}
|
|
|
|
|
2009-09-07 10:22:53 +00:00
|
|
|
#endif /* !UNIV_HOTBACKUP */
|