mirror/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2025-01-30 18:41:56 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions
mariadb/storage/innobase/include/btr0btr.h
Marko Mäkelä de4030e4d4 MDEV-30400 Assertion height == btr_page_get_level(...) on INSERT 
This also fixes part of MDEV-29835 Partial server freeze
which is caused by violations of the latching order that was
defined in https://dev.mysql.com/worklog/task/?id=6326
(WL#6326: InnoDB: fix index->lock contention). Unless the
current thread is holding an exclusive dict_index_t::lock,
it must acquire page latches in a strict parent-to-child,
left-to-right order. Not all cases of MDEV-29835 are fixed yet.
Failure to follow the correct latching order will cause deadlocks
of threads due to lock order inversion.

As part of these changes, the BTR_MODIFY_TREE mode is modified
so that an Update latch (U a.k.a. SX) will be acquired on the
root page, and eXclusive latches (X) will be acquired on all pages
leading to the leaf page, as well as any left and right siblings
of the pages along the path. The DEBUG_SYNC test innodb.innodb_wl6326
will be removed, because at the time the DEBUG_SYNC point is hit,
the thread is actually holding several page latches that will be
blocking a concurrent SELECT statement.

We also remove double bookkeeping that was caused due to excessive
information hiding in mtr_t::m_memo. We simply let mtr_t::m_memo
store information of latched pages, and ensure that
mtr_memo_slot_t::object is never a null pointer.
The tree_blocks[] and tree_savepoints[] were redundant.

buf_page_get_low(): If innodb_change_buffering_debug=1, to avoid
a hang, do not try to evict blocks if we are holding a latch on
a modified page. The test innodb.innodb-change-buffer-recovery
will be removed, because change buffering may no longer be forced
by debug injection when the change buffer comprises multiple pages.
Remove a debug assertion that could fail when
innodb_change_buffering_debug=1 fails to evict a page.
For other cases, the assertion is redundant, because we already
checked that right after the got_block: label. The test
innodb.innodb-change-buffering-recovery will be removed, because
due to this change, we will be unable to evict the desired page.

mtr_t::lock_register(): Register a change of a page latch
on an unmodified buffer-fixed block.

mtr_t::x_latch_at_savepoint(), mtr_t::sx_latch_at_savepoint():
Replaced by the use of mtr_t::upgrade_buffer_fix(), which now
also handles RW_S_LATCH.

mtr_t::set_modified(): For temporary tables, invoke
buf_page_t::set_modified() here and not in mtr_t::commit().
We will never set the MTR_MEMO_MODIFY flag on other than
persistent data pages, nor set mtr_t::m_modifications when
temporary data pages are modified.

mtr_t::commit(): Only invoke the buf_flush_note_modification() loop
if persistent data pages were modified.

mtr_t::get_already_latched(): Look up a latched page in mtr_t::m_memo.
This avoids many redundant entries in mtr_t::m_memo, as well as
redundant calls to buf_page_get_gen() for blocks that had already
been looked up in a mini-transaction.

btr_get_latched_root(): Return a pointer to an already latched root page.
This replaces btr_root_block_get() in cases where the mini-transaction
has already latched the root page.

btr_page_get_parent(): Fetch a parent page that was already latched
in BTR_MODIFY_TREE, by invoking mtr_t::get_already_latched().
If needed, upgrade the root page U latch to X.
This avoids bloating mtr_t::m_memo as well as performing redundant
buf_pool.page_hash lookups. For non-QUICK CHECK TABLE as well as for
B-tree defragmentation, we will invoke btr_cur_search_to_nth_level().

btr_cur_search_to_nth_level(): This will only be used for non-leaf
(level>0) B-tree searches that were formerly named BTR_CONT_SEARCH_TREE
or BTR_CONT_MODIFY_TREE. In MDEV-29835, this function could be
removed altogether, or retained for the case of
CHECK TABLE without QUICK.

btr_cur_t::left_block: Remove. btr_pcur_move_backward_from_page()
can retrieve the left sibling from the end of mtr_t::m_memo.

btr_cur_t::open_leaf(): Some clean-up.

btr_cur_t::search_leaf(): Replaces btr_cur_search_to_nth_level()
for searches to level=0 (the leaf level). We will never release
parent page latches before acquiring leaf page latches. If we need to
temporarily release the level=1 page latch in the BTR_SEARCH_PREV or
BTR_MODIFY_PREV latch_mode, we will reposition the cursor on the
child node pointer so that we will land on the correct leaf page.

btr_cur_t::pessimistic_search_leaf(): Implement new BTR_MODIFY_TREE
latching logic in the case that page splits or merges will be needed.
The parent pages (and their siblings) should already be latched on
the first dive to the leaf and be present in mtr_t::m_memo; there
should be no need for BTR_CONT_MODIFY_TREE. This pre-latching almost
suffices; it must be revised in MDEV-29835 and work-arounds removed
for cases where mtr_t::get_already_latched() fails to find a block.

rtr_search_to_nth_level(): A SPATIAL INDEX version of
btr_search_to_nth_level() that can search to any level
(including the leaf level).

rtr_search_leaf(), rtr_insert_leaf(): Wrappers for
rtr_search_to_nth_level().

rtr_search(): Replaces rtr_pcur_open().

rtr_latch_leaves(): Replaces btr_cur_latch_leaves(). Note that unlike
in the B-tree code, there is no error handling in case the sibling
pages are corrupted.

rtr_cur_restore_position(): Remove an unused constant parameter.

btr_pcur_open_on_user_rec(): Remove the constant parameter
mode=PAGE_CUR_GE.

row_ins_clust_index_entry_low(): Use a new
mode=BTR_MODIFY_ROOT_AND_LEAF to gain access to the root page
when mode!=BTR_MODIFY_TREE, to write the PAGE_ROOT_AUTO_INC.

BTR_SEARCH_TREE, BTR_CONT_SEARCH_TREE: Remove.

BTR_CONT_MODIFY_TREE: Note that this is only used by
rtr_search_to_nth_level().

btr_pcur_optimistic_latch_leaves(): Replaces
btr_cur_optimistic_latch_leaves().

ibuf_delete_rec(): Acquire exclusive ibuf.index->lock in order
to avoid a deadlock with ibuf_insert_low(BTR_MODIFY_PREV).

btr_blob_log_check_t(): Acquire a U latch on the root page,
so that btr_page_alloc() in btr_store_big_rec_extern_fields()
will avoid a deadlock.

btr_store_big_rec_extern_fields(): Assert that the root page latch
is being held.

Tested by: Matthias Leich
Reviewed by: Vladislav Lesin
2023-01-24 14:09:21 +02:00

543 lines
21 KiB
C++
Raw Blame History

/*****************************************************************************
Copyright (c) 1994, 2016, Oracle and/or its affiliates. All Rights Reserved.
Copyright (c) 2012, Facebook Inc.
Copyright (c) 2014, 2023, MariaDB Corporation.
This program is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free Software
Foundation; version 2 of the License.
This program is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with
this program; if not, write to the Free Software Foundation, Inc.,
51 Franklin Street, Fifth Floor, Boston, MA 02110-1335 USA
*****************************************************************************/
/**************************************************//**
@file include/btr0btr.h
The B-tree
Created 6/2/1994 Heikki Tuuri
*******************************************************/
#pragma once
#include "dict0dict.h"
#include "data0data.h"
#include "rem0types.h"
#include "page0cur.h"
#include "btr0types.h"
#include "gis0type.h"
#define BTR_MAX_NODE_LEVEL 50 /*!< Maximum B-tree page level
(not really a hard limit).
Used in debug assertions
in btr_page_set_level and
btr_page_get_level */
/** Maximum record size which can be stored on a page, without using the
special big record storage structure */
#define BTR_PAGE_MAX_REC_SIZE (srv_page_size / 2 - 200)
/** @brief Maximum depth of a B-tree in InnoDB.
Note that this isn't a maximum as such; none of the tree operations
avoid producing trees bigger than this. It is instead a "max depth
that other code must work with", useful for e.g. fixed-size arrays
that must store some information about each level in a tree. In other
words: if a B-tree with bigger depth than this is encountered, it is
not acceptable for it to lead to mysterious memory corruption, but it
is acceptable for the program to die with a clear assert failure. */
#define BTR_MAX_LEVELS 100
#define BTR_LATCH_MODE_WITHOUT_FLAGS(latch_mode) \
btr_latch_mode((latch_mode) & ~(BTR_INSERT \
| BTR_DELETE_MARK \
| BTR_RTREE_UNDO_INS \
| BTR_RTREE_DELETE_MARK \
| BTR_DELETE \
| BTR_IGNORE_SEC_UNIQUE \
| BTR_ALREADY_S_LATCHED \
| BTR_LATCH_FOR_INSERT \
| BTR_LATCH_FOR_DELETE))
#define BTR_LATCH_MODE_WITHOUT_INTENTION(latch_mode) \
btr_latch_mode((latch_mode) \
& ~(BTR_LATCH_FOR_INSERT | BTR_LATCH_FOR_DELETE))
/**************************************************************//**
Checks and adjusts the root node of a tree during IMPORT TABLESPACE.
@return error code, or DB_SUCCESS */
dberr_t
btr_root_adjust_on_import(
/*======================*/
const dict_index_t* index) /*!< in: index tree */
MY_ATTRIBUTE((warn_unused_result));
/** Report a decryption failure. */
ATTRIBUTE_COLD void btr_decryption_failed(const dict_index_t &index);
/** Get an index page and declare its latching order level.
@param[in] index index tree
@param[in] page page number
@param[in] mode latch mode
@param[in] merge whether change buffer merge should be attempted
@param[in,out] mtr mini-transaction
@param[out] err error code
@return block */
buf_block_t *btr_block_get(const dict_index_t &index,
uint32_t page, ulint mode, bool merge,
mtr_t *mtr, dberr_t *err= nullptr);
/**************************************************************//**
Gets the index id field of a page.
@return index id */
UNIV_INLINE
index_id_t
btr_page_get_index_id(
/*==================*/
const page_t* page) /*!< in: index page */
MY_ATTRIBUTE((warn_unused_result));
/** Read the B-tree or R-tree PAGE_LEVEL.
@param page B-tree or R-tree page
@return number of child page links to reach the leaf level
@retval 0 for leaf pages */
inline uint16_t btr_page_get_level(const page_t *page)
{
uint16_t level= mach_read_from_2(my_assume_aligned<2>
(PAGE_HEADER + PAGE_LEVEL + page));
ut_ad(level <= BTR_MAX_NODE_LEVEL);
return level;
} MY_ATTRIBUTE((warn_unused_result))
/** Read FIL_PAGE_NEXT.
@param page buffer pool page
@return previous page number */
inline uint32_t btr_page_get_next(const page_t* page)
{
return mach_read_from_4(my_assume_aligned<4>(page + FIL_PAGE_NEXT));
}
/** Read FIL_PAGE_PREV.
@param page buffer pool page
@return previous page number */
inline uint32_t btr_page_get_prev(const page_t* page)
{
return mach_read_from_4(my_assume_aligned<4>(page + FIL_PAGE_PREV));
}
/**************************************************************//**
Gets the child node file address in a node pointer.
NOTE: the offsets array must contain all offsets for the record since
we read the last field according to offsets and assume that it contains
the child page number. In other words offsets must have been retrieved
with rec_get_offsets(n_fields=ULINT_UNDEFINED).
@return child node address */
UNIV_INLINE
uint32_t
btr_node_ptr_get_child_page_no(
/*===========================*/
const rec_t* rec, /*!< in: node pointer record */
const rec_offs* offsets)/*!< in: array returned by rec_get_offsets() */
MY_ATTRIBUTE((warn_unused_result));
/** Create the root node for a new index tree.
@param[in] type type of the index
@param[in,out] space tablespace where created
@param[in] index_id index id
@param[in] index index, or NULL to create a system table
@param[in,out] mtr mini-transaction
@param[out] err error code
@return page number of the created root
@retval FIL_NULL if did not succeed */
uint32_t
btr_create(
ulint type,
fil_space_t* space,
index_id_t index_id,
dict_index_t* index,
mtr_t* mtr,
dberr_t* err)
MY_ATTRIBUTE((nonnull(2,5,6), warn_unused_result));
/** Free a persistent index tree if it exists.
@param[in,out] space tablespce
@param[in] page root page number
@param[in] index_id PAGE_INDEX_ID contents
@param[in,out] mtr mini-transaction */
void btr_free_if_exists(fil_space_t *space, uint32_t page,
index_id_t index_id, mtr_t *mtr);
/** Drop a temporary table
@param table temporary table */
void btr_drop_temporary_table(const dict_table_t &table);
/** Read the last used AUTO_INCREMENT value from PAGE_ROOT_AUTO_INC.
@param[in,out] index clustered index
@return the last used AUTO_INCREMENT value
@retval 0 on error or if no AUTO_INCREMENT value was used yet */
ib_uint64_t
btr_read_autoinc(dict_index_t* index)
MY_ATTRIBUTE((nonnull, warn_unused_result));
/** Read the last used AUTO_INCREMENT value from PAGE_ROOT_AUTO_INC,
or fall back to MAX(auto_increment_column).
@param[in] table table containing an AUTO_INCREMENT column
@param[in] col_no index of the AUTO_INCREMENT column
@return the AUTO_INCREMENT value
@retval 0 on error or if no AUTO_INCREMENT value was used yet */
ib_uint64_t
btr_read_autoinc_with_fallback(const dict_table_t* table, unsigned col_no)
MY_ATTRIBUTE((nonnull, warn_unused_result));
/** Write the next available AUTO_INCREMENT value to PAGE_ROOT_AUTO_INC.
@param[in,out] index clustered index
@param[in] autoinc the AUTO_INCREMENT value
@param[in] reset whether to reset the AUTO_INCREMENT
to a possibly smaller value than currently
exists in the page */
void
btr_write_autoinc(dict_index_t* index, ib_uint64_t autoinc, bool reset = false)
MY_ATTRIBUTE((nonnull));
/** Write instant ALTER TABLE metadata to a root page.
@param[in,out] root clustered index root page
@param[in] index clustered index with instant ALTER TABLE
@param[in,out] mtr mini-transaction */
void btr_set_instant(buf_block_t* root, const dict_index_t& index, mtr_t* mtr);
ATTRIBUTE_COLD __attribute__((nonnull))
/** Reset the table to the canonical format on ROLLBACK of instant ALTER TABLE.
@param[in] index clustered index with instant ALTER TABLE
@param[in] all whether to reset FIL_PAGE_TYPE as well
@param[in,out] mtr mini-transaction */
void btr_reset_instant(const dict_index_t &index, bool all, mtr_t *mtr);
/*************************************************************//**
Makes tree one level higher by splitting the root, and inserts
the tuple. It is assumed that mtr contains an x-latch on the tree.
NOTE that the operation of this function must always succeed,
we cannot reverse it: therefore enough free disk space must be
guaranteed to be available before this function is called.
@return inserted record */
rec_t*
btr_root_raise_and_insert(
/*======================*/
ulint flags, /*!< in: undo logging and locking flags */
btr_cur_t* cursor, /*!< in: cursor at which to insert: must be
on the root page; when the function returns,
the cursor is positioned on the predecessor
of the inserted record */
rec_offs** offsets,/*!< out: offsets on inserted record */
mem_heap_t** heap, /*!< in/out: pointer to memory heap
that can be emptied, or NULL */
const dtuple_t* tuple, /*!< in: tuple to insert */
ulint n_ext, /*!< in: number of externally stored columns */
mtr_t* mtr, /*!< in: mtr */
dberr_t* err) /*!< out: error code */
MY_ATTRIBUTE((nonnull, warn_unused_result));
/*************************************************************//**
Reorganizes an index page.
IMPORTANT: On success, the caller will have to update IBUF_BITMAP_FREE
if this is a compressed leaf page in a secondary index. This has to
be done either within the same mini-transaction, or by invoking
ibuf_reset_free_bits() before mtr_commit(). On uncompressed pages,
IBUF_BITMAP_FREE is unaffected by reorganization.
@param cursor page cursor
@param mtr mini-transaction
@return error code
@retval DB_FAIL if reorganizing a ROW_FORMAT=COMPRESSED page failed */
dberr_t btr_page_reorganize(page_cur_t *cursor, mtr_t *mtr)
MY_ATTRIBUTE((nonnull, warn_unused_result));
/** Decide if the page should be split at the convergence point of inserts
converging to the left.
@param[in] cursor insert position
@return the first record to be moved to the right half page
@retval NULL if no split is recommended */
rec_t* btr_page_get_split_rec_to_left(const btr_cur_t* cursor);
/** Decide if the page should be split at the convergence point of inserts
converging to the right.
@param[in] cursor insert position
@param[out] split_rec if split recommended, the first record
on the right half page, or
NULL if the to-be-inserted record
should be first
@return whether split is recommended */
bool
btr_page_get_split_rec_to_right(const btr_cur_t* cursor, rec_t** split_rec);
/*************************************************************//**
Splits an index page to halves and inserts the tuple. It is assumed
that mtr holds an x-latch to the index tree. NOTE: the tree x-latch is
released within this function! NOTE that the operation of this
function must always succeed, we cannot reverse it: therefore enough
free disk space (2 pages) must be guaranteed to be available before
this function is called.
@return inserted record */
rec_t*
btr_page_split_and_insert(
/*======================*/
ulint flags, /*!< in: undo logging and locking flags */
btr_cur_t* cursor, /*!< in: cursor at which to insert; when the
function returns, the cursor is positioned
on the predecessor of the inserted record */
rec_offs** offsets,/*!< out: offsets on inserted record */
mem_heap_t** heap, /*!< in/out: pointer to memory heap
that can be emptied, or NULL */
const dtuple_t* tuple, /*!< in: tuple to insert */
ulint n_ext, /*!< in: number of externally stored columns */
mtr_t* mtr, /*!< in: mtr */
dberr_t* err) /*!< out: error code */
MY_ATTRIBUTE((nonnull, warn_unused_result));
/*******************************************************//**
Inserts a data tuple to a tree on a non-leaf level. It is assumed
that mtr holds an x-latch on the tree. */
dberr_t
btr_insert_on_non_leaf_level(
ulint flags, /*!< in: undo logging and locking flags */
dict_index_t* index, /*!< in: index */
ulint level, /*!< in: level, must be > 0 */
dtuple_t* tuple, /*!< in: the record to be inserted */
mtr_t* mtr) /*!< in: mtr */
MY_ATTRIBUTE((nonnull, warn_unused_result));
/** Set a child page pointer record as the predefined minimum record.
@tparam has_prev whether the page is supposed to have a left sibling
@param[in,out] rec leftmost record on a leftmost non-leaf page
@param[in,out] block buffer pool block
@param[in,out] mtr mini-transaction */
template<bool has_prev= false>
inline void btr_set_min_rec_mark(rec_t *rec, const buf_block_t &block,
mtr_t *mtr)
{
ut_ad(block.page.frame == page_align(rec));
ut_ad(!page_is_leaf(block.page.frame));
ut_ad(has_prev == page_has_prev(block.page.frame));
rec-= page_rec_is_comp(rec) ? REC_NEW_INFO_BITS : REC_OLD_INFO_BITS;
if (block.page.zip.data)
/* This flag is computed from other contents on a ROW_FORMAT=COMPRESSED
page. We are not modifying the compressed page frame at all. */
*rec|= REC_INFO_MIN_REC_FLAG;
else
mtr->write<1>(block, rec, *rec | REC_INFO_MIN_REC_FLAG);
}
/** Seek to the parent page of a B-tree page.
@param[in,out] mtr mini-transaction
@param[in,out] cursor cursor pointing to the x-latched parent page
@return whether the cursor was successfully positioned */
bool btr_page_get_father(mtr_t* mtr, btr_cur_t* cursor)
MY_ATTRIBUTE((nonnull,warn_unused_result));
#ifdef UNIV_DEBUG
/************************************************************//**
Checks that the node pointer to a page is appropriate.
@return TRUE */
ibool
btr_check_node_ptr(
/*===============*/
dict_index_t* index, /*!< in: index tree */
buf_block_t* block, /*!< in: index page */
mtr_t* mtr) /*!< in: mtr */
MY_ATTRIBUTE((warn_unused_result));
#endif /* UNIV_DEBUG */
/*************************************************************//**
Tries to merge the page first to the left immediate brother if such a
brother exists, and the node pointers to the current page and to the
brother reside on the same page. If the left brother does not satisfy these
conditions, looks at the right brother. If the page is the only one on that
level lifts the records of the page to the father page, thus reducing the
tree height. It is assumed that mtr holds an x-latch on the tree and on the
page. If cursor is on the leaf level, mtr must also hold x-latches to
the brothers, if they exist.
@return error code
@retval DB_FAIL if the tree could not be merged */
dberr_t
btr_compress(
/*=========*/
btr_cur_t* cursor, /*!< in/out: cursor on the page to merge
or lift; the page must not be empty:
when deleting records, use btr_discard_page()
if the page would become empty */
bool adjust, /*!< in: whether the cursor position should be
adjusted even when compression occurs */
mtr_t* mtr) /*!< in/out: mini-transaction */
MY_ATTRIBUTE((nonnull, warn_unused_result));
/*************************************************************//**
Discards a page from a B-tree. This is used to remove the last record from
a B-tree page: the whole page must be removed at the same time. This cannot
be used for the root page, which is allowed to be empty. */
dberr_t
btr_discard_page(
/*=============*/
btr_cur_t* cursor, /*!< in: cursor on the page to discard: not on
the root page */
mtr_t* mtr); /*!< in: mtr */
/**************************************************************//**
Allocates a new file page to be used in an index tree. NOTE: we assume
that the caller has made the reservation for free extents!
@retval NULL if no page could be allocated */
buf_block_t*
btr_page_alloc(
/*===========*/
dict_index_t* index, /*!< in: index tree */
uint32_t hint_page_no, /*!< in: hint of a good page */
byte file_direction, /*!< in: direction where a possible
page split is made */
ulint level, /*!< in: level where the page is placed
in the tree */
mtr_t* mtr, /*!< in/out: mini-transaction
for the allocation */
mtr_t* init_mtr, /*!< in/out: mini-transaction
for x-latching and initializing
the page */
dberr_t* err) /*!< out: error code */
MY_ATTRIBUTE((warn_unused_result));
/** Empty an index page (possibly the root page). @see btr_page_create().
@param[in,out] block page to be emptied
@param[in,out] page_zip compressed page frame, or NULL
@param[in] index index of the page
@param[in] level B-tree level of the page (0=leaf)
@param[in,out] mtr mini-transaction */
void
btr_page_empty(
buf_block_t* block,
page_zip_des_t* page_zip,
dict_index_t* index,
ulint level,
mtr_t* mtr)
MY_ATTRIBUTE((nonnull(1, 3, 5)));
/**************************************************************//**
Creates a new index page (not the root, and also not
used in page reorganization). @see btr_page_empty(). */
void
btr_page_create(
/*============*/
buf_block_t* block, /*!< in/out: page to be created */
page_zip_des_t* page_zip,/*!< in/out: compressed page, or NULL */
dict_index_t* index, /*!< in: index */
ulint level, /*!< in: the B-tree level of the page */
mtr_t* mtr); /*!< in: mtr */
/** Free an index page.
@param[in,out] index index tree
@param[in,out] block block to be freed
@param[in,out] mtr mini-transaction
@param[in] blob whether this is freeing a BLOB page
@param[in] latched whether index->table->space->x_lock() was called */
MY_ATTRIBUTE((nonnull))
dberr_t btr_page_free(dict_index_t *index, buf_block_t *block, mtr_t *mtr,
bool blob= false, bool space_latched= false);
/**************************************************************//**
Gets the root node of a tree and x- or s-latches it.
@return root page, x- or s-latched */
buf_block_t*
btr_root_block_get(
/*===============*/
const dict_index_t* index, /*!< in: index tree */
rw_lock_type_t mode, /*!< in: either RW_S_LATCH
or RW_X_LATCH */
mtr_t* mtr, /*!< in: mtr */
dberr_t* err); /*!< out: error code */
/*************************************************************//**
Reorganizes an index page.
IMPORTANT: On success, the caller will have to update IBUF_BITMAP_FREE
if this is a compressed leaf page in a secondary index. This has to
be done either within the same mini-transaction, or by invoking
ibuf_reset_free_bits() before mtr_commit(). On uncompressed pages,
IBUF_BITMAP_FREE is unaffected by reorganization.
@return error code
@retval DB_FAIL if reorganizing a ROW_FORMAT=COMPRESSED page failed */
dberr_t btr_page_reorganize_block(
ulint z_level,/*!< in: compression level to be used
if dealing with compressed page */
buf_block_t* block, /*!< in/out: B-tree page */
dict_index_t* index, /*!< in: the index tree of the page */
mtr_t* mtr) /*!< in/out: mini-transaction */
__attribute__((nonnull, warn_unused_result));
#ifdef UNIV_BTR_PRINT
/*************************************************************//**
Prints size info of a B-tree. */
void
btr_print_size(
/*===========*/
dict_index_t* index) /*!< in: index tree */
MY_ATTRIBUTE((nonnull));
/**************************************************************//**
Prints directories and other info of all nodes in the index. */
void
btr_print_index(
/*============*/
dict_index_t* index, /*!< in: index */
ulint width) /*!< in: print this many entries from start
and end */
MY_ATTRIBUTE((nonnull));
#endif /* UNIV_BTR_PRINT */
/************************************************************//**
Checks the size and number of fields in a record based on the definition of
the index.
@return TRUE if ok */
ibool
btr_index_rec_validate(
/*===================*/
const rec_t* rec, /*!< in: index record */
const dict_index_t* index, /*!< in: index */
ibool dump_on_error) /*!< in: TRUE if the function
should print hex dump of record
and page on error */
MY_ATTRIBUTE((warn_unused_result));
/**************************************************************//**
Checks the consistency of an index tree.
@return DB_SUCCESS if ok, error code if not */
dberr_t
btr_validate_index(
/*===============*/
dict_index_t* index, /*!< in: index */
const trx_t* trx) /*!< in: transaction or 0 */
MY_ATTRIBUTE((warn_unused_result));
/** Remove a page from the level list of pages.
@param[in] block page to remove
@param[in] index index tree
@param[in,out] mtr mini-transaction */
dberr_t btr_level_list_remove(const buf_block_t& block,
const dict_index_t& index, mtr_t* mtr)
MY_ATTRIBUTE((warn_unused_result));
/*************************************************************//**
If page is the only on its level, this function moves its records to the
father page, thus reducing the tree height.
@return father block */
buf_block_t*
btr_lift_page_up(
dict_index_t* index, /*!< in: index tree */
buf_block_t* block, /*!< in: page which is the only on its level;
must not be empty: use
btr_discard_only_page_on_level if the last
record from the page should be removed */
mtr_t* mtr, /*!< in/out: mini-transaction */
dberr_t* err) /*!< out: error code */
__attribute__((nonnull));
#define BTR_N_LEAF_PAGES 1
#define BTR_TOTAL_SIZE 2
#include "btr0btr.inl"
/****************************************************************
Global variable controlling if scrubbing should be performed */
extern my_bool srv_immediate_scrub_data_uncompressed;
Reference in a new issue View git blame Copy permalink