mariadb/include/ibuf0ibuf.h
marko 372c4132fd branches/innodb+: Merge revisions 4006:4060 from branches/zip:
------------------------------------------------------------------------
  r4008 | vasil | 2009-01-20 17:01:08 +0200 (Tue, 20 Jan 2009) | 4 lines

  branches/zip:

  Add ChangeLog entries for the bugfixes in r4004 and r4005.

  ------------------------------------------------------------------------
  r4027 | marko | 2009-01-23 11:04:49 +0200 (Fri, 23 Jan 2009) | 1 line

  branches/zip: Remove some redundant #include statements.
  ------------------------------------------------------------------------
  r4028 | marko | 2009-01-23 11:26:10 +0200 (Fri, 23 Jan 2009) | 13 lines

  branches/zip: Enclose some more unused code in #ifdef UNIV_LOG_ARCHIVE.
  This will help trim the dependencies of InnoDB Hot Backup.

  recv_recovery_from_checkpoint_start(): Rename to
  recv_recovery_from_checkpoint_start_func(), and remove the two first
  parameters unless UNIV_LOG_ARCHIVE is defined.  Define and use
  the auxiliary macros TYPE_CHECKPOINT and LIMIT_LSN in the function.

  struct recv_sys_struct: Remove archive_group unless UNIV_LOG_ARCHIVE
  is defined.

  Do not define LOG_ARCHIVE unless UNIV_LOG_ARCHIVE is defined.
  ------------------------------------------------------------------------
  r4029 | marko | 2009-01-23 14:07:38 +0200 (Fri, 23 Jan 2009) | 1 line

  branches/zip: Remove some redundant #include directives.
  ------------------------------------------------------------------------
  r4030 | marko | 2009-01-23 15:31:36 +0200 (Fri, 23 Jan 2009) | 2 lines

  branches/zip: buf_page_get_gen(): Remove the unused mode BUF_GET_NOWAIT.
  This was noticed while investigating Issue #160.
  ------------------------------------------------------------------------
  r4033 | marko | 2009-01-23 15:49:04 +0200 (Fri, 23 Jan 2009) | 16 lines

  branches/zip: Merge revisions 4005:4032 from branches/5.1:

    ------------------------------------------------------------------------
    r4032 | marko | 2009-01-23 15:43:51 +0200 (Fri, 23 Jan 2009) | 10 lines

    branches/5.1: Merge r4031 from branches/5.0:

    btr_search_drop_page_hash_when_freed(): Check if buf_page_get_gen()
    returns NULL.  The page may have been evicted from the buffer pool
    between buf_page_peek_if_search_hashed() and buf_page_get_gen(),
    because the buffer pool mutex will be released between these two calls.
    (Bug #42279, Issue #160)

    rb://82 approved by Heikki Tuuri
    ------------------------------------------------------------------------
  ------------------------------------------------------------------------
  r4034 | marko | 2009-01-26 16:16:39 +0200 (Mon, 26 Jan 2009) | 2 lines

  branches/zip: buf_page_get_gen(): Fix a "possibly uninitialized" warning
  that was introduced in r4030.
  ------------------------------------------------------------------------
  r4036 | marko | 2009-01-26 22:33:20 +0200 (Mon, 26 Jan 2009) | 22 lines

  branches/zip: In r988, the function buf_block_align() was enclosed
  within UNIV_DEBUG. The two remaining callers in non-debug builds,
  btr_search_guess_on_hash() and btr_search_validate(), were rewritten
  to call buf_page_hash_get().

  To implement support for a resizeable buffer pool, the function
  buf_block_align() had been rewritten to perform a page hash lookup in
  the buffer pool. The caller was also made responsible for holding the
  buffer pool mutex.

  Because the page hash lookup is expensive and it has to be done while
  holding the buffer pool mutex, implement buf_block_align() by pointer
  arithmetics again, and make btr_search_guess_on_hash() call it. Note
  that this will have to be adjusted if the interface to the resizeable
  buffer pool is actually implemented.

  rb://83 approved by Heikki Tuuri, to address Issue #161.

  As a deviation from the approved patch, this patch also makes
  btr_search_validate() (invoked by CHECK TABLE) check that
  buf_pool->page_hash is consistent with buf_block_align().
  ------------------------------------------------------------------------
  r4039 | vasil | 2009-01-27 08:04:17 +0200 (Tue, 27 Jan 2009) | 5 lines

  branches/zip:

  Adjust the paths in innodb_file_per_table.diff with a recent rename of
  the test/result files in the MySQL tree.
  ------------------------------------------------------------------------
  r4042 | marko | 2009-01-27 10:05:24 +0200 (Tue, 27 Jan 2009) | 2 lines

  branches/zip: buf_LRU_invalidate_tablespace(): Fix a race condition:
  read zip_size while still holding block_mutex.
  ------------------------------------------------------------------------
  r4045 | marko | 2009-01-28 00:31:17 +0200 (Wed, 28 Jan 2009) | 8 lines

  branches/zip: btr_search_validate(): Fix a bogus UNIV_DEBUG
  assertion failure that was accidentally introduced in r4036.
  Instead of calling buf_block_get_frame(), which asserts that the
  block must be buffer-fixed, access block->frame directly.  That
  is safe, because changes of block->page.state are protected by
  the buffer pool mutex, which we are holding.

  This bug was reported by Michael.
  ------------------------------------------------------------------------
  r4046 | marko | 2009-01-28 00:33:20 +0200 (Wed, 28 Jan 2009) | 2 lines

  branches/zip: Revert the change to univ.i that was accidentally
  committed in r4045.
  ------------------------------------------------------------------------
  r4047 | marko | 2009-01-28 00:46:13 +0200 (Wed, 28 Jan 2009) | 6 lines

  branches/zip: btr_search_validate(): Fix an assertion failure that was
  introduced in r4036.
  Do not call buf_block_get_space(), buf_block_get_page_no()
  unless the block state is BUF_BLOCK_FILE_PAGE.

  This bug was reported by Michael.
  ------------------------------------------------------------------------
  r4050 | vasil | 2009-01-28 08:21:44 +0200 (Wed, 28 Jan 2009) | 5 lines

  branches/zip:

  Adjust the paths in innodb_lock_wait_timeout.diff with a recent rename of
  the test/result files in the MySQL tree.
  ------------------------------------------------------------------------
  r4051 | marko | 2009-01-28 14:35:49 +0200 (Wed, 28 Jan 2009) | 1 line

  branches/zip: trx0sys.ic: Remove unnecessary #include <data0type.h>.
  ------------------------------------------------------------------------
  r4052 | marko | 2009-01-28 15:21:45 +0200 (Wed, 28 Jan 2009) | 5 lines

  branches/zip: Enclose some functions inside #ifdef UNIV_HOTBACKUP:

  ut_sprintf_timestamp_without_extra_chars(), ut_get_year_month_day(),
  log_reset_first_header_and_checkpoint(): These functions are only used
  in InnoDB Hot Backup.
  ------------------------------------------------------------------------
  r4056 | calvin | 2009-01-29 03:06:41 +0200 (Thu, 29 Jan 2009) | 33 lines

  branches/zip: Merge revisions 4032:4035 from branches/5.1

  All InnoDB related tests passed on Windows, except
  known failure in partition_innodb_semi_consistent.

  The inadvertent change to btr0sea.c in this commit is reverted in r4060.

    ------------------------------------------------------------------------
    r4035 | vasil | 2009-01-26 09:26:25 -0600 (Mon, 26 Jan 2009) | 23 lines

    branches/5.1:

    Merge a change from MySQL:

      ------------------------------------------------------------
      revno: 2646.161.4
      committer: Tatiana A. Nurnberg <azundris@mysql.com>
      branch nick: 51-31177v2
      timestamp: Mon 2009-01-12 06:32:49 +0100
      message:
        Bug#31177: Server variables can't be set to their current values

        Bounds-checks and blocksize corrections were applied to user-input,
        but constants in the server were trusted implicitly. If these values
        did not actually meet the requirements, the user could not set change
        a variable, then set it back to the (wonky) factory default or maximum
        by explicitly specifying it (SET <var>=<value> vs SET <var>=DEFAULT).

        Now checks also apply to the server's presets. Wonky values and maxima
        get corrected at startup. Consequently all non-offsetted values the user
        sees are valid, and users can set the variable to that exact value if
        they so desire.
  ------------------------------------------------------------------------
  r4060 | marko | 2009-01-29 09:39:04 +0200 (Thu, 29 Jan 2009) | 1 line

  branches/zip: btr0sea.c: Revert the inadvertent change made in r4056.
  ------------------------------------------------------------------------
2009-01-29 09:47:07 +00:00

375 lines
14 KiB
C

/******************************************************
Insert buffer
(c) 1997 Innobase Oy
Created 7/19/1997 Heikki Tuuri
*******************************************************/
#ifndef ibuf0ibuf_h
#define ibuf0ibuf_h
#include "univ.i"
#include "dict0mem.h"
#include "mtr0mtr.h"
#include "que0types.h"
#include "ibuf0types.h"
#include "fsp0fsp.h"
/* Possible operations buffered in the insert/whatever buffer. See
ibuf_insert(). DO NOT CHANGE THE VALUES OF THESE, THEY ARE STORED ON DISK. */
typedef enum {
IBUF_OP_INSERT = 0,
IBUF_OP_DELETE_MARK = 1,
IBUF_OP_DELETE = 2,
/* Number of different operation types. */
IBUF_OP_COUNT = 3,
} ibuf_op_t;
/** Combinations of operations that can be buffered. */
typedef enum {
IBUF_USE_NONE = 0,
IBUF_USE_INSERT, /* insert */
IBUF_USE_DELETE_MARK, /* delete */
IBUF_USE_INSERT_DELETE_MARK, /* insert+delete */
IBUF_USE_DELETE, /* delete+purge */
IBUF_USE_ALL /* insert+delete+purge */
} ibuf_use_t;
/** Operations that can currently be buffered. */
extern ibuf_use_t ibuf_use;
/** The insert buffer control structure */
extern ibuf_t* ibuf;
/* The purpose of the insert buffer is to reduce random disk access.
When we wish to insert a record into a non-unique secondary index and
the B-tree leaf page where the record belongs to is not in the buffer
pool, we insert the record into the insert buffer B-tree, indexed by
(space_id, page_no). When the page is eventually read into the buffer
pool, we look up the insert buffer B-tree for any modifications to the
page, and apply these upon the completion of the read operation. This
is called the insert buffer merge. */
/* The insert buffer merge must always succeed. To guarantee this,
the insert buffer subsystem keeps track of the free space in pages for
which it can buffer operations. Two bits per page in the insert
buffer bitmap indicate the available space in coarse increments. The
free bits in the insert buffer bitmap must never exceed the free space
on a page. It is safe to decrement or reset the bits in the bitmap in
a mini-transaction that is committed before the mini-transaction that
affects the free space. It is unsafe to increment the bits in a
separately committed mini-transaction, because in crash recovery, the
free bits could momentarily be set too high. */
/**********************************************************************
Creates the insert buffer data structure at a database startup. */
UNIV_INTERN
void
ibuf_init_at_db_start(void);
/*=======================*/
/*************************************************************************
Reads the biggest tablespace id from the high end of the insert buffer
tree and updates the counter in fil_system. */
UNIV_INTERN
void
ibuf_update_max_tablespace_id(void);
/*===============================*/
/*************************************************************************
Initializes an ibuf bitmap page. */
UNIV_INTERN
void
ibuf_bitmap_page_init(
/*==================*/
buf_block_t* block, /* in: bitmap page */
mtr_t* mtr); /* in: mtr */
/****************************************************************************
Resets the free bits of the page in the ibuf bitmap. This is done in a
separate mini-transaction, hence this operation does not restrict
further work to only ibuf bitmap operations, which would result if the
latch to the bitmap page were kept. NOTE: The free bits in the insert
buffer bitmap must never exceed the free space on a page. It is safe
to decrement or reset the bits in the bitmap in a mini-transaction
that is committed before the mini-transaction that affects the free
space. */
UNIV_INTERN
void
ibuf_reset_free_bits(
/*=================*/
buf_block_t* block); /* in: index page; free bits are set to 0
if the index is a non-clustered
non-unique, and page level is 0 */
/****************************************************************************
Updates the free bits of an uncompressed page in the ibuf bitmap if
there is not enough free on the page any more. This is done in a
separate mini-transaction, hence this operation does not restrict
further work to only ibuf bitmap operations, which would result if the
latch to the bitmap page were kept. NOTE: The free bits in the insert
buffer bitmap must never exceed the free space on a page. It is
unsafe to increment the bits in a separately committed
mini-transaction, because in crash recovery, the free bits could
momentarily be set too high. It is only safe to use this function for
decrementing the free bits. Should more free space become available,
we must not update the free bits here, because that would break crash
recovery. */
UNIV_INLINE
void
ibuf_update_free_bits_if_full(
/*==========================*/
buf_block_t* block, /* in: index page to which we have added new
records; the free bits are updated if the
index is non-clustered and non-unique and
the page level is 0, and the page becomes
fuller */
ulint max_ins_size,/* in: value of maximum insert size with
reorganize before the latest operation
performed to the page */
ulint increase);/* in: upper limit for the additional space
used in the latest operation, if known, or
ULINT_UNDEFINED */
/**************************************************************************
Updates the free bits for an uncompressed page to reflect the present
state. Does this in the mtr given, which means that the latching
order rules virtually prevent any further operations for this OS
thread until mtr is committed. NOTE: The free bits in the insert
buffer bitmap must never exceed the free space on a page. It is safe
to set the free bits in the same mini-transaction that updated the
page. */
UNIV_INTERN
void
ibuf_update_free_bits_low(
/*======================*/
const buf_block_t* block, /* in: index page */
ulint max_ins_size, /* in: value of
maximum insert size
with reorganize before
the latest operation
performed to the page */
mtr_t* mtr); /* in/out: mtr */
/**************************************************************************
Updates the free bits for a compressed page to reflect the present
state. Does this in the mtr given, which means that the latching
order rules virtually prevent any further operations for this OS
thread until mtr is committed. NOTE: The free bits in the insert
buffer bitmap must never exceed the free space on a page. It is safe
to set the free bits in the same mini-transaction that updated the
page. */
UNIV_INTERN
void
ibuf_update_free_bits_zip(
/*======================*/
buf_block_t* block, /* in/out: index page */
mtr_t* mtr); /* in/out: mtr */
/**************************************************************************
Updates the free bits for the two pages to reflect the present state.
Does this in the mtr given, which means that the latching order rules
virtually prevent any further operations until mtr is committed.
NOTE: The free bits in the insert buffer bitmap must never exceed the
free space on a page. It is safe to set the free bits in the same
mini-transaction that updated the pages. */
UNIV_INTERN
void
ibuf_update_free_bits_for_two_pages_low(
/*====================================*/
ulint zip_size,/* in: compressed page size in bytes;
0 for uncompressed pages */
buf_block_t* block1, /* in: index page */
buf_block_t* block2, /* in: index page */
mtr_t* mtr); /* in: mtr */
/**************************************************************************
A basic partial test if an insert to the insert buffer could be possible and
recommended. */
UNIV_INLINE
ibool
ibuf_should_try(
/*============*/
dict_index_t* index, /* in: index where to insert */
ulint ignore_sec_unique); /* in: if != 0, we should
ignore UNIQUE constraint on
a secondary index when we
decide */
/**********************************************************************
Returns TRUE if the current OS thread is performing an insert buffer
routine. */
UNIV_INTERN
ibool
ibuf_inside(void);
/*=============*/
/* out: TRUE if inside an insert buffer routine: for instance,
a read-ahead of non-ibuf pages is then forbidden */
/***************************************************************************
Checks if a page address is an ibuf bitmap page (level 3 page) address. */
UNIV_INLINE
ibool
ibuf_bitmap_page(
/*=============*/
/* out: TRUE if a bitmap page */
ulint zip_size,/* in: compressed page size in bytes;
0 for uncompressed pages */
ulint page_no);/* in: page number */
/***************************************************************************
Checks if a page is a level 2 or 3 page in the ibuf hierarchy of pages.
Must not be called when recv_no_ibuf_operations==TRUE. */
UNIV_INTERN
ibool
ibuf_page(
/*======*/
/* out: TRUE if level 2 or level 3 page */
ulint space, /* in: space id */
ulint zip_size,/* in: compressed page size in bytes, or 0 */
ulint page_no,/* in: page number */
mtr_t* mtr); /* in: mtr which will contain an x-latch to the
bitmap page if the page is not one of the fixed
address ibuf pages, or NULL, in which case a new
transaction is created. */
/***************************************************************************
Frees excess pages from the ibuf free list. This function is called when an OS
thread calls fsp services to allocate a new file segment, or a new page to a
file segment, and the thread did not own the fsp latch before this call. */
UNIV_INTERN
void
ibuf_free_excess_pages(void);
/*========================*/
/*************************************************************************
Buffer an operation in the insert/delete buffer, instead of doing it
directly to the disk page, if this is possible. Does not do it if the index
is clustered or unique. */
UNIV_INTERN
ibool
ibuf_insert(
/*========*/
/* out: TRUE if success */
ibuf_op_t op, /* in: operation type */
const dtuple_t* entry, /* in: index entry to insert */
dict_index_t* index, /* in: index where to insert */
ulint space, /* in: space id where to insert */
ulint zip_size,/* in: compressed page size in bytes, or 0 */
ulint page_no,/* in: page number where to insert */
que_thr_t* thr); /* in: query thread */
/*************************************************************************
When an index page is read from a disk to the buffer pool, this function
applies any buffered operations to the page and deletes the entries from the
insert buffer. If the page is not read, but created in the buffer pool, this
function deletes its buffered entries from the insert buffer; there can
exist entries for such a page if the page belonged to an index which
subsequently was dropped. */
UNIV_INTERN
void
ibuf_merge_or_delete_for_page(
/*==========================*/
buf_block_t* block, /* in: if page has been read from
disk, pointer to the page x-latched,
else NULL */
ulint space, /* in: space id of the index page */
ulint page_no,/* in: page number of the index page */
ulint zip_size,/* in: compressed page size in bytes,
or 0 */
ibool update_ibuf_bitmap);/* in: normally this is set
to TRUE, but if we have deleted or are
deleting the tablespace, then we
naturally do not want to update a
non-existent bitmap page */
/*************************************************************************
Deletes all entries in the insert buffer for a given space id. This is used
in DISCARD TABLESPACE and IMPORT TABLESPACE.
NOTE: this does not update the page free bitmaps in the space. The space will
become CORRUPT when you call this function! */
UNIV_INTERN
void
ibuf_delete_for_discarded_space(
/*============================*/
ulint space); /* in: space id */
/*************************************************************************
Contracts insert buffer trees by reading pages to the buffer pool. */
UNIV_INTERN
ulint
ibuf_contract(
/*==========*/
/* out: a lower limit for the combined size in bytes
of entries which will be merged from ibuf trees to the
pages read, 0 if ibuf is empty */
ibool sync); /* in: TRUE if the caller wants to wait for the
issued read with the highest tablespace address
to complete */
/*************************************************************************
Contracts insert buffer trees by reading pages to the buffer pool. */
UNIV_INTERN
ulint
ibuf_contract_for_n_pages(
/*======================*/
/* out: a lower limit for the combined size in bytes
of entries which will be merged from ibuf trees to the
pages read, 0 if ibuf is empty */
ibool sync, /* in: TRUE if the caller wants to wait for the
issued read with the highest tablespace address
to complete */
ulint n_pages);/* in: try to read at least this many pages to
the buffer pool and merge the ibuf contents to
them */
/*************************************************************************
Parses a redo log record of an ibuf bitmap page init. */
UNIV_INTERN
byte*
ibuf_parse_bitmap_init(
/*===================*/
/* out: end of log record or NULL */
byte* ptr, /* in: buffer */
byte* end_ptr,/* in: buffer end */
buf_block_t* block, /* in: block or NULL */
mtr_t* mtr); /* in: mtr or NULL */
#ifdef UNIV_IBUF_COUNT_DEBUG
/**********************************************************************
Gets the ibuf count for a given page. */
UNIV_INTERN
ulint
ibuf_count_get(
/*===========*/
/* out: number of entries in the insert buffer
currently buffered for this page */
ulint space, /* in: space id */
ulint page_no);/* in: page number */
#endif
/**********************************************************************
Looks if the insert buffer is empty. */
UNIV_INTERN
ibool
ibuf_is_empty(void);
/*===============*/
/* out: TRUE if empty */
/**********************************************************************
Prints info of ibuf. */
UNIV_INTERN
void
ibuf_print(
/*=======*/
FILE* file); /* in: file where to print */
/********************************************************************
Read the first two bytes from a record's fourth field (counter field in new
records; something else in older records). */
UNIV_INTERN
ulint
ibuf_rec_get_counter(
/*=================*/
/* out: "counter" field,
or ULINT_UNDEFINED if for
some reason it can't be read */
const rec_t* rec); /* in: ibuf record */
#define IBUF_HEADER_PAGE_NO FSP_IBUF_HEADER_PAGE_NO
#define IBUF_TREE_ROOT_PAGE_NO FSP_IBUF_TREE_ROOT_PAGE_NO
/* The ibuf header page currently contains only the file segment header
for the file segment from which the pages for the ibuf tree are allocated */
#define IBUF_HEADER PAGE_DATA
#define IBUF_TREE_SEG_HEADER 0 /* fseg header for ibuf tree */
/* The insert buffer tree itself is always located in space 0. */
#define IBUF_SPACE_ID 0
#ifndef UNIV_NONINL
#include "ibuf0ibuf.ic"
#endif
#endif