diff options
Diffstat (limited to 'storage/innodb_plugin/include')
29 files changed, 574 insertions, 212 deletions
diff --git a/storage/innodb_plugin/include/btr0btr.h b/storage/innodb_plugin/include/btr0btr.h index 5aa02694e0e..476ad29adac 100644 --- a/storage/innodb_plugin/include/btr0btr.h +++ b/storage/innodb_plugin/include/btr0btr.h @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1994, 2010, Innobase Oy. All Rights Reserved. +Copyright (c) 1994, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -188,26 +188,45 @@ btr_block_get_func( ulint mode, /*!< in: latch mode */ const char* file, /*!< in: file name */ ulint line, /*!< in: line where called */ - mtr_t* mtr) /*!< in/out: mtr */ - __attribute__((nonnull)); +# ifdef UNIV_SYNC_DEBUG + const dict_index_t* index, /*!< in: index tree, may be NULL + if it is not an insert buffer tree */ +# endif /* UNIV_SYNC_DEBUG */ + mtr_t* mtr); /*!< in/out: mini-transaction */ +# ifdef UNIV_SYNC_DEBUG +/** Gets a buffer page and declares its latching order level. +@param space tablespace identifier +@param zip_size compressed page size in bytes or 0 for uncompressed pages +@param page_no page number +@param mode latch mode +@param index index tree, may be NULL if not the insert buffer tree +@param mtr mini-transaction handle +@return the block descriptor */ +# define btr_block_get(space,zip_size,page_no,mode,index,mtr) \ + btr_block_get_func(space,zip_size,page_no,mode, \ + __FILE__,__LINE__,index,mtr) +# else /* UNIV_SYNC_DEBUG */ /** Gets a buffer page and declares its latching order level. @param space tablespace identifier @param zip_size compressed page size in bytes or 0 for uncompressed pages @param page_no page number @param mode latch mode +@param idx index tree, may be NULL if not the insert buffer tree @param mtr mini-transaction handle @return the block descriptor */ -# define btr_block_get(space,zip_size,page_no,mode,mtr) \ +# define btr_block_get(space,zip_size,page_no,mode,idx,mtr) \ btr_block_get_func(space,zip_size,page_no,mode,__FILE__,__LINE__,mtr) +# endif /* UNIV_SYNC_DEBUG */ /** Gets a buffer page and declares its latching order level. @param space tablespace identifier @param zip_size compressed page size in bytes or 0 for uncompressed pages @param page_no page number @param mode latch mode +@param idx index tree, may be NULL if not the insert buffer tree @param mtr mini-transaction handle @return the uncompressed page frame */ -# define btr_page_get(space,zip_size,page_no,mode,mtr) \ - buf_block_get_frame(btr_block_get(space,zip_size,page_no,mode,mtr)) +# define btr_page_get(space,zip_size,page_no,mode,idx,mtr) \ + buf_block_get_frame(btr_block_get(space,zip_size,page_no,mode,idx,mtr)) #endif /* !UNIV_HOTBACKUP */ /**************************************************************//** Gets the index id field of a page. @@ -333,8 +352,7 @@ btr_free_root( ulint zip_size, /*!< in: compressed page size in bytes or 0 for uncompressed pages */ ulint root_page_no, /*!< in: root page number */ - mtr_t* mtr); /*!< in: a mini-transaction which has already - been started */ + mtr_t* mtr); /*!< in/out: mini-transaction */ /*************************************************************//** 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. @@ -470,11 +488,14 @@ UNIV_INTERN ibool btr_compress( /*=========*/ - btr_cur_t* cursor, /*!< in: cursor on the page to merge or lift; - the page must not be empty: in record delete - use btr_discard_page if the page would become - empty */ - mtr_t* mtr); /*!< in: mtr */ + 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 */ + ibool adjust, /*!< in: TRUE if should adjust the + cursor position even if compression occurs */ + mtr_t* mtr) /*!< in/out: mini-transaction */ + __attribute__((nonnull)); /*************************************************************//** 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 @@ -536,7 +557,12 @@ btr_page_alloc( page split is made */ ulint level, /*!< in: level where the page is placed in the tree */ - mtr_t* mtr); /*!< in: mtr */ + 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 */ + __attribute__((nonnull, warn_unused_result)); /**************************************************************//** Frees a file page used in an index tree. NOTE: cannot free field external storage pages because the page must contain info on its level. */ @@ -559,6 +585,33 @@ btr_page_free_low( buf_block_t* block, /*!< in: block to be freed, x-latched */ ulint level, /*!< in: page level */ mtr_t* mtr); /*!< in: mtr */ +/**************************************************************//** +Marks all MTR_MEMO_FREE_CLUST_LEAF pages nonfree or free. +For invoking btr_store_big_rec_extern_fields() after an update, +we must temporarily mark freed clustered index pages allocated, so +that off-page columns will not be allocated from them. Between the +btr_store_big_rec_extern_fields() and mtr_commit() we have to +mark the pages free again, so that no pages will be leaked. */ +UNIV_INTERN +void +btr_mark_freed_leaves( +/*==================*/ + dict_index_t* index, /*!< in/out: clustered index */ + mtr_t* mtr, /*!< in/out: mini-transaction */ + ibool nonfree)/*!< in: TRUE=mark nonfree, FALSE=mark freed */ + __attribute__((nonnull)); +#ifdef UNIV_DEBUG +/**************************************************************//** +Validates all pages marked MTR_MEMO_FREE_CLUST_LEAF. +@see btr_mark_freed_leaves() +@return TRUE */ +UNIV_INTERN +ibool +btr_freed_leaves_validate( +/*======================*/ + mtr_t* mtr) /*!< in: mini-transaction */ + __attribute__((nonnull, warn_unused_result)); +#endif /* UNIV_DEBUG */ #ifdef UNIV_BTR_PRINT /*************************************************************//** Prints size info of a B-tree. */ diff --git a/storage/innodb_plugin/include/btr0btr.ic b/storage/innodb_plugin/include/btr0btr.ic index 83eb3627abb..1837d091ccf 100644 --- a/storage/innodb_plugin/include/btr0btr.ic +++ b/storage/innodb_plugin/include/btr0btr.ic @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1994, 2010, Innobase Oy. All Rights Reserved. +Copyright (c) 1994, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -48,6 +48,10 @@ btr_block_get_func( ulint mode, /*!< in: latch mode */ const char* file, /*!< in: file name */ ulint line, /*!< in: line where called */ +#ifdef UNIV_SYNC_DEBUG + const dict_index_t* index, /*!< in: index tree, may be NULL + if it is not an insert buffer tree */ +#endif /* UNIV_SYNC_DEBUG */ mtr_t* mtr) /*!< in/out: mtr */ { buf_block_t* block; @@ -57,7 +61,9 @@ btr_block_get_func( if (mode != RW_NO_LATCH) { - buf_block_dbg_add_level(block, SYNC_TREE_NODE); + buf_block_dbg_add_level( + block, index != NULL && dict_index_is_ibuf(index) + ? SYNC_IBUF_TREE_NODE : SYNC_TREE_NODE); } return(block); diff --git a/storage/innodb_plugin/include/btr0cur.h b/storage/innodb_plugin/include/btr0cur.h index ece3621fa97..1d97c5b9452 100644 --- a/storage/innodb_plugin/include/btr0cur.h +++ b/storage/innodb_plugin/include/btr0cur.h @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1994, 2010, Innobase Oy. All Rights Reserved. +Copyright (c) 1994, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -36,6 +36,9 @@ Created 10/16/1994 Heikki Tuuri #define BTR_NO_LOCKING_FLAG 2 /* do no record lock checking */ #define BTR_KEEP_SYS_FLAG 4 /* sys fields will be found from the update vector or inserted entry */ +#define BTR_KEEP_POS_FLAG 8 /* btr_cur_pessimistic_update() + must keep cursor position when + moving columns to big_rec */ #ifndef UNIV_HOTBACKUP #include "que0types.h" @@ -309,7 +312,9 @@ btr_cur_pessimistic_update( /*=======================*/ ulint flags, /*!< in: undo logging, locking, and rollback flags */ - btr_cur_t* cursor, /*!< in: cursor on the record to update */ + btr_cur_t* cursor, /*!< in/out: cursor on the record to update; + cursor may become invalid if *big_rec == NULL + || !(flags & BTR_KEEP_POS_FLAG) */ mem_heap_t** heap, /*!< in/out: pointer to memory heap, or NULL */ big_rec_t** big_rec,/*!< out: big rec vector whose fields have to be stored externally by the caller, or NULL */ @@ -376,10 +381,13 @@ UNIV_INTERN ibool btr_cur_compress_if_useful( /*=======================*/ - btr_cur_t* cursor, /*!< in: cursor on the page to compress; + btr_cur_t* cursor, /*!< in/out: cursor on the page to compress; cursor does not stay valid if compression occurs */ - mtr_t* mtr); /*!< in: mtr */ + ibool adjust, /*!< in: TRUE if should adjust the + cursor position even if compression occurs */ + mtr_t* mtr) /*!< in/out: mini-transaction */ + __attribute__((nonnull)); /*******************************************************//** Removes the record on which the tree cursor is positioned. It is assumed that the mtr has an x-latch on the page where the cursor is positioned, @@ -522,6 +530,8 @@ btr_store_big_rec_extern_fields_func( the "external storage" flags in offsets will not correspond to rec when this function returns */ + const big_rec_t*big_rec_vec, /*!< in: vector containing fields + to be stored externally */ #ifdef UNIV_DEBUG mtr_t* local_mtr, /*!< in: mtr containing the latch to rec and to the tree */ @@ -530,9 +540,12 @@ btr_store_big_rec_extern_fields_func( ibool update_in_place,/*! in: TRUE if the record is updated in place (not delete+insert) */ #endif /* UNIV_DEBUG || UNIV_BLOB_LIGHT_DEBUG */ - const big_rec_t*big_rec_vec) /*!< in: vector containing fields - to be stored externally */ - __attribute__((nonnull)); + mtr_t* alloc_mtr) /*!< in/out: in an insert, NULL; + in an update, local_mtr for + allocating BLOB pages and + updating BLOB pointers; alloc_mtr + must not have freed any leaf pages */ + __attribute__((nonnull(1,2,3,4,5), warn_unused_result)); /** Stores the fields in big_rec_vec to the tablespace and puts pointers to them in rec. The extern flags in rec will have to be set beforehand. @@ -541,21 +554,22 @@ file segment of the index tree. @param index in: clustered index; MUST be X-latched by mtr @param b in/out: block containing rec; MUST be X-latched by mtr @param rec in/out: clustered index record -@param offsets in: rec_get_offsets(rec, index); +@param offs in: rec_get_offsets(rec, index); the "external storage" flags in offsets will not be adjusted +@param big in: vector containing fields to be stored externally @param mtr in: mini-transaction that holds x-latch on index and b @param upd in: TRUE if the record is updated in place (not delete+insert) -@param big in: vector containing fields to be stored externally +@param rmtr in/out: in updates, the mini-transaction that holds rec @return DB_SUCCESS or DB_OUT_OF_FILE_SPACE */ #ifdef UNIV_DEBUG -# define btr_store_big_rec_extern_fields(index,b,rec,offsets,mtr,upd,big) \ - btr_store_big_rec_extern_fields_func(index,b,rec,offsets,mtr,upd,big) +# define btr_store_big_rec_extern_fields(index,b,rec,offs,big,mtr,upd,rmtr) \ + btr_store_big_rec_extern_fields_func(index,b,rec,offs,big,mtr,upd,rmtr) #elif defined UNIV_BLOB_LIGHT_DEBUG -# define btr_store_big_rec_extern_fields(index,b,rec,offsets,mtr,upd,big) \ - btr_store_big_rec_extern_fields_func(index,b,rec,offsets,upd,big) +# define btr_store_big_rec_extern_fields(index,b,rec,offs,big,mtr,upd,rmtr) \ + btr_store_big_rec_extern_fields_func(index,b,rec,offs,big,upd,rmtr) #else -# define btr_store_big_rec_extern_fields(index,b,rec,offsets,mtr,upd,big) \ - btr_store_big_rec_extern_fields_func(index,b,rec,offsets,big) +# define btr_store_big_rec_extern_fields(index,b,rec,offs,big,mtr,upd,rmtr) \ + btr_store_big_rec_extern_fields_func(index,b,rec,offs,big,rmtr) #endif /*******************************************************************//** diff --git a/storage/innodb_plugin/include/btr0cur.ic b/storage/innodb_plugin/include/btr0cur.ic index 280583f6ccf..c833b3e8572 100644 --- a/storage/innodb_plugin/include/btr0cur.ic +++ b/storage/innodb_plugin/include/btr0cur.ic @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1994, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1994, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -139,7 +139,7 @@ btr_cur_compress_recommendation( btr_cur_t* cursor, /*!< in: btr cursor */ mtr_t* mtr) /*!< in: mtr */ { - page_t* page; + const page_t* page; ut_ad(mtr_memo_contains(mtr, btr_cur_get_block(cursor), MTR_MEMO_PAGE_X_FIX)); diff --git a/storage/innodb_plugin/include/buf0buddy.h b/storage/innodb_plugin/include/buf0buddy.h index 7648950d5d1..d218a7112f1 100644 --- a/storage/innodb_plugin/include/buf0buddy.h +++ b/storage/innodb_plugin/include/buf0buddy.h @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 2006, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 2006, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -37,24 +37,19 @@ Created December 2006 by Marko Makela /**********************************************************************//** Allocate a block. The thread calling this function must hold buf_pool_mutex and must not hold buf_pool_zip_mutex or any -block->mutex. The buf_pool_mutex may only be released and reacquired -if lru != NULL. This function should only be used for allocating -compressed page frames or control blocks (buf_page_t). Allocated -control blocks must be properly initialized immediately after -buf_buddy_alloc() has returned the memory, before releasing -buf_pool_mutex. -@return allocated block, possibly NULL if lru == NULL */ +block->mutex. The buf_pool_mutex may be released and reacquired. +This function should only be used for allocating compressed page frames. +@return allocated block, never NULL */ UNIV_INLINE void* buf_buddy_alloc( /*============*/ - ulint size, /*!< in: block size, up to UNIV_PAGE_SIZE */ + ulint size, /*!< in: compressed page size + (between PAGE_ZIP_MIN_SIZE and UNIV_PAGE_SIZE) */ ibool* lru) /*!< in: pointer to a variable that will be assigned TRUE if storage was allocated from the LRU list - and buf_pool_mutex was temporarily released, - or NULL if the LRU list should not be used */ - __attribute__((malloc)); - + and buf_pool_mutex was temporarily released */ + __attribute__((malloc, nonnull)); /**********************************************************************//** Release a block. */ UNIV_INLINE diff --git a/storage/innodb_plugin/include/buf0buddy.ic b/storage/innodb_plugin/include/buf0buddy.ic index c419a2374d9..1087b45ee61 100644 --- a/storage/innodb_plugin/include/buf0buddy.ic +++ b/storage/innodb_plugin/include/buf0buddy.ic @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 2006, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 2006, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -36,8 +36,8 @@ Created December 2006 by Marko Makela /**********************************************************************//** Allocate a block. The thread calling this function must hold buf_pool_mutex and must not hold buf_pool_zip_mutex or any block->mutex. -The buf_pool_mutex may only be released and reacquired if lru != NULL. -@return allocated block, possibly NULL if lru==NULL */ +The buf_pool_mutex may be released and reacquired. +@return allocated block, never NULL */ UNIV_INTERN void* buf_buddy_alloc_low( @@ -46,9 +46,8 @@ buf_buddy_alloc_low( or BUF_BUDDY_SIZES */ ibool* lru) /*!< in: pointer to a variable that will be assigned TRUE if storage was allocated from the LRU list - and buf_pool_mutex was temporarily released, - or NULL if the LRU list should not be used */ - __attribute__((malloc)); + and buf_pool_mutex was temporarily released */ + __attribute__((malloc, nonnull)); /**********************************************************************//** Deallocate a block. */ @@ -74,6 +73,8 @@ buf_buddy_get_slot( ulint i; ulint s; + ut_ad(size >= PAGE_ZIP_MIN_SIZE); + for (i = 0, s = BUF_BUDDY_LOW; s < size; i++, s <<= 1) { } @@ -84,26 +85,25 @@ buf_buddy_get_slot( /**********************************************************************//** Allocate a block. The thread calling this function must hold buf_pool_mutex and must not hold buf_pool_zip_mutex or any -block->mutex. The buf_pool_mutex may only be released and reacquired -if lru != NULL. This function should only be used for allocating -compressed page frames or control blocks (buf_page_t). Allocated -control blocks must be properly initialized immediately after -buf_buddy_alloc() has returned the memory, before releasing -buf_pool_mutex. -@return allocated block, possibly NULL if lru == NULL */ +block->mutex. The buf_pool_mutex may be released and reacquired. +This function should only be used for allocating compressed page frames. +@return allocated block, never NULL */ UNIV_INLINE void* buf_buddy_alloc( /*============*/ - ulint size, /*!< in: block size, up to UNIV_PAGE_SIZE */ + ulint size, /*!< in: compressed page size + (between PAGE_ZIP_MIN_SIZE and UNIV_PAGE_SIZE) */ ibool* lru) /*!< in: pointer to a variable that will be assigned TRUE if storage was allocated from the LRU list - and buf_pool_mutex was temporarily released, - or NULL if the LRU list should not be used */ + and buf_pool_mutex was temporarily released */ { ut_ad(buf_pool_mutex_own()); + ut_ad(ut_is_2pow(size)); + ut_ad(size >= PAGE_ZIP_MIN_SIZE); + ut_ad(size <= UNIV_PAGE_SIZE); - return(buf_buddy_alloc_low(buf_buddy_get_slot(size), lru)); + return((byte*) buf_buddy_alloc_low(buf_buddy_get_slot(size), lru)); } /**********************************************************************//** @@ -117,6 +117,9 @@ buf_buddy_free( ulint size) /*!< in: block size, up to UNIV_PAGE_SIZE */ { ut_ad(buf_pool_mutex_own()); + ut_ad(ut_is_2pow(size)); + ut_ad(size >= PAGE_ZIP_MIN_SIZE); + ut_ad(size <= UNIV_PAGE_SIZE); buf_buddy_free_low(buf, buf_buddy_get_slot(size)); } diff --git a/storage/innodb_plugin/include/buf0buf.h b/storage/innodb_plugin/include/buf0buf.h index 05dead5ac9e..557bc17d311 100644 --- a/storage/innodb_plugin/include/buf0buf.h +++ b/storage/innodb_plugin/include/buf0buf.h @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1995, 2010, Innobase Oy. All Rights Reserved. +Copyright (c) 1995, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -141,12 +141,6 @@ buf_relocate( BUF_BLOCK_ZIP_DIRTY or BUF_BLOCK_ZIP_PAGE */ buf_page_t* dpage) /*!< in/out: destination control block */ __attribute__((nonnull)); -/********************************************************************//** -Resizes the buffer pool. */ -UNIV_INTERN -void -buf_pool_resize(void); -/*=================*/ /*********************************************************************//** Gets the current size of buffer buf_pool in bytes. @return size in bytes */ @@ -163,6 +157,23 @@ ib_uint64_t buf_pool_get_oldest_modification(void); /*==================================*/ /********************************************************************//** +Allocates a buf_page_t descriptor. This function must succeed. In case +of failure we assert in this function. */ +UNIV_INLINE +buf_page_t* +buf_page_alloc_descriptor(void) +/*===========================*/ + __attribute__((malloc)); +/********************************************************************//** +Free a buf_page_t descriptor. */ +UNIV_INLINE +void +buf_page_free_descriptor( +/*=====================*/ + buf_page_t* bpage) /*!< in: bpage descriptor to free. */ + __attribute__((nonnull)); + +/********************************************************************//** Allocates a buffer block. @return own: the allocated block, in state BUF_BLOCK_MEMORY */ UNIV_INLINE @@ -361,15 +372,6 @@ buf_page_peek( /*==========*/ ulint space, /*!< in: space id */ ulint offset);/*!< in: page number */ -/********************************************************************//** -Resets the check_index_page_at_flush field of a page if found in the buffer -pool. */ -UNIV_INTERN -void -buf_reset_check_index_page_at_flush( -/*================================*/ - ulint space, /*!< in: space id */ - ulint offset);/*!< in: page number */ #if defined UNIV_DEBUG_FILE_ACCESSES || defined UNIV_DEBUG /********************************************************************//** Sets file_page_was_freed TRUE if the page is found in the buffer pool. @@ -416,6 +418,18 @@ buf_block_get_freed_page_clock( __attribute__((pure)); /********************************************************************//** +Tells if a block is still close enough to the MRU end of the LRU list +meaning that it is not in danger of getting evicted and also implying +that it has been accessed recently. +Note that this is for heuristics only and does not reserve buffer pool +mutex. +@return TRUE if block is close to MRU end of LRU */ +UNIV_INLINE +ibool +buf_page_peek_if_young( +/*===================*/ + const buf_page_t* bpage); /*!< in: block */ +/********************************************************************//** Recommends a move of a block to the start of the LRU list if there is danger of dropping from the buffer pool. NOTE: does not reserve the buffer pool mutex. @@ -467,6 +481,31 @@ buf_block_get_modify_clock( #else /* !UNIV_HOTBACKUP */ # define buf_block_modify_clock_inc(block) ((void) 0) #endif /* !UNIV_HOTBACKUP */ +/*******************************************************************//** +Increments the bufferfix count. */ +UNIV_INLINE +void +buf_block_buf_fix_inc_func( +/*=======================*/ +#ifdef UNIV_SYNC_DEBUG + const char* file, /*!< in: file name */ + ulint line, /*!< in: line */ +#endif /* UNIV_SYNC_DEBUG */ + buf_block_t* block) /*!< in/out: block to bufferfix */ + __attribute__((nonnull)); +#ifdef UNIV_SYNC_DEBUG +/** Increments the bufferfix count. +@param b in/out: block to bufferfix +@param f in: file name where requested +@param l in: line number where requested */ +# define buf_block_buf_fix_inc(b,f,l) buf_block_buf_fix_inc_func(f,l,b) +#else /* UNIV_SYNC_DEBUG */ +/** Increments the bufferfix count. +@param b in/out: block to bufferfix +@param f in: file name where requested +@param l in: line number where requested */ +# define buf_block_buf_fix_inc(b,f,l) buf_block_buf_fix_inc_func(b) +#endif /* UNIV_SYNC_DEBUG */ /********************************************************************//** Calculates a page checksum which is stored to the page when it is written to a file. Note that we must be careful to calculate the same value @@ -1298,6 +1337,8 @@ struct buf_pool_stat_struct{ ulint n_pages_written;/*!< number write operations */ ulint n_pages_created;/*!< number of pages created in the pool with no read */ + ulint n_ra_pages_read_rnd;/*!< number of pages read in + as part of random read ahead */ ulint n_ra_pages_read;/*!< number of pages read in as part of read ahead */ ulint n_ra_pages_evicted;/*!< number of read ahead @@ -1421,8 +1462,10 @@ struct buf_pool_struct{ frames and buf_page_t descriptors of blocks that exist in the buffer pool only in compressed form. */ /* @{ */ +#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG UT_LIST_BASE_NODE_T(buf_page_t) zip_clean; /*!< unmodified compressed pages */ +#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */ UT_LIST_BASE_NODE_T(buf_page_t) zip_free[BUF_BUDDY_SIZES]; /*!< buddy free lists */ #if BUF_BUDDY_HIGH != UNIV_PAGE_SIZE diff --git a/storage/innodb_plugin/include/buf0buf.ic b/storage/innodb_plugin/include/buf0buf.ic index 0025bef5aac..2f3d65e80bd 100644 --- a/storage/innodb_plugin/include/buf0buf.ic +++ b/storage/innodb_plugin/include/buf0buf.ic @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1995, 2010, Innobase Oy. All Rights Reserved. +Copyright (c) 1995, 2011, Oracle and/or its affiliates. All Rights Reserved. Copyright (c) 2008, Google Inc. Portions of this file contain modifications contributed and copyrighted by @@ -62,6 +62,27 @@ buf_block_get_freed_page_clock( } /********************************************************************//** +Tells if a block is still close enough to the MRU end of the LRU list +meaning that it is not in danger of getting evicted and also implying +that it has been accessed recently. +Note that this is for heuristics only and does not reserve buffer pool +mutex. +@return TRUE if block is close to MRU end of LRU */ +UNIV_INLINE +ibool +buf_page_peek_if_young( +/*===================*/ + const buf_page_t* bpage) /*!< in: block */ +{ + /* FIXME: bpage->freed_page_clock is 31 bits */ + return((buf_pool->freed_page_clock & ((1UL << 31) - 1)) + < ((ulint) bpage->freed_page_clock + + (buf_pool->curr_size + * (BUF_LRU_OLD_RATIO_DIV - buf_LRU_old_ratio) + / (BUF_LRU_OLD_RATIO_DIV * 4)))); +} + +/********************************************************************//** Recommends a move of a block to the start of the LRU list if there is danger of dropping from the buffer pool. NOTE: does not reserve the buffer pool mutex. @@ -89,12 +110,7 @@ buf_page_peek_if_too_old( buf_pool->stat.n_pages_not_made_young++; return(FALSE); } else { - /* FIXME: bpage->freed_page_clock is 31 bits */ - return((buf_pool->freed_page_clock & ((1UL << 31) - 1)) - > ((ulint) bpage->freed_page_clock - + (buf_pool->curr_size - * (BUF_LRU_OLD_RATIO_DIV - buf_LRU_old_ratio) - / (BUF_LRU_OLD_RATIO_DIV * 4)))); + return(!buf_page_peek_if_young(bpage)); } } @@ -715,6 +731,35 @@ buf_block_get_lock_hash_val( } /********************************************************************//** +Allocates a buf_page_t descriptor. This function must succeed. In case +of failure we assert in this function. +@return: the allocated descriptor. */ +UNIV_INLINE +buf_page_t* +buf_page_alloc_descriptor(void) +/*===========================*/ +{ + buf_page_t* bpage; + + bpage = (buf_page_t*) ut_malloc(sizeof *bpage); + ut_d(memset(bpage, 0, sizeof *bpage)); + UNIV_MEM_ALLOC(bpage, sizeof *bpage); + + return(bpage); +} + +/********************************************************************//** +Free a buf_page_t descriptor. */ +UNIV_INLINE +void +buf_page_free_descriptor( +/*=====================*/ + buf_page_t* bpage) /*!< in: bpage descriptor to free. */ +{ + ut_free(bpage); +} + +/********************************************************************//** Allocates a buffer block. @return own: the allocated block, in state BUF_BLOCK_MEMORY */ UNIV_INLINE @@ -871,19 +916,6 @@ buf_block_buf_fix_inc_func( block->page.buf_fix_count++; } -#ifdef UNIV_SYNC_DEBUG -/** Increments the bufferfix count. -@param b in/out: block to bufferfix -@param f in: file name where requested -@param l in: line number where requested */ -# define buf_block_buf_fix_inc(b,f,l) buf_block_buf_fix_inc_func(f,l,b) -#else /* UNIV_SYNC_DEBUG */ -/** Increments the bufferfix count. -@param b in/out: block to bufferfix -@param f in: file name where requested -@param l in: line number where requested */ -# define buf_block_buf_fix_inc(b,f,l) buf_block_buf_fix_inc_func(b) -#endif /* UNIV_SYNC_DEBUG */ /*******************************************************************//** Decrements the bufferfix count. */ @@ -1071,7 +1103,7 @@ buf_block_dbg_add_level( where we have acquired latch */ ulint level) /*!< in: latching order level */ { - sync_thread_add_level(&block->lock, level); + sync_thread_add_level(&block->lock, level, FALSE); } #endif /* UNIV_SYNC_DEBUG */ #endif /* !UNIV_HOTBACKUP */ diff --git a/storage/innodb_plugin/include/buf0lru.h b/storage/innodb_plugin/include/buf0lru.h index d543bce53cd..0c2102b9549 100644 --- a/storage/innodb_plugin/include/buf0lru.h +++ b/storage/innodb_plugin/include/buf0lru.h @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1995, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1995, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -30,18 +30,6 @@ Created 11/5/1995 Heikki Tuuri #include "ut0byte.h" #include "buf0types.h" -/** The return type of buf_LRU_free_block() */ -enum buf_lru_free_block_status { - /** freed */ - BUF_LRU_FREED = 0, - /** not freed because the caller asked to remove the - uncompressed frame but the control block cannot be - relocated */ - BUF_LRU_CANNOT_RELOCATE, - /** not freed because of some other reason */ - BUF_LRU_NOT_FREED -}; - /******************************************************************//** Tries to remove LRU flushed blocks from the end of the LRU list and put them to the free list. This is beneficial for the efficiency of the insert buffer @@ -84,6 +72,7 @@ void buf_LRU_invalidate_tablespace( /*==========================*/ ulint id); /*!< in: space id */ +#if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG /********************************************************************//** Insert a compressed block into buf_pool->zip_clean in the LRU order. */ UNIV_INTERN @@ -91,22 +80,22 @@ void buf_LRU_insert_zip_clean( /*=====================*/ buf_page_t* bpage); /*!< in: pointer to the block in question */ +#endif /* UNIV_DEBUG || UNIV_BUF_DEBUG */ /******************************************************************//** Try to free a block. If bpage is a descriptor of a compressed-only page, the descriptor object will be freed as well. -NOTE: If this function returns BUF_LRU_FREED, it will temporarily +NOTE: If this function returns TRUE, it will temporarily release buf_pool_mutex. Furthermore, the page frame will no longer be accessible via bpage. The caller must hold buf_pool_mutex and buf_page_get_mutex(bpage) and release these two mutexes after the call. No other buf_page_get_mutex() may be held when calling this function. -@return BUF_LRU_FREED if freed, BUF_LRU_CANNOT_RELOCATE or -BUF_LRU_NOT_FREED otherwise. */ +@return TRUE if freed, FALSE otherwise. */ UNIV_INTERN -enum buf_lru_free_block_status +ibool buf_LRU_free_block( /*===============*/ buf_page_t* bpage, /*!< in: block to be freed */ diff --git a/storage/innodb_plugin/include/buf0types.h b/storage/innodb_plugin/include/buf0types.h index bfae6477135..4fe0b4483c8 100644 --- a/storage/innodb_plugin/include/buf0types.h +++ b/storage/innodb_plugin/include/buf0types.h @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1995, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1995, 2011, Oracle and/or its affiliates. All Rights Reserved 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 @@ -26,6 +26,8 @@ Created 11/17/1995 Heikki Tuuri #ifndef buf0types_h #define buf0types_h +#include "page0types.h" + /** Buffer page (uncompressed or compressed) */ typedef struct buf_page_struct buf_page_t; /** Buffer block for which an uncompressed page exists */ @@ -58,17 +60,10 @@ enum buf_io_fix { /** Parameters of binary buddy system for compressed pages (buf0buddy.h) */ /* @{ */ -#if UNIV_WORD_SIZE <= 4 /* 32-bit system */ -/** Base-2 logarithm of the smallest buddy block size */ -# define BUF_BUDDY_LOW_SHIFT 6 -#else /* 64-bit system */ -/** Base-2 logarithm of the smallest buddy block size */ -# define BUF_BUDDY_LOW_SHIFT 7 -#endif +#define BUF_BUDDY_LOW_SHIFT PAGE_ZIP_MIN_SIZE_SHIFT + #define BUF_BUDDY_LOW (1 << BUF_BUDDY_LOW_SHIFT) - /*!< minimum block size in the binary - buddy system; must be at least - sizeof(buf_page_t) */ + #define BUF_BUDDY_SIZES (UNIV_PAGE_SIZE_SHIFT - BUF_BUDDY_LOW_SHIFT) /*!< number of buddy sizes */ diff --git a/storage/innodb_plugin/include/fsp0fsp.h b/storage/innodb_plugin/include/fsp0fsp.h index 7abd3914eda..2221380c9a2 100644 --- a/storage/innodb_plugin/include/fsp0fsp.h +++ b/storage/innodb_plugin/include/fsp0fsp.h @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1995, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1995, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -176,19 +176,18 @@ fseg_n_reserved_pages( Allocates a single free page from a segment. This function implements the intelligent allocation strategy which tries to minimize file space fragmentation. -@return the allocated page offset FIL_NULL if no page could be allocated */ -UNIV_INTERN -ulint -fseg_alloc_free_page( -/*=================*/ - fseg_header_t* seg_header, /*!< in: segment header */ - ulint hint, /*!< in: hint of which page would be desirable */ - byte direction, /*!< in: if the new page is needed because +@param[in/out] seg_header segment header +@param[in] hint hint of which page would be desirable +@param[in] direction if the new page is needed because of an index page split, and records are inserted there in order, into which direction they go alphabetically: FSP_DOWN, - FSP_UP, FSP_NO_DIR */ - mtr_t* mtr); /*!< in: mtr handle */ + FSP_UP, FSP_NO_DIR +@param[in/out] mtr mini-transaction +@return the allocated page offset FIL_NULL if no page could be allocated */ +#define fseg_alloc_free_page(seg_header, hint, direction, mtr) \ + fseg_alloc_free_page_general(seg_header, hint, direction, \ + FALSE, mtr, mtr) /**********************************************************************//** Allocates a single free page from a segment. This function implements the intelligent allocation strategy which tries to minimize file space @@ -198,7 +197,7 @@ UNIV_INTERN ulint fseg_alloc_free_page_general( /*=========================*/ - fseg_header_t* seg_header,/*!< in: segment header */ + fseg_header_t* seg_header,/*!< in/out: segment header */ ulint hint, /*!< in: hint of which page would be desirable */ byte direction,/*!< in: if the new page is needed because of an index page split, and records are @@ -210,7 +209,12 @@ fseg_alloc_free_page_general( with fsp_reserve_free_extents, then there is no need to do the check for this individual page */ - mtr_t* mtr); /*!< in: mtr handle */ + mtr_t* mtr, /*!< in/out: mini-transaction */ + mtr_t* init_mtr)/*!< in/out: mtr or another mini-transaction + in which the page should be initialized, + or NULL if this is a "fake allocation" of + a page that was previously freed in mtr */ + __attribute__((warn_unused_result, nonnull(1,5))); /**********************************************************************//** Reserves free pages from a tablespace. All mini-transactions which may use several pages from the tablespace should call this function beforehand diff --git a/storage/innodb_plugin/include/mtr0mtr.h b/storage/innodb_plugin/include/mtr0mtr.h index bc3f1951be9..2a561131c09 100644 --- a/storage/innodb_plugin/include/mtr0mtr.h +++ b/storage/innodb_plugin/include/mtr0mtr.h @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1995, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1995, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -53,6 +53,8 @@ first 3 values must be RW_S_LATCH, RW_X_LATCH, RW_NO_LATCH */ #define MTR_MEMO_MODIFY 54 #define MTR_MEMO_S_LOCK 55 #define MTR_MEMO_X_LOCK 56 +/** The mini-transaction freed a clustered index leaf page. */ +#define MTR_MEMO_FREE_CLUST_LEAF 57 /** @name Log item types The log items are declared 'byte' so that the compiler can warn if val @@ -387,9 +389,12 @@ struct mtr_struct{ #endif dyn_array_t memo; /*!< memo stack for locks etc. */ dyn_array_t log; /*!< mini-transaction log */ - ibool modifications; - /* TRUE if the mtr made modifications to - buffer pool pages */ + unsigned modifications:1; + /*!< TRUE if the mini-transaction + modified buffer pool pages */ + unsigned freed_clust_leaf:1; + /*!< TRUE if MTR_MEMO_FREE_CLUST_LEAF + was logged in the mini-transaction */ ulint n_log_recs; /* count of how many page initial log records have been written to the mtr log */ diff --git a/storage/innodb_plugin/include/mtr0mtr.ic b/storage/innodb_plugin/include/mtr0mtr.ic index 18f8e87b3cf..9c0ddff9132 100644 --- a/storage/innodb_plugin/include/mtr0mtr.ic +++ b/storage/innodb_plugin/include/mtr0mtr.ic @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1995, 2010, Innobase Oy. All Rights Reserved. +Copyright (c) 1995, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -44,6 +44,7 @@ mtr_start( mtr->log_mode = MTR_LOG_ALL; mtr->modifications = FALSE; + mtr->freed_clust_leaf = FALSE; mtr->n_log_recs = 0; ut_d(mtr->state = MTR_ACTIVE); @@ -67,7 +68,8 @@ mtr_memo_push( ut_ad(object); ut_ad(type >= MTR_MEMO_PAGE_S_FIX); - ut_ad(type <= MTR_MEMO_X_LOCK); + ut_ad(type <= MTR_MEMO_FREE_CLUST_LEAF); + ut_ad(type != MTR_MEMO_FREE_CLUST_LEAF || mtr->freed_clust_leaf); ut_ad(mtr); ut_ad(mtr->magic_n == MTR_MAGIC_N); ut_ad(mtr->state == MTR_ACTIVE); diff --git a/storage/innodb_plugin/include/page0cur.ic b/storage/innodb_plugin/include/page0cur.ic index 3520677dfb3..81474fa35f5 100644 --- a/storage/innodb_plugin/include/page0cur.ic +++ b/storage/innodb_plugin/include/page0cur.ic @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1994, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1994, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -27,6 +27,8 @@ Created 10/4/1994 Heikki Tuuri #include "buf0types.h" #ifdef UNIV_DEBUG +# include "rem0cmp.h" + /*********************************************************//** Gets pointer to the page frame where the cursor is positioned. @return page */ @@ -268,6 +270,7 @@ page_cur_tuple_insert( index, rec, offsets, mtr); } + ut_ad(!rec || !cmp_dtuple_rec(tuple, rec, offsets)); mem_heap_free(heap); return(rec); } diff --git a/storage/innodb_plugin/include/page0page.h b/storage/innodb_plugin/include/page0page.h index 3899499fb6a..caf4cee2c57 100644 --- a/storage/innodb_plugin/include/page0page.h +++ b/storage/innodb_plugin/include/page0page.h @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1994, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1994, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -68,10 +68,7 @@ typedef byte page_header_t; #define PAGE_MAX_TRX_ID 18 /* highest id of a trx which may have modified a record on the page; a dulint; defined only in secondary indexes and in the insert buffer - tree; NOTE: this may be modified only - when the thread has an x-latch to the page, - and ALSO an x-latch to btr_search_latch - if there is a hash index to the page! */ + tree */ #define PAGE_HEADER_PRIV_END 26 /* end of private data structure of the page header which are set in a page create */ /*----*/ @@ -284,16 +281,42 @@ page_get_supremum_offset( const page_t* page); /*!< in: page which must have record(s) */ #define page_get_infimum_rec(page) ((page) + page_get_infimum_offset(page)) #define page_get_supremum_rec(page) ((page) + page_get_supremum_offset(page)) + /************************************************************//** -Returns the middle record of record list. If there are an even number -of records in the list, returns the first record of upper half-list. -@return middle record */ +Returns the nth record of the record list. +This is the inverse function of page_rec_get_n_recs_before(). +@return nth record */ UNIV_INTERN +const rec_t* +page_rec_get_nth_const( +/*===================*/ + const page_t* page, /*!< in: page */ + ulint nth) /*!< in: nth record */ + __attribute__((nonnull, warn_unused_result)); +/************************************************************//** +Returns the nth record of the record list. +This is the inverse function of page_rec_get_n_recs_before(). +@return nth record */ +UNIV_INLINE +rec_t* +page_rec_get_nth( +/*=============*/ + page_t* page, /*< in: page */ + ulint nth) /*!< in: nth record */ + __attribute__((nonnull, warn_unused_result)); + +#ifndef UNIV_HOTBACKUP +/************************************************************//** +Returns the middle record of the records on the page. If there is an +even number of records in the list, returns the first record of the +upper half-list. +@return middle record */ +UNIV_INLINE rec_t* page_get_middle_rec( /*================*/ - page_t* page); /*!< in: page */ -#ifndef UNIV_HOTBACKUP + page_t* page) /*!< in: page */ + __attribute__((nonnull, warn_unused_result)); /*************************************************************//** Compares a data tuple to a physical record. Differs from the function cmp_dtuple_rec_with_match in the way that the record must reside on an @@ -348,6 +371,7 @@ page_get_n_recs( /***************************************************************//** Returns the number of records before the given record in chain. The number includes infimum and supremum records. +This is the inverse function of page_rec_get_nth(). @return number of records */ UNIV_INTERN ulint diff --git a/storage/innodb_plugin/include/page0page.ic b/storage/innodb_plugin/include/page0page.ic index 8f794410f20..1450b0892b3 100644 --- a/storage/innodb_plugin/include/page0page.ic +++ b/storage/innodb_plugin/include/page0page.ic @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1994, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1994, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -420,7 +420,37 @@ page_rec_is_infimum( return(page_rec_is_infimum_low(page_offset(rec))); } +/************************************************************//** +Returns the nth record of the record list. +This is the inverse function of page_rec_get_n_recs_before(). +@return nth record */ +UNIV_INLINE +rec_t* +page_rec_get_nth( +/*=============*/ + page_t* page, /*!< in: page */ + ulint nth) /*!< in: nth record */ +{ + return((rec_t*) page_rec_get_nth_const(page, nth)); +} + #ifndef UNIV_HOTBACKUP +/************************************************************//** +Returns the middle record of the records on the page. If there is an +even number of records in the list, returns the first record of the +upper half-list. +@return middle record */ +UNIV_INLINE +rec_t* +page_get_middle_rec( +/*================*/ + page_t* page) /*!< in: page */ +{ + ulint middle = (page_get_n_recs(page) + PAGE_HEAP_NO_USER_LOW) / 2; + + return(page_rec_get_nth(page, middle)); +} + /*************************************************************//** Compares a data tuple to a physical record. Differs from the function cmp_dtuple_rec_with_match in the way that the record must reside on an diff --git a/storage/innodb_plugin/include/rem0rec.h b/storage/innodb_plugin/include/rem0rec.h index 17d08afabb9..06de23be757 100644 --- a/storage/innodb_plugin/include/rem0rec.h +++ b/storage/innodb_plugin/include/rem0rec.h @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1994, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1994, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -480,6 +480,18 @@ ulint rec_offs_any_extern( /*================*/ const ulint* offsets);/*!< in: array returned by rec_get_offsets() */ +#if defined UNIV_DEBUG || defined UNIV_BLOB_LIGHT_DEBUG +/******************************************************//** +Determine if the offsets are for a record containing null BLOB pointers. +@return first field containing a null BLOB pointer, or NULL if none found */ +UNIV_INLINE +const byte* +rec_offs_any_null_extern( +/*=====================*/ + const rec_t* rec, /*!< in: record */ + const ulint* offsets) /*!< in: rec_get_offsets(rec) */ + __attribute__((nonnull, warn_unused_result)); +#endif /* UNIV_DEBUG || UNIV_BLOB_LIGHT_DEBUG */ /******************************************************//** Returns nonzero if the extern bit is set in nth field of rec. @return nonzero if externally stored */ diff --git a/storage/innodb_plugin/include/rem0rec.ic b/storage/innodb_plugin/include/rem0rec.ic index 8e5bd9a7fcd..7cff36fee6c 100644 --- a/storage/innodb_plugin/include/rem0rec.ic +++ b/storage/innodb_plugin/include/rem0rec.ic @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1994, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1994, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -26,6 +26,7 @@ Created 5/30/1994 Heikki Tuuri #include "mach0data.h" #include "ut0byte.h" #include "dict0dict.h" +#include "btr0types.h" /* Compact flag ORed to the extra size returned by rec_get_offsets() */ #define REC_OFFS_COMPACT ((ulint) 1 << 31) @@ -1087,6 +1088,44 @@ rec_offs_any_extern( return(UNIV_UNLIKELY(*rec_offs_base(offsets) & REC_OFFS_EXTERNAL)); } +#if defined UNIV_DEBUG || defined UNIV_BLOB_LIGHT_DEBUG +/******************************************************//** +Determine if the offsets are for a record containing null BLOB pointers. +@return first field containing a null BLOB pointer, or NULL if none found */ +UNIV_INLINE +const byte* +rec_offs_any_null_extern( +/*=====================*/ + const rec_t* rec, /*!< in: record */ + const ulint* offsets) /*!< in: rec_get_offsets(rec) */ +{ + ulint i; + ut_ad(rec_offs_validate(rec, NULL, offsets)); + + if (!rec_offs_any_extern(offsets)) { + return(NULL); + } + + for (i = 0; i < rec_offs_n_fields(offsets); i++) { + if (rec_offs_nth_extern(offsets, i)) { + ulint len; + const byte* field + = rec_get_nth_field(rec, offsets, i, &len); + + ut_a(len >= BTR_EXTERN_FIELD_REF_SIZE); + if (!memcmp(field + len + - BTR_EXTERN_FIELD_REF_SIZE, + field_ref_zero, + BTR_EXTERN_FIELD_REF_SIZE)) { + return(field); + } + } + } + + return(NULL); +} +#endif /* UNIV_DEBUG || UNIV_BLOB_LIGHT_DEBUG */ + /******************************************************//** Returns nonzero if the extern bit is set in nth field of rec. @return nonzero if externally stored */ diff --git a/storage/innodb_plugin/include/row0row.h b/storage/innodb_plugin/include/row0row.h index 723b7b53395..36fb26482ce 100644 --- a/storage/innodb_plugin/include/row0row.h +++ b/storage/innodb_plugin/include/row0row.h @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1996, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1996, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -38,16 +38,16 @@ Created 4/20/1996 Heikki Tuuri #include "btr0types.h" /*********************************************************************//** -Gets the offset of the trx id field, in bytes relative to the origin of +Gets the offset of the DB_TRX_ID field, in bytes relative to the origin of a clustered index record. @return offset of DATA_TRX_ID */ -UNIV_INTERN +UNIV_INLINE ulint row_get_trx_id_offset( /*==================*/ - const rec_t* rec, /*!< in: record */ - dict_index_t* index, /*!< in: clustered index */ - const ulint* offsets);/*!< in: rec_get_offsets(rec, index) */ + const dict_index_t* index, /*!< in: clustered index */ + const ulint* offsets)/*!< in: record offsets */ + __attribute__((nonnull, warn_unused_result)); /*********************************************************************//** Reads the trx id field from a clustered index record. @return value of the field */ @@ -55,9 +55,10 @@ UNIV_INLINE trx_id_t row_get_rec_trx_id( /*===============*/ - const rec_t* rec, /*!< in: record */ - dict_index_t* index, /*!< in: clustered index */ - const ulint* offsets);/*!< in: rec_get_offsets(rec, index) */ + const rec_t* rec, /*!< in: record */ + const dict_index_t* index, /*!< in: clustered index */ + const ulint* offsets)/*!< in: rec_get_offsets(rec, index) */ + __attribute__((nonnull, warn_unused_result)); /*********************************************************************//** Reads the roll pointer field from a clustered index record. @return value of the field */ @@ -65,9 +66,10 @@ UNIV_INLINE roll_ptr_t row_get_rec_roll_ptr( /*=================*/ - const rec_t* rec, /*!< in: record */ - dict_index_t* index, /*!< in: clustered index */ - const ulint* offsets);/*!< in: rec_get_offsets(rec, index) */ + const rec_t* rec, /*!< in: record */ + const dict_index_t* index, /*!< in: clustered index */ + const ulint* offsets)/*!< in: rec_get_offsets(rec, index) */ + __attribute__((nonnull, warn_unused_result)); /*****************************************************************//** When an insert or purge to a table is performed, this function builds the entry to be inserted into or purged from an index on the table. diff --git a/storage/innodb_plugin/include/row0row.ic b/storage/innodb_plugin/include/row0row.ic index 05c007641af..0b9ca982af8 100644 --- a/storage/innodb_plugin/include/row0row.ic +++ b/storage/innodb_plugin/include/row0row.ic @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1996, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1996, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -28,15 +28,42 @@ Created 4/20/1996 Heikki Tuuri #include "trx0undo.h" /*********************************************************************//** +Gets the offset of the DB_TRX_ID field, in bytes relative to the origin of +a clustered index record. +@return offset of DATA_TRX_ID */ +UNIV_INLINE +ulint +row_get_trx_id_offset( +/*==================*/ + const dict_index_t* index, /*!< in: clustered index */ + const ulint* offsets)/*!< in: record offsets */ +{ + ulint pos; + ulint offset; + ulint len; + + ut_ad(dict_index_is_clust(index)); + ut_ad(rec_offs_validate(NULL, index, offsets)); + + pos = dict_index_get_sys_col_pos(index, DATA_TRX_ID); + + offset = rec_get_nth_field_offs(offsets, pos, &len); + + ut_ad(len == DATA_TRX_ID_LEN); + + return(offset); +} + +/*********************************************************************//** Reads the trx id field from a clustered index record. @return value of the field */ UNIV_INLINE trx_id_t row_get_rec_trx_id( /*===============*/ - const rec_t* rec, /*!< in: record */ - dict_index_t* index, /*!< in: clustered index */ - const ulint* offsets)/*!< in: rec_get_offsets(rec, index) */ + const rec_t* rec, /*!< in: record */ + const dict_index_t* index, /*!< in: clustered index */ + const ulint* offsets)/*!< in: rec_get_offsets(rec, index) */ { ulint offset; @@ -46,7 +73,7 @@ row_get_rec_trx_id( offset = index->trx_id_offset; if (!offset) { - offset = row_get_trx_id_offset(rec, index, offsets); + offset = row_get_trx_id_offset(index, offsets); } return(trx_read_trx_id(rec + offset)); @@ -59,9 +86,9 @@ UNIV_INLINE roll_ptr_t row_get_rec_roll_ptr( /*=================*/ - const rec_t* rec, /*!< in: record */ - dict_index_t* index, /*!< in: clustered index */ - const ulint* offsets)/*!< in: rec_get_offsets(rec, index) */ + const rec_t* rec, /*!< in: record */ + const dict_index_t* index, /*!< in: clustered index */ + const ulint* offsets)/*!< in: rec_get_offsets(rec, index) */ { ulint offset; @@ -71,7 +98,7 @@ row_get_rec_roll_ptr( offset = index->trx_id_offset; if (!offset) { - offset = row_get_trx_id_offset(rec, index, offsets); + offset = row_get_trx_id_offset(index, offsets); } return(trx_read_roll_ptr(rec + offset + DATA_TRX_ID_LEN)); diff --git a/storage/innodb_plugin/include/row0upd.ic b/storage/innodb_plugin/include/row0upd.ic index 18e22f1eca9..11db82f64da 100644 --- a/storage/innodb_plugin/include/row0upd.ic +++ b/storage/innodb_plugin/include/row0upd.ic @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1996, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1996, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -157,11 +157,6 @@ row_upd_rec_sys_fields( { ut_ad(dict_index_is_clust(index)); ut_ad(rec_offs_validate(rec, index, offsets)); -#ifdef UNIV_SYNC_DEBUG - if (!rw_lock_own(&btr_search_latch, RW_LOCK_EX)) { - ut_ad(!buf_block_align(rec)->is_hashed); - } -#endif /* UNIV_SYNC_DEBUG */ if (UNIV_LIKELY_NULL(page_zip)) { ulint pos = dict_index_get_sys_col_pos(index, DATA_TRX_ID); @@ -171,7 +166,7 @@ row_upd_rec_sys_fields( ulint offset = index->trx_id_offset; if (!offset) { - offset = row_get_trx_id_offset(rec, index, offsets); + offset = row_get_trx_id_offset(index, offsets); } #if DATA_TRX_ID + 1 != DATA_ROLL_PTR diff --git a/storage/innodb_plugin/include/srv0srv.h b/storage/innodb_plugin/include/srv0srv.h index 91ae895040c..7c63a5f0d45 100644 --- a/storage/innodb_plugin/include/srv0srv.h +++ b/storage/innodb_plugin/include/srv0srv.h @@ -143,6 +143,7 @@ extern ulint srv_mem_pool_size; extern ulint srv_lock_table_size; extern ulint srv_n_file_io_threads; +extern my_bool srv_random_read_ahead; extern ulong srv_read_ahead_threshold; extern ulint srv_n_read_io_threads; extern ulint srv_n_write_io_threads; @@ -618,6 +619,7 @@ struct export_var_struct{ ulint innodb_buffer_pool_wait_free; /*!< srv_buf_pool_wait_free */ ulint innodb_buffer_pool_pages_flushed; /*!< srv_buf_pool_flushed */ ulint innodb_buffer_pool_write_requests;/*!< srv_buf_pool_write_requests */ + ulint innodb_buffer_pool_read_ahead_rnd;/*!< srv_read_ahead_rnd */ ulint innodb_buffer_pool_read_ahead; /*!< srv_read_ahead */ ulint innodb_buffer_pool_read_ahead_evicted;/*!< srv_read_ahead evicted*/ ulint innodb_dblwr_pages_written; /*!< srv_dblwr_pages_written */ diff --git a/storage/innodb_plugin/include/sync0rw.ic b/storage/innodb_plugin/include/sync0rw.ic index 7116f1b7c9b..485a63a1b18 100644 --- a/storage/innodb_plugin/include/sync0rw.ic +++ b/storage/innodb_plugin/include/sync0rw.ic @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1995, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1995, 2011, Oracle and/or its affiliates. All Rights Reserved. Copyright (c) 2008, Google Inc. Portions of this file contain modifications contributed and copyrighted by @@ -603,16 +603,16 @@ rw_lock_x_unlock_direct( ut_ad((lock->lock_word % X_LOCK_DECR) == 0); -#ifdef UNIV_SYNC_DEBUG - rw_lock_remove_debug_info(lock, 0, RW_LOCK_EX); -#endif - if (lock->lock_word == 0) { lock->recursive = FALSE; UNIV_MEM_INVALID(&lock->writer_thread, sizeof lock->writer_thread); } +#ifdef UNIV_SYNC_DEBUG + rw_lock_remove_debug_info(lock, 0, RW_LOCK_EX); +#endif + lock->lock_word += X_LOCK_DECR; ut_ad(!lock->waiters); diff --git a/storage/innodb_plugin/include/sync0sync.h b/storage/innodb_plugin/include/sync0sync.h index 71c9920a10b..f2013c202c1 100644 --- a/storage/innodb_plugin/include/sync0sync.h +++ b/storage/innodb_plugin/include/sync0sync.h @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1995, 2010, Innobase Oy. All Rights Reserved. +Copyright (c) 1995, 2011, Oracle and/or its affiliates. All Rights Reserved. Copyright (c) 2008, Google Inc. Portions of this file contain modifications contributed and copyrighted by @@ -219,8 +219,10 @@ void sync_thread_add_level( /*==================*/ void* latch, /*!< in: pointer to a mutex or an rw-lock */ - ulint level); /*!< in: level in the latching order; if + ulint level, /*!< in: level in the latching order; if SYNC_LEVEL_VARYING, nothing is done */ + ibool relock) /*!< in: TRUE if re-entering an x-lock */ + __attribute__((nonnull)); /******************************************************************//** Removes a latch from the thread level array if it is found there. @return TRUE if found in the array; it is no error if the latch is @@ -446,10 +448,6 @@ or row lock! */ #define SYNC_DICT_HEADER 995 #define SYNC_IBUF_HEADER 914 #define SYNC_IBUF_PESS_INSERT_MUTEX 912 -#define SYNC_IBUF_MUTEX 910 /* ibuf mutex is really below - SYNC_FSP_PAGE: we assign a value this - high only to make the program to pass - the debug checks */ /*-------------------------------*/ #define SYNC_INDEX_TREE 900 #define SYNC_TREE_NODE_NEW 892 @@ -466,8 +464,11 @@ or row lock! */ #define SYNC_FSP 400 #define SYNC_FSP_PAGE 395 /*------------------------------------- Insert buffer headers */ -/*------------------------------------- ibuf_mutex */ +#define SYNC_IBUF_MUTEX 370 /* ibuf_mutex */ /*------------------------------------- Insert buffer tree */ +#define SYNC_IBUF_INDEX_TREE 360 +#define SYNC_IBUF_TREE_NODE_NEW 359 +#define SYNC_IBUF_TREE_NODE 358 #define SYNC_IBUF_BITMAP_MUTEX 351 #define SYNC_IBUF_BITMAP 350 /*------------------------------------- MySQL query cache mutex */ diff --git a/storage/innodb_plugin/include/trx0sys.h b/storage/innodb_plugin/include/trx0sys.h index cbb89689748..78bb6fc349b 100644 --- a/storage/innodb_plugin/include/trx0sys.h +++ b/storage/innodb_plugin/include/trx0sys.h @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1996, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1996, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -284,6 +284,17 @@ ibool trx_in_trx_list( /*============*/ trx_t* in_trx);/*!< in: trx */ +#if defined UNIV_DEBUG || defined UNIV_BLOB_LIGHT_DEBUG +/***********************************************************//** +Assert that a transaction has been recovered. +@return TRUE */ +UNIV_INLINE +ibool +trx_assert_recovered( +/*=================*/ + trx_id_t trx_id) /*!< in: transaction identifier */ + __attribute__((warn_unused_result)); +#endif /* UNIV_DEBUG || UNIV_BLOB_LIGHT_DEBUG */ /*****************************************************************//** Updates the offset information about the end of the MySQL binlog entry which corresponds to the transaction just being committed. In a MySQL diff --git a/storage/innodb_plugin/include/trx0sys.ic b/storage/innodb_plugin/include/trx0sys.ic index 820d31d0692..6246debac0a 100644 --- a/storage/innodb_plugin/include/trx0sys.ic +++ b/storage/innodb_plugin/include/trx0sys.ic @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1996, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1996, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -277,6 +277,28 @@ trx_get_on_id( return(NULL); } +#if defined UNIV_DEBUG || defined UNIV_BLOB_LIGHT_DEBUG +/***********************************************************//** +Assert that a transaction has been recovered. +@return TRUE */ +UNIV_INLINE +ibool +trx_assert_recovered( +/*=================*/ + trx_id_t trx_id) /*!< in: transaction identifier */ +{ + trx_t* trx; + + mutex_enter(&kernel_mutex); + trx = trx_get_on_id(trx_id); + ut_a(trx); + ut_a(trx->is_recovered); + mutex_exit(&kernel_mutex); + + return(TRUE); +} +#endif /* UNIV_DEBUG || UNIV_BLOB_LIGHT_DEBUG */ + /****************************************************************//** Returns the minumum trx id in trx list. This is the smallest id for which the trx can possibly be active. (But, you must look at the trx->conc_state to diff --git a/storage/innodb_plugin/include/trx0trx.h b/storage/innodb_plugin/include/trx0trx.h index 833bae4a4ff..4bf3e75a5ee 100644 --- a/storage/innodb_plugin/include/trx0trx.h +++ b/storage/innodb_plugin/include/trx0trx.h @@ -44,6 +44,9 @@ extern sess_t* trx_dummy_sess; /** Number of transactions currently allocated for MySQL: protected by the kernel mutex */ extern ulint trx_n_mysql_transactions; +/** Number of transactions currently in the XA PREPARED state: protected by +the kernel mutex */ +extern ulint trx_n_prepared; /********************************************************************//** Releases the search latch if trx has reserved it. */ @@ -108,6 +111,14 @@ trx_free( /*=====*/ trx_t* trx); /*!< in, own: trx object */ /********************************************************************//** +At shutdown, frees a transaction object that is in the PREPARED state. */ +UNIV_INTERN +void +trx_free_prepared( +/*==============*/ + trx_t* trx) /*!< in, own: trx object */ + __attribute__((nonnull)); +/********************************************************************//** Frees a transaction object for MySQL. */ UNIV_INTERN void diff --git a/storage/innodb_plugin/include/trx0undo.h b/storage/innodb_plugin/include/trx0undo.h index a084f2394b5..c95f99d6417 100644 --- a/storage/innodb_plugin/include/trx0undo.h +++ b/storage/innodb_plugin/include/trx0undo.h @@ -1,6 +1,6 @@ /***************************************************************************** -Copyright (c) 1996, 2009, Innobase Oy. All Rights Reserved. +Copyright (c) 1996, 2011, Oracle and/or its affiliates. All Rights Reserved. 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 @@ -204,17 +204,51 @@ trx_undo_add_page( mtr_t* mtr); /*!< in: mtr which does not have a latch to any undo log page; the caller must have reserved the rollback segment mutex */ +/********************************************************************//** +Frees the last undo log page. +The caller must hold the rollback segment mutex. */ +UNIV_INTERN +void +trx_undo_free_last_page_func( +/*==========================*/ +#ifdef UNIV_DEBUG + const trx_t* trx, /*!< in: transaction */ +#endif /* UNIV_DEBUG */ + trx_undo_t* undo, /*!< in/out: undo log memory copy */ + mtr_t* mtr) /*!< in/out: mini-transaction which does not + have a latch to any undo log page or which + has allocated the undo log page */ + __attribute__((nonnull)); +#ifdef UNIV_DEBUG +# define trx_undo_free_last_page(trx,undo,mtr) \ + trx_undo_free_last_page_func(trx,undo,mtr) +#else /* UNIV_DEBUG */ +# define trx_undo_free_last_page(trx,undo,mtr) \ + trx_undo_free_last_page_func(undo,mtr) +#endif /* UNIV_DEBUG */ + /***********************************************************************//** Truncates an undo log from the end. This function is used during a rollback to free space from an undo log. */ UNIV_INTERN void -trx_undo_truncate_end( -/*==================*/ - trx_t* trx, /*!< in: transaction whose undo log it is */ - trx_undo_t* undo, /*!< in: undo log */ - undo_no_t limit); /*!< in: all undo records with undo number +trx_undo_truncate_end_func( +/*=======================*/ +#ifdef UNIV_DEBUG + const trx_t* trx, /*!< in: transaction whose undo log it is */ +#endif /* UNIV_DEBUG */ + trx_undo_t* undo, /*!< in/out: undo log */ + undo_no_t limit) /*!< in: all undo records with undo number >= this value should be truncated */ + __attribute__((nonnull)); +#ifdef UNIV_DEBUG +# define trx_undo_truncate_end(trx,undo,limit) \ + trx_undo_truncate_end_func(trx,undo,limit) +#else /* UNIV_DEBUG */ +# define trx_undo_truncate_end(trx,undo,limit) \ + trx_undo_truncate_end_func(undo,limit) +#endif /* UNIV_DEBUG */ + /***********************************************************************//** Truncates an undo log from the start. This function is used during a purge operation. */ @@ -298,6 +332,15 @@ void trx_undo_insert_cleanup( /*====================*/ trx_t* trx); /*!< in: transaction handle */ + +/********************************************************************//** +At shutdown, frees the undo logs of a PREPARED transaction. */ +UNIV_INTERN +void +trx_undo_free_prepared( +/*===================*/ + trx_t* trx) /*!< in/out: PREPARED transaction */ + __attribute__((nonnull)); #endif /* !UNIV_HOTBACKUP */ /***********************************************************//** Parses the redo log entry of an undo log page initialization. diff --git a/storage/innodb_plugin/include/univ.i b/storage/innodb_plugin/include/univ.i index 319db93d137..8a4d8a55370 100644 --- a/storage/innodb_plugin/include/univ.i +++ b/storage/innodb_plugin/include/univ.i @@ -1,8 +1,7 @@ /***************************************************************************** -Copyright (c) 1994, 2010, Innobase Oy. All Rights Reserved. +Copyright (c) 1994, 2011, Oracle and/or its affiliates. All Rights Reserved. Copyright (c) 2008, Google Inc. -Copyright (c) 2009, Sun Microsystems, Inc. Portions of this file contain modifications contributed and copyrighted by Google, Inc. Those modifications are gratefully acknowledged and are described @@ -46,7 +45,7 @@ Created 1/20/1994 Heikki Tuuri #define INNODB_VERSION_MAJOR 1 #define INNODB_VERSION_MINOR 0 -#define INNODB_VERSION_BUGFIX 16 +#define INNODB_VERSION_BUGFIX 17 /* The following is the InnoDB version as shown in SELECT plugin_version FROM information_schema.plugins; |