Autocommit - Updating local tree

mysql-test/r/variables-big.result

@@ -1,3 +1,4 @@
+SET @def_var= @@session.transaction_prealloc_size;
 SET SESSION transaction_prealloc_size=1024*1024*1024*1;
 SHOW PROCESSLIST;
 Id	User	Host	db	Command	Time	State	Info
@@ -18,3 +19,4 @@ SET SESSION transaction_prealloc_size=1024*1024*1024*5;
 SHOW PROCESSLIST;
 Id	User	Host	db	Command	Time	State	Info
 <Id>	root	<Host>	test	Query	<Time>	NULL	SHOW PROCESSLIST
+SET @@session.transaction_prealloc_size= @def_var;

mysql-test/t/variables-big.test

@@ -34,6 +34,7 @@
 # There is a significant probablitity that this tests fails with testcase
 # timeout if the testing box is not powerful enough.
 #
+SET @def_var= @@session.transaction_prealloc_size;
 
 --disable_warnings
 SET SESSION transaction_prealloc_size=1024*1024*1024*1;
@@ -53,3 +54,5 @@ SET SESSION transaction_prealloc_size=1024*1024*1024*5;
 SHOW PROCESSLIST;
 --enable_warnings
 
+SET @@session.transaction_prealloc_size= @def_var;
+

storage/innobase/btr/btr0cur.c

@@ -3264,9 +3264,14 @@ btr_estimate_n_rows_in_range_on_level(
 
 		mtr_start(&mtr);
 
-		/* fetch the page */
-		block = buf_page_get(space, zip_size, page_no, RW_S_LATCH,
-				     &mtr);
+		/* Fetch the page. Because we are not holding the
+		index->lock, the tree may have changed and we may be
+		attempting to read a page that is no longer part of
+		the B-tree. We pass BUF_GET_POSSIBLY_FREED in order to
+		silence a debug assertion about this. */
+		block = buf_page_get_gen(space, zip_size, page_no, RW_S_LATCH,
+					 NULL, BUF_GET_POSSIBLY_FREED,
+					 __FILE__, __LINE__, &mtr);
 
 		page = buf_block_get_frame(block);
 
@@ -3285,6 +3290,13 @@ btr_estimate_n_rows_in_range_on_level(
 			goto inexact;
 		}
 
+		/* It is possible but highly unlikely that the page was
+		originally written by an old version of InnoDB that did
+		not initialize FIL_PAGE_TYPE on other than B-tree pages.
+		For example, this could be an almost-empty BLOB page
+		that happens to contain the magic values in the fields
+		that we checked above. */
+
 		n_pages_read++;
 
 		if (page_no != slot1->page_no) {

storage/innobase/buf/buf0buf.c

@@ -2747,6 +2747,7 @@ buf_page_get_gen(
 	case BUF_GET_IF_IN_POOL:
 	case BUF_PEEK_IF_IN_POOL:
 	case BUF_GET_IF_IN_POOL_OR_WATCH:
+	case BUF_GET_POSSIBLY_FREED:
 		break;
 	default:
 		ut_error;
@@ -3062,7 +3063,10 @@ wait_until_unfixed:
 #endif /* UNIV_DEBUG || UNIV_IBUF_DEBUG */
 
 	buf_block_buf_fix_inc(block, file, line);
-
+#if defined UNIV_DEBUG_FILE_ACCESSES || defined UNIV_DEBUG
+	ut_a(mode == BUF_GET_POSSIBLY_FREED
+	     || !block->page.file_page_was_freed);
+#endif
 	mutex_exit(&block->mutex);
 
 	/* Check if this is the first access to the page */
@@ -3075,10 +3079,6 @@ wait_until_unfixed:
 		buf_page_set_accessed_make_young(&block->page, access_time);
 	}
 
-#if defined UNIV_DEBUG_FILE_ACCESSES || defined UNIV_DEBUG
-	ut_a(!block->page.file_page_was_freed);
-#endif
-
 #if defined UNIV_DEBUG || defined UNIV_BUF_DEBUG
 	ut_a(++buf_dbg_counter % 5771 || buf_validate());
 	ut_a(block->page.buf_fix_count > 0);

storage/innobase/include/buf0buf.h

@@ -53,6 +53,9 @@ Created 11/5/1995 Heikki Tuuri
 					/*!< Get the page only if it's in the
 					buffer pool, if not then set a watch
 					on the page. */
+#define BUF_GET_POSSIBLY_FREED		16
+					/*!< Like BUF_GET, but do not mind
+					if the file page has been freed. */
 /* @} */
 /** @name Modes for buf_page_get_known_nowait */
 /* @{ */

storage/innobase/include/lock0lock.h

@@ -72,9 +72,10 @@ UNIV_INLINE
 trx_t*
 lock_clust_rec_some_has_impl(
 /*=========================*/
-	const rec_t*	rec,	/*!< in: user record */
-	dict_index_t*	index,	/*!< in: clustered index */
-	const ulint*	offsets);/*!< in: rec_get_offsets(rec, index) */
+	const rec_t*		rec,	/*!< in: user record */
+	const dict_index_t*	index,	/*!< in: clustered index */
+	const ulint*		offsets)/*!< in: rec_get_offsets(rec, index) */
+	__attribute__((nonnull, warn_unused_result));
 /*********************************************************************//**
 Gets the heap_no of the smallest user record on a page.
 @return	heap_no of smallest user record, or PAGE_HEAP_NO_SUPREMUM */

storage/innobase/include/lock0lock.ic

@@ -75,9 +75,9 @@ UNIV_INLINE
 trx_t*
 lock_clust_rec_some_has_impl(
 /*=========================*/
-	const rec_t*	rec,	/*!< in: user record */
-	dict_index_t*	index,	/*!< in: clustered index */
-	const ulint*	offsets)/*!< in: rec_get_offsets(rec, index) */
+	const rec_t*		rec,	/*!< in: user record */
+	const dict_index_t*	index,	/*!< in: clustered index */
+	const ulint*		offsets)/*!< in: rec_get_offsets(rec, index) */
 {
 	trx_id_t	trx_id;
 

storage/innobase/include/row0row.h

@@ -41,13 +41,24 @@ Created 4/20/1996 Heikki Tuuri
 Gets the offset of the 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) */
+row_get_trx_id_offset_func(
+/*=======================*/
+#ifdef UNIV_DEBUG
+	const rec_t*		rec,	/*!< in: record */
+#endif /* UNIV_DEBUG */
+	const dict_index_t*	index,	/*!< in: clustered index */
+	const ulint*		offsets)/*!< in: rec_get_offsets(rec, index) */
+	__attribute__((nonnull, warn_unused_result));
+#ifdef UNIV_DEBUG
+# define row_get_trx_id_offset(rec, index, offsets)	\
+	row_get_trx_id_offset_func(rec, index, offsets)
+#else /* UNIV_DEBUG */
+# define row_get_trx_id_offset(rec, index, offsets)	\
+	row_get_trx_id_offset_func(index, offsets)
+#endif /* UNIV_DEBUG */
+
 /*********************************************************************//**
 Reads the trx id field from a clustered index record.
 @return	value of the field */
@@ -55,9 +66,10 @@ 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 */

storage/innobase/include/row0row.ic

@@ -27,6 +27,36 @@ Created 4/20/1996 Heikki Tuuri
 #include "rem0rec.h"
 #include "trx0undo.h"
 
+/*********************************************************************//**
+Gets the offset of 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_func(
+/*=======================*/
+#ifdef UNIV_DEBUG
+	const rec_t*		rec,	/*!< in: record */
+#endif /* UNIV_DEBUG */
+	const dict_index_t*	index,	/*!< in: clustered index */
+	const ulint*		offsets)/*!< in: rec_get_offsets(rec, index) */
+{
+	ulint	pos;
+	ulint	offset;
+	ulint	len;
+
+	ut_ad(dict_index_is_clust(index));
+	ut_ad(rec_offs_validate(rec, 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 */
@@ -34,9 +64,9 @@ 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;
 

storage/innobase/lock/lock0lock.c

@@ -352,6 +352,7 @@ ibool
 lock_validate(void);
 /*===============*/
 
+# ifdef UNIV_DEBUG_LOCK_VALIDATE
 /*********************************************************************//**
 Validates the record lock queues on a page.
 @return	TRUE if ok */
@@ -359,10 +360,9 @@ static
 ibool
 lock_rec_validate_page(
 /*===================*/
-	ulint	space,	/*!< in: space id */
-	ulint	zip_size,/*!< in: compressed page size in bytes
-			or 0 for uncompressed pages */
-	ulint	page_no);/*!< in: page number */
+	const buf_block_t*	block)	/*!< in: buffer block */
+	__attribute__((nonnull, warn_unused_result));
+# endif /* UNIV_DEBUG_LOCK_VALIDATE */
 #endif /* UNIV_DEBUG */
 
 /* The lock system */
@@ -1100,10 +1100,10 @@ lock_rec_reset_nth_bit(
 Gets the first or next record lock on a page.
 @return	next lock, NULL if none exists */
 UNIV_INLINE
-lock_t*
-lock_rec_get_next_on_page(
-/*======================*/
-	lock_t*	lock)	/*!< in: a record lock */
+const lock_t*
+lock_rec_get_next_on_page_const(
+/*============================*/
+	const lock_t*	lock)	/*!< in: a record lock */
 {
 	ulint	space;
 	ulint	page_no;
@@ -1132,6 +1132,18 @@ lock_rec_get_next_on_page(
 	return(lock);
 }
 
+/*********************************************************************//**
+Gets the first or next record lock on a page.
+@return	next lock, NULL if none exists */
+UNIV_INLINE
+lock_t*
+lock_rec_get_next_on_page(
+/*======================*/
+	lock_t*	lock)	/*!< in: a record lock */
+{
+	return((lock_t*) lock_rec_get_next_on_page_const(lock));
+}
+
 /*********************************************************************//**
 Gets the first record lock on a page, where the page is identified by its
 file address.
@@ -2645,9 +2657,7 @@ lock_move_reorganize_page(
 	mem_heap_free(heap);
 
 #ifdef UNIV_DEBUG_LOCK_VALIDATE
-	ut_ad(lock_rec_validate_page(buf_block_get_space(block),
-				     buf_block_get_zip_size(block),
-				     buf_block_get_page_no(block)));
+	ut_ad(lock_rec_validate_page(block));
 #endif
 }
 
@@ -2735,12 +2745,8 @@ lock_move_rec_list_end(
 	lock_mutex_exit_kernel();
 
 #ifdef UNIV_DEBUG_LOCK_VALIDATE
-	ut_ad(lock_rec_validate_page(buf_block_get_space(block),
-				     buf_block_get_zip_size(block),
-				     buf_block_get_page_no(block)));
-	ut_ad(lock_rec_validate_page(buf_block_get_space(new_block),
-				     buf_block_get_zip_size(block),
-				     buf_block_get_page_no(new_block)));
+	ut_ad(lock_rec_validate_page(block));
+	ut_ad(lock_rec_validate_page(new_block));
 #endif
 }
 
@@ -2848,9 +2854,7 @@ lock_move_rec_list_start(
 	lock_mutex_exit_kernel();
 
 #ifdef UNIV_DEBUG_LOCK_VALIDATE
-	ut_ad(lock_rec_validate_page(buf_block_get_space(block),
-				     buf_block_get_zip_size(block),
-				     buf_block_get_page_no(block)));
+	ut_ad(lock_rec_validate_page(block));
 #endif
 }
 
@@ -3833,17 +3837,18 @@ Checks if other transactions have an incompatible mode lock request in
 the lock queue.
 @return	lock or NULL */
 UNIV_INLINE
-lock_t*
+const lock_t*
 lock_table_other_has_incompatible(
 /*==============================*/
-	trx_t*		trx,	/*!< in: transaction, or NULL if all
-				transactions should be included */
-	ulint		wait,	/*!< in: LOCK_WAIT if also waiting locks are
-				taken into account, or 0 if not */
-	dict_table_t*	table,	/*!< in: table */
-	enum lock_mode	mode)	/*!< in: lock mode */
+	const trx_t*		trx,	/*!< in: transaction, or NULL if all
+					transactions should be included */
+	ulint			wait,	/*!< in: LOCK_WAIT if also
+					waiting locks are taken into
+					account, or 0 if not */
+	const dict_table_t*	table,	/*!< in: table */
+	enum lock_mode		mode)	/*!< in: lock mode */
 {
-	lock_t*	lock;
+	const lock_t*	lock;
 
 	ut_ad(mutex_own(&kernel_mutex));
 
@@ -3934,10 +3939,10 @@ static
 ibool
 lock_table_has_to_wait_in_queue(
 /*============================*/
-	lock_t*	wait_lock)	/*!< in: waiting table lock */
+	const lock_t*	wait_lock)	/*!< in: waiting table lock */
 {
-	dict_table_t*	table;
-	lock_t*		lock;
+	const dict_table_t*	table;
+	const lock_t*		lock;
 
 	ut_ad(mutex_own(&kernel_mutex));
 	ut_ad(lock_get_wait(wait_lock));
@@ -4677,9 +4682,9 @@ static
 ibool
 lock_table_queue_validate(
 /*======================*/
-	dict_table_t*	table)	/*!< in: table */
+	const dict_table_t*	table)	/*!< in: table */
 {
-	lock_t*	lock;
+	const lock_t*	lock;
 
 	ut_ad(mutex_own(&kernel_mutex));
 
@@ -4715,7 +4720,7 @@ lock_rec_queue_validate(
 /*====================*/
 	const buf_block_t*	block,	/*!< in: buffer block containing rec */
 	const rec_t*		rec,	/*!< in: record to look at */
-	dict_index_t*		index,	/*!< in: index, or NULL if not known */
+	const dict_index_t*	index,	/*!< in: index, or NULL if not known */
 	const ulint*		offsets)/*!< in: rec_get_offsets(rec, index) */
 {
 	trx_t*	impl_trx;
@@ -4860,42 +4865,28 @@ lock_rec_queue_validate(
 /*********************************************************************//**
 Validates the record lock queues on a page.
 @return	TRUE if ok */
-static
+static __attribute__((nonnull, warn_unused_result))
 ibool
-lock_rec_validate_page(
-/*===================*/
-	ulint	space,	/*!< in: space id */
-	ulint	zip_size,/*!< in: compressed page size in bytes
-			or 0 for uncompressed pages */
-	ulint	page_no)/*!< in: page number */
+lock_rec_validate_page_low(
+/*=======================*/
+	const buf_block_t*	block)	/*!< in: buffer block */
 {
-	dict_index_t*	index;
-	buf_block_t*	block;
-	const page_t*	page;
-	lock_t*		lock;
+	const lock_t*	lock;
 	const rec_t*	rec;
 	ulint		nth_lock	= 0;
 	ulint		nth_bit		= 0;
 	ulint		i;
-	mtr_t		mtr;
 	mem_heap_t*	heap		= NULL;
 	ulint		offsets_[REC_OFFS_NORMAL_SIZE];
 	ulint*		offsets		= offsets_;
 	rec_offs_init(offsets_);
 
-	ut_ad(!mutex_own(&kernel_mutex));
+	ut_ad(mutex_own(&kernel_mutex));
+	ut_ad(buf_block_get_state(block) == BUF_BLOCK_FILE_PAGE);
 
-	mtr_start(&mtr);
-
-	ut_ad(zip_size != ULINT_UNDEFINED);
-	block = buf_page_get(space, zip_size, page_no, RW_X_LATCH, &mtr);
-	buf_block_dbg_add_level(block, SYNC_NO_ORDER_CHECK);
-
-	page = block->frame;
-
-	lock_mutex_enter_kernel();
 loop:
-	lock = lock_rec_get_first_on_page_addr(space, page_no);
+	lock = lock_rec_get_first_on_page_addr(buf_block_get_space(block),
+					       buf_block_get_page_no(block));
 
 	if (!lock) {
 		goto function_exit;
@@ -4903,7 +4894,7 @@ loop:
 
 	for (i = 0; i < nth_lock; i++) {
 
-		lock = lock_rec_get_next_on_page(lock);
+		lock = lock_rec_get_next_on_page_const(lock);
 
 		if (!lock) {
 			goto function_exit;
@@ -4926,15 +4917,14 @@ loop:
 
 		if (i == 1 || lock_rec_get_nth_bit(lock, i)) {
 
-			index = lock->index;
-			rec = page_find_rec_with_heap_no(page, i);
+			rec = page_find_rec_with_heap_no(block->frame, i);
 			ut_a(rec);
-			offsets = rec_get_offsets(rec, index, offsets,
+			offsets = rec_get_offsets(rec, lock->index, offsets,
 						  ULINT_UNDEFINED, &heap);
 #if 0
 			fprintf(stderr,
-				"Validating %lu %lu\n",
-				(ulong) space, (ulong) page_no);
+				"Validating %u %u\n",
+				block->page.space, block->page.offset);
 #endif
 			lock_mutex_exit_kernel();
 
@@ -4943,7 +4933,8 @@ loop:
 			check WILL break the latching order and may
 			cause a deadlock of threads. */
 
-			lock_rec_queue_validate(block, rec, index, offsets);
+			lock_rec_queue_validate(block, rec, lock->index,
+						offsets);
 
 			lock_mutex_enter_kernel();
 
@@ -4959,16 +4950,32 @@ loop:
 	goto loop;
 
 function_exit:
-	lock_mutex_exit_kernel();
-
-	mtr_commit(&mtr);
-
 	if (UNIV_LIKELY_NULL(heap)) {
 		mem_heap_free(heap);
 	}
 	return(TRUE);
 }
 
+#ifdef UNIV_DEBUG_LOCK_VALIDATE
+/*********************************************************************//**
+Validates the record lock queues on a page.
+@return	TRUE if ok */
+static
+ibool
+lock_rec_validate_page(
+/*===================*/
+	const buf_block_t*	block)	/*!< in: buffer block */
+{
+	ibool	valid;
+
+	lock_mutex_enter_kernel();
+	valid = lock_rec_validate_page_low(block);
+	lock_mutex_exit_kernel();
+
+	return(valid);
+}
+#endif /* UNIV_DEBUG_LOCK_VALIDATE */
+
 /*********************************************************************//**
 Validates the lock system.
 @return	TRUE if ok */
@@ -4977,11 +4984,8 @@ ibool
 lock_validate(void)
 /*===============*/
 {
-	lock_t*		lock;
-	trx_t*		trx;
-	ib_uint64_t	limit;
-	ulint		space;
-	ulint		page_no;
+	const lock_t*	lock;
+	const trx_t*	trx;
 	ulint		i;
 
 	lock_mutex_enter_kernel();
@@ -5006,9 +5010,14 @@ lock_validate(void)
 
 	for (i = 0; i < hash_get_n_cells(lock_sys->rec_hash); i++) {
 
-		limit = 0;
+		ulint		space;
+		ulint		page_no;
+		ib_uint64_t	limit	= 0;
 
 		for (;;) {
+			mtr_t		mtr;
+			buf_block_t*	block;
+
 			lock = HASH_GET_FIRST(lock_sys->rec_hash, i);
 
 			while (lock) {
@@ -5032,15 +5041,16 @@ lock_validate(void)
 				break;
 			}
 
-			lock_mutex_exit_kernel();
+			mtr_start(&mtr);
+			block = buf_page_get(
+				space, fil_space_get_zip_size(space),
+				page_no, RW_X_LATCH, &mtr);
+			buf_block_dbg_add_level(block, SYNC_NO_ORDER_CHECK);
 
-			lock_rec_validate_page(space,
-					       fil_space_get_zip_size(space),
-					       page_no);
+			ut_ad(lock_rec_validate_page_low(block));
+			mtr_commit(&mtr);
 
-			lock_mutex_enter_kernel();
-
-			limit = ut_ull_create(space, page_no + 1);
+			limit++;
 		}
 	}
 

storage/innobase/row/row0row.c

@@ -47,35 +47,6 @@ Created 4/20/1996 Heikki Tuuri
 #include "read0read.h"
 #include "ut0mem.h"
 
-/*********************************************************************//**
-Gets the offset of trx id field, in bytes relative to the origin of
-a clustered index record.
-@return	offset of DATA_TRX_ID */
-UNIV_INTERN
-ulint
-row_get_trx_id_offset(
-/*==================*/
-	const rec_t*	rec __attribute__((unused)),
-				/*!< in: record */
-	dict_index_t*	index,	/*!< in: clustered index */
-	const ulint*	offsets)/*!< in: rec_get_offsets(rec, index) */
-{
-	ulint	pos;
-	ulint	offset;
-	ulint	len;
-
-	ut_ad(dict_index_is_clust(index));
-	ut_ad(rec_offs_validate(rec, 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);
-}
-
 /*****************************************************************//**
 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.