mirror/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2026-05-06 23:25:34 +02:00
Code Issues Projects Releases Packages Wiki Activity

MDEV-12353: Replace MLOG_*LIST_*_DELETE and MLOG_*REC_DELETE

Browse source 
No longer write the following redo log records:
MLOG_COMP_LIST_END_DELETE, MLOG_LIST_END_DELETE,
MLOG_COMP_LIST_START_DELETE, MLOG_LIST_START_DELETE,
MLOG_REC_DELETE,MLOG_COMP_REC_DELETE.

Each individual deleted record will be logged separately
using physical log records.

page_dir_slot_set_n_owned(),
page_zip_rec_set_owned(), page_zip_dir_delete(), page_zip_clear_rec():
Add the parameter mtr, and write redo log.

page_dir_slot_set_rec(): Remove. Replaced with lower-level operations
that write redo log when necessary.

page_rec_set_n_owned(): Replaces rec_set_n_owned_old(),
rec_set_n_owned_new().

rec_set_heap_no(): Replaces rec_set_heap_no_old(), rec_set_heap_no_new().

page_mem_free(), page_dir_split_slot(), page_dir_balance_slot():
Add the parameter mtr.

page_dir_set_n_slots(): Merge with the caller page_dir_split_slot().

page_dir_slot_set_rec(): Merge with the callers page_dir_split_slot()
and page_dir_balance_slot().

page_cur_insert_rec_low(), page_cur_insert_rec_zip():
Suppress the logging of lower-level operations.

page_cur_delete_rec_write_log(): Remove.

page_cur_delete_rec(): Do not tolerate mtr=NULL.

rec_convert_dtuple_to_rec_old(), rec_convert_dtuple_to_rec_comp():
Replace rec_set_heap_no_old() and rec_set_heap_no_new() with direct
access that does not involve redo logging.

mtr_t::memcpy(): Do allow non-redo-logged writes to uncompressed pages
of ROW_FORMAT=COMPRESSED pages.

buf_page_io_complete(): Evict the uncompressed page of
a ROW_FORMAT=COMPRESSED page after recovery. Because we no longer
write logical log records for deleting index records, but instead
write physical records that may refer directly to the compressed
page frame of a ROW_FORMAT=COMPRESSED page, and because on recovery
we will only apply the changes to the ROW_FORMAT=COMPRESSED page,
the uncompressed page frame can be stale until page_zip_decompress()
is executed.

recv_parse_or_apply_log_rec_body(): After applying MLOG_ZIP_WRITE_STRING,
ensure that the FIL_PAGE_TYPE of the uncompressed page matches the
compressed page, because buf_flush_init_for_writing() assumes that
field to be valid.

mlog_init_t::mark_ibuf_exist(): Invoke page_zip_decompress(), because
the uncompressed page after buf_page_create() is not necessarily
up to date.

buf_LRU_block_remove_hashed(): Bypass a page_zip_validate() check
during redo log apply.

recv_apply_hashed_log_recs(): Invoke mlog_init.mark_ibuf_exist()
also for the last batch, to ensure that page_zip_decompress() will
be called for freshly initialized pages.
This commit is contained in:
Marko Mäkelä 2020-02-13 11:41:54 +02:00
commit 2a77b2a510
18 changed files with 615 additions and 561 deletions
Download patch file Download diff file Expand all files Collapse all files

58
storage/innobase/include/rem0rec.ic
View file

@ -503,19 +503,6 @@ rec_get_n_owned_old(
REC_N_OWNED_MASK, REC_N_OWNED_SHIFT));
}
/******************************************************//**
The following function is used to set the number of owned records. */
UNIV_INLINE
void
rec_set_n_owned_old(
/*================*/
rec_t* rec, /*!< in: old-style physical record */
ulint n_owned) /*!< in: the number of owned */
{
rec_set_bit_field_1(rec, n_owned, REC_OLD_N_OWNED,
REC_N_OWNED_MASK, REC_N_OWNED_SHIFT);
}
/******************************************************//**
The following function is used to get the number of records owned by the
previous directory record.
@ -530,23 +517,6 @@ rec_get_n_owned_new(
REC_N_OWNED_MASK, REC_N_OWNED_SHIFT));
}
/******************************************************//**
The following function is used to set the number of owned records. */
UNIV_INLINE
void
rec_set_n_owned_new(
/*================*/
rec_t* rec, /*!< in/out: new-style physical record */
page_zip_des_t* page_zip,/*!< in/out: compressed page, or NULL */
ulint n_owned)/*!< in: the number of owned */
{
rec_set_bit_field_1(rec, n_owned, REC_NEW_N_OWNED,
REC_N_OWNED_MASK, REC_N_OWNED_SHIFT);
if (page_zip && rec_get_status(rec) != REC_STATUS_SUPREMUM) {
page_zip_rec_set_owned(page_zip, rec, n_owned);
}
}
/******************************************************//**
The following function is used to retrieve the info bits of a record.
@return info bits */
@ -674,20 +644,6 @@ rec_get_heap_no_old(
REC_HEAP_NO_MASK, REC_HEAP_NO_SHIFT));
}
/******************************************************//**
The following function is used to set the heap number
field in an old-style record. */
UNIV_INLINE
void
rec_set_heap_no_old(
/*================*/
rec_t* rec, /*!< in: physical record */
ulint heap_no)/*!< in: the heap number */
{
rec_set_bit_field_2(rec, heap_no, REC_OLD_HEAP_NO,
REC_HEAP_NO_MASK, REC_HEAP_NO_SHIFT);
}
/******************************************************//**
The following function is used to get the order number
of a new-style record in the heap of the index page.
@ -702,20 +658,6 @@ rec_get_heap_no_new(
REC_HEAP_NO_MASK, REC_HEAP_NO_SHIFT));
}
/******************************************************//**
The following function is used to set the heap number
field in a new-style record. */
UNIV_INLINE
void
rec_set_heap_no_new(
/*================*/
rec_t* rec, /*!< in/out: physical record */
ulint heap_no)/*!< in: the heap number */
{
rec_set_bit_field_2(rec, heap_no, REC_NEW_HEAP_NO,
REC_HEAP_NO_MASK, REC_HEAP_NO_SHIFT);
}
/******************************************************//**
The following function is used to test whether the data offsets in the record
are stored in one-byte or two-byte format.