mariadb/storage/innobase/fut/fut0lst.c
aivanov@mysql.com d33b523271 Applied innodb-5.1-ss269 snapshot.
Fixed BUGS:
  #3300: "UPDATE statement with no index column in where condition locks
    all rows"
    Implement semi-consistent read to reduce lock conflicts at the cost
    of breaking serializability.
    ha_innobase::unlock_row(): reset the "did semi consistent read" flag
    ha_innobase::was_semi_consistent_read(),
    ha_innobase::try_semi_consistent_read(): new methods
    row_prebuilt_t, row_create_prebuilt(): add field row_read_type for
    keeping track of semi-consistent reads
    row_vers_build_for_semi_consistent_read(),
    row_sel_build_committed_vers_for_mysql(): new functions
    row_search_for_mysql(): implement semi-consistent reads

  #9802: "Foreign key checks disallow alter table".
    Added test cases.

  #12456: "Cursor shows incorrect data - DML does not affect,
    probably caching"
    This patch implements a high-granularity read view to be used with
    cursors. In this high-granularity consistent read view modifications 
    done by the creating transaction after the cursor is created or 
   future transactions are not visible. But those modifications that 
   transaction did before the cursor was created are visible.

  #12701: "Support >4GB buffer pool and log files on 64-bit Windows"
    Do not call os_file_create_tmpfile() at runtime. Instead, create all
    tempfiles at startup and guard access to them with mutexes.

  #13778: "If FOREIGN_KEY_CHECKS=0, one can create inconsistent FOREIGN KEYs".
    When FOREIGN_KEY_CHECKS=0 we still need to check that datatypes between
    foreign key references are compatible.

  #14189: "VARBINARY and BINARY variables: trailing space ignored with InnoDB"
    innobase_init(): Assert that
    DATA_MYSQL_BINARY_CHARSET_COLL == my_charset_bin.number.
    dtype_get_pad_char(): Do not pad VARBINARY or BINARY columns.
    row_ins_cascade_calc_update_vec(): Refuse ON UPDATE CASCADE when trying
    to change the length of a VARBINARY column that refers to or is referenced
    by a BINARY column. BINARY columns are no longer padded on comparison,
    and thus they cannot be padded on storage either.

  #14747: "Race condition can cause btr_search_drop_page_hash_index() to crash"
    Note that buf_block_t::index should be protected by btr_search_latch
    or an s-latch or x-latch on the index page.
    btr_search_drop_page_hash_index(): Read block->index while holding
    btr_search_latch and use the cached value in the loop.  Remove some
    redundant assertions.

  #15108: "mysqld crashes when innodb_log_file_size is set > 4G"

  #15308: "Problem of Order with Enum Column in Primary Key"

  #15550: "mysqld crashes in printing a FOREIGN KEY error in InnoDB"
    row_ins_foreign_report_add_err(): When printing the parent record,
    use the index in the parent table rather than the index in the child table.

  #15653: "Slow inserts to InnoDB if many thousands of .ibd files"
    Keep track on unflushed modifications to file spaces.  When there are tens
    of thousands of file spaces, flushing all files in fil_flush_file_spaces()
    would be very slow.
    fil_flush_file_spaces(): Only flush unflushed file spaces.
    fil_space_t, fil_system_t: Add a list of unflushed spaces.

  #15991: "innodb-file-per-table + symlink database + rename = cr"
   os_file_handle_error(): Map the error codes EXDEV, ENOTDIR, and EISDIR
   to the new code OS_FILE_PATH_ERROR. Treat this code as OS_FILE_PATH_ERROR.
   This fixes the crash on RENAME TABLE when the .ibd file is a symbolic link
   to a different file system.

  #16157: "InnoDB crashes when main location settings are empty"
    This patch is from Heikki.

  #16298: "InnoDB segfaults in INSERTs in upgrade of 4.0 -> 5.0 tables
    with VARCHAR BINARY"
    dict_load_columns(): Set the charset-collation code
    DATA_MYSQL_BINARY_CHARSET_COLL for those binary string columns
    that lack a charset-collation code, i.e., the tables were created
    with an older version of MySQL/InnoDB than 4.1.2.

  #16229: "MySQL/InnoDB uses full explicit table locks in trigger processing"
    Take a InnoDB table lock only if user has explicitly requested a table
    lock. Added some additional comments to store_lock() and external_lock().

  #16387: "InnoDB crash when dropping a foreign key <table>_ibfk_0"
    Do not mistake TABLENAME_ibfk_0 for auto-generated id.
    dict_table_get_highest_foreign_id(): Ignore foreign constraint
    identifiers starting with the pattern TABLENAME_ibfk_0.

  #16582: "InnoDB: Error in an adaptive hash index pointer to page"
    Account for a race condition when dropping the adaptive hash index
    for a B-tree page.
    btr_search_drop_page_hash_index(): Retry the operation if a hash index
    with different parameters was built meanwhile.  Add diagnostics for the
    case that hash node pointers to the page remain.
    btr_search_info_update_hash(), btr_search_info_update_slow():
    Document the parameter "info" as in/out.

  #16814: "SHOW INNODB STATUS format error in LATEST FOREIGN KEY ERROR
    section"
    Add a missing newline to the LAST FOREIGN KEY ERROR section in SHOW
    INNODB STATUS output.
    dict_foreign_error_report(): Always print a newline after invoking
    dict_print_info_on_foreign_key_in_create_format().

  #16827: "Better InnoDB error message if ibdata files omitted from my.cnf"

  #17126: "CHECK TABLE on InnoDB causes a short hang during check of adaptive
    hash"
    CHECK TABLE blocking other queries, by releasing the btr_search_latch
    periodically during the adaptive hash table validation.

  #17405: "Valgrind: conditional jump or move depends on unititialised values"
    buf_block_init(): Reset magic_n, buf_fix_count and io_fix to avoid
    testing uninitialized variables.
2006-03-10 19:22:21 +03:00

518 lines
14 KiB
C

/**********************************************************************
File-based list utilities
(c) 1995 Innobase Oy
Created 11/28/1995 Heikki Tuuri
***********************************************************************/
#include "fut0lst.h"
#ifdef UNIV_NONINL
#include "fut0lst.ic"
#endif
#include "buf0buf.h"
/************************************************************************
Adds a node to an empty list. */
static
void
flst_add_to_empty(
/*==============*/
flst_base_node_t* base, /* in: pointer to base node of
empty list */
flst_node_t* node, /* in: node to add */
mtr_t* mtr) /* in: mini-transaction handle */
{
ulint space;
fil_addr_t node_addr;
ulint len;
ut_ad(mtr && base && node);
ut_ad(base != node);
ut_ad(mtr_memo_contains(mtr, buf_block_align(base),
MTR_MEMO_PAGE_X_FIX));
ut_ad(mtr_memo_contains(mtr, buf_block_align(node),
MTR_MEMO_PAGE_X_FIX));
len = flst_get_len(base, mtr);
ut_a(len == 0);
buf_ptr_get_fsp_addr(node, &space, &node_addr);
/* Update first and last fields of base node */
flst_write_addr(base + FLST_FIRST, node_addr, mtr);
flst_write_addr(base + FLST_LAST, node_addr, mtr);
/* Set prev and next fields of node to add */
flst_write_addr(node + FLST_PREV, fil_addr_null, mtr);
flst_write_addr(node + FLST_NEXT, fil_addr_null, mtr);
/* Update len of base node */
mlog_write_ulint(base + FLST_LEN, len + 1, MLOG_4BYTES, mtr);
}
/************************************************************************
Adds a node as the last node in a list. */
void
flst_add_last(
/*==========*/
flst_base_node_t* base, /* in: pointer to base node of list */
flst_node_t* node, /* in: node to add */
mtr_t* mtr) /* in: mini-transaction handle */
{
ulint space;
fil_addr_t node_addr;
ulint len;
fil_addr_t last_addr;
flst_node_t* last_node;
ut_ad(mtr && base && node);
ut_ad(base != node);
ut_ad(mtr_memo_contains(mtr, buf_block_align(base),
MTR_MEMO_PAGE_X_FIX));
ut_ad(mtr_memo_contains(mtr, buf_block_align(node),
MTR_MEMO_PAGE_X_FIX));
len = flst_get_len(base, mtr);
last_addr = flst_get_last(base, mtr);
buf_ptr_get_fsp_addr(node, &space, &node_addr);
/* If the list is not empty, call flst_insert_after */
if (len != 0) {
if (last_addr.page == node_addr.page) {
last_node = buf_frame_align(node) + last_addr.boffset;
} else {
last_node = fut_get_ptr(space, last_addr, RW_X_LATCH,
mtr);
}
flst_insert_after(base, last_node, node, mtr);
} else {
/* else call flst_add_to_empty */
flst_add_to_empty(base, node, mtr);
}
}
/************************************************************************
Adds a node as the first node in a list. */
void
flst_add_first(
/*===========*/
flst_base_node_t* base, /* in: pointer to base node of list */
flst_node_t* node, /* in: node to add */
mtr_t* mtr) /* in: mini-transaction handle */
{
ulint space;
fil_addr_t node_addr;
ulint len;
fil_addr_t first_addr;
flst_node_t* first_node;
ut_ad(mtr && base && node);
ut_ad(base != node);
ut_ad(mtr_memo_contains(mtr, buf_block_align(base),
MTR_MEMO_PAGE_X_FIX));
ut_ad(mtr_memo_contains(mtr, buf_block_align(node),
MTR_MEMO_PAGE_X_FIX));
len = flst_get_len(base, mtr);
first_addr = flst_get_first(base, mtr);
buf_ptr_get_fsp_addr(node, &space, &node_addr);
/* If the list is not empty, call flst_insert_before */
if (len != 0) {
if (first_addr.page == node_addr.page) {
first_node = buf_frame_align(node)
+ first_addr.boffset;
} else {
first_node = fut_get_ptr(space, first_addr,
RW_X_LATCH, mtr);
}
flst_insert_before(base, node, first_node, mtr);
} else {
/* else call flst_add_to_empty */
flst_add_to_empty(base, node, mtr);
}
}
/************************************************************************
Inserts a node after another in a list. */
void
flst_insert_after(
/*==============*/
flst_base_node_t* base, /* in: pointer to base node of list */
flst_node_t* node1, /* in: node to insert after */
flst_node_t* node2, /* in: node to add */
mtr_t* mtr) /* in: mini-transaction handle */
{
ulint space;
fil_addr_t node1_addr;
fil_addr_t node2_addr;
flst_node_t* node3;
fil_addr_t node3_addr;
ulint len;
ut_ad(mtr && node1 && node2 && base);
ut_ad(base != node1);
ut_ad(base != node2);
ut_ad(node2 != node1);
ut_ad(mtr_memo_contains(mtr, buf_block_align(base),
MTR_MEMO_PAGE_X_FIX));
ut_ad(mtr_memo_contains(mtr, buf_block_align(node1),
MTR_MEMO_PAGE_X_FIX));
ut_ad(mtr_memo_contains(mtr, buf_block_align(node2),
MTR_MEMO_PAGE_X_FIX));
buf_ptr_get_fsp_addr(node1, &space, &node1_addr);
buf_ptr_get_fsp_addr(node2, &space, &node2_addr);
node3_addr = flst_get_next_addr(node1, mtr);
/* Set prev and next fields of node2 */
flst_write_addr(node2 + FLST_PREV, node1_addr, mtr);
flst_write_addr(node2 + FLST_NEXT, node3_addr, mtr);
if (!fil_addr_is_null(node3_addr)) {
/* Update prev field of node3 */
node3 = fut_get_ptr(space, node3_addr, RW_X_LATCH, mtr);
flst_write_addr(node3 + FLST_PREV, node2_addr, mtr);
} else {
/* node1 was last in list: update last field in base */
flst_write_addr(base + FLST_LAST, node2_addr, mtr);
}
/* Set next field of node1 */
flst_write_addr(node1 + FLST_NEXT, node2_addr, mtr);
/* Update len of base node */
len = flst_get_len(base, mtr);
mlog_write_ulint(base + FLST_LEN, len + 1, MLOG_4BYTES, mtr);
}
/************************************************************************
Inserts a node before another in a list. */
void
flst_insert_before(
/*===============*/
flst_base_node_t* base, /* in: pointer to base node of list */
flst_node_t* node2, /* in: node to insert */
flst_node_t* node3, /* in: node to insert before */
mtr_t* mtr) /* in: mini-transaction handle */
{
ulint space;
flst_node_t* node1;
fil_addr_t node1_addr;
fil_addr_t node2_addr;
fil_addr_t node3_addr;
ulint len;
ut_ad(mtr && node2 && node3 && base);
ut_ad(base != node2);
ut_ad(base != node3);
ut_ad(node2 != node3);
ut_ad(mtr_memo_contains(mtr, buf_block_align(base),
MTR_MEMO_PAGE_X_FIX));
ut_ad(mtr_memo_contains(mtr, buf_block_align(node2),
MTR_MEMO_PAGE_X_FIX));
ut_ad(mtr_memo_contains(mtr, buf_block_align(node3),
MTR_MEMO_PAGE_X_FIX));
buf_ptr_get_fsp_addr(node2, &space, &node2_addr);
buf_ptr_get_fsp_addr(node3, &space, &node3_addr);
node1_addr = flst_get_prev_addr(node3, mtr);
/* Set prev and next fields of node2 */
flst_write_addr(node2 + FLST_PREV, node1_addr, mtr);
flst_write_addr(node2 + FLST_NEXT, node3_addr, mtr);
if (!fil_addr_is_null(node1_addr)) {
/* Update next field of node1 */
node1 = fut_get_ptr(space, node1_addr, RW_X_LATCH, mtr);
flst_write_addr(node1 + FLST_NEXT, node2_addr, mtr);
} else {
/* node3 was first in list: update first field in base */
flst_write_addr(base + FLST_FIRST, node2_addr, mtr);
}
/* Set prev field of node3 */
flst_write_addr(node3 + FLST_PREV, node2_addr, mtr);
/* Update len of base node */
len = flst_get_len(base, mtr);
mlog_write_ulint(base + FLST_LEN, len + 1, MLOG_4BYTES, mtr);
}
/************************************************************************
Removes a node. */
void
flst_remove(
/*========*/
flst_base_node_t* base, /* in: pointer to base node of list */
flst_node_t* node2, /* in: node to remove */
mtr_t* mtr) /* in: mini-transaction handle */
{
ulint space;
flst_node_t* node1;
fil_addr_t node1_addr;
fil_addr_t node2_addr;
flst_node_t* node3;
fil_addr_t node3_addr;
ulint len;
ut_ad(mtr && node2 && base);
ut_ad(mtr_memo_contains(mtr, buf_block_align(base),
MTR_MEMO_PAGE_X_FIX));
ut_ad(mtr_memo_contains(mtr, buf_block_align(node2),
MTR_MEMO_PAGE_X_FIX));
buf_ptr_get_fsp_addr(node2, &space, &node2_addr);
node1_addr = flst_get_prev_addr(node2, mtr);
node3_addr = flst_get_next_addr(node2, mtr);
if (!fil_addr_is_null(node1_addr)) {
/* Update next field of node1 */
if (node1_addr.page == node2_addr.page) {
node1 = buf_frame_align(node2) + node1_addr.boffset;
} else {
node1 = fut_get_ptr(space, node1_addr, RW_X_LATCH,
mtr);
}
ut_ad(node1 != node2);
flst_write_addr(node1 + FLST_NEXT, node3_addr, mtr);
} else {
/* node2 was first in list: update first field in base */
flst_write_addr(base + FLST_FIRST, node3_addr, mtr);
}
if (!fil_addr_is_null(node3_addr)) {
/* Update prev field of node3 */
if (node3_addr.page == node2_addr.page) {
node3 = buf_frame_align(node2) + node3_addr.boffset;
} else {
node3 = fut_get_ptr(space, node3_addr, RW_X_LATCH,
mtr);
}
ut_ad(node2 != node3);
flst_write_addr(node3 + FLST_PREV, node1_addr, mtr);
} else {
/* node2 was last in list: update last field in base */
flst_write_addr(base + FLST_LAST, node1_addr, mtr);
}
/* Update len of base node */
len = flst_get_len(base, mtr);
ut_ad(len > 0);
mlog_write_ulint(base + FLST_LEN, len - 1, MLOG_4BYTES, mtr);
}
/************************************************************************
Cuts off the tail of the list, including the node given. The number of
nodes which will be removed must be provided by the caller, as this function
does not measure the length of the tail. */
void
flst_cut_end(
/*=========*/
flst_base_node_t* base, /* in: pointer to base node of list */
flst_node_t* node2, /* in: first node to remove */
ulint n_nodes,/* in: number of nodes to remove,
must be >= 1 */
mtr_t* mtr) /* in: mini-transaction handle */
{
ulint space;
flst_node_t* node1;
fil_addr_t node1_addr;
fil_addr_t node2_addr;
ulint len;
ut_ad(mtr && node2 && base);
ut_ad(mtr_memo_contains(mtr, buf_block_align(base),
MTR_MEMO_PAGE_X_FIX));
ut_ad(mtr_memo_contains(mtr, buf_block_align(node2),
MTR_MEMO_PAGE_X_FIX));
ut_ad(n_nodes > 0);
buf_ptr_get_fsp_addr(node2, &space, &node2_addr);
node1_addr = flst_get_prev_addr(node2, mtr);
if (!fil_addr_is_null(node1_addr)) {
/* Update next field of node1 */
if (node1_addr.page == node2_addr.page) {
node1 = buf_frame_align(node2) + node1_addr.boffset;
} else {
node1 = fut_get_ptr(space, node1_addr, RW_X_LATCH,
mtr);
}
flst_write_addr(node1 + FLST_NEXT, fil_addr_null, mtr);
} else {
/* node2 was first in list: update the field in base */
flst_write_addr(base + FLST_FIRST, fil_addr_null, mtr);
}
flst_write_addr(base + FLST_LAST, node1_addr, mtr);
/* Update len of base node */
len = flst_get_len(base, mtr);
ut_ad(len >= n_nodes);
mlog_write_ulint(base + FLST_LEN, len - n_nodes, MLOG_4BYTES, mtr);
}
/************************************************************************
Cuts off the tail of the list, not including the given node. The number of
nodes which will be removed must be provided by the caller, as this function
does not measure the length of the tail. */
void
flst_truncate_end(
/*==============*/
flst_base_node_t* base, /* in: pointer to base node of list */
flst_node_t* node2, /* in: first node not to remove */
ulint n_nodes,/* in: number of nodes to remove */
mtr_t* mtr) /* in: mini-transaction handle */
{
fil_addr_t node2_addr;
ulint len;
ulint space;
ut_ad(mtr && node2 && base);
ut_ad(mtr_memo_contains(mtr, buf_block_align(base),
MTR_MEMO_PAGE_X_FIX));
ut_ad(mtr_memo_contains(mtr, buf_block_align(node2),
MTR_MEMO_PAGE_X_FIX));
if (n_nodes == 0) {
ut_ad(fil_addr_is_null(flst_get_next_addr(node2, mtr)));
return;
}
buf_ptr_get_fsp_addr(node2, &space, &node2_addr);
/* Update next field of node2 */
flst_write_addr(node2 + FLST_NEXT, fil_addr_null, mtr);
flst_write_addr(base + FLST_LAST, node2_addr, mtr);
/* Update len of base node */
len = flst_get_len(base, mtr);
ut_ad(len >= n_nodes);
mlog_write_ulint(base + FLST_LEN, len - n_nodes, MLOG_4BYTES, mtr);
}
/************************************************************************
Validates a file-based list. */
ibool
flst_validate(
/*==========*/
/* out: TRUE if ok */
flst_base_node_t* base, /* in: pointer to base node of list */
mtr_t* mtr1) /* in: mtr */
{
ulint space;
flst_node_t* node;
fil_addr_t node_addr;
fil_addr_t base_addr;
ulint len;
ulint i;
mtr_t mtr2;
ut_ad(base);
ut_ad(mtr_memo_contains(mtr1, buf_block_align(base),
MTR_MEMO_PAGE_X_FIX));
/* We use two mini-transaction handles: the first is used to
lock the base node, and prevent other threads from modifying the
list. The second is used to traverse the list. We cannot run the
second mtr without committing it at times, because if the list
is long, then the x-locked pages could fill the buffer resulting
in a deadlock. */
/* Find out the space id */
buf_ptr_get_fsp_addr(base, &space, &base_addr);
len = flst_get_len(base, mtr1);
node_addr = flst_get_first(base, mtr1);
for (i = 0; i < len; i++) {
mtr_start(&mtr2);
node = fut_get_ptr(space, node_addr, RW_X_LATCH, &mtr2);
node_addr = flst_get_next_addr(node, &mtr2);
mtr_commit(&mtr2); /* Commit mtr2 each round to prevent buffer
becoming full */
}
ut_a(fil_addr_is_null(node_addr));
node_addr = flst_get_last(base, mtr1);
for (i = 0; i < len; i++) {
mtr_start(&mtr2);
node = fut_get_ptr(space, node_addr, RW_X_LATCH, &mtr2);
node_addr = flst_get_prev_addr(node, &mtr2);
mtr_commit(&mtr2); /* Commit mtr2 each round to prevent buffer
becoming full */
}
ut_a(fil_addr_is_null(node_addr));
return(TRUE);
}
/************************************************************************
Prints info of a file-based list. */
void
flst_print(
/*=======*/
flst_base_node_t* base, /* in: pointer to base node of list */
mtr_t* mtr) /* in: mtr */
{
buf_frame_t* frame;
ulint len;
ut_ad(base && mtr);
ut_ad(mtr_memo_contains(mtr, buf_block_align(base),
MTR_MEMO_PAGE_X_FIX));
frame = buf_frame_align(base);
len = flst_get_len(base, mtr);
fprintf(stderr,
"FILE-BASED LIST:\n"
"Base node in space %lu page %lu byte offset %lu; len %lu\n",
(ulong) buf_frame_get_space_id(frame),
(ulong) buf_frame_get_page_no(frame),
(ulong) (base - frame), (ulong) len);
}