Commit e56854d8 authored by Inaam Rana's avatar Inaam Rana

Bug#13789853 SHOW ENGINE INNODB STATUS HANGS DUE TO EXCESSIVE WORK

IN LOCK_VALIDATE()

rb://917
approved by: Marko Makela

In lock_validate() the limit is used to release the kernel_mutex during
the validation, to obey the latching order.
If we do the limit++ then we are rechecking the same lock most times on
each iteration because limit is being incremented by one and
<space, page_no> will nearly always be > limit. If we set the limit
correctly to (space, page+1) then we are actually making progress
during the iteration.
parent a54e6e00
......@@ -4964,6 +4964,74 @@ function_exit:
return(TRUE);
}
/*********************************************************************//**
Validate record locks up to a limit.
@return lock at limit or NULL if no more locks in the hash bucket */
static __attribute__((nonnull, warn_unused_result))
const lock_t*
lock_rec_validate(
/*==============*/
ulint start, /*!< in: lock_sys->rec_hash
bucket */
ib_uint64_t* limit) /*!< in/out: upper limit of
(space, page_no) */
{
lock_t* lock;
ut_ad(mutex_own(&kernel_mutex));
for (lock = HASH_GET_FIRST(lock_sys->rec_hash, start);
lock != NULL;
lock = HASH_GET_NEXT(hash, lock)) {
ib_uint64_t current;
ut_a(trx_in_trx_list(lock->trx));
ut_a(lock_get_type(lock) == LOCK_REC);
current = ut_ull_create(
lock->un_member.rec_lock.space,
lock->un_member.rec_lock.page_no);
if (current > *limit) {
*limit = current + 1;
return(lock);
}
}
return(NULL);
}
/*********************************************************************//**
Validate a record lock's block */
static
void
lock_rec_block_validate(
/*====================*/
ulint space,
ulint page_no)
{
/* The lock and the block that it is referring to may be freed at
this point. We pass BUF_GET_POSSIBLY_FREED to skip a debug check.
If the lock exists in lock_rec_validate_page() we assert
!block->page.file_page_was_freed. */
mtr_t mtr;
buf_block_t* block;
mtr_start(&mtr);
block = buf_page_get_gen(
space, fil_space_get_zip_size(space),
page_no, RW_X_LATCH, NULL,
BUF_GET_POSSIBLY_FREED,
__FILE__, __LINE__, &mtr);
buf_block_dbg_add_level(block, SYNC_NO_ORDER_CHECK);
ut_ad(lock_rec_validate_page(block));
mtr_commit(&mtr);
}
/*********************************************************************//**
Validates the lock system.
@return TRUE if ok */
......@@ -4996,60 +5064,21 @@ lock_validate(void)
trx = UT_LIST_GET_NEXT(trx_list, trx);
}
for (i = 0; i < hash_get_n_cells(lock_sys->rec_hash); i++) {
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) {
ib_uint64_t space_page;
ut_a(trx_in_trx_list(lock->trx));
space = lock->un_member.rec_lock.space;
page_no = lock->un_member.rec_lock.page_no;
/* Iterate over all the record locks and validate the locks. We
don't want to hog the lock_sys_t::mutex and the trx_sys_t::mutex.
Release both mutexes during the validation check. */
space_page = ut_ull_create(space, page_no);
if (space_page >= limit) {
break;
}
lock = HASH_GET_NEXT(hash, lock);
}
for (i = 0; i < hash_get_n_cells(lock_sys->rec_hash); i++) {
const lock_t* lock;
ib_uint64_t limit = 0;
if (!lock) {
while ((lock = lock_rec_validate(i, &limit)) != NULL) {
break;
}
ulint space = lock->un_member.rec_lock.space;
ulint page_no = lock->un_member.rec_lock.page_no;
lock_mutex_exit_kernel();
/* The lock and the block that it is referring
to may be freed at this point. We pass
BUF_GET_POSSIBLY_FREED to skip a debug check.
If the lock exists in lock_rec_validate_page()
we assert !block->page.file_page_was_freed. */
mtr_start(&mtr);
block = buf_page_get_gen(
space, fil_space_get_zip_size(space),
page_no, RW_X_LATCH, NULL,
BUF_GET_POSSIBLY_FREED,
__FILE__, __LINE__, &mtr);
buf_block_dbg_add_level(block, SYNC_NO_ORDER_CHECK);
ut_ad(lock_rec_validate_page(block));
mtr_commit(&mtr);
limit++;
lock_rec_block_validate(space, page_no);
lock_mutex_enter_kernel();
}
}
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment