Fix multiple-sector journal writes, add assertions to not miss any SQEs

Vitaliy Filippov 2021-02-02 01:29:11 +03:00
parent e66ed47515
commit 9b5d8b9ad4
6 changed files with 49 additions and 66 deletions

View File

@ -25,6 +25,10 @@ int blockstore_journal_check_t::check_available(blockstore_op_t *op, int entries
: (bs->journal.block_size - next_in_pos) / size;
if (fits > 0)
{
if (fits > required)
{
fits = required;
}
if (first_sector == -1)
{
first_sector = next_sector;
@ -116,12 +120,10 @@ int blockstore_journal_check_t::check_available(blockstore_op_t *op, int entries
journal_entry* prefill_single_journal_entry(journal_t & journal, uint16_t type, uint32_t size)
{
if (journal.block_size - journal.in_sector_pos < size ||
journal.no_same_sector_overwrites && journal.sector_info[journal.cur_sector].written)
if (!journal.entry_fits(size))
{
assert(!journal.sector_info[journal.cur_sector].dirty);
// Move to the next journal sector
journal.sector_info[journal.cur_sector].written = false;
if (journal.sector_info[journal.cur_sector].usage_count > 0)
{
// Also select next sector buffer in memory
@ -132,6 +134,7 @@ journal_entry* prefill_single_journal_entry(journal_t & journal, uint16_t type,
{
journal.dirty_start = journal.next_free;
}
journal.sector_info[journal.cur_sector].written = false;
journal.sector_info[journal.cur_sector].offset = journal.next_free;
journal.in_sector_pos = 0;
journal.next_free = (journal.next_free+journal.block_size) < journal.len ? journal.next_free + journal.block_size : journal.block_size;

View File

@ -133,7 +133,7 @@ inline uint32_t je_crc32(journal_entry *je)
struct journal_sector_info_t
{
uint64_t offset;
uint64_t usage_count;
uint64_t usage_count; // flusher_count!
bool written;
bool dirty;
};
@ -170,13 +170,18 @@ struct journal_t
~journal_t();
bool trim();
uint64_t get_trim_pos();
inline bool entry_fits(int size)
{
return !(block_size - in_sector_pos < size ||
no_same_sector_overwrites && sector_info[cur_sector].written);
}
};
struct blockstore_journal_check_t
{
blockstore_impl_t *bs;
uint64_t next_pos, next_sector, next_in_pos;
int sectors_required, first_sector;
int sectors_required, first_sector; // "sectors to write"
bool right_dir; // writing to the end or the beginning of the ring buffer
blockstore_journal_check_t(blockstore_impl_t *bs);

View File

@ -83,35 +83,27 @@ skip_ov:
// Prepare and submit journal entries
auto cb = [this, op](ring_data_t *data) { handle_rollback_event(data, op); };
int s = 0, cur_sector = -1;
if ((journal_block_size - journal.in_sector_pos) < sizeof(journal_entry_rollback) &&
journal.sector_info[journal.cur_sector].dirty)
{
PRIV(op)->min_flushed_journal_sector = 1 + journal.cur_sector;
prepare_journal_sector_write(journal, journal.cur_sector, sqe[s++], cb);
cur_sector = ((journal.cur_sector + 1) % journal.sector_count);
}
for (i = 0, v = (obj_ver_id*)op->buf; i < op->len; i++, v++)
{
if (!journal.entry_fits(sizeof(journal_entry_rollback)) &&
journal.sector_info[journal.cur_sector].dirty)
{
if (cur_sector == -1)
PRIV(op)->min_flushed_journal_sector = 1 + journal.cur_sector;
prepare_journal_sector_write(journal, journal.cur_sector, sqe[s++], cb);
cur_sector = journal.cur_sector;
}
journal_entry_rollback *je = (journal_entry_rollback*)
prefill_single_journal_entry(journal, JE_ROLLBACK, sizeof(journal_entry_rollback));
journal.sector_info[journal.cur_sector].dirty = false;
je->oid = v->oid;
je->version = v->version;
je->crc32 = je_crc32((journal_entry*)je);
journal.crc32_last = je->crc32;
if (cur_sector != journal.cur_sector)
{
// Write previous sector. We should write the sector only after filling it,
// because otherwise we'll write a lot more sectors in the "no_same_sector_overwrite" mode
if (cur_sector != -1)
prepare_journal_sector_write(journal, cur_sector, sqe[s++], cb);
else
}
prepare_journal_sector_write(journal, journal.cur_sector, sqe[s++], cb);
assert(s == space_check.sectors_required);
if (cur_sector == -1)
PRIV(op)->min_flushed_journal_sector = 1 + journal.cur_sector;
cur_sector = journal.cur_sector;
}
}
if (cur_sector != -1)
prepare_journal_sector_write(journal, cur_sector, sqe[s++], cb);
PRIV(op)->max_flushed_journal_sector = 1 + journal.cur_sector;
PRIV(op)->pending_ops = s;
PRIV(op)->op_state = 1;

View File

@ -106,36 +106,28 @@ int blockstore_impl_t::dequeue_stable(blockstore_op_t *op)
// Prepare and submit journal entries
auto cb = [this, op](ring_data_t *data) { handle_stable_event(data, op); };
int s = 0, cur_sector = -1;
if ((journal_block_size - journal.in_sector_pos) < sizeof(journal_entry_stable) &&
journal.sector_info[journal.cur_sector].dirty)
{
PRIV(op)->min_flushed_journal_sector = 1 + journal.cur_sector;
prepare_journal_sector_write(journal, journal.cur_sector, sqe[s++], cb);
cur_sector = ((journal.cur_sector + 1) % journal.sector_count);
}
for (i = 0, v = (obj_ver_id*)op->buf; i < op->len; i++, v++)
{
// FIXME: Only stabilize versions that aren't stable yet
if (!journal.entry_fits(sizeof(journal_entry_stable)) &&
journal.sector_info[journal.cur_sector].dirty)
{
if (cur_sector == -1)
PRIV(op)->min_flushed_journal_sector = 1 + journal.cur_sector;
prepare_journal_sector_write(journal, journal.cur_sector, sqe[s++], cb);
cur_sector = journal.cur_sector;
}
journal_entry_stable *je = (journal_entry_stable*)
prefill_single_journal_entry(journal, JE_STABLE, sizeof(journal_entry_stable));
journal.sector_info[journal.cur_sector].dirty = false;
je->oid = v->oid;
je->version = v->version;
je->crc32 = je_crc32((journal_entry*)je);
journal.crc32_last = je->crc32;
if (cur_sector != journal.cur_sector)
{
// Write previous sector. We should write the sector only after filling it,
// because otherwise we'll write a lot more sectors in the "no_same_sector_overwrite" mode
if (cur_sector != -1)
prepare_journal_sector_write(journal, cur_sector, sqe[s++], cb);
else
}
prepare_journal_sector_write(journal, journal.cur_sector, sqe[s++], cb);
assert(s == space_check.sectors_required);
if (cur_sector == -1)
PRIV(op)->min_flushed_journal_sector = 1 + journal.cur_sector;
cur_sector = journal.cur_sector;
}
}
if (cur_sector != -1)
prepare_journal_sector_write(journal, cur_sector, sqe[s++], cb);
PRIV(op)->max_flushed_journal_sector = 1 + journal.cur_sector;
PRIV(op)->pending_ops = s;
PRIV(op)->op_state = 1;

View File

@ -120,21 +120,21 @@ int blockstore_impl_t::continue_sync(blockstore_op_t *op)
// Prepare and submit journal entries
auto it = PRIV(op)->sync_big_writes.begin();
int s = 0, cur_sector = -1;
if ((journal_block_size - journal.in_sector_pos) < sizeof(journal_entry_big_write) &&
journal.sector_info[journal.cur_sector].dirty)
{
PRIV(op)->min_flushed_journal_sector = 1 + journal.cur_sector;
prepare_journal_sector_write(journal, journal.cur_sector, sqe[s++], cb);
cur_sector = ((journal.cur_sector + 1) % journal.sector_count);
}
while (it != PRIV(op)->sync_big_writes.end())
{
if (!journal.entry_fits(sizeof(journal_entry_big_write)) &&
journal.sector_info[journal.cur_sector].dirty)
{
if (cur_sector == -1)
PRIV(op)->min_flushed_journal_sector = 1 + journal.cur_sector;
prepare_journal_sector_write(journal, journal.cur_sector, sqe[s++], cb);
cur_sector = journal.cur_sector;
}
journal_entry_big_write *je = (journal_entry_big_write*)prefill_single_journal_entry(
journal, (dirty_db[*it].state & BS_ST_INSTANT) ? JE_BIG_WRITE_INSTANT : JE_BIG_WRITE,
sizeof(journal_entry_big_write)
);
dirty_db[*it].journal_sector = journal.sector_info[journal.cur_sector].offset;
journal.sector_info[journal.cur_sector].dirty = false;
journal.used_sectors[journal.sector_info[journal.cur_sector].offset]++;
#ifdef BLOCKSTORE_DEBUG
printf(
@ -151,19 +151,11 @@ int blockstore_impl_t::continue_sync(blockstore_op_t *op)
je->crc32 = je_crc32((journal_entry*)je);
journal.crc32_last = je->crc32;
it++;
if (cur_sector != journal.cur_sector)
{
// Write the previous sector. We should write the sector only after filling it,
// because otherwise we'll write a lot more sectors in the "no_same_sector_overwrite" mode
if (cur_sector != -1)
prepare_journal_sector_write(journal, cur_sector, sqe[s++], cb);
else
}
prepare_journal_sector_write(journal, journal.cur_sector, sqe[s++], cb);
assert(s == space_check.sectors_required);
if (cur_sector == -1)
PRIV(op)->min_flushed_journal_sector = 1 + journal.cur_sector;
cur_sector = journal.cur_sector;
}
}
if (cur_sector != -1)
prepare_journal_sector_write(journal, cur_sector, sqe[s++], cb);
PRIV(op)->max_flushed_journal_sector = 1 + journal.cur_sector;
PRIV(op)->pending_ops = s;
PRIV(op)->op_state = SYNC_JOURNAL_WRITE_SENT;

View File

@ -377,7 +377,6 @@ resume_2:
sizeof(journal_entry_big_write)
);
dirty_it->second.journal_sector = journal.sector_info[journal.cur_sector].offset;
journal.sector_info[journal.cur_sector].dirty = false;
journal.used_sectors[journal.sector_info[journal.cur_sector].offset]++;
#ifdef BLOCKSTORE_DEBUG
printf(