Browse Source

Fix metadata area size calculation, print free space, wait for free space

FIXME: Now it crashes with -ENOSPC on linear overwrite
blocking-uring-test
Vitaliy Filippov 2 years ago
parent
commit
b6fff5a77e
  1. 11
      allocator.cpp
  2. 2
      allocator.h
  3. 8
      blockstore.cpp
  4. 2
      blockstore.h
  5. 10
      blockstore_init.cpp
  6. 4
      blockstore_init.h
  7. 2
      blockstore_open.cpp
  8. 10
      blockstore_write.cpp

11
allocator.cpp

@ -19,7 +19,7 @@ allocator::allocator(uint64_t blocks)
total -= p2;
total += (blocks+63) / 64;
mask = new uint64_t[2 + total];
size = blocks;
size = free = blocks;
last_one_mask = (blocks % 64) == 0
? UINT64_MAX
: ~(UINT64_MAX << (64 - blocks % 64));
@ -55,6 +55,10 @@ void allocator::set(uint64_t addr, bool value)
uint64_t bit = cur_addr % 64;
if (((mask[last] >> bit) & 1) != value64)
{
if (is_last)
{
free += value ? -1 : 1;
}
if (value)
{
mask[last] = mask[last] | (1l << bit);
@ -120,3 +124,8 @@ uint64_t allocator::find_free()
}
return addr;
}
uint64_t allocator::get_free_count()
{
return free;
}

2
allocator.h

@ -6,6 +6,7 @@
class allocator
{
uint64_t size;
uint64_t free;
uint64_t last_one_mask;
uint64_t *mask;
public:
@ -13,4 +14,5 @@ public:
~allocator();
void set(uint64_t addr, bool value);
uint64_t find_free();
uint64_t get_free_count();
};

8
blockstore.cpp

@ -260,6 +260,14 @@ void blockstore::check_wait(blockstore_operation *op)
}
op->wait_for = 0;
}
else if (op->wait_for == WAIT_FREE)
{
if (!data_alloc->get_free_count() && !flusher->is_active())
{
return;
}
op->wait_for = 0;
}
else
{
throw std::runtime_error("BUG: op->wait_for value is unexpected");

2
blockstore.h

@ -198,6 +198,8 @@ public:
#define WAIT_JOURNAL 3
// Suspend operation until the next journal sector buffer is free
#define WAIT_JOURNAL_BUFFER 4
// Suspend operation until there is some free space on the data device
#define WAIT_FREE 5
struct blockstore_operation
{

10
blockstore_init.cpp

@ -72,7 +72,7 @@ int blockstore_init_meta::loop()
}
}
// metadata read finished
printf("Metadata entries loaded: %d\n", entries_loaded);
printf("Metadata entries loaded: %lu, free blocks: %lu / %lu\n", entries_loaded, bs->data_alloc->get_free_count(), bs->block_count);
free(metadata_buffer);
metadata_buffer = NULL;
return 0;
@ -93,10 +93,9 @@ void blockstore_init_meta::handle_entries(struct clean_disk_entry* entries, int
if (clean_it != bs->clean_db.end())
{
// free the previous block
bs->data_alloc->set(clean_it->second.version >> block_order, false);
bs->data_alloc->set(clean_it->second.location >> block_order, false);
}
else
entries_loaded++;
entries_loaded++;
bs->data_alloc->set(done_cnt+i, true);
bs->clean_db[entries[i].oid] = (struct clean_entry){
.version = entries[i].version,
@ -287,7 +286,7 @@ resume_1:
}
}
}
printf("Journal entries loaded: %d\n", entries_loaded);
printf("Journal entries loaded: %lu, free blocks: %lu / %lu\n", entries_loaded, bs->data_alloc->get_free_count(), bs->block_count);
if (!bs->journal.inmemory)
{
free(journal_buffer);
@ -397,6 +396,7 @@ int blockstore_init_journal::handle_journal_part(void *buf, uint64_t len)
.len = bs->block_size,
.journal_sector = proc_pos,
});
bs->data_alloc->set(je->big_write.location >> bs->block_order, true);
bs->journal.used_sectors[proc_pos]++;
auto & unstab = bs->unstable_writes[ov.oid];
unstab = unstab < ov.version ? ov.version : unstab;

4
blockstore_init.h

@ -7,7 +7,7 @@ class blockstore_init_meta
uint8_t *metadata_buffer = NULL;
uint64_t metadata_read = 0;
int prev = 0, prev_done = 0, done_len = 0, submitted = 0, done_cnt = 0;
int entries_loaded = 0;
uint64_t entries_loaded = 0;
struct io_uring_sqe *sqe;
struct ring_data_t *data;
void handle_entries(struct clean_disk_entry* entries, int count, int block_order);
@ -21,7 +21,7 @@ class blockstore_init_journal
{
blockstore *bs;
int wait_state = 0, wait_count = 0;
int entries_loaded = 0;
uint64_t entries_loaded = 0;
void *journal_buffer = NULL;
uint32_t crc32_last = 0;
bool started = false;

2
blockstore_open.cpp

@ -37,7 +37,7 @@ void blockstore::calc_lengths(blockstore_config_t & config)
}
// required metadata size
block_count = data_len / block_size;
meta_len = (block_count / (512 / sizeof(clean_disk_entry))) * 512;
meta_len = ((block_count - 1 + 512 / sizeof(clean_disk_entry)) / (512 / sizeof(clean_disk_entry))) * 512;
if (meta_area < meta_len)
{
throw std::runtime_error("Metadata area is too small");

10
blockstore_write.cpp

@ -49,7 +49,6 @@ void blockstore::enqueue_write(blockstore_operation *op)
// First step of the write algorithm: dequeue operation and submit initial write(s)
int blockstore::dequeue_write(blockstore_operation *op)
{
auto cb = [this, op](ring_data_t *data) { handle_write_event(data, op); };
auto dirty_it = dirty_db.find((obj_ver_id){
.oid = op->oid,
.version = op->version,
@ -61,6 +60,12 @@ int blockstore::dequeue_write(blockstore_operation *op)
if (loc == UINT64_MAX)
{
// no space
if (flusher->is_active())
{
// hope that some space will be available after flush
op->wait_for = WAIT_FREE;
return 0;
}
op->retval = -ENOSPC;
op->callback(op);
return 1;
@ -87,7 +92,7 @@ int blockstore::dequeue_write(blockstore_operation *op)
op->iov_zerofill[0] = (struct iovec){ op->buf, op->len };
data->iov.iov_len = op->len; // to check it in the callback
}
data->callback = cb;
data->callback = [this, op](ring_data_t *data) { handle_write_event(data, op); };
my_uring_prep_writev(
sqe, data_fd, op->iov_zerofill, vcnt, data_offset + (loc << block_order)
);
@ -134,6 +139,7 @@ int blockstore::dequeue_write(blockstore_operation *op)
je->crc32_data = crc32c(0, op->buf, op->len);
je->crc32 = je_crc32((journal_entry*)je);
journal.crc32_last = je->crc32;
auto cb = [this, op](ring_data_t *data) { handle_write_event(data, op); };
prepare_journal_sector_write(journal, sqe1, cb);
op->min_used_journal_sector = op->max_used_journal_sector = 1 + journal.cur_sector;
// Prepare journal data write

Loading…
Cancel
Save