static void _logger_log_evictions(logentry *e, item *it) {
struct logentry_eviction *le = (struct logentry_eviction *) e->data;
le->exptime = (it->exptime > 0) ? (long long int)(it->exptime - current_time) : (long long int) -1;
le->latime = current_time - it->time;
le->it_flags = it->it_flags;
le->nkey = it->nkey;
le->clsid = ITEM_clsid(it);
memcpy(le->key, ITEM_key(it), it->nkey);
e->size = sizeof(struct logentry_eviction) + le->nkey;
}
/* TODO: When more logging endpoints are done and the extstore API has matured
* more, this could be merged with above and print different types of
* expulsion events.
*/
static void _logger_log_ext_write(logentry *e, item *it, uint8_t bucket) {
struct logentry_ext_write *le = (struct logentry_ext_write *) e->data;
le->exptime = (it->exptime > 0) ? (long long int)(it->exptime - current_time) : (long long int) -1;
le->latime = current_time - it->time;
le->it_flags = it->it_flags;
le->nkey = it->nkey;
le->clsid = ITEM_clsid(it);
le->bucket = bucket;
memcpy(le->key, ITEM_key(it), it->nkey);
e->size = sizeof(struct logentry_ext_write) + le->nkey;
}
void item_free(item *it) {
size_t ntotal = ITEM_ntotal(it);
unsigned int clsid;
assert((it->it_flags & ITEM_LINKED) == 0);
assert(it != heads[it->slabs_clsid]);
assert(it != tails[it->slabs_clsid]);
assert(it->refcount == 0);
/* so slab size changer can tell later if item is already free or not */
clsid = ITEM_clsid(it);
DEBUG_REFCNT(it, 'F');
slabs_free(it, ntotal, clsid);
}
/* Returns number of items remove, expired, or evicted.
* Callable from worker threads or the LRU maintainer thread */
static int lru_pull_tail(const int orig_id, const int cur_lru,
const uint64_t total_bytes, uint8_t flags) {
item *it = NULL;
int id = orig_id;
int removed = 0;
if (id == 0)
return 0;
int tries = 5;
item *search;
item *next_it;
void *hold_lock = NULL;
unsigned int move_to_lru = 0;
uint64_t limit = 0;
id |= cur_lru;
pthread_mutex_lock(&lru_locks[id]);
search = tails[id];
/* We walk up *only* for locked items, and if bottom is expired. */
for (; tries > 0 && search != NULL; tries--, search=next_it) {
/* we might relink search mid-loop, so search->prev isn't reliable */
next_it = search->prev;
if (search->nbytes == 0 && search->nkey == 0 && search->it_flags == 1) {
/* We are a crawler, ignore it. */
if (flags & LRU_PULL_CRAWL_BLOCKS) {
pthread_mutex_unlock(&lru_locks[id]);
return 0;
}
tries++;
continue;
}
uint32_t hv = hash(ITEM_key(search), search->nkey);
/* Attempt to hash item lock the "search" item. If locked, no
* other callers can incr the refcount. Also skip ourselves. */
if ((hold_lock = item_trylock(hv)) == NULL)
continue;
/* Now see if the item is refcount locked */
if (refcount_incr(&search->refcount) != 2) {
/* Note pathological case with ref'ed items in tail.
* Can still unlink the item, but it won't be reusable yet */
itemstats[id].lrutail_reflocked++;
/* In case of refcount leaks, enable for quick workaround. */
/* WARNING: This can cause terrible corruption */
if (settings.tail_repair_time &&
search->time + settings.tail_repair_time < current_time) {
itemstats[id].tailrepairs++;
search->refcount = 1;
/* This will call item_remove -> item_free since refcnt is 1 */
do_item_unlink_nolock(search, hv);
item_trylock_unlock(hold_lock);
continue;
}
}
/* Expired or flushed */
if ((search->exptime != 0 && search->exptime < current_time)
|| item_is_flushed(search)) {
itemstats[id].reclaimed++;
if ((search->it_flags & ITEM_FETCHED) == 0) {
itemstats[id].expired_unfetched++;
}
/* refcnt 2 -> 1 */
do_item_unlink_nolock(search, hv);
/* refcnt 1 -> 0 -> item_free */
do_item_remove(search);
item_trylock_unlock(hold_lock);
removed++;
/* If all we're finding are expired, can keep going */
continue;
}
/* If we're HOT_LRU or WARM_LRU and over size limit, send to COLD_LRU.
* If we're COLD_LRU, send to WARM_LRU unless we need to evict
*/
switch (cur_lru) {
case HOT_LRU:
limit = total_bytes * settings.hot_lru_pct / 100;
case WARM_LRU:
if (limit == 0)
limit = total_bytes * settings.warm_lru_pct / 100;
if (sizes_bytes[id] > limit) {
itemstats[id].moves_to_cold++;
move_to_lru = COLD_LRU;
do_item_unlink_q(search);
it = search;
removed++;
break;
} else if ((search->it_flags & ITEM_ACTIVE) != 0) {
/* Only allow ACTIVE relinking if we're not too large. */
itemstats[id].moves_within_lru++;
search->it_flags &= ~ITEM_ACTIVE;
do_item_update_nolock(search);
do_item_remove(search);
item_trylock_unlock(hold_lock);
} else {
/* Don't want to move to COLD, not active, bail out */
it = search;
}
break;
case COLD_LRU:
it = search; /* No matter what, we're stopping */
if (flags & LRU_PULL_EVICT) {
if (settings.evict_to_free == 0) {
/* Don't think we need a counter for this. It'll OOM. */
break;
}
itemstats[id].evicted++;
itemstats[id].evicted_time = current_time - search->time;
if (search->exptime != 0)
itemstats[id].evicted_nonzero++;
if ((search->it_flags & ITEM_FETCHED) == 0) {
itemstats[id].evicted_unfetched++;
}
LOGGER_LOG(NULL, LOG_EVICTIONS, LOGGER_EVICTION, search);
do_item_unlink_nolock(search, hv);
removed++;
if (settings.slab_automove == 2) {
slabs_reassign(-1, orig_id);
}
} else if ((search->it_flags & ITEM_ACTIVE) != 0
&& settings.lru_maintainer_thread) {
itemstats[id].moves_to_warm++;
search->it_flags &= ~ITEM_ACTIVE;
move_to_lru = WARM_LRU;
do_item_unlink_q(search);
removed++;
}
break;
}
if (it != NULL)
break;
}
pthread_mutex_unlock(&lru_locks[id]);
if (it != NULL) {
if (move_to_lru) {
it->slabs_clsid = ITEM_clsid(it);
it->slabs_clsid |= move_to_lru;
item_link_q(it);
}
do_item_remove(it);
item_trylock_unlock(hold_lock);
}
return removed;
}
static int storage_write(void *storage, const int clsid, const int item_age) {
int did_moves = 0;
struct lru_pull_tail_return it_info;
it_info.it = NULL;
lru_pull_tail(clsid, COLD_LRU, 0, LRU_PULL_RETURN_ITEM, 0, &it_info);
/* Item is locked, and we have a reference to it. */
if (it_info.it == NULL) {
return did_moves;
}
obj_io io;
item *it = it_info.it;
/* First, storage for the header object */
size_t orig_ntotal = ITEM_ntotal(it);
uint32_t flags;
if ((it->it_flags & ITEM_HDR) == 0 &&
(item_age == 0 || current_time - it->time > item_age)) {
FLAGS_CONV(it, flags);
item *hdr_it = do_item_alloc(ITEM_key(it), it->nkey, flags, it->exptime, sizeof(item_hdr));
/* Run the storage write understanding the start of the item is dirty.
* We will fill it (time/exptime/etc) from the header item on read.
*/
if (hdr_it != NULL) {
int bucket = (it->it_flags & ITEM_CHUNKED) ?
PAGE_BUCKET_CHUNKED : PAGE_BUCKET_DEFAULT;
// Compress soon to expire items into similar pages.
if (it->exptime - current_time < settings.ext_low_ttl) {
bucket = PAGE_BUCKET_LOWTTL;
}
hdr_it->it_flags |= ITEM_HDR;
io.len = orig_ntotal;
io.mode = OBJ_IO_WRITE;
// NOTE: when the item is read back in, the slab mover
// may see it. Important to have refcount>=2 or ~ITEM_LINKED
assert(it->refcount >= 2);
// NOTE: write bucket vs free page bucket will disambiguate once
// lowttl feature is better understood.
if (extstore_write_request(storage, bucket, bucket, &io) == 0) {
// cuddle the hash value into the time field so we don't have
// to recalculate it.
item *buf_it = (item *) io.buf;
buf_it->time = it_info.hv;
// copy from past the headers + time headers.
// TODO: should be in items.c
if (it->it_flags & ITEM_CHUNKED) {
// Need to loop through the item and copy
item_chunk *sch = (item_chunk *) ITEM_schunk(it);
int remain = orig_ntotal;
int copied = 0;
// copy original header
int hdrtotal = ITEM_ntotal(it) - it->nbytes;
memcpy((char *)io.buf+STORE_OFFSET, (char *)it+STORE_OFFSET, hdrtotal - STORE_OFFSET);
copied = hdrtotal;
// copy data in like it were one large object.
while (sch && remain) {
assert(remain >= sch->used);
memcpy((char *)io.buf+copied, sch->data, sch->used);
// FIXME: use one variable?
remain -= sch->used;
copied += sch->used;
sch = sch->next;
}
} else {
memcpy((char *)io.buf+STORE_OFFSET, (char *)it+STORE_OFFSET, io.len-STORE_OFFSET);
}
// crc what we copied so we can do it sequentially.
buf_it->it_flags &= ~ITEM_LINKED;
buf_it->exptime = crc32c(0, (char*)io.buf+STORE_OFFSET, orig_ntotal-STORE_OFFSET);
extstore_write(storage, &io);
item_hdr *hdr = (item_hdr *) ITEM_data(hdr_it);
hdr->page_version = io.page_version;
hdr->page_id = io.page_id;
hdr->offset = io.offset;
// overload nbytes for the header it
hdr_it->nbytes = it->nbytes;
/* success! Now we need to fill relevant data into the new
* header and replace. Most of this requires the item lock
*/
/* CAS gets set while linking. Copy post-replace */
item_replace(it, hdr_it, it_info.hv);
ITEM_set_cas(hdr_it, ITEM_get_cas(it));
do_item_remove(hdr_it);
did_moves = 1;
LOGGER_LOG(NULL, LOG_EVICTIONS, LOGGER_EXTSTORE_WRITE, it, bucket);
} else {
/* Failed to write for some reason, can't continue. */
slabs_free(hdr_it, ITEM_ntotal(hdr_it), ITEM_clsid(hdr_it));
}
}
}
do_item_remove(it);
item_unlock(it_info.hv);
return did_moves;
}
