/* expires items that are more recent than the oldest_live setting. */
void do_item_flush_expired(void) {
int i;
item *iter, *next;
if (settings.oldest_live == 0)
return;
for (i = 0; i < LARGEST_ID; i++) {
/* The LRU is sorted in decreasing time order, and an item's timestamp
* is never newer than its last access time, so we only need to walk
* back until we hit an item older than the oldest_live time.
* The oldest_live checking will auto-expire the remaining items.
*/
for (iter = heads[i]; iter != NULL; iter = next) {
/* iter->time of 0 are magic objects. */
if (iter->time != 0 && iter->time >= settings.oldest_live) {
next = iter->next;
if ((iter->it_flags & ITEM_SLABBED) == 0) {
do_item_unlink_nolock(iter, hash(ITEM_key(iter), iter->nkey));
}
} else {
/* We've hit the first old item. Continue to the next queue. */
break;
}
}
}
}
/* I pulled this out to make the main thread clearer, but it reaches into the
* main thread's values too much. Should rethink again.
*/
static void item_crawler_evaluate(item *search, uint32_t hv, int i) {
rel_time_t oldest_live = settings.oldest_live;
if ((search->exptime != 0 && search->exptime < current_time)
|| (search->time <= oldest_live && oldest_live <= current_time)) {
itemstats[i].crawler_reclaimed++;
if (settings.verbose > 1) {
int ii;
char *key = ITEM_key(search);
fprintf(stderr, "LRU crawler found an expired item (flags: %d, slab: %d): ",
search->it_flags, search->slabs_clsid);
for (ii = 0; ii < search->nkey; ++ii) {
fprintf(stderr, "%c", key[ii]);
}
fprintf(stderr, "\n");
}
if ((search->it_flags & ITEM_FETCHED) == 0) {
itemstats[i].expired_unfetched++;
}
do_item_unlink_nolock(search, hv);
do_item_remove(search);
assert(search->slabs_clsid == 0);
} else {
refcount_decr(&search->refcount);
}
}
enum store_item_type do_store_item(item *it, const uint32_t hv) {
enum store_item_type stored;
char *key = ITEM_key(it);
item *old_it = do_item_get(key, it->nkey, hv);
if (old_it != NULL) {
assert(old_it != it);
before_write(old_it);
do_item_unlink_nolock(old_it, hv);
//do_item_remove(old_it); /* release our reference */
item_free(old_it);
assert((old_it->it_flags & ITEM_LINKED) == 0);
assert((old_it->it_flags & ITEM_SLABBED) != 0);
after_write(old_it);
}
before_write(it);
int ret = do_item_link_nolock(it, hv);
after_write(it);
if (ret == 0) {
stored = NOT_STORED;
printf("not stored\n");
} else {
stored = STORED;
}
return stored;
}
/** wrapper around assoc_find which does the lazy expiration logic */
item *do_item_get(const char *key, const size_t nkey, const uint32_t hv) {
mutex_lock(&cache_lock);
item *it = assoc_find(key, nkey, hv);
if (it != NULL) {
refcount_incr(&it->refcount);
/* Optimization for slab reassignment. prevents popular items from
* jamming in busy wait. Can only do this here to satisfy lock order
* of item_lock, cache_lock, slabs_lock. */
if (slab_rebalance_signal &&
((void *)it >= slab_rebal.slab_start && (void *)it < slab_rebal.slab_end)) {
do_item_unlink_nolock(it, hv);
do_item_remove(it);
it = NULL;
}
}
pthread_mutex_unlock(&cache_lock);
int was_found = 0;
if (settings.verbose > 2) {
if (it == NULL) {
fprintf(stderr, "> NOT FOUND %s", key);
} else {
fprintf(stderr, "> FOUND KEY %s", ITEM_key(it));
was_found++;
}
}
if (it != NULL) {
if (settings.oldest_live != 0 && settings.oldest_live <= current_time &&
it->time <= settings.oldest_live) {
do_item_unlink(it, hv);
do_item_remove(it);
it = NULL;
if (was_found) {
fprintf(stderr, " -nuked by flush");
}
} else if (it->exptime != 0 && it->exptime <= current_time) {
if (it->exptime + 10 < current_time) {
do_item_unlink(it, hv);
do_item_remove(it);
if (was_found) {
fprintf(stderr, " -nuked by expire");
}
} else {
/* re-active just expired items, to anti miss-storm */
it->exptime = current_time + 10;
}
it = NULL;
} else {
it->it_flags |= ITEM_FETCHED;
DEBUG_REFCNT(it, '+');
}
}
if (settings.verbose > 2)
fprintf(stderr, "\n");
return it;
}
/* refcount == 0 is safe since nobody can incr while cache_lock is held.
* refcount != 0 is impossible since flags/etc can be modified in other
* threads. instead, note we found a busy one and bail. logic in do_item_get
* will prevent busy items from continuing to be busy
*/
static int slab_rebalance_move(void) {
slabclass_t *s_cls;
int x;
int was_busy = 0;
int refcount = 0;
enum move_status status = MOVE_PASS;
pthread_mutex_lock(&cache_lock);
pthread_mutex_lock(&slabs_lock);
s_cls = &slabclass[slab_rebal.s_clsid];
for (x = 0; x < slab_bulk_check; x++) {
item *it = slab_rebal.slab_pos;
status = MOVE_PASS;
if (it->slabs_clsid != 255) {
void *hold_lock = NULL;
uint32_t hv = hash(ITEM_key(it), it->nkey, 0);
if ((hold_lock = item_trylock(hv)) == NULL) {
status = MOVE_LOCKED;
} else {
refcount = refcount_incr(&it->refcount);
if (refcount == 1) { /* item is unlinked, unused */
if (it->it_flags & ITEM_SLABBED) {
/* remove from slab freelist */
if (s_cls->slots == it) {
s_cls->slots = it->next;
}
if (it->next) it->next->prev = it->prev;
if (it->prev) it->prev->next = it->next;
s_cls->sl_curr--;
status = MOVE_DONE;
} else {
status = MOVE_BUSY;
}
} else if (refcount == 2) { /* item is linked but not busy */
if ((it->it_flags & ITEM_LINKED) != 0) {
do_item_unlink_nolock(it, hv);
status = MOVE_DONE;
} else {
/* refcount == 1 + !ITEM_LINKED means the item is being
* uploaded to, or was just unlinked but hasn't been freed
* yet. Let it bleed off on its own and try again later */
status = MOVE_BUSY;
}
} else {
if (settings.verbose > 2) {
fprintf(stderr, "Slab reassign hit a busy item: refcount: %d (%d -> %d)\n",
it->refcount, slab_rebal.s_clsid, slab_rebal.d_clsid);
}
status = MOVE_BUSY;
}
item_trylock_unlock(hold_lock);
}
}
switch (status) {
case MOVE_DONE:
it->refcount = 0;
it->it_flags = 0;
it->slabs_clsid = 255;
break;
case MOVE_BUSY:
refcount_decr(&it->refcount);
case MOVE_LOCKED:
slab_rebal.busy_items++;
was_busy++;
break;
case MOVE_PASS:
break;
}
slab_rebal.slab_pos = (char *)slab_rebal.slab_pos + s_cls->size;
if (slab_rebal.slab_pos >= slab_rebal.slab_end)
break;
}
if (slab_rebal.slab_pos >= slab_rebal.slab_end) {
/* Some items were busy, start again from the top */
if (slab_rebal.busy_items) {
slab_rebal.slab_pos = slab_rebal.slab_start;
slab_rebal.busy_items = 0;
} else {
slab_rebal.done++;
}
}
pthread_mutex_unlock(&slabs_lock);
pthread_mutex_unlock(&cache_lock);
return was_busy;
}
/* 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;
}
/* 从 slab 系统分配一个空闲 item */
item *do_item_alloc(char *key, const size_t nkey, const int flags, const rel_time_t exptime, const int nbytes,
const uint32_t cur_hv)
{
uint8_t nsuffix;
item *it = NULL;
char suffix[40];
size_t ntotal = item_make_header(nkey + 1, flags, nbytes, suffix, &nsuffix);
if (settings.use_cas) {
ntotal += sizeof(uint64_t);
}
unsigned int id = slabs_clsid(ntotal);
if (id == 0)
return 0;
mutex_lock(&cache_lock);
/* do a quick check if we have any expired items in the tail.. */
int tries = 5;
int tried_alloc = 0;
item *search;
void *hold_lock = NULL;
rel_time_t oldest_live = settings.oldest_live;
search = tails[id];
/* We walk up *only* for locked items. Never searching for expired.
* Waste of CPU for almost all deployments */
for (; tries > 0 && search != NULL; tries--, search=search->prev)
{
uint32_t hv = hash(ITEM_key(search), search->nkey, 0);
/* Attempt to hash item lock the "search" item. If locked, no
* other callers can incr the refcount
*/
/* FIXME: I think we need to mask the hv here for comparison? */
if (hv != cur_hv && (hold_lock = item_trylock(hv)) == NULL)
continue;
/* Now see if the item is refcount locked */
if (refcount_incr(&search->refcount) != 2)
{
refcount_decr(&search->refcount);
/* Old rare bug could cause a refcount leak. We haven't seen
* it in years, but we leave this code in to prevent failures
* just in case */
if (search->time + TAIL_REPAIR_TIME < current_time)
{
itemstats[id].tailrepairs++;
search->refcount = 1;
do_item_unlink_nolock(search, hv);
}
if (hold_lock)
item_trylock_unlock(hold_lock);
continue;
}
/* 先检查 LRU 队列最后一个 item 是否超时, 超时的话就把这个 item 分配给用户 */
if ((search->exptime != 0 && search->exptime < current_time)
|| (search->time <= oldest_live && oldest_live <= current_time))
{
itemstats[id].reclaimed++;
if ((search->it_flags & ITEM_FETCHED) == 0)
{
itemstats[id].expired_unfetched++;
}
it = search;
slabs_adjust_mem_requested(it->slabs_clsid, ITEM_ntotal(it), ntotal);
/* 把这个 item 从 LRU 队列和哈希表中移除 */
do_item_unlink_nolock(it, hv);
/* Initialize the item block: */
it->slabs_clsid = 0;
}
/* 没有超时的item, 那就尝试从slabclass分配, 运气不好的话, 分配失败, 那就把 LRU 队列最后一个 item 剔除, 然后分配给用户 */
else if ((it = slabs_alloc(ntotal, id)) == NULL)
{
tried_alloc = 1;
if (settings.evict_to_free == 0)
{
itemstats[id].outofmemory++;//显示出的统计信息
}
else
{
itemstats[id].evicted++;//这个slab的分配失败次数加1，后面的分析统计信息的线程会用到这个统计信息
itemstats[id].evicted_time = current_time - search->time;//显示的统计信息
if (search->exptime != 0)
itemstats[id].evicted_nonzero++;
if ((search->it_flags & ITEM_FETCHED) == 0)
{
itemstats[id].evicted_unfetched++;
}
it = search;
slabs_adjust_mem_requested(it->slabs_clsid, ITEM_ntotal(it), ntotal);//不用请求新的item了，减少相关的统计信息
/* 把老的item从hash表和lru队列中删除 */
do_item_unlink_nolock(it, hv);
/* Initialize the item block: */
it->slabs_clsid = 0;
/* If we've just 回收 an item, and the automover is set to angry bird mode, attempt to rip memory into this
slab class. TODO: Move valid object detection into a function, and on a "successful" memory pull, look
behind and see if the next alloc would be an eviction. Then kick off the slab mover before the eviction
happens.可以看到如果slab_automove=2(默认是1)，这样会导致angry模式，就是只要分配失败了，马上就选择一个slab(旧的slagclass
释放的)，把这个slab移动到当前slab-class中（不会有通过统计信息有选择的移动slab）*/
if (settings.slab_automove == 2)
slabs_reassign(-1, id);
}
}
refcount_decr(&search->refcount);
/* If hash values were equal, we don't grab a second lock */
if (hold_lock)
item_trylock_unlock(hold_lock);
break;
}
if (!tried_alloc && (tries == 0 || search == NULL)) it = slabs_alloc(ntotal, id);
if (it == NULL)
{
itemstats[id].outofmemory++;
mutex_unlock(&cache_lock);
return NULL;
}
assert(it->slabs_clsid == 0);
assert(it != heads[id]);
/* Item initialization can happen outside of the lock; the item's already been removed from the slab LRU. */
it->refcount = 1; /* the caller will have a reference */
mutex_unlock(&cache_lock);
it->next = it->prev = it->h_next = 0;
it->slabs_clsid = id;
DEBUG_REFCNT(it, '*');
it->it_flags = settings.use_cas ? ITEM_CAS : 0;
it->nkey = nkey;
it->nbytes = nbytes;
memcpy(ITEM_key(it), key, nkey);
it->exptime = exptime;
memcpy(ITEM_suffix(it), suffix, (size_t)nsuffix);
it->nsuffix = nsuffix;
return it;
}
/*@null@*/
item *do_item_alloc(char *key, const size_t nkey, const int flags,
const rel_time_t exptime, const int nbytes,
const uint32_t cur_hv) {
uint8_t nsuffix;
ck_spinlock_mcs_context_t second_lock;
item *it = NULL;
char suffix[40];
size_t ntotal = item_make_header(nkey + 1, flags, nbytes, suffix, &nsuffix);
if (settings.use_cas) {
ntotal += sizeof(uint64_t);
}
unsigned int id = slabs_clsid(ntotal);
if (id == 0)
return 0;
LOCK_CLOCK();
/* Avoid hangs if a slab has nothing but refcounted stuff in it. */
/* int tries_lrutail_reflocked = 1000; */
item *search;
item *next_it;
void *hold_lock = NULL;
/* We have no expiration. Try alloc a new one first. */
if ((it = slabs_alloc(ntotal, id)) == NULL) {
printf("item slab alloc fails\n");
assert(0);
/* doing CLOCK eviction */
search = hand[id];
if (!search) {
/* no mem from alloc or replace */
UNLOCK_CLOCK();
return NULL;
}
/* scan loop of the clock, which could be potentially
* unbounded -- we may want an upper limit for it. */
for (search = hand[id]; search != NULL; search = next_it) {
assert(search);
/* we might relink search mid-loop, so search->prev isn't reliable */
next_it = search->prev;
// if (*key == 101) printf("aaa %d\n", sizes[id]);
if (search->nbytes == 0 && search->nkey == 0 && search->it_flags == 1) {
/* We are a crawler, ignore it. */
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
*/
/* Don't accidentally grab ourselves, or bail if we can't quicklock */
if (hv == cur_hv || (hold_lock = item_try_mcslock(hv, &second_lock)) == NULL)
continue;
/* Now see if the item is refcount locked */
if (refcount_incr(&search->refcount) != 2) {
/* Avoid pathological case with ref'ed items in tail */
do_item_update_nolock(search);
/* tries_lrutail_reflocked--; */
refcount_decr(&search->refcount);
itemstats[id].lrutail_reflocked++;
/* Old rare bug could cause a refcount leak. We haven't seen
* it in years, but we leave this code in to prevent failures
* just in case */
if (settings.tail_repair_time &&
search->time + settings.tail_repair_time < current_time) {
itemstats[id].tailrepairs++;
search->refcount = 1;
do_item_unlink_nolock(search, hv);
}
if (hold_lock)
item_try_mcsunlock(hold_lock, &second_lock);
/* if (tries_lrutail_reflocked < 1) */
/* break; */
continue;
}
if (search->recency) {
/* recently accessed. clear bit and continue. */
search->recency = 0;
continue;
}
// printf("aaa %d, %d\n", sizes[id], *key);
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++;
}
it = search;
slabs_adjust_mem_requested(it->slabs_clsid, ITEM_ntotal(it), ntotal);
do_item_unlink_nolock(it, hv);
/* Initialize the item block: */
it->slabs_clsid = 0;
refcount_decr(&search->refcount);
/* If hash values were equal, we don't grab a second lock */
if (hold_lock)
item_try_mcsunlock(hold_lock, &second_lock);
break;
}
/* end of loop*/
}
/* end of allocation / eviction */
if (it == NULL) {
itemstats[id].outofmemory++;
UNLOCK_CLOCK();
return NULL;
}
assert(it->slabs_clsid == 0);
/* Item initialization can happen outside of the lock; the item's already
* been removed from the slab LRU.
*/
it->refcount = 1; /* the caller will have a reference */
UNLOCK_CLOCK();
it->next = it->prev = it->h_next = 0;
it->slabs_clsid = id;
DEBUG_REFCNT(it, '*');
it->it_flags = settings.use_cas ? ITEM_CAS : 0;
it->nkey = nkey;
it->nbytes = nbytes;
memcpy(ITEM_key(it), key, nkey);
it->exptime = 0; //exptime; /* disable expiration. */
memcpy(ITEM_suffix(it), suffix, (size_t)nsuffix);
it->nsuffix = nsuffix;
return it;
}
/*@null@*/
item *do_item_alloc(char *key, const size_t nkey, const int flags,
const rel_time_t exptime, const int nbytes,
const uint32_t cur_hv) {
uint8_t nsuffix;
item *it = NULL;
char suffix[40];
//计算这个item的空间
size_t ntotal = item_make_header(nkey + 1, flags, nbytes, suffix, &nsuffix);
if (settings.use_cas) {
ntotal += sizeof(uint64_t);
}
//根据大小判断从属于哪个slab
unsigned int id = slabs_clsid(ntotal);
if (id == 0)
return 0;
mutex_lock(&cache_lock);
/* do a quick check if we have any expired items in the tail.. */
//在LRU中尝试5次还没合适的空间，则执行申请空间的操作
int tries = 5;
int tried_alloc = 0;
item *search;
void *hold_lock = NULL;
rel_time_t oldest_live = settings.oldest_live;
search = tails[id];
/* We walk up *only* for locked items. Never searching for expired.
* Waste of CPU for almost all deployments */
for (; tries > 0 && search != NULL; tries--, search=search->prev) {
uint32_t hv = hash(ITEM_key(search), search->nkey, 0);
/* Attempt to hash item lock the "search" item. If locked, no
* other callers can incr the refcount
*/
/* FIXME: I think we need to mask the hv here for comparison? */
if (hv != cur_hv && (hold_lock = item_trylock(hv)) == NULL)
continue;
/* Now see if the item is refcount locked */
if (refcount_incr(&search->refcount) != 2) {
refcount_decr(&search->refcount);
/* Old rare bug could cause a refcount leak. We haven't seen
* it in years, but we leave this code in to prevent failures
* just in case */
if (search->time + TAIL_REPAIR_TIME < current_time) {
itemstats[id].tailrepairs++;
search->refcount = 1;
do_item_unlink_nolock(search, hv);
}
if (hold_lock)
item_trylock_unlock(hold_lock);
continue;
}
/* Expired or flushed */
// search指向的item过期了，则直接复用这块内存
if ((search->exptime != 0 && search->exptime < current_time)
|| (search->time <= oldest_live && oldest_live <= current_time)) {
itemstats[id].reclaimed++;
if ((search->it_flags & ITEM_FETCHED) == 0) {
itemstats[id].expired_unfetched++;
}
it = search;
slabs_adjust_mem_requested(it->slabs_clsid, ITEM_ntotal(it), ntotal);
do_item_unlink_nolock(it, hv);
/* Initialize the item block: */
it->slabs_clsid = 0;
}
//此刻，过期失效的item没有找到，申请内存又失败了。看来只能使用
//LRU淘汰一个item(即使这个item并没有过期失效)
else if ((it = slabs_alloc(ntotal, id)) == NULL) {
tried_alloc = 1;
if (settings.evict_to_free == 0) {
itemstats[id].outofmemory++;
} else {
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++;
}
it = search;
slabs_adjust_mem_requested(it->slabs_clsid, ITEM_ntotal(it), ntotal);
do_item_unlink_nolock(it, hv);
/* Initialize the item block: */
it->slabs_clsid = 0;
/* If we've just evicted an item, and the automover is set to
* angry bird mode, attempt to rip memory into this slab class.
* TODO: Move valid object detection into a function, and on a
* "successful" memory pull, look behind and see if the next alloc
* would be an eviction. Then kick off the slab mover before the
* eviction happens.
*/
if (settings.slab_automove == 2)
slabs_reassign(-1, id);
}
}
refcount_decr(&search->refcount);
/* If hash values were equal, we don't grab a second lock */
if (hold_lock)
item_trylock_unlock(hold_lock);
break;
}
//从slab分配器中申请内存
if (!tried_alloc && (tries == 0 || search == NULL))
it = slabs_alloc(ntotal, id);
if (it == NULL) {
itemstats[id].outofmemory++;
mutex_unlock(&cache_lock);
return NULL;
}
assert(it->slabs_clsid == 0);
assert(it != heads[id]);
/* Item initialization can happen outside of the lock; the item's already
* been removed from the slab LRU.
*/
it->refcount = 1; /* the caller will have a reference */
mutex_unlock(&cache_lock);
it->next = it->prev = it->h_next = 0;
it->slabs_clsid = id;
DEBUG_REFCNT(it, '*');
it->it_flags = settings.use_cas ? ITEM_CAS : 0;
it->nkey = nkey;
it->nbytes = nbytes;
memcpy(ITEM_key(it), key, nkey);
it->exptime = exptime;
memcpy(ITEM_suffix(it), suffix, (size_t)nsuffix);
it->nsuffix = nsuffix;
return it;
}
/** wrapper around assoc_find which does the lazy expiration logic */
item *do_item_get(const char *key, const size_t nkey, const uint32_t hv) {
//mutex_lock(&cache_lock);
item *it = assoc_find(key, nkey, hv);
if (it != NULL) {
refcount_incr(&it->refcount);
/* Optimization for slab reassignment. prevents popular items from
* jamming in busy wait. Can only do this here to satisfy lock order
* of item_lock, cache_lock, slabs_lock. */
if (slab_rebalance_signal &&
((void *)it >= slab_rebal.slab_start && (void *)it < slab_rebal.slab_end)) {
do_item_unlink_nolock(it, hv);
do_item_remove(it);
it = NULL;
}
}
//mutex_unlock(&cache_lock);
int was_found = 0;
if (settings.verbose > 2) {
int ii;
if (it == NULL) {
fprintf(stderr, "> NOT FOUND ");
} else {
fprintf(stderr, "> FOUND KEY ");
was_found++;
}
for (ii = 0; ii < nkey; ++ii) {
fprintf(stderr, "%c", key[ii]);
}
}
if (it != NULL) {
if (settings.oldest_live != 0 && settings.oldest_live <= current_time &&
it->time <= settings.oldest_live) {
do_item_unlink(it, hv);
do_item_remove(it);
it = NULL;
if (was_found) {
fprintf(stderr, " -nuked by flush");
}
} else if (it->exptime != 0 && it->exptime <= current_time) {
do_item_unlink(it, hv);
do_item_remove(it);
it = NULL;
if (was_found) {
fprintf(stderr, " -nuked by expire");
}
} else {
it->it_flags |= ITEM_FETCHED;
DEBUG_REFCNT(it, '+');
if (settings.anti_stampede > 0 && !(it->it_flags & ITEM_FAKE_MISSED)
&& it->exptime != 0
&& it->exptime < current_time + settings.anti_stampede
&& it->exptime <= current_time + rand() % settings.anti_stampede) {
it->it_flags |= ITEM_FAKE_MISSED;
if (settings.verbose > 2)
fprintf(stderr," - FAKE MISS (stampede protection)");
refcount_decr(&it->refcount);
it = NULL;
}
}
}
if (settings.verbose > 2)
fprintf(stderr, "\n");
return it;
}
/*@null@*/
item *do_item_alloc(char *key, const size_t nkey, const int flags,
const rel_time_t exptime, const int nbytes,
const uint32_t cur_hv) {
uint8_t nsuffix;
item *it = NULL;
char suffix[40];
size_t ntotal = item_make_header(nkey + 1, flags, nbytes, suffix, &nsuffix);
if (settings.use_cas) {
ntotal += sizeof(uint64_t);
}
unsigned int id = slabs_clsid(ntotal);
if (id == 0)
return 0;
mutex_lock(&cache_lock);
/* do a quick check if we have any expired items in the tail.. */
int tries = 5;
/* Avoid hangs if a slab has nothing but refcounted stuff in it. */
int tries_lrutail_reflocked = 1000;
int tried_alloc = 0;
item *search;
item *next_it;
void *hold_lock = NULL;
rel_time_t oldest_live = settings.oldest_live;
search = tails[id];
/* We walk up *only* for locked items. Never searching for expired.
* Waste of CPU for almost all deployments */
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. */
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
*/
/* Don't accidentally grab ourselves, or bail if we can't quicklock */
if (hv == cur_hv || (hold_lock = item_trylock(hv)) == NULL)
continue;
/* Now see if the item is refcount locked */
if (refcount_incr(&search->refcount) != 2) {
/* Avoid pathological case with ref'ed items in tail */
do_item_update_nolock(search);
tries_lrutail_reflocked--;
tries++;
refcount_decr(&search->refcount);
itemstats[id].lrutail_reflocked++;
/* Old rare bug could cause a refcount leak. We haven't seen
* it in years, but we leave this code in to prevent failures
* just in case */
if (settings.tail_repair_time &&
search->time + settings.tail_repair_time < current_time) {
itemstats[id].tailrepairs++;
search->refcount = 1;
do_item_unlink_nolock(search, hv);
}
if (hold_lock)
item_trylock_unlock(hold_lock);
if (tries_lrutail_reflocked < 1)
break;
continue;
}
/* Expired or flushed */
if ((search->exptime != 0 && search->exptime < current_time)
|| (search->time <= oldest_live && oldest_live <= current_time)) {
itemstats[id].reclaimed++;
if ((search->it_flags & ITEM_FETCHED) == 0) {
itemstats[id].expired_unfetched++;
}
it = search;
slabs_adjust_mem_requested(it->slabs_clsid, ITEM_ntotal(it), ntotal);
do_item_unlink_nolock(it, hv);
/* Initialize the item block: */
it->slabs_clsid = 0;
} else if ((it = slabs_alloc(ntotal, id)) == NULL) {
tried_alloc = 1;
if (settings.evict_to_free == 0) {
itemstats[id].outofmemory++;
} else {
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++;
}
shadow_item* new_shadow_it = create_shadow_item(search);
hv = hash(new_shadow_it->key, new_shadow_it->nkey);
shadow_assoc_insert(new_shadow_it, hv);
insert_shadowq_item(new_shadow_it,new_shadow_it->slabs_clsid);
it = search;
slabs_adjust_mem_requested(it->slabs_clsid, ITEM_ntotal(it), ntotal);
do_item_unlink_nolock(it, hv);
/* Initialize the item block: */
it->slabs_clsid = 0;
/* If we've just evicted an item, and the automover is set to
* angry bird mode, attempt to rip memory into this slab class.
* TODO: Move valid object detection into a function, and on a
* "successful" memory pull, look behind and see if the next alloc
* would be an eviction. Then kick off the slab mover before the
* eviction happens.
*/
if (settings.slab_automove == 2)
slabs_reassign(-1, id);
}
}
refcount_decr(&search->refcount);
/* If hash values were equal, we don't grab a second lock */
if (hold_lock)
item_trylock_unlock(hold_lock);
break;
}
if (!tried_alloc && (tries == 0 || search == NULL))
it = slabs_alloc(ntotal, id);
if (it == NULL) {
itemstats[id].outofmemory++;
mutex_unlock(&cache_lock);
return NULL;
}
assert(it->slabs_clsid == 0);
assert(it != heads[id]);
/* Item initialization can happen outside of the lock; the item's already
* been removed from the slab LRU.
*/
it->refcount = 1; /* the caller will have a reference */
mutex_unlock(&cache_lock);
it->next = it->prev = it->h_next = 0;
it->slabs_clsid = id;
DEBUG_REFCNT(it, '*');
it->it_flags = settings.use_cas ? ITEM_CAS : 0;
it->nkey = nkey;
it->nbytes = nbytes;
memcpy(ITEM_key(it), key, nkey);
it->exptime = exptime;
memcpy(ITEM_suffix(it), suffix, (size_t)nsuffix);
it->nsuffix = nsuffix;
return it;
}
/** wrapper around assoc_find which does the lazy expiration logic */
item *do_item_get(const char *key, const size_t nkey, const uint32_t hv) {
//mutex_lock(&cache_lock);
// 查找 key 是否存在
item *it = assoc_find(key, nkey, hv);
if (it != NULL) {
refcount_incr(&it->refcount);
/* Optimization for slab reassignment. prevents popular items from
* jamming in busy wait. Can only do this here to satisfy lock order
* of item_lock, cache_lock, slabs_lock. */
if (slab_rebalance_signal &&
((void *)it >= slab_rebal.slab_start && (void *)it < slab_rebal.slab_end)) {
do_item_unlink_nolock(it, hv);
do_item_remove(it);
it = NULL;
}
}
//mutex_unlock(&cache_lock);
int was_found = 0;
if (settings.verbose > 2) {
if (it == NULL) {
fprintf(stderr, "> NOT FOUND %s", key);
} else {
fprintf(stderr, "> FOUND KEY %s", ITEM_key(it));
was_found++;
}
}
if (it != NULL) {
//settings.oldest_live初始化值为0
//检测用户是否使用过flush_all命令，删除所有item
//it->time <= settings.oldest_live就说明用户在使用flush_all命令的时候
//就已经存在该item了。那么该item是要删除的
if (settings.oldest_live != 0 && settings.oldest_live <= current_time &&
it->time <= settings.oldest_live) {
do_item_unlink(it, hv);
do_item_remove(it);
it = NULL;
if (was_found) {
fprintf(stderr, " -nuked by flush");
}
}
//该item已经过期失效了
else if (it->exptime != 0 && it->exptime <= current_time) {
do_item_unlink(it, hv);
do_item_remove(it);
it = NULL;
if (was_found) {
fprintf(stderr, " -nuked by expire");
}
} else {
it->it_flags |= ITEM_FETCHED;
DEBUG_REFCNT(it, '+');
}
}
if (settings.verbose > 2)
fprintf(stderr, "\n");
return it;
}
/* 根据 key 找对应的 item, 为了加快查找速度, memcached 使用一个哈希表对 key 和 item 所在的内存地址做映射. item_get直接从哈希表中
查找就可以了, 当然找到了还要检查 item 是否超时. 超时了的 item将从哈希表和 LRU 队列中删除掉 */
item *do_item_get(const char *key, const size_t nkey, const uint32_t hv)
{
item *it = assoc_find(key, nkey, hv);/* 从哈希表中找 item */
if (it != NULL)
{
refcount_incr(&it->refcount);//item的引用次数+1
/* Optimization for slab reassignment. prevents popular items from
* jamming in busy wait. Can only do this here to satisfy lock order
* of item_lock, cache_lock, slabs_lock. */
if (slab_rebalance_signal && //如果正在进行slab调整，且该item是调整的对象
((void *)it >= slab_rebal.slab_start && (void *)it < slab_rebal.slab_end))
{
do_item_unlink_nolock(it, hv);//将item从hashtable和LRU链中移除
do_item_remove(it);//删除item
it = NULL;//置为空
}
}
//mutex_unlock(&cache_lock);
int was_found = 0;
//打印调试信息
if (settings.verbose > 2)
{
if (it == NULL)
{
fprintf(stderr, "> NOT FOUND %s", key);
}
else
{
fprintf(stderr, "> FOUND KEY %s", ITEM_key(it));
was_found++;
}
}
/* 找到了, 然后检查是否超时 */
if (it != NULL)
{
//判断Memcached初始化是否开启过期删除机制，如果开启，则执行删除相关操作
if (settings.oldest_live != 0 && settings.oldest_live <= current_time && it->time <= settings.oldest_live)
{
do_item_unlink(it, hv);//将item从hashtable和LRU链中移除
do_item_remove(it);//删除item
it = NULL;
if (was_found)
{
fprintf(stderr, " -nuked by flush");
}
}
//判断item是否过期
else if (it->exptime != 0 && it->exptime <= current_time)
{
do_item_unlink(it, hv);//将item从hashtable和LRU链中移除
do_item_remove(it);//删除item
it = NULL;
if (was_found)
{
fprintf(stderr, " -nuked by expire");
}
}
else
{
it->it_flags |= ITEM_FETCHED;//item的标识修改为已经读取
DEBUG_REFCNT(it, '+');
}
}
if (settings.verbose > 2)fprintf(stderr, "\n");
return it;
}
base_item* LRU_list::do_item_alloc(const char* key, const size_t nkey, const int flags, const unsigned int exptime, const int nbytes, const uint32_t cur_hv) {
base_item *it = 0;
//base_item nkey nbytes
int ntotal = sizeof(base_item) + nkey + 1 + nbytes;
std::cout << slab->mem_base.size() << std::endl;
unsigned int id = slab->slabs_clsid(ntotal);
if (id == 0)
return 0;
mutex_lock(&cache_lock);
//尝试从LRU链尾中找5次，看是否有合适的item被替换(没有被锁定&&淘汰时间到了的)
int tries = 5;
int tried_alloc = 0;
base_item* search = tails[id];
base_item* next_it;
std::mutex* hold_lock;
for( ; tries > 0 && search != 0; tries--, search = next_it) {
next_it = search->prev;
//不满足条件的item跳过
if (search->nbytes == 0 && search->nkey == 0 && search->item_flag == 1) {
tries++;
continue;
}
//算出哈希值
uint32_t hv = hashtable.hash(search->data, search->nkey);
//此处有很大争议
//if (hv == cur_hv || (hold_lock = item_trylock(hv)))
//if (hv != cur_hv && (hold_lock = item_trylock(hv)))
//是否被锁定
if ((hold_lock = hashtable.item_trylock(hv)) == 0)
continue;
if (std::atomic_fetch_add(&(search->refcount), 1u) != 2) {
std::atomic_fetch_sub(&(search->refcount), 1u);
if (search->realtime + TAIL_REPAIR_TIME < current_time()) {
search->refcount = 1;
do_item_unlink_nolock(search, hv);
}
if (hold_lock)
hashtable.item_trylock_unlock(hold_lock);
continue;
}
if ((search->exptime != 0 && search->exptime < current_time())
//少了一个和最老时间的比较
) {
it = search;
//这里的item_total与ntotal区别在?
slab->slabs_adjust_mem_requested(it->slabs_clsid, it->item_total(), ntotal);
do_item_unlink_nolock(it, hv);
it->slabs_clsid = 0;
} else if ((it = (base_item*)(slab->slabs_alloc(ntotal, id))) == 0) {
//这个if语句都是来记录item各种状态的
tried_alloc = 1;
if (mem_setting.evict_to_free == 0) {
//记录超出内存
}
else {
if (search->exptime != 0)
it = search;
slab->slabs_adjust_mem_requested(it->slabs_clsid, it->item_total(), ntotal);
do_item_unlink_nolock(it, hv);
it->slabs_clsid = 0;
//if (mem_setting.slab_automove == 2)
}
}
//解除原子引用锁定
std::atomic_fetch_sub(&(search->refcount), 1u);
if (hold_lock)
hashtable.item_trylock_unlock(hold_lock);
break;
}
if (!tried_alloc && (tries == 0 || search == 0))
it = (base_item*)slab->slabs_alloc(ntotal, id);
cache_lock.unlock();
return it;
}
//分配一个item，这个函数包含了memcached具体item分配的逻辑
item *do_item_alloc(char *key, const size_t nkey, const int flags,
const rel_time_t exptime, const int nbytes,
const uint32_t cur_hv) {
uint8_t nsuffix;
item *it = NULL;
char suffix[40];
size_t ntotal = item_make_header(nkey + 1, flags, nbytes, suffix, &nsuffix); //item总大小
if (settings.use_cas) {
ntotal += sizeof(uint64_t); //如果有用到cas，那么item大小还要加上unit64_t的size
}
unsigned int id = slabs_clsid(ntotal); //根据item的大小，找到适合的slabclass
if (id == 0)
return 0;
mutex_lock(&cache_lock); //cache锁
/* do a quick check if we have any expired items in the tail.. */
int tries = 5;
int tried_alloc = 0;
item *search;
void *hold_lock = NULL;
rel_time_t oldest_live = settings.oldest_live;
search = tails[id]; //全局变量，tails[x]是id为x的slabclass lru链表的尾部
/* We walk up *only* for locked items. Never searching for expired.
* Waste of CPU for almost all deployments */
//首先从lru链表尾部查找有没有过期的item，tries = 5，最多循环5次
//注意这里是最多查找5次，只要找到一个没有被其他地方引用的item，那么就不再继续查找，如果这个item过期，就使用这个item的空间，否则创建新的slab
for (; tries > 0 && search != NULL; tries--, search=search->prev) {
if (search->nbytes == 0 && search->nkey == 0 && search->it_flags == 1) {
/* We are a crawler, ignore it. */
//这里只是搜索过期的item，对于异常的item，直接忽略继续查找
tries++;
continue;
}
//计算item的hash值，hv有两个作用：1.用于hash表保存item 2.用于item lock表中锁住item，通过hv计算出item_lock中哪个锁对当前item加锁
//不同item的hash值可能相同，hash表中用链表的方式解决冲突；item lock中多个item共享一个锁
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
*/
/* Don't accidentally grab ourselves, or bail if we can't quicklock */
//锁住当前item
if (hv == cur_hv || (hold_lock = item_trylock(hv)) == NULL)
continue;
/* Now see if the item is refcount locked */
//检查这个指向的这个item是否被其他地方引用，如果是的话，继续向前查找
if (refcount_incr(&search->refcount) != 2) {
refcount_decr(&search->refcount);
/* Old rare bug could cause a refcount leak. We haven't seen
* it in years, but we leave this code in to prevent failures
* just in case */
if (settings.tail_repair_time &&
search->time + settings.tail_repair_time < current_time) {
itemstats[id].tailrepairs++;
search->refcount = 1;
do_item_unlink_nolock(search, hv);
}
if (hold_lock)
item_trylock_unlock(hold_lock);
continue;
}
/* Expired or flushed */
//如果找到过期的item
if ((search->exptime != 0 && search->exptime < current_time)
|| (search->time <= oldest_live && oldest_live <= current_time)) {
itemstats[id].reclaimed++;
if ((search->it_flags & ITEM_FETCHED) == 0) {
itemstats[id].expired_unfetched++;
}
it = search;
//更新统计数据
slabs_adjust_mem_requested(it->slabs_clsid, ITEM_ntotal(it), ntotal);
//把旧的item从hash表和LRU链表中移除
do_item_unlink_nolock(it, hv);
/* Initialize the item block: */
it->slabs_clsid = 0;
}
//如果没有找到过期的item，则调用slabs_alloc分配空间
//如果slabs_alloc返回null，表示分配失败，内存空间已满
//需要按LRU进行淘汰
else if ((it = slabs_alloc(ntotal, id)) == NULL) {
tried_alloc = 1; //标记一下，表示有尝试调用slabs_alloc分配空间
//记录被淘汰item的信息, 使用memcached经常会查看的evicted_time就是在这里赋值的
if (settings.evict_to_free == 0) {
itemstats[id].outofmemory++;
} else {
itemstats[id].evicted++;
itemstats[id].evicted_time = current_time - search->time; //被淘汰item距离上次使用的时间
if (search->exptime != 0)
itemstats[id].evicted_nonzero++;
if ((search->it_flags & ITEM_FETCHED) == 0) {
itemstats[id].evicted_unfetched++;
}
it = search;
slabs_adjust_mem_requested(it->slabs_clsid, ITEM_ntotal(it), ntotal); //更新统计数据
do_item_unlink_nolock(it, hv); //从hash表和LRU链表中移除
/* Initialize the item block: */
it->slabs_clsid = 0;
/* If we've just evicted an item, and the automover is set to
* angry bird mode, attempt to rip memory into this slab class.
* TODO: Move valid object detection into a function, and on a
* "successful" memory pull, look behind and see if the next alloc
* would be an eviction. Then kick off the slab mover before the
* eviction happens.
*/
//默认情况下，slab_automove=1，会合理地更具淘汰统计数据来分析怎么进行slabclass空间的分配
//如果slab_automove=2，只要分配失败了，马上进行slabclass空间的重分配
if (settings.slab_automove == 2)
slabs_reassign(-1, id);
}
}
refcount_decr(&search->refcount);
/* If hash values were equal, we don't grab a second lock */
if (hold_lock)
item_trylock_unlock(hold_lock);
break;
}
//查找5次过期的item都失败，并且也没有淘汰可用且没有过期的item
//分配新的内存空间
if (!tried_alloc && (tries == 0 || search == NULL))
it = slabs_alloc(ntotal, id);
//分配失败，返回null
if (it == NULL) {
itemstats[id].outofmemory++;
mutex_unlock(&cache_lock);
return NULL;
}
assert(it->slabs_clsid == 0);
assert(it != heads[id]);
/* Item initialization can happen outside of the lock; the item's already
* been removed from the slab LRU.
*/
//item内存空间分配成功，做一些初始化工作
it->refcount = 1; /* the caller will have a reference */
mutex_unlock(&cache_lock);
it->next = it->prev = it->h_next = 0;
it->slabs_clsid = id;
DEBUG_REFCNT(it, '*');
it->it_flags = settings.use_cas ? ITEM_CAS : 0;
it->nkey = nkey;
it->nbytes = nbytes;
memcpy(ITEM_key(it), key, nkey);
it->exptime = exptime;
memcpy(ITEM_suffix(it), suffix, (size_t)nsuffix);
it->nsuffix = nsuffix;
return it;
}
/* refcount == 0 is safe since nobody can incr while cache_lock is held.
* refcount != 0 is impossible since flags/etc can be modified in other
* threads. instead, note we found a busy one and bail. logic in do_item_get
* will prevent busy items from continuing to be busy
*/
static int slab_rebalance_move(void) {
slabclass_t *s_cls;
int x;
int was_busy = 0;
int refcount = 0;
enum move_status status = MOVE_PASS;
pthread_mutex_lock(&cache_lock);
pthread_mutex_lock(&slabs_lock);
s_cls = &slabclass[slab_rebal.s_clsid];
//会在start_slab_maintenance_thread函数中读取环境变量设置slab_bulk_check
//默认值为1.同样这里也是采用分期处理的方案处理一个页上的多个item
for (x = 0; x < slab_bulk_check; x++) {
item *it = slab_rebal.slab_pos;
status = MOVE_PASS;
if (it->slabs_clsid != 255) {
void *hold_lock = NULL;
uint32_t hv = hash(ITEM_key(it), it->nkey);
if ((hold_lock = item_trylock(hv)) == NULL) {
status = MOVE_LOCKED;
} else {
refcount = refcount_incr(&it->refcount);
if (refcount == 1) { /* item is unlinked, unused */
//如果it_flags&ITEM_SLABBED为真，那么就说明这个item
//根本就没有分配出去。如果为假，那么说明这个item被分配
//出去了，但处于归还途中。参考do_item_get函数里面的
//判断语句，有slab_rebalance_signal作为判断条件的那个。
if (it->it_flags & ITEM_SLABBED) {//没有分配出去
/* remove from slab freelist */
if (s_cls->slots == it) {
s_cls->slots = it->next;
}
if (it->next) it->next->prev = it->prev;
if (it->prev) it->prev->next = it->next;
s_cls->sl_curr--;
status = MOVE_DONE;//这个item处理成功
} else {//此时还有另外一个worker线程在归还这个item
status = MOVE_BUSY;
}
} else if (refcount == 2) { /* item is linked but not busy */
//没有worker线程引用这个item
if ((it->it_flags & ITEM_LINKED) != 0) {
//直接把这个item从哈希表和LRU队列中删除
do_item_unlink_nolock(it, hv);
status = MOVE_DONE;
} else {
/* refcount == 1 + !ITEM_LINKED means the item is being
* uploaded to, or was just unlinked but hasn't been freed
* yet. Let it bleed off on its own and try again later */
//现在有worker线程正在引用这个item
status = MOVE_BUSY;
}
} else {
if (settings.verbose > 2) {
fprintf(stderr, "Slab reassign hit a busy item: refcount: %d (%d -> %d)\n",
it->refcount, slab_rebal.s_clsid, slab_rebal.d_clsid);
}
status = MOVE_BUSY;
}
item_trylock_unlock(hold_lock);
}
}
switch (status) {
case MOVE_DONE:
it->refcount = 0;//引用计数清零
it->it_flags = 0;//清零所有属性
it->slabs_clsid = 255;
break;
case MOVE_BUSY:
refcount_decr(&it->refcount);//注意这里没有break
case MOVE_LOCKED:
slab_rebal.busy_items++;
was_busy++;//记录是否有不能马上处理的item
break;
case MOVE_PASS:
break;
}
//处理这个页的下一个item
slab_rebal.slab_pos = (char *)slab_rebal.slab_pos + s_cls->size;
if (slab_rebal.slab_pos >= slab_rebal.slab_end)//遍历完了这个页
break;
}
//遍历完了这个页的所有item
if (slab_rebal.slab_pos >= slab_rebal.slab_end) {
/* Some items were busy, start again from the top */
//在处理的时候，跳过了一些item(因为有worker线程在引用)
if (slab_rebal.busy_items) {//此时需要从头再扫描一次这个页
slab_rebal.slab_pos = slab_rebal.slab_start;
slab_rebal.busy_items = 0;
} else {
slab_rebal.done++;//标志已经处理完这个页的所有item
}
}
pthread_mutex_unlock(&slabs_lock);
pthread_mutex_unlock(&cache_lock);
return was_busy;//返回记录
}
/*@null@*/
item *do_item_alloc(char *key, const size_t nkey, const int flags, const rel_time_t exptime, const int nbytes) {
uint8_t nsuffix;
item *it = NULL;
char suffix[40];
size_t ntotal = item_make_header(nkey + 1, flags, nbytes, suffix, &nsuffix);
if (settings.use_cas) {
ntotal += sizeof(uint64_t);
}
unsigned int id = slabs_clsid(ntotal);
if (id == 0)
return 0;
mutex_lock(&cache_lock);
/* do a quick check if we have any expired items in the tail.. */
item *search;
rel_time_t oldest_live = settings.oldest_live;
search = tails[id];
if (search != NULL && (refcount_incr(&search->refcount) == 2)) {
if ((search->exptime != 0 && search->exptime < current_time)
|| (search->time <= oldest_live && oldest_live <= current_time)) { // dead by flush
STATS_LOCK();
stats.reclaimed++;
STATS_UNLOCK();
itemstats[id].reclaimed++;
if ((search->it_flags & ITEM_FETCHED) == 0) {
STATS_LOCK();
stats.expired_unfetched++;
STATS_UNLOCK();
itemstats[id].expired_unfetched++;
}
it = search;
slabs_adjust_mem_requested(it->slabs_clsid, ITEM_ntotal(it), ntotal);
do_item_unlink_nolock(it, hash(ITEM_key(it), it->nkey, 0));
/* Initialize the item block: */
it->slabs_clsid = 0;
} else if ((it = slabs_alloc(ntotal, id)) == NULL) {
if (settings.evict_to_free == 0) {
itemstats[id].outofmemory++;
mutex_unlock(&cache_lock);
return NULL;
}
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) {
STATS_LOCK();
stats.evicted_unfetched++;
STATS_UNLOCK();
itemstats[id].evicted_unfetched++;
}
STATS_LOCK();
stats.evictions++;
STATS_UNLOCK();
it = search;
slabs_adjust_mem_requested(it->slabs_clsid, ITEM_ntotal(it), ntotal);
do_item_unlink_nolock(it, hash(ITEM_key(it), it->nkey, 0));
/* Initialize the item block: */
it->slabs_clsid = 0;
/* If we've just evicted an item, and the automover is set to
* angry bird mode, attempt to rip memory into this slab class.
* TODO: Move valid object detection into a function, and on a
* "successful" memory pull, look behind and see if the next alloc
* would be an eviction. Then kick off the slab mover before the
* eviction happens.
*/
if (settings.slab_automove == 2)
slabs_reassign(-1, id, 1);
} else {
refcount_decr(&search->refcount);
}
} else {
/* If the LRU is empty or locked, attempt to allocate memory */
it = slabs_alloc(ntotal, id);
if (search != NULL)
refcount_decr(&search->refcount);
}
if (it == NULL) {
itemstats[id].outofmemory++;
/* Last ditch effort. There was a very rare bug which caused
* refcount leaks. We leave this just in case they ever happen again.
* We can reasonably assume no item can stay locked for more than
* three hours, so if we find one in the tail which is that old,
* free it anyway.
*/
if (search != NULL &&
search->refcount != 2 &&
search->time + TAIL_REPAIR_TIME < current_time) {
itemstats[id].tailrepairs++;
search->refcount = 1;
do_item_unlink_nolock(search, hash(ITEM_key(search), search->nkey, 0));
}
mutex_unlock(&cache_lock);
return NULL;
}
assert(it->slabs_clsid == 0);
assert(it != heads[id]);
/* Item initialization can happen outside of the lock; the item's already
* been removed from the slab LRU.
*/
it->refcount = 1; /* the caller will have a reference */
mutex_unlock(&cache_lock);
it->next = it->prev = it->h_next = 0;
it->slabs_clsid = id;
DEBUG_REFCNT(it, '*');
it->it_flags = settings.use_cas ? ITEM_CAS : 0;
it->nkey = nkey;
it->nbytes = nbytes;
memcpy(ITEM_key(it), key, nkey);
it->exptime = exptime;
memcpy(ITEM_suffix(it), suffix, (size_t)nsuffix);
it->nsuffix = nsuffix;
return it;
}