Optimize fast path allocations by storing bucket headers in the per-cpu

cache area.  This allows us to check on bucket space for all per-cpu
buckets with a single cacheline access and fewer branches.

Reviewed by:	markj, rlibby
Differential Revision:	https://reviews.freebsd.org/D22825
This commit is contained in:
Jeff Roberson 2019-12-25 20:50:53 +00:00
parent 3639ac42e5
commit 376b1ba394
Notes: svn2git 2020-12-20 02:59:44 +00:00
svn path=/head/; revision=356080
3 changed files with 225 additions and 139 deletions

View File

@ -438,28 +438,9 @@ memstat_kvm_uma(struct memory_type_list *list, void *kvm_handle)
mtp->mt_numallocs += ucp->uc_allocs;
mtp->mt_numfrees += ucp->uc_frees;
if (ucp->uc_allocbucket != NULL) {
ret = kread(kvm, ucp->uc_allocbucket,
&ub, sizeof(ub), 0);
if (ret != 0) {
free(ucp_array);
_memstat_mtl_empty(list);
list->mtl_error = ret;
return (-1);
}
mtp->mt_free += ub.ub_cnt;
}
if (ucp->uc_freebucket != NULL) {
ret = kread(kvm, ucp->uc_freebucket,
&ub, sizeof(ub), 0);
if (ret != 0) {
free(ucp_array);
_memstat_mtl_empty(list);
list->mtl_error = ret;
return (-1);
}
mtp->mt_free += ub.ub_cnt;
}
mtp->mt_free += ucp->uc_allocbucket.ucb_cnt;
mtp->mt_free += ucp->uc_freebucket.ucb_cnt;
mtp->mt_free += ucp->uc_crossbucket.ucb_cnt;
}
skip_percpu:
mtp->mt_size = kz.uk_size;

View File

@ -533,6 +533,144 @@ zone_put_bucket(uma_zone_t zone, uma_zone_domain_t zdom, uma_bucket_t bucket,
zone->uz_bkt_count += bucket->ub_cnt;
}
/* Pops an item out of a per-cpu cache bucket. */
static inline void *
cache_bucket_pop(uma_cache_t cache, uma_cache_bucket_t bucket)
{
void *item;
CRITICAL_ASSERT(curthread);
bucket->ucb_cnt--;
item = bucket->ucb_bucket->ub_bucket[bucket->ucb_cnt];
#ifdef INVARIANTS
bucket->ucb_bucket->ub_bucket[bucket->ucb_cnt] = NULL;
KASSERT(item != NULL, ("uma_zalloc: Bucket pointer mangled."));
#endif
cache->uc_allocs++;
return (item);
}
/* Pushes an item into a per-cpu cache bucket. */
static inline void
cache_bucket_push(uma_cache_t cache, uma_cache_bucket_t bucket, void *item)
{
CRITICAL_ASSERT(curthread);
KASSERT(bucket->ucb_bucket->ub_bucket[bucket->ucb_cnt] == NULL,
("uma_zfree: Freeing to non free bucket index."));
bucket->ucb_bucket->ub_bucket[bucket->ucb_cnt] = item;
bucket->ucb_cnt++;
cache->uc_frees++;
}
/*
* Unload a UMA bucket from a per-cpu cache.
*/
static inline uma_bucket_t
cache_bucket_unload(uma_cache_bucket_t bucket)
{
uma_bucket_t b;
b = bucket->ucb_bucket;
if (b != NULL) {
MPASS(b->ub_entries == bucket->ucb_entries);
b->ub_cnt = bucket->ucb_cnt;
bucket->ucb_bucket = NULL;
bucket->ucb_entries = bucket->ucb_cnt = 0;
}
return (b);
}
static inline uma_bucket_t
cache_bucket_unload_alloc(uma_cache_t cache)
{
return (cache_bucket_unload(&cache->uc_allocbucket));
}
static inline uma_bucket_t
cache_bucket_unload_free(uma_cache_t cache)
{
return (cache_bucket_unload(&cache->uc_freebucket));
}
static inline uma_bucket_t
cache_bucket_unload_cross(uma_cache_t cache)
{
return (cache_bucket_unload(&cache->uc_crossbucket));
}
/*
* Load a bucket into a per-cpu cache bucket.
*/
static inline void
cache_bucket_load(uma_cache_bucket_t bucket, uma_bucket_t b)
{
CRITICAL_ASSERT(curthread);
MPASS(bucket->ucb_bucket == NULL);
bucket->ucb_bucket = b;
bucket->ucb_cnt = b->ub_cnt;
bucket->ucb_entries = b->ub_entries;
}
static inline void
cache_bucket_load_alloc(uma_cache_t cache, uma_bucket_t b)
{
cache_bucket_load(&cache->uc_allocbucket, b);
}
static inline void
cache_bucket_load_free(uma_cache_t cache, uma_bucket_t b)
{
cache_bucket_load(&cache->uc_freebucket, b);
}
#ifdef UMA_XDOMAIN
static inline void
cache_bucket_load_cross(uma_cache_t cache, uma_bucket_t b)
{
cache_bucket_load(&cache->uc_crossbucket, b);
}
#endif
/*
* Copy and preserve ucb_spare.
*/
static inline void
cache_bucket_copy(uma_cache_bucket_t b1, uma_cache_bucket_t b2)
{
b1->ucb_bucket = b2->ucb_bucket;
b1->ucb_entries = b2->ucb_entries;
b1->ucb_cnt = b2->ucb_cnt;
}
/*
* Swap two cache buckets.
*/
static inline void
cache_bucket_swap(uma_cache_bucket_t b1, uma_cache_bucket_t b2)
{
struct uma_cache_bucket b3;
CRITICAL_ASSERT(curthread);
cache_bucket_copy(&b3, b1);
cache_bucket_copy(b1, b2);
cache_bucket_copy(b2, &b3);
}
static void
zone_log_warning(uma_zone_t zone)
{
@ -801,6 +939,7 @@ static void
cache_drain(uma_zone_t zone)
{
uma_cache_t cache;
uma_bucket_t bucket;
int cpu;
/*
@ -817,18 +956,21 @@ cache_drain(uma_zone_t zone)
*/
CPU_FOREACH(cpu) {
cache = &zone->uz_cpu[cpu];
bucket_drain(zone, cache->uc_allocbucket);
if (cache->uc_allocbucket != NULL)
bucket_free(zone, cache->uc_allocbucket, NULL);
cache->uc_allocbucket = NULL;
bucket_drain(zone, cache->uc_freebucket);
if (cache->uc_freebucket != NULL)
bucket_free(zone, cache->uc_freebucket, NULL);
cache->uc_freebucket = NULL;
bucket_drain(zone, cache->uc_crossbucket);
if (cache->uc_crossbucket != NULL)
bucket_free(zone, cache->uc_crossbucket, NULL);
cache->uc_crossbucket = NULL;
bucket = cache_bucket_unload_alloc(cache);
if (bucket != NULL) {
bucket_drain(zone, bucket);
bucket_free(zone, bucket, NULL);
}
bucket = cache_bucket_unload_free(cache);
if (bucket != NULL) {
bucket_drain(zone, bucket);
bucket_free(zone, bucket, NULL);
}
bucket = cache_bucket_unload_cross(cache);
if (bucket != NULL) {
bucket_drain(zone, bucket);
bucket_free(zone, bucket, NULL);
}
}
ZONE_LOCK(zone);
bucket_cache_reclaim(zone, true);
@ -866,24 +1008,17 @@ cache_drain_safe_cpu(uma_zone_t zone, void *unused)
else
domain = 0;
cache = &zone->uz_cpu[curcpu];
if (cache->uc_allocbucket) {
if (cache->uc_allocbucket->ub_cnt != 0)
zone_put_bucket(zone, &zone->uz_domain[domain],
cache->uc_allocbucket, false);
else
b1 = cache->uc_allocbucket;
cache->uc_allocbucket = NULL;
b1 = cache_bucket_unload_alloc(cache);
if (b1 != NULL && b1->ub_cnt != 0) {
zone_put_bucket(zone, &zone->uz_domain[domain], b1, false);
b1 = NULL;
}
if (cache->uc_freebucket) {
if (cache->uc_freebucket->ub_cnt != 0)
zone_put_bucket(zone, &zone->uz_domain[domain],
cache->uc_freebucket, false);
else
b2 = cache->uc_freebucket;
cache->uc_freebucket = NULL;
b2 = cache_bucket_unload_free(cache);
if (b2 != NULL && b2->ub_cnt != 0) {
zone_put_bucket(zone, &zone->uz_domain[domain], b2, false);
b2 = NULL;
}
b3 = cache->uc_crossbucket;
cache->uc_crossbucket = NULL;
b3 = cache_bucket_unload_cross(cache);
critical_exit();
ZONE_UNLOCK(zone);
if (b1)
@ -2666,33 +2801,6 @@ uma_zfree_pcpu_arg(uma_zone_t zone, void *item, void *udata)
uma_zfree_arg(zone, item, udata);
}
static inline void *
bucket_pop(uma_zone_t zone, uma_cache_t cache, uma_bucket_t bucket)
{
void *item;
bucket->ub_cnt--;
item = bucket->ub_bucket[bucket->ub_cnt];
#ifdef INVARIANTS
bucket->ub_bucket[bucket->ub_cnt] = NULL;
KASSERT(item != NULL, ("uma_zalloc: Bucket pointer mangled."));
#endif
cache->uc_allocs++;
return (item);
}
static inline void
bucket_push(uma_zone_t zone, uma_cache_t cache, uma_bucket_t bucket,
void *item)
{
KASSERT(bucket->ub_bucket[bucket->ub_cnt] == NULL,
("uma_zfree: Freeing to non free bucket index."));
bucket->ub_bucket[bucket->ub_cnt] = item;
bucket->ub_cnt++;
cache->uc_frees++;
}
static void *
item_ctor(uma_zone_t zone, void *udata, int flags, void *item)
{
@ -2749,7 +2857,7 @@ item_dtor(uma_zone_t zone, void *item, void *udata, enum zfreeskip skip)
void *
uma_zalloc_arg(uma_zone_t zone, void *udata, int flags)
{
uma_bucket_t bucket;
uma_cache_bucket_t bucket;
uma_cache_t cache;
void *item;
int cpu, domain;
@ -2806,9 +2914,9 @@ uma_zalloc_arg(uma_zone_t zone, void *udata, int flags)
do {
cpu = curcpu;
cache = &zone->uz_cpu[cpu];
bucket = cache->uc_allocbucket;
if (__predict_true(bucket != NULL && bucket->ub_cnt != 0)) {
item = bucket_pop(zone, cache, bucket);
bucket = &cache->uc_allocbucket;
if (__predict_true(bucket->ucb_cnt != 0)) {
item = cache_bucket_pop(cache, bucket);
critical_exit();
return (item_ctor(zone, udata, flags, item));
}
@ -2846,18 +2954,15 @@ cache_alloc(uma_zone_t zone, uma_cache_t cache, void *udata, int flags)
* If we have run out of items in our alloc bucket see
* if we can switch with the free bucket.
*/
bucket = cache->uc_freebucket;
if (bucket != NULL && bucket->ub_cnt != 0) {
cache->uc_freebucket = cache->uc_allocbucket;
cache->uc_allocbucket = bucket;
if (cache->uc_freebucket.ucb_cnt != 0) {
cache_bucket_swap(&cache->uc_freebucket, &cache->uc_allocbucket);
return (true);
}
/*
* Discard any empty allocation bucket while we hold no locks.
*/
bucket = cache->uc_allocbucket;
cache->uc_allocbucket = NULL;
bucket = cache_bucket_unload_alloc(cache);
critical_exit();
if (bucket != NULL)
bucket_free(zone, bucket, udata);
@ -2887,7 +2992,7 @@ cache_alloc(uma_zone_t zone, uma_cache_t cache, void *udata, int flags)
cache = &zone->uz_cpu[cpu];
/* See if we lost the race to fill the cache. */
if (cache->uc_allocbucket != NULL) {
if (cache->uc_allocbucket.ucb_bucket != NULL) {
ZONE_UNLOCK(zone);
return (true);
}
@ -2907,7 +3012,7 @@ cache_alloc(uma_zone_t zone, uma_cache_t cache, void *udata, int flags)
ZONE_UNLOCK(zone);
KASSERT(bucket->ub_cnt != 0,
("uma_zalloc_arg: Returning an empty bucket."));
cache->uc_allocbucket = bucket;
cache_bucket_load_alloc(cache, bucket);
return (true);
}
/* We are no longer associated with this CPU. */
@ -2937,10 +3042,10 @@ cache_alloc(uma_zone_t zone, uma_cache_t cache, void *udata, int flags)
*/
cpu = curcpu;
cache = &zone->uz_cpu[cpu];
if (cache->uc_allocbucket == NULL &&
if (cache->uc_allocbucket.ucb_bucket == NULL &&
((zone->uz_flags & UMA_ZONE_NUMA) == 0 ||
domain == PCPU_GET(domain))) {
cache->uc_allocbucket = bucket;
cache_bucket_load_alloc(cache, bucket);
zdom->uzd_imax += bucket->ub_cnt;
} else if (zone->uz_bkt_count >= zone->uz_bkt_max) {
critical_exit();
@ -3361,7 +3466,7 @@ void
uma_zfree_arg(uma_zone_t zone, void *item, void *udata)
{
uma_cache_t cache;
uma_bucket_t bucket;
uma_cache_bucket_t bucket;
int cpu, domain, itemdomain;
/* Enable entropy collection for RANDOM_ENABLE_UMA kernel option */
@ -3411,7 +3516,7 @@ uma_zfree_arg(uma_zone_t zone, void *item, void *udata)
do {
cpu = curcpu;
cache = &zone->uz_cpu[cpu];
bucket = cache->uc_allocbucket;
bucket = &cache->uc_allocbucket;
#ifdef UMA_XDOMAIN
if ((zone->uz_flags & UMA_ZONE_NUMA) != 0) {
itemdomain = _vm_phys_domain(pmap_kextract((vm_offset_t)item));
@ -3419,7 +3524,7 @@ uma_zfree_arg(uma_zone_t zone, void *item, void *udata)
}
if ((zone->uz_flags & UMA_ZONE_NUMA) != 0 &&
domain != itemdomain) {
bucket = cache->uc_crossbucket;
bucket = &cache->uc_crossbucket;
} else
#endif
@ -3428,11 +3533,10 @@ uma_zfree_arg(uma_zone_t zone, void *item, void *udata)
* for cache-hot datastructures. Spill over into the freebucket
* if necessary. Alloc will swap them if one runs dry.
*/
if (bucket == NULL || bucket->ub_cnt >= bucket->ub_entries)
bucket = cache->uc_freebucket;
if (__predict_true(bucket != NULL &&
bucket->ub_cnt < bucket->ub_entries)) {
bucket_push(zone, cache, bucket, item);
if (__predict_false(bucket->ucb_cnt >= bucket->ucb_entries))
bucket = &cache->uc_freebucket;
if (__predict_true(bucket->ucb_cnt < bucket->ucb_entries)) {
cache_bucket_push(cache, bucket, item);
critical_exit();
return;
}
@ -3536,16 +3640,12 @@ cache_free(uma_zone_t zone, uma_cache_t cache, void *udata, void *item,
itemdomain = domain = 0;
#ifdef UMA_XDOMAIN
if (domain != itemdomain) {
bucket = cache->uc_crossbucket;
cache->uc_crossbucket = NULL;
bucket = cache_bucket_unload_cross(cache);
if (bucket != NULL)
atomic_add_64(&zone->uz_xdomain, bucket->ub_cnt);
} else
#endif
{
bucket = cache->uc_freebucket;
cache->uc_freebucket = NULL;
}
bucket = cache_bucket_unload_free(cache);
/* We are no longer associated with this CPU. */
@ -3570,8 +3670,9 @@ cache_free(uma_zone_t zone, uma_cache_t cache, void *udata, void *item,
*/
if ((zone->uz_flags & UMA_ZONE_NUMA) != 0) {
domain = PCPU_GET(domain);
if (domain != itemdomain && cache->uc_crossbucket == NULL) {
cache->uc_crossbucket = bucket;
if (domain != itemdomain &&
cache->uc_crossbucket.ucb_bucket == NULL) {
cache_bucket_load_cross(cache, bucket);
return (true);
}
}
@ -3579,12 +3680,12 @@ cache_free(uma_zone_t zone, uma_cache_t cache, void *udata, void *item,
/*
* We may have lost the race to fill the bucket or switched CPUs.
*/
if (cache->uc_freebucket != NULL) {
if (cache->uc_freebucket.ucb_bucket != NULL) {
critical_exit();
bucket_free(zone, bucket, udata);
critical_enter();
} else
cache->uc_freebucket = bucket;
cache_bucket_load_free(cache, bucket);
return (true);
}
@ -4175,9 +4276,6 @@ uma_avail(void)
* Note: does not update the zone statistics, as it can't safely clear the
* per-CPU cache statistic.
*
* XXXRW: Following the uc_allocbucket and uc_freebucket pointers here isn't
* safe from off-CPU; we should modify the caches to track this information
* directly so that we don't have to.
*/
static void
uma_zone_sumstat(uma_zone_t z, long *cachefreep, uint64_t *allocsp,
@ -4191,14 +4289,10 @@ uma_zone_sumstat(uma_zone_t z, long *cachefreep, uint64_t *allocsp,
cachefree = 0;
CPU_FOREACH(cpu) {
cache = &z->uz_cpu[cpu];
if (cache->uc_allocbucket != NULL)
cachefree += cache->uc_allocbucket->ub_cnt;
if (cache->uc_freebucket != NULL)
cachefree += cache->uc_freebucket->ub_cnt;
if (cache->uc_crossbucket != NULL) {
xdomain += cache->uc_crossbucket->ub_cnt;
cachefree += cache->uc_crossbucket->ub_cnt;
}
cachefree += cache->uc_allocbucket.ucb_cnt;
cachefree += cache->uc_freebucket.ucb_cnt;
xdomain += cache->uc_crossbucket.ucb_cnt;
cachefree += cache->uc_crossbucket.ucb_cnt;
allocs += cache->uc_allocs;
frees += cache->uc_frees;
}
@ -4244,7 +4338,6 @@ uma_vm_zone_stats(struct uma_type_header *uth, uma_zone_t z, struct sbuf *sbuf,
struct uma_percpu_stat *ups, bool internal)
{
uma_zone_domain_t zdom;
uma_bucket_t bucket;
uma_cache_t cache;
int i;
@ -4272,15 +4365,9 @@ uma_vm_zone_stats(struct uma_type_header *uth, uma_zone_t z, struct sbuf *sbuf,
if (internal || CPU_ABSENT(i))
continue;
cache = &z->uz_cpu[i];
bucket = (uma_bucket_t)atomic_load_ptr(&cache->uc_allocbucket);
if (bucket != NULL)
ups[i].ups_cache_free += bucket->ub_cnt;
bucket = (uma_bucket_t)atomic_load_ptr(&cache->uc_freebucket);
if (bucket != NULL)
ups[i].ups_cache_free += bucket->ub_cnt;
bucket = (uma_bucket_t)atomic_load_ptr(&cache->uc_crossbucket);
if (bucket != NULL)
ups[i].ups_cache_free += bucket->ub_cnt;
ups[i].ups_cache_free += cache->uc_allocbucket.ucb_cnt;
ups[i].ups_cache_free += cache->uc_freebucket.ucb_cnt;
ups[i].ups_cache_free += cache->uc_crossbucket.ucb_cnt;
ups[i].ups_allocs = cache->uc_allocs;
ups[i].ups_frees = cache->uc_frees;
}

View File

@ -171,13 +171,14 @@ struct uma_hash {
#if defined(__amd64__) || defined(__powerpc64__)
#define UMA_ALIGN __aligned(128)
#else
#define UMA_ALIGN
#define UMA_ALIGN __aligned(CACHE_LINE_SIZE)
#endif
/*
* Structures for per cpu queues.
* The uma_bucket structure is used to queue and manage buckets divorced
* from per-cpu caches. They are loaded into uma_cache_bucket structures
* for use.
*/
struct uma_bucket {
TAILQ_ENTRY(uma_bucket) ub_link; /* Link into the zone */
int16_t ub_cnt; /* Count of items in bucket. */
@ -187,12 +188,29 @@ struct uma_bucket {
typedef struct uma_bucket * uma_bucket_t;
/*
* The uma_cache_bucket structure is statically allocated on each per-cpu
* cache. Its use reduces branches and cache misses in the fast path.
*/
struct uma_cache_bucket {
uma_bucket_t ucb_bucket;
int16_t ucb_cnt;
int16_t ucb_entries;
uint32_t ucb_spare;
};
typedef struct uma_cache_bucket * uma_cache_bucket_t;
/*
* The uma_cache structure is allocated for each cpu for every zone
* type. This optimizes synchronization out of the allocator fast path.
*/
struct uma_cache {
uma_bucket_t uc_freebucket; /* Bucket we're freeing to */
uma_bucket_t uc_allocbucket; /* Bucket to allocate from */
uma_bucket_t uc_crossbucket; /* cross domain bucket */
uint64_t uc_allocs; /* Count of allocations */
uint64_t uc_frees; /* Count of frees */
struct uma_cache_bucket uc_freebucket; /* Bucket we're freeing to */
struct uma_cache_bucket uc_allocbucket; /* Bucket to allocate from */
struct uma_cache_bucket uc_crossbucket; /* cross domain bucket */
uint64_t uc_allocs; /* Count of allocations */
uint64_t uc_frees; /* Count of frees */
} UMA_ALIGN;
typedef struct uma_cache * uma_cache_t;
@ -200,7 +218,7 @@ typedef struct uma_cache * uma_cache_t;
LIST_HEAD(slabhead, uma_slab);
/*
* Per-domain memory list. Embedded in the kegs.
* Per-domain slab lists. Embedded in the kegs.
*/
struct uma_domain {
struct slabhead ud_part_slab; /* partially allocated slabs */