RepoMirrors/haproxy
1
0
Fork 0
mirror of http://git.haproxy.org/git/haproxy.git/ synced 2025-03-11 05:48:41 +00:00
Code Issues Packages Projects Releases Wiki Activity

OPTIM: pools: use exponential back-off on shared pool allocation/release

Browse Source 
Running a stick-table stress with -dMglobal under 56 threads shows
extreme contention on the pool's free_list because it has to be
processed in two phases and only used to implement a cpu_relax() on
the retry path.

Let's at least implement exponential back-off here to limit the neighbor's
noise and reduce the time needed to successfully acquire the pointer. Just
doing so shows there's still contention but almost doubled the performance,
from 1.1 to 2.1M req/s.
This commit is contained in:
Willy Tarreau 2023-07-24 15:12:31 +02:00
parent f0d188f6ed
commit feeda4132b
1 changed files with 2 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
src/pool.c
View File

@ -632,8 +632,7 @@ void pool_refill_local_from_shared(struct pool_head *pool, struct pool_cache_hea
ret = _HA_ATOMIC_LOAD(&pool->free_list);
do {
while (unlikely(ret == POOL_BUSY)) {
__ha_cpu_relax();
ret = _HA_ATOMIC_LOAD(&pool->free_list);
ret = (void*)pl_wait_new_long((ulong*)&pool->free_list, (ulong)ret);
}
if (ret == NULL)
return;
@ -678,8 +677,7 @@ void pool_put_to_shared_cache(struct pool_head *pool, struct pool_item *item, ui
free_list = _HA_ATOMIC_LOAD(&pool->free_list);
do {
while (unlikely(free_list == POOL_BUSY)) {
__ha_cpu_relax();
free_list = _HA_ATOMIC_LOAD(&pool->free_list);
free_list = (void*)pl_wait_new_long((ulong*)&pool->free_list, (ulong)free_list);
}
_HA_ATOMIC_STORE(&item->next, free_list);
__ha_barrier_atomic_store();