mirrors/haproxy
1
0
Fork 0
mirror of https://git.haproxy.org/git/haproxy.git/ synced 2025-08-06 15:17:01 +02:00
Code Issues Projects Releases Wiki Activity

MEDIUM: ring: significant boost in the loop by checking the ring queue ptr first

Browse Source 
By doing that and placing the cpu_relax at the right places, the ARM
reaches 6.0M/s on 80 threads. On x86_64, at 3C6T the EPYC sees a small
increase from 4.45M to 4.57M but at 24C48T it sees a drop from 3.82M
to 3.33M due to the write contention hidden behind the CAS that
implements the FETCH_OR(), that we'll address next.
This commit is contained in:
Willy Tarreau 2024-03-17 10:20:56 +01:00
parent 1e2311edbc
commit 30a659c355
1 changed files with 11 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
src/ring.c
View File

@ -272,21 +272,23 @@ ssize_t ring_write(struct ring *ring, size_t maxlen, const struct ist pfx[], siz
* we must detect a new leader ASAP so that the fewest possible
* threads check the tail.
*/
while ((tail_ofs = HA_ATOMIC_LOAD(tail_ptr)) & RING_TAIL_LOCK) {
next_cell = HA_ATOMIC_LOAD(ring_queue_ptr);
if (next_cell != &cell)
goto wait_for_flush; // another thread arrived, we should go to wait now
__ha_cpu_relax_for_read();
}
/* the tail is available again and we're still the leader, try
* again.
*/
if (HA_ATOMIC_LOAD(ring_queue_ptr) != &cell)
goto wait_for_flush; // another thread arrived, we should go to wait now
while (1) {
next_cell = HA_ATOMIC_LOAD(ring_queue_ptr);
if (next_cell != &cell)
goto wait_for_flush; // FIXME: another thread arrived, we should go to wait now
__ha_cpu_relax_for_read();
tail_ofs = HA_ATOMIC_FETCH_OR(tail_ptr, RING_TAIL_LOCK);
if (!(tail_ofs & RING_TAIL_LOCK))
break;
__ha_cpu_relax_for_read();
}
/* OK the queue is locked, let's attempt to get the tail lock */
tail_ofs = HA_ATOMIC_FETCH_OR(tail_ptr, RING_TAIL_LOCK);
/* did we get it ? */
if (!(tail_ofs & RING_TAIL_LOCK)) {