From: Willy Tarreau <w@1wt.eu>
Date: Sat, 17 Oct 2020 17:45:42 +0000 (+0200)
Subject: MINOR: lb/first: use a read lock in fas_get_next_server()
X-Git-Tag: v2.3-dev8~48
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=f76a21f78c2066f74fb04875006e08bc7a9413db;p=thirdparty%2Fhaproxy.git

MINOR: lb/first: use a read lock in fas_get_next_server()

The "first" algorithm creates a lot of contention because all threads
focus on the same server by definition (the first available one). By
turning the exclusive lock to a read lock in fas_get_next_server(),
the request rate increases by 16% for 8 threads when many servers are
getting close to their maxconn.
---

diff --git a/src/lb_fas.c b/src/lb_fas.c
index 43ff6a5822..e30606507c 100644
--- a/src/lb_fas.c
+++ b/src/lb_fas.c
@@ -300,7 +300,7 @@ struct server *fas_get_next_server(struct proxy *p, struct server *srvtoavoid)
 
 	srv = avoided = NULL;
 
-	HA_RWLOCK_WRLOCK(LBPRM_LOCK, &p->lbprm.lock);
+	HA_RWLOCK_RDLOCK(LBPRM_LOCK, &p->lbprm.lock);
 	if (p->srv_act)
 		node = eb32_first(&p->lbprm.fas.act);
 	else if (p->lbprm.fbck) {
@@ -336,7 +336,7 @@ struct server *fas_get_next_server(struct proxy *p, struct server *srvtoavoid)
 	if (!srv)
 		srv = avoided;
   out:
-	HA_RWLOCK_WRUNLOCK(LBPRM_LOCK, &p->lbprm.lock);
+	HA_RWLOCK_RDUNLOCK(LBPRM_LOCK, &p->lbprm.lock);
 	return srv;
 }