Add a new global rwlock, in_ifaddr_lock, which will synchronize use of the

in_ifaddrhead and INADDR_HASH address lists.

Previously, these lists were used unsynchronized as they were effectively
never changed in steady state, but we've seen increasing reports of
writer-writer races on very busy VPN servers as core count has gone up
(and similar configurations where address lists change frequently and
concurrently).

For the time being, use rwlocks rather than rmlocks in order to take
advantage of their better lock debugging support.  As a result, we don't
enable ip_input()'s read-locking of INADDR_HASH until an rmlock conversion
is complete and a performance analysis has been done.  This means that one
class of reader-writer races still exists.

MFC after:      6 weeks
Reviewed by:    bz

1 files changed, 11 insertions, 1 deletions

diff --git a/sys/netinet/if_ether.c b/sys/netinet/if_ether.c
index 9c80ea0..97ea108 100644
--- a/sys/netinet/if_ether.c
+++ b/sys/netinet/if_ether.c
@@ -509,11 +509,13 @@ in_arpinput(struct mbuf *m)
 	 * request for the virtual host ip.
 	 * XXX: This is really ugly!
 	 */
+	IN_IFADDR_RLOCK();
 	LIST_FOREACH(ia, INADDR_HASH(itaddr.s_addr), ia_hash) {
 		if (((bridged && ia->ia_ifp->if_bridge != NULL) ||
 		    ia->ia_ifp == ifp) &&
 		    itaddr.s_addr == ia->ia_addr.sin_addr.s_addr) {
 			ifa_ref(&ia->ia_ifa);
+			IN_IFADDR_RUNLOCK();
 			goto match;
 		}
 #ifdef DEV_CARP
@@ -522,6 +524,7 @@ in_arpinput(struct mbuf *m)
 		    itaddr.s_addr == ia->ia_addr.sin_addr.s_addr) {
 			carp_match = 1;
 			ifa_ref(&ia->ia_ifa);
+			IN_IFADDR_RUNLOCK();
 			goto match;
 		}
 #endif
@@ -531,6 +534,7 @@ in_arpinput(struct mbuf *m)
 		    ia->ia_ifp == ifp) &&
 		    isaddr.s_addr == ia->ia_addr.sin_addr.s_addr) {
 			ifa_ref(&ia->ia_ifa);
+			IN_IFADDR_RUNLOCK();
 			goto match;
 		}
 
@@ -549,11 +553,13 @@ in_arpinput(struct mbuf *m)
 			if (BDG_MEMBER_MATCHES_ARP(itaddr.s_addr, ifp, ia)) {
 				ifa_ref(&ia->ia_ifa);
 				ifp = ia->ia_ifp;
+				IN_IFADDR_RUNLOCK();
 				goto match;
 			}
 		}
 	}
 #undef BDG_MEMBER_MATCHES_ARP
+	IN_IFADDR_RUNLOCK();
 
 	/*
 	 * No match, use the first inet address on the receive interface
@@ -572,9 +578,13 @@ in_arpinput(struct mbuf *m)
 	/*
 	 * If bridging, fall back to using any inet address.
 	 */
-	if (!bridged || (ia = TAILQ_FIRST(&V_in_ifaddrhead)) == NULL)
+	IN_IFADDR_RLOCK();
+	if (!bridged || (ia = TAILQ_FIRST(&V_in_ifaddrhead)) == NULL) {
+		IN_IFADDR_RUNLOCK();
 		goto drop;
+	}
 	ifa_ref(&ia->ia_ifa);
+	IN_IFADDR_RUNLOCK();
 match:
 	if (!enaddr)
 		enaddr = (u_int8_t *)IF_LLADDR(ifp);