Rework NFS Duplicate Request Cache cleanup logic.

 - Introduce additional hash to group requests by hash of sockref.  This
allows to process TCP acknowledgements without looping though all the cache,
and as result allows to do it every time.
 - Indroduce additional callbacks to notify application layer about sockets
disconnection.  Without this last few requests processed just before socket
disconnection never processed their ACKs and stuck in cache for many hours.
 - Implement transport-specific method for tracking reply acknowledgements.
New implementation does not cross multiple stack layers to get the data and
does not have race conditions that previously made some requests stuck
in cache.  This could be done more efficiently at sockbuf layer, but that
would broke some KBIs, while I don't know other consumers for it aside NFS.
 - Instead of traversing all DRC twice per request, run cleaning only once
per request, and except in some conditions traverse only single hash slot
at a time.

Together this limits NFS DRC growth only to situations of real connectivity
problems.  If network is working well, and so all replies are acknowledged,
cache remains almost empty even after hours of heavy load.  Without this
change on the same test cache was growing to many thousand requests even
with perfectly working local network.

As another result this reduces CPU time spent on the DRC handling during
SPEC NFS benchmark from about 10% to 0.5%.

Sponsored by:	iXsystems, Inc.

1 files changed, 63 insertions, 4 deletions

diff --git a/sys/rpc/svc.c b/sys/rpc/svc.c
index b42bafb..bc82cd4 100644
--- a/sys/rpc/svc.c
+++ b/sys/rpc/svc.c
@@ -56,6 +56,7 @@ __FBSDID("$FreeBSD$");
 #include <sys/queue.h>
 #include <sys/socketvar.h>
 #include <sys/systm.h>
+#include <sys/sx.h>
 #include <sys/ucred.h>
 
 #include <rpc/rpc.h>
@@ -93,6 +94,7 @@ svcpool_create(const char *name, struct sysctl_oid_list *sysctl_base)
 	TAILQ_INIT(&pool->sp_xlist);
 	TAILQ_INIT(&pool->sp_active);
 	TAILQ_INIT(&pool->sp_callouts);
+	TAILQ_INIT(&pool->sp_lcallouts);
 	LIST_INIT(&pool->sp_threads);
 	LIST_INIT(&pool->sp_idlethreads);
 	pool->sp_minthreads = 1;
@@ -158,6 +160,7 @@ svcpool_destroy(SVCPOOL *pool)
 {
 	SVCXPRT *xprt, *nxprt;
 	struct svc_callout *s;
+	struct svc_loss_callout *sl;
 	struct svcxprt_list cleanup;
 
 	TAILQ_INIT(&cleanup);
@@ -169,12 +172,16 @@ svcpool_destroy(SVCPOOL *pool)
 		TAILQ_INSERT_TAIL(&cleanup, xprt, xp_link);
 	}
 
-	while (TAILQ_FIRST(&pool->sp_callouts)) {
-		s = TAILQ_FIRST(&pool->sp_callouts);
+	while ((s = TAILQ_FIRST(&pool->sp_callouts)) != NULL) {
 		mtx_unlock(&pool->sp_lock);
 		svc_unreg(pool, s->sc_prog, s->sc_vers);
 		mtx_lock(&pool->sp_lock);
 	}
+	while ((sl = TAILQ_FIRST(&pool->sp_lcallouts)) != NULL) {
+		mtx_unlock(&pool->sp_lock);
+		svc_loss_unreg(pool, sl->slc_dispatch);
+		mtx_lock(&pool->sp_lock);
+	}
 	mtx_unlock(&pool->sp_lock);
 
 	TAILQ_FOREACH_SAFE(xprt, &cleanup, xp_link, nxprt) {
@@ -511,6 +518,55 @@ svc_unreg(SVCPOOL *pool, const rpcprog_t prog, const rpcvers_t vers)
 	mtx_unlock(&pool->sp_lock);
 }
 
+/*
+ * Add a service connection loss program to the callout list.
+ * The dispatch routine will be called when some port in ths pool die.
+ */
+bool_t
+svc_loss_reg(SVCXPRT *xprt, void (*dispatch)(SVCXPRT *))
+{
+	SVCPOOL *pool = xprt->xp_pool;
+	struct svc_loss_callout *s;
+
+	mtx_lock(&pool->sp_lock);
+	TAILQ_FOREACH(s, &pool->sp_lcallouts, slc_link) {
+		if (s->slc_dispatch == dispatch)
+			break;
+	}
+	if (s != NULL) {
+		mtx_unlock(&pool->sp_lock);
+		return (TRUE);
+	}
+	s = malloc(sizeof (struct svc_callout), M_RPC, M_NOWAIT);
+	if (s == NULL) {
+		mtx_unlock(&pool->sp_lock);
+		return (FALSE);
+	}
+	s->slc_dispatch = dispatch;
+	TAILQ_INSERT_TAIL(&pool->sp_lcallouts, s, slc_link);
+	mtx_unlock(&pool->sp_lock);
+	return (TRUE);
+}
+
+/*
+ * Remove a service connection loss program from the callout list.
+ */
+void
+svc_loss_unreg(SVCPOOL *pool, void (*dispatch)(SVCXPRT *))
+{
+	struct svc_loss_callout *s;
+
+	mtx_lock(&pool->sp_lock);
+	TAILQ_FOREACH(s, &pool->sp_lcallouts, slc_link) {
+		if (s->slc_dispatch == dispatch) {
+			TAILQ_REMOVE(&pool->sp_lcallouts, s, slc_link);
+			free(s, M_RPC);
+			break;
+		}
+	}
+	mtx_unlock(&pool->sp_lock);
+}
+
 /* ********************** CALLOUT list related stuff ************* */
 
 /*
@@ -554,7 +610,7 @@ svc_sendreply_common(struct svc_req *rqstp, struct rpc_msg *rply,
 	if (!SVCAUTH_WRAP(&rqstp->rq_auth, &body))
 		return (FALSE);
 
-	ok = SVC_REPLY(xprt, rply, rqstp->rq_addr, body); 
+	ok = SVC_REPLY(xprt, rply, rqstp->rq_addr, body, &rqstp->rq_reply_seq);
 	if (rqstp->rq_addr) {
 		free(rqstp->rq_addr, M_SONAME);
 		rqstp->rq_addr = NULL;
@@ -803,6 +859,7 @@ svc_getreq(SVCXPRT *xprt, struct svc_req **rqstp_ret)
 	struct svc_req *r;
 	struct rpc_msg msg;
 	struct mbuf *args;
+	struct svc_loss_callout *s;
 	enum xprt_stat stat;
 
 	/* now receive msgs from xprtprt (support batch calls) */
@@ -831,7 +888,7 @@ svc_getreq(SVCXPRT *xprt, struct svc_req **rqstp_ret)
 				break;
 			case RS_DONE:
 				SVC_REPLY(xprt, &repmsg, r->rq_addr,
-				    repbody);
+				    repbody, &r->rq_reply_seq);
 				if (r->rq_addr) {
 					free(r->rq_addr, M_SONAME);
 					r->rq_addr = NULL;
@@ -881,6 +938,8 @@ call_done:
 		r = NULL;
 	}
 	if ((stat = SVC_STAT(xprt)) == XPRT_DIED) {
+		TAILQ_FOREACH(s, &pool->sp_lcallouts, slc_link)
+			(*s->slc_dispatch)(xprt);
 		xprt_unregister(xprt);
 	}