MFC r264146:

Fix a race between kqueue_register() and kqueue_scan() setting KN_INFLUX
flag while knlist is not locked, which caused lost notifications from
parallel knote().

2 files changed, 33 insertions, 26 deletions

diff --git a/sys/kern/kern_event.c b/sys/kern/kern_event.c
index d23452a..3ad5738 100644
--- a/sys/kern/kern_event.c
+++ b/sys/kern/kern_event.c
@@ -468,7 +468,7 @@ knote_fork(struct knlist *list, int pid)
 			continue;
 		kq = kn->kn_kq;
 		KQ_LOCK(kq);
-		if ((kn->kn_status & KN_INFLUX) == KN_INFLUX) {
+		if ((kn->kn_status & (KN_INFLUX | KN_SCAN)) == KN_INFLUX) {
 			KQ_UNLOCK(kq);
 			continue;
 		}
@@ -1145,7 +1145,7 @@ findkn:
 	 * but doing so will not reset any filter which has already been
 	 * triggered.
 	 */
-	kn->kn_status |= KN_INFLUX;
+	kn->kn_status |= KN_INFLUX | KN_SCAN;
 	KQ_UNLOCK(kq);
 	KN_LIST_LOCK(kn);
 	kn->kn_kevent.udata = kev->udata;
@@ -1168,7 +1168,7 @@ done_ev_add:
 	KQ_LOCK(kq);
 	if (event)
 		KNOTE_ACTIVATE(kn, 1);
-	kn->kn_status &= ~KN_INFLUX;
+	kn->kn_status &= ~(KN_INFLUX | KN_SCAN);
 	KN_LIST_UNLOCK(kn);
 
 	if ((kev->flags & EV_DISABLE) &&
@@ -1477,7 +1477,7 @@ retry:
 			KQ_LOCK(kq);
 			kn = NULL;
 		} else {
-			kn->kn_status |= KN_INFLUX;
+			kn->kn_status |= KN_INFLUX | KN_SCAN;
 			KQ_UNLOCK(kq);
 			if ((kn->kn_status & KN_KQUEUE) == KN_KQUEUE)
 				KQ_GLOBAL_LOCK(&kq_global, haskqglobal);
@@ -1486,7 +1486,8 @@ retry:
 				KQ_LOCK(kq);
 				KQ_GLOBAL_UNLOCK(&kq_global, haskqglobal);
 				kn->kn_status &=
-				    ~(KN_QUEUED | KN_ACTIVE | KN_INFLUX);
+				    ~(KN_QUEUED | KN_ACTIVE | KN_INFLUX |
+				    KN_SCAN);
 				kq->kq_count--;
 				KN_LIST_UNLOCK(kn);
 				influx = 1;
@@ -1516,7 +1517,7 @@ retry:
 			} else
 				TAILQ_INSERT_TAIL(&kq->kq_head, kn, kn_tqe);
 			
-			kn->kn_status &= ~(KN_INFLUX);
+			kn->kn_status &= ~(KN_INFLUX | KN_SCAN);
 			KN_LIST_UNLOCK(kn);
 			influx = 1;
 		}
@@ -1834,28 +1835,33 @@ knote(struct knlist *list, long hint, int lockflags)
 	 */
 	SLIST_FOREACH(kn, &list->kl_list, kn_selnext) {
 		kq = kn->kn_kq;
-		if ((kn->kn_status & KN_INFLUX) != KN_INFLUX) {
+		KQ_LOCK(kq);
+		if ((kn->kn_status & (KN_INFLUX | KN_SCAN)) == KN_INFLUX) {
+			/*
+			 * Do not process the influx notes, except for
+			 * the influx coming from the kq unlock in the
+			 * kqueue_scan().  In the later case, we do
+			 * not interfere with the scan, since the code
+			 * fragment in kqueue_scan() locks the knlist,
+			 * and cannot proceed until we finished.
+			 */
+			KQ_UNLOCK(kq);
+		} else if ((lockflags & KNF_NOKQLOCK) != 0) {
+			kn->kn_status |= KN_INFLUX;
+			KQ_UNLOCK(kq);
+			error = kn->kn_fop->f_event(kn, hint);
 			KQ_LOCK(kq);
-			if ((kn->kn_status & KN_INFLUX) == KN_INFLUX) {
-				KQ_UNLOCK(kq);
-			} else if ((lockflags & KNF_NOKQLOCK) != 0) {
-				kn->kn_status |= KN_INFLUX;
-				KQ_UNLOCK(kq);
-				error = kn->kn_fop->f_event(kn, hint);
-				KQ_LOCK(kq);
-				kn->kn_status &= ~KN_INFLUX;
-				if (error)
-					KNOTE_ACTIVATE(kn, 1);
-				KQ_UNLOCK_FLUX(kq);
-			} else {
-				kn->kn_status |= KN_HASKQLOCK;
-				if (kn->kn_fop->f_event(kn, hint))
-					KNOTE_ACTIVATE(kn, 1);
-				kn->kn_status &= ~KN_HASKQLOCK;
-				KQ_UNLOCK(kq);
-			}
+			kn->kn_status &= ~KN_INFLUX;
+			if (error)
+				KNOTE_ACTIVATE(kn, 1);
+			KQ_UNLOCK_FLUX(kq);
+		} else {
+			kn->kn_status |= KN_HASKQLOCK;
+			if (kn->kn_fop->f_event(kn, hint))
+				KNOTE_ACTIVATE(kn, 1);
+			kn->kn_status &= ~KN_HASKQLOCK;
+			KQ_UNLOCK(kq);
 		}
-		kq = NULL;
 	}
 	if ((lockflags & KNF_LISTLOCKED) == 0)
 		list->kl_unlock(list->kl_lockarg); 
diff --git a/sys/sys/event.h b/sys/sys/event.h
index 03bd7b9..b2f9c76 100644
--- a/sys/sys/event.h
+++ b/sys/sys/event.h
@@ -206,6 +206,7 @@ struct knote {
 #define KN_MARKER	0x20			/* ignore this knote */
 #define KN_KQUEUE	0x40			/* this knote belongs to a kq */
 #define KN_HASKQLOCK	0x80			/* for _inevent */
+#define	KN_SCAN		0x100			/* flux set in kqueue_scan() */
 	int			kn_sfflags;	/* saved filter flags */
 	intptr_t		kn_sdata;	/* saved data field */
 	union {