]> git.saurik.com Git - apple/xnu.git/blobdiff - bsd/kern/kern_event.c
xnu-2782.40.9.tar.gz
[apple/xnu.git] / bsd / kern / kern_event.c
index f298f525b42f3098def80ad2e23816dd3e6bc989..708aef4747ad4ddd1bbeef79d7c7fdf43cedaa9f 100644 (file)
@@ -1,17 +1,20 @@
 /*
- * Copyright (c) 2000 Apple Computer, Inc. All rights reserved.
+ * Copyright (c) 2000-2014 Apple Inc. All rights reserved.
+ *
+ * @APPLE_OSREFERENCE_LICENSE_HEADER_START@
  *
- * @APPLE_LICENSE_HEADER_START@
- * 
- * Copyright (c) 1999-2003 Apple Computer, Inc.  All Rights Reserved.
- * 
  * This file contains Original Code and/or Modifications of Original Code
  * as defined in and that are subject to the Apple Public Source License
  * Version 2.0 (the 'License'). You may not use this file except in
- * compliance with the License. Please obtain a copy of the License at
- * http://www.opensource.apple.com/apsl/ and read it before using this
- * file.
- * 
+ * compliance with the License. The rights granted to you under the License
+ * may not be used to create, or enable the creation or redistribution of,
+ * unlawful or unlicensed copies of an Apple operating system, or to
+ * circumvent, violate, or enable the circumvention or violation of, any
+ * terms of an Apple operating system software license agreement.
+ *
+ * Please obtain a copy of the License at
+ * http://www.opensource.apple.com/apsl/ and read it before using this file.
+ *
  * The Original Code and all software distributed under the License are
  * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
  * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
  * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
  * Please see the License for the specific language governing rights and
  * limitations under the License.
- * 
- * @APPLE_LICENSE_HEADER_END@
+ *
+ * @APPLE_OSREFERENCE_LICENSE_HEADER_END@
  *
  */
+/*-
+ * Copyright (c) 1999,2000,2001 Jonathan Lemon <jlemon@FreeBSD.org>
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
 /*
  *     @(#)kern_event.c       1.0 (3/31/2000)
  */
+#include <stdint.h>
 
 #include <sys/param.h>
-#include <sys/socket.h>
-#include <sys/protosw.h>
-#include <sys/domain.h>
-#include <sys/mbuf.h>
-#include <sys/kern_event.h>
+#include <sys/systm.h>
+#include <sys/filedesc.h>
+#include <sys/kernel.h>
+#include <sys/proc_internal.h>
+#include <sys/kauth.h>
 #include <sys/malloc.h>
-#include <sys/sys_domain.h>
-#include <sys/syslog.h>
+#include <sys/unistd.h>
+#include <sys/file_internal.h>
+#include <sys/fcntl.h>
+#include <sys/select.h>
+#include <sys/queue.h>
+#include <sys/event.h>
+#include <sys/eventvar.h>
+#include <sys/protosw.h>
+#include <sys/socket.h>
+#include <sys/socketvar.h>
+#include <sys/stat.h>
+#include <sys/sysctl.h>
+#include <sys/uio.h>
+#include <sys/sysproto.h>
+#include <sys/user.h>
+#include <sys/vnode_internal.h>
+#include <string.h>
+#include <sys/proc_info.h>
+#include <sys/codesign.h>
 
+#include <kern/locks.h>
+#include <kern/clock.h>
+#include <kern/thread_call.h>
+#include <kern/sched_prim.h>
+#include <kern/wait_queue.h>
+#include <kern/zalloc.h>
+#include <kern/assert.h>
 
-int    raw_usrreq();
-struct pr_usrreqs event_usrreqs;
+#include <libkern/libkern.h>
+#include "net/net_str_id.h"
 
-struct protosw eventsw[] = {
-     {
-         SOCK_RAW,             &systemdomain,  SYSPROTO_EVENT,         PR_ATOMIC,
-         0,            0,              0,              0,
-         0,
-         0,            0,              0,              0,
-         0,            &event_usrreqs
-     }
+#include <mach/task.h>
+
+#if VM_PRESSURE_EVENTS
+#include <kern/vm_pressure.h>
+#endif
+
+#if CONFIG_MEMORYSTATUS
+#include <sys/kern_memorystatus.h>
+#endif
+
+MALLOC_DEFINE(M_KQUEUE, "kqueue", "memory for kqueue system");
+
+#define        KQ_EVENT        NULL
+
+static inline void kqlock(struct kqueue *kq);
+static inline void kqunlock(struct kqueue *kq);
+
+static int kqlock2knoteuse(struct kqueue *kq, struct knote *kn);
+static int kqlock2knoteusewait(struct kqueue *kq, struct knote *kn);
+static int kqlock2knotedrop(struct kqueue *kq, struct knote *kn);
+static int knoteuse2kqlock(struct kqueue *kq, struct knote *kn);
+
+static void kqueue_wakeup(struct kqueue *kq, int closed);
+static int kqueue_read(struct fileproc *fp, struct uio *uio,
+    int flags, vfs_context_t ctx);
+static int kqueue_write(struct fileproc *fp, struct uio *uio,
+    int flags, vfs_context_t ctx);
+static int kqueue_ioctl(struct fileproc *fp, u_long com, caddr_t data,
+    vfs_context_t ctx);
+static int kqueue_select(struct fileproc *fp, int which, void *wql,
+    vfs_context_t ctx);
+static int kqueue_close(struct fileglob *fg, vfs_context_t ctx);
+static int kqueue_kqfilter(struct fileproc *fp, struct knote *kn,
+       vfs_context_t ctx);
+static int kqueue_drain(struct fileproc *fp, vfs_context_t ctx);
+
+static const struct fileops kqueueops = {
+       .fo_type = DTYPE_KQUEUE,
+       .fo_read = kqueue_read,
+       .fo_write = kqueue_write,
+       .fo_ioctl = kqueue_ioctl,
+       .fo_select = kqueue_select,
+       .fo_close = kqueue_close,
+       .fo_kqfilter = kqueue_kqfilter,
+       .fo_drain = kqueue_drain,
 };
 
-static
-struct kern_event_head kern_event_head;
+static int kevent_internal(struct proc *p, int iskev64, user_addr_t changelist,
+    int nchanges, user_addr_t eventlist, int nevents, int fd,
+    user_addr_t utimeout, unsigned int flags, int32_t *retval);
+static int kevent_copyin(user_addr_t *addrp, struct kevent64_s *kevp,
+    struct proc *p, int iskev64);
+static int kevent_copyout(struct kevent64_s *kevp, user_addr_t *addrp,
+    struct proc *p, int iskev64);
+char * kevent_description(struct kevent64_s *kevp, char *s, size_t n);
+
+static int kevent_callback(struct kqueue *kq, struct kevent64_s *kevp,
+    void *data);
+static void kevent_continue(struct kqueue *kq, void *data, int error);
+static void kqueue_scan_continue(void *contp, wait_result_t wait_result);
+static int kqueue_process(struct kqueue *kq, kevent_callback_t callback,
+    void *data, int *countp, struct proc *p);
+static int kqueue_begin_processing(struct kqueue *kq);
+static void kqueue_end_processing(struct kqueue *kq);
+static int knote_process(struct knote *kn, kevent_callback_t callback,
+    void *data, struct kqtailq *inprocessp, struct proc *p);
+static void knote_put(struct knote *kn);
+static int knote_fdpattach(struct knote *kn, struct filedesc *fdp,
+    struct proc *p);
+static void knote_drop(struct knote *kn, struct proc *p);
+static void knote_activate(struct knote *kn, int);
+static void knote_deactivate(struct knote *kn);
+static void knote_enqueue(struct knote *kn);
+static void knote_dequeue(struct knote *kn);
+static struct knote *knote_alloc(void);
+static void knote_free(struct knote *kn);
+
+static int filt_fileattach(struct knote *kn);
+static struct filterops file_filtops = {
+       .f_isfd = 1,
+       .f_attach = filt_fileattach,
+};
+
+static void filt_kqdetach(struct knote *kn);
+static int filt_kqueue(struct knote *kn, long hint);
+static struct filterops kqread_filtops = {
+       .f_isfd = 1,
+       .f_detach = filt_kqdetach,
+       .f_event = filt_kqueue,
+};
+
+/* placeholder for not-yet-implemented filters */
+static int filt_badattach(struct knote *kn);
+static struct filterops bad_filtops = {
+       .f_attach = filt_badattach,
+};
+
+static int filt_procattach(struct knote *kn);
+static void filt_procdetach(struct knote *kn);
+static int filt_proc(struct knote *kn, long hint);
+static struct filterops proc_filtops = {
+       .f_attach = filt_procattach,
+       .f_detach = filt_procdetach,
+       .f_event = filt_proc,
+};
+
+#if VM_PRESSURE_EVENTS
+static int filt_vmattach(struct knote *kn);
+static void filt_vmdetach(struct knote *kn);
+static int filt_vm(struct knote *kn, long hint);
+static struct filterops vm_filtops = {
+       .f_attach = filt_vmattach,
+       .f_detach = filt_vmdetach,
+       .f_event = filt_vm,
+};
+#endif /* VM_PRESSURE_EVENTS */
+
+#if CONFIG_MEMORYSTATUS
+extern struct filterops memorystatus_filtops;
+#endif /* CONFIG_MEMORYSTATUS */
+
+extern struct filterops fs_filtops;
+
+extern struct filterops sig_filtops;
+
+/* Timer filter */
+static int filt_timerattach(struct knote *kn);
+static void filt_timerdetach(struct knote *kn);
+static int filt_timer(struct knote *kn, long hint);
+static void filt_timertouch(struct knote *kn, struct kevent64_s *kev,
+    long type);
+static struct filterops timer_filtops = {
+       .f_attach = filt_timerattach,
+       .f_detach = filt_timerdetach,
+       .f_event = filt_timer,
+       .f_touch = filt_timertouch,
+};
+
+/* Helpers */
+static void filt_timerexpire(void *knx, void *param1);
+static int filt_timervalidate(struct knote *kn);
+static void filt_timerupdate(struct knote *kn);
+static void filt_timercancel(struct knote *kn);
+
+#define        TIMER_RUNNING           0x1
+#define        TIMER_CANCELWAIT        0x2
+
+static lck_mtx_t _filt_timerlock;
+static void filt_timerlock(void);
+static void filt_timerunlock(void);
+
+static zone_t knote_zone;
+
+#define        KN_HASH(val, mask)      (((val) ^ (val >> 8)) & (mask))
+
+#if 0
+extern struct filterops aio_filtops;
+#endif
 
-static u_long static_event_id = 0;
+/* Mach portset filter */
+extern struct filterops machport_filtops;
+
+/* User filter */
+static int filt_userattach(struct knote *kn);
+static void filt_userdetach(struct knote *kn);
+static int filt_user(struct knote *kn, long hint);
+static void filt_usertouch(struct knote *kn, struct kevent64_s *kev,
+    long type);
+static struct filterops user_filtops = {
+       .f_attach = filt_userattach,
+       .f_detach = filt_userdetach,
+       .f_event = filt_user,
+       .f_touch = filt_usertouch,
+};
 
 /*
- * Install the protosw's for the NKE manager.  Invoked at
- *  extension load time
+ * Table for all system-defined filters.
  */
-int
-kern_event_init(void)
+static struct filterops *sysfilt_ops[] = {
+       &file_filtops,                  /* EVFILT_READ */
+       &file_filtops,                  /* EVFILT_WRITE */
+#if 0
+       &aio_filtops,                   /* EVFILT_AIO */
+#else
+       &bad_filtops,                   /* EVFILT_AIO */
+#endif
+       &file_filtops,                  /* EVFILT_VNODE */
+       &proc_filtops,                  /* EVFILT_PROC */
+       &sig_filtops,                   /* EVFILT_SIGNAL */
+       &timer_filtops,                 /* EVFILT_TIMER */
+       &machport_filtops,              /* EVFILT_MACHPORT */
+       &fs_filtops,                    /* EVFILT_FS */
+       &user_filtops,                  /* EVFILT_USER */
+       &bad_filtops,                   /* unused */
+#if VM_PRESSURE_EVENTS
+       &vm_filtops,                    /* EVFILT_VM */
+#else
+       &bad_filtops,                   /* EVFILT_VM */
+#endif
+       &file_filtops,                  /* EVFILT_SOCK */
+#if CONFIG_MEMORYSTATUS
+       &memorystatus_filtops,  /* EVFILT_MEMORYSTATUS */
+#else
+       &bad_filtops,                   /* EVFILT_MEMORYSTATUS */
+#endif
+};
+
+/*
+ * kqueue/note lock attributes and implementations
+ *
+ *     kqueues have locks, while knotes have use counts
+ *     Most of the knote state is guarded by the object lock.
+ *     the knote "inuse" count and status use the kqueue lock.
+ */
+lck_grp_attr_t * kq_lck_grp_attr;
+lck_grp_t * kq_lck_grp;
+lck_attr_t * kq_lck_attr;
+
+static inline void
+kqlock(struct kqueue *kq)
+{
+       lck_spin_lock(&kq->kq_lock);
+}
+
+static inline void
+kqunlock(struct kqueue *kq)
+{
+       lck_spin_unlock(&kq->kq_lock);
+}
+
+/*
+ * Convert a kq lock to a knote use referece.
+ *
+ *     If the knote is being dropped, we can't get
+ *     a use reference, so just return with it
+ *     still locked.
+ *     - kq locked at entry
+ *     - unlock on exit if we get the use reference
+ */
+static int
+kqlock2knoteuse(struct kqueue *kq, struct knote *kn)
+{
+       if (kn->kn_status & KN_DROPPING)
+               return (0);
+       kn->kn_inuse++;
+       kqunlock(kq);
+       return (1);
+}
+
+/*
+ * Convert a kq lock to a knote use referece,
+ * but wait for attach and drop events to complete.
+ *
+ *     If the knote is being dropped, we can't get
+ *     a use reference, so just return with it
+ *     still locked.
+ *     - kq locked at entry
+ *     - kq always unlocked on exit
+ */
+static int
+kqlock2knoteusewait(struct kqueue *kq, struct knote *kn)
+{
+       if ((kn->kn_status & (KN_DROPPING | KN_ATTACHING)) != 0) {
+               kn->kn_status |= KN_USEWAIT;
+               wait_queue_assert_wait((wait_queue_t)kq->kq_wqs,
+                   &kn->kn_status, THREAD_UNINT, 0);
+               kqunlock(kq);
+               thread_block(THREAD_CONTINUE_NULL);
+               return (0);
+       }
+       kn->kn_inuse++;
+       kqunlock(kq);
+       return (1);
+}
+
+/*
+ * Convert from a knote use reference back to kq lock.
+ *
+ *     Drop a use reference and wake any waiters if
+ *     this is the last one.
+ *
+ *     The exit return indicates if the knote is
+ *     still alive - but the kqueue lock is taken
+ *     unconditionally.
+ */
+static int
+knoteuse2kqlock(struct kqueue *kq, struct knote *kn)
+{
+       kqlock(kq);
+       if (--kn->kn_inuse == 0) {
+               if ((kn->kn_status & KN_ATTACHING) != 0) {
+                       kn->kn_status &= ~KN_ATTACHING;
+               }
+               if ((kn->kn_status & KN_USEWAIT) != 0) {
+                       kn->kn_status &= ~KN_USEWAIT;
+                       wait_queue_wakeup_all((wait_queue_t)kq->kq_wqs,
+                           &kn->kn_status, THREAD_AWAKENED);
+               }
+       }
+       return ((kn->kn_status & KN_DROPPING) == 0);
+}
+
+/*
+ * Convert a kq lock to a knote drop reference.
+ *
+ *     If the knote is in use, wait for the use count
+ *     to subside.  We first mark our intention to drop
+ *     it - keeping other users from "piling on."
+ *     If we are too late, we have to wait for the
+ *     other drop to complete.
+ *
+ *     - kq locked at entry
+ *     - always unlocked on exit.
+ *     - caller can't hold any locks that would prevent
+ *       the other dropper from completing.
+ */
+static int
+kqlock2knotedrop(struct kqueue *kq, struct knote *kn)
+{
+       int oktodrop;
+
+       oktodrop = ((kn->kn_status & (KN_DROPPING | KN_ATTACHING)) == 0);
+       kn->kn_status &= ~KN_STAYQUEUED;
+       kn->kn_status |= KN_DROPPING;
+       if (oktodrop) {
+               if (kn->kn_inuse == 0) {
+                       kqunlock(kq);
+                       return (oktodrop);
+               }
+       }
+       kn->kn_status |= KN_USEWAIT;
+       wait_queue_assert_wait((wait_queue_t)kq->kq_wqs, &kn->kn_status,
+           THREAD_UNINT, 0);
+       kqunlock(kq);
+       thread_block(THREAD_CONTINUE_NULL);
+       return (oktodrop);
+}
+
+/*
+ * Release a knote use count reference.
+ */
+static void
+knote_put(struct knote *kn)
+{
+       struct kqueue *kq = kn->kn_kq;
+
+       kqlock(kq);
+       if (--kn->kn_inuse == 0) {
+               if ((kn->kn_status & KN_USEWAIT) != 0) {
+                       kn->kn_status &= ~KN_USEWAIT;
+                       wait_queue_wakeup_all((wait_queue_t)kq->kq_wqs,
+                           &kn->kn_status, THREAD_AWAKENED);
+               }
+       }
+       kqunlock(kq);
+}
+
+static int
+filt_fileattach(struct knote *kn)
+{
+       return (fo_kqfilter(kn->kn_fp, kn, vfs_context_current()));
+}
+
+#define        f_flag f_fglob->fg_flag
+#define        f_msgcount f_fglob->fg_msgcount
+#define        f_cred f_fglob->fg_cred
+#define        f_ops f_fglob->fg_ops
+#define        f_offset f_fglob->fg_offset
+#define        f_data f_fglob->fg_data
+
+static void
+filt_kqdetach(struct knote *kn)
+{
+       struct kqueue *kq = (struct kqueue *)kn->kn_fp->f_data;
+
+       kqlock(kq);
+       KNOTE_DETACH(&kq->kq_sel.si_note, kn);
+       kqunlock(kq);
+}
+
+/*ARGSUSED*/
+static int
+filt_kqueue(struct knote *kn, __unused long hint)
+{
+       struct kqueue *kq = (struct kqueue *)kn->kn_fp->f_data;
+
+       kn->kn_data = kq->kq_count;
+       return (kn->kn_data > 0);
+}
+
+static int
+filt_procattach(struct knote *kn)
+{
+       struct proc *p;
+
+       assert(PID_MAX < NOTE_PDATAMASK);
+
+       if ((kn->kn_sfflags & (NOTE_TRACK | NOTE_TRACKERR | NOTE_CHILD)) != 0)
+               return (ENOTSUP);
+
+       p = proc_find(kn->kn_id);
+       if (p == NULL) {
+               return (ESRCH);
+       }
+
+       const int NoteExitStatusBits = NOTE_EXIT | NOTE_EXITSTATUS;
+
+       if ((kn->kn_sfflags & NoteExitStatusBits) == NoteExitStatusBits)
+               do {
+                       pid_t selfpid = proc_selfpid();
+
+                       if (p->p_ppid == selfpid)
+                               break;  /* parent => ok */
+
+                       if ((p->p_lflag & P_LTRACED) != 0 &&
+                           (p->p_oppid == selfpid))
+                               break;  /* parent-in-waiting => ok */
+
+                       proc_rele(p);
+                       return (EACCES);
+               } while (0);
+
+       proc_klist_lock();
+
+       kn->kn_flags |= EV_CLEAR;       /* automatically set */
+       kn->kn_ptr.p_proc = p;          /* store the proc handle */
+
+       KNOTE_ATTACH(&p->p_klist, kn);
+
+       proc_klist_unlock();
+
+       proc_rele(p);
+
+       return (0);
+}
+
+/*
+ * The knote may be attached to a different process, which may exit,
+ * leaving nothing for the knote to be attached to.  In that case,
+ * the pointer to the process will have already been nulled out.
+ */
+static void
+filt_procdetach(struct knote *kn)
+{
+       struct proc *p;
+
+       proc_klist_lock();
+
+       p = kn->kn_ptr.p_proc;
+       if (p != PROC_NULL) {
+               kn->kn_ptr.p_proc = PROC_NULL;
+               KNOTE_DETACH(&p->p_klist, kn);
+       }
+
+       proc_klist_unlock();
+}
+
+static int
+filt_proc(struct knote *kn, long hint)
+{
+       /*
+        * Note: a lot of bits in hint may be obtained from the knote
+        * To free some of those bits, see <rdar://problem/12592988> Freeing up
+        * bits in hint for filt_proc
+        */
+       /* hint is 0 when called from above */
+       if (hint != 0) {
+               u_int event;
+
+               /* ALWAYS CALLED WITH proc_klist_lock when (hint != 0) */
+
+               /*
+                * mask off extra data
+                */
+               event = (u_int)hint & NOTE_PCTRLMASK;
+
+               /*
+                * termination lifecycle events can happen while a debugger
+                * has reparented a process, in which case notifications
+                * should be quashed except to the tracing parent. When
+                * the debugger reaps the child (either via wait4(2) or
+                * process exit), the child will be reparented to the original
+                * parent and these knotes re-fired.
+                */
+               if (event & NOTE_EXIT) {
+                       if ((kn->kn_ptr.p_proc->p_oppid != 0)
+                               && (kn->kn_kq->kq_p->p_pid != kn->kn_ptr.p_proc->p_ppid)) {
+                               /*
+                                * This knote is not for the current ptrace(2) parent, ignore.
+                                */
+                               return 0;
+                       }
+               }                                       
+
+               /*
+                * if the user is interested in this event, record it.
+                */
+               if (kn->kn_sfflags & event)
+                       kn->kn_fflags |= event;
+
+#pragma clang diagnostic push
+#pragma clang diagnostic ignored "-Wdeprecated-declarations"
+               if ((event == NOTE_REAP) || ((event == NOTE_EXIT) && !(kn->kn_sfflags & NOTE_REAP))) {
+                       kn->kn_flags |= (EV_EOF | EV_ONESHOT);
+               }
+#pragma clang diagnostic pop
+
+
+               /*
+                * The kernel has a wrapper in place that returns the same data
+                * as is collected here, in kn_data.  Any changes to how 
+                * NOTE_EXITSTATUS and NOTE_EXIT_DETAIL are collected
+                * should also be reflected in the proc_pidnoteexit() wrapper.
+                */
+               if (event == NOTE_EXIT) {
+                       kn->kn_data = 0;
+                       if ((kn->kn_sfflags & NOTE_EXITSTATUS) != 0) {
+                               kn->kn_fflags |= NOTE_EXITSTATUS;
+                               kn->kn_data |= (hint & NOTE_PDATAMASK);
+                       }
+                       if ((kn->kn_sfflags & NOTE_EXIT_DETAIL) != 0) {
+                               kn->kn_fflags |= NOTE_EXIT_DETAIL;
+                               if ((kn->kn_ptr.p_proc->p_lflag &
+                                   P_LTERM_DECRYPTFAIL) != 0) {
+                                       kn->kn_data |= NOTE_EXIT_DECRYPTFAIL; 
+                               }
+                               if ((kn->kn_ptr.p_proc->p_lflag &
+                                   P_LTERM_JETSAM) != 0) {
+                                       kn->kn_data |= NOTE_EXIT_MEMORY;
+                                       switch (kn->kn_ptr.p_proc->p_lflag &
+                                           P_JETSAM_MASK) {
+                                               case P_JETSAM_VMPAGESHORTAGE:
+                                                       kn->kn_data |= NOTE_EXIT_MEMORY_VMPAGESHORTAGE;
+                                                       break;
+                                               case P_JETSAM_VMTHRASHING:
+                                                       kn->kn_data |= NOTE_EXIT_MEMORY_VMTHRASHING;
+                                                       break;
+                                               case P_JETSAM_FCTHRASHING:
+                                                       kn->kn_data |= NOTE_EXIT_MEMORY_FCTHRASHING;
+                                                       break;
+                                               case P_JETSAM_VNODE:
+                                                       kn->kn_data |= NOTE_EXIT_MEMORY_VNODE;
+                                                       break;
+                                               case P_JETSAM_HIWAT:
+                                                       kn->kn_data |= NOTE_EXIT_MEMORY_HIWAT;
+                                                       break;
+                                               case P_JETSAM_PID:
+                                                       kn->kn_data |= NOTE_EXIT_MEMORY_PID;
+                                                       break;
+                                               case P_JETSAM_IDLEEXIT:
+                                                       kn->kn_data |= NOTE_EXIT_MEMORY_IDLE;
+                                                       break;
+                                       }
+                               }
+                               if ((kn->kn_ptr.p_proc->p_csflags &
+                                   CS_KILLED) != 0) {
+                                       kn->kn_data |= NOTE_EXIT_CSERROR;
+                               }
+                       }
+               }
+       }
+
+       /* atomic check, no locking need when called from above */
+       return (kn->kn_fflags != 0);
+}
+
+#if VM_PRESSURE_EVENTS
+/*
+ * Virtual memory kevents
+ *
+ * author: Matt Jacobson [matthew_jacobson@apple.com]
+ */
+
+static int
+filt_vmattach(struct knote *kn)
+{
+       /*
+        * The note will be cleared once the information has been flushed to
+        * the client. If there is still pressure, we will be re-alerted.
+        */
+       kn->kn_flags |= EV_CLEAR;
+       return (vm_knote_register(kn));
+}
+
+static void
+filt_vmdetach(struct knote *kn)
+{
+       vm_knote_unregister(kn);
+}
+
+static int
+filt_vm(struct knote *kn, long hint)
+{
+       /* hint == 0 means this is just an alive? check (always true) */
+       if (hint != 0) {
+               const pid_t pid = (pid_t)hint;
+               if ((kn->kn_sfflags & NOTE_VM_PRESSURE) &&
+                   (kn->kn_kq->kq_p->p_pid == pid)) {
+                       kn->kn_fflags |= NOTE_VM_PRESSURE;
+               }
+       }
+
+       return (kn->kn_fflags != 0);
+}
+#endif /* VM_PRESSURE_EVENTS */
+
+/*
+ * filt_timervalidate - process data from user
+ *
+ *     Converts to either interval or deadline format.
+ *
+ *     The saved-data field in the knote contains the
+ *     time value.  The saved filter-flags indicates
+ *     the unit of measurement.
+ *
+ *     After validation, either the saved-data field
+ *     contains the interval in absolute time, or ext[0]
+ *     contains the expected deadline. If that deadline
+ *     is in the past, ext[0] is 0.
+ *
+ *     Returns EINVAL for unrecognized units of time.
+ *
+ *     Timer filter lock is held.
+ *
+ */
+static int
+filt_timervalidate(struct knote *kn)
+{
+       uint64_t multiplier;
+       uint64_t raw = 0;
+
+       switch (kn->kn_sfflags & (NOTE_SECONDS|NOTE_USECONDS|NOTE_NSECONDS)) {
+       case NOTE_SECONDS:
+               multiplier = NSEC_PER_SEC;
+               break;
+       case NOTE_USECONDS:
+               multiplier = NSEC_PER_USEC;
+               break;
+       case NOTE_NSECONDS:
+               multiplier = 1;
+               break;
+       case 0: /* milliseconds (default) */
+               multiplier = NSEC_PER_SEC / 1000;
+               break;
+       default:
+               return (EINVAL);
+       }
+
+       /* transform the slop delta(leeway) in kn_ext[1] if passed to same time scale */
+       if(kn->kn_sfflags & NOTE_LEEWAY){
+               nanoseconds_to_absolutetime((uint64_t)kn->kn_ext[1] * multiplier, &raw);
+               kn->kn_ext[1] = raw;
+       }
+
+       nanoseconds_to_absolutetime((uint64_t)kn->kn_sdata * multiplier, &raw);
+
+       kn->kn_ext[0] = 0;
+       kn->kn_sdata = 0;
+
+       if (kn->kn_sfflags & NOTE_ABSOLUTE) {
+               clock_sec_t seconds;
+               clock_nsec_t nanoseconds;
+               uint64_t now;
+
+               clock_get_calendar_nanotime(&seconds, &nanoseconds);
+               nanoseconds_to_absolutetime((uint64_t)seconds * NSEC_PER_SEC +
+                   nanoseconds, &now);
+
+               if (raw < now) {
+                       /* time has already passed */
+                       kn->kn_ext[0] = 0;
+               } else {
+                       raw -= now;
+                       clock_absolutetime_interval_to_deadline(raw,
+                           &kn->kn_ext[0]);
+               }
+       } else {
+               kn->kn_sdata = raw;
+       }
+
+       return (0);
+}
+
+/*
+ * filt_timerupdate - compute the next deadline
+ *
+ *     Repeating timers store their interval in kn_sdata. Absolute
+ *     timers have already calculated the deadline, stored in ext[0].
+ *
+ *     On return, the next deadline (or zero if no deadline is needed)
+ *     is stored in kn_ext[0].
+ *
+ *     Timer filter lock is held.
+ */
+static void
+filt_timerupdate(struct knote *kn)
+{
+       /* if there's no interval, deadline is just in kn_ext[0] */
+       if (kn->kn_sdata == 0)
+               return;
+
+       /* if timer hasn't fired before, fire in interval nsecs */
+       if (kn->kn_ext[0] == 0) {
+               clock_absolutetime_interval_to_deadline(kn->kn_sdata,
+                   &kn->kn_ext[0]);
+       } else {
+               /*
+                * If timer has fired before, schedule the next pop
+                * relative to the last intended deadline.
+                *
+                * We could check for whether the deadline has expired,
+                * but the thread call layer can handle that.
+                */
+               kn->kn_ext[0] += kn->kn_sdata;
+       }
+}
+
+/*
+ * filt_timerexpire - the timer callout routine
+ *
+ * Just propagate the timer event into the knote
+ * filter routine (by going through the knote
+ * synchronization point).  Pass a hint to
+ * indicate this is a real event, not just a
+ * query from above.
+ */
+static void
+filt_timerexpire(void *knx, __unused void *spare)
+{
+       struct klist timer_list;
+       struct knote *kn = knx;
+
+       filt_timerlock();
+
+       kn->kn_hookid &= ~TIMER_RUNNING;
+
+       /* no "object" for timers, so fake a list */
+       SLIST_INIT(&timer_list);
+       SLIST_INSERT_HEAD(&timer_list, kn, kn_selnext);
+       KNOTE(&timer_list, 1);
+
+       /* if someone is waiting for timer to pop */
+       if (kn->kn_hookid & TIMER_CANCELWAIT) {
+               struct kqueue *kq = kn->kn_kq;
+               wait_queue_wakeup_all((wait_queue_t)kq->kq_wqs, &kn->kn_hook,
+                   THREAD_AWAKENED);
+       }
+
+       filt_timerunlock();
+}
+
+/*
+ * Cancel a running timer (or wait for the pop).
+ * Timer filter lock is held.
+ */
+static void
+filt_timercancel(struct knote *kn)
+{
+       struct kqueue *kq = kn->kn_kq;
+       thread_call_t callout = kn->kn_hook;
+       boolean_t cancelled;
+
+       if (kn->kn_hookid & TIMER_RUNNING) {
+               /* cancel the callout if we can */
+               cancelled = thread_call_cancel(callout);
+               if (cancelled) {
+                       kn->kn_hookid &= ~TIMER_RUNNING;
+               } else {
+                       /* we have to wait for the expire routine.  */
+                       kn->kn_hookid |= TIMER_CANCELWAIT;
+                       wait_queue_assert_wait((wait_queue_t)kq->kq_wqs,
+                           &kn->kn_hook, THREAD_UNINT, 0);
+                       filt_timerunlock();
+                       thread_block(THREAD_CONTINUE_NULL);
+                       filt_timerlock();
+                       assert((kn->kn_hookid & TIMER_RUNNING) == 0);
+               }
+       }
+}
+
+/*
+ * Allocate a thread call for the knote's lifetime, and kick off the timer.
+ */
+static int
+filt_timerattach(struct knote *kn)
 {
-    int retval;
+       thread_call_t callout;
+       int error;
+
+       callout = thread_call_allocate(filt_timerexpire, kn);
+       if (NULL == callout)
+               return (ENOMEM);
+
+       filt_timerlock();
+       error = filt_timervalidate(kn);
+       if (error != 0) {
+               filt_timerunlock();
+               return (error);
+       }
+
+       kn->kn_hook = (void*)callout;
+       kn->kn_hookid = 0;
 
-    if ((retval = net_add_proto(eventsw, &systemdomain)) == 0)
-            return(KERN_SUCCESS);
-    
-    log(LOG_WARNING, "Can't install kernel events protocol (%d)\n", retval);
-    return(retval);
+       /* absolute=EV_ONESHOT */
+       if (kn->kn_sfflags & NOTE_ABSOLUTE)
+               kn->kn_flags |= EV_ONESHOT;
+
+       filt_timerupdate(kn);
+       if (kn->kn_ext[0]) {
+               kn->kn_flags |= EV_CLEAR;
+               unsigned int timer_flags = 0;
+               if (kn->kn_sfflags & NOTE_CRITICAL)
+                       timer_flags |= THREAD_CALL_DELAY_USER_CRITICAL;
+               else if (kn->kn_sfflags & NOTE_BACKGROUND)
+                       timer_flags |= THREAD_CALL_DELAY_USER_BACKGROUND;
+               else
+                       timer_flags |= THREAD_CALL_DELAY_USER_NORMAL;
+
+               if (kn->kn_sfflags & NOTE_LEEWAY)
+                       timer_flags |= THREAD_CALL_DELAY_LEEWAY;
+
+               thread_call_enter_delayed_with_leeway(callout, NULL,
+                               kn->kn_ext[0], kn->kn_ext[1], timer_flags);
+
+               kn->kn_hookid |= TIMER_RUNNING;
+       } else {
+               /* fake immediate */
+               kn->kn_data = 1;
+       }
+
+       filt_timerunlock();
+       return (0);
 }
 
-int kev_attach(struct socket *so, int proto, struct proc *p)
+/*
+ * Shut down the timer if it's running, and free the callout.
+ */
+static void
+filt_timerdetach(struct knote *kn)
 {
-     int error;
-     struct kern_event_pcb  *ev_pcb;
+       thread_call_t callout;
 
-     ev_pcb = _MALLOC(sizeof(struct kern_event_pcb), M_PCB, M_WAITOK);
-     if (ev_pcb == 0)
-         return ENOBUFS;
+       filt_timerlock();
 
-     ev_pcb->ev_socket = so;
-     ev_pcb->vendor_code_filter = 0xffffffff;
+       callout = (thread_call_t)kn->kn_hook;
+       filt_timercancel(kn);
 
-     so->so_pcb = (caddr_t) ev_pcb;
-     LIST_INSERT_HEAD(&kern_event_head, ev_pcb, ev_link);
-     error = soreserve(so, KEV_SNDSPACE, KEV_RECVSPACE);
-     if (error)
-         return error;
+       filt_timerunlock();
 
-     return 0;
+       thread_call_free(callout);
 }
 
 
-int kev_detach(struct socket *so)
+
+static int
+filt_timer(struct knote *kn, long hint)
 {
-     struct kern_event_pcb *ev_pcb = (struct kern_event_pcb *) so->so_pcb;
+       int result;
+
+       if (hint) {
+               /* real timer pop -- timer lock held by filt_timerexpire */
+               kn->kn_data++;
+
+               if (((kn->kn_hookid & TIMER_CANCELWAIT) == 0) &&
+                               ((kn->kn_flags & EV_ONESHOT) == 0)) {
+
+                       /* evaluate next time to fire */
+                       filt_timerupdate(kn);
+
+                       if (kn->kn_ext[0]) {
+                               unsigned int timer_flags = 0;
+
+                               /* keep the callout and re-arm */
+                               if (kn->kn_sfflags & NOTE_CRITICAL)
+                                       timer_flags |= THREAD_CALL_DELAY_USER_CRITICAL;
+                               else if (kn->kn_sfflags & NOTE_BACKGROUND)
+                                       timer_flags |= THREAD_CALL_DELAY_USER_BACKGROUND;
+                               else
+                                       timer_flags |= THREAD_CALL_DELAY_USER_NORMAL;
+
+                               if (kn->kn_sfflags & NOTE_LEEWAY)
+                                       timer_flags |= THREAD_CALL_DELAY_LEEWAY;
+
+                               thread_call_enter_delayed_with_leeway(kn->kn_hook, NULL,
+                                               kn->kn_ext[0], kn->kn_ext[1], timer_flags);
+
+                               kn->kn_hookid |= TIMER_RUNNING;
+                       }
+               }
 
-     LIST_REMOVE(ev_pcb, ev_link);
-     if (ev_pcb)
-         FREE(ev_pcb, M_PCB);
+               return (1);
+       }
 
-     return 0;
+       /* user-query */
+       filt_timerlock();
+
+       result = (kn->kn_data != 0);
+
+       filt_timerunlock();
+
+       return (result);
 }
 
 
-int  kev_post_msg(struct kev_msg *event_msg)
+/*
+ * filt_timertouch - update knote with new user input
+ *
+ * Cancel and restart the timer based on new user data. When
+ * the user picks up a knote, clear the count of how many timer
+ * pops have gone off (in kn_data).
+ */
+static void
+filt_timertouch(struct knote *kn, struct kevent64_s *kev, long type)
 {
-     struct mbuf *m, *m2;
-     struct kern_event_pcb  *ev_pcb;
-     struct kern_event_msg  *ev;
-     char              *tmp;
-     int               total_size;
-     int               i;
+       int error;
+       filt_timerlock();
 
+       switch (type) {
+       case EVENT_REGISTER:
+               /* cancel current call */
+               filt_timercancel(kn);
 
-     m = m_get(M_DONTWAIT, MT_DATA);
-     if (m == 0)
-         return ENOBUFS;
+               /* recalculate deadline */
+               kn->kn_sdata = kev->data;
+               kn->kn_sfflags = kev->fflags;
+               kn->kn_ext[0] = kev->ext[0];
+               kn->kn_ext[1] = kev->ext[1];
 
-     ev = mtod(m, struct kern_event_msg *);
-     total_size = KEV_MSG_HEADER_SIZE;
+               error = filt_timervalidate(kn);
+               if (error) {
+                       /* no way to report error, so mark it in the knote */
+                       kn->kn_flags |= EV_ERROR;
+                       kn->kn_data = error;
+                       break;
+               }
 
-     tmp = (char *) &ev->event_data[0];
-     for (i = 0; i < 5; i++) {
-         if (event_msg->dv[i].data_length == 0)
-              break;
+               /* start timer if necessary */
+               filt_timerupdate(kn);
 
-         total_size += event_msg->dv[i].data_length;
-         bcopy(event_msg->dv[i].data_ptr, tmp, 
-               event_msg->dv[i].data_length);
-         tmp += event_msg->dv[i].data_length;
-     }
+               if (kn->kn_ext[0]) {
+                       unsigned int timer_flags = 0;
+                       if (kn->kn_sfflags & NOTE_CRITICAL)
+                               timer_flags |= THREAD_CALL_DELAY_USER_CRITICAL;
+                       else if (kn->kn_sfflags & NOTE_BACKGROUND)
+                               timer_flags |= THREAD_CALL_DELAY_USER_BACKGROUND;
+                       else
+                               timer_flags |= THREAD_CALL_DELAY_USER_NORMAL;
 
+                       if (kn->kn_sfflags & NOTE_LEEWAY)
+                               timer_flags |= THREAD_CALL_DELAY_LEEWAY;
 
-     ev->id = ++static_event_id;
-     ev->total_size   = total_size;
-     ev->vendor_code  = event_msg->vendor_code;
-     ev->kev_class    = event_msg->kev_class;
-     ev->kev_subclass = event_msg->kev_subclass;
-     ev->event_code   = event_msg->event_code;
+                       thread_call_enter_delayed_with_leeway(kn->kn_hook, NULL,
+                                       kn->kn_ext[0], kn->kn_ext[1], timer_flags);
 
-     m->m_len = total_size;
-     for (ev_pcb = LIST_FIRST(&kern_event_head); 
-         ev_pcb; 
-         ev_pcb = LIST_NEXT(ev_pcb, ev_link)) {
+                       kn->kn_hookid |= TIMER_RUNNING;
+               } else {
+                       /* pretend the timer has fired */
+                       kn->kn_data = 1;
+               }
 
-         if (ev_pcb->vendor_code_filter != KEV_ANY_VENDOR) {
-              if (ev_pcb->vendor_code_filter != ev->vendor_code)
-                   continue;
+               break;
 
-              if (ev_pcb->class_filter != KEV_ANY_CLASS) {
-                   if (ev_pcb->class_filter != ev->kev_class)
-                        continue;
+       case EVENT_PROCESS:
+               /* reset the timer pop count in kn_data */
+               *kev = kn->kn_kevent;
+               kev->ext[0] = 0;
+               kn->kn_data = 0;
+               if (kn->kn_flags & EV_CLEAR)
+                       kn->kn_fflags = 0;
+               break;
+       default:
+               panic("%s: - invalid type (%ld)", __func__, type);
+               break;
+       }
 
-                   if ((ev_pcb->subclass_filter != KEV_ANY_SUBCLASS) &&
-                       (ev_pcb->subclass_filter != ev->kev_subclass))
-                        continue;
-              }
-         }
+       filt_timerunlock();
+}
 
-         m2 = m_copym(m, 0, m->m_len, M_NOWAIT);
-         if (m2 == 0) {
-              m_free(m);
-              return ENOBUFS;
-         }
+static void
+filt_timerlock(void)
+{
+       lck_mtx_lock(&_filt_timerlock);
+}
 
-         sbappendrecord(&ev_pcb->ev_socket->so_rcv, m2);
-         sorwakeup(ev_pcb->ev_socket);
-     }
+static void
+filt_timerunlock(void)
+{
+       lck_mtx_unlock(&_filt_timerlock);
+}
 
+static int
+filt_userattach(struct knote *kn)
+{
+       /* EVFILT_USER knotes are not attached to anything in the kernel */
+       kn->kn_hook = NULL;
+       if (kn->kn_fflags & NOTE_TRIGGER) {
+               kn->kn_hookid = 1;
+       } else {
+               kn->kn_hookid = 0;
+       }
+       return (0);
+}
 
-     m_free(m);
-     return 0;
+static void
+filt_userdetach(__unused struct knote *kn)
+{
+       /* EVFILT_USER knotes are not attached to anything in the kernel */
 }
 
+static int
+filt_user(struct knote *kn, __unused long hint)
+{
+       return (kn->kn_hookid);
+}
 
-int kev_control(so, cmd, data, ifp, p)
-    struct socket *so;
-    u_long cmd;
-    caddr_t data;
-    register struct ifnet *ifp;
-    struct proc *p;
+static void
+filt_usertouch(struct knote *kn, struct kevent64_s *kev, long type)
 {
-     struct kev_request *kev_req = (struct kev_request *) data;
-     int  stat = 0;
-     struct kern_event_pcb  *ev_pcb;
-     u_long  *id_value = (u_long *) data;
+       uint32_t ffctrl;
+       switch (type) {
+       case EVENT_REGISTER:
+               if (kev->fflags & NOTE_TRIGGER) {
+                       kn->kn_hookid = 1;
+               }
 
+               ffctrl = kev->fflags & NOTE_FFCTRLMASK;
+               kev->fflags &= NOTE_FFLAGSMASK;
+               switch (ffctrl) {
+               case NOTE_FFNOP:
+                       break;
+               case NOTE_FFAND:
+                       OSBitAndAtomic(kev->fflags, &kn->kn_sfflags);
+                       break;
+               case NOTE_FFOR:
+                       OSBitOrAtomic(kev->fflags, &kn->kn_sfflags);
+                       break;
+               case NOTE_FFCOPY:
+                       kn->kn_sfflags = kev->fflags;
+                       break;
+               }
+               kn->kn_sdata = kev->data;
+               break;
+       case EVENT_PROCESS:
+               *kev = kn->kn_kevent;
+               kev->fflags = (volatile UInt32)kn->kn_sfflags;
+               kev->data = kn->kn_sdata;
+               if (kn->kn_flags & EV_CLEAR) {
+                       kn->kn_hookid = 0;
+                       kn->kn_data = 0;
+                       kn->kn_fflags = 0;
+               }
+               break;
+       default:
+               panic("%s: - invalid type (%ld)", __func__, type);
+               break;
+       }
+}
 
-     switch (cmd) {
+/*
+ * JMM - placeholder for not-yet-implemented filters
+ */
+static int
+filt_badattach(__unused struct knote *kn)
+{
+       return (ENOTSUP);
+}
 
-     case SIOCGKEVID:
-         *id_value = static_event_id;
-         break;
+struct kqueue *
+kqueue_alloc(struct proc *p)
+{
+       struct filedesc *fdp = p->p_fd;
+       struct kqueue *kq;
 
-     case SIOCSKEVFILT:
-         ev_pcb = (struct kern_event_pcb *) so->so_pcb;
-         ev_pcb->vendor_code_filter = kev_req->vendor_code;
-         ev_pcb->class_filter     = kev_req->kev_class;
-         ev_pcb->subclass_filter  = kev_req->kev_subclass;
-         break;
+       MALLOC_ZONE(kq, struct kqueue *, sizeof (struct kqueue), M_KQUEUE,
+           M_WAITOK);
+       if (kq != NULL) {
+               wait_queue_set_t wqs;
 
-     case SIOCGKEVFILT:
-         ev_pcb = (struct kern_event_pcb *) so->so_pcb;
-         kev_req->vendor_code = ev_pcb->vendor_code_filter;
-         kev_req->kev_class   = ev_pcb->class_filter;
-         kev_req->kev_subclass = ev_pcb->subclass_filter;
-         break;
+               wqs = wait_queue_set_alloc(SYNC_POLICY_FIFO |
+                   SYNC_POLICY_PREPOST);
+               if (wqs != NULL) {
+                       bzero(kq, sizeof (struct kqueue));
+                       lck_spin_init(&kq->kq_lock, kq_lck_grp, kq_lck_attr);
+                       TAILQ_INIT(&kq->kq_head);
+                       kq->kq_wqs = wqs;
+                       kq->kq_p = p;
+               } else {
+                       FREE_ZONE(kq, sizeof (struct kqueue), M_KQUEUE);
+                       kq = NULL;
+               }
+       }
 
-     default:
-         return EOPNOTSUPP;
-     }
+       if (fdp->fd_knlistsize < 0) {
+               proc_fdlock(p);
+               if (fdp->fd_knlistsize < 0)
+                       fdp->fd_knlistsize = 0; /* this process has had a kq */
+               proc_fdunlock(p);
+       }
 
-     return 0;
+       return (kq);
 }
 
+/*
+ * kqueue_dealloc - detach all knotes from a kqueue and free it
+ *
+ *     We walk each list looking for knotes referencing this
+ *     this kqueue.  If we find one, we try to drop it.  But
+ *     if we fail to get a drop reference, that will wait
+ *     until it is dropped.  So, we can just restart again
+ *     safe in the assumption that the list will eventually
+ *     not contain any more references to this kqueue (either
+ *     we dropped them all, or someone else did).
+ *
+ *     Assumes no new events are being added to the kqueue.
+ *     Nothing locked on entry or exit.
+ */
+void
+kqueue_dealloc(struct kqueue *kq)
+{
+       struct proc *p = kq->kq_p;
+       struct filedesc *fdp = p->p_fd;
+       struct knote *kn;
+       int i;
+
+       proc_fdlock(p);
+       for (i = 0; i < fdp->fd_knlistsize; i++) {
+               kn = SLIST_FIRST(&fdp->fd_knlist[i]);
+               while (kn != NULL) {
+                       if (kq == kn->kn_kq) {
+                               kqlock(kq);
+                               proc_fdunlock(p);
+                               /* drop it ourselves or wait */
+                               if (kqlock2knotedrop(kq, kn)) {
+                                       kn->kn_fop->f_detach(kn);
+                                       knote_drop(kn, p);
+                               }
+                               proc_fdlock(p);
+                               /* start over at beginning of list */
+                               kn = SLIST_FIRST(&fdp->fd_knlist[i]);
+                               continue;
+                       }
+                       kn = SLIST_NEXT(kn, kn_link);
+               }
+       }
+       if (fdp->fd_knhashmask != 0) {
+               for (i = 0; i < (int)fdp->fd_knhashmask + 1; i++) {
+                       kn = SLIST_FIRST(&fdp->fd_knhash[i]);
+                       while (kn != NULL) {
+                               if (kq == kn->kn_kq) {
+                                       kqlock(kq);
+                                       proc_fdunlock(p);
+                                       /* drop it ourselves or wait */
+                                       if (kqlock2knotedrop(kq, kn)) {
+                                               kn->kn_fop->f_detach(kn);
+                                               knote_drop(kn, p);
+                                       }
+                                       proc_fdlock(p);
+                                       /* start over at beginning of list */
+                                       kn = SLIST_FIRST(&fdp->fd_knhash[i]);
+                                       continue;
+                               }
+                               kn = SLIST_NEXT(kn, kn_link);
+                       }
+               }
+       }
+       proc_fdunlock(p);
 
-struct pr_usrreqs event_usrreqs = {
-     pru_abort_notsupp, pru_accept_notsupp, kev_attach, pru_bind_notsupp, pru_connect_notsupp,
-     pru_connect2_notsupp, kev_control, kev_detach, pru_disconnect_notsupp,
-     pru_listen_notsupp, pru_peeraddr_notsupp, pru_rcvd_notsupp, pru_rcvoob_notsupp,
-     pru_send_notsupp, pru_sense_null, pru_shutdown_notsupp, pru_sockaddr_notsupp,
-     pru_sosend_notsupp, soreceive, sopoll
-};
+       /*
+        * before freeing the wait queue set for this kqueue,
+        * make sure it is unlinked from all its containing (select) sets.
+        */
+       wait_queue_unlink_all((wait_queue_t)kq->kq_wqs);
+       wait_queue_set_free(kq->kq_wqs);
+       lck_spin_destroy(&kq->kq_lock, kq_lck_grp);
+       FREE_ZONE(kq, sizeof (struct kqueue), M_KQUEUE);
+}
 
+int
+kqueue_body(struct proc *p, fp_allocfn_t fp_zalloc, void *cra, int32_t *retval)
+{
+       struct kqueue *kq;
+       struct fileproc *fp;
+       int fd, error;
 
+       error = falloc_withalloc(p,
+           &fp, &fd, vfs_context_current(), fp_zalloc, cra);
+       if (error) {
+               return (error);
+       }
 
+       kq = kqueue_alloc(p);
+       if (kq == NULL) {
+               fp_free(p, fd, fp);
+               return (ENOMEM);
+       }
+
+       fp->f_flag = FREAD | FWRITE;
+       fp->f_ops = &kqueueops;
+       fp->f_data = kq;
+
+       proc_fdlock(p);
+       *fdflags(p, fd) |= UF_EXCLOSE;
+       procfdtbl_releasefd(p, fd, NULL);
+       fp_drop(p, fd, fp, 1);
+       proc_fdunlock(p);
+
+       *retval = fd;
+       return (error);
+}
+
+int
+kqueue(struct proc *p, __unused struct kqueue_args *uap, int32_t *retval)
+{
+       return (kqueue_body(p, fileproc_alloc_init, NULL, retval));
+}
+
+static int
+kevent_copyin(user_addr_t *addrp, struct kevent64_s *kevp, struct proc *p,
+    int iskev64)
+{
+       int advance;
+       int error;
+
+       if (iskev64) {
+               advance = sizeof (struct kevent64_s);
+               error = copyin(*addrp, (caddr_t)kevp, advance);
+       } else if (IS_64BIT_PROCESS(p)) {
+               struct user64_kevent kev64;
+               bzero(kevp, sizeof (struct kevent64_s));
+
+               advance = sizeof (kev64);
+               error = copyin(*addrp, (caddr_t)&kev64, advance);
+               if (error)
+                       return (error);
+               kevp->ident = kev64.ident;
+               kevp->filter = kev64.filter;
+               kevp->flags = kev64.flags;
+               kevp->fflags = kev64.fflags;
+               kevp->data = kev64.data;
+               kevp->udata = kev64.udata;
+       } else {
+               struct user32_kevent kev32;
+               bzero(kevp, sizeof (struct kevent64_s));
+
+               advance = sizeof (kev32);
+               error = copyin(*addrp, (caddr_t)&kev32, advance);
+               if (error)
+                       return (error);
+               kevp->ident = (uintptr_t)kev32.ident;
+               kevp->filter = kev32.filter;
+               kevp->flags = kev32.flags;
+               kevp->fflags = kev32.fflags;
+               kevp->data = (intptr_t)kev32.data;
+               kevp->udata = CAST_USER_ADDR_T(kev32.udata);
+       }
+       if (!error)
+               *addrp += advance;
+       return (error);
+}
+
+static int
+kevent_copyout(struct kevent64_s *kevp, user_addr_t *addrp, struct proc *p,
+    int iskev64)
+{
+       int advance;
+       int error;
+
+       if (iskev64) {
+               advance = sizeof (struct kevent64_s);
+               error = copyout((caddr_t)kevp, *addrp, advance);
+       } else if (IS_64BIT_PROCESS(p)) {
+               struct user64_kevent kev64;
+
+               /*
+                * deal with the special case of a user-supplied
+                * value of (uintptr_t)-1.
+                */
+               kev64.ident = (kevp->ident == (uintptr_t)-1) ?
+                   (uint64_t)-1LL : (uint64_t)kevp->ident;
+
+               kev64.filter = kevp->filter;
+               kev64.flags = kevp->flags;
+               kev64.fflags = kevp->fflags;
+               kev64.data = (int64_t) kevp->data;
+               kev64.udata = kevp->udata;
+               advance = sizeof (kev64);
+               error = copyout((caddr_t)&kev64, *addrp, advance);
+       } else {
+               struct user32_kevent kev32;
+
+               kev32.ident = (uint32_t)kevp->ident;
+               kev32.filter = kevp->filter;
+               kev32.flags = kevp->flags;
+               kev32.fflags = kevp->fflags;
+               kev32.data = (int32_t)kevp->data;
+               kev32.udata = kevp->udata;
+               advance = sizeof (kev32);
+               error = copyout((caddr_t)&kev32, *addrp, advance);
+       }
+       if (!error)
+               *addrp += advance;
+       return (error);
+}
+
+/*
+ * kevent_continue - continue a kevent syscall after blocking
+ *
+ *     assume we inherit a use count on the kq fileglob.
+ */
+
+static void
+kevent_continue(__unused struct kqueue *kq, void *data, int error)
+{
+       struct _kevent *cont_args;
+       struct fileproc *fp;
+       int32_t *retval;
+       int noutputs;
+       int fd;
+       struct proc *p = current_proc();
+
+       cont_args = (struct _kevent *)data;
+       noutputs = cont_args->eventout;
+       retval = cont_args->retval;
+       fd = cont_args->fd;
+       fp = cont_args->fp;
+
+       fp_drop(p, fd, fp, 0);
+
+       /* don't restart after signals... */
+       if (error == ERESTART)
+               error = EINTR;
+       else if (error == EWOULDBLOCK)
+               error = 0;
+       if (error == 0)
+               *retval = noutputs;
+       unix_syscall_return(error);
+}
+
+/*
+ * kevent - [syscall] register and wait for kernel events
+ *
+ */
+int
+kevent(struct proc *p, struct kevent_args *uap, int32_t *retval)
+{
+       return (kevent_internal(p,
+           0,
+           uap->changelist,
+           uap->nchanges,
+           uap->eventlist,
+           uap->nevents,
+           uap->fd,
+           uap->timeout,
+           0, /* no flags from old kevent() call */
+           retval));
+}
+
+int
+kevent64(struct proc *p, struct kevent64_args *uap, int32_t *retval)
+{
+       return (kevent_internal(p,
+           1,
+           uap->changelist,
+           uap->nchanges,
+           uap->eventlist,
+           uap->nevents,
+           uap->fd,
+           uap->timeout,
+           uap->flags,
+           retval));
+}
+
+static int
+kevent_internal(struct proc *p, int iskev64, user_addr_t changelist,
+    int nchanges, user_addr_t ueventlist, int nevents, int fd,
+    user_addr_t utimeout, __unused unsigned int flags,
+    int32_t *retval)
+{
+       struct _kevent *cont_args;
+       uthread_t ut;
+       struct kqueue *kq;
+       struct fileproc *fp;
+       struct kevent64_s kev;
+       int error, noutputs;
+       struct timeval atv;
+
+       /* convert timeout to absolute - if we have one */
+       if (utimeout != USER_ADDR_NULL) {
+               struct timeval rtv;
+               if (IS_64BIT_PROCESS(p)) {
+                       struct user64_timespec ts;
+                       error = copyin(utimeout, &ts, sizeof(ts));
+                       if ((ts.tv_sec & 0xFFFFFFFF00000000ull) != 0)
+                               error = EINVAL;
+                       else
+                               TIMESPEC_TO_TIMEVAL(&rtv, &ts);
+               } else {
+                       struct user32_timespec ts;
+                       error = copyin(utimeout, &ts, sizeof(ts));
+                       TIMESPEC_TO_TIMEVAL(&rtv, &ts);
+               }
+               if (error)
+                       return (error);
+               if (itimerfix(&rtv))
+                       return (EINVAL);
+               getmicrouptime(&atv);
+               timevaladd(&atv, &rtv);
+       } else {
+               atv.tv_sec = 0;
+               atv.tv_usec = 0;
+       }
+
+       /* get a usecount for the kq itself */
+       if ((error = fp_getfkq(p, fd, &fp, &kq)) != 0)
+               return (error);
+
+       /* each kq should only be used for events of one type */
+       kqlock(kq);
+       if (kq->kq_state & (KQ_KEV32 | KQ_KEV64)) {
+               if (((iskev64 && (kq->kq_state & KQ_KEV32)) ||
+                       (!iskev64 && (kq->kq_state & KQ_KEV64)))) {
+                       error = EINVAL;
+                       kqunlock(kq);
+                       goto errorout;
+               }
+       } else {
+               kq->kq_state |= (iskev64 ? KQ_KEV64 : KQ_KEV32);
+       }
+       kqunlock(kq);
+
+       /* register all the change requests the user provided... */
+       noutputs = 0;
+       while (nchanges > 0 && error == 0) {
+               error = kevent_copyin(&changelist, &kev, p, iskev64);
+               if (error)
+                       break;
+
+               kev.flags &= ~EV_SYSFLAGS;
+               error = kevent_register(kq, &kev, p);
+               if ((error || (kev.flags & EV_RECEIPT)) && nevents > 0) {
+                       kev.flags = EV_ERROR;
+                       kev.data = error;
+                       error = kevent_copyout(&kev, &ueventlist, p, iskev64);
+                       if (error == 0) {
+                               nevents--;
+                               noutputs++;
+                       }
+               }
+               nchanges--;
+       }
+
+       /* store the continuation/completion data in the uthread */
+       ut = (uthread_t)get_bsdthread_info(current_thread());
+       cont_args = &ut->uu_kevent.ss_kevent;
+       cont_args->fp = fp;
+       cont_args->fd = fd;
+       cont_args->retval = retval;
+       cont_args->eventlist = ueventlist;
+       cont_args->eventcount = nevents;
+       cont_args->eventout = noutputs;
+       cont_args->eventsize = iskev64;
+
+       if (nevents > 0 && noutputs == 0 && error == 0)
+               error = kqueue_scan(kq, kevent_callback,
+                   kevent_continue, cont_args,
+                   &atv, p);
+       kevent_continue(kq, cont_args, error);
+
+errorout:
+       fp_drop(p, fd, fp, 0);
+       return (error);
+}
+
+
+/*
+ * kevent_callback - callback for each individual event
+ *
+ * called with nothing locked
+ * caller holds a reference on the kqueue
+ */
+static int
+kevent_callback(__unused struct kqueue *kq, struct kevent64_s *kevp,
+    void *data)
+{
+       struct _kevent *cont_args;
+       int error;
+       int iskev64;
+
+       cont_args = (struct _kevent *)data;
+       assert(cont_args->eventout < cont_args->eventcount);
+
+       iskev64 = cont_args->eventsize;
+
+       /*
+        * Copy out the appropriate amount of event data for this user.
+        */
+       error = kevent_copyout(kevp, &cont_args->eventlist, current_proc(),
+           iskev64);
+
+       /*
+        * If there isn't space for additional events, return
+        * a harmless error to stop the processing here
+        */
+       if (error == 0 && ++cont_args->eventout == cont_args->eventcount)
+               error = EWOULDBLOCK;
+       return (error);
+}
+
+/*
+ * kevent_description - format a description of a kevent for diagnostic output
+ *
+ * called with a 128-byte string buffer
+ */
+
+char *
+kevent_description(struct kevent64_s *kevp, char *s, size_t n)
+{
+       snprintf(s, n,
+           "kevent="
+           "{.ident=%#llx, .filter=%d, .flags=%#x, .fflags=%#x, .data=%#llx, .udata=%#llx, .ext[0]=%#llx, .ext[1]=%#llx}",
+           kevp->ident,
+           kevp->filter,
+           kevp->flags,
+           kevp->fflags,
+           kevp->data,
+           kevp->udata,
+           kevp->ext[0],
+           kevp->ext[1]);
+
+       return (s);
+}
+
+/*
+ * kevent_register - add a new event to a kqueue
+ *
+ *     Creates a mapping between the event source and
+ *     the kqueue via a knote data structure.
+ *
+ *     Because many/most the event sources are file
+ *     descriptor related, the knote is linked off
+ *     the filedescriptor table for quick access.
+ *
+ *     called with nothing locked
+ *     caller holds a reference on the kqueue
+ */
+
+int
+kevent_register(struct kqueue *kq, struct kevent64_s *kev,
+    __unused struct proc *ctxp)
+{
+       struct proc *p = kq->kq_p;
+       struct filedesc *fdp = p->p_fd;
+       struct filterops *fops;
+       struct fileproc *fp = NULL;
+       struct knote *kn = NULL;
+       int error = 0;
+
+       if (kev->filter < 0) {
+               if (kev->filter + EVFILT_SYSCOUNT < 0)
+                       return (EINVAL);
+               fops = sysfilt_ops[~kev->filter];       /* to 0-base index */
+       } else {
+               /*
+                * XXX
+                * filter attach routine is responsible for insuring that
+                * the identifier can be attached to it.
+                */
+               printf("unknown filter: %d\n", kev->filter);
+               return (EINVAL);
+       }
+
+restart:
+       /* this iocount needs to be dropped if it is not registered */
+       proc_fdlock(p);
+       if (fops->f_isfd && (error = fp_lookup(p, kev->ident, &fp, 1)) != 0) {
+               proc_fdunlock(p);
+               return (error);
+       }
+
+       if (fops->f_isfd) {
+               /* fd-based knotes are linked off the fd table */
+               if (kev->ident < (u_int)fdp->fd_knlistsize) {
+                       SLIST_FOREACH(kn, &fdp->fd_knlist[kev->ident], kn_link)
+                               if (kq == kn->kn_kq &&
+                                   kev->filter == kn->kn_filter)
+                                       break;
+               }
+       } else {
+               /* hash non-fd knotes here too */
+               if (fdp->fd_knhashmask != 0) {
+                       struct klist *list;
+
+                       list = &fdp->fd_knhash[
+                           KN_HASH((u_long)kev->ident, fdp->fd_knhashmask)];
+                       SLIST_FOREACH(kn, list, kn_link)
+                               if (kev->ident == kn->kn_id &&
+                                   kq == kn->kn_kq &&
+                                   kev->filter == kn->kn_filter)
+                                       break;
+               }
+       }
+
+       /*
+        * kn now contains the matching knote, or NULL if no match
+        */
+       if (kn == NULL) {
+               if ((kev->flags & (EV_ADD|EV_DELETE)) == EV_ADD) {
+                       kn = knote_alloc();
+                       if (kn == NULL) {
+                               proc_fdunlock(p);
+                               error = ENOMEM;
+                               goto done;
+                       }
+                       kn->kn_fp = fp;
+                       kn->kn_kq = kq;
+                       kn->kn_tq = &kq->kq_head;
+                       kn->kn_fop = fops;
+                       kn->kn_sfflags = kev->fflags;
+                       kn->kn_sdata = kev->data;
+                       kev->fflags = 0;
+                       kev->data = 0;
+                       kn->kn_kevent = *kev;
+                       kn->kn_inuse = 1;  /* for f_attach() */
+                       kn->kn_status = KN_ATTACHING;
+
+                       /* before anyone can find it */
+                       if (kev->flags & EV_DISABLE)
+                               kn->kn_status |= KN_DISABLED;
+
+                       error = knote_fdpattach(kn, fdp, p);
+                       proc_fdunlock(p);
+
+                       if (error) {
+                               knote_free(kn);
+                               goto done;
+                       }
+
+                       /*
+                        * apply reference count to knote structure, and
+                        * do not release it at the end of this routine.
+                        */
+                       fp = NULL;
+
+                       error = fops->f_attach(kn);
+
+                       kqlock(kq);
+
+                       if (error != 0) {
+                               /*
+                                * Failed to attach correctly, so drop.
+                                * All other possible users/droppers
+                                * have deferred to us.
+                                */
+                               kn->kn_status |= KN_DROPPING;
+                               kqunlock(kq);
+                               knote_drop(kn, p);
+                               goto done;
+                       } else if (kn->kn_status & KN_DROPPING) {
+                               /*
+                                * Attach succeeded, but someone else
+                                * deferred their drop - now we have
+                                * to do it for them (after detaching).
+                                */
+                               kqunlock(kq);
+                               kn->kn_fop->f_detach(kn);
+                               knote_drop(kn, p);
+                               goto done;
+                       }
+                       kn->kn_status &= ~KN_ATTACHING;
+                       kqunlock(kq);
+               } else {
+                       proc_fdunlock(p);
+                       error = ENOENT;
+                       goto done;
+               }
+       } else {
+               /* existing knote - get kqueue lock */
+               kqlock(kq);
+               proc_fdunlock(p);
+
+               if (kev->flags & EV_DELETE) {
+                       knote_dequeue(kn);
+                       kn->kn_status |= KN_DISABLED;
+                       if (kqlock2knotedrop(kq, kn)) {
+                               kn->kn_fop->f_detach(kn);
+                               knote_drop(kn, p);
+                       }
+                       goto done;
+               }
+
+               /* update status flags for existing knote */
+               if (kev->flags & EV_DISABLE) {
+                       knote_dequeue(kn);
+                       kn->kn_status |= KN_DISABLED;
+               } else if (kev->flags & EV_ENABLE) {
+                       kn->kn_status &= ~KN_DISABLED;
+                       if (kn->kn_status & KN_ACTIVE)
+                               knote_enqueue(kn);
+               }
+
+               /*
+                * The user may change some filter values after the
+                * initial EV_ADD, but doing so will not reset any
+                * filter which have already been triggered.
+                */
+               kn->kn_kevent.udata = kev->udata;
+               if (fops->f_isfd || fops->f_touch == NULL) {
+                       kn->kn_sfflags = kev->fflags;
+                       kn->kn_sdata = kev->data;
+               }
+
+               /*
+                * If somebody is in the middle of dropping this
+                * knote - go find/insert a new one.  But we have
+                * wait for this one to go away first. Attaches
+                * running in parallel may also drop/modify the
+                * knote.  Wait for those to complete as well and
+                * then start over if we encounter one.
+                */
+               if (!kqlock2knoteusewait(kq, kn)) {
+                       /* kqueue, proc_fdlock both unlocked */
+                       goto restart;
+               }
+
+               /*
+                * Call touch routine to notify filter of changes
+                * in filter values.
+                */
+               if (!fops->f_isfd && fops->f_touch != NULL)
+                       fops->f_touch(kn, kev, EVENT_REGISTER);
+       }
+       /* still have use ref on knote */
+
+       /*
+        * If the knote is not marked to always stay enqueued,
+        * invoke the filter routine to see if it should be
+        * enqueued now.
+        */
+       if ((kn->kn_status & KN_STAYQUEUED) == 0 && kn->kn_fop->f_event(kn, 0)) {
+               if (knoteuse2kqlock(kq, kn))
+                       knote_activate(kn, 1);
+               kqunlock(kq);
+       } else {
+               knote_put(kn);
+       }
+
+done:
+       if (fp != NULL)
+               fp_drop(p, kev->ident, fp, 0);
+       return (error);
+}
+
+
+/*
+ * knote_process - process a triggered event
+ *
+ *     Validate that it is really still a triggered event
+ *     by calling the filter routines (if necessary).  Hold
+ *     a use reference on the knote to avoid it being detached.
+ *     If it is still considered triggered, invoke the callback
+ *     routine provided and move it to the provided inprocess
+ *     queue.
+ *
+ *     caller holds a reference on the kqueue.
+ *     kqueue locked on entry and exit - but may be dropped
+ */
+static int
+knote_process(struct knote *kn,
+    kevent_callback_t callback,
+    void *data,
+    struct kqtailq *inprocessp,
+    struct proc *p)
+{
+       struct kqueue *kq = kn->kn_kq;
+       struct kevent64_s kev;
+       int touch;
+       int result;
+       int error;
+
+       /*
+        * Determine the kevent state we want to return.
+        *
+        * Some event states need to be revalidated before returning
+        * them, others we take the snapshot at the time the event
+        * was enqueued.
+        *
+        * Events with non-NULL f_touch operations must be touched.
+        * Triggered events must fill in kev for the callback.
+        *
+        * Convert our lock to a use-count and call the event's
+        * filter routine(s) to update.
+        */
+       if ((kn->kn_status & KN_DISABLED) != 0) {
+               result = 0;
+               touch = 0;
+       } else {
+               int revalidate;
+
+               result = 1;
+               revalidate = ((kn->kn_status & KN_STAYQUEUED) != 0 ||
+                   (kn->kn_flags & EV_ONESHOT) == 0);
+               touch = (!kn->kn_fop->f_isfd && kn->kn_fop->f_touch != NULL);
+
+               if (revalidate || touch) {
+                       if (revalidate)
+                               knote_deactivate(kn);
+
+                       /* call the filter/touch routines with just a ref */
+                       if (kqlock2knoteuse(kq, kn)) {
+                               /* if we have to revalidate, call the filter */
+                               if (revalidate) {
+                                       result = kn->kn_fop->f_event(kn, 0);
+                               }
+
+                               /*
+                                * capture the kevent data - using touch if
+                                * specified
+                                */
+                               if (result && touch) {
+                                       kn->kn_fop->f_touch(kn, &kev,
+                                           EVENT_PROCESS);
+                               }
+
+                               /*
+                                * convert back to a kqlock - bail if the knote
+                                * went away
+                                */
+                               if (!knoteuse2kqlock(kq, kn)) {
+                                       return (EJUSTRETURN);
+                               } else if (result) {
+                                       /*
+                                        * if revalidated as alive, make sure
+                                        * it's active
+                                        */
+                                       if (!(kn->kn_status & KN_ACTIVE)) {
+                                               knote_activate(kn, 0);
+                                       }
+
+                                       /*
+                                        * capture all events that occurred
+                                        * during filter
+                                        */
+                                       if (!touch) {
+                                               kev = kn->kn_kevent;
+                                       }
+
+                               } else if ((kn->kn_status & KN_STAYQUEUED) == 0) {
+                                       /*
+                                        * was already dequeued, so just bail on
+                                        * this one
+                                        */
+                                       return (EJUSTRETURN);
+                               }
+                       } else {
+                               return (EJUSTRETURN);
+                       }
+               } else {
+                       kev = kn->kn_kevent;
+               }
+       }
+
+       /* move knote onto inprocess queue */
+       assert(kn->kn_tq == &kq->kq_head);
+       TAILQ_REMOVE(&kq->kq_head, kn, kn_tqe);
+       kn->kn_tq = inprocessp;
+       TAILQ_INSERT_TAIL(inprocessp, kn, kn_tqe);
+
+       /*
+        * Determine how to dispatch the knote for future event handling.
+        * not-fired: just return (do not callout).
+        * One-shot: deactivate it.
+        * Clear: deactivate and clear the state.
+        * Dispatch: don't clear state, just deactivate it and mark it disabled.
+        * All others: just leave where they are.
+        */
+
+       if (result == 0) {
+               return (EJUSTRETURN);
+       } else if ((kn->kn_flags & EV_ONESHOT) != 0) {
+               knote_deactivate(kn);
+               if (kqlock2knotedrop(kq, kn)) {
+                       kn->kn_fop->f_detach(kn);
+                       knote_drop(kn, p);
+               }
+       } else if ((kn->kn_flags & (EV_CLEAR | EV_DISPATCH)) != 0) {
+               if ((kn->kn_flags & EV_DISPATCH) != 0) {
+                       /* deactivate and disable all dispatch knotes */
+                       knote_deactivate(kn);
+                       kn->kn_status |= KN_DISABLED;
+               } else if (!touch || kn->kn_fflags == 0) {
+                       /* only deactivate if nothing since the touch */
+                       knote_deactivate(kn);
+               }
+               if (!touch && (kn->kn_flags & EV_CLEAR) != 0) {
+                       /* manually clear non-touch knotes */
+                       kn->kn_data = 0;
+                       kn->kn_fflags = 0;
+               }
+               kqunlock(kq);
+       } else {
+               /*
+                * leave on inprocess queue.  We'll
+                * move all the remaining ones back
+                * the kq queue and wakeup any
+                * waiters when we are done.
+                */
+               kqunlock(kq);
+       }
+
+       /* callback to handle each event as we find it */
+       error = (callback)(kq, &kev, data);
+
+       kqlock(kq);
+       return (error);
+}
+
+/*
+ * Return 0 to indicate that processing should proceed,
+ * -1 if there is nothing to process.
+ *
+ * Called with kqueue locked and returns the same way,
+ * but may drop lock temporarily.
+ */
+static int
+kqueue_begin_processing(struct kqueue *kq)
+{
+       for (;;) {
+               if (kq->kq_count == 0) {
+                       return (-1);
+               }
+
+               /* if someone else is processing the queue, wait */
+               if (kq->kq_nprocess != 0) {
+                       wait_queue_assert_wait((wait_queue_t)kq->kq_wqs,
+                           &kq->kq_nprocess, THREAD_UNINT, 0);
+                       kq->kq_state |= KQ_PROCWAIT;
+                       kqunlock(kq);
+                       thread_block(THREAD_CONTINUE_NULL);
+                       kqlock(kq);
+               } else {
+                       kq->kq_nprocess = 1;
+                       return (0);
+               }
+       }
+}
+
+/*
+ * Called with kqueue lock held.
+ */
+static void
+kqueue_end_processing(struct kqueue *kq)
+{
+       kq->kq_nprocess = 0;
+       if (kq->kq_state & KQ_PROCWAIT) {
+               kq->kq_state &= ~KQ_PROCWAIT;
+               wait_queue_wakeup_all((wait_queue_t)kq->kq_wqs,
+                   &kq->kq_nprocess, THREAD_AWAKENED);
+       }
+}
+
+/*
+ * kqueue_process - process the triggered events in a kqueue
+ *
+ *     Walk the queued knotes and validate that they are
+ *     really still triggered events by calling the filter
+ *     routines (if necessary).  Hold a use reference on
+ *     the knote to avoid it being detached. For each event
+ *     that is still considered triggered, invoke the
+ *     callback routine provided.
+ *
+ *     caller holds a reference on the kqueue.
+ *     kqueue locked on entry and exit - but may be dropped
+ *     kqueue list locked (held for duration of call)
+ */
+
+static int
+kqueue_process(struct kqueue *kq,
+    kevent_callback_t callback,
+    void *data,
+    int *countp,
+    struct proc *p)
+{
+       struct kqtailq inprocess;
+       struct knote *kn;
+       int nevents;
+       int error;
+
+       TAILQ_INIT(&inprocess);
+
+       if (kqueue_begin_processing(kq) == -1) {
+               *countp = 0;
+               /* Nothing to process */
+               return (0);
+       }
+
+       /*
+        * Clear any pre-posted status from previous runs, so we
+        * only detect events that occur during this run.
+        */
+       wait_queue_sub_clearrefs(kq->kq_wqs);
+
+       /*
+        * loop through the enqueued knotes, processing each one and
+        * revalidating those that need it. As they are processed,
+        * they get moved to the inprocess queue (so the loop can end).
+        */
+       error = 0;
+       nevents = 0;
+
+       while (error == 0 &&
+           (kn = TAILQ_FIRST(&kq->kq_head)) != NULL) {
+               error = knote_process(kn, callback, data, &inprocess, p);
+               if (error == EJUSTRETURN)
+                       error = 0;
+               else
+                       nevents++;
+       }
+
+       /*
+        * With the kqueue still locked, move any knotes
+        * remaining on the inprocess queue back to the
+        * kq's queue and wake up any waiters.
+        */
+       while ((kn = TAILQ_FIRST(&inprocess)) != NULL) {
+               assert(kn->kn_tq == &inprocess);
+               TAILQ_REMOVE(&inprocess, kn, kn_tqe);
+               kn->kn_tq = &kq->kq_head;
+               TAILQ_INSERT_TAIL(&kq->kq_head, kn, kn_tqe);
+       }
+
+       kqueue_end_processing(kq);
+
+       *countp = nevents;
+       return (error);
+}
+
+
+static void
+kqueue_scan_continue(void *data, wait_result_t wait_result)
+{
+       thread_t self = current_thread();
+       uthread_t ut = (uthread_t)get_bsdthread_info(self);
+       struct _kqueue_scan * cont_args = &ut->uu_kevent.ss_kqueue_scan;
+       struct kqueue *kq = (struct kqueue *)data;
+       int error;
+       int count;
+
+       /* convert the (previous) wait_result to a proper error */
+       switch (wait_result) {
+       case THREAD_AWAKENED:
+               kqlock(kq);
+               error = kqueue_process(kq, cont_args->call, cont_args, &count,
+                   current_proc());
+               if (error == 0 && count == 0) {
+                       wait_queue_assert_wait((wait_queue_t)kq->kq_wqs,
+                           KQ_EVENT, THREAD_ABORTSAFE, cont_args->deadline);
+                       kq->kq_state |= KQ_SLEEP;
+                       kqunlock(kq);
+                       thread_block_parameter(kqueue_scan_continue, kq);
+                       /* NOTREACHED */
+               }
+               kqunlock(kq);
+               break;
+       case THREAD_TIMED_OUT:
+               error = EWOULDBLOCK;
+               break;
+       case THREAD_INTERRUPTED:
+               error = EINTR;
+               break;
+       default:
+               panic("%s: - invalid wait_result (%d)", __func__,
+                   wait_result);
+               error = 0;
+       }
+
+       /* call the continuation with the results */
+       assert(cont_args->cont != NULL);
+       (cont_args->cont)(kq, cont_args->data, error);
+}
+
+
+/*
+ * kqueue_scan - scan and wait for events in a kqueue
+ *
+ *     Process the triggered events in a kqueue.
+ *
+ *     If there are no events triggered arrange to
+ *     wait for them. If the caller provided a
+ *     continuation routine, then kevent_scan will
+ *     also.
+ *
+ *     The callback routine must be valid.
+ *     The caller must hold a use-count reference on the kq.
+ */
+
+int
+kqueue_scan(struct kqueue *kq,
+           kevent_callback_t callback,
+           kqueue_continue_t continuation,
+           void *data,
+           struct timeval *atvp,
+           struct proc *p)
+{
+       thread_continue_t cont = THREAD_CONTINUE_NULL;
+       uint64_t deadline;
+       int error;
+       int first;
+
+       assert(callback != NULL);
+
+       first = 1;
+       for (;;) {
+               wait_result_t wait_result;
+               int count;
+
+               /*
+                * Make a pass through the kq to find events already
+                * triggered.
+                */
+               kqlock(kq);
+               error = kqueue_process(kq, callback, data, &count, p);
+               if (error || count)
+                       break; /* lock still held */
+
+               /* looks like we have to consider blocking */
+               if (first) {
+                       first = 0;
+                       /* convert the timeout to a deadline once */
+                       if (atvp->tv_sec || atvp->tv_usec) {
+                               uint64_t now;
+
+                               clock_get_uptime(&now);
+                               nanoseconds_to_absolutetime((uint64_t)atvp->tv_sec * NSEC_PER_SEC +
+                                                           atvp->tv_usec * (long)NSEC_PER_USEC,
+                                                           &deadline);
+                               if (now >= deadline) {
+                                       /* non-blocking call */
+                                       error = EWOULDBLOCK;
+                                       break; /* lock still held */
+                               }
+                               deadline -= now;
+                               clock_absolutetime_interval_to_deadline(deadline, &deadline);
+                       } else {
+                               deadline = 0;   /* block forever */
+                       }
+
+                       if (continuation) {
+                               uthread_t ut = (uthread_t)get_bsdthread_info(current_thread());
+                               struct _kqueue_scan *cont_args = &ut->uu_kevent.ss_kqueue_scan;
+
+                               cont_args->call = callback;
+                               cont_args->cont = continuation;
+                               cont_args->deadline = deadline;
+                               cont_args->data = data;
+                               cont = kqueue_scan_continue;
+                       }
+               }
+
+               /* go ahead and wait */
+               wait_queue_assert_wait_with_leeway((wait_queue_t)kq->kq_wqs,
+                   KQ_EVENT, THREAD_ABORTSAFE, TIMEOUT_URGENCY_USER_NORMAL,
+                   deadline, 0);
+               kq->kq_state |= KQ_SLEEP;
+               kqunlock(kq);
+               wait_result = thread_block_parameter(cont, kq);
+               /* NOTREACHED if (continuation != NULL) */
+
+               switch (wait_result) {
+               case THREAD_AWAKENED:
+                       continue;
+               case THREAD_TIMED_OUT:
+                       return (EWOULDBLOCK);
+               case THREAD_INTERRUPTED:
+                       return (EINTR);
+               default:
+                       panic("%s: - bad wait_result (%d)", __func__,
+                           wait_result);
+                       error = 0;
+               }
+       }
+       kqunlock(kq);
+       return (error);
+}
+
+
+/*
+ * XXX
+ * This could be expanded to call kqueue_scan, if desired.
+ */
+/*ARGSUSED*/
+static int
+kqueue_read(__unused struct fileproc *fp,
+    __unused struct uio *uio,
+    __unused int flags,
+    __unused vfs_context_t ctx)
+{
+       return (ENXIO);
+}
+
+/*ARGSUSED*/
+static int
+kqueue_write(__unused struct fileproc *fp,
+    __unused struct uio *uio,
+    __unused int flags,
+    __unused vfs_context_t ctx)
+{
+       return (ENXIO);
+}
+
+/*ARGSUSED*/
+static int
+kqueue_ioctl(__unused struct fileproc *fp,
+    __unused u_long com,
+    __unused caddr_t data,
+    __unused vfs_context_t ctx)
+{
+       return (ENOTTY);
+}
+
+/*ARGSUSED*/
+static int
+kqueue_select(struct fileproc *fp, int which, void *wql,
+    __unused vfs_context_t ctx)
+{
+       struct kqueue *kq = (struct kqueue *)fp->f_data;
+       struct knote *kn;
+       struct kqtailq inprocessq;
+       int retnum = 0;
+
+       if (which != FREAD)
+               return (0);
+
+       TAILQ_INIT(&inprocessq);
+
+       kqlock(kq);
+       /*
+        * If this is the first pass, link the wait queue associated with the
+        * the kqueue onto the wait queue set for the select().  Normally we
+        * use selrecord() for this, but it uses the wait queue within the
+        * selinfo structure and we need to use the main one for the kqueue to
+        * catch events from KN_STAYQUEUED sources. So we do the linkage manually.
+        * (The select() call will unlink them when it ends).
+        */
+       if (wql != NULL) {
+               thread_t cur_act = current_thread();
+               struct uthread * ut = get_bsdthread_info(cur_act);
+
+               kq->kq_state |= KQ_SEL;
+               wait_queue_link_noalloc((wait_queue_t)kq->kq_wqs, ut->uu_wqset,
+                   (wait_queue_link_t)wql);
+       }
+
+       if (kqueue_begin_processing(kq) == -1) {
+               kqunlock(kq);
+               return (0);
+       }
+
+       if (kq->kq_count != 0) {
+               /*
+                * there is something queued - but it might be a
+                * KN_STAYQUEUED knote, which may or may not have
+                * any events pending.  So, we have to walk the
+                * list of knotes to see, and peek at the stay-
+                * queued ones to be really sure.
+                */
+               while ((kn = (struct knote *)TAILQ_FIRST(&kq->kq_head)) != NULL) {
+                       if ((kn->kn_status & KN_STAYQUEUED) == 0) {
+                               retnum = 1;
+                               goto out;
+                       }
+
+                       TAILQ_REMOVE(&kq->kq_head, kn, kn_tqe);
+                       TAILQ_INSERT_TAIL(&inprocessq, kn, kn_tqe);
+
+                       if (kqlock2knoteuse(kq, kn)) {
+                               unsigned peek;
+
+                               peek = kn->kn_fop->f_peek(kn);
+                               if (knoteuse2kqlock(kq, kn)) {
+                                       if (peek > 0) {
+                                               retnum = 1;
+                                               goto out;
+                                       }
+                               } else {
+                                       retnum = 0;
+                               }
+                       }
+               }
+       }
+
+out:
+       /* Return knotes to active queue */
+       while ((kn = TAILQ_FIRST(&inprocessq)) != NULL) {
+               TAILQ_REMOVE(&inprocessq, kn, kn_tqe);
+               kn->kn_tq = &kq->kq_head;
+               TAILQ_INSERT_TAIL(&kq->kq_head, kn, kn_tqe);
+       }
+
+       kqueue_end_processing(kq);
+       kqunlock(kq);
+       return (retnum);
+}
+
+/*
+ * kqueue_close -
+ */
+/*ARGSUSED*/
+static int
+kqueue_close(struct fileglob *fg, __unused vfs_context_t ctx)
+{
+       struct kqueue *kq = (struct kqueue *)fg->fg_data;
+
+       kqueue_dealloc(kq);
+       fg->fg_data = NULL;
+       return (0);
+}
+
+/*ARGSUSED*/
+/*
+ * The callers has taken a use-count reference on this kqueue and will donate it
+ * to the kqueue we are being added to.  This keeps the kqueue from closing until
+ * that relationship is torn down.
+ */
+static int
+kqueue_kqfilter(__unused struct fileproc *fp, struct knote *kn, __unused vfs_context_t ctx)
+{
+       struct kqueue *kq = (struct kqueue *)kn->kn_fp->f_data;
+       struct kqueue *parentkq = kn->kn_kq;
+
+       if (parentkq == kq ||
+           kn->kn_filter != EVFILT_READ)
+               return (1);
+
+       /*
+        * We have to avoid creating a cycle when nesting kqueues
+        * inside another.  Rather than trying to walk the whole
+        * potential DAG of nested kqueues, we just use a simple
+        * ceiling protocol.  When a kqueue is inserted into another,
+        * we check that the (future) parent is not already nested
+        * into another kqueue at a lower level than the potenial
+        * child (because it could indicate a cycle).  If that test
+        * passes, we just mark the nesting levels accordingly.
+        */
+
+       kqlock(parentkq);
+       if (parentkq->kq_level > 0 &&
+           parentkq->kq_level < kq->kq_level)
+       {
+               kqunlock(parentkq);
+               return (1);
+       } else {
+               /* set parent level appropriately */
+               if (parentkq->kq_level == 0)
+                       parentkq->kq_level = 2;
+               if (parentkq->kq_level < kq->kq_level + 1)
+                       parentkq->kq_level = kq->kq_level + 1;
+               kqunlock(parentkq);
+
+               kn->kn_fop = &kqread_filtops;
+               kqlock(kq);
+               KNOTE_ATTACH(&kq->kq_sel.si_note, kn);
+               /* indicate nesting in child, if needed */
+               if (kq->kq_level == 0)
+                       kq->kq_level = 1;
+               kqunlock(kq);
+               return (0);
+       }
+}
+
+/*
+ * kqueue_drain - called when kq is closed
+ */
+/*ARGSUSED*/
+static int
+kqueue_drain(struct fileproc *fp, __unused vfs_context_t ctx)
+{
+       struct kqueue *kq = (struct kqueue *)fp->f_fglob->fg_data;
+       kqlock(kq);
+       kqueue_wakeup(kq, 1);
+       kqunlock(kq);
+       return (0);
+}
+
+/*ARGSUSED*/
+int
+kqueue_stat(struct kqueue *kq, void *ub, int isstat64, proc_t p)
+{
+       kqlock(kq);
+       if (isstat64 != 0) {
+               struct stat64 *sb64 = (struct stat64 *)ub;
+
+               bzero((void *)sb64, sizeof(*sb64));
+               sb64->st_size = kq->kq_count;
+               if (kq->kq_state & KQ_KEV64)
+                       sb64->st_blksize = sizeof(struct kevent64_s);
+               else
+                       sb64->st_blksize = IS_64BIT_PROCESS(p) ? sizeof(struct user64_kevent) : sizeof(struct user32_kevent);
+               sb64->st_mode = S_IFIFO;
+       } else {
+               struct stat *sb = (struct stat *)ub;
+
+               bzero((void *)sb, sizeof(*sb));
+               sb->st_size = kq->kq_count;
+               if (kq->kq_state & KQ_KEV64)
+                       sb->st_blksize = sizeof(struct kevent64_s);
+               else
+                       sb->st_blksize = IS_64BIT_PROCESS(p) ? sizeof(struct user64_kevent) : sizeof(struct user32_kevent);
+               sb->st_mode = S_IFIFO;
+       }
+       kqunlock(kq);
+       return (0);
+}
+
+/*
+ * Called with the kqueue locked
+ */
+static void
+kqueue_wakeup(struct kqueue *kq, int closed)
+{
+       if ((kq->kq_state & (KQ_SLEEP | KQ_SEL)) != 0 || kq->kq_nprocess > 0) {
+               kq->kq_state &= ~(KQ_SLEEP | KQ_SEL);
+               wait_queue_wakeup_all((wait_queue_t)kq->kq_wqs, KQ_EVENT,
+                   (closed) ? THREAD_INTERRUPTED : THREAD_AWAKENED);
+       }
+}
+
+void
+klist_init(struct klist *list)
+{
+       SLIST_INIT(list);
+}
+
+
+/*
+ * Query/Post each knote in the object's list
+ *
+ *     The object lock protects the list. It is assumed
+ *     that the filter/event routine for the object can
+ *     determine that the object is already locked (via
+ *     the hint) and not deadlock itself.
+ *
+ *     The object lock should also hold off pending
+ *     detach/drop operations.  But we'll prevent it here
+ *     too - just in case.
+ */
+void
+knote(struct klist *list, long hint)
+{
+       struct knote *kn;
+
+       SLIST_FOREACH(kn, list, kn_selnext) {
+               struct kqueue *kq = kn->kn_kq;
+
+               kqlock(kq);
+               if (kqlock2knoteuse(kq, kn)) {
+                       int result;
+
+                       /* call the event with only a use count */
+                       result = kn->kn_fop->f_event(kn, hint);
+
+                       /* if its not going away and triggered */
+                       if (knoteuse2kqlock(kq, kn) && result)
+                               knote_activate(kn, 1);
+                       /* lock held again */
+               }
+               kqunlock(kq);
+       }
+}
+
+/*
+ * attach a knote to the specified list.  Return true if this is the first entry.
+ * The list is protected by whatever lock the object it is associated with uses.
+ */
+int
+knote_attach(struct klist *list, struct knote *kn)
+{
+       int ret = SLIST_EMPTY(list);
+       SLIST_INSERT_HEAD(list, kn, kn_selnext);
+       return (ret);
+}
+
+/*
+ * detach a knote from the specified list.  Return true if that was the last entry.
+ * The list is protected by whatever lock the object it is associated with uses.
+ */
+int
+knote_detach(struct klist *list, struct knote *kn)
+{
+       SLIST_REMOVE(list, kn, knote, kn_selnext);
+       return (SLIST_EMPTY(list));
+}
+
+/*
+ * For a given knote, link a provided wait queue directly with the kqueue.
+ * Wakeups will happen via recursive wait queue support.  But nothing will move
+ * the knote to the active list at wakeup (nothing calls knote()).  Instead,
+ * we permanently enqueue them here.
+ *
+ * kqueue and knote references are held by caller.
+ *
+ * caller provides the wait queue link structure.
+ */
+int
+knote_link_wait_queue(struct knote *kn, struct wait_queue *wq, wait_queue_link_t wql)
+{
+       struct kqueue *kq = kn->kn_kq;
+       kern_return_t kr;
+
+       kr = wait_queue_link_noalloc(wq, kq->kq_wqs, wql);
+       if (kr == KERN_SUCCESS) {
+               knote_markstayqueued(kn);
+               return (0);
+       } else {
+               return (EINVAL);
+       }
+}
+
+/*
+ * Unlink the provided wait queue from the kqueue associated with a knote.
+ * Also remove it from the magic list of directly attached knotes.
+ *
+ * Note that the unlink may have already happened from the other side, so
+ * ignore any failures to unlink and just remove it from the kqueue list.
+ *
+ * On success, caller is responsible for the link structure
+ */
+int
+knote_unlink_wait_queue(struct knote *kn, struct wait_queue *wq, wait_queue_link_t *wqlp)
+{
+       struct kqueue *kq = kn->kn_kq;
+       kern_return_t kr;
+
+       kr = wait_queue_unlink_nofree(wq, kq->kq_wqs, wqlp);
+       knote_clearstayqueued(kn);
+       return ((kr != KERN_SUCCESS) ? EINVAL : 0);
+}
+
+/*
+ * remove all knotes referencing a specified fd
+ *
+ * Essentially an inlined knote_remove & knote_drop
+ * when we know for sure that the thing is a file
+ *
+ * Entered with the proc_fd lock already held.
+ * It returns the same way, but may drop it temporarily.
+ */
+void
+knote_fdclose(struct proc *p, int fd)
+{
+       struct filedesc *fdp = p->p_fd;
+       struct klist *list;
+       struct knote *kn;
+
+       list = &fdp->fd_knlist[fd];
+       while ((kn = SLIST_FIRST(list)) != NULL) {
+               struct kqueue *kq = kn->kn_kq;
+
+               if (kq->kq_p != p)
+                       panic("%s: proc mismatch (kq->kq_p=%p != p=%p)",
+                           __func__, kq->kq_p, p);
+
+               kqlock(kq);
+               proc_fdunlock(p);
+
+               /*
+                * Convert the lock to a drop ref.
+                * If we get it, go ahead and drop it.
+                * Otherwise, we waited for it to
+                * be dropped by the other guy, so
+                * it is safe to move on in the list.
+                */
+               if (kqlock2knotedrop(kq, kn)) {
+                       kn->kn_fop->f_detach(kn);
+                       knote_drop(kn, p);
+               }
+
+               proc_fdlock(p);
+
+               /* the fd tables may have changed - start over */
+               list = &fdp->fd_knlist[fd];
+       }
+}
+
+/* proc_fdlock held on entry (and exit) */
+static int
+knote_fdpattach(struct knote *kn, struct filedesc *fdp, struct proc *p)
+{
+       struct klist *list = NULL;
+
+       if (! kn->kn_fop->f_isfd) {
+               if (fdp->fd_knhashmask == 0)
+                       fdp->fd_knhash = hashinit(CONFIG_KN_HASHSIZE, M_KQUEUE,
+                           &fdp->fd_knhashmask);
+               list = &fdp->fd_knhash[KN_HASH(kn->kn_id, fdp->fd_knhashmask)];
+       } else {
+               if ((u_int)fdp->fd_knlistsize <= kn->kn_id) {
+                       u_int size = 0;
+
+                       if (kn->kn_id >= (uint64_t)p->p_rlimit[RLIMIT_NOFILE].rlim_cur
+                           || kn->kn_id >= (uint64_t)maxfiles)
+                               return (EINVAL);
+
+                       /* have to grow the fd_knlist */
+                       size = fdp->fd_knlistsize;
+                       while (size <= kn->kn_id)
+                               size += KQEXTENT;
+
+                       if (size >= (UINT_MAX/sizeof(struct klist *)))
+                               return (EINVAL);
+
+                       MALLOC(list, struct klist *,
+                           size * sizeof(struct klist *), M_KQUEUE, M_WAITOK);
+                       if (list == NULL)
+                               return (ENOMEM);
+
+                       bcopy((caddr_t)fdp->fd_knlist, (caddr_t)list,
+                           fdp->fd_knlistsize * sizeof(struct klist *));
+                       bzero((caddr_t)list +
+                           fdp->fd_knlistsize * sizeof(struct klist *),
+                           (size - fdp->fd_knlistsize) * sizeof(struct klist *));
+                       FREE(fdp->fd_knlist, M_KQUEUE);
+                       fdp->fd_knlist = list;
+                       fdp->fd_knlistsize = size;
+               }
+               list = &fdp->fd_knlist[kn->kn_id];
+       }
+       SLIST_INSERT_HEAD(list, kn, kn_link);
+       return (0);
+}
+
+
+
+/*
+ * should be called at spl == 0, since we don't want to hold spl
+ * while calling fdrop and free.
+ */
+static void
+knote_drop(struct knote *kn, __unused struct proc *ctxp)
+{
+       struct kqueue *kq = kn->kn_kq;
+       struct proc *p = kq->kq_p;
+       struct filedesc *fdp = p->p_fd;
+       struct klist *list;
+       int needswakeup;
+
+       proc_fdlock(p);
+       if (kn->kn_fop->f_isfd)
+               list = &fdp->fd_knlist[kn->kn_id];
+       else
+               list = &fdp->fd_knhash[KN_HASH(kn->kn_id, fdp->fd_knhashmask)];
+
+       SLIST_REMOVE(list, kn, knote, kn_link);
+       kqlock(kq);
+       knote_dequeue(kn);
+       needswakeup = (kn->kn_status & KN_USEWAIT);
+       kqunlock(kq);
+       proc_fdunlock(p);
+
+       if (needswakeup)
+               wait_queue_wakeup_all((wait_queue_t)kq->kq_wqs, &kn->kn_status,
+                   THREAD_AWAKENED);
+
+       if (kn->kn_fop->f_isfd)
+               fp_drop(p, kn->kn_id, kn->kn_fp, 0);
+
+       knote_free(kn);
+}
+
+/* called with kqueue lock held */
+static void
+knote_activate(struct knote *kn, int propagate)
+{
+       struct kqueue *kq = kn->kn_kq;
+
+       kn->kn_status |= KN_ACTIVE;
+       knote_enqueue(kn);
+       kqueue_wakeup(kq, 0);
+
+       /* this is a real event: wake up the parent kq, too */
+       if (propagate)
+               KNOTE(&kq->kq_sel.si_note, 0);
+}
+
+/* called with kqueue lock held */
+static void
+knote_deactivate(struct knote *kn)
+{
+       kn->kn_status &= ~KN_ACTIVE;
+       knote_dequeue(kn);
+}
+
+/* called with kqueue lock held */
+static void
+knote_enqueue(struct knote *kn)
+{
+       if ((kn->kn_status & (KN_QUEUED | KN_STAYQUEUED)) == KN_STAYQUEUED ||
+           (kn->kn_status & (KN_QUEUED | KN_STAYQUEUED | KN_DISABLED)) == 0) {
+               struct kqtailq *tq = kn->kn_tq;
+               struct kqueue *kq = kn->kn_kq;
+
+               TAILQ_INSERT_TAIL(tq, kn, kn_tqe);
+               kn->kn_status |= KN_QUEUED;
+               kq->kq_count++;
+       }
+}
+
+/* called with kqueue lock held */
+static void
+knote_dequeue(struct knote *kn)
+{
+       struct kqueue *kq = kn->kn_kq;
+
+       if ((kn->kn_status & (KN_QUEUED | KN_STAYQUEUED)) == KN_QUEUED) {
+               struct kqtailq *tq = kn->kn_tq;
+
+               TAILQ_REMOVE(tq, kn, kn_tqe);
+               kn->kn_tq = &kq->kq_head;
+               kn->kn_status &= ~KN_QUEUED;
+               kq->kq_count--;
+       }
+}
+
+void
+knote_init(void)
+{
+       knote_zone = zinit(sizeof(struct knote), 8192*sizeof(struct knote),
+           8192, "knote zone");
+
+       /* allocate kq lock group attribute and group */
+       kq_lck_grp_attr = lck_grp_attr_alloc_init();
+
+       kq_lck_grp = lck_grp_alloc_init("kqueue",  kq_lck_grp_attr);
+
+       /* Allocate kq lock attribute */
+       kq_lck_attr = lck_attr_alloc_init();
+
+       /* Initialize the timer filter lock */
+       lck_mtx_init(&_filt_timerlock, kq_lck_grp, kq_lck_attr);
+
+#if VM_PRESSURE_EVENTS
+       /* Initialize the vm pressure list lock */
+       vm_pressure_init(kq_lck_grp, kq_lck_attr);
+#endif
+
+#if CONFIG_MEMORYSTATUS
+       /* Initialize the memorystatus list lock */
+       memorystatus_kevent_init(kq_lck_grp, kq_lck_attr);
+#endif
+}
+SYSINIT(knote, SI_SUB_PSEUDO, SI_ORDER_ANY, knote_init, NULL)
+
+static struct knote *
+knote_alloc(void)
+{
+       return ((struct knote *)zalloc(knote_zone));
+}
+
+static void
+knote_free(struct knote *kn)
+{
+       zfree(knote_zone, kn);
+}
+
+#if SOCKETS
+#include <sys/param.h>
+#include <sys/socket.h>
+#include <sys/protosw.h>
+#include <sys/domain.h>
+#include <sys/mbuf.h>
+#include <sys/kern_event.h>
+#include <sys/malloc.h>
+#include <sys/sys_domain.h>
+#include <sys/syslog.h>
+
+#ifndef ROUNDUP64
+#define        ROUNDUP64(x) P2ROUNDUP((x), sizeof (u_int64_t))
+#endif
+
+#ifndef ADVANCE64
+#define        ADVANCE64(p, n) (void*)((char *)(p) + ROUNDUP64(n))
+#endif
+
+static lck_grp_attr_t *kev_lck_grp_attr;
+static lck_attr_t *kev_lck_attr;
+static lck_grp_t *kev_lck_grp;
+static decl_lck_rw_data(,kev_lck_data);
+static lck_rw_t *kev_rwlock = &kev_lck_data;
+
+static int kev_attach(struct socket *so, int proto, struct proc *p);
+static int kev_detach(struct socket *so);
+static int kev_control(struct socket *so, u_long cmd, caddr_t data,
+    struct ifnet *ifp, struct proc *p);
+static lck_mtx_t * event_getlock(struct socket *, int);
+static int event_lock(struct socket *, int, void *);
+static int event_unlock(struct socket *, int, void *);
+
+static int event_sofreelastref(struct socket *);
+static void kev_delete(struct kern_event_pcb *);
+
+static struct pr_usrreqs event_usrreqs = {
+       .pru_attach =           kev_attach,
+       .pru_control =          kev_control,
+       .pru_detach =           kev_detach,
+       .pru_soreceive =        soreceive,
+};
+
+static struct protosw eventsw[] = {
+{
+       .pr_type =              SOCK_RAW,
+       .pr_protocol =          SYSPROTO_EVENT,
+       .pr_flags =             PR_ATOMIC,
+       .pr_usrreqs =           &event_usrreqs,
+       .pr_lock =              event_lock,
+       .pr_unlock =            event_unlock,
+       .pr_getlock =           event_getlock,
+}
+};
+
+__private_extern__ int kevt_getstat SYSCTL_HANDLER_ARGS;
+__private_extern__ int kevt_pcblist SYSCTL_HANDLER_ARGS;
+
+SYSCTL_NODE(_net_systm, OID_AUTO, kevt,
+       CTLFLAG_RW|CTLFLAG_LOCKED, 0, "Kernel event family");
+
+struct kevtstat kevtstat;
+SYSCTL_PROC(_net_systm_kevt, OID_AUTO, stats,
+    CTLTYPE_STRUCT | CTLFLAG_RD | CTLFLAG_LOCKED, 0, 0,
+    kevt_getstat, "S,kevtstat", "");
+
+SYSCTL_PROC(_net_systm_kevt, OID_AUTO, pcblist,
+       CTLTYPE_STRUCT | CTLFLAG_RD | CTLFLAG_LOCKED, 0, 0,
+       kevt_pcblist, "S,xkevtpcb", "");
+
+static lck_mtx_t *
+event_getlock(struct socket *so, int locktype)
+{
+#pragma unused(locktype)
+       struct kern_event_pcb *ev_pcb = (struct kern_event_pcb *)so->so_pcb;
+
+       if (so->so_pcb != NULL)  {
+               if (so->so_usecount < 0)
+                       panic("%s: so=%p usecount=%d lrh= %s\n", __func__,
+                           so, so->so_usecount, solockhistory_nr(so));
+                       /* NOTREACHED */
+       } else {
+               panic("%s: so=%p NULL NO so_pcb %s\n", __func__,
+                   so, solockhistory_nr(so));
+               /* NOTREACHED */
+       }
+       return (&ev_pcb->evp_mtx);
+}
+
+static int
+event_lock(struct socket *so, int refcount, void *lr)
+{
+       void *lr_saved;
+
+       if (lr == NULL)
+               lr_saved = __builtin_return_address(0);
+       else
+               lr_saved = lr;
+
+       if (so->so_pcb != NULL) {
+               lck_mtx_lock(&((struct kern_event_pcb *)so->so_pcb)->evp_mtx);
+       } else  {
+               panic("%s: so=%p NO PCB! lr=%p lrh= %s\n", __func__,
+                   so, lr_saved, solockhistory_nr(so));
+               /* NOTREACHED */
+       }
+
+       if (so->so_usecount < 0) {
+               panic("%s: so=%p so_pcb=%p lr=%p ref=%d lrh= %s\n", __func__,
+                   so, so->so_pcb, lr_saved, so->so_usecount,
+                   solockhistory_nr(so));
+               /* NOTREACHED */
+       }
+
+       if (refcount)
+               so->so_usecount++;
+
+       so->lock_lr[so->next_lock_lr] = lr_saved;
+       so->next_lock_lr = (so->next_lock_lr+1) % SO_LCKDBG_MAX;
+       return (0);
+}
+
+static int
+event_unlock(struct socket *so, int refcount, void *lr)
+{
+       void *lr_saved;
+       lck_mtx_t *mutex_held;
+
+       if (lr == NULL)
+               lr_saved = __builtin_return_address(0);
+       else
+               lr_saved = lr;
+
+       if (refcount)
+               so->so_usecount--;
+
+       if (so->so_usecount < 0) {
+               panic("%s: so=%p usecount=%d lrh= %s\n", __func__,
+                   so, so->so_usecount, solockhistory_nr(so));
+               /* NOTREACHED */
+       }
+       if (so->so_pcb == NULL) {
+               panic("%s: so=%p NO PCB usecount=%d lr=%p lrh= %s\n", __func__,
+                   so, so->so_usecount, (void *)lr_saved,
+                   solockhistory_nr(so));
+               /* NOTREACHED */
+       }
+       mutex_held = (&((struct kern_event_pcb *)so->so_pcb)->evp_mtx);
+
+       lck_mtx_assert(mutex_held, LCK_MTX_ASSERT_OWNED);
+       so->unlock_lr[so->next_unlock_lr] = lr_saved;
+       so->next_unlock_lr = (so->next_unlock_lr+1) % SO_LCKDBG_MAX;
+
+       if (so->so_usecount == 0) {
+               VERIFY(so->so_flags & SOF_PCBCLEARING);
+               event_sofreelastref(so);
+       } else {
+               lck_mtx_unlock(mutex_held);
+       }
+
+       return (0);
+}
+
+static int
+event_sofreelastref(struct socket *so)
+{
+       struct kern_event_pcb *ev_pcb = (struct kern_event_pcb *)so->so_pcb;
+
+       lck_mtx_assert(&(ev_pcb->evp_mtx), LCK_MTX_ASSERT_OWNED);
+
+       so->so_pcb = NULL;
+
+       /*
+        * Disable upcall in the event another thread is in kev_post_msg()
+        * appending record to the receive socket buffer, since sbwakeup()
+        * may release the socket lock otherwise.
+        */
+       so->so_rcv.sb_flags &= ~SB_UPCALL;
+       so->so_snd.sb_flags &= ~SB_UPCALL;
+       so->so_event = sonullevent;
+       lck_mtx_unlock(&(ev_pcb->evp_mtx));
+
+       lck_mtx_assert(&(ev_pcb->evp_mtx), LCK_MTX_ASSERT_NOTOWNED);
+       lck_rw_lock_exclusive(kev_rwlock);
+       LIST_REMOVE(ev_pcb, evp_link);
+       kevtstat.kes_pcbcount--;
+       kevtstat.kes_gencnt++;
+       lck_rw_done(kev_rwlock);
+       kev_delete(ev_pcb);
+
+       sofreelastref(so, 1);
+       return (0);
+}
+
+static int event_proto_count = (sizeof (eventsw) / sizeof (struct protosw));
+
+static
+struct kern_event_head kern_event_head;
+
+static u_int32_t static_event_id = 0;
+
+#define        EVPCB_ZONE_MAX          65536
+#define        EVPCB_ZONE_NAME         "kerneventpcb"
+static struct zone *ev_pcb_zone;
+
+/*
+ * Install the protosw's for the NKE manager.  Invoked at extension load time
+ */
+void
+kern_event_init(struct domain *dp)
+{
+       struct protosw *pr;
+       int i;
+
+       VERIFY(!(dp->dom_flags & DOM_INITIALIZED));
+       VERIFY(dp == systemdomain);
+
+       kev_lck_grp_attr = lck_grp_attr_alloc_init();
+       if (kev_lck_grp_attr == NULL) {
+               panic("%s: lck_grp_attr_alloc_init failed\n", __func__);
+               /* NOTREACHED */
+       }
+
+       kev_lck_grp = lck_grp_alloc_init("Kernel Event Protocol",
+           kev_lck_grp_attr);
+       if (kev_lck_grp == NULL) {
+               panic("%s: lck_grp_alloc_init failed\n", __func__);
+               /* NOTREACHED */
+       }
+
+       kev_lck_attr = lck_attr_alloc_init();
+       if (kev_lck_attr == NULL) {
+               panic("%s: lck_attr_alloc_init failed\n", __func__);
+               /* NOTREACHED */
+       }
+
+       lck_rw_init(kev_rwlock, kev_lck_grp, kev_lck_attr);
+       if (kev_rwlock == NULL) {
+               panic("%s: lck_mtx_alloc_init failed\n", __func__);
+               /* NOTREACHED */
+       }
+
+       for (i = 0, pr = &eventsw[0]; i < event_proto_count; i++, pr++)
+               net_add_proto(pr, dp, 1);
+
+       ev_pcb_zone = zinit(sizeof(struct kern_event_pcb),
+           EVPCB_ZONE_MAX * sizeof(struct kern_event_pcb), 0, EVPCB_ZONE_NAME);
+       if (ev_pcb_zone == NULL) {
+               panic("%s: failed allocating ev_pcb_zone", __func__);
+               /* NOTREACHED */
+       }
+       zone_change(ev_pcb_zone, Z_EXPAND, TRUE);
+       zone_change(ev_pcb_zone, Z_CALLERACCT, TRUE);
+}
+
+static int
+kev_attach(struct socket *so, __unused int proto, __unused struct proc *p)
+{
+       int error = 0;
+       struct kern_event_pcb *ev_pcb;
+
+       error = soreserve(so, KEV_SNDSPACE, KEV_RECVSPACE);
+       if (error != 0)
+               return (error);
+
+       if ((ev_pcb = (struct kern_event_pcb *)zalloc(ev_pcb_zone)) == NULL) {
+               return (ENOBUFS);
+       }
+       bzero(ev_pcb, sizeof(struct kern_event_pcb));
+       lck_mtx_init(&ev_pcb->evp_mtx, kev_lck_grp, kev_lck_attr);
+
+       ev_pcb->evp_socket = so;
+       ev_pcb->evp_vendor_code_filter = 0xffffffff;
+
+       so->so_pcb = (caddr_t) ev_pcb;
+       lck_rw_lock_exclusive(kev_rwlock);
+       LIST_INSERT_HEAD(&kern_event_head, ev_pcb, evp_link);
+       kevtstat.kes_pcbcount++;
+       kevtstat.kes_gencnt++;
+       lck_rw_done(kev_rwlock);
+
+       return (error);
+}
+
+static void
+kev_delete(struct kern_event_pcb *ev_pcb)
+{
+       VERIFY(ev_pcb != NULL);
+       lck_mtx_destroy(&ev_pcb->evp_mtx, kev_lck_grp);
+       zfree(ev_pcb_zone, ev_pcb);
+}
+
+static int
+kev_detach(struct socket *so)
+{
+       struct kern_event_pcb *ev_pcb = (struct kern_event_pcb *) so->so_pcb;
+
+       if (ev_pcb != NULL) {
+               soisdisconnected(so);
+               so->so_flags |= SOF_PCBCLEARING;
+       }
+
+       return (0);
+}
+
+/*
+ * For now, kev_vendor_code and mbuf_tags use the same
+ * mechanism.
+ */
+errno_t kev_vendor_code_find(
+       const char      *string,
+       u_int32_t       *out_vendor_code)
+{
+       if (strlen(string) >= KEV_VENDOR_CODE_MAX_STR_LEN) {
+               return (EINVAL);
+       }
+       return (net_str_id_find_internal(string, out_vendor_code,
+           NSI_VENDOR_CODE, 1));
+}
+
+errno_t
+kev_msg_post(struct kev_msg *event_msg)
+{
+       mbuf_tag_id_t min_vendor, max_vendor;
+
+       net_str_id_first_last(&min_vendor, &max_vendor, NSI_VENDOR_CODE);
+
+       if (event_msg == NULL)
+               return (EINVAL);
+
+       /* 
+        * Limit third parties to posting events for registered vendor codes
+        * only
+        */
+       if (event_msg->vendor_code < min_vendor ||
+           event_msg->vendor_code > max_vendor) {
+               OSIncrementAtomic64((SInt64 *)&kevtstat.kes_badvendor);
+               return (EINVAL);
+       }
+       return (kev_post_msg(event_msg));
+}
+
+int
+kev_post_msg(struct kev_msg *event_msg)
+{
+       struct mbuf *m, *m2;
+       struct kern_event_pcb *ev_pcb;
+       struct kern_event_msg *ev;
+       char *tmp;
+       u_int32_t total_size;
+       int i;
+
+       /* Verify the message is small enough to fit in one mbuf w/o cluster */
+       total_size = KEV_MSG_HEADER_SIZE;
+
+       for (i = 0; i < 5; i++) {
+               if (event_msg->dv[i].data_length == 0)
+                       break;
+               total_size += event_msg->dv[i].data_length;
+       }
+
+       if (total_size > MLEN) {
+               OSIncrementAtomic64((SInt64 *)&kevtstat.kes_toobig);
+               return (EMSGSIZE);
+       }
+
+       m = m_get(M_DONTWAIT, MT_DATA);
+       if (m == 0) {
+               OSIncrementAtomic64((SInt64 *)&kevtstat.kes_nomem);
+               return (ENOMEM);
+       }
+       ev = mtod(m, struct kern_event_msg *);
+       total_size = KEV_MSG_HEADER_SIZE;
+
+       tmp = (char *) &ev->event_data[0];
+       for (i = 0; i < 5; i++) {
+               if (event_msg->dv[i].data_length == 0)
+                       break;
+
+               total_size += event_msg->dv[i].data_length;
+               bcopy(event_msg->dv[i].data_ptr, tmp,
+                   event_msg->dv[i].data_length);
+               tmp += event_msg->dv[i].data_length;
+       }
+
+       ev->id = ++static_event_id;
+       ev->total_size   = total_size;
+       ev->vendor_code  = event_msg->vendor_code;
+       ev->kev_class    = event_msg->kev_class;
+       ev->kev_subclass = event_msg->kev_subclass;
+       ev->event_code   = event_msg->event_code;
+
+       m->m_len = total_size;
+       lck_rw_lock_shared(kev_rwlock);
+       for (ev_pcb = LIST_FIRST(&kern_event_head);
+           ev_pcb;
+           ev_pcb = LIST_NEXT(ev_pcb, evp_link)) {
+               lck_mtx_lock(&ev_pcb->evp_mtx);
+               if (ev_pcb->evp_socket->so_pcb == NULL) {
+                       lck_mtx_unlock(&ev_pcb->evp_mtx);
+                       continue;
+               }
+               if (ev_pcb->evp_vendor_code_filter != KEV_ANY_VENDOR) {
+                       if (ev_pcb->evp_vendor_code_filter != ev->vendor_code) {
+                               lck_mtx_unlock(&ev_pcb->evp_mtx);
+                               continue;
+                       }
+
+                       if (ev_pcb->evp_class_filter != KEV_ANY_CLASS) {
+                               if (ev_pcb->evp_class_filter != ev->kev_class) {
+                                       lck_mtx_unlock(&ev_pcb->evp_mtx);
+                                       continue;
+                               }
+
+                               if ((ev_pcb->evp_subclass_filter !=
+                                   KEV_ANY_SUBCLASS) &&
+                                   (ev_pcb->evp_subclass_filter !=
+                                   ev->kev_subclass)) {
+                                       lck_mtx_unlock(&ev_pcb->evp_mtx);
+                                       continue;
+                               }
+                       }
+               }
+
+               m2 = m_copym(m, 0, m->m_len, M_NOWAIT);
+               if (m2 == 0) {
+                       OSIncrementAtomic64((SInt64 *)&kevtstat.kes_nomem);
+                       m_free(m);
+                       lck_mtx_unlock(&ev_pcb->evp_mtx);
+                       lck_rw_done(kev_rwlock);
+                       return (ENOMEM);
+               }
+               if (sbappendrecord(&ev_pcb->evp_socket->so_rcv, m2)) {
+                       /*
+                        * We use "m" for the socket stats as it would be
+                        * unsafe to use "m2"
+                        */
+                       so_inc_recv_data_stat(ev_pcb->evp_socket,
+                           1, m->m_len, SO_TC_BE);
+
+                       sorwakeup(ev_pcb->evp_socket);
+                       OSIncrementAtomic64((SInt64 *)&kevtstat.kes_posted);
+               } else {
+                       OSIncrementAtomic64((SInt64 *)&kevtstat.kes_fullsock);
+               }
+               lck_mtx_unlock(&ev_pcb->evp_mtx);
+       }
+       m_free(m);
+       lck_rw_done(kev_rwlock);
+
+       return (0);
+}
+
+static int
+kev_control(struct socket *so,
+    u_long cmd,
+    caddr_t data,
+    __unused struct ifnet *ifp,
+    __unused struct proc *p)
+{
+       struct kev_request *kev_req = (struct kev_request *) data;
+       struct kern_event_pcb  *ev_pcb;
+       struct kev_vendor_code *kev_vendor;
+       u_int32_t  *id_value = (u_int32_t *) data;
+
+       switch (cmd) {
+               case SIOCGKEVID:
+                       *id_value = static_event_id;
+                       break;
+               case SIOCSKEVFILT:
+                       ev_pcb = (struct kern_event_pcb *) so->so_pcb;
+                       ev_pcb->evp_vendor_code_filter = kev_req->vendor_code;
+                       ev_pcb->evp_class_filter = kev_req->kev_class;
+                       ev_pcb->evp_subclass_filter  = kev_req->kev_subclass;
+                       break;
+               case SIOCGKEVFILT:
+                       ev_pcb = (struct kern_event_pcb *) so->so_pcb;
+                       kev_req->vendor_code = ev_pcb->evp_vendor_code_filter;
+                       kev_req->kev_class   = ev_pcb->evp_class_filter;
+                       kev_req->kev_subclass = ev_pcb->evp_subclass_filter;
+                       break;
+               case SIOCGKEVVENDOR:
+                       kev_vendor = (struct kev_vendor_code *)data;
+                       /* Make sure string is NULL terminated */
+                       kev_vendor->vendor_string[KEV_VENDOR_CODE_MAX_STR_LEN-1] = 0;
+                       return (net_str_id_find_internal(kev_vendor->vendor_string,
+                           &kev_vendor->vendor_code, NSI_VENDOR_CODE, 0));
+               default:
+                       return (ENOTSUP);
+       }
+
+       return (0);
+}
+
+int
+kevt_getstat SYSCTL_HANDLER_ARGS
+{
+#pragma unused(oidp, arg1, arg2)
+       int error = 0;
+
+       lck_rw_lock_shared(kev_rwlock);
+
+       if (req->newptr != USER_ADDR_NULL) {
+               error = EPERM;
+               goto done;
+       }
+       if (req->oldptr == USER_ADDR_NULL) {
+               req->oldidx = sizeof(struct kevtstat);
+               goto done;
+       }
+
+       error = SYSCTL_OUT(req, &kevtstat,
+           MIN(sizeof(struct kevtstat), req->oldlen));
+done:
+       lck_rw_done(kev_rwlock);
+
+       return (error);
+}
+
+__private_extern__ int
+kevt_pcblist SYSCTL_HANDLER_ARGS
+{
+#pragma unused(oidp, arg1, arg2)
+       int error = 0;
+       int n, i;
+       struct xsystmgen xsg;
+       void *buf = NULL;
+       size_t item_size = ROUNDUP64(sizeof (struct xkevtpcb)) +
+               ROUNDUP64(sizeof (struct xsocket_n)) +
+               2 * ROUNDUP64(sizeof (struct xsockbuf_n)) +
+               ROUNDUP64(sizeof (struct xsockstat_n));
+       struct kern_event_pcb  *ev_pcb;
+
+       buf = _MALLOC(item_size, M_TEMP, M_WAITOK | M_ZERO);
+       if (buf == NULL)
+               return (ENOMEM);
+
+       lck_rw_lock_shared(kev_rwlock);
+
+       n = kevtstat.kes_pcbcount;
+
+       if (req->oldptr == USER_ADDR_NULL) {
+               req->oldidx = (n + n/8) * item_size;
+               goto done;
+       }
+       if (req->newptr != USER_ADDR_NULL) {
+               error = EPERM;
+               goto done;
+       }
+       bzero(&xsg, sizeof (xsg));
+       xsg.xg_len = sizeof (xsg);
+       xsg.xg_count = n;
+       xsg.xg_gen = kevtstat.kes_gencnt;
+       xsg.xg_sogen = so_gencnt;
+       error = SYSCTL_OUT(req, &xsg, sizeof (xsg));
+       if (error) {
+               goto done;
+       }
+       /*
+        * We are done if there is no pcb
+        */
+       if (n == 0) {
+               goto done;
+       }
+
+       i = 0;
+       for (i = 0, ev_pcb = LIST_FIRST(&kern_event_head);
+           i < n && ev_pcb != NULL;
+           i++, ev_pcb = LIST_NEXT(ev_pcb, evp_link)) {
+               struct xkevtpcb *xk = (struct xkevtpcb *)buf;
+               struct xsocket_n *xso = (struct xsocket_n *)
+                       ADVANCE64(xk, sizeof (*xk));
+               struct xsockbuf_n *xsbrcv = (struct xsockbuf_n *)
+                       ADVANCE64(xso, sizeof (*xso));
+               struct xsockbuf_n *xsbsnd = (struct xsockbuf_n *)
+                       ADVANCE64(xsbrcv, sizeof (*xsbrcv));
+               struct xsockstat_n *xsostats = (struct xsockstat_n *)
+                       ADVANCE64(xsbsnd, sizeof (*xsbsnd));
+
+               bzero(buf, item_size);
+
+               lck_mtx_lock(&ev_pcb->evp_mtx);
+
+               xk->kep_len = sizeof(struct xkevtpcb);
+               xk->kep_kind = XSO_EVT;
+               xk->kep_evtpcb = (uint64_t)VM_KERNEL_ADDRPERM(ev_pcb);
+               xk->kep_vendor_code_filter = ev_pcb->evp_vendor_code_filter;
+               xk->kep_class_filter = ev_pcb->evp_class_filter;
+               xk->kep_subclass_filter = ev_pcb->evp_subclass_filter;
+
+               sotoxsocket_n(ev_pcb->evp_socket, xso);
+               sbtoxsockbuf_n(ev_pcb->evp_socket ?
+                       &ev_pcb->evp_socket->so_rcv : NULL, xsbrcv);
+               sbtoxsockbuf_n(ev_pcb->evp_socket ?
+                       &ev_pcb->evp_socket->so_snd : NULL, xsbsnd);
+               sbtoxsockstat_n(ev_pcb->evp_socket, xsostats);
+
+               lck_mtx_unlock(&ev_pcb->evp_mtx);
+
+               error = SYSCTL_OUT(req, buf, item_size);
+       }
+
+       if (error == 0) {
+               /*
+                * Give the user an updated idea of our state.
+                * If the generation differs from what we told
+                * her before, she knows that something happened
+                * while we were processing this request, and it
+                * might be necessary to retry.
+                */
+               bzero(&xsg, sizeof (xsg));
+               xsg.xg_len = sizeof (xsg);
+               xsg.xg_count = n;
+               xsg.xg_gen = kevtstat.kes_gencnt;
+               xsg.xg_sogen = so_gencnt;
+               error = SYSCTL_OUT(req, &xsg, sizeof (xsg));
+               if (error) {
+                       goto done;
+               }
+       }
+
+done:
+       lck_rw_done(kev_rwlock);
+
+       return (error);
+}
+
+#endif /* SOCKETS */
+
+
+int
+fill_kqueueinfo(struct kqueue *kq, struct kqueue_info * kinfo)
+{
+       struct vinfo_stat * st;
+
+       st = &kinfo->kq_stat;
+
+       st->vst_size = kq->kq_count;
+       if (kq->kq_state & KQ_KEV64)
+               st->vst_blksize = sizeof(struct kevent64_s);
+       else
+               st->vst_blksize = sizeof(struct kevent);
+       st->vst_mode = S_IFIFO;
+       if (kq->kq_state & KQ_SEL)
+               kinfo->kq_state |=  PROC_KQUEUE_SELECT;
+       if (kq->kq_state & KQ_SLEEP)
+               kinfo->kq_state |= PROC_KQUEUE_SLEEP;
+
+       return (0);
+}
+
+
+void
+knote_markstayqueued(struct knote *kn)
+{
+       kqlock(kn->kn_kq);
+       kn->kn_status |= KN_STAYQUEUED;
+       knote_enqueue(kn);
+       kqunlock(kn->kn_kq);
+}
+
+void
+knote_clearstayqueued(struct knote *kn)
+{
+       kqlock(kn->kn_kq);
+       kn->kn_status &= ~KN_STAYQUEUED;
+       knote_dequeue(kn);
+       kqunlock(kn->kn_kq);
+}