]> git.saurik.com Git - apple/xnu.git/blobdiff - bsd/net/bpf.c
xnu-7195.81.3.tar.gz
[apple/xnu.git] / bsd / net / bpf.c
index 3d025ccbc089284c42b446a1098069d0a53b84b2..37d4c7d6ab77321a453b700ccc0aab067b1227fe 100644 (file)
@@ -1,23 +1,29 @@
 /*
 /*
- * Copyright (c) 2000-2004 Apple Computer, Inc. All rights reserved.
+ * Copyright (c) 2000-2020 Apple Inc. All rights reserved.
  *
  *
- * @APPLE_LICENSE_HEADER_START@
- * 
- * The contents of this file constitute Original Code as defined in and
- * are subject to the Apple Public Source License Version 1.1 (the
- * "License").  You may not use this file except in compliance with the
- * License.  Please obtain a copy of the License at
- * http://www.apple.com/publicsource and read it before using this file.
- * 
- * This Original Code and all software distributed under the License are
- * distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY KIND, EITHER
+ * @APPLE_OSREFERENCE_LICENSE_HEADER_START@
+ *
+ * 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. 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,
  * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
  * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
  * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE OR NON-INFRINGEMENT.  Please see the
- * License for the specific language governing rights and limitations
- * under the License.
- * 
- * @APPLE_LICENSE_HEADER_END@
+ * 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_OSREFERENCE_LICENSE_HEADER_END@
  */
 /*
  * Copyright (c) 1990, 1991, 1993
  */
 /*
  * Copyright (c) 1990, 1991, 1993
  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  * SUCH DAMAGE.
  *
  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  * SUCH DAMAGE.
  *
- *      @(#)bpf.c      8.2 (Berkeley) 3/28/94
+ *     @(#)bpf.c       8.2 (Berkeley) 3/28/94
  *
  * $FreeBSD: src/sys/net/bpf.c,v 1.59.2.5 2001/01/05 04:49:09 jdp Exp $
  */
  *
  * $FreeBSD: src/sys/net/bpf.c,v 1.59.2.5 2001/01/05 04:49:09 jdp Exp $
  */
+/*
+ * NOTICE: This file was modified by SPARTA, Inc. in 2005 to introduce
+ * support for mandatory and extensible security protections.  This notice
+ * is included in support of clause 2.2 (b) of the Apple Public License,
+ * Version 2.0.
+ */
 
 #include "bpf.h"
 
 
 #include "bpf.h"
 
 #include <sys/ttycom.h>
 #include <sys/filedesc.h>
 #include <sys/uio_internal.h>
 #include <sys/ttycom.h>
 #include <sys/filedesc.h>
 #include <sys/uio_internal.h>
+#include <sys/file_internal.h>
+#include <sys/event.h>
 
 
-#if defined(sparc) && BSD < 199103
-#include <sys/stream.h>
-#endif
 #include <sys/poll.h>
 
 #include <sys/socket.h>
 #include <sys/poll.h>
 
 #include <sys/socket.h>
+#include <sys/socketvar.h>
 #include <sys/vnode.h>
 
 #include <net/if.h>
 #include <sys/vnode.h>
 
 #include <net/if.h>
 #include <net/bpfdesc.h>
 
 #include <netinet/in.h>
 #include <net/bpfdesc.h>
 
 #include <netinet/in.h>
+#include <netinet/ip.h>
+#include <netinet/ip6.h>
+#include <netinet/in_pcb.h>
+#include <netinet/in_var.h>
+#include <netinet/ip_var.h>
+#include <netinet/tcp.h>
+#include <netinet/tcp_var.h>
+#include <netinet/udp.h>
+#include <netinet/udp_var.h>
 #include <netinet/if_ether.h>
 #include <netinet/if_ether.h>
+#include <netinet/isakmp.h>
+#include <netinet6/esp.h>
 #include <sys/kernel.h>
 #include <sys/sysctl.h>
 #include <net/firewire.h>
 
 #include <sys/kernel.h>
 #include <sys/sysctl.h>
 #include <net/firewire.h>
 
-#include <machine/spl.h>
 #include <miscfs/devfs/devfs.h>
 #include <net/dlil.h>
 #include <miscfs/devfs/devfs.h>
 #include <net/dlil.h>
+#include <net/pktap.h>
 
 #include <kern/locks.h>
 
 #include <kern/locks.h>
+#include <kern/thread_call.h>
+#include <libkern/section_keywords.h>
 
 
-extern int tvtohz(struct timeval *);
+#include <os/log.h>
 
 
-#if NBPFILTER > 0
+extern int tvtohz(struct timeval *);
 
 
-/*
- * Older BSDs don't have kernel malloc.
- */
-#if BSD < 199103
-extern bcopy();
-static caddr_t bpf_alloc();
-#include <net/bpf_compat.h>
-#define BPF_BUFSIZE (MCLBYTES-8)
-#define UIOMOVE(cp, len, code, uio) uiomove(cp, len, code, uio)
-#else
 #define BPF_BUFSIZE 4096
 #define UIOMOVE(cp, len, code, uio) uiomove(cp, len, uio)
 #define BPF_BUFSIZE 4096
 #define UIOMOVE(cp, len, code, uio) uiomove(cp, len, uio)
-#endif
 
 
+#define PRINET  26                      /* interruptible */
+
+#define ISAKMP_HDR_SIZE (sizeof(struct isakmp) + sizeof(struct isakmp_gen))
+#define ESP_HDR_SIZE sizeof(struct newesp)
 
 
-#define PRINET  26                     /* interruptible */
+typedef void (*pktcopyfunc_t)(const void *, void *, size_t);
 
 /*
  * The default read buffer size is patchable.
  */
 static unsigned int bpf_bufsize = BPF_BUFSIZE;
 
 /*
  * The default read buffer size is patchable.
  */
 static unsigned int bpf_bufsize = BPF_BUFSIZE;
-SYSCTL_INT(_debug, OID_AUTO, bpf_bufsize, CTLFLAG_RW, 
-       &bpf_bufsize, 0, "");
-static unsigned int bpf_maxbufsize = BPF_MAXBUFSIZE;
-SYSCTL_INT(_debug, OID_AUTO, bpf_maxbufsize, CTLFLAG_RW, 
-       &bpf_maxbufsize, 0, "");
+SYSCTL_INT(_debug, OID_AUTO, bpf_bufsize, CTLFLAG_RW | CTLFLAG_LOCKED,
+    &bpf_bufsize, 0, "");
+
+static int sysctl_bpf_maxbufsize SYSCTL_HANDLER_ARGS;
+extern const int copysize_limit_panic;
+#define BPF_MAXSIZE_CAP (copysize_limit_panic >> 1)
+__private_extern__ unsigned int bpf_maxbufsize = BPF_MAXBUFSIZE;
+SYSCTL_PROC(_debug, OID_AUTO, bpf_maxbufsize, CTLTYPE_INT | CTLFLAG_RW | CTLFLAG_LOCKED,
+    &bpf_maxbufsize, 0,
+    sysctl_bpf_maxbufsize, "I", "Default BPF max buffer size");
+
 static unsigned int bpf_maxdevices = 256;
 static unsigned int bpf_maxdevices = 256;
-SYSCTL_UINT(_debug, OID_AUTO, bpf_maxdevices, CTLFLAG_RW, 
-       &bpf_maxdevices, 0, "");
+SYSCTL_UINT(_debug, OID_AUTO, bpf_maxdevices, CTLFLAG_RW | CTLFLAG_LOCKED,
+    &bpf_maxdevices, 0, "");
+/*
+ * bpf_wantpktap controls the defaul visibility of DLT_PKTAP
+ * For OS X is off by default so process need to use the ioctl BPF_WANT_PKTAP
+ * explicitly to be able to use DLT_PKTAP.
+ */
+#if !XNU_TARGET_OS_OSX
+static unsigned int bpf_wantpktap = 1;
+#else /* XNU_TARGET_OS_OSX */
+static unsigned int bpf_wantpktap = 0;
+#endif /* XNU_TARGET_OS_OSX */
+SYSCTL_UINT(_debug, OID_AUTO, bpf_wantpktap, CTLFLAG_RW | CTLFLAG_LOCKED,
+    &bpf_wantpktap, 0, "");
+
+static int bpf_debug = 0;
+SYSCTL_INT(_debug, OID_AUTO, bpf_debug, CTLFLAG_RW | CTLFLAG_LOCKED,
+    &bpf_debug, 0, "");
 
 /*
  *  bpf_iflist is the list of interfaces; each corresponds to an ifnet
  *  bpf_dtab holds pointer to the descriptors, indexed by minor device #
  */
 
 /*
  *  bpf_iflist is the list of interfaces; each corresponds to an ifnet
  *  bpf_dtab holds pointer to the descriptors, indexed by minor device #
  */
-static struct bpf_if   *bpf_iflist;
+static struct bpf_if    *bpf_iflist;
 #ifdef __APPLE__
 /*
  * BSD now stores the bpf_d in the dev_t which is a struct
 #ifdef __APPLE__
 /*
  * BSD now stores the bpf_d in the dev_t which is a struct
@@ -153,88 +193,167 @@ static struct bpf_if     *bpf_iflist;
  * the bpf_d in a separate table indexed by minor device #.
  *
  * The value stored in bpf_dtab[n] represent three states:
  * the bpf_d in a separate table indexed by minor device #.
  *
  * The value stored in bpf_dtab[n] represent three states:
- *  0: device not opened
- *  1: device opening or closing
+ *  NULL: device not opened
+ *  BPF_DEV_RESERVED: device opening or closing
  *  other: device <n> opened with pointer to storage
  */
  *  other: device <n> opened with pointer to storage
  */
-static struct bpf_d    **bpf_dtab = NULL;
+#define BPF_DEV_RESERVED ((struct bpf_d *)(uintptr_t)1)
+static struct bpf_d     **bpf_dtab = NULL;
 static unsigned int bpf_dtab_size = 0;
 static unsigned int bpf_dtab_size = 0;
-static unsigned int    nbpfilter = 0;
+static unsigned int     nbpfilter = 0;
 
 
-static lck_mtx_t               *bpf_mlock;
-static lck_grp_t               *bpf_mlock_grp;
-static lck_grp_attr_t  *bpf_mlock_grp_attr;
-static lck_attr_t              *bpf_mlock_attr;
+decl_lck_mtx_data(static, bpf_mlock_data);
+static lck_mtx_t                *bpf_mlock = &bpf_mlock_data;
+static lck_grp_t                *bpf_mlock_grp;
+static lck_grp_attr_t   *bpf_mlock_grp_attr;
+static lck_attr_t               *bpf_mlock_attr;
 
 
-/*
- * Mark a descriptor free by making it point to itself.
- * This is probably cheaper than marking with a constant since
- * the address should be in a register anyway.
- */
 #endif /* __APPLE__ */
 
 #endif /* __APPLE__ */
 
-static int     bpf_allocbufs(struct bpf_d *);
-static void    bpf_attachd(struct bpf_d *d, struct bpf_if *bp);
-static void    bpf_detachd(struct bpf_d *d);
-static void    bpf_freed(struct bpf_d *);
-static void    bpf_mcopy(const void *, void *, size_t);
-static int     bpf_movein(struct uio *, int,
-                   struct mbuf **, struct sockaddr *, int *);
-static int     bpf_setif(struct bpf_d *, struct ifreq *);
-static void bpf_wakeup(struct bpf_d *);
-static void    catchpacket(struct bpf_d *, u_char *, u_int,
-                   u_int, void (*)(const void *, void *, size_t));
-static void    reset_d(struct bpf_d *);
-static int bpf_setf(struct bpf_d *, struct user_bpf_program *);
-
-/*static  void *bpf_devfs_token[MAXBPFILTER];*/
+static int      bpf_allocbufs(struct bpf_d *);
+static errno_t  bpf_attachd(struct bpf_d *d, struct bpf_if *bp);
+static int      bpf_detachd(struct bpf_d *d, int);
+static void     bpf_freed(struct bpf_d *);
+static int      bpf_movein(struct uio *, int,
+    struct mbuf **, struct sockaddr *, int *);
+static int      bpf_setif(struct bpf_d *, ifnet_t ifp, bool, bool);
+static void     bpf_timed_out(void *, void *);
+static void     bpf_wakeup(struct bpf_d *);
+static u_int    get_pkt_trunc_len(u_char *, u_int);
+static void     catchpacket(struct bpf_d *, struct bpf_packet *, u_int, int);
+static void     reset_d(struct bpf_d *);
+static int      bpf_setf(struct bpf_d *, u_int, user_addr_t, u_long);
+static int      bpf_getdltlist(struct bpf_d *, caddr_t, struct proc *);
+static int      bpf_setdlt(struct bpf_d *, u_int);
+static int      bpf_set_traffic_class(struct bpf_d *, int);
+static void     bpf_set_packet_service_class(struct mbuf *, int);
+
+static void     bpf_acquire_d(struct bpf_d *);
+static void     bpf_release_d(struct bpf_d *);
 
 static  int bpf_devsw_installed;
 
 void bpf_init(void *unused);
 
 static  int bpf_devsw_installed;
 
 void bpf_init(void *unused);
-int bpf_tap_callback(struct ifnet *ifp, struct mbuf *m);
+static int bpf_tap_callback(struct ifnet *ifp, struct mbuf *m);
 
 /*
  * Darwin differs from BSD here, the following are static
  * on BSD and not static on Darwin.
  */
 
 /*
  * Darwin differs from BSD here, the following are static
  * on BSD and not static on Darwin.
  */
-       d_open_t        bpfopen;
-       d_close_t       bpfclose;
-       d_read_t        bpfread;
-       d_write_t       bpfwrite;
-        ioctl_fcn_t    bpfioctl;
-       select_fcn_t    bpfpoll;
-
+d_open_t            bpfopen;
+d_close_t           bpfclose;
+d_read_t            bpfread;
+d_write_t           bpfwrite;
+ioctl_fcn_t         bpfioctl;
+select_fcn_t        bpfselect;
 
 /* Darwin's cdevsw struct differs slightly from BSDs */
 #define CDEV_MAJOR 23
 
 /* Darwin's cdevsw struct differs slightly from BSDs */
 #define CDEV_MAJOR 23
-static struct cdevsw bpf_cdevsw = {
-       /* open */      bpfopen,
-       /* close */     bpfclose,
-       /* read */      bpfread,
-       /* write */     bpfwrite,
-       /* ioctl */     bpfioctl,
-       /* stop */              eno_stop,
-       /* reset */             eno_reset,
-       /* tty */               NULL,
-       /* select */    bpfpoll,
-       /* mmap */              eno_mmap,
-       /* strategy*/   eno_strat,
-       /* getc */              eno_getc,
-       /* putc */              eno_putc,
-       /* type */              0
+static const struct cdevsw bpf_cdevsw = {
+       .d_open       = bpfopen,
+       .d_close      = bpfclose,
+       .d_read       = bpfread,
+       .d_write      = bpfwrite,
+       .d_ioctl      = bpfioctl,
+       .d_stop       = eno_stop,
+       .d_reset      = eno_reset,
+       .d_ttys       = NULL,
+       .d_select     = bpfselect,
+       .d_mmap       = eno_mmap,
+       .d_strategy   = eno_strat,
+       .d_reserved_1 = eno_getc,
+       .d_reserved_2 = eno_putc,
+       .d_type       = 0
 };
 
 };
 
-#define SOCKADDR_HDR_LEN          offsetof(struct sockaddr, sa_data)
+#define SOCKADDR_HDR_LEN           offsetof(struct sockaddr, sa_data)
 
 static int
 
 static int
-bpf_movein(struct uio *uio, int linktype, struct mbuf **mp, struct sockaddr *sockp, int *datlen)
+bpf_movein(struct uio *uio, int linktype, struct mbuf **mp,
+    struct sockaddr *sockp, int *datlen)
 {
        struct mbuf *m;
        int error;
        int len;
 {
        struct mbuf *m;
        int error;
        int len;
+       uint8_t sa_family;
        int hlen;
 
        int hlen;
 
+       switch (linktype) {
+#if SLIP
+       case DLT_SLIP:
+               sa_family = AF_INET;
+               hlen = 0;
+               break;
+#endif /* SLIP */
+
+       case DLT_EN10MB:
+               sa_family = AF_UNSPEC;
+               /* XXX Would MAXLINKHDR be better? */
+               hlen = sizeof(struct ether_header);
+               break;
+
+#if FDDI
+       case DLT_FDDI:
+#if defined(__FreeBSD__) || defined(__bsdi__)
+               sa_family = AF_IMPLINK;
+               hlen = 0;
+#else
+               sa_family = AF_UNSPEC;
+               /* XXX 4(FORMAC)+6(dst)+6(src)+3(LLC)+5(SNAP) */
+               hlen = 24;
+#endif
+               break;
+#endif /* FDDI */
+
+       case DLT_RAW:
+       case DLT_NULL:
+               sa_family = AF_UNSPEC;
+               hlen = 0;
+               break;
+
+#ifdef __FreeBSD__
+       case DLT_ATM_RFC1483:
+               /*
+                * en atm driver requires 4-byte atm pseudo header.
+                * though it isn't standard, vpi:vci needs to be
+                * specified anyway.
+                */
+               sa_family = AF_UNSPEC;
+               hlen = 12;      /* XXX 4(ATM_PH) + 3(LLC) + 5(SNAP) */
+               break;
+#endif
+
+       case DLT_PPP:
+               sa_family = AF_UNSPEC;
+               hlen = 4;       /* This should match PPP_HDRLEN */
+               break;
+
+       case DLT_APPLE_IP_OVER_IEEE1394:
+               sa_family = AF_UNSPEC;
+               hlen = sizeof(struct firewire_header);
+               break;
+
+       case DLT_IEEE802_11:            /* IEEE 802.11 wireless */
+               sa_family = AF_IEEE80211;
+               hlen = 0;
+               break;
+
+       case DLT_IEEE802_11_RADIO:
+               sa_family = AF_IEEE80211;
+               hlen = 0;
+               break;
+
+       default:
+               return EIO;
+       }
+
+       // LP64todo - fix this!
+       len = uio_resid(uio);
+       *datlen = len - hlen;
+       if ((unsigned)len > MCLBYTES) {
+               return EIO;
+       }
+
        if (sockp) {
                /*
                 * Build a sockaddr based on the data link layer type.
        if (sockp) {
                /*
                 * Build a sockaddr based on the data link layer type.
@@ -245,85 +364,27 @@ bpf_movein(struct uio *uio, int linktype, struct mbuf **mp, struct sockaddr *soc
                 * Also, we are careful to leave room at the front of the mbuf
                 * for the link level header.
                 */
                 * Also, we are careful to leave room at the front of the mbuf
                 * for the link level header.
                 */
-               switch (linktype) {
-       
-               case DLT_SLIP:
-                       sockp->sa_family = AF_INET;
-                       hlen = 0;
-                       break;
-       
-               case DLT_EN10MB:
-                       sockp->sa_family = AF_UNSPEC;
-                       /* XXX Would MAXLINKHDR be better? */
-                       hlen = sizeof(struct ether_header);
-                       break;
-       
-               case DLT_FDDI:
-       #if defined(__FreeBSD__) || defined(__bsdi__)
-                       sockp->sa_family = AF_IMPLINK;
-                       hlen = 0;
-       #else
-                       sockp->sa_family = AF_UNSPEC;
-                       /* XXX 4(FORMAC)+6(dst)+6(src)+3(LLC)+5(SNAP) */
-                       hlen = 24;
-       #endif
-                       break;
-       
-               case DLT_RAW:
-               case DLT_NULL:
-                       sockp->sa_family = AF_UNSPEC;
-                       hlen = 0;
-                       break;
-       
-       #ifdef __FreeBSD__
-               case DLT_ATM_RFC1483:
-                       /*
-                        * en atm driver requires 4-byte atm pseudo header.
-                        * though it isn't standard, vpi:vci needs to be
-                        * specified anyway.
-                        */
-                       sockp->sa_family = AF_UNSPEC;
-                       hlen = 12;      /* XXX 4(ATM_PH) + 3(LLC) + 5(SNAP) */
-                       break;
-       #endif
-               case DLT_PPP:
-                       sockp->sa_family = AF_UNSPEC;
-                       hlen = 4;       /* This should match PPP_HDRLEN */
-                       break;
-       
-               case DLT_APPLE_IP_OVER_IEEE1394:
-                       sockp->sa_family = AF_UNSPEC;
-                       hlen = sizeof(struct firewire_header);
-                       break;
-       
-               default:
-                       return (EIO);
-               }
                if ((hlen + SOCKADDR_HDR_LEN) > sockp->sa_len) {
                if ((hlen + SOCKADDR_HDR_LEN) > sockp->sa_len) {
-                       return (EIO);
+                       return EIO;
                }
                }
-       }
-       else {
+               sockp->sa_family = sa_family;
+       } else {
+               /*
+                * We're directly sending the packet data supplied by
+                * the user; we don't need to make room for the link
+                * header, and don't need the header length value any
+                * more, so set it to 0.
+                */
                hlen = 0;
        }
                hlen = 0;
        }
-       
-       // LP64todo - fix this!
-       len = uio_resid(uio);
-       *datlen = len - hlen;
-       if ((unsigned)len > MCLBYTES)
-               return (EIO);
 
        MGETHDR(m, M_WAIT, MT_DATA);
 
        MGETHDR(m, M_WAIT, MT_DATA);
-       if (m == 0)
-               return (ENOBUFS);
+       if (m == 0) {
+               return ENOBUFS;
+       }
        if ((unsigned)len > MHLEN) {
        if ((unsigned)len > MHLEN) {
-#if BSD >= 199103
                MCLGET(m, M_WAIT);
                if ((m->m_flags & M_EXT) == 0) {
                MCLGET(m, M_WAIT);
                if ((m->m_flags & M_EXT) == 0) {
-#else
-               MCLGET(m);
-               if (m->m_len != MCLBYTES) {
-#endif
                        error = ENOBUFS;
                        goto bad;
                }
                        error = ENOBUFS;
                        goto bad;
                }
@@ -331,95 +392,105 @@ bpf_movein(struct uio *uio, int linktype, struct mbuf **mp, struct sockaddr *soc
        m->m_pkthdr.len = m->m_len = len;
        m->m_pkthdr.rcvif = NULL;
        *mp = m;
        m->m_pkthdr.len = m->m_len = len;
        m->m_pkthdr.rcvif = NULL;
        *mp = m;
+
        /*
         * Make room for link header.
         */
        if (hlen != 0) {
                m->m_pkthdr.len -= hlen;
                m->m_len -= hlen;
        /*
         * Make room for link header.
         */
        if (hlen != 0) {
                m->m_pkthdr.len -= hlen;
                m->m_len -= hlen;
-#if BSD >= 199103
                m->m_data += hlen; /* XXX */
                m->m_data += hlen; /* XXX */
-#else
-               m->m_off += hlen;
-#endif
                error = UIOMOVE((caddr_t)sockp->sa_data, hlen, UIO_WRITE, uio);
                error = UIOMOVE((caddr_t)sockp->sa_data, hlen, UIO_WRITE, uio);
-               if (error)
+               if (error) {
                        goto bad;
                        goto bad;
+               }
        }
        error = UIOMOVE(mtod(m, caddr_t), len - hlen, UIO_WRITE, uio);
        }
        error = UIOMOVE(mtod(m, caddr_t), len - hlen, UIO_WRITE, uio);
-       if (!error)
-               return (0);
- bad:
+       if (error) {
+               goto bad;
+       }
+
+       /* Check for multicast destination */
+       switch (linktype) {
+       case DLT_EN10MB: {
+               struct ether_header *eh;
+
+               eh = mtod(m, struct ether_header *);
+               if (ETHER_IS_MULTICAST(eh->ether_dhost)) {
+                       if (_ether_cmp(etherbroadcastaddr,
+                           eh->ether_dhost) == 0) {
+                               m->m_flags |= M_BCAST;
+                       } else {
+                               m->m_flags |= M_MCAST;
+                       }
+               }
+               break;
+       }
+       }
+
+       return 0;
+bad:
        m_freem(m);
        m_freem(m);
-       return (error);
+       return error;
 }
 
 #ifdef __APPLE__
 }
 
 #ifdef __APPLE__
-/* Callback registered with Ethernet driver. */
-int bpf_tap_callback(struct ifnet *ifp, struct mbuf *m)
-{
-    /*
-     * Do nothing if the BPF tap has been turned off.
-     * This is to protect from a potential race where this
-     * call blocks on the lock. And in the meantime
-     * BPF is turned off, which will clear if_bpf.
-     */
-    if (ifp->if_bpf)
-        bpf_mtap(ifp, m);
-    return 0;
-}
 
 /*
 
 /*
- * The dynamic addition of a new device node must block all processes that are opening 
- * the last device so that no process will get an unexpected ENOENT 
+ * The dynamic addition of a new device node must block all processes that
+ * are opening the last device so that no process will get an unexpected
+ * ENOENT
  */
 static void
 bpf_make_dev_t(int maj)
 {
  */
 static void
 bpf_make_dev_t(int maj)
 {
-       static int              bpf_growing = 0;
-       unsigned int    cur_size = nbpfilter, i;
+       static int              bpf_growing = 0;
+       unsigned int    cur_size = nbpfilter, i;
 
 
-       if (nbpfilter >= bpf_maxdevices)
+       if (nbpfilter >= bpf_maxdevices) {
                return;
                return;
+       }
 
        while (bpf_growing) {
                /* Wait until new device has been created */
 
        while (bpf_growing) {
                /* Wait until new device has been created */
-               (void)tsleep((caddr_t)&bpf_growing, PZERO, "bpf_growing", 0);
+               (void) tsleep((caddr_t)&bpf_growing, PZERO, "bpf_growing", 0);
        }
        if (nbpfilter > cur_size) {
                /* other thread grew it already */
                return;
        }
        bpf_growing = 1;
        }
        if (nbpfilter > cur_size) {
                /* other thread grew it already */
                return;
        }
        bpf_growing = 1;
-       
+
        /* need to grow bpf_dtab first */
        if (nbpfilter == bpf_dtab_size) {
                int new_dtab_size;
                struct bpf_d **new_dtab = NULL;
                struct bpf_d **old_dtab = NULL;
        /* need to grow bpf_dtab first */
        if (nbpfilter == bpf_dtab_size) {
                int new_dtab_size;
                struct bpf_d **new_dtab = NULL;
                struct bpf_d **old_dtab = NULL;
-               
-               new_dtab_size = bpf_dtab_size + NBPFILTER;      
-               new_dtab = (struct bpf_d **)_MALLOC(sizeof(struct bpf_d *) * new_dtab_size, M_DEVBUF, M_WAIT);
+
+               new_dtab_size = bpf_dtab_size + NBPFILTER;
+               new_dtab = (struct bpf_d **)_MALLOC(
+                       sizeof(struct bpf_d *) * new_dtab_size, M_DEVBUF, M_WAIT);
                if (new_dtab == 0) {
                        printf("bpf_make_dev_t: malloc bpf_dtab failed\n");
                        goto done;
                }
                if (bpf_dtab) {
                if (new_dtab == 0) {
                        printf("bpf_make_dev_t: malloc bpf_dtab failed\n");
                        goto done;
                }
                if (bpf_dtab) {
-                       bcopy(bpf_dtab, new_dtab, 
-                                 sizeof(struct bpf_d *) * bpf_dtab_size);
+                       bcopy(bpf_dtab, new_dtab,
+                           sizeof(struct bpf_d *) * bpf_dtab_size);
                }
                }
-               bzero(new_dtab + bpf_dtab_size, 
-                         sizeof(struct bpf_d *) * NBPFILTER);
+               bzero(new_dtab + bpf_dtab_size,
+                   sizeof(struct bpf_d *) * NBPFILTER);
                old_dtab = bpf_dtab;
                bpf_dtab = new_dtab;
                bpf_dtab_size = new_dtab_size;
                old_dtab = bpf_dtab;
                bpf_dtab = new_dtab;
                bpf_dtab_size = new_dtab_size;
-               if (old_dtab != NULL)
+               if (old_dtab != NULL) {
                        _FREE(old_dtab, M_DEVBUF);
                        _FREE(old_dtab, M_DEVBUF);
+               }
        }
        i = nbpfilter++;
        (void) devfs_make_node(makedev(maj, i),
        }
        i = nbpfilter++;
        (void) devfs_make_node(makedev(maj, i),
-                               DEVFS_CHAR, UID_ROOT, GID_WHEEL, 0600,
-                               "bpf%d", i);
+           DEVFS_CHAR, UID_ROOT, GID_WHEEL, 0600,
+           "bpf%d", i);
 done:
        bpf_growing = 0;
        wakeup((caddr_t)&bpf_growing);
 done:
        bpf_growing = 0;
        wakeup((caddr_t)&bpf_growing);
@@ -429,11 +500,13 @@ done:
 
 /*
  * Attach file to the bpf interface, i.e. make d listen on bp.
 
 /*
  * Attach file to the bpf interface, i.e. make d listen on bp.
- * Must be called at splimp.
  */
  */
-static void
+static errno_t
 bpf_attachd(struct bpf_d *d, struct bpf_if *bp)
 {
 bpf_attachd(struct bpf_d *d, struct bpf_if *bp)
 {
+       int first = bp->bif_dlist == NULL;
+       int     error = 0;
+
        /*
         * Point d at bp, and add d to the interface's list of listeners.
         * Finally, point the driver's bpf cookie at the interface so
        /*
         * Point d at bp, and add d to the interface's list of listeners.
         * Finally, point the driver's bpf cookie at the interface so
@@ -443,123 +516,341 @@ bpf_attachd(struct bpf_d *d, struct bpf_if *bp)
        d->bd_next = bp->bif_dlist;
        bp->bif_dlist = d;
 
        d->bd_next = bp->bif_dlist;
        bp->bif_dlist = d;
 
-       bp->bif_ifp->if_bpf = bp;
+       /*
+        * Take a reference on the device even if an error is returned
+        * because we keep the device in the interface's list of listeners
+        */
+       bpf_acquire_d(d);
+
+       if (first) {
+               /* Find the default bpf entry for this ifp */
+               if (bp->bif_ifp->if_bpf == NULL) {
+                       struct bpf_if   *tmp, *primary = NULL;
+
+                       for (tmp = bpf_iflist; tmp; tmp = tmp->bif_next) {
+                               if (tmp->bif_ifp == bp->bif_ifp) {
+                                       primary = tmp;
+                                       break;
+                               }
+                       }
+                       bp->bif_ifp->if_bpf = primary;
+               }
+               /* Only call dlil_set_bpf_tap for primary dlt */
+               if (bp->bif_ifp->if_bpf == bp) {
+                       dlil_set_bpf_tap(bp->bif_ifp, BPF_TAP_INPUT_OUTPUT,
+                           bpf_tap_callback);
+               }
+
+               if (bp->bif_tap != NULL) {
+                       error = bp->bif_tap(bp->bif_ifp, bp->bif_dlt,
+                           BPF_TAP_INPUT_OUTPUT);
+               }
+       }
 
 
-#ifdef __APPLE__
-       dlil_set_bpf_tap(bp->bif_ifp, BPF_TAP_INPUT_OUTPUT, bpf_tap_callback);
-#endif
+       /*
+        * Reset the detach flags in case we previously detached an interface
+        */
+       d->bd_flags &= ~(BPF_DETACHING | BPF_DETACHED);
+
+       if (bp->bif_dlt == DLT_PKTAP) {
+               d->bd_flags |= BPF_FINALIZE_PKTAP;
+       } else {
+               d->bd_flags &= ~BPF_FINALIZE_PKTAP;
+       }
+       return error;
 }
 
 /*
  * Detach a file from its interface.
 }
 
 /*
  * Detach a file from its interface.
+ *
+ * Return 1 if was closed by some thread, 0 otherwise
  */
  */
-static void
-bpf_detachd(struct bpf_d *d)
+static int
+bpf_detachd(struct bpf_d *d, int closing)
 {
        struct bpf_d **p;
        struct bpf_if *bp;
 {
        struct bpf_d **p;
        struct bpf_if *bp;
-#ifdef __APPLE__
        struct ifnet  *ifp;
 
        struct ifnet  *ifp;
 
+       int bpf_closed = d->bd_flags & BPF_CLOSING;
+       /*
+        * Some other thread already detached
+        */
+       if ((d->bd_flags & (BPF_DETACHED | BPF_DETACHING)) != 0) {
+               goto done;
+       }
+       /*
+        * This thread is doing the detach
+        */
+       d->bd_flags |= BPF_DETACHING;
+
        ifp = d->bd_bif->bif_ifp;
        ifp = d->bd_bif->bif_ifp;
+       bp = d->bd_bif;
 
 
-#endif
+       if (bpf_debug != 0) {
+               printf("%s: %llx %s%s\n",
+                   __func__, (uint64_t)VM_KERNEL_ADDRPERM(d),
+                   if_name(ifp), closing ? " closing" : "");
+       }
 
 
-       bp = d->bd_bif;
+       /* Remove d from the interface's descriptor list. */
+       p = &bp->bif_dlist;
+       while (*p != d) {
+               p = &(*p)->bd_next;
+               if (*p == 0) {
+                       panic("bpf_detachd: descriptor not in list");
+               }
+       }
+       *p = (*p)->bd_next;
+       if (bp->bif_dlist == 0) {
+               /*
+                * Let the driver know that there are no more listeners.
+                */
+               /* Only call dlil_set_bpf_tap for primary dlt */
+               if (bp->bif_ifp->if_bpf == bp) {
+                       dlil_set_bpf_tap(ifp, BPF_TAP_DISABLE, NULL);
+               }
+               if (bp->bif_tap) {
+                       bp->bif_tap(ifp, bp->bif_dlt, BPF_TAP_DISABLE);
+               }
+
+               for (bp = bpf_iflist; bp; bp = bp->bif_next) {
+                       if (bp->bif_ifp == ifp && bp->bif_dlist != 0) {
+                               break;
+                       }
+               }
+               if (bp == NULL) {
+                       ifp->if_bpf = NULL;
+               }
+       }
+       d->bd_bif = NULL;
        /*
         * Check if this descriptor had requested promiscuous mode.
         * If so, turn it off.
         */
        if (d->bd_promisc) {
                d->bd_promisc = 0;
        /*
         * Check if this descriptor had requested promiscuous mode.
         * If so, turn it off.
         */
        if (d->bd_promisc) {
                d->bd_promisc = 0;
-               if (ifnet_set_promiscuous(bp->bif_ifp, 0))
+               lck_mtx_unlock(bpf_mlock);
+               if (ifnet_set_promiscuous(ifp, 0)) {
                        /*
                         * Something is really wrong if we were able to put
                         * the driver into promiscuous mode, but can't
                         * take it out.
                         * Most likely the network interface is gone.
                         */
                        /*
                         * Something is really wrong if we were able to put
                         * the driver into promiscuous mode, but can't
                         * take it out.
                         * Most likely the network interface is gone.
                         */
-                       printf("bpf: ifnet_set_promiscuous failed");
+                       printf("%s: ifnet_set_promiscuous failed\n", __func__);
+               }
+               lck_mtx_lock(bpf_mlock);
        }
        }
-       /* Remove d from the interface's descriptor list. */
-       p = &bp->bif_dlist;
-       while (*p != d) {
-               p = &(*p)->bd_next;
-               if (*p == 0)
-                       panic("bpf_detachd: descriptor not in list");
+
+       /*
+        * Wake up other thread that are waiting for this thread to finish
+        * detaching
+        */
+       d->bd_flags &= ~BPF_DETACHING;
+       d->bd_flags |= BPF_DETACHED;
+
+       /* Refresh the local variable as d could have been modified */
+       bpf_closed = d->bd_flags & BPF_CLOSING;
+       /*
+        * Note that We've kept the reference because we may have dropped
+        * the lock when turning off promiscuous mode
+        */
+       bpf_release_d(d);
+
+done:
+       /*
+        * When closing makes sure no other thread refer to the bpf_d
+        */
+       if (bpf_debug != 0) {
+               printf("%s: %llx done\n",
+                   __func__, (uint64_t)VM_KERNEL_ADDRPERM(d));
        }
        }
-       *p = (*p)->bd_next;
-       if (bp->bif_dlist == 0) {
+       /*
+        * Let the caller know the bpf_d is closed
+        */
+       if (bpf_closed) {
+               return 1;
+       } else {
+               return 0;
+       }
+}
+
+/*
+ * Start asynchronous timer, if necessary.
+ * Must be called with bpf_mlock held.
+ */
+static void
+bpf_start_timer(struct bpf_d *d)
+{
+       uint64_t deadline;
+       struct timeval tv;
+
+       if (d->bd_rtout > 0 && d->bd_state == BPF_IDLE) {
+               tv.tv_sec = d->bd_rtout / hz;
+               tv.tv_usec = (d->bd_rtout % hz) * tick;
+
+               clock_interval_to_deadline(
+                       (uint64_t)tv.tv_sec * USEC_PER_SEC + tv.tv_usec,
+                       NSEC_PER_USEC, &deadline);
                /*
                /*
-                * Let the driver know that there are no more listeners.
+                * The state is BPF_IDLE, so the timer hasn't
+                * been started yet, and hasn't gone off yet;
+                * there is no thread call scheduled, so this
+                * won't change the schedule.
+                *
+                * XXX - what if, by the time it gets entered,
+                * the deadline has already passed?
                 */
                 */
-               if (ifp->if_set_bpf_tap)
-                       (*ifp->if_set_bpf_tap)(ifp, BPF_TAP_DISABLE, 0);
-               d->bd_bif->bif_ifp->if_bpf = 0;
+               thread_call_enter_delayed(d->bd_thread_call, deadline);
+               d->bd_state = BPF_WAITING;
        }
        }
-       d->bd_bif = 0;
 }
 
 }
 
+/*
+ * Cancel asynchronous timer.
+ * Must be called with bpf_mlock held.
+ */
+static boolean_t
+bpf_stop_timer(struct bpf_d *d)
+{
+       /*
+        * If the timer has already gone off, this does nothing.
+        * Our caller is expected to set d->bd_state to BPF_IDLE,
+        * with the bpf_mlock, after we are called. bpf_timed_out()
+        * also grabs bpf_mlock, so, if the timer has gone off and
+        * bpf_timed_out() hasn't finished, it's waiting for the
+        * lock; when this thread releases the lock, it will
+        * find the state is BPF_IDLE, and just release the
+        * lock and return.
+        */
+       return thread_call_cancel(d->bd_thread_call);
+}
+
+void
+bpf_acquire_d(struct bpf_d *d)
+{
+       void *lr_saved =  __builtin_return_address(0);
+
+       LCK_MTX_ASSERT(bpf_mlock, LCK_MTX_ASSERT_OWNED);
+
+       d->bd_refcnt += 1;
+
+       d->bd_ref_lr[d->bd_next_ref_lr] = lr_saved;
+       d->bd_next_ref_lr = (d->bd_next_ref_lr + 1) % BPF_REF_HIST;
+}
+
+void
+bpf_release_d(struct bpf_d *d)
+{
+       void *lr_saved =  __builtin_return_address(0);
+
+       LCK_MTX_ASSERT(bpf_mlock, LCK_MTX_ASSERT_OWNED);
+
+       if (d->bd_refcnt <= 0) {
+               panic("%s: %p refcnt <= 0", __func__, d);
+       }
+
+       d->bd_refcnt -= 1;
+
+       d->bd_unref_lr[d->bd_next_unref_lr] = lr_saved;
+       d->bd_next_unref_lr = (d->bd_next_unref_lr + 1) % BPF_REF_HIST;
+
+       if (d->bd_refcnt == 0) {
+               /* Assert the device is detached */
+               if ((d->bd_flags & BPF_DETACHED) == 0) {
+                       panic("%s: %p BPF_DETACHED not set", __func__, d);
+               }
+
+               _FREE(d, M_DEVBUF);
+       }
+}
 
 /*
  * Open ethernet device.  Returns ENXIO for illegal minor device number,
  * EBUSY if file is open by another process.
  */
 /* ARGSUSED */
 
 /*
  * Open ethernet device.  Returns ENXIO for illegal minor device number,
  * EBUSY if file is open by another process.
  */
 /* ARGSUSED */
-       int
-bpfopen(dev_t dev, __unused int flags, __unused int fmt, __unused struct proc *p)
+int
+bpfopen(dev_t dev, int flags, __unused int fmt,
+    struct proc *p)
 {
 {
-       register struct bpf_d *d;
-
-       if ((unsigned int) minor(dev) >= nbpfilter)
-               return (ENXIO);
-               
-       /* 
-        * New device nodes are created on demand when opening the last one. 
-        * The programming model is for processes to loop on the minor starting at 0 
-        * as long as EBUSY is returned. The loop stops when either the open succeeds or 
-        * an error other that EBUSY is returned. That means that bpf_make_dev_t() must 
-        * block all processes that are opening the last  node. If not all 
-        * processes are blocked, they could unexpectedly get ENOENT and abort their 
-        * opening loop.
-        */
-       if ((unsigned int) minor(dev) == (nbpfilter - 1))
+       struct bpf_d *d;
+
+       lck_mtx_lock(bpf_mlock);
+       if ((unsigned int) minor(dev) >= nbpfilter) {
+               lck_mtx_unlock(bpf_mlock);
+               return ENXIO;
+       }
+       /*
+        * New device nodes are created on demand when opening the last one.
+        * The programming model is for processes to loop on the minor starting
+        * at 0 as long as EBUSY is returned. The loop stops when either the
+        * open succeeds or an error other that EBUSY is returned. That means
+        * that bpf_make_dev_t() must block all processes that are opening the
+        * last  node. If not all processes are blocked, they could unexpectedly
+        * get ENOENT and abort their opening loop.
+        */
+       if ((unsigned int) minor(dev) == (nbpfilter - 1)) {
                bpf_make_dev_t(major(dev));
                bpf_make_dev_t(major(dev));
+       }
 
        /*
 
        /*
-        * Each minor can be opened by only one process.  If the requested 
+        * Each minor can be opened by only one process.  If the requested
         * minor is in use, return EBUSY.
         *
         * minor is in use, return EBUSY.
         *
-        * Important: bpfopen() and bpfclose() have to check and set the status of a device
-        * in the same lockin context otherwise the device may be leaked because the vnode use count 
-        * will be unpextectly greater than 1 when close() is called.
+        * Important: bpfopen() and bpfclose() have to check and set the status
+        * of a device in the same lockin context otherwise the device may be
+        * leaked because the vnode use count will be unpextectly greater than 1
+        * when close() is called.
         */
         */
-       if (bpf_dtab[minor(dev)] == 0)
-               bpf_dtab[minor(dev)] = (void *)1;       /* Mark opening */
-       else
-               return (EBUSY);
-
-       d = (struct bpf_d *)_MALLOC(sizeof(struct bpf_d), M_DEVBUF, M_WAIT);
+       if (bpf_dtab[minor(dev)] == NULL) {
+               /* Reserve while opening */
+               bpf_dtab[minor(dev)] = BPF_DEV_RESERVED;
+       } else {
+               lck_mtx_unlock(bpf_mlock);
+               return EBUSY;
+       }
+       d = (struct bpf_d *)_MALLOC(sizeof(struct bpf_d), M_DEVBUF,
+           M_WAIT | M_ZERO);
        if (d == NULL) {
                /* this really is a catastrophic failure */
                printf("bpfopen: malloc bpf_d failed\n");
        if (d == NULL) {
                /* this really is a catastrophic failure */
                printf("bpfopen: malloc bpf_d failed\n");
-               bpf_dtab[minor(dev)] = 0;
+               bpf_dtab[minor(dev)] = NULL;
+               lck_mtx_unlock(bpf_mlock);
                return ENOMEM;
        }
                return ENOMEM;
        }
-       bzero(d, sizeof(struct bpf_d));
-       
-       /*
-        * It is not necessary to take the BPF lock here because no other 
-        * thread can access the device until it is marked opened...
-        */
-       
+
        /* Mark "in use" and do most initialization. */
        /* Mark "in use" and do most initialization. */
+       bpf_acquire_d(d);
        d->bd_bufsize = bpf_bufsize;
        d->bd_sig = SIGIO;
        d->bd_seesent = 1;
        d->bd_bufsize = bpf_bufsize;
        d->bd_sig = SIGIO;
        d->bd_seesent = 1;
-       bpf_dtab[minor(dev)] = d;                               /* Mark opened */
+       d->bd_oflags = flags;
+       d->bd_state = BPF_IDLE;
+       d->bd_traffic_class = SO_TC_BE;
+       d->bd_flags |= BPF_DETACHED;
+       if (bpf_wantpktap) {
+               d->bd_flags |= BPF_WANT_PKTAP;
+       } else {
+               d->bd_flags &= ~BPF_WANT_PKTAP;
+       }
+       d->bd_thread_call = thread_call_allocate(bpf_timed_out, d);
+       if (d->bd_thread_call == NULL) {
+               printf("bpfopen: malloc thread call failed\n");
+               bpf_dtab[minor(dev)] = NULL;
+               bpf_release_d(d);
+               lck_mtx_unlock(bpf_mlock);
+
+               return ENOMEM;
+       }
+       d->bd_opened_by = p;
+       uuid_generate(d->bd_uuid);
+
+       bpf_dtab[minor(dev)] = d; /* Mark opened */
+       lck_mtx_unlock(bpf_mlock);
 
 
-       return (0);
+       return 0;
 }
 
 /*
 }
 
 /*
@@ -567,50 +858,146 @@ bpfopen(dev_t dev, __unused int flags, __unused int fmt, __unused struct proc *p
  * deallocating its buffers, and marking it free.
  */
 /* ARGSUSED */
  * deallocating its buffers, and marking it free.
  */
 /* ARGSUSED */
-       int
-bpfclose(dev_t dev, __unused int flags, __unused int fmt, __unused struct proc *p)
+int
+bpfclose(dev_t dev, __unused int flags, __unused int fmt,
+    __unused struct proc *p)
 {
 {
-       register struct bpf_d *d;
-
-       d = bpf_dtab[minor(dev)];
-       if (d == 0 || d == (void *)1)
-               return (ENXIO);
-       
-       bpf_dtab[minor(dev)] = (void *)1;               /* Mark closing */
+       struct bpf_d *d;
 
        /* Take BPF lock to ensure no other thread is using the device */
        lck_mtx_lock(bpf_mlock);
 
 
        /* Take BPF lock to ensure no other thread is using the device */
        lck_mtx_lock(bpf_mlock);
 
-       if (d->bd_bif)
-               bpf_detachd(d);
-       selthreadclear(&d->bd_sel);
-       bpf_freed(d);
-
-       lck_mtx_unlock(bpf_mlock);
-       
-       /* Mark free in same context as bpfopen comes to check */
-       bpf_dtab[minor(dev)] = 0;                               /* Mark closed */
-       _FREE(d, M_DEVBUF);
-       
-       return (0);
-}
+       d = bpf_dtab[minor(dev)];
+       if (d == NULL || d == BPF_DEV_RESERVED) {
+               lck_mtx_unlock(bpf_mlock);
+               return ENXIO;
+       }
 
 
+       /*
+        * Other threads may call bpd_detachd() if we drop the bpf_mlock
+        */
+       d->bd_flags |= BPF_CLOSING;
 
 
-#define BPF_SLEEP bpf_sleep
+       if (bpf_debug != 0) {
+               printf("%s: %llx\n",
+                   __func__, (uint64_t)VM_KERNEL_ADDRPERM(d));
+       }
 
 
-static int
-bpf_sleep(struct bpf_d *d, int pri, const char *wmesg, int timo)
-{
-       register int st;
+       bpf_dtab[minor(dev)] = BPF_DEV_RESERVED; /* Reserve while closing */
 
 
-       lck_mtx_unlock(bpf_mlock);
-       
-       st = tsleep((caddr_t)d, pri, wmesg, timo);
-       
-       lck_mtx_lock(bpf_mlock);
-       
-       return st;
-}
+       /*
+        * Deal with any in-progress timeouts.
+        */
+       switch (d->bd_state) {
+       case BPF_IDLE:
+               /*
+                * Not waiting for a timeout, and no timeout happened.
+                */
+               break;
+
+       case BPF_WAITING:
+               /*
+                * Waiting for a timeout.
+                * Cancel any timer that has yet to go off,
+                * and mark the state as "closing".
+                * Then drop the lock to allow any timers that
+                * *have* gone off to run to completion, and wait
+                * for them to finish.
+                */
+               if (!bpf_stop_timer(d)) {
+                       /*
+                        * There was no pending call, so the call must
+                        * have been in progress. Wait for the call to
+                        * complete; we have to drop the lock while
+                        * waiting. to let the in-progrss call complete
+                        */
+                       d->bd_state = BPF_DRAINING;
+                       while (d->bd_state == BPF_DRAINING) {
+                               msleep((caddr_t)d, bpf_mlock, PRINET,
+                                   "bpfdraining", NULL);
+                       }
+               }
+               d->bd_state = BPF_IDLE;
+               break;
+
+       case BPF_TIMED_OUT:
+               /*
+                * Timer went off, and the timeout routine finished.
+                */
+               d->bd_state = BPF_IDLE;
+               break;
+
+       case BPF_DRAINING:
+               /*
+                * Another thread is blocked on a close waiting for
+                * a timeout to finish.
+                * This "shouldn't happen", as the first thread to enter
+                * bpfclose() will set bpf_dtab[minor(dev)] to 1, and
+                * all subsequent threads should see that and fail with
+                * ENXIO.
+                */
+               panic("Two threads blocked in a BPF close");
+               break;
+       }
+
+       if (d->bd_bif) {
+               bpf_detachd(d, 1);
+       }
+       selthreadclear(&d->bd_sel);
+       thread_call_free(d->bd_thread_call);
+
+       while (d->bd_hbuf_read != 0) {
+               msleep((caddr_t)d, bpf_mlock, PRINET, "bpf_reading", NULL);
+       }
+
+       bpf_freed(d);
+
+       /* Mark free in same context as bpfopen comes to check */
+       bpf_dtab[minor(dev)] = NULL;                    /* Mark closed */
+
+       bpf_release_d(d);
+
+       lck_mtx_unlock(bpf_mlock);
+
+       return 0;
+}
+
+#define BPF_SLEEP bpf_sleep
+
+static int
+bpf_sleep(struct bpf_d *d, int pri, const char *wmesg, int timo)
+{
+       u_int64_t abstime = 0;
+
+       if (timo != 0) {
+               clock_interval_to_deadline(timo, NSEC_PER_SEC / hz, &abstime);
+       }
+
+       return msleep1((caddr_t)d, bpf_mlock, pri, wmesg, abstime);
+}
+
+static void
+bpf_finalize_pktap(struct bpf_hdr *hp, struct pktap_header *pktaphdr)
+{
+       if (pktaphdr->pth_flags & PTH_FLAG_V2_HDR) {
+               struct pktap_v2_hdr *pktap_v2_hdr;
+
+               pktap_v2_hdr = (struct pktap_v2_hdr *)pktaphdr;
+
+               if (pktap_v2_hdr->pth_flags & PTH_FLAG_DELAY_PKTAP) {
+                       pktap_v2_finalize_proc_info(pktap_v2_hdr);
+               }
+       } else {
+               if (pktaphdr->pth_flags & PTH_FLAG_DELAY_PKTAP) {
+                       pktap_finalize_proc_info(pktaphdr);
+               }
+
+               if (pktaphdr->pth_flags & PTH_FLAG_TSTAMP) {
+                       hp->bh_tstamp.tv_sec = pktaphdr->pth_tstamp.tv_sec;
+                       hp->bh_tstamp.tv_usec = pktaphdr->pth_tstamp.tv_usec;
+               }
+       }
+}
 
 /*
  * Rotate the packet buffers in descriptor d.  Move the store buffer
 
 /*
  * Rotate the packet buffers in descriptor d.  Move the store buffer
@@ -618,48 +1005,78 @@ bpf_sleep(struct bpf_d *d, int pri, const char *wmesg, int timo)
  * Zero the length of the new store buffer.
  */
 #define ROTATE_BUFFERS(d) \
  * Zero the length of the new store buffer.
  */
 #define ROTATE_BUFFERS(d) \
+       if (d->bd_hbuf_read != 0) \
+               panic("rotating bpf buffers during read"); \
        (d)->bd_hbuf = (d)->bd_sbuf; \
        (d)->bd_hlen = (d)->bd_slen; \
        (d)->bd_hbuf = (d)->bd_sbuf; \
        (d)->bd_hlen = (d)->bd_slen; \
+       (d)->bd_hcnt = (d)->bd_scnt; \
        (d)->bd_sbuf = (d)->bd_fbuf; \
        (d)->bd_slen = 0; \
        (d)->bd_sbuf = (d)->bd_fbuf; \
        (d)->bd_slen = 0; \
-       (d)->bd_fbuf = 0;
+       (d)->bd_scnt = 0; \
+       (d)->bd_fbuf = NULL;
 /*
  *  bpfread - read next chunk of packets from buffers
  */
 /*
  *  bpfread - read next chunk of packets from buffers
  */
-       int
+int
 bpfread(dev_t dev, struct uio *uio, int ioflag)
 {
 bpfread(dev_t dev, struct uio *uio, int ioflag)
 {
-       register struct bpf_d *d;
+       struct bpf_d *d;
+       caddr_t hbuf;
+       int timed_out, hbuf_len;
        int error;
        int error;
-       int s;
-
-       d = bpf_dtab[minor(dev)];
-       if (d == 0 || d == (void *)1)
-               return (ENXIO);
+       int flags;
 
        lck_mtx_lock(bpf_mlock);
 
 
        lck_mtx_lock(bpf_mlock);
 
+       d = bpf_dtab[minor(dev)];
+       if (d == NULL || d == BPF_DEV_RESERVED ||
+           (d->bd_flags & BPF_CLOSING) != 0) {
+               lck_mtx_unlock(bpf_mlock);
+               return ENXIO;
+       }
+
+       bpf_acquire_d(d);
 
        /*
         * Restrict application to use a buffer the same size as
         * as kernel buffers.
         */
 
        /*
         * Restrict application to use a buffer the same size as
         * as kernel buffers.
         */
-       // LP64todo - fix this
-       if (uio->uio_resid != d->bd_bufsize) {
+       if (uio_resid(uio) != d->bd_bufsize) {
+               bpf_release_d(d);
                lck_mtx_unlock(bpf_mlock);
                lck_mtx_unlock(bpf_mlock);
-               return (EINVAL);
+               return EINVAL;
+       }
+
+       if (d->bd_state == BPF_WAITING) {
+               bpf_stop_timer(d);
+       }
+
+       timed_out = (d->bd_state == BPF_TIMED_OUT);
+       d->bd_state = BPF_IDLE;
+
+       while (d->bd_hbuf_read != 0) {
+               msleep((caddr_t)d, bpf_mlock, PRINET, "bpf_reading", NULL);
        }
 
        }
 
-       s = splimp();
+       if ((d->bd_flags & BPF_CLOSING) != 0) {
+               bpf_release_d(d);
+               lck_mtx_unlock(bpf_mlock);
+               return ENXIO;
+       }
        /*
         * If the hold buffer is empty, then do a timed sleep, which
         * ends when the timeout expires or when enough packets
         * have arrived to fill the store buffer.
         */
        while (d->bd_hbuf == 0) {
        /*
         * If the hold buffer is empty, then do a timed sleep, which
         * ends when the timeout expires or when enough packets
         * have arrived to fill the store buffer.
         */
        while (d->bd_hbuf == 0) {
-               if (d->bd_immediate && d->bd_slen != 0) {
+               if ((d->bd_immediate || timed_out || (ioflag & IO_NDELAY)) &&
+                   d->bd_slen != 0) {
                        /*
                        /*
-                        * A packet(s) either arrived since the previous
+                        * We're in immediate mode, or are reading
+                        * in non-blocking mode, or a timer was
+                        * started before the read (e.g., by select()
+                        * or poll()) and has expired and a packet(s)
+                        * either arrived since the previous
                         * read or arrived while we were asleep.
                         * Rotate the buffers and return what's here.
                         */
                         * read or arrived while we were asleep.
                         * Rotate the buffers and return what's here.
                         */
@@ -674,20 +1091,64 @@ bpfread(dev_t dev, struct uio *uio, int ioflag)
                 * it before using it again.
                 */
                if (d->bd_bif == NULL) {
                 * it before using it again.
                 */
                if (d->bd_bif == NULL) {
-                       splx(s);
+                       bpf_release_d(d);
+                       lck_mtx_unlock(bpf_mlock);
+                       return ENXIO;
+               }
+               if (ioflag & IO_NDELAY) {
+                       bpf_release_d(d);
+                       lck_mtx_unlock(bpf_mlock);
+                       return EWOULDBLOCK;
+               }
+               error = BPF_SLEEP(d, PRINET | PCATCH, "bpf", d->bd_rtout);
+               /*
+                * Make sure device is still opened
+                */
+               if ((d->bd_flags & BPF_CLOSING) != 0) {
+                       bpf_release_d(d);
+                       lck_mtx_unlock(bpf_mlock);
+                       return ENXIO;
+               }
+
+               while (d->bd_hbuf_read != 0) {
+                       msleep((caddr_t)d, bpf_mlock, PRINET, "bpf_reading",
+                           NULL);
+               }
+
+               if ((d->bd_flags & BPF_CLOSING) != 0) {
+                       bpf_release_d(d);
                        lck_mtx_unlock(bpf_mlock);
                        lck_mtx_unlock(bpf_mlock);
-                       return (ENXIO);
-               }
-               
-               if (ioflag & IO_NDELAY)
-                       error = EWOULDBLOCK;
-               else
-                       error = BPF_SLEEP(d, PRINET|PCATCH, "bpf",
-                                         d->bd_rtout);
+                       return ENXIO;
+               }
+
                if (error == EINTR || error == ERESTART) {
                if (error == EINTR || error == ERESTART) {
-                       splx(s);
+                       if (d->bd_hbuf != NULL) {
+                               /*
+                                * Because we msleep, the hold buffer might
+                                * be filled when we wake up.  Avoid rotating
+                                * in this case.
+                                */
+                               break;
+                       }
+                       if (d->bd_slen != 0) {
+                               /*
+                                * Sometimes we may be interrupted often and
+                                * the sleep above will not timeout.
+                                * Regardless, we should rotate the buffers
+                                * if there's any new data pending and
+                                * return it.
+                                */
+                               ROTATE_BUFFERS(d);
+                               break;
+                       }
+                       bpf_release_d(d);
                        lck_mtx_unlock(bpf_mlock);
                        lck_mtx_unlock(bpf_mlock);
-                       return (error);
+                       if (error == ERESTART) {
+                               printf("%s: %llx ERESTART to EINTR\n",
+                                   __func__, (uint64_t)VM_KERNEL_ADDRPERM(d));
+                               error = EINTR;
+                       }
+                       return error;
                }
                if (error == EWOULDBLOCK) {
                        /*
                }
                if (error == EWOULDBLOCK) {
                        /*
@@ -695,18 +1156,19 @@ bpfread(dev_t dev, struct uio *uio, int ioflag)
                         * which may be nothing.  If there is something
                         * in the store buffer, we can rotate the buffers.
                         */
                         * which may be nothing.  If there is something
                         * in the store buffer, we can rotate the buffers.
                         */
-                       if (d->bd_hbuf)
+                       if (d->bd_hbuf) {
                                /*
                                 * We filled up the buffer in between
                                 * getting the timeout and arriving
                                 * here, so we don't need to rotate.
                                 */
                                break;
                                /*
                                 * We filled up the buffer in between
                                 * getting the timeout and arriving
                                 * here, so we don't need to rotate.
                                 */
                                break;
+                       }
 
                        if (d->bd_slen == 0) {
 
                        if (d->bd_slen == 0) {
-                               splx(s);
+                               bpf_release_d(d);
                                lck_mtx_unlock(bpf_mlock);
                                lck_mtx_unlock(bpf_mlock);
-                               return (0);
+                               return 0;
                        }
                        ROTATE_BUFFERS(d);
                        break;
                        }
                        ROTATE_BUFFERS(d);
                        break;
@@ -715,126 +1177,439 @@ bpfread(dev_t dev, struct uio *uio, int ioflag)
        /*
         * At this point, we know we have something in the hold slot.
         */
        /*
         * At this point, we know we have something in the hold slot.
         */
-       splx(s);
+
+       /*
+        * Set the hold buffer read. So we do not
+        * rotate the buffers until the hold buffer
+        * read is complete. Also to avoid issues resulting
+        * from page faults during disk sleep (<rdar://problem/13436396>).
+        */
+       d->bd_hbuf_read = 1;
+       hbuf = d->bd_hbuf;
+       hbuf_len = d->bd_hlen;
+       flags = d->bd_flags;
+       lck_mtx_unlock(bpf_mlock);
+
+#ifdef __APPLE__
+       /*
+        * Before we move data to userland, we fill out the extended
+        * header fields.
+        */
+       if (flags & BPF_EXTENDED_HDR) {
+               char *p;
+
+               p = hbuf;
+               while (p < hbuf + hbuf_len) {
+                       struct bpf_hdr_ext *ehp;
+                       uint32_t flowid;
+                       struct so_procinfo soprocinfo;
+                       int found = 0;
+
+                       ehp = (struct bpf_hdr_ext *)(void *)p;
+                       if ((flowid = ehp->bh_flowid) != 0) {
+                               if (ehp->bh_proto == IPPROTO_TCP) {
+                                       found = inp_findinpcb_procinfo(&tcbinfo,
+                                           flowid, &soprocinfo);
+                               } else if (ehp->bh_proto == IPPROTO_UDP) {
+                                       found = inp_findinpcb_procinfo(&udbinfo,
+                                           flowid, &soprocinfo);
+                               }
+                               if (found == 1) {
+                                       ehp->bh_pid = soprocinfo.spi_pid;
+                                       strlcpy(&ehp->bh_comm[0], &soprocinfo.spi_proc_name[0], sizeof(ehp->bh_comm));
+                               }
+                               ehp->bh_flowid = 0;
+                       }
+
+                       if (flags & BPF_FINALIZE_PKTAP) {
+                               struct pktap_header *pktaphdr;
+
+                               pktaphdr = (struct pktap_header *)(void *)
+                                   (p + BPF_WORDALIGN(ehp->bh_hdrlen));
+
+                               bpf_finalize_pktap((struct bpf_hdr *) ehp,
+                                   pktaphdr);
+                       }
+                       p += BPF_WORDALIGN(ehp->bh_hdrlen + ehp->bh_caplen);
+               }
+       } else if (flags & BPF_FINALIZE_PKTAP) {
+               char *p;
+
+               p = hbuf;
+               while (p < hbuf + hbuf_len) {
+                       struct bpf_hdr *hp;
+                       struct pktap_header *pktaphdr;
+
+                       hp = (struct bpf_hdr *)(void *)p;
+                       pktaphdr = (struct pktap_header *)(void *)
+                           (p + BPF_WORDALIGN(hp->bh_hdrlen));
+
+                       bpf_finalize_pktap(hp, pktaphdr);
+
+                       p += BPF_WORDALIGN(hp->bh_hdrlen + hp->bh_caplen);
+               }
+       }
+#endif
 
        /*
         * Move data from hold buffer into user space.
         * We know the entire buffer is transferred since
         * we checked above that the read buffer is bpf_bufsize bytes.
         */
 
        /*
         * Move data from hold buffer into user space.
         * We know the entire buffer is transferred since
         * we checked above that the read buffer is bpf_bufsize bytes.
         */
-       error = UIOMOVE(d->bd_hbuf, d->bd_hlen, UIO_READ, uio);
+       error = UIOMOVE(hbuf, hbuf_len, UIO_READ, uio);
+
+       lck_mtx_lock(bpf_mlock);
+       /*
+        * Make sure device is still opened
+        */
+       if ((d->bd_flags & BPF_CLOSING) != 0) {
+               bpf_release_d(d);
+               lck_mtx_unlock(bpf_mlock);
+               return ENXIO;
+       }
 
 
-       s = splimp();
+       d->bd_hbuf_read = 0;
        d->bd_fbuf = d->bd_hbuf;
        d->bd_fbuf = d->bd_hbuf;
-       d->bd_hbuf = 0;
+       d->bd_hbuf = NULL;
        d->bd_hlen = 0;
        d->bd_hlen = 0;
-       splx(s);
+       d->bd_hcnt = 0;
+       wakeup((caddr_t)d);
+
+       bpf_release_d(d);
        lck_mtx_unlock(bpf_mlock);
        lck_mtx_unlock(bpf_mlock);
-       return (error);
+       return error;
 }
 
 }
 
-
 /*
  * If there are processes sleeping on this descriptor, wake them up.
  */
 static void
 bpf_wakeup(struct bpf_d *d)
 {
 /*
  * If there are processes sleeping on this descriptor, wake them up.
  */
 static void
 bpf_wakeup(struct bpf_d *d)
 {
+       if (d->bd_state == BPF_WAITING) {
+               bpf_stop_timer(d);
+               d->bd_state = BPF_IDLE;
+       }
        wakeup((caddr_t)d);
        wakeup((caddr_t)d);
-       if (d->bd_async && d->bd_sig && d->bd_sigio)
-               pgsigio(d->bd_sigio, d->bd_sig, 0);
+       if (d->bd_async && d->bd_sig && d->bd_sigio) {
+               pgsigio(d->bd_sigio, d->bd_sig);
+       }
 
 
-#if BSD >= 199103
        selwakeup(&d->bd_sel);
        selwakeup(&d->bd_sel);
-#ifndef __APPLE__
-       /* XXX */
-       d->bd_sel.si_pid = 0;
-#endif
-#else
-       if (d->bd_selproc) {
-               selwakeup(d->bd_selproc, (int)d->bd_selcoll);
-               d->bd_selcoll = 0;
-               d->bd_selproc = 0;
+       if ((d->bd_flags & BPF_KNOTE)) {
+               KNOTE(&d->bd_sel.si_note, 1);
        }
        }
-#endif
+}
+
+static void
+bpf_timed_out(void *arg, __unused void *dummy)
+{
+       struct bpf_d *d = (struct bpf_d *)arg;
+
+       lck_mtx_lock(bpf_mlock);
+       if (d->bd_state == BPF_WAITING) {
+               /*
+                * There's a select or kqueue waiting for this; if there's
+                * now stuff to read, wake it up.
+                */
+               d->bd_state = BPF_TIMED_OUT;
+               if (d->bd_slen != 0) {
+                       bpf_wakeup(d);
+               }
+       } else if (d->bd_state == BPF_DRAINING) {
+               /*
+                * A close is waiting for this to finish.
+                * Mark it as finished, and wake the close up.
+                */
+               d->bd_state = BPF_IDLE;
+               bpf_wakeup(d);
+       }
+       lck_mtx_unlock(bpf_mlock);
 }
 
 /* keep in sync with bpf_movein above: */
 }
 
 /* keep in sync with bpf_movein above: */
-#define MAX_DATALINK_HDR_LEN   (sizeof(struct firewire_header))
+#define MAX_DATALINK_HDR_LEN    (sizeof(struct firewire_header))
 
 
-       int
+int
 bpfwrite(dev_t dev, struct uio *uio, __unused int ioflag)
 {
 bpfwrite(dev_t dev, struct uio *uio, __unused int ioflag)
 {
-       register struct bpf_d *d;
+       struct bpf_d *d;
        struct ifnet *ifp;
        struct ifnet *ifp;
-       struct mbuf *m;
+       struct mbuf *m = NULL;
        int error;
        int error;
-       char              dst_buf[SOCKADDR_HDR_LEN + MAX_DATALINK_HDR_LEN];
-       int datlen;
+       char              dst_buf[SOCKADDR_HDR_LEN + MAX_DATALINK_HDR_LEN];
+       int datlen = 0;
+       int bif_dlt;
+       int bd_hdrcmplt;
 
 
-       d = bpf_dtab[minor(dev)];
-       if (d == 0 || d == (void *)1)
-               return (ENXIO);
-       
        lck_mtx_lock(bpf_mlock);
        lck_mtx_lock(bpf_mlock);
-       
+
+       d = bpf_dtab[minor(dev)];
+       if (d == NULL || d == BPF_DEV_RESERVED ||
+           (d->bd_flags & BPF_CLOSING) != 0) {
+               lck_mtx_unlock(bpf_mlock);
+               return ENXIO;
+       }
+
+       bpf_acquire_d(d);
+
        if (d->bd_bif == 0) {
        if (d->bd_bif == 0) {
+               bpf_release_d(d);
                lck_mtx_unlock(bpf_mlock);
                lck_mtx_unlock(bpf_mlock);
-            return (ENXIO);
+               return ENXIO;
        }
 
        ifp = d->bd_bif->bif_ifp;
 
        }
 
        ifp = d->bd_bif->bif_ifp;
 
-       if (uio->uio_resid == 0) {
+       if ((ifp->if_flags & IFF_UP) == 0) {
+               bpf_release_d(d);
+               lck_mtx_unlock(bpf_mlock);
+               return ENETDOWN;
+       }
+       if (uio_resid(uio) == 0) {
+               bpf_release_d(d);
                lck_mtx_unlock(bpf_mlock);
                lck_mtx_unlock(bpf_mlock);
-            return (0);
+               return 0;
        }
        ((struct sockaddr *)dst_buf)->sa_len = sizeof(dst_buf);
        }
        ((struct sockaddr *)dst_buf)->sa_len = sizeof(dst_buf);
-       error = bpf_movein(uio, (int)d->bd_bif->bif_dlt, &m, 
-                          d->bd_hdrcmplt ? 0 : (struct sockaddr *)dst_buf, &datlen);
+
+       /*
+        * fix for PR-6849527
+        * geting variables onto stack before dropping lock for bpf_movein()
+        */
+       bif_dlt = (int)d->bd_bif->bif_dlt;
+       bd_hdrcmplt  = d->bd_hdrcmplt;
+
+       /* bpf_movein allocating mbufs; drop lock */
+       lck_mtx_unlock(bpf_mlock);
+
+       error = bpf_movein(uio, bif_dlt, &m,
+           bd_hdrcmplt ? NULL : (struct sockaddr *)dst_buf,
+           &datlen);
+
+       /* take the lock again */
+       lck_mtx_lock(bpf_mlock);
        if (error) {
        if (error) {
+               bpf_release_d(d);
+               lck_mtx_unlock(bpf_mlock);
+               return error;
+       }
+
+       /* verify the device is still open */
+       if ((d->bd_flags & BPF_CLOSING) != 0) {
+               bpf_release_d(d);
+               lck_mtx_unlock(bpf_mlock);
+               m_freem(m);
+               return ENXIO;
+       }
+
+       if (d->bd_bif == NULL) {
+               bpf_release_d(d);
                lck_mtx_unlock(bpf_mlock);
                lck_mtx_unlock(bpf_mlock);
-            return (error);
+               m_free(m);
+               return ENXIO;
        }
 
        if ((unsigned)datlen > ifp->if_mtu) {
        }
 
        if ((unsigned)datlen > ifp->if_mtu) {
+               bpf_release_d(d);
                lck_mtx_unlock(bpf_mlock);
                lck_mtx_unlock(bpf_mlock);
-            return (EMSGSIZE);
+               m_freem(m);
+               return EMSGSIZE;
        }
 
        }
 
+       bpf_set_packet_service_class(m, d->bd_traffic_class);
+
        lck_mtx_unlock(bpf_mlock);
 
        lck_mtx_unlock(bpf_mlock);
 
-       if (d->bd_hdrcmplt) {
-               error = dlil_output(ifp, 0, m, NULL, NULL, 1);
-       }
-       else {
-               error = dlil_output(ifp, PF_INET, m, NULL, (struct sockaddr *)dst_buf, 0);
-       }
-       
        /*
         * The driver frees the mbuf.
         */
        /*
         * The driver frees the mbuf.
         */
-       return (error);
+       if (d->bd_hdrcmplt) {
+               if (d->bd_bif->bif_send) {
+                       error = d->bd_bif->bif_send(ifp, d->bd_bif->bif_dlt, m);
+               } else {
+                       error = dlil_output(ifp, 0, m, NULL, NULL, 1, NULL);
+               }
+       } else {
+               error = dlil_output(ifp, PF_INET, m, NULL,
+                   (struct sockaddr *)dst_buf, 0, NULL);
+       }
+
+       lck_mtx_lock(bpf_mlock);
+       bpf_release_d(d);
+       lck_mtx_unlock(bpf_mlock);
+
+       return error;
 }
 
 /*
  * Reset a descriptor by flushing its packet buffer and clearing the
 }
 
 /*
  * Reset a descriptor by flushing its packet buffer and clearing the
- * receive and drop counts.  Should be called at splimp.
+ * receive and drop counts.
  */
 static void
 reset_d(struct bpf_d *d)
 {
  */
 static void
 reset_d(struct bpf_d *d)
 {
+       if (d->bd_hbuf_read != 0) {
+               panic("resetting buffers during read");
+       }
+
        if (d->bd_hbuf) {
                /* Free the hold buffer. */
                d->bd_fbuf = d->bd_hbuf;
        if (d->bd_hbuf) {
                /* Free the hold buffer. */
                d->bd_fbuf = d->bd_hbuf;
-               d->bd_hbuf = 0;
+               d->bd_hbuf = NULL;
        }
        d->bd_slen = 0;
        d->bd_hlen = 0;
        }
        d->bd_slen = 0;
        d->bd_hlen = 0;
+       d->bd_scnt = 0;
+       d->bd_hcnt = 0;
        d->bd_rcount = 0;
        d->bd_dcount = 0;
 }
 
        d->bd_rcount = 0;
        d->bd_dcount = 0;
 }
 
+static struct bpf_d *
+bpf_get_device_from_uuid(uuid_t uuid)
+{
+       unsigned int i;
+
+       for (i = 0; i < nbpfilter; i++) {
+               struct bpf_d *d = bpf_dtab[i];
+
+               if (d == NULL || d == BPF_DEV_RESERVED ||
+                   (d->bd_flags & BPF_CLOSING) != 0) {
+                       continue;
+               }
+               if (uuid_compare(uuid, d->bd_uuid) == 0) {
+                       return d;
+               }
+       }
+
+       return NULL;
+}
+
+/*
+ * The BIOCSETUP command "atomically" attach to the interface and
+ * copy the buffer from another interface. This minimizes the risk
+ * of missing packet because this is done while holding
+ * the BPF global lock
+ */
+static int
+bpf_setup(struct bpf_d *d_to, uuid_t uuid_from, ifnet_t ifp)
+{
+       struct bpf_d *d_from;
+       int error = 0;
+
+       LCK_MTX_ASSERT(bpf_mlock, LCK_MTX_ASSERT_OWNED);
+
+       /*
+        * Sanity checks
+        */
+       d_from = bpf_get_device_from_uuid(uuid_from);
+       if (d_from == NULL) {
+               error = ENOENT;
+               os_log_info(OS_LOG_DEFAULT,
+                   "%s: uuids not found error %d",
+                   __func__, error);
+               return error;
+       }
+       if (d_from->bd_opened_by != d_to->bd_opened_by) {
+               error = EACCES;
+               os_log_info(OS_LOG_DEFAULT,
+                   "%s: processes not matching error %d",
+                   __func__, error);
+               return error;
+       }
+
+       /*
+        * Prevent any read while copying
+        */
+       while (d_to->bd_hbuf_read != 0) {
+               msleep((caddr_t)d_to, bpf_mlock, PRINET, __func__, NULL);
+       }
+       d_to->bd_hbuf_read = 1;
+
+       while (d_from->bd_hbuf_read != 0) {
+               msleep((caddr_t)d_from, bpf_mlock, PRINET, __func__, NULL);
+       }
+       d_from->bd_hbuf_read = 1;
+
+       /*
+        * Verify the devices have not been closed
+        */
+       if (d_to->bd_flags & BPF_CLOSING) {
+               error = ENXIO;
+               os_log_info(OS_LOG_DEFAULT,
+                   "%s: d_to is closing error %d",
+                   __func__, error);
+               goto done;
+       }
+       if (d_from->bd_flags & BPF_CLOSING) {
+               error = ENXIO;
+               os_log_info(OS_LOG_DEFAULT,
+                   "%s: d_from is closing error %d",
+                   __func__, error);
+               goto done;
+       }
+
+       /*
+        * For now require the same buffer size
+        */
+       if (d_from->bd_bufsize != d_to->bd_bufsize) {
+               error = EINVAL;
+               os_log_info(OS_LOG_DEFAULT,
+                   "%s: bufsizes not matching error %d",
+                   __func__, error);
+               goto done;
+       }
+
+       /*
+        * Attach to the interface
+        */
+       error = bpf_setif(d_to, ifp, false, true);
+       if (error != 0) {
+               os_log_info(OS_LOG_DEFAULT,
+                   "%s: bpf_setif() failed error %d",
+                   __func__, error);
+               goto done;
+       }
+
+       /*
+        * Make sure the buffers are setup as expected by bpf_setif()
+        */
+       ASSERT(d_to->bd_hbuf == NULL);
+       ASSERT(d_to->bd_sbuf != NULL);
+       ASSERT(d_to->bd_fbuf != NULL);
+
+       /*
+        * Copy the buffers and update the pointers and counts
+        */
+       memcpy(d_to->bd_sbuf, d_from->bd_sbuf, d_from->bd_slen);
+       d_to->bd_slen = d_from->bd_slen;
+       d_to->bd_scnt = d_from->bd_scnt;
+
+       if (d_from->bd_hbuf != NULL) {
+               d_to->bd_hbuf = d_to->bd_fbuf;
+               d_to->bd_fbuf = NULL;
+               memcpy(d_to->bd_hbuf, d_from->bd_hbuf, d_from->bd_hlen);
+       }
+       d_to->bd_hlen = d_from->bd_hlen;
+       d_to->bd_hcnt = d_from->bd_hcnt;
+
+       if (bpf_debug > 0) {
+               os_log_info(OS_LOG_DEFAULT,
+                   "%s: done slen %u scnt %u hlen %u hcnt %u",
+                   __func__, d_to->bd_slen, d_to->bd_scnt,
+                   d_to->bd_hlen, d_to->bd_hcnt);
+       }
+done:
+       d_from->bd_hbuf_read = 0;
+       wakeup((caddr_t)d_from);
+
+       d_to->bd_hbuf_read = 0;
+       wakeup((caddr_t)d_to);
+
+       return error;
+}
+
 /*
  *  FIONREAD           Check for read packet available.
  *  SIOCGIFADDR                Get interface address - convenient hook to driver.
 /*
  *  FIONREAD           Check for read packet available.
  *  SIOCGIFADDR                Get interface address - convenient hook to driver.
@@ -854,22 +1629,39 @@ reset_d(struct bpf_d *d)
  *  BIOCSHDRCMPLT      Set "header already complete" flag
  *  BIOCGSEESENT       Get "see packets sent" flag
  *  BIOCSSEESENT       Set "see packets sent" flag
  *  BIOCSHDRCMPLT      Set "header already complete" flag
  *  BIOCGSEESENT       Get "see packets sent" flag
  *  BIOCSSEESENT       Set "see packets sent" flag
+ *  BIOCSETTC          Set traffic class.
+ *  BIOCGETTC          Get traffic class.
+ *  BIOCSEXTHDR                Set "extended header" flag
+ *  BIOCSHEADDROP      Drop head of the buffer if user is not reading
+ *  BIOCGHEADDROP      Get "head-drop" flag
  */
 /* ARGSUSED */
 int
  */
 /* ARGSUSED */
 int
-bpfioctl(dev_t dev, u_long cmd, caddr_t addr, __unused int flags, struct proc *p)
+bpfioctl(dev_t dev, u_long cmd, caddr_t addr, __unused int flags,
+    struct proc *p)
 {
 {
-       register struct bpf_d *d;
-       int s, error = 0;
+       struct bpf_d *d;
+       int error = 0;
+       u_int int_arg;
+       struct ifreq ifr;
+
+       lck_mtx_lock(bpf_mlock);
 
        d = bpf_dtab[minor(dev)];
 
        d = bpf_dtab[minor(dev)];
-       if (d == 0 || d == (void *)1)
-               return (ENXIO);
+       if (d == NULL || d == BPF_DEV_RESERVED ||
+           (d->bd_flags & BPF_CLOSING) != 0) {
+               lck_mtx_unlock(bpf_mlock);
+               return ENXIO;
+       }
 
 
-       lck_mtx_lock(bpf_mlock);
+       bpf_acquire_d(d);
 
 
-       switch (cmd) {
+       if (d->bd_state == BPF_WAITING) {
+               bpf_stop_timer(d);
+       }
+       d->bd_state = BPF_IDLE;
 
 
+       switch (cmd) {
        default:
                error = EINVAL;
                break;
        default:
                error = EINVAL;
                break;
@@ -877,86 +1669,119 @@ bpfioctl(dev_t dev, u_long cmd, caddr_t addr, __unused int flags, struct proc *p
        /*
         * Check for read packet available.
         */
        /*
         * Check for read packet available.
         */
-       case FIONREAD:
-               {
-                       int n;
-
-                       s = splimp();
-                       n = d->bd_slen;
-                       if (d->bd_hbuf)
-                               n += d->bd_hlen;
-                       splx(s);
+       case FIONREAD:                  /* int */
+       {
+               int n;
 
 
-                       *(int *)addr = n;
-                       break;
+               n = d->bd_slen;
+               if (d->bd_hbuf && d->bd_hbuf_read == 0) {
+                       n += d->bd_hlen;
                }
 
                }
 
-       case SIOCGIFADDR:
-               {
-                       struct ifnet *ifp;
+               bcopy(&n, addr, sizeof(n));
+               break;
+       }
+
+       case SIOCGIFADDR:               /* struct ifreq */
+       {
+               struct ifnet *ifp;
 
 
-                       if (d->bd_bif == 0)
-                               error = EINVAL;
-                       else {
-                               ifp = d->bd_bif->bif_ifp;
-                               error = dlil_ioctl(0, ifp, cmd, addr);
-                       }
-                       break;
+               if (d->bd_bif == 0) {
+                       error = EINVAL;
+               } else {
+                       ifp = d->bd_bif->bif_ifp;
+                       error = ifnet_ioctl(ifp, 0, cmd, addr);
                }
                }
+               break;
+       }
 
        /*
         * Get buffer len [for read()].
         */
 
        /*
         * Get buffer len [for read()].
         */
-       case BIOCGBLEN:
-               *(u_int *)addr = d->bd_bufsize;
+       case BIOCGBLEN:                 /* u_int */
+               bcopy(&d->bd_bufsize, addr, sizeof(u_int));
                break;
 
        /*
         * Set buffer length.
         */
                break;
 
        /*
         * Set buffer length.
         */
-       case BIOCSBLEN:
-#if BSD < 199103
-               error = EINVAL;
-#else
-               if (d->bd_bif != 0)
+       case BIOCSBLEN: {               /* u_int */
+               u_int size;
+               unsigned int maxbufsize = bpf_maxbufsize;
+
+               /*
+                * Allow larger buffer in head drop mode to with the
+                * assumption the reading process may be low priority but
+                * is interested in the most recent traffic
+                */
+               if (d->bd_headdrop != 0) {
+                       maxbufsize = 2 * bpf_maxbufsize;
+               }
+
+               if (d->bd_bif != 0 || (d->bd_flags & BPF_DETACHING)) {
+                       /*
+                        * Interface already attached, unable to change buffers
+                        */
                        error = EINVAL;
                        error = EINVAL;
-               else {
-                       register u_int size = *(u_int *)addr;
+                       break;
+               }
+               bcopy(addr, &size, sizeof(size));
+
+               if (size > maxbufsize) {
+                       d->bd_bufsize = maxbufsize;
 
 
-                       if (size > bpf_maxbufsize)
-                               *(u_int *)addr = size = bpf_maxbufsize;
-                       else if (size < BPF_MINBUFSIZE)
-                               *(u_int *)addr = size = BPF_MINBUFSIZE;
+                       os_log_info(OS_LOG_DEFAULT,
+                           "%s bufsize capped to %u from %u",
+                           __func__, d->bd_bufsize, size);
+               } else if (size < BPF_MINBUFSIZE) {
+                       d->bd_bufsize = BPF_MINBUFSIZE;
+
+                       os_log_info(OS_LOG_DEFAULT,
+                           "%s bufsize bumped to %u from %u",
+                           __func__, d->bd_bufsize, size);
+               } else {
                        d->bd_bufsize = size;
                }
                        d->bd_bufsize = size;
                }
-#endif
-               break;
 
 
+               /* It's a read/write ioctl */
+               bcopy(&d->bd_bufsize, addr, sizeof(u_int));
+               break;
+       }
        /*
         * Set link layer read filter.
         */
        /*
         * Set link layer read filter.
         */
-       case BIOCSETF:
-               if (proc_is64bit(p)) {
-                       error = bpf_setf(d, (struct user_bpf_program *)addr);
-               }
-               else {
-                       struct bpf_program *    tmpp;
-                       struct user_bpf_program tmp;
+       case BIOCSETF32:
+       case BIOCSETFNR32: {            /* struct bpf_program32 */
+               struct bpf_program32 prg32;
 
 
-                       tmpp = (struct bpf_program *)addr;
-                       tmp.bf_len = tmpp->bf_len;
-                       tmp.bf_insns = CAST_USER_ADDR_T(tmpp->bf_insns);
-                       error = bpf_setf(d, &tmp);
-               }
+               bcopy(addr, &prg32, sizeof(prg32));
+               error = bpf_setf(d, prg32.bf_len,
+                   CAST_USER_ADDR_T(prg32.bf_insns), cmd);
+               break;
+       }
+
+       case BIOCSETF64:
+       case BIOCSETFNR64: {            /* struct bpf_program64 */
+               struct bpf_program64 prg64;
+
+               bcopy(addr, &prg64, sizeof(prg64));
+               error = bpf_setf(d, prg64.bf_len, prg64.bf_insns, cmd);
                break;
                break;
+       }
 
        /*
         * Flush read packet buffer.
         */
        case BIOCFLUSH:
 
        /*
         * Flush read packet buffer.
         */
        case BIOCFLUSH:
-               s = splimp();
+               while (d->bd_hbuf_read != 0) {
+                       msleep((caddr_t)d, bpf_mlock, PRINET, "bpf_reading",
+                           NULL);
+               }
+               if ((d->bd_flags & BPF_CLOSING) != 0) {
+                       error = ENXIO;
+                       break;
+               }
                reset_d(d);
                reset_d(d);
-               splx(s);
                break;
 
        /*
                break;
 
        /*
@@ -970,136 +1795,232 @@ bpfioctl(dev_t dev, u_long cmd, caddr_t addr, __unused int flags, struct proc *p
                        error = EINVAL;
                        break;
                }
                        error = EINVAL;
                        break;
                }
-               s = splimp();
                if (d->bd_promisc == 0) {
                if (d->bd_promisc == 0) {
+                       lck_mtx_unlock(bpf_mlock);
                        error = ifnet_set_promiscuous(d->bd_bif->bif_ifp, 1);
                        error = ifnet_set_promiscuous(d->bd_bif->bif_ifp, 1);
-                       if (error == 0)
+                       lck_mtx_lock(bpf_mlock);
+                       if (error == 0) {
                                d->bd_promisc = 1;
                                d->bd_promisc = 1;
+                       }
                }
                }
-               splx(s);
                break;
 
        /*
         * Get device parameters.
         */
                break;
 
        /*
         * Get device parameters.
         */
-       case BIOCGDLT:
-               if (d->bd_bif == 0)
+       case BIOCGDLT:                  /* u_int */
+               if (d->bd_bif == 0) {
+                       error = EINVAL;
+               } else {
+                       bcopy(&d->bd_bif->bif_dlt, addr, sizeof(u_int));
+               }
+               break;
+
+       /*
+        * Get a list of supported data link types.
+        */
+       case BIOCGDLTLIST:              /* struct bpf_dltlist */
+               if (d->bd_bif == NULL) {
+                       error = EINVAL;
+               } else {
+                       error = bpf_getdltlist(d, addr, p);
+               }
+               break;
+
+       /*
+        * Set data link type.
+        */
+       case BIOCSDLT:                  /* u_int */
+               if (d->bd_bif == NULL) {
                        error = EINVAL;
                        error = EINVAL;
-               else
-                       *(u_int *)addr = d->bd_bif->bif_dlt;
+               } else {
+                       u_int dlt;
+
+                       bcopy(addr, &dlt, sizeof(dlt));
+
+                       if (dlt == DLT_PKTAP &&
+                           !(d->bd_flags & BPF_WANT_PKTAP)) {
+                               dlt = DLT_RAW;
+                       }
+                       error = bpf_setdlt(d, dlt);
+               }
                break;
 
        /*
         * Get interface name.
         */
                break;
 
        /*
         * Get interface name.
         */
-       case BIOCGETIF:
-               if (d->bd_bif == 0)
+       case BIOCGETIF:                 /* struct ifreq */
+               if (d->bd_bif == 0) {
                        error = EINVAL;
                        error = EINVAL;
-               else {
+               else {
                        struct ifnet *const ifp = d->bd_bif->bif_ifp;
                        struct ifnet *const ifp = d->bd_bif->bif_ifp;
-                       struct ifreq *const ifr = (struct ifreq *)addr;
 
 
-                       snprintf(ifr->ifr_name, sizeof(ifr->ifr_name),
-                           "%s%d", ifp->if_name, ifp->if_unit);
+                       snprintf(((struct ifreq *)(void *)addr)->ifr_name,
+                           sizeof(ifr.ifr_name), "%s", if_name(ifp));
                }
                break;
 
        /*
         * Set interface.
         */
                }
                break;
 
        /*
         * Set interface.
         */
-       case BIOCSETIF:
-               error = bpf_setif(d, (struct ifreq *)addr);
+       case BIOCSETIF: {               /* struct ifreq */
+               ifnet_t ifp;
+
+               bcopy(addr, &ifr, sizeof(ifr));
+               ifr.ifr_name[IFNAMSIZ - 1] = '\0';
+               ifp = ifunit(ifr.ifr_name);
+               if (ifp == NULL) {
+                       error = ENXIO;
+               } else {
+                       error = bpf_setif(d, ifp, true, false);
+               }
                break;
                break;
+       }
 
        /*
         * Set read timeout.
         */
 
        /*
         * Set read timeout.
         */
-       case BIOCSRTIMEOUT:
-               {
-                       struct timeval *tv = (struct timeval *)addr;
+       case BIOCSRTIMEOUT32: {         /* struct user32_timeval */
+               struct user32_timeval _tv;
+               struct timeval tv;
 
 
-                       /*
-                        * Subtract 1 tick from tvtohz() since this isn't
-                        * a one-shot timer.
-                        */
-                       if ((error = itimerfix(tv)) == 0)
-                               d->bd_rtout = tvtohz(tv) - 1;
-                       break;
+               bcopy(addr, &_tv, sizeof(_tv));
+               tv.tv_sec  = _tv.tv_sec;
+               tv.tv_usec = _tv.tv_usec;
+
+               /*
+                * Subtract 1 tick from tvtohz() since this isn't
+                * a one-shot timer.
+                */
+               if ((error = itimerfix(&tv)) == 0) {
+                       d->bd_rtout = tvtohz(&tv) - 1;
+               }
+               break;
+       }
+
+       case BIOCSRTIMEOUT64: {         /* struct user64_timeval */
+               struct user64_timeval _tv;
+               struct timeval tv;
+
+               bcopy(addr, &_tv, sizeof(_tv));
+               tv.tv_sec  = _tv.tv_sec;
+               tv.tv_usec = _tv.tv_usec;
+
+               /*
+                * Subtract 1 tick from tvtohz() since this isn't
+                * a one-shot timer.
+                */
+               if ((error = itimerfix(&tv)) == 0) {
+                       d->bd_rtout = tvtohz(&tv) - 1;
                }
                }
+               break;
+       }
 
        /*
         * Get read timeout.
         */
 
        /*
         * Get read timeout.
         */
-       case BIOCGRTIMEOUT:
-               {
-                       struct timeval *tv = (struct timeval *)addr;
+       case BIOCGRTIMEOUT32: {         /* struct user32_timeval */
+               struct user32_timeval tv;
 
 
-                       tv->tv_sec = d->bd_rtout / hz;
-                       tv->tv_usec = (d->bd_rtout % hz) * tick;
-                       break;
-               }
+               bzero(&tv, sizeof(tv));
+               tv.tv_sec = d->bd_rtout / hz;
+               tv.tv_usec = (d->bd_rtout % hz) * tick;
+               bcopy(&tv, addr, sizeof(tv));
+               break;
+       }
+
+       case BIOCGRTIMEOUT64: {         /* struct user64_timeval */
+               struct user64_timeval tv;
+
+               bzero(&tv, sizeof(tv));
+               tv.tv_sec = d->bd_rtout / hz;
+               tv.tv_usec = (d->bd_rtout % hz) * tick;
+               bcopy(&tv, addr, sizeof(tv));
+               break;
+       }
 
        /*
         * Get packet stats.
         */
 
        /*
         * Get packet stats.
         */
-       case BIOCGSTATS:
-               {
-                       struct bpf_stat *bs = (struct bpf_stat *)addr;
+       case BIOCGSTATS: {              /* struct bpf_stat */
+               struct bpf_stat bs;
 
 
-                       bs->bs_recv = d->bd_rcount;
-                       bs->bs_drop = d->bd_dcount;
-                       break;
-               }
+               bzero(&bs, sizeof(bs));
+               bs.bs_recv = d->bd_rcount;
+               bs.bs_drop = d->bd_dcount;
+               bcopy(&bs, addr, sizeof(bs));
+               break;
+       }
 
        /*
         * Set immediate mode.
         */
 
        /*
         * Set immediate mode.
         */
-       case BIOCIMMEDIATE:
-               d->bd_immediate = *(u_int *)addr;
+       case BIOCIMMEDIATE:             /* u_int */
+               d->bd_immediate = *(u_int *)(void *)addr;
                break;
 
                break;
 
-       case BIOCVERSION:
-               {
-                       struct bpf_version *bv = (struct bpf_version *)addr;
+       case BIOCVERSION: {             /* struct bpf_version */
+               struct bpf_version bv;
 
 
-                       bv->bv_major = BPF_MAJOR_VERSION;
-                       bv->bv_minor = BPF_MINOR_VERSION;
-                       break;
-               }
+               bzero(&bv, sizeof(bv));
+               bv.bv_major = BPF_MAJOR_VERSION;
+               bv.bv_minor = BPF_MINOR_VERSION;
+               bcopy(&bv, addr, sizeof(bv));
+               break;
+       }
 
        /*
         * Get "header already complete" flag
         */
 
        /*
         * Get "header already complete" flag
         */
-       case BIOCGHDRCMPLT:
-               *(u_int *)addr = d->bd_hdrcmplt;
+       case BIOCGHDRCMPLT:             /* u_int */
+               bcopy(&d->bd_hdrcmplt, addr, sizeof(u_int));
                break;
 
        /*
         * Set "header already complete" flag
         */
                break;
 
        /*
         * Set "header already complete" flag
         */
-       case BIOCSHDRCMPLT:
-               d->bd_hdrcmplt = *(u_int *)addr ? 1 : 0;
+       case BIOCSHDRCMPLT:             /* u_int */
+               bcopy(addr, &int_arg, sizeof(int_arg));
+               d->bd_hdrcmplt = int_arg ? 1 : 0;
                break;
 
        /*
         * Get "see sent packets" flag
         */
                break;
 
        /*
         * Get "see sent packets" flag
         */
-       case BIOCGSEESENT:
-               *(u_int *)addr = d->bd_seesent;
+       case BIOCGSEESENT:              /* u_int */
+               bcopy(&d->bd_seesent, addr, sizeof(u_int));
                break;
 
        /*
         * Set "see sent packets" flag
         */
                break;
 
        /*
         * Set "see sent packets" flag
         */
-       case BIOCSSEESENT:
-               d->bd_seesent = *(u_int *)addr;
+       case BIOCSSEESENT:              /* u_int */
+               bcopy(addr, &d->bd_seesent, sizeof(u_int));
+               break;
+
+       /*
+        * Set traffic service class
+        */
+       case BIOCSETTC: {               /* int */
+               int tc;
+
+               bcopy(addr, &tc, sizeof(int));
+               error = bpf_set_traffic_class(d, tc);
+               break;
+       }
+
+       /*
+        * Get traffic service class
+        */
+       case BIOCGETTC:                 /* int */
+               bcopy(&d->bd_traffic_class, addr, sizeof(int));
                break;
 
                break;
 
-       case FIONBIO:           /* Non-blocking I/O */
+       case FIONBIO:           /* Non-blocking I/O; int */
                break;
 
                break;
 
-       case FIOASYNC:          /* Send signal on receive packets */
-               d->bd_async = *(int *)addr;
+       case FIOASYNC:          /* Send signal on receive packets; int */
+               bcopy(addr, &d->bd_async, sizeof(int));
                break;
 #ifndef __APPLE__
        case FIOSETOWN:
                break;
 #ifndef __APPLE__
        case FIOSETOWN:
@@ -1120,288 +2041,1253 @@ bpfioctl(dev_t dev, u_long cmd, caddr_t addr, __unused int flags, struct proc *p
                *(int *)addr = -fgetown(d->bd_sigio);
                break;
 #endif
                *(int *)addr = -fgetown(d->bd_sigio);
                break;
 #endif
-       case BIOCSRSIG:         /* Set receive signal */
-               {
-                       u_int sig;
+       case BIOCSRSIG: {       /* Set receive signal; u_int */
+               u_int sig;
+
+               bcopy(addr, &sig, sizeof(u_int));
+
+               if (sig >= NSIG) {
+                       error = EINVAL;
+               } else {
+                       d->bd_sig = sig;
+               }
+               break;
+       }
+       case BIOCGRSIG:                 /* u_int */
+               bcopy(&d->bd_sig, addr, sizeof(u_int));
+               break;
+#ifdef __APPLE__
+       case BIOCSEXTHDR:               /* u_int */
+               bcopy(addr, &int_arg, sizeof(int_arg));
+               if (int_arg) {
+                       d->bd_flags |= BPF_EXTENDED_HDR;
+               } else {
+                       d->bd_flags &= ~BPF_EXTENDED_HDR;
+               }
+               break;
+
+       case BIOCGIFATTACHCOUNT: {              /* struct ifreq */
+               ifnet_t ifp;
+               struct bpf_if *bp;
+
+               bcopy(addr, &ifr, sizeof(ifr));
+               ifr.ifr_name[IFNAMSIZ - 1] = '\0';
+               ifp = ifunit(ifr.ifr_name);
+               if (ifp == NULL) {
+                       error = ENXIO;
+                       break;
+               }
+               ifr.ifr_intval = 0;
+               for (bp = bpf_iflist; bp != 0; bp = bp->bif_next) {
+                       struct bpf_d *bpf_d;
+
+                       if (bp->bif_ifp == NULL || bp->bif_ifp != ifp) {
+                               continue;
+                       }
+                       for (bpf_d = bp->bif_dlist; bpf_d;
+                           bpf_d = bpf_d->bd_next) {
+                               ifr.ifr_intval += 1;
+                       }
+               }
+               bcopy(&ifr, addr, sizeof(ifr));
+               break;
+       }
+       case BIOCGWANTPKTAP:                    /* u_int */
+               int_arg = d->bd_flags & BPF_WANT_PKTAP ? 1 : 0;
+               bcopy(&int_arg, addr, sizeof(int_arg));
+               break;
+
+       case BIOCSWANTPKTAP:                    /* u_int */
+               bcopy(addr, &int_arg, sizeof(int_arg));
+               if (int_arg) {
+                       d->bd_flags |= BPF_WANT_PKTAP;
+               } else {
+                       d->bd_flags &= ~BPF_WANT_PKTAP;
+               }
+               break;
+#endif
+
+       case BIOCSHEADDROP:
+               bcopy(addr, &int_arg, sizeof(int_arg));
+               d->bd_headdrop = int_arg ? 1 : 0;
+               break;
+
+       case BIOCGHEADDROP:
+               bcopy(&d->bd_headdrop, addr, sizeof(int));
+               break;
+
+       case BIOCSTRUNCATE:
+               bcopy(addr, &int_arg, sizeof(int_arg));
+               if (int_arg) {
+                       d->bd_flags |=  BPF_TRUNCATE;
+               } else {
+                       d->bd_flags &= ~BPF_TRUNCATE;
+               }
+               break;
 
 
-                       sig = *(u_int *)addr;
+       case BIOCGETUUID:
+               bcopy(&d->bd_uuid, addr, sizeof(uuid_t));
+               break;
 
 
-                       if (sig >= NSIG)
-                               error = EINVAL;
-                       else
-                               d->bd_sig = sig;
+       case BIOCSETUP: {
+               struct bpf_setup_args bsa;
+               ifnet_t ifp;
+
+               bcopy(addr, &bsa, sizeof(struct bpf_setup_args));
+               bsa.bsa_ifname[IFNAMSIZ - 1] = 0;
+               ifp = ifunit(bsa.bsa_ifname);
+               if (ifp == NULL) {
+                       error = ENXIO;
+                       os_log_info(OS_LOG_DEFAULT,
+                           "%s: ifnet not found for %s error %d",
+                           __func__, bsa.bsa_ifname, error);
                        break;
                }
                        break;
                }
-       case BIOCGRSIG:
-               *(u_int *)addr = d->bd_sig;
+
+               error = bpf_setup(d, bsa.bsa_uuid, ifp);
+               break;
+       }
+       case BIOCSPKTHDRV2:
+               bcopy(addr, &int_arg, sizeof(int_arg));
+               if (int_arg != 0) {
+                       d->bd_flags |= BPF_PKTHDRV2;
+               } else {
+                       d->bd_flags &= ~BPF_PKTHDRV2;
+               }
+               break;
+
+       case BIOCGPKTHDRV2:
+               int_arg = d->bd_flags & BPF_PKTHDRV2 ? 1 : 0;
+               bcopy(&int_arg, addr, sizeof(int));
+               break;
+       }
+
+       bpf_release_d(d);
+       lck_mtx_unlock(bpf_mlock);
+
+       return error;
+}
+
+/*
+ * Set d's packet filter program to fp.  If this file already has a filter,
+ * free it and replace it.  Returns EINVAL for bogus requests.
+ */
+static int
+bpf_setf(struct bpf_d *d, u_int bf_len, user_addr_t bf_insns,
+    u_long cmd)
+{
+       struct bpf_insn *fcode, *old;
+       u_int flen, size;
+
+       while (d->bd_hbuf_read != 0) {
+               msleep((caddr_t)d, bpf_mlock, PRINET, "bpf_reading", NULL);
+       }
+
+       if ((d->bd_flags & BPF_CLOSING) != 0) {
+               return ENXIO;
+       }
+
+       old = d->bd_filter;
+       if (bf_insns == USER_ADDR_NULL) {
+               if (bf_len != 0) {
+                       return EINVAL;
+               }
+               d->bd_filter = NULL;
+               reset_d(d);
+               if (old != 0) {
+                       FREE(old, M_DEVBUF);
+               }
+               return 0;
+       }
+       flen = bf_len;
+       if (flen > BPF_MAXINSNS) {
+               return EINVAL;
+       }
+
+       size = flen * sizeof(struct bpf_insn);
+       fcode = (struct bpf_insn *) _MALLOC(size, M_DEVBUF, M_WAIT);
+#ifdef __APPLE__
+       if (fcode == NULL) {
+               return ENOBUFS;
+       }
+#endif
+       if (copyin(bf_insns, (caddr_t)fcode, size) == 0 &&
+           bpf_validate(fcode, (int)flen)) {
+               d->bd_filter = fcode;
+
+               if (cmd == BIOCSETF32 || cmd == BIOCSETF64) {
+                       reset_d(d);
+               }
+
+               if (old != 0) {
+                       FREE(old, M_DEVBUF);
+               }
+
+               return 0;
+       }
+       FREE(fcode, M_DEVBUF);
+       return EINVAL;
+}
+
+/*
+ * Detach a file from its current interface (if attached at all) and attach
+ * to the interface indicated by the name stored in ifr.
+ * Return an errno or 0.
+ */
+static int
+bpf_setif(struct bpf_d *d, ifnet_t theywant, bool do_reset, bool has_hbuf_read)
+{
+       struct bpf_if *bp;
+       int error;
+
+       while (d->bd_hbuf_read != 0 && !has_hbuf_read) {
+               msleep((caddr_t)d, bpf_mlock, PRINET, "bpf_reading", NULL);
+       }
+
+       if ((d->bd_flags & BPF_CLOSING) != 0) {
+               return ENXIO;
+       }
+
+       /*
+        * Look through attached interfaces for the named one.
+        */
+       for (bp = bpf_iflist; bp != 0; bp = bp->bif_next) {
+               struct ifnet *ifp = bp->bif_ifp;
+
+               if (ifp == 0 || ifp != theywant) {
+                       continue;
+               }
+               /*
+                * Do not use DLT_PKTAP, unless requested explicitly
+                */
+               if (bp->bif_dlt == DLT_PKTAP && !(d->bd_flags & BPF_WANT_PKTAP)) {
+                       continue;
+               }
+               /*
+                * Skip the coprocessor interface
+                */
+               if (!intcoproc_unrestricted && IFNET_IS_INTCOPROC(ifp)) {
+                       continue;
+               }
+               /*
+                * We found the requested interface.
+                * Allocate the packet buffers.
+                */
+               error = bpf_allocbufs(d);
+               if (error != 0) {
+                       return error;
+               }
+               /*
+                * Detach if attached to something else.
+                */
+               if (bp != d->bd_bif) {
+                       if (d->bd_bif != NULL) {
+                               if (bpf_detachd(d, 0) != 0) {
+                                       return ENXIO;
+                               }
+                       }
+                       if (bpf_attachd(d, bp) != 0) {
+                               return ENXIO;
+                       }
+               }
+               if (do_reset) {
+                       reset_d(d);
+               }
+               return 0;
+       }
+       /* Not found. */
+       return ENXIO;
+}
+
+/*
+ * Get a list of available data link type of the interface.
+ */
+static int
+bpf_getdltlist(struct bpf_d *d, caddr_t addr, struct proc *p)
+{
+       u_int           n;
+       int             error;
+       struct ifnet    *ifp;
+       struct bpf_if   *bp;
+       user_addr_t     dlist;
+       struct bpf_dltlist bfl;
+
+       bcopy(addr, &bfl, sizeof(bfl));
+       if (proc_is64bit(p)) {
+               dlist = (user_addr_t)bfl.bfl_u.bflu_pad;
+       } else {
+               dlist = CAST_USER_ADDR_T(bfl.bfl_u.bflu_list);
+       }
+
+       ifp = d->bd_bif->bif_ifp;
+       n = 0;
+       error = 0;
+
+       for (bp = bpf_iflist; bp; bp = bp->bif_next) {
+               if (bp->bif_ifp != ifp) {
+                       continue;
+               }
+               /*
+                * Do not use DLT_PKTAP, unless requested explicitly
+                */
+               if (bp->bif_dlt == DLT_PKTAP && !(d->bd_flags & BPF_WANT_PKTAP)) {
+                       continue;
+               }
+               if (dlist != USER_ADDR_NULL) {
+                       if (n >= bfl.bfl_len) {
+                               return ENOMEM;
+                       }
+                       error = copyout(&bp->bif_dlt, dlist,
+                           sizeof(bp->bif_dlt));
+                       if (error != 0) {
+                               break;
+                       }
+                       dlist += sizeof(bp->bif_dlt);
+               }
+               n++;
+       }
+       bfl.bfl_len = n;
+       bcopy(&bfl, addr, sizeof(bfl));
+
+       return error;
+}
+
+/*
+ * Set the data link type of a BPF instance.
+ */
+static int
+bpf_setdlt(struct bpf_d *d, uint32_t dlt)
+{
+       int error, opromisc;
+       struct ifnet *ifp;
+       struct bpf_if *bp;
+
+       if (d->bd_bif->bif_dlt == dlt) {
+               return 0;
+       }
+
+       while (d->bd_hbuf_read != 0) {
+               msleep((caddr_t)d, bpf_mlock, PRINET, "bpf_reading", NULL);
+       }
+
+       if ((d->bd_flags & BPF_CLOSING) != 0) {
+               return ENXIO;
+       }
+
+       ifp = d->bd_bif->bif_ifp;
+       for (bp = bpf_iflist; bp; bp = bp->bif_next) {
+               if (bp->bif_ifp == ifp && bp->bif_dlt == dlt) {
+                       /*
+                        * Do not use DLT_PKTAP, unless requested explicitly
+                        */
+                       if (bp->bif_dlt == DLT_PKTAP &&
+                           !(d->bd_flags & BPF_WANT_PKTAP)) {
+                               continue;
+                       }
+                       break;
+               }
+       }
+       if (bp != NULL) {
+               opromisc = d->bd_promisc;
+               if (bpf_detachd(d, 0) != 0) {
+                       return ENXIO;
+               }
+               error = bpf_attachd(d, bp);
+               if (error) {
+                       printf("bpf_setdlt: bpf_attachd %s%d failed (%d)\n",
+                           ifnet_name(bp->bif_ifp), ifnet_unit(bp->bif_ifp),
+                           error);
+                       return error;
+               }
+               reset_d(d);
+               if (opromisc) {
+                       lck_mtx_unlock(bpf_mlock);
+                       error = ifnet_set_promiscuous(bp->bif_ifp, 1);
+                       lck_mtx_lock(bpf_mlock);
+                       if (error) {
+                               printf("%s: ifpromisc %s%d failed (%d)\n",
+                                   __func__, ifnet_name(bp->bif_ifp),
+                                   ifnet_unit(bp->bif_ifp), error);
+                       } else {
+                               d->bd_promisc = 1;
+                       }
+               }
+       }
+       return bp == NULL ? EINVAL : 0;
+}
+
+static int
+bpf_set_traffic_class(struct bpf_d *d, int tc)
+{
+       int error = 0;
+
+       if (!SO_VALID_TC(tc)) {
+               error = EINVAL;
+       } else {
+               d->bd_traffic_class = tc;
+       }
+
+       return error;
+}
+
+static void
+bpf_set_packet_service_class(struct mbuf *m, int tc)
+{
+       if (!(m->m_flags & M_PKTHDR)) {
+               return;
+       }
+
+       VERIFY(SO_VALID_TC(tc));
+       (void) m_set_service_class(m, so_tc2msc(tc));
+}
+
+/*
+ * Support for select()
+ *
+ * Return true iff the specific operation will not block indefinitely.
+ * Otherwise, return false but make a note that a selwakeup() must be done.
+ */
+int
+bpfselect(dev_t dev, int which, void * wql, struct proc *p)
+{
+       struct bpf_d *d;
+       int ret = 0;
+
+       lck_mtx_lock(bpf_mlock);
+
+       d = bpf_dtab[minor(dev)];
+       if (d == NULL || d == BPF_DEV_RESERVED ||
+           (d->bd_flags & BPF_CLOSING) != 0) {
+               lck_mtx_unlock(bpf_mlock);
+               return ENXIO;
+       }
+
+       bpf_acquire_d(d);
+
+       if (d->bd_bif == NULL) {
+               bpf_release_d(d);
+               lck_mtx_unlock(bpf_mlock);
+               return ENXIO;
+       }
+
+       while (d->bd_hbuf_read != 0) {
+               msleep((caddr_t)d, bpf_mlock, PRINET, "bpf_reading", NULL);
+       }
+
+       if ((d->bd_flags & BPF_CLOSING) != 0) {
+               bpf_release_d(d);
+               lck_mtx_unlock(bpf_mlock);
+               return ENXIO;
+       }
+
+       switch (which) {
+       case FREAD:
+               if (d->bd_hlen != 0 ||
+                   ((d->bd_immediate ||
+                   d->bd_state == BPF_TIMED_OUT) && d->bd_slen != 0)) {
+                       ret = 1;         /* read has data to return */
+               } else {
+                       /*
+                        * Read has no data to return.
+                        * Make the select wait, and start a timer if
+                        * necessary.
+                        */
+                       selrecord(p, &d->bd_sel, wql);
+                       bpf_start_timer(d);
+               }
+               break;
+
+       case FWRITE:
+               /* can't determine whether a write would block */
+               ret = 1;
+               break;
+       }
+
+       bpf_release_d(d);
+       lck_mtx_unlock(bpf_mlock);
+
+       return ret;
+}
+
+/*
+ * Support for kevent() system call.  Register EVFILT_READ filters and
+ * reject all others.
+ */
+int bpfkqfilter(dev_t dev, struct knote *kn);
+static void filt_bpfdetach(struct knote *);
+static int filt_bpfread(struct knote *, long);
+static int filt_bpftouch(struct knote *kn, struct kevent_qos_s *kev);
+static int filt_bpfprocess(struct knote *kn, struct kevent_qos_s *kev);
+
+SECURITY_READ_ONLY_EARLY(struct filterops) bpfread_filtops = {
+       .f_isfd = 1,
+       .f_detach = filt_bpfdetach,
+       .f_event = filt_bpfread,
+       .f_touch = filt_bpftouch,
+       .f_process = filt_bpfprocess,
+};
+
+static int
+filt_bpfread_common(struct knote *kn, struct kevent_qos_s *kev, struct bpf_d *d)
+{
+       int ready = 0;
+       int64_t data = 0;
+
+       if (d->bd_immediate) {
+               /*
+                * If there's data in the hold buffer, it's the
+                * amount of data a read will return.
+                *
+                * If there's no data in the hold buffer, but
+                * there's data in the store buffer, a read will
+                * immediately rotate the store buffer to the
+                * hold buffer, the amount of data in the store
+                * buffer is the amount of data a read will
+                * return.
+                *
+                * If there's no data in either buffer, we're not
+                * ready to read.
+                */
+               data = (d->bd_hlen == 0 || d->bd_hbuf_read != 0 ?
+                   d->bd_slen : d->bd_hlen);
+               int64_t lowwat = knote_low_watermark(kn);
+               if (lowwat > d->bd_bufsize) {
+                       lowwat = d->bd_bufsize;
+               }
+               ready = (data >= lowwat);
+       } else {
+               /*
+                * If there's data in the hold buffer, it's the
+                * amount of data a read will return.
+                *
+                * If there's no data in the hold buffer, but
+                * there's data in the store buffer, if the
+                * timer has expired a read will immediately
+                * rotate the store buffer to the hold buffer,
+                * so the amount of data in the store buffer is
+                * the amount of data a read will return.
+                *
+                * If there's no data in either buffer, or there's
+                * no data in the hold buffer and the timer hasn't
+                * expired, we're not ready to read.
+                */
+               data = ((d->bd_hlen == 0 || d->bd_hbuf_read != 0) &&
+                   d->bd_state == BPF_TIMED_OUT ? d->bd_slen : d->bd_hlen);
+               ready = (data > 0);
+       }
+       if (!ready) {
+               bpf_start_timer(d);
+       } else if (kev) {
+               knote_fill_kevent(kn, kev, data);
+       }
+
+       return ready;
+}
+
+int
+bpfkqfilter(dev_t dev, struct knote *kn)
+{
+       struct bpf_d *d;
+       int res;
+
+       /*
+        * Is this device a bpf?
+        */
+       if (major(dev) != CDEV_MAJOR || kn->kn_filter != EVFILT_READ) {
+               knote_set_error(kn, EINVAL);
+               return 0;
+       }
+
+       lck_mtx_lock(bpf_mlock);
+
+       d = bpf_dtab[minor(dev)];
+
+       if (d == NULL || d == BPF_DEV_RESERVED ||
+           (d->bd_flags & BPF_CLOSING) != 0 ||
+           d->bd_bif == NULL) {
+               lck_mtx_unlock(bpf_mlock);
+               knote_set_error(kn, ENXIO);
+               return 0;
+       }
+
+       kn->kn_hook = d;
+       kn->kn_filtid = EVFILTID_BPFREAD;
+       KNOTE_ATTACH(&d->bd_sel.si_note, kn);
+       d->bd_flags |= BPF_KNOTE;
+
+       /* capture the current state */
+       res = filt_bpfread_common(kn, NULL, d);
+
+       lck_mtx_unlock(bpf_mlock);
+
+       return res;
+}
+
+static void
+filt_bpfdetach(struct knote *kn)
+{
+       struct bpf_d *d = (struct bpf_d *)kn->kn_hook;
+
+       lck_mtx_lock(bpf_mlock);
+       if (d->bd_flags & BPF_KNOTE) {
+               KNOTE_DETACH(&d->bd_sel.si_note, kn);
+               d->bd_flags &= ~BPF_KNOTE;
+       }
+       lck_mtx_unlock(bpf_mlock);
+}
+
+static int
+filt_bpfread(struct knote *kn, long hint)
+{
+#pragma unused(hint)
+       struct bpf_d *d = (struct bpf_d *)kn->kn_hook;
+
+       return filt_bpfread_common(kn, NULL, d);
+}
+
+static int
+filt_bpftouch(struct knote *kn, struct kevent_qos_s *kev)
+{
+       struct bpf_d *d = (struct bpf_d *)kn->kn_hook;
+       int res;
+
+       lck_mtx_lock(bpf_mlock);
+
+       /* save off the lowat threshold and flag */
+       kn->kn_sdata = kev->data;
+       kn->kn_sfflags = kev->fflags;
+
+       /* output data will be re-generated here */
+       res = filt_bpfread_common(kn, NULL, d);
+
+       lck_mtx_unlock(bpf_mlock);
+
+       return res;
+}
+
+static int
+filt_bpfprocess(struct knote *kn, struct kevent_qos_s *kev)
+{
+       struct bpf_d *d = (struct bpf_d *)kn->kn_hook;
+       int res;
+
+       lck_mtx_lock(bpf_mlock);
+       res = filt_bpfread_common(kn, kev, d);
+       lck_mtx_unlock(bpf_mlock);
+
+       return res;
+}
+
+/*
+ * Copy data from an mbuf chain into a buffer. This code is derived
+ * from m_copydata in kern/uipc_mbuf.c.
+ */
+static void
+bpf_mcopy(struct mbuf * m, void *dst_arg, size_t len)
+{
+       u_int count;
+       u_char *dst;
+
+       dst = dst_arg;
+       while (len > 0) {
+               if (m == 0) {
+                       panic("bpf_mcopy");
+               }
+               count = min(m->m_len, len);
+               bcopy(mbuf_data(m), dst, count);
+               m = m->m_next;
+               dst += count;
+               len -= count;
+       }
+}
+
+static inline void
+bpf_tap_imp(
+       ifnet_t         ifp,
+       u_int32_t       dlt,
+       struct bpf_packet *bpf_pkt,
+       int             outbound)
+{
+       struct bpf_d    *d;
+       u_int slen;
+       struct bpf_if *bp;
+
+       /*
+        * It's possible that we get here after the bpf descriptor has been
+        * detached from the interface; in such a case we simply return.
+        * Lock ordering is important since we can be called asynchronously
+        * (from IOKit) to process an inbound packet; when that happens
+        * we would have been holding its "gateLock" and will be acquiring
+        * "bpf_mlock" upon entering this routine.  Due to that, we release
+        * "bpf_mlock" prior to calling ifnet_set_promiscuous (which will
+        * acquire "gateLock" in the IOKit), in order to avoid a deadlock
+        * when a ifnet_set_promiscuous request simultaneously collides with
+        * an inbound packet being passed into the tap callback.
+        */
+       lck_mtx_lock(bpf_mlock);
+       if (ifp->if_bpf == NULL) {
+               lck_mtx_unlock(bpf_mlock);
+               return;
+       }
+       for (bp = ifp->if_bpf; bp != NULL; bp = bp->bif_next) {
+               if (bp->bif_ifp != ifp) {
+                       /* wrong interface */
+                       bp = NULL;
+                       break;
+               }
+               if (dlt == 0 || bp->bif_dlt == dlt) {
+                       /* tapping default DLT or DLT matches */
+                       break;
+               }
+       }
+       if (bp == NULL) {
+               goto done;
+       }
+       for (d = bp->bif_dlist; d; d = d->bd_next) {
+               struct bpf_packet *bpf_pkt_saved = bpf_pkt;
+               struct bpf_packet bpf_pkt_tmp;
+               struct pktap_header_buffer bpfp_header_tmp;
+
+               if (outbound && !d->bd_seesent) {
+                       continue;
+               }
+
+               ++d->bd_rcount;
+               slen = bpf_filter(d->bd_filter, (u_char *)bpf_pkt,
+                   bpf_pkt->bpfp_total_length, 0);
+               if (bp->bif_ifp->if_type == IFT_PKTAP &&
+                   bp->bif_dlt == DLT_PKTAP) {
+                       /*
+                        * Need to copy the bpf_pkt because the conversion
+                        * to v2 pktap header modifies the content of the
+                        * bpfp_header
+                        */
+                       if ((d->bd_flags & BPF_PKTHDRV2) &&
+                           bpf_pkt->bpfp_header_length <= sizeof(bpfp_header_tmp)) {
+                               bpf_pkt_tmp = *bpf_pkt;
+
+                               bpf_pkt = &bpf_pkt_tmp;
+
+                               memcpy(&bpfp_header_tmp, bpf_pkt->bpfp_header,
+                                   bpf_pkt->bpfp_header_length);
+
+                               bpf_pkt->bpfp_header = &bpfp_header_tmp;
+
+                               convert_to_pktap_header_to_v2(bpf_pkt,
+                                   !!(d->bd_flags & BPF_TRUNCATE));
+                       }
+
+                       if (d->bd_flags & BPF_TRUNCATE) {
+                               slen = min(slen,
+                                   get_pkt_trunc_len((u_char *)bpf_pkt,
+                                   bpf_pkt->bpfp_total_length));
+                       }
+               }
+               if (slen != 0) {
+                       catchpacket(d, bpf_pkt, slen, outbound);
+               }
+               bpf_pkt = bpf_pkt_saved;
+       }
+
+done:
+       lck_mtx_unlock(bpf_mlock);
+}
+
+static inline void
+bpf_tap_mbuf(
+       ifnet_t         ifp,
+       u_int32_t       dlt,
+       mbuf_t          m,
+       void*           hdr,
+       size_t          hlen,
+       int             outbound)
+{
+       struct bpf_packet bpf_pkt;
+       struct mbuf *m0;
+
+       if (ifp->if_bpf == NULL) {
+               /* quickly check without taking lock */
+               return;
+       }
+       bpf_pkt.bpfp_type = BPF_PACKET_TYPE_MBUF;
+       bpf_pkt.bpfp_mbuf = m;
+       bpf_pkt.bpfp_total_length = 0;
+       for (m0 = m; m0 != NULL; m0 = m0->m_next) {
+               bpf_pkt.bpfp_total_length += m0->m_len;
+       }
+       bpf_pkt.bpfp_header = hdr;
+       if (hdr != NULL) {
+               bpf_pkt.bpfp_total_length += hlen;
+               bpf_pkt.bpfp_header_length = hlen;
+       } else {
+               bpf_pkt.bpfp_header_length = 0;
+       }
+       bpf_tap_imp(ifp, dlt, &bpf_pkt, outbound);
+}
+
+void
+bpf_tap_out(
+       ifnet_t         ifp,
+       u_int32_t       dlt,
+       mbuf_t          m,
+       void*           hdr,
+       size_t          hlen)
+{
+       bpf_tap_mbuf(ifp, dlt, m, hdr, hlen, 1);
+}
+
+void
+bpf_tap_in(
+       ifnet_t         ifp,
+       u_int32_t       dlt,
+       mbuf_t          m,
+       void*           hdr,
+       size_t          hlen)
+{
+       bpf_tap_mbuf(ifp, dlt, m, hdr, hlen, 0);
+}
+
+/* Callback registered with Ethernet driver. */
+static int
+bpf_tap_callback(struct ifnet *ifp, struct mbuf *m)
+{
+       bpf_tap_mbuf(ifp, 0, m, NULL, 0, mbuf_pkthdr_rcvif(m) == NULL);
+
+       return 0;
+}
+
+
+static errno_t
+bpf_copydata(struct bpf_packet *pkt, size_t off, size_t len, void* out_data)
+{
+       errno_t err = 0;
+       if (pkt->bpfp_type == BPF_PACKET_TYPE_MBUF) {
+               err = mbuf_copydata(pkt->bpfp_mbuf, off, len, out_data);
+       } else {
+               err = EINVAL;
+       }
+
+       return err;
+}
+
+static void
+copy_bpf_packet(struct bpf_packet * pkt, void * dst, size_t len)
+{
+       /* copy the optional header */
+       if (pkt->bpfp_header_length != 0) {
+               size_t  count = min(len, pkt->bpfp_header_length);
+               bcopy(pkt->bpfp_header, dst, count);
+               len -= count;
+               dst += count;
+       }
+       if (len == 0) {
+               /* nothing past the header */
+               return;
+       }
+       /* copy the packet */
+       switch (pkt->bpfp_type) {
+       case BPF_PACKET_TYPE_MBUF:
+               bpf_mcopy(pkt->bpfp_mbuf, dst, len);
+               break;
+       default:
+               break;
+       }
+}
+
+static uint16_t
+get_esp_trunc_len(__unused struct bpf_packet *pkt, __unused uint16_t off,
+    const uint16_t remaining_caplen)
+{
+       /*
+        * For some reason tcpdump expects to have one byte beyond the ESP header
+        */
+       uint16_t trunc_len = ESP_HDR_SIZE + 1;
+
+       if (trunc_len > remaining_caplen) {
+               return remaining_caplen;
+       }
+
+       return trunc_len;
+}
+
+static uint16_t
+get_isakmp_trunc_len(__unused struct bpf_packet *pkt, __unused uint16_t off,
+    const uint16_t remaining_caplen)
+{
+       /*
+        * Include the payload generic header
+        */
+       uint16_t trunc_len = ISAKMP_HDR_SIZE;
+
+       if (trunc_len > remaining_caplen) {
+               return remaining_caplen;
+       }
+
+       return trunc_len;
+}
+
+static uint16_t
+get_isakmp_natt_trunc_len(struct bpf_packet *pkt, uint16_t off,
+    const uint16_t remaining_caplen)
+{
+       int err = 0;
+       uint16_t trunc_len = 0;
+       char payload[remaining_caplen];
+
+       err = bpf_copydata(pkt, off, remaining_caplen, payload);
+       if (err != 0) {
+               return remaining_caplen;
+       }
+       /*
+        * They are three cases:
+        * - IKE: payload start with 4 bytes header set to zero before ISAKMP header
+        * - keep alive: 1 byte payload
+        * - otherwise it's ESP
+        */
+       if (remaining_caplen >= 4 &&
+           payload[0] == 0 && payload[1] == 0 &&
+           payload[2] == 0 && payload[3] == 0) {
+               trunc_len = 4 + get_isakmp_trunc_len(pkt, off + 4, remaining_caplen - 4);
+       } else if (remaining_caplen == 1) {
+               trunc_len = 1;
+       } else {
+               trunc_len = get_esp_trunc_len(pkt, off, remaining_caplen);
+       }
+
+       if (trunc_len > remaining_caplen) {
+               return remaining_caplen;
+       }
+
+       return trunc_len;
+}
+
+static uint16_t
+get_udp_trunc_len(struct bpf_packet *pkt, uint16_t off, const uint16_t remaining_caplen)
+{
+       int err = 0;
+       uint16_t trunc_len = sizeof(struct udphdr); /* By default no UDP payload */
+
+       if (trunc_len >= remaining_caplen) {
+               return remaining_caplen;
+       }
+
+       struct udphdr udphdr;
+       err = bpf_copydata(pkt, off, sizeof(struct udphdr), &udphdr);
+       if (err != 0) {
+               return remaining_caplen;
+       }
+
+       u_short sport, dport;
+
+       sport = EXTRACT_SHORT(&udphdr.uh_sport);
+       dport = EXTRACT_SHORT(&udphdr.uh_dport);
+
+       if (dport == PORT_DNS || sport == PORT_DNS) {
+               /*
+                * Full UDP payload for DNS
+                */
+               trunc_len = remaining_caplen;
+       } else if ((sport == PORT_BOOTPS && dport == PORT_BOOTPC) ||
+           (sport == PORT_BOOTPC && dport == PORT_BOOTPS)) {
+               /*
+                * Full UDP payload for BOOTP and DHCP
+                */
+               trunc_len = remaining_caplen;
+       } else if (dport == PORT_ISAKMP && sport == PORT_ISAKMP) {
+               /*
+                * Return the ISAKMP header
+                */
+               trunc_len += get_isakmp_trunc_len(pkt, off + sizeof(struct udphdr),
+                   remaining_caplen - sizeof(struct udphdr));
+       } else if (dport == PORT_ISAKMP_NATT && sport == PORT_ISAKMP_NATT) {
+               trunc_len += get_isakmp_natt_trunc_len(pkt, off + sizeof(struct udphdr),
+                   remaining_caplen - sizeof(struct udphdr));
+       }
+       if (trunc_len >= remaining_caplen) {
+               return remaining_caplen;
+       }
+
+       return trunc_len;
+}
+
+static uint16_t
+get_tcp_trunc_len(struct bpf_packet *pkt, uint16_t off, const uint16_t remaining_caplen)
+{
+       int err = 0;
+       uint16_t trunc_len = sizeof(struct tcphdr); /* By default no TCP payload */
+       if (trunc_len >= remaining_caplen) {
+               return remaining_caplen;
+       }
+
+       struct tcphdr tcphdr;
+       err = bpf_copydata(pkt, off, sizeof(struct tcphdr), &tcphdr);
+       if (err != 0) {
+               return remaining_caplen;
+       }
+
+       u_short sport, dport;
+       sport = EXTRACT_SHORT(&tcphdr.th_sport);
+       dport = EXTRACT_SHORT(&tcphdr.th_dport);
+
+       if (dport == PORT_DNS || sport == PORT_DNS) {
+               /*
+                * Full TCP payload  for DNS
+                */
+               trunc_len = remaining_caplen;
+       } else {
+               trunc_len = tcphdr.th_off << 2;
+       }
+       if (trunc_len >= remaining_caplen) {
+               return remaining_caplen;
+       }
+
+       return trunc_len;
+}
+
+static uint16_t
+get_proto_trunc_len(uint8_t proto, struct bpf_packet *pkt, uint16_t off, const uint16_t remaining_caplen)
+{
+       uint16_t trunc_len;
+
+       switch (proto) {
+       case IPPROTO_ICMP: {
+               /*
+                * Full IMCP payload
+                */
+               trunc_len = remaining_caplen;
+               break;
+       }
+       case IPPROTO_ICMPV6: {
+               /*
+                * Full IMCPV6 payload
+                */
+               trunc_len = remaining_caplen;
+               break;
+       }
+       case IPPROTO_IGMP: {
+               /*
+                * Full IGMP payload
+                */
+               trunc_len = remaining_caplen;
+               break;
+       }
+       case IPPROTO_UDP: {
+               trunc_len = get_udp_trunc_len(pkt, off, remaining_caplen);
+               break;
+       }
+       case IPPROTO_TCP: {
+               trunc_len = get_tcp_trunc_len(pkt, off, remaining_caplen);
+               break;
+       }
+       case IPPROTO_ESP: {
+               trunc_len = get_esp_trunc_len(pkt, off, remaining_caplen);
                break;
        }
                break;
        }
-       
-       lck_mtx_unlock(bpf_mlock);
-       
-       return (error);
-}
-
-/*
- * Set d's packet filter program to fp.  If this file already has a filter,
- * free it and replace it.  Returns EINVAL for bogus requests.
- */
-static int
-bpf_setf(struct bpf_d *d, struct user_bpf_program *fp)
-{
-       struct bpf_insn *fcode, *old;
-       u_int flen, size;
-       int s;
-
-       old = d->bd_filter;
-       if (fp->bf_insns == USER_ADDR_NULL) {
-               if (fp->bf_len != 0)
-                       return (EINVAL);
-               s = splimp();
-               d->bd_filter = 0;
-               reset_d(d);
-               splx(s);
-               if (old != 0)
-                       FREE((caddr_t)old, M_DEVBUF);
-               return (0);
+       default: {
+               /*
+                * By default we only include the IP header
+                */
+               trunc_len = 0;
+               break;
        }
        }
-       flen = fp->bf_len;
-       if (flen > BPF_MAXINSNS)
-               return (EINVAL);
-
-       size = flen * sizeof(struct bpf_insn);
-       fcode = (struct bpf_insn *) _MALLOC(size, M_DEVBUF, M_WAIT);
-#ifdef __APPLE__
-       if (fcode == NULL)
-               return (ENOBUFS);
-#endif
-       if (copyin(fp->bf_insns, (caddr_t)fcode, size) == 0 &&
-           bpf_validate(fcode, (int)flen)) {
-               s = splimp();
-               d->bd_filter = fcode;
-               reset_d(d);
-               splx(s);
-               if (old != 0)
-                       FREE((caddr_t)old, M_DEVBUF);
-
-               return (0);
        }
        }
-       FREE((caddr_t)fcode, M_DEVBUF);
-       return (EINVAL);
+       if (trunc_len >= remaining_caplen) {
+               return remaining_caplen;
+       }
+
+       return trunc_len;
 }
 
 }
 
-/*
- * Detach a file from its current interface (if attached at all) and attach
- * to the interface indicated by the name stored in ifr.
- * Return an errno or 0.
- */
-static int
-bpf_setif(struct bpf_d *d, struct ifreq *ifr)
+static uint16_t
+get_ip_trunc_len(struct bpf_packet *pkt, uint16_t off, const uint16_t remaining_caplen)
 {
 {
-       struct bpf_if *bp;
-       int s, error;
-       struct ifnet *theywant;
+       int err = 0;
+       uint16_t iplen = sizeof(struct ip);
+       if (iplen >= remaining_caplen) {
+               return remaining_caplen;
+       }
 
 
-       theywant = ifunit(ifr->ifr_name);
-       if (theywant == 0)
-               return ENXIO;
+       struct ip iphdr;
+       err =  bpf_copydata(pkt, off, sizeof(struct ip), &iphdr);
+       if (err != 0) {
+               return remaining_caplen;
+       }
 
 
-       /*
-        * Look through attached interfaces for the named one.
-        */
-       for (bp = bpf_iflist; bp != 0; bp = bp->bif_next) {
-               struct ifnet *ifp = bp->bif_ifp;
+       uint8_t proto = 0;
 
 
-               if (ifp == 0 || ifp != theywant)
-                       continue;
-               /*
-                * We found the requested interface.
-                * If it's not up, return an error.
-                * Allocate the packet buffers if we need to.
-                * If we're already attached to requested interface,
-                * just flush the buffer.
-                */
-               if ((ifp->if_flags & IFF_UP) == 0)
-                       return (ENETDOWN);
+       iplen = iphdr.ip_hl << 2;
+       if (iplen >= remaining_caplen) {
+               return remaining_caplen;
+       }
 
 
-               if (d->bd_sbuf == 0) {
-                       error = bpf_allocbufs(d);
-                       if (error != 0)
-                               return (error);
-               }
-               s = splimp();
-               if (bp != d->bd_bif) {
-                       if (d->bd_bif)
-                               /*
-                                * Detach if attached to something else.
-                                */
-                               bpf_detachd(d);
+       proto = iphdr.ip_p;
+       iplen += get_proto_trunc_len(proto, pkt, off + iplen, remaining_caplen - iplen);
 
 
-                       bpf_attachd(d, bp);
-               }
-               reset_d(d);
-               splx(s);
-               return (0);
+       if (iplen >= remaining_caplen) {
+               return remaining_caplen;
        }
        }
-       /* Not found. */
-       return (ENXIO);
+
+       return iplen;
 }
 
 }
 
-/*
- * Support for select() and poll() system calls
- *
- * Return true iff the specific operation will not block indefinitely.
- * Otherwise, return false but make a note that a selwakeup() must be done.
- */
-int
-bpfpoll(dev_t dev, int events, void * wql, struct proc *p)
+static uint16_t
+get_ip6_trunc_len(struct bpf_packet *pkt, uint16_t off, const uint16_t remaining_caplen)
 {
 {
-       register struct bpf_d *d;
-       register int s;
-       int revents = 0;
+       int err = 0;
+       uint16_t iplen = sizeof(struct ip6_hdr);
+       if (iplen >= remaining_caplen) {
+               return remaining_caplen;
+       }
 
 
-       d = bpf_dtab[minor(dev)];
-       if (d == 0 || d == (void *)1)
-               return (ENXIO);
+       struct ip6_hdr ip6hdr;
+       err = bpf_copydata(pkt, off, sizeof(struct ip6_hdr), &ip6hdr);
+       if (err != 0) {
+               return remaining_caplen;
+       }
 
 
-       lck_mtx_lock(bpf_mlock);
+       uint8_t proto = 0;
 
        /*
 
        /*
-        * An imitation of the FIONREAD ioctl code.
+        * TBD: process the extension headers
         */
         */
-       if (d->bd_bif == NULL) {
-               lck_mtx_unlock(bpf_mlock);
-               return (ENXIO);
-       }
+       proto = ip6hdr.ip6_nxt;
+       iplen += get_proto_trunc_len(proto, pkt, off + iplen, remaining_caplen - iplen);
 
 
-       s = splimp();
-       if (events & (POLLIN | POLLRDNORM)) {
-               if (d->bd_hlen != 0 || (d->bd_immediate && d->bd_slen != 0))
-                       revents |= events & (POLLIN | POLLRDNORM);
-               else
-                       selrecord(p, &d->bd_sel, wql);
+       if (iplen >= remaining_caplen) {
+               return remaining_caplen;
        }
        }
-       splx(s);
 
 
-       lck_mtx_unlock(bpf_mlock);
-       return (revents);
+       return iplen;
 }
 
 }
 
-/*
- * Incoming linkage from device drivers.  Process the packet pkt, of length
- * pktlen, which is stored in a contiguous buffer.  The packet is parsed
- * by each process' filter, and if accepted, stashed into the corresponding
- * buffer.
- */
-void
-bpf_tap(struct ifnet *ifp, u_char *pkt, u_int pktlen)
+static uint16_t
+get_ether_trunc_len(struct bpf_packet *pkt, int off, const uint16_t remaining_caplen)
 {
 {
-       struct bpf_if *bp;
-       register struct bpf_d *d;
-       register u_int slen;
-       /*
-        * Note that the ipl does not have to be raised at this point.
-        * The only problem that could arise here is that if two different
-        * interfaces shared any data.  This is not the case.
-        */
-       lck_mtx_lock(bpf_mlock);
-       
-       bp = ifp->if_bpf;
-#ifdef __APPLE__
-       if (bp) {
-#endif
-               for (d = bp->bif_dlist; d != 0; d = d->bd_next) {
-                       ++d->bd_rcount;
-                       slen = bpf_filter(d->bd_filter, pkt, pktlen, pktlen);
-                       if (slen != 0)
-                               catchpacket(d, pkt, pktlen, slen, bcopy);
-               }
-#ifdef __APPLE__
-    }
-       lck_mtx_unlock(bpf_mlock);
-#endif
-}
+       int err = 0;
+       uint16_t ethlen = sizeof(struct ether_header);
+       if (ethlen >= remaining_caplen) {
+               return remaining_caplen;
+       }
 
 
-/*
- * Copy data from an mbuf chain into a buffer.  This code is derived
- * from m_copydata in sys/uipc_mbuf.c.
- */
-static void
-bpf_mcopy(const void *src_arg, void *dst_arg, size_t len)
-{
-       const struct mbuf *m;
-       u_int count;
-       u_char *dst;
+       struct ether_header eh;
+       u_short type;
+       err = bpf_copydata(pkt, off, sizeof(struct ether_header), &eh);
+       if (err != 0) {
+               return remaining_caplen;
+       }
 
 
-       m = src_arg;
-       dst = dst_arg;
-       while (len > 0) {
-               if (m == 0)
-                       panic("bpf_mcopy");
-               count = min(m->m_len, len);
-               bcopy(mtod(m, const void *), dst, count);
-               m = m->m_next;
-               dst += count;
-               len -= count;
+       type = EXTRACT_SHORT(&eh.ether_type);
+       /* Include full ARP */
+       if (type == ETHERTYPE_ARP) {
+               ethlen = remaining_caplen;
+       } else if (type != ETHERTYPE_IP && type != ETHERTYPE_IPV6) {
+               ethlen = min(BPF_MIN_PKT_SIZE, remaining_caplen);
+       } else {
+               if (type == ETHERTYPE_IP) {
+                       ethlen += get_ip_trunc_len(pkt, sizeof(struct ether_header),
+                           remaining_caplen);
+               } else if (type == ETHERTYPE_IPV6) {
+                       ethlen += get_ip6_trunc_len(pkt, sizeof(struct ether_header),
+                           remaining_caplen);
+               }
        }
        }
+       return ethlen;
 }
 
 }
 
-/*
- * Incoming linkage from device drivers, when packet is in an mbuf chain.
- */
-void
-bpf_mtap(struct ifnet *ifp, struct mbuf *m)
+static uint32_t
+get_pkt_trunc_len(u_char *p, u_int len)
 {
 {
-       struct bpf_if *bp;
-       struct bpf_d *d;
-       u_int pktlen, slen;
-       struct mbuf *m0;
+       struct bpf_packet *pkt = (struct bpf_packet *)(void *) p;
+       struct pktap_header *pktap = (struct pktap_header *) (pkt->bpfp_header);
+       uint32_t out_pkt_len = 0, tlen = 0;
+       /*
+        * pktap->pth_frame_pre_length is L2 header length and accounts
+        * for both pre and pre_adjust.
+        * pktap->pth_length is sizeof(pktap_header) (excl the pre/pre_adjust)
+        * pkt->bpfp_header_length is (pktap->pth_length + pre_adjust)
+        * pre is the offset to the L3 header after the bpfp_header, or length
+        * of L2 header after bpfp_header, if present.
+        */
+       int32_t pre = pktap->pth_frame_pre_length -
+           (pkt->bpfp_header_length - pktap->pth_length);
+
+       /* Length of the input packet starting from  L3 header */
+       uint32_t in_pkt_len = len - pkt->bpfp_header_length - pre;
+       if (pktap->pth_protocol_family == AF_INET ||
+           pktap->pth_protocol_family == AF_INET6) {
+               /* Contains L2 header */
+               if (pre > 0) {
+                       if (pre < (int32_t)sizeof(struct ether_header)) {
+                               goto too_short;
+                       }
 
 
-       lck_mtx_lock(bpf_mlock);
-       
-       bp = ifp->if_bpf;
-       if (bp) {
-       pktlen = 0;
-       for (m0 = m; m0 != 0; m0 = m0->m_next)
-               pktlen += m0->m_len;
-       
-               for (d = bp->bif_dlist; d != 0; d = d->bd_next) {
-                       if (!d->bd_seesent && (m->m_pkthdr.rcvif == NULL))
-                               continue;
-                       ++d->bd_rcount;
-                       slen = bpf_filter(d->bd_filter, (u_char *)m, pktlen, 0);
-                       if (slen != 0)
-                               catchpacket(d, (u_char *)m, pktlen, slen, bpf_mcopy);
+                       out_pkt_len = get_ether_trunc_len(pkt, 0, in_pkt_len);
+               } else if (pre == 0) {
+                       if (pktap->pth_protocol_family == AF_INET) {
+                               out_pkt_len = get_ip_trunc_len(pkt, pre, in_pkt_len);
+                       } else if (pktap->pth_protocol_family == AF_INET6) {
+                               out_pkt_len = get_ip6_trunc_len(pkt, pre, in_pkt_len);
+                       }
+               } else {
+                       /* Ideally pre should be >= 0. This is an exception */
+                       out_pkt_len = min(BPF_MIN_PKT_SIZE, in_pkt_len);
+               }
+       } else {
+               if (pktap->pth_iftype == IFT_ETHER) {
+                       if (in_pkt_len < sizeof(struct ether_header)) {
+                               goto too_short;
+                       }
+                       /* At most include the Ethernet header and 16 bytes */
+                       out_pkt_len = MIN(sizeof(struct ether_header) + 16,
+                           in_pkt_len);
+               } else {
+                       /*
+                        * For unknown protocols include at most 16 bytes
+                        */
+                       out_pkt_len = MIN(16, in_pkt_len);
                }
        }
                }
        }
-       
-       lck_mtx_unlock(bpf_mlock);
+done:
+       tlen = pkt->bpfp_header_length + out_pkt_len + pre;
+       return tlen;
+too_short:
+       out_pkt_len = in_pkt_len;
+       goto done;
 }
 
 /*
  * Move the packet data from interface memory (pkt) into the
  * store buffer.  Return 1 if it's time to wakeup a listener (buffer full),
 }
 
 /*
  * Move the packet data from interface memory (pkt) into the
  * store buffer.  Return 1 if it's time to wakeup a listener (buffer full),
- * otherwise 0.  "copy" is the routine called to do the actual data
- * transfer.  bcopy is passed in to copy contiguous chunks, while
- * bpf_mcopy is passed in to copy mbuf chains.  In the latter case,
- * pkt is really an mbuf.
+ * otherwise 0.
  */
 static void
  */
 static void
-catchpacket(struct bpf_d *d, u_char *pkt, u_int pktlen, u_int snaplen, 
-       void (*cpfn)(const void *, void *, size_t))
+catchpacket(struct bpf_d *d, struct bpf_packet * pkt,
+    u_int snaplen, int outbound)
 {
 {
-       register struct bpf_hdr *hp;
-       register int totlen, curlen;
-       register int hdrlen = d->bd_bif->bif_hdrlen;
+       struct bpf_hdr *hp;
+       struct bpf_hdr_ext *ehp;
+       int totlen, curlen;
+       int hdrlen, caplen;
+       int do_wakeup = 0;
+       u_char *payload;
+       struct timeval tv;
+
+       hdrlen = (d->bd_flags & BPF_EXTENDED_HDR) ? d->bd_bif->bif_exthdrlen :
+           d->bd_bif->bif_hdrlen;
        /*
         * Figure out how many bytes to move.  If the packet is
         * greater or equal to the snapshot length, transfer that
         * much.  Otherwise, transfer the whole packet (unless
         * we hit the buffer size limit).
         */
        /*
         * Figure out how many bytes to move.  If the packet is
         * greater or equal to the snapshot length, transfer that
         * much.  Otherwise, transfer the whole packet (unless
         * we hit the buffer size limit).
         */
-       totlen = hdrlen + min(snaplen, pktlen);
-       if (totlen > d->bd_bufsize)
+       totlen = hdrlen + min(snaplen, pkt->bpfp_total_length);
+       if (totlen > d->bd_bufsize) {
                totlen = d->bd_bufsize;
                totlen = d->bd_bufsize;
+       }
+
+       if (hdrlen > totlen) {
+               return;
+       }
 
        /*
         * Round up the end of the previous packet to the next longword.
 
        /*
         * Round up the end of the previous packet to the next longword.
@@ -1412,44 +3298,117 @@ catchpacket(struct bpf_d *d, u_char *pkt, u_int pktlen, u_int snaplen,
                 * This packet will overflow the storage buffer.
                 * Rotate the buffers if we can, then wakeup any
                 * pending reads.
                 * This packet will overflow the storage buffer.
                 * Rotate the buffers if we can, then wakeup any
                 * pending reads.
+                *
+                * We cannot rotate buffers if a read is in progress
+                * so drop the packet
                 */
                 */
-               if (d->bd_fbuf == 0) {
-                       /*
-                        * We haven't completed the previous read yet,
-                        * so drop the packet.
-                        */
+               if (d->bd_hbuf_read != 0) {
                        ++d->bd_dcount;
                        return;
                }
                        ++d->bd_dcount;
                        return;
                }
-               ROTATE_BUFFERS(d);
-               bpf_wakeup(d);
+
+               if (d->bd_fbuf == NULL) {
+                       if (d->bd_headdrop == 0) {
+                               /*
+                                * We haven't completed the previous read yet,
+                                * so drop the packet.
+                                */
+                               ++d->bd_dcount;
+                               return;
+                       }
+                       /*
+                        * Drop the hold buffer as it contains older packets
+                        */
+                       d->bd_dcount += d->bd_hcnt;
+                       d->bd_fbuf = d->bd_hbuf;
+                       ROTATE_BUFFERS(d);
+               } else {
+                       ROTATE_BUFFERS(d);
+               }
+               do_wakeup = 1;
                curlen = 0;
                curlen = 0;
-       }
-       else if (d->bd_immediate)
+       } else if (d->bd_immediate || d->bd_state == BPF_TIMED_OUT) {
                /*
                /*
-                * Immediate mode is set.  A packet arrived so any
-                * reads should be woken up.
+                * Immediate mode is set, or the read timeout has
+                * already expired during a select call. A packet
+                * arrived, so the reader should be woken up.
                 */
                 */
-               bpf_wakeup(d);
+               do_wakeup = 1;
+       }
 
        /*
         * Append the bpf header.
         */
 
        /*
         * Append the bpf header.
         */
-       hp = (struct bpf_hdr *)(d->bd_sbuf + curlen);
-#if BSD >= 199103
-       microtime(&hp->bh_tstamp);
-#elif defined(sun)
-       uniqtime(&hp->bh_tstamp);
-#else
-       hp->bh_tstamp = time;
-#endif
-       hp->bh_datalen = pktlen;
-       hp->bh_hdrlen = hdrlen;
+       microtime(&tv);
+       if (d->bd_flags & BPF_EXTENDED_HDR) {
+               struct mbuf *m;
+
+               m = (pkt->bpfp_type == BPF_PACKET_TYPE_MBUF)
+                   ? pkt->bpfp_mbuf : NULL;
+               ehp = (struct bpf_hdr_ext *)(void *)(d->bd_sbuf + curlen);
+               memset(ehp, 0, sizeof(*ehp));
+               ehp->bh_tstamp.tv_sec = tv.tv_sec;
+               ehp->bh_tstamp.tv_usec = tv.tv_usec;
+
+               ehp->bh_datalen = pkt->bpfp_total_length;
+               ehp->bh_hdrlen = hdrlen;
+               caplen = ehp->bh_caplen = totlen - hdrlen;
+               if (m == NULL) {
+                       if (outbound) {
+                               ehp->bh_flags |= BPF_HDR_EXT_FLAGS_DIR_OUT;
+                       } else {
+                               ehp->bh_flags |= BPF_HDR_EXT_FLAGS_DIR_IN;
+                       }
+               } else if (outbound) {
+                       ehp->bh_flags |= BPF_HDR_EXT_FLAGS_DIR_OUT;
+
+                       /* only do lookups on non-raw INPCB */
+                       if ((m->m_pkthdr.pkt_flags & (PKTF_FLOW_ID |
+                           PKTF_FLOW_LOCALSRC | PKTF_FLOW_RAWSOCK)) ==
+                           (PKTF_FLOW_ID | PKTF_FLOW_LOCALSRC) &&
+                           m->m_pkthdr.pkt_flowsrc == FLOWSRC_INPCB) {
+                               ehp->bh_flowid = m->m_pkthdr.pkt_flowid;
+                               ehp->bh_proto = m->m_pkthdr.pkt_proto;
+                       }
+                       ehp->bh_svc = so_svc2tc(m->m_pkthdr.pkt_svc);
+                       if (m->m_pkthdr.pkt_flags & PKTF_TCP_REXMT) {
+                               ehp->bh_pktflags |= BPF_PKTFLAGS_TCP_REXMT;
+                       }
+                       if (m->m_pkthdr.pkt_flags & PKTF_START_SEQ) {
+                               ehp->bh_pktflags |= BPF_PKTFLAGS_START_SEQ;
+                       }
+                       if (m->m_pkthdr.pkt_flags & PKTF_LAST_PKT) {
+                               ehp->bh_pktflags |= BPF_PKTFLAGS_LAST_PKT;
+                       }
+                       if (m->m_pkthdr.pkt_flags & PKTF_VALID_UNSENT_DATA) {
+                               ehp->bh_unsent_bytes =
+                                   m->m_pkthdr.bufstatus_if;
+                               ehp->bh_unsent_snd =
+                                   m->m_pkthdr.bufstatus_sndbuf;
+                       }
+               } else {
+                       ehp->bh_flags |= BPF_HDR_EXT_FLAGS_DIR_IN;
+               }
+               payload = (u_char *)ehp + hdrlen;
+       } else {
+               hp = (struct bpf_hdr *)(void *)(d->bd_sbuf + curlen);
+               hp->bh_tstamp.tv_sec = tv.tv_sec;
+               hp->bh_tstamp.tv_usec = tv.tv_usec;
+               hp->bh_datalen = pkt->bpfp_total_length;
+               hp->bh_hdrlen = hdrlen;
+               caplen = hp->bh_caplen = totlen - hdrlen;
+               payload = (u_char *)hp + hdrlen;
+       }
        /*
         * Copy the packet data into the store buffer and update its length.
         */
        /*
         * Copy the packet data into the store buffer and update its length.
         */
-       (*cpfn)(pkt, (u_char *)hp + hdrlen, (hp->bh_caplen = totlen - hdrlen));
+       copy_bpf_packet(pkt, payload, caplen);
        d->bd_slen = curlen + totlen;
        d->bd_slen = curlen + totlen;
+       d->bd_scnt += 1;
+
+       if (do_wakeup) {
+               bpf_wakeup(d);
+       }
 }
 
 /*
 }
 
 /*
@@ -1458,18 +3417,35 @@ catchpacket(struct bpf_d *d, u_char *pkt, u_int pktlen, u_int snaplen,
 static int
 bpf_allocbufs(struct bpf_d *d)
 {
 static int
 bpf_allocbufs(struct bpf_d *d)
 {
+       if (d->bd_sbuf != NULL) {
+               FREE(d->bd_sbuf, M_DEVBUF);
+               d->bd_sbuf = NULL;
+       }
+       if (d->bd_hbuf != NULL) {
+               FREE(d->bd_hbuf, M_DEVBUF);
+               d->bd_hbuf = NULL;
+       }
+       if (d->bd_fbuf != NULL) {
+               FREE(d->bd_fbuf, M_DEVBUF);
+               d->bd_fbuf = NULL;
+       }
+
        d->bd_fbuf = (caddr_t) _MALLOC(d->bd_bufsize, M_DEVBUF, M_WAIT);
        d->bd_fbuf = (caddr_t) _MALLOC(d->bd_bufsize, M_DEVBUF, M_WAIT);
-       if (d->bd_fbuf == 0)
-               return (ENOBUFS);
+       if (d->bd_fbuf == NULL) {
+               return ENOBUFS;
+       }
 
        d->bd_sbuf = (caddr_t) _MALLOC(d->bd_bufsize, M_DEVBUF, M_WAIT);
 
        d->bd_sbuf = (caddr_t) _MALLOC(d->bd_bufsize, M_DEVBUF, M_WAIT);
-       if (d->bd_sbuf == 0) {
+       if (d->bd_sbuf == NULL) {
                FREE(d->bd_fbuf, M_DEVBUF);
                FREE(d->bd_fbuf, M_DEVBUF);
-               return (ENOBUFS);
+               d->bd_fbuf = NULL;
+               return ENOBUFS;
        }
        d->bd_slen = 0;
        d->bd_hlen = 0;
        }
        d->bd_slen = 0;
        d->bd_hlen = 0;
-       return (0);
+       d->bd_scnt = 0;
+       d->bd_hcnt = 0;
+       return 0;
 }
 
 /*
 }
 
 /*
@@ -1484,40 +3460,116 @@ bpf_freed(struct bpf_d *d)
         * been detached from its interface and it yet hasn't been marked
         * free.
         */
         * been detached from its interface and it yet hasn't been marked
         * free.
         */
+       if (d->bd_hbuf_read != 0) {
+               panic("bpf buffer freed during read");
+       }
+
        if (d->bd_sbuf != 0) {
                FREE(d->bd_sbuf, M_DEVBUF);
        if (d->bd_sbuf != 0) {
                FREE(d->bd_sbuf, M_DEVBUF);
-               if (d->bd_hbuf != 0)
+               if (d->bd_hbuf != 0) {
                        FREE(d->bd_hbuf, M_DEVBUF);
                        FREE(d->bd_hbuf, M_DEVBUF);
-               if (d->bd_fbuf != 0)
+               }
+               if (d->bd_fbuf != 0) {
                        FREE(d->bd_fbuf, M_DEVBUF);
                        FREE(d->bd_fbuf, M_DEVBUF);
+               }
+       }
+       if (d->bd_filter) {
+               FREE(d->bd_filter, M_DEVBUF);
        }
        }
-       if (d->bd_filter)
-               FREE((caddr_t)d->bd_filter, M_DEVBUF);
 }
 
 /*
 }
 
 /*
- * Attach an interface to bpf.  driverp is a pointer to a (struct bpf_if *)
+ * Attach an interface to bpf. driverp is a pointer to a (struct bpf_if *)
  * in the driver's softc; dlt is the link layer type; hdrlen is the fixed
  * size of the link header (variable length headers not yet supported).
  */
 void
 bpfattach(struct ifnet *ifp, u_int dlt, u_int hdrlen)
  * in the driver's softc; dlt is the link layer type; hdrlen is the fixed
  * size of the link header (variable length headers not yet supported).
  */
 void
 bpfattach(struct ifnet *ifp, u_int dlt, u_int hdrlen)
+{
+       bpf_attach(ifp, dlt, hdrlen, NULL, NULL);
+}
+
+errno_t
+bpf_attach(
+       ifnet_t                 ifp,
+       u_int32_t               dlt,
+       u_int32_t               hdrlen,
+       bpf_send_func   send,
+       bpf_tap_func    tap)
 {
        struct bpf_if *bp;
 {
        struct bpf_if *bp;
-       bp = (struct bpf_if *) _MALLOC(sizeof(*bp), M_DEVBUF, M_WAIT);
-       if (bp == 0)
+       struct bpf_if *bp_new;
+       struct bpf_if *bp_before_first = NULL;
+       struct bpf_if *bp_first = NULL;
+       struct bpf_if *bp_last = NULL;
+       boolean_t found;
+
+       bp_new = (struct bpf_if *) _MALLOC(sizeof(*bp_new), M_DEVBUF,
+           M_WAIT | M_ZERO);
+       if (bp_new == 0) {
                panic("bpfattach");
                panic("bpfattach");
+       }
 
        lck_mtx_lock(bpf_mlock);
 
 
        lck_mtx_lock(bpf_mlock);
 
-       bp->bif_dlist = 0;
-       bp->bif_ifp = ifp;
-       bp->bif_dlt = dlt;
+       /*
+        * Check if this interface/dlt is already attached. Remember the
+        * first and last attachment for this interface, as well as the
+        * element before the first attachment.
+        */
+       found = FALSE;
+       for (bp = bpf_iflist; bp != NULL; bp = bp->bif_next) {
+               if (bp->bif_ifp != ifp) {
+                       if (bp_first != NULL) {
+                               /* no more elements for this interface */
+                               break;
+                       }
+                       bp_before_first = bp;
+               } else {
+                       if (bp->bif_dlt == dlt) {
+                               found = TRUE;
+                               break;
+                       }
+                       if (bp_first == NULL) {
+                               bp_first = bp;
+                       }
+                       bp_last = bp;
+               }
+       }
+       if (found) {
+               lck_mtx_unlock(bpf_mlock);
+               printf("bpfattach - %s with dlt %d is already attached\n",
+                   if_name(ifp), dlt);
+               FREE(bp_new, M_DEVBUF);
+               return EEXIST;
+       }
 
 
-       bp->bif_next = bpf_iflist;
-       bpf_iflist = bp;
+       bp_new->bif_ifp = ifp;
+       bp_new->bif_dlt = dlt;
+       bp_new->bif_send = send;
+       bp_new->bif_tap = tap;
 
 
-       bp->bif_ifp->if_bpf = 0;
+       if (bp_first == NULL) {
+               /* No other entries for this ifp */
+               bp_new->bif_next = bpf_iflist;
+               bpf_iflist = bp_new;
+       } else {
+               if (ifnet_type(ifp) == IFT_ETHER && dlt == DLT_EN10MB) {
+                       /* Make this the first entry for this interface */
+                       if (bp_before_first != NULL) {
+                               /*  point the previous to us */
+                               bp_before_first->bif_next = bp_new;
+                       } else {
+                               /* we're the new head */
+                               bpf_iflist = bp_new;
+                       }
+                       bp_new->bif_next = bp_first;
+               } else {
+                       /* Add this after the last entry for this interface */
+                       bp_new->bif_next = bp_last->bif_next;
+                       bp_last->bif_next = bp_new;
+               }
+       }
 
        /*
         * Compute the length of the bpf header.  This is not necessarily
 
        /*
         * Compute the length of the bpf header.  This is not necessarily
@@ -1525,17 +3577,22 @@ bpfattach(struct ifnet *ifp, u_int dlt, u_int hdrlen)
         * that the network layer header begins on a longword boundary (for
         * performance reasons and to alleviate alignment restrictions).
         */
         * that the network layer header begins on a longword boundary (for
         * performance reasons and to alleviate alignment restrictions).
         */
-       bp->bif_hdrlen = BPF_WORDALIGN(hdrlen + SIZEOF_BPF_HDR) - hdrlen;
-       
+       bp_new->bif_hdrlen = BPF_WORDALIGN(hdrlen + SIZEOF_BPF_HDR) - hdrlen;
+       bp_new->bif_exthdrlen = BPF_WORDALIGN(hdrlen +
+           sizeof(struct bpf_hdr_ext)) - hdrlen;
+
        /* Take a reference on the interface */
        /* Take a reference on the interface */
-       ifp_reference(ifp);
+       ifnet_reference(ifp);
 
        lck_mtx_unlock(bpf_mlock);
 
 #ifndef __APPLE__
 
        lck_mtx_unlock(bpf_mlock);
 
 #ifndef __APPLE__
-       if (bootverbose)
-               printf("bpf: %s%d attached\n", ifp->if_name, ifp->if_unit);
+       if (bootverbose) {
+               printf("bpf: %s attached\n", if_name(ifp));
+       }
 #endif
 #endif
+
+       return 0;
 }
 
 /*
 }
 
 /*
@@ -1547,107 +3604,90 @@ bpfattach(struct ifnet *ifp, u_int dlt, u_int hdrlen)
 void
 bpfdetach(struct ifnet *ifp)
 {
 void
 bpfdetach(struct ifnet *ifp)
 {
-       struct bpf_if   *bp, *bp_prev;
-       struct bpf_d    *d;
-       int     s;
+       struct bpf_if   *bp, *bp_prev, *bp_next;
+       struct bpf_d    *d;
+
+       if (bpf_debug != 0) {
+               printf("%s: %s\n", __func__, if_name(ifp));
+       }
 
 
-       s = splimp();
-       
        lck_mtx_lock(bpf_mlock);
 
        lck_mtx_lock(bpf_mlock);
 
-       /* Locate BPF interface information */
+       /*
+        * Build the list of devices attached to that interface
+        * that we need to free while keeping the lock to maintain
+        * the integrity of the interface list
+        */
        bp_prev = NULL;
        bp_prev = NULL;
-       for (bp = bpf_iflist; bp != NULL; bp = bp->bif_next) {
-               if (ifp == bp->bif_ifp)
-                       break;
-               bp_prev = bp;
-       }
-
-#ifdef __APPLE__
-       /* Check for no BPF interface information */
-       if (bp == NULL) {
-               return;
-       }
-#endif
+       for (bp = bpf_iflist; bp != NULL; bp = bp_next) {
+               bp_next = bp->bif_next;
 
 
-       /* Interface wasn't attached */
-       if (bp->bif_ifp == NULL) {
-               splx(s);
-#ifndef __APPLE__
-               printf("bpfdetach: %s%d was not attached\n", ifp->if_name,
-                   ifp->if_unit);
-#endif
-               return;
-       }
+               if (ifp != bp->bif_ifp) {
+                       bp_prev = bp;
+                       continue;
+               }
+               /* Unlink from the interface list */
+               if (bp_prev) {
+                       bp_prev->bif_next = bp->bif_next;
+               } else {
+                       bpf_iflist = bp->bif_next;
+               }
 
 
-       while ((d = bp->bif_dlist) != NULL) {
-               bpf_detachd(d);
-               bpf_wakeup(d);
+               /* Detach the devices attached to the interface */
+               while ((d = bp->bif_dlist) != NULL) {
+                       /*
+                        * Take an extra reference to prevent the device
+                        * from being freed when bpf_detachd() releases
+                        * the reference for the interface list
+                        */
+                       bpf_acquire_d(d);
+                       bpf_detachd(d, 0);
+                       bpf_wakeup(d);
+                       bpf_release_d(d);
+               }
+               ifnet_release(ifp);
        }
 
        }
 
-       if (bp_prev) {
-               bp_prev->bif_next = bp->bif_next;
-       } else {
-               bpf_iflist = bp->bif_next;
-       }
-       
-       ifp_release(ifp);
-       
        lck_mtx_unlock(bpf_mlock);
        lck_mtx_unlock(bpf_mlock);
-
-       FREE(bp, M_DEVBUF);
-
-       splx(s);
 }
 
 void
 bpf_init(__unused void *unused)
 {
 #ifdef __APPLE__
 }
 
 void
 bpf_init(__unused void *unused)
 {
 #ifdef __APPLE__
-       int     i;
-       int     maj;
+       int     i;
+       int     maj;
 
        if (bpf_devsw_installed == 0) {
                bpf_devsw_installed = 1;
 
        if (bpf_devsw_installed == 0) {
                bpf_devsw_installed = 1;
-
-        bpf_mlock_grp_attr = lck_grp_attr_alloc_init();
-        lck_grp_attr_setdefault(bpf_mlock_grp_attr);
-
-        bpf_mlock_grp = lck_grp_alloc_init("bpf", bpf_mlock_grp_attr);
-
-        bpf_mlock_attr = lck_attr_alloc_init();
-        lck_attr_setdefault(bpf_mlock_attr);
-
-        bpf_mlock = lck_mtx_alloc_init(bpf_mlock_grp, bpf_mlock_attr);
-
-               if (bpf_mlock == 0) {
-                       printf("bpf_init: failed to allocate bpf_mlock\n");
-                       bpf_devsw_installed = 0;
-                       return;
-               }
-               
+               bpf_mlock_grp_attr = lck_grp_attr_alloc_init();
+               bpf_mlock_grp = lck_grp_alloc_init("bpf", bpf_mlock_grp_attr);
+               bpf_mlock_attr = lck_attr_alloc_init();
+               lck_mtx_init(bpf_mlock, bpf_mlock_grp, bpf_mlock_attr);
                maj = cdevsw_add(CDEV_MAJOR, &bpf_cdevsw);
                if (maj == -1) {
                maj = cdevsw_add(CDEV_MAJOR, &bpf_cdevsw);
                if (maj == -1) {
-                       if (bpf_mlock)
-                               lck_mtx_free(bpf_mlock, bpf_mlock_grp);
-                       if (bpf_mlock_attr)
+                       if (bpf_mlock_attr) {
                                lck_attr_free(bpf_mlock_attr);
                                lck_attr_free(bpf_mlock_attr);
-                       if (bpf_mlock_grp)
+                       }
+                       if (bpf_mlock_grp) {
                                lck_grp_free(bpf_mlock_grp);
                                lck_grp_free(bpf_mlock_grp);
-                       if (bpf_mlock_grp_attr)
+                       }
+                       if (bpf_mlock_grp_attr) {
                                lck_grp_attr_free(bpf_mlock_grp_attr);
                                lck_grp_attr_free(bpf_mlock_grp_attr);
-                       
-                       bpf_mlock = 0;
-                       bpf_mlock_attr = 0;
-                       bpf_mlock_grp = 0;
-                       bpf_mlock_grp_attr = 0;
+                       }
+
+                       bpf_mlock = NULL;
+                       bpf_mlock_attr = NULL;
+                       bpf_mlock_grp = NULL;
+                       bpf_mlock_grp_attr = NULL;
                        bpf_devsw_installed = 0;
                        bpf_devsw_installed = 0;
-                       printf("bpf_init: failed to allocate a major number!\n");
+                       printf("bpf_init: failed to allocate a major number\n");
                        return;
                }
 
                        return;
                }
 
-               for (i = 0 ; i < NBPFILTER; i++)
+               for (i = 0; i < NBPFILTER; i++) {
                        bpf_make_dev_t(maj);
                        bpf_make_dev_t(maj);
+               }
        }
 #else
        cdevsw_add(&bpf_cdevsw);
        }
 #else
        cdevsw_add(&bpf_cdevsw);
@@ -1655,54 +3695,26 @@ bpf_init(__unused void *unused)
 }
 
 #ifndef __APPLE__
 }
 
 #ifndef __APPLE__
-SYSINIT(bpfdev,SI_SUB_DRIVERS,SI_ORDER_MIDDLE+CDEV_MAJOR,bpf_drvinit,NULL)
+SYSINIT(bpfdev, SI_SUB_DRIVERS, SI_ORDER_MIDDLE + CDEV_MAJOR, bpf_drvinit, NULL);
 #endif
 
 #endif
 
-#else /* !BPF */
-#ifndef __APPLE__
-/*
- * NOP stubs to allow bpf-using drivers to load and function.
- *
- * A 'better' implementation would allow the core bpf functionality
- * to be loaded at runtime.
- */
-
-void
-bpf_tap(ifp, pkt, pktlen)
-       struct ifnet *ifp;
-       register u_char *pkt;
-       register u_int pktlen;
+static int
+sysctl_bpf_maxbufsize SYSCTL_HANDLER_ARGS
 {
 {
-}
+#pragma unused(arg1, arg2)
+       int i, err;
 
 
-void
-bpf_mtap(ifp, m)
-       struct ifnet *ifp;
-       struct mbuf *m;
-{
-}
+       i = bpf_maxbufsize;
 
 
-void
-bpfattach(ifp, dlt, hdrlen)
-       struct ifnet *ifp;
-       u_int dlt, hdrlen;
-{
-}
+       err = sysctl_handle_int(oidp, &i, 0, req);
+       if (err != 0 || req->newptr == USER_ADDR_NULL) {
+               return err;
+       }
 
 
-void
-bpfdetach(ifp)
-       struct ifnet *ifp;
-{
-}
+       if (i < 0 || i > BPF_MAXSIZE_CAP) {
+               i = BPF_MAXSIZE_CAP;
+       }
 
 
-u_int
-bpf_filter(pc, p, wirelen, buflen)
-       register const struct bpf_insn *pc;
-       register u_char *p;
-       u_int wirelen;
-       register u_int buflen;
-{
-       return -1;      /* "no filter" behaviour */
+       bpf_maxbufsize = i;
+       return err;
 }
 }
-#endif /* !defined(__APPLE__) */
-#endif /* NBPFILTER > 0 */