81062ad740
Update libpcap from 1.9.0 to 1.9.1. MFC after: 2 weeks
3339 lines
84 KiB
C
3339 lines
84 KiB
C
/*
|
|
* Copyright (c) 1993, 1994, 1995, 1996, 1998
|
|
* The Regents of the University of California. All rights reserved.
|
|
*
|
|
* Redistribution and use in source and binary forms, with or without
|
|
* modification, are permitted provided that: (1) source code distributions
|
|
* retain the above copyright notice and this paragraph in its entirety, (2)
|
|
* distributions including binary code include the above copyright notice and
|
|
* this paragraph in its entirety in the documentation or other materials
|
|
* provided with the distribution, and (3) all advertising materials mentioning
|
|
* features or use of this software display the following acknowledgement:
|
|
* ``This product includes software developed by the University of California,
|
|
* Lawrence Berkeley Laboratory and its contributors.'' Neither the name of
|
|
* the University nor the names of its contributors may be used to endorse
|
|
* or promote products derived from this software without specific prior
|
|
* written permission.
|
|
* THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED
|
|
* WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF
|
|
* MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
|
|
*/
|
|
|
|
#ifdef HAVE_CONFIG_H
|
|
#include <config.h>
|
|
#endif
|
|
|
|
#include <sys/param.h> /* optionally get BSD define */
|
|
#include <sys/socket.h>
|
|
#include <time.h>
|
|
/*
|
|
* <net/bpf.h> defines ioctls, but doesn't include <sys/ioccom.h>.
|
|
*
|
|
* We include <sys/ioctl.h> as it might be necessary to declare ioctl();
|
|
* at least on *BSD and macOS, it also defines various SIOC ioctls -
|
|
* we could include <sys/sockio.h>, but if we're already including
|
|
* <sys/ioctl.h>, which includes <sys/sockio.h> on those platforms,
|
|
* there's not much point in doing so.
|
|
*
|
|
* If we have <sys/ioccom.h>, we include it as well, to handle systems
|
|
* such as Solaris which don't arrange to include <sys/ioccom.h> if you
|
|
* include <sys/ioctl.h>
|
|
*/
|
|
#include <sys/ioctl.h>
|
|
#ifdef HAVE_SYS_IOCCOM_H
|
|
#include <sys/ioccom.h>
|
|
#endif
|
|
#include <sys/utsname.h>
|
|
|
|
#if defined(__FreeBSD__) && defined(SIOCIFCREATE2)
|
|
/*
|
|
* Add support for capturing on FreeBSD usbusN interfaces.
|
|
*/
|
|
static const char usbus_prefix[] = "usbus";
|
|
#define USBUS_PREFIX_LEN (sizeof(usbus_prefix) - 1)
|
|
#include <dirent.h>
|
|
#endif
|
|
|
|
#include <net/if.h>
|
|
|
|
#ifdef _AIX
|
|
|
|
/*
|
|
* Make "pcap.h" not include "pcap/bpf.h"; we are going to include the
|
|
* native OS version, as we need "struct bpf_config" from it.
|
|
*/
|
|
#define PCAP_DONT_INCLUDE_PCAP_BPF_H
|
|
|
|
#include <sys/types.h>
|
|
|
|
/*
|
|
* Prevent bpf.h from redefining the DLT_ values to their
|
|
* IFT_ values, as we're going to return the standard libpcap
|
|
* values, not IBM's non-standard IFT_ values.
|
|
*/
|
|
#undef _AIX
|
|
#include <net/bpf.h>
|
|
#define _AIX
|
|
|
|
/*
|
|
* If both BIOCROTZBUF and BPF_BUFMODE_ZBUF are defined, we have
|
|
* zero-copy BPF.
|
|
*/
|
|
#if defined(BIOCROTZBUF) && defined(BPF_BUFMODE_ZBUF)
|
|
#define HAVE_ZEROCOPY_BPF
|
|
#include <sys/mman.h>
|
|
#include <machine/atomic.h>
|
|
#endif
|
|
|
|
#include <net/if_types.h> /* for IFT_ values */
|
|
#include <sys/sysconfig.h>
|
|
#include <sys/device.h>
|
|
#include <sys/cfgodm.h>
|
|
#include <cf.h>
|
|
|
|
#ifdef __64BIT__
|
|
#define domakedev makedev64
|
|
#define getmajor major64
|
|
#define bpf_hdr bpf_hdr32
|
|
#else /* __64BIT__ */
|
|
#define domakedev makedev
|
|
#define getmajor major
|
|
#endif /* __64BIT__ */
|
|
|
|
#define BPF_NAME "bpf"
|
|
#define BPF_MINORS 4
|
|
#define DRIVER_PATH "/usr/lib/drivers"
|
|
#define BPF_NODE "/dev/bpf"
|
|
static int bpfloadedflag = 0;
|
|
static int odmlockid = 0;
|
|
|
|
static int bpf_load(char *errbuf);
|
|
|
|
#else /* _AIX */
|
|
|
|
#include <net/bpf.h>
|
|
|
|
#endif /* _AIX */
|
|
|
|
#include <ctype.h>
|
|
#include <fcntl.h>
|
|
#include <errno.h>
|
|
#include <netdb.h>
|
|
#include <stdio.h>
|
|
#include <stdlib.h>
|
|
#include <string.h>
|
|
#include <unistd.h>
|
|
|
|
#ifdef SIOCGIFMEDIA
|
|
# include <net/if_media.h>
|
|
#endif
|
|
|
|
#include "pcap-int.h"
|
|
|
|
#ifdef HAVE_OS_PROTO_H
|
|
#include "os-proto.h"
|
|
#endif
|
|
|
|
/*
|
|
* Later versions of NetBSD stick padding in front of FDDI frames
|
|
* to align the IP header on a 4-byte boundary.
|
|
*/
|
|
#if defined(__NetBSD__) && __NetBSD_Version__ > 106000000
|
|
#define PCAP_FDDIPAD 3
|
|
#endif
|
|
|
|
/*
|
|
* Private data for capturing on BPF devices.
|
|
*/
|
|
struct pcap_bpf {
|
|
#ifdef HAVE_ZEROCOPY_BPF
|
|
/*
|
|
* Zero-copy read buffer -- for zero-copy BPF. 'buffer' above will
|
|
* alternative between these two actual mmap'd buffers as required.
|
|
* As there is a header on the front size of the mmap'd buffer, only
|
|
* some of the buffer is exposed to libpcap as a whole via bufsize;
|
|
* zbufsize is the true size. zbuffer tracks the current zbuf
|
|
* assocated with buffer so that it can be used to decide which the
|
|
* next buffer to read will be.
|
|
*/
|
|
u_char *zbuf1, *zbuf2, *zbuffer;
|
|
u_int zbufsize;
|
|
u_int zerocopy;
|
|
u_int interrupted;
|
|
struct timespec firstsel;
|
|
/*
|
|
* If there's currently a buffer being actively processed, then it is
|
|
* referenced here; 'buffer' is also pointed at it, but offset by the
|
|
* size of the header.
|
|
*/
|
|
struct bpf_zbuf_header *bzh;
|
|
int nonblock; /* true if in nonblocking mode */
|
|
#endif /* HAVE_ZEROCOPY_BPF */
|
|
|
|
char *device; /* device name */
|
|
int filtering_in_kernel; /* using kernel filter */
|
|
int must_do_on_close; /* stuff we must do when we close */
|
|
};
|
|
|
|
/*
|
|
* Stuff to do when we close.
|
|
*/
|
|
#define MUST_CLEAR_RFMON 0x00000001 /* clear rfmon (monitor) mode */
|
|
#define MUST_DESTROY_USBUS 0x00000002 /* destroy usbusN interface */
|
|
|
|
#ifdef BIOCGDLTLIST
|
|
# if (defined(HAVE_NET_IF_MEDIA_H) && defined(IFM_IEEE80211)) && !defined(__APPLE__)
|
|
#define HAVE_BSD_IEEE80211
|
|
|
|
/*
|
|
* The ifm_ulist member of a struct ifmediareq is an int * on most systems,
|
|
* but it's a uint64_t on newer versions of OpenBSD.
|
|
*
|
|
* We check this by checking whether IFM_GMASK is defined and > 2^32-1.
|
|
*/
|
|
# if defined(IFM_GMASK) && IFM_GMASK > 0xFFFFFFFF
|
|
# define IFM_ULIST_TYPE uint64_t
|
|
# else
|
|
# define IFM_ULIST_TYPE int
|
|
# endif
|
|
# endif
|
|
|
|
# if defined(__APPLE__) || defined(HAVE_BSD_IEEE80211)
|
|
static int find_802_11(struct bpf_dltlist *);
|
|
|
|
# ifdef HAVE_BSD_IEEE80211
|
|
static int monitor_mode(pcap_t *, int);
|
|
# endif
|
|
|
|
# if defined(__APPLE__)
|
|
static void remove_non_802_11(pcap_t *);
|
|
static void remove_802_11(pcap_t *);
|
|
# endif
|
|
|
|
# endif /* defined(__APPLE__) || defined(HAVE_BSD_IEEE80211) */
|
|
|
|
#endif /* BIOCGDLTLIST */
|
|
|
|
#if defined(sun) && defined(LIFNAMSIZ) && defined(lifr_zoneid)
|
|
#include <zone.h>
|
|
#endif
|
|
|
|
/*
|
|
* We include the OS's <net/bpf.h>, not our "pcap/bpf.h", so we probably
|
|
* don't get DLT_DOCSIS defined.
|
|
*/
|
|
#ifndef DLT_DOCSIS
|
|
#define DLT_DOCSIS 143
|
|
#endif
|
|
|
|
/*
|
|
* In some versions of macOS, we might not even get any of the
|
|
* 802.11-plus-radio-header DLT_'s defined, even though some
|
|
* of them are used by various Airport drivers in those versions.
|
|
*/
|
|
#ifndef DLT_PRISM_HEADER
|
|
#define DLT_PRISM_HEADER 119
|
|
#endif
|
|
#ifndef DLT_AIRONET_HEADER
|
|
#define DLT_AIRONET_HEADER 120
|
|
#endif
|
|
#ifndef DLT_IEEE802_11_RADIO
|
|
#define DLT_IEEE802_11_RADIO 127
|
|
#endif
|
|
#ifndef DLT_IEEE802_11_RADIO_AVS
|
|
#define DLT_IEEE802_11_RADIO_AVS 163
|
|
#endif
|
|
|
|
static int pcap_can_set_rfmon_bpf(pcap_t *p);
|
|
static int pcap_activate_bpf(pcap_t *p);
|
|
static int pcap_setfilter_bpf(pcap_t *p, struct bpf_program *fp);
|
|
static int pcap_setdirection_bpf(pcap_t *, pcap_direction_t);
|
|
static int pcap_set_datalink_bpf(pcap_t *p, int dlt);
|
|
|
|
/*
|
|
* For zerocopy bpf, the setnonblock/getnonblock routines need to modify
|
|
* pb->nonblock so we don't call select(2) if the pcap handle is in non-
|
|
* blocking mode.
|
|
*/
|
|
static int
|
|
pcap_getnonblock_bpf(pcap_t *p)
|
|
{
|
|
#ifdef HAVE_ZEROCOPY_BPF
|
|
struct pcap_bpf *pb = p->priv;
|
|
|
|
if (pb->zerocopy)
|
|
return (pb->nonblock);
|
|
#endif
|
|
return (pcap_getnonblock_fd(p));
|
|
}
|
|
|
|
static int
|
|
pcap_setnonblock_bpf(pcap_t *p, int nonblock)
|
|
{
|
|
#ifdef HAVE_ZEROCOPY_BPF
|
|
struct pcap_bpf *pb = p->priv;
|
|
|
|
if (pb->zerocopy) {
|
|
pb->nonblock = nonblock;
|
|
return (0);
|
|
}
|
|
#endif
|
|
return (pcap_setnonblock_fd(p, nonblock));
|
|
}
|
|
|
|
#ifdef HAVE_ZEROCOPY_BPF
|
|
/*
|
|
* Zero-copy BPF buffer routines to check for and acknowledge BPF data in
|
|
* shared memory buffers.
|
|
*
|
|
* pcap_next_zbuf_shm(): Check for a newly available shared memory buffer,
|
|
* and set up p->buffer and cc to reflect one if available. Notice that if
|
|
* there was no prior buffer, we select zbuf1 as this will be the first
|
|
* buffer filled for a fresh BPF session.
|
|
*/
|
|
static int
|
|
pcap_next_zbuf_shm(pcap_t *p, int *cc)
|
|
{
|
|
struct pcap_bpf *pb = p->priv;
|
|
struct bpf_zbuf_header *bzh;
|
|
|
|
if (pb->zbuffer == pb->zbuf2 || pb->zbuffer == NULL) {
|
|
bzh = (struct bpf_zbuf_header *)pb->zbuf1;
|
|
if (bzh->bzh_user_gen !=
|
|
atomic_load_acq_int(&bzh->bzh_kernel_gen)) {
|
|
pb->bzh = bzh;
|
|
pb->zbuffer = (u_char *)pb->zbuf1;
|
|
p->buffer = pb->zbuffer + sizeof(*bzh);
|
|
*cc = bzh->bzh_kernel_len;
|
|
return (1);
|
|
}
|
|
} else if (pb->zbuffer == pb->zbuf1) {
|
|
bzh = (struct bpf_zbuf_header *)pb->zbuf2;
|
|
if (bzh->bzh_user_gen !=
|
|
atomic_load_acq_int(&bzh->bzh_kernel_gen)) {
|
|
pb->bzh = bzh;
|
|
pb->zbuffer = (u_char *)pb->zbuf2;
|
|
p->buffer = pb->zbuffer + sizeof(*bzh);
|
|
*cc = bzh->bzh_kernel_len;
|
|
return (1);
|
|
}
|
|
}
|
|
*cc = 0;
|
|
return (0);
|
|
}
|
|
|
|
/*
|
|
* pcap_next_zbuf() -- Similar to pcap_next_zbuf_shm(), except wait using
|
|
* select() for data or a timeout, and possibly force rotation of the buffer
|
|
* in the event we time out or are in immediate mode. Invoke the shared
|
|
* memory check before doing system calls in order to avoid doing avoidable
|
|
* work.
|
|
*/
|
|
static int
|
|
pcap_next_zbuf(pcap_t *p, int *cc)
|
|
{
|
|
struct pcap_bpf *pb = p->priv;
|
|
struct bpf_zbuf bz;
|
|
struct timeval tv;
|
|
struct timespec cur;
|
|
fd_set r_set;
|
|
int data, r;
|
|
int expire, tmout;
|
|
|
|
#define TSTOMILLI(ts) (((ts)->tv_sec * 1000) + ((ts)->tv_nsec / 1000000))
|
|
/*
|
|
* Start out by seeing whether anything is waiting by checking the
|
|
* next shared memory buffer for data.
|
|
*/
|
|
data = pcap_next_zbuf_shm(p, cc);
|
|
if (data)
|
|
return (data);
|
|
/*
|
|
* If a previous sleep was interrupted due to signal delivery, make
|
|
* sure that the timeout gets adjusted accordingly. This requires
|
|
* that we analyze when the timeout should be been expired, and
|
|
* subtract the current time from that. If after this operation,
|
|
* our timeout is less then or equal to zero, handle it like a
|
|
* regular timeout.
|
|
*/
|
|
tmout = p->opt.timeout;
|
|
if (tmout)
|
|
(void) clock_gettime(CLOCK_MONOTONIC, &cur);
|
|
if (pb->interrupted && p->opt.timeout) {
|
|
expire = TSTOMILLI(&pb->firstsel) + p->opt.timeout;
|
|
tmout = expire - TSTOMILLI(&cur);
|
|
#undef TSTOMILLI
|
|
if (tmout <= 0) {
|
|
pb->interrupted = 0;
|
|
data = pcap_next_zbuf_shm(p, cc);
|
|
if (data)
|
|
return (data);
|
|
if (ioctl(p->fd, BIOCROTZBUF, &bz) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf,
|
|
PCAP_ERRBUF_SIZE, errno, "BIOCROTZBUF");
|
|
return (PCAP_ERROR);
|
|
}
|
|
return (pcap_next_zbuf_shm(p, cc));
|
|
}
|
|
}
|
|
/*
|
|
* No data in the buffer, so must use select() to wait for data or
|
|
* the next timeout. Note that we only call select if the handle
|
|
* is in blocking mode.
|
|
*/
|
|
if (!pb->nonblock) {
|
|
FD_ZERO(&r_set);
|
|
FD_SET(p->fd, &r_set);
|
|
if (tmout != 0) {
|
|
tv.tv_sec = tmout / 1000;
|
|
tv.tv_usec = (tmout * 1000) % 1000000;
|
|
}
|
|
r = select(p->fd + 1, &r_set, NULL, NULL,
|
|
p->opt.timeout != 0 ? &tv : NULL);
|
|
if (r < 0 && errno == EINTR) {
|
|
if (!pb->interrupted && p->opt.timeout) {
|
|
pb->interrupted = 1;
|
|
pb->firstsel = cur;
|
|
}
|
|
return (0);
|
|
} else if (r < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "select");
|
|
return (PCAP_ERROR);
|
|
}
|
|
}
|
|
pb->interrupted = 0;
|
|
/*
|
|
* Check again for data, which may exist now that we've either been
|
|
* woken up as a result of data or timed out. Try the "there's data"
|
|
* case first since it doesn't require a system call.
|
|
*/
|
|
data = pcap_next_zbuf_shm(p, cc);
|
|
if (data)
|
|
return (data);
|
|
/*
|
|
* Try forcing a buffer rotation to dislodge timed out or immediate
|
|
* data.
|
|
*/
|
|
if (ioctl(p->fd, BIOCROTZBUF, &bz) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCROTZBUF");
|
|
return (PCAP_ERROR);
|
|
}
|
|
return (pcap_next_zbuf_shm(p, cc));
|
|
}
|
|
|
|
/*
|
|
* Notify kernel that we are done with the buffer. We don't reset zbuffer so
|
|
* that we know which buffer to use next time around.
|
|
*/
|
|
static int
|
|
pcap_ack_zbuf(pcap_t *p)
|
|
{
|
|
struct pcap_bpf *pb = p->priv;
|
|
|
|
atomic_store_rel_int(&pb->bzh->bzh_user_gen,
|
|
pb->bzh->bzh_kernel_gen);
|
|
pb->bzh = NULL;
|
|
p->buffer = NULL;
|
|
return (0);
|
|
}
|
|
#endif /* HAVE_ZEROCOPY_BPF */
|
|
|
|
pcap_t *
|
|
pcap_create_interface(const char *device _U_, char *ebuf)
|
|
{
|
|
pcap_t *p;
|
|
|
|
p = pcap_create_common(ebuf, sizeof (struct pcap_bpf));
|
|
if (p == NULL)
|
|
return (NULL);
|
|
|
|
p->activate_op = pcap_activate_bpf;
|
|
p->can_set_rfmon_op = pcap_can_set_rfmon_bpf;
|
|
#ifdef BIOCSTSTAMP
|
|
/*
|
|
* We claim that we support microsecond and nanosecond time
|
|
* stamps.
|
|
*/
|
|
p->tstamp_precision_count = 2;
|
|
p->tstamp_precision_list = malloc(2 * sizeof(u_int));
|
|
if (p->tstamp_precision_list == NULL) {
|
|
pcap_fmt_errmsg_for_errno(ebuf, PCAP_ERRBUF_SIZE, errno,
|
|
"malloc");
|
|
free(p);
|
|
return (NULL);
|
|
}
|
|
p->tstamp_precision_list[0] = PCAP_TSTAMP_PRECISION_MICRO;
|
|
p->tstamp_precision_list[1] = PCAP_TSTAMP_PRECISION_NANO;
|
|
#endif /* BIOCSTSTAMP */
|
|
return (p);
|
|
}
|
|
|
|
/*
|
|
* On success, returns a file descriptor for a BPF device.
|
|
* On failure, returns a PCAP_ERROR_ value, and sets p->errbuf.
|
|
*/
|
|
static int
|
|
bpf_open(char *errbuf)
|
|
{
|
|
int fd = -1;
|
|
static const char cloning_device[] = "/dev/bpf";
|
|
int n = 0;
|
|
char device[sizeof "/dev/bpf0000000000"];
|
|
static int no_cloning_bpf = 0;
|
|
|
|
#ifdef _AIX
|
|
/*
|
|
* Load the bpf driver, if it isn't already loaded,
|
|
* and create the BPF device entries, if they don't
|
|
* already exist.
|
|
*/
|
|
if (bpf_load(errbuf) == PCAP_ERROR)
|
|
return (PCAP_ERROR);
|
|
#endif
|
|
|
|
/*
|
|
* First, unless we've already tried opening /dev/bpf and
|
|
* gotten ENOENT, try opening /dev/bpf.
|
|
* If it fails with ENOENT, remember that, so we don't try
|
|
* again, and try /dev/bpfN.
|
|
*/
|
|
if (!no_cloning_bpf &&
|
|
(fd = open(cloning_device, O_RDWR)) == -1 &&
|
|
((errno != EACCES && errno != ENOENT) ||
|
|
(fd = open(cloning_device, O_RDONLY)) == -1)) {
|
|
if (errno != ENOENT) {
|
|
if (errno == EACCES)
|
|
fd = PCAP_ERROR_PERM_DENIED;
|
|
else
|
|
fd = PCAP_ERROR;
|
|
pcap_fmt_errmsg_for_errno(errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "(cannot open device) %s", cloning_device);
|
|
return (fd);
|
|
}
|
|
no_cloning_bpf = 1;
|
|
}
|
|
|
|
if (no_cloning_bpf) {
|
|
/*
|
|
* We don't have /dev/bpf.
|
|
* Go through all the /dev/bpfN minors and find one
|
|
* that isn't in use.
|
|
*/
|
|
do {
|
|
(void)pcap_snprintf(device, sizeof(device), "/dev/bpf%d", n++);
|
|
/*
|
|
* Initially try a read/write open (to allow the inject
|
|
* method to work). If that fails due to permission
|
|
* issues, fall back to read-only. This allows a
|
|
* non-root user to be granted specific access to pcap
|
|
* capabilities via file permissions.
|
|
*
|
|
* XXX - we should have an API that has a flag that
|
|
* controls whether to open read-only or read-write,
|
|
* so that denial of permission to send (or inability
|
|
* to send, if sending packets isn't supported on
|
|
* the device in question) can be indicated at open
|
|
* time.
|
|
*/
|
|
fd = open(device, O_RDWR);
|
|
if (fd == -1 && errno == EACCES)
|
|
fd = open(device, O_RDONLY);
|
|
} while (fd < 0 && errno == EBUSY);
|
|
}
|
|
|
|
/*
|
|
* XXX better message for all minors used
|
|
*/
|
|
if (fd < 0) {
|
|
switch (errno) {
|
|
|
|
case ENOENT:
|
|
fd = PCAP_ERROR;
|
|
if (n == 1) {
|
|
/*
|
|
* /dev/bpf0 doesn't exist, which
|
|
* means we probably have no BPF
|
|
* devices.
|
|
*/
|
|
pcap_snprintf(errbuf, PCAP_ERRBUF_SIZE,
|
|
"(there are no BPF devices)");
|
|
} else {
|
|
/*
|
|
* We got EBUSY on at least one
|
|
* BPF device, so we have BPF
|
|
* devices, but all the ones
|
|
* that exist are busy.
|
|
*/
|
|
pcap_snprintf(errbuf, PCAP_ERRBUF_SIZE,
|
|
"(all BPF devices are busy)");
|
|
}
|
|
break;
|
|
|
|
case EACCES:
|
|
/*
|
|
* Got EACCES on the last device we tried,
|
|
* and EBUSY on all devices before that,
|
|
* if any.
|
|
*/
|
|
fd = PCAP_ERROR_PERM_DENIED;
|
|
pcap_fmt_errmsg_for_errno(errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "(cannot open BPF device) %s", device);
|
|
break;
|
|
|
|
default:
|
|
/*
|
|
* Some other problem.
|
|
*/
|
|
fd = PCAP_ERROR;
|
|
pcap_fmt_errmsg_for_errno(errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "(cannot open BPF device) %s", device);
|
|
break;
|
|
}
|
|
}
|
|
|
|
return (fd);
|
|
}
|
|
|
|
/*
|
|
* Open and bind to a device; used if we're not actually going to use
|
|
* the device, but are just testing whether it can be opened, or opening
|
|
* it to get information about it.
|
|
*
|
|
* Returns an error code on failure (always negative), and an FD for
|
|
* the now-bound BPF device on success (always non-negative).
|
|
*/
|
|
static int
|
|
bpf_open_and_bind(const char *name, char *errbuf)
|
|
{
|
|
int fd;
|
|
struct ifreq ifr;
|
|
|
|
/*
|
|
* First, open a BPF device.
|
|
*/
|
|
fd = bpf_open(errbuf);
|
|
if (fd < 0)
|
|
return (fd); /* fd is the appropriate error code */
|
|
|
|
/*
|
|
* Now bind to the device.
|
|
*/
|
|
(void)strncpy(ifr.ifr_name, name, sizeof(ifr.ifr_name));
|
|
if (ioctl(fd, BIOCSETIF, (caddr_t)&ifr) < 0) {
|
|
switch (errno) {
|
|
|
|
case ENXIO:
|
|
/*
|
|
* There's no such device.
|
|
*/
|
|
close(fd);
|
|
return (PCAP_ERROR_NO_SUCH_DEVICE);
|
|
|
|
case ENETDOWN:
|
|
/*
|
|
* Return a "network down" indication, so that
|
|
* the application can report that rather than
|
|
* saying we had a mysterious failure and
|
|
* suggest that they report a problem to the
|
|
* libpcap developers.
|
|
*/
|
|
close(fd);
|
|
return (PCAP_ERROR_IFACE_NOT_UP);
|
|
|
|
default:
|
|
pcap_fmt_errmsg_for_errno(errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCSETIF: %s", name);
|
|
close(fd);
|
|
return (PCAP_ERROR);
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Success.
|
|
*/
|
|
return (fd);
|
|
}
|
|
|
|
#ifdef BIOCGDLTLIST
|
|
static int
|
|
get_dlt_list(int fd, int v, struct bpf_dltlist *bdlp, char *ebuf)
|
|
{
|
|
memset(bdlp, 0, sizeof(*bdlp));
|
|
if (ioctl(fd, BIOCGDLTLIST, (caddr_t)bdlp) == 0) {
|
|
u_int i;
|
|
int is_ethernet;
|
|
|
|
bdlp->bfl_list = (u_int *) malloc(sizeof(u_int) * (bdlp->bfl_len + 1));
|
|
if (bdlp->bfl_list == NULL) {
|
|
pcap_fmt_errmsg_for_errno(ebuf, PCAP_ERRBUF_SIZE,
|
|
errno, "malloc");
|
|
return (PCAP_ERROR);
|
|
}
|
|
|
|
if (ioctl(fd, BIOCGDLTLIST, (caddr_t)bdlp) < 0) {
|
|
pcap_fmt_errmsg_for_errno(ebuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCGDLTLIST");
|
|
free(bdlp->bfl_list);
|
|
return (PCAP_ERROR);
|
|
}
|
|
|
|
/*
|
|
* OK, for real Ethernet devices, add DLT_DOCSIS to the
|
|
* list, so that an application can let you choose it,
|
|
* in case you're capturing DOCSIS traffic that a Cisco
|
|
* Cable Modem Termination System is putting out onto
|
|
* an Ethernet (it doesn't put an Ethernet header onto
|
|
* the wire, it puts raw DOCSIS frames out on the wire
|
|
* inside the low-level Ethernet framing).
|
|
*
|
|
* A "real Ethernet device" is defined here as a device
|
|
* that has a link-layer type of DLT_EN10MB and that has
|
|
* no alternate link-layer types; that's done to exclude
|
|
* 802.11 interfaces (which might or might not be the
|
|
* right thing to do, but I suspect it is - Ethernet <->
|
|
* 802.11 bridges would probably badly mishandle frames
|
|
* that don't have Ethernet headers).
|
|
*
|
|
* On Solaris with BPF, Ethernet devices also offer
|
|
* DLT_IPNET, so we, if DLT_IPNET is defined, we don't
|
|
* treat it as an indication that the device isn't an
|
|
* Ethernet.
|
|
*/
|
|
if (v == DLT_EN10MB) {
|
|
is_ethernet = 1;
|
|
for (i = 0; i < bdlp->bfl_len; i++) {
|
|
if (bdlp->bfl_list[i] != DLT_EN10MB
|
|
#ifdef DLT_IPNET
|
|
&& bdlp->bfl_list[i] != DLT_IPNET
|
|
#endif
|
|
) {
|
|
is_ethernet = 0;
|
|
break;
|
|
}
|
|
}
|
|
if (is_ethernet) {
|
|
/*
|
|
* We reserved one more slot at the end of
|
|
* the list.
|
|
*/
|
|
bdlp->bfl_list[bdlp->bfl_len] = DLT_DOCSIS;
|
|
bdlp->bfl_len++;
|
|
}
|
|
}
|
|
} else {
|
|
/*
|
|
* EINVAL just means "we don't support this ioctl on
|
|
* this device"; don't treat it as an error.
|
|
*/
|
|
if (errno != EINVAL) {
|
|
pcap_fmt_errmsg_for_errno(ebuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCGDLTLIST");
|
|
return (PCAP_ERROR);
|
|
}
|
|
}
|
|
return (0);
|
|
}
|
|
#endif
|
|
|
|
#if defined(__APPLE__)
|
|
static int
|
|
pcap_can_set_rfmon_bpf(pcap_t *p)
|
|
{
|
|
struct utsname osinfo;
|
|
struct ifreq ifr;
|
|
int fd;
|
|
#ifdef BIOCGDLTLIST
|
|
struct bpf_dltlist bdl;
|
|
#endif
|
|
|
|
/*
|
|
* The joys of monitor mode on Mac OS X/OS X/macOS.
|
|
*
|
|
* Prior to 10.4, it's not supported at all.
|
|
*
|
|
* In 10.4, if adapter enN supports monitor mode, there's a
|
|
* wltN adapter corresponding to it; you open it, instead of
|
|
* enN, to get monitor mode. You get whatever link-layer
|
|
* headers it supplies.
|
|
*
|
|
* In 10.5, and, we assume, later releases, if adapter enN
|
|
* supports monitor mode, it offers, among its selectable
|
|
* DLT_ values, values that let you get the 802.11 header;
|
|
* selecting one of those values puts the adapter into monitor
|
|
* mode (i.e., you can't get 802.11 headers except in monitor
|
|
* mode, and you can't get Ethernet headers in monitor mode).
|
|
*/
|
|
if (uname(&osinfo) == -1) {
|
|
/*
|
|
* Can't get the OS version; just say "no".
|
|
*/
|
|
return (0);
|
|
}
|
|
/*
|
|
* We assume osinfo.sysname is "Darwin", because
|
|
* __APPLE__ is defined. We just check the version.
|
|
*/
|
|
if (osinfo.release[0] < '8' && osinfo.release[1] == '.') {
|
|
/*
|
|
* 10.3 (Darwin 7.x) or earlier.
|
|
* Monitor mode not supported.
|
|
*/
|
|
return (0);
|
|
}
|
|
if (osinfo.release[0] == '8' && osinfo.release[1] == '.') {
|
|
/*
|
|
* 10.4 (Darwin 8.x). s/en/wlt/, and check
|
|
* whether the device exists.
|
|
*/
|
|
if (strncmp(p->opt.device, "en", 2) != 0) {
|
|
/*
|
|
* Not an enN device; no monitor mode.
|
|
*/
|
|
return (0);
|
|
}
|
|
fd = socket(AF_INET, SOCK_DGRAM, 0);
|
|
if (fd == -1) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "socket");
|
|
return (PCAP_ERROR);
|
|
}
|
|
pcap_strlcpy(ifr.ifr_name, "wlt", sizeof(ifr.ifr_name));
|
|
pcap_strlcat(ifr.ifr_name, p->opt.device + 2, sizeof(ifr.ifr_name));
|
|
if (ioctl(fd, SIOCGIFFLAGS, (char *)&ifr) < 0) {
|
|
/*
|
|
* No such device?
|
|
*/
|
|
close(fd);
|
|
return (0);
|
|
}
|
|
close(fd);
|
|
return (1);
|
|
}
|
|
|
|
#ifdef BIOCGDLTLIST
|
|
/*
|
|
* Everything else is 10.5 or later; for those,
|
|
* we just open the enN device, and check whether
|
|
* we have any 802.11 devices.
|
|
*
|
|
* First, open a BPF device.
|
|
*/
|
|
fd = bpf_open(p->errbuf);
|
|
if (fd < 0)
|
|
return (fd); /* fd is the appropriate error code */
|
|
|
|
/*
|
|
* Now bind to the device.
|
|
*/
|
|
(void)strncpy(ifr.ifr_name, p->opt.device, sizeof(ifr.ifr_name));
|
|
if (ioctl(fd, BIOCSETIF, (caddr_t)&ifr) < 0) {
|
|
switch (errno) {
|
|
|
|
case ENXIO:
|
|
/*
|
|
* There's no such device.
|
|
*/
|
|
close(fd);
|
|
return (PCAP_ERROR_NO_SUCH_DEVICE);
|
|
|
|
case ENETDOWN:
|
|
/*
|
|
* Return a "network down" indication, so that
|
|
* the application can report that rather than
|
|
* saying we had a mysterious failure and
|
|
* suggest that they report a problem to the
|
|
* libpcap developers.
|
|
*/
|
|
close(fd);
|
|
return (PCAP_ERROR_IFACE_NOT_UP);
|
|
|
|
default:
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCSETIF: %s", p->opt.device);
|
|
close(fd);
|
|
return (PCAP_ERROR);
|
|
}
|
|
}
|
|
|
|
/*
|
|
* We know the default link type -- now determine all the DLTs
|
|
* this interface supports. If this fails with EINVAL, it's
|
|
* not fatal; we just don't get to use the feature later.
|
|
* (We don't care about DLT_DOCSIS, so we pass DLT_NULL
|
|
* as the default DLT for this adapter.)
|
|
*/
|
|
if (get_dlt_list(fd, DLT_NULL, &bdl, p->errbuf) == PCAP_ERROR) {
|
|
close(fd);
|
|
return (PCAP_ERROR);
|
|
}
|
|
if (find_802_11(&bdl) != -1) {
|
|
/*
|
|
* We have an 802.11 DLT, so we can set monitor mode.
|
|
*/
|
|
free(bdl.bfl_list);
|
|
close(fd);
|
|
return (1);
|
|
}
|
|
free(bdl.bfl_list);
|
|
close(fd);
|
|
#endif /* BIOCGDLTLIST */
|
|
return (0);
|
|
}
|
|
#elif defined(HAVE_BSD_IEEE80211)
|
|
static int
|
|
pcap_can_set_rfmon_bpf(pcap_t *p)
|
|
{
|
|
int ret;
|
|
|
|
ret = monitor_mode(p, 0);
|
|
if (ret == PCAP_ERROR_RFMON_NOTSUP)
|
|
return (0); /* not an error, just a "can't do" */
|
|
if (ret == 0)
|
|
return (1); /* success */
|
|
return (ret);
|
|
}
|
|
#else
|
|
static int
|
|
pcap_can_set_rfmon_bpf(pcap_t *p _U_)
|
|
{
|
|
return (0);
|
|
}
|
|
#endif
|
|
|
|
static int
|
|
pcap_stats_bpf(pcap_t *p, struct pcap_stat *ps)
|
|
{
|
|
struct bpf_stat s;
|
|
|
|
/*
|
|
* "ps_recv" counts packets handed to the filter, not packets
|
|
* that passed the filter. This includes packets later dropped
|
|
* because we ran out of buffer space.
|
|
*
|
|
* "ps_drop" counts packets dropped inside the BPF device
|
|
* because we ran out of buffer space. It doesn't count
|
|
* packets dropped by the interface driver. It counts
|
|
* only packets that passed the filter.
|
|
*
|
|
* Both statistics include packets not yet read from the kernel
|
|
* by libpcap, and thus not yet seen by the application.
|
|
*/
|
|
if (ioctl(p->fd, BIOCGSTATS, (caddr_t)&s) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCGSTATS");
|
|
return (PCAP_ERROR);
|
|
}
|
|
|
|
ps->ps_recv = s.bs_recv;
|
|
ps->ps_drop = s.bs_drop;
|
|
ps->ps_ifdrop = 0;
|
|
return (0);
|
|
}
|
|
|
|
static int
|
|
pcap_read_bpf(pcap_t *p, int cnt, pcap_handler callback, u_char *user)
|
|
{
|
|
struct pcap_bpf *pb = p->priv;
|
|
int cc;
|
|
int n = 0;
|
|
register u_char *bp, *ep;
|
|
u_char *datap;
|
|
#ifdef PCAP_FDDIPAD
|
|
register u_int pad;
|
|
#endif
|
|
#ifdef HAVE_ZEROCOPY_BPF
|
|
int i;
|
|
#endif
|
|
|
|
again:
|
|
/*
|
|
* Has "pcap_breakloop()" been called?
|
|
*/
|
|
if (p->break_loop) {
|
|
/*
|
|
* Yes - clear the flag that indicates that it
|
|
* has, and return PCAP_ERROR_BREAK to indicate
|
|
* that we were told to break out of the loop.
|
|
*/
|
|
p->break_loop = 0;
|
|
return (PCAP_ERROR_BREAK);
|
|
}
|
|
cc = p->cc;
|
|
if (p->cc == 0) {
|
|
/*
|
|
* When reading without zero-copy from a file descriptor, we
|
|
* use a single buffer and return a length of data in the
|
|
* buffer. With zero-copy, we update the p->buffer pointer
|
|
* to point at whatever underlying buffer contains the next
|
|
* data and update cc to reflect the data found in the
|
|
* buffer.
|
|
*/
|
|
#ifdef HAVE_ZEROCOPY_BPF
|
|
if (pb->zerocopy) {
|
|
if (p->buffer != NULL)
|
|
pcap_ack_zbuf(p);
|
|
i = pcap_next_zbuf(p, &cc);
|
|
if (i == 0)
|
|
goto again;
|
|
if (i < 0)
|
|
return (PCAP_ERROR);
|
|
} else
|
|
#endif
|
|
{
|
|
cc = read(p->fd, p->buffer, p->bufsize);
|
|
}
|
|
if (cc < 0) {
|
|
/* Don't choke when we get ptraced */
|
|
switch (errno) {
|
|
|
|
case EINTR:
|
|
goto again;
|
|
|
|
#ifdef _AIX
|
|
case EFAULT:
|
|
/*
|
|
* Sigh. More AIX wonderfulness.
|
|
*
|
|
* For some unknown reason the uiomove()
|
|
* operation in the bpf kernel extension
|
|
* used to copy the buffer into user
|
|
* space sometimes returns EFAULT. I have
|
|
* no idea why this is the case given that
|
|
* a kernel debugger shows the user buffer
|
|
* is correct. This problem appears to
|
|
* be mostly mitigated by the memset of
|
|
* the buffer before it is first used.
|
|
* Very strange.... Shaun Clowes
|
|
*
|
|
* In any case this means that we shouldn't
|
|
* treat EFAULT as a fatal error; as we
|
|
* don't have an API for returning
|
|
* a "some packets were dropped since
|
|
* the last packet you saw" indication,
|
|
* we just ignore EFAULT and keep reading.
|
|
*/
|
|
goto again;
|
|
#endif
|
|
|
|
case EWOULDBLOCK:
|
|
return (0);
|
|
|
|
case ENXIO: /* FreeBSD, DragonFly BSD, and Darwin */
|
|
case EIO: /* OpenBSD */
|
|
/* NetBSD appears not to return an error in this case */
|
|
/*
|
|
* The device on which we're capturing
|
|
* went away.
|
|
*
|
|
* XXX - we should really return
|
|
* an appropriate error for that,
|
|
* but pcap_dispatch() etc. aren't
|
|
* documented as having error returns
|
|
* other than PCAP_ERROR or PCAP_ERROR_BREAK.
|
|
*/
|
|
pcap_snprintf(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
"The interface disappeared");
|
|
return (PCAP_ERROR);
|
|
|
|
#if defined(sun) && !defined(BSD) && !defined(__svr4__) && !defined(__SVR4)
|
|
/*
|
|
* Due to a SunOS bug, after 2^31 bytes, the kernel
|
|
* file offset overflows and read fails with EINVAL.
|
|
* The lseek() to 0 will fix things.
|
|
*/
|
|
case EINVAL:
|
|
if (lseek(p->fd, 0L, SEEK_CUR) +
|
|
p->bufsize < 0) {
|
|
(void)lseek(p->fd, 0L, SEEK_SET);
|
|
goto again;
|
|
}
|
|
/* fall through */
|
|
#endif
|
|
}
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "read");
|
|
return (PCAP_ERROR);
|
|
}
|
|
bp = (u_char *)p->buffer;
|
|
} else
|
|
bp = p->bp;
|
|
|
|
/*
|
|
* Loop through each packet.
|
|
*/
|
|
#ifdef BIOCSTSTAMP
|
|
#define bhp ((struct bpf_xhdr *)bp)
|
|
#else
|
|
#define bhp ((struct bpf_hdr *)bp)
|
|
#endif
|
|
ep = bp + cc;
|
|
#ifdef PCAP_FDDIPAD
|
|
pad = p->fddipad;
|
|
#endif
|
|
while (bp < ep) {
|
|
register u_int caplen, hdrlen;
|
|
|
|
/*
|
|
* Has "pcap_breakloop()" been called?
|
|
* If so, return immediately - if we haven't read any
|
|
* packets, clear the flag and return PCAP_ERROR_BREAK
|
|
* to indicate that we were told to break out of the loop,
|
|
* otherwise leave the flag set, so that the *next* call
|
|
* will break out of the loop without having read any
|
|
* packets, and return the number of packets we've
|
|
* processed so far.
|
|
*/
|
|
if (p->break_loop) {
|
|
p->bp = bp;
|
|
p->cc = ep - bp;
|
|
/*
|
|
* ep is set based on the return value of read(),
|
|
* but read() from a BPF device doesn't necessarily
|
|
* return a value that's a multiple of the alignment
|
|
* value for BPF_WORDALIGN(). However, whenever we
|
|
* increment bp, we round up the increment value by
|
|
* a value rounded up by BPF_WORDALIGN(), so we
|
|
* could increment bp past ep after processing the
|
|
* last packet in the buffer.
|
|
*
|
|
* We treat ep < bp as an indication that this
|
|
* happened, and just set p->cc to 0.
|
|
*/
|
|
if (p->cc < 0)
|
|
p->cc = 0;
|
|
if (n == 0) {
|
|
p->break_loop = 0;
|
|
return (PCAP_ERROR_BREAK);
|
|
} else
|
|
return (n);
|
|
}
|
|
|
|
caplen = bhp->bh_caplen;
|
|
hdrlen = bhp->bh_hdrlen;
|
|
datap = bp + hdrlen;
|
|
/*
|
|
* Short-circuit evaluation: if using BPF filter
|
|
* in kernel, no need to do it now - we already know
|
|
* the packet passed the filter.
|
|
*
|
|
#ifdef PCAP_FDDIPAD
|
|
* Note: the filter code was generated assuming
|
|
* that p->fddipad was the amount of padding
|
|
* before the header, as that's what's required
|
|
* in the kernel, so we run the filter before
|
|
* skipping that padding.
|
|
#endif
|
|
*/
|
|
if (pb->filtering_in_kernel ||
|
|
bpf_filter(p->fcode.bf_insns, datap, bhp->bh_datalen, caplen)) {
|
|
struct pcap_pkthdr pkthdr;
|
|
#ifdef BIOCSTSTAMP
|
|
struct bintime bt;
|
|
|
|
bt.sec = bhp->bh_tstamp.bt_sec;
|
|
bt.frac = bhp->bh_tstamp.bt_frac;
|
|
if (p->opt.tstamp_precision == PCAP_TSTAMP_PRECISION_NANO) {
|
|
struct timespec ts;
|
|
|
|
bintime2timespec(&bt, &ts);
|
|
pkthdr.ts.tv_sec = ts.tv_sec;
|
|
pkthdr.ts.tv_usec = ts.tv_nsec;
|
|
} else {
|
|
struct timeval tv;
|
|
|
|
bintime2timeval(&bt, &tv);
|
|
pkthdr.ts.tv_sec = tv.tv_sec;
|
|
pkthdr.ts.tv_usec = tv.tv_usec;
|
|
}
|
|
#else
|
|
pkthdr.ts.tv_sec = bhp->bh_tstamp.tv_sec;
|
|
#ifdef _AIX
|
|
/*
|
|
* AIX's BPF returns seconds/nanoseconds time
|
|
* stamps, not seconds/microseconds time stamps.
|
|
*/
|
|
pkthdr.ts.tv_usec = bhp->bh_tstamp.tv_usec/1000;
|
|
#else
|
|
pkthdr.ts.tv_usec = bhp->bh_tstamp.tv_usec;
|
|
#endif
|
|
#endif /* BIOCSTSTAMP */
|
|
#ifdef PCAP_FDDIPAD
|
|
if (caplen > pad)
|
|
pkthdr.caplen = caplen - pad;
|
|
else
|
|
pkthdr.caplen = 0;
|
|
if (bhp->bh_datalen > pad)
|
|
pkthdr.len = bhp->bh_datalen - pad;
|
|
else
|
|
pkthdr.len = 0;
|
|
datap += pad;
|
|
#else
|
|
pkthdr.caplen = caplen;
|
|
pkthdr.len = bhp->bh_datalen;
|
|
#endif
|
|
(*callback)(user, &pkthdr, datap);
|
|
bp += BPF_WORDALIGN(caplen + hdrlen);
|
|
if (++n >= cnt && !PACKET_COUNT_IS_UNLIMITED(cnt)) {
|
|
p->bp = bp;
|
|
p->cc = ep - bp;
|
|
/*
|
|
* See comment above about p->cc < 0.
|
|
*/
|
|
if (p->cc < 0)
|
|
p->cc = 0;
|
|
return (n);
|
|
}
|
|
} else {
|
|
/*
|
|
* Skip this packet.
|
|
*/
|
|
bp += BPF_WORDALIGN(caplen + hdrlen);
|
|
}
|
|
}
|
|
#undef bhp
|
|
p->cc = 0;
|
|
return (n);
|
|
}
|
|
|
|
static int
|
|
pcap_inject_bpf(pcap_t *p, const void *buf, size_t size)
|
|
{
|
|
int ret;
|
|
|
|
ret = write(p->fd, buf, size);
|
|
#ifdef __APPLE__
|
|
if (ret == -1 && errno == EAFNOSUPPORT) {
|
|
/*
|
|
* In some versions of macOS, there's a bug wherein setting
|
|
* the BIOCSHDRCMPLT flag causes writes to fail; see, for
|
|
* example:
|
|
*
|
|
* http://cerberus.sourcefire.com/~jeff/archives/patches/macosx/BIOCSHDRCMPLT-10.3.3.patch
|
|
*
|
|
* So, if, on macOS, we get EAFNOSUPPORT from the write, we
|
|
* assume it's due to that bug, and turn off that flag
|
|
* and try again. If we succeed, it either means that
|
|
* somebody applied the fix from that URL, or other patches
|
|
* for that bug from
|
|
*
|
|
* http://cerberus.sourcefire.com/~jeff/archives/patches/macosx/
|
|
*
|
|
* and are running a Darwin kernel with those fixes, or
|
|
* that Apple fixed the problem in some macOS release.
|
|
*/
|
|
u_int spoof_eth_src = 0;
|
|
|
|
if (ioctl(p->fd, BIOCSHDRCMPLT, &spoof_eth_src) == -1) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "send: can't turn off BIOCSHDRCMPLT");
|
|
return (PCAP_ERROR);
|
|
}
|
|
|
|
/*
|
|
* Now try the write again.
|
|
*/
|
|
ret = write(p->fd, buf, size);
|
|
}
|
|
#endif /* __APPLE__ */
|
|
if (ret == -1) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "send");
|
|
return (PCAP_ERROR);
|
|
}
|
|
return (ret);
|
|
}
|
|
|
|
#ifdef _AIX
|
|
static int
|
|
bpf_odminit(char *errbuf)
|
|
{
|
|
char *errstr;
|
|
|
|
if (odm_initialize() == -1) {
|
|
if (odm_err_msg(odmerrno, &errstr) == -1)
|
|
errstr = "Unknown error";
|
|
pcap_snprintf(errbuf, PCAP_ERRBUF_SIZE,
|
|
"bpf_load: odm_initialize failed: %s",
|
|
errstr);
|
|
return (PCAP_ERROR);
|
|
}
|
|
|
|
if ((odmlockid = odm_lock("/etc/objrepos/config_lock", ODM_WAIT)) == -1) {
|
|
if (odm_err_msg(odmerrno, &errstr) == -1)
|
|
errstr = "Unknown error";
|
|
pcap_snprintf(errbuf, PCAP_ERRBUF_SIZE,
|
|
"bpf_load: odm_lock of /etc/objrepos/config_lock failed: %s",
|
|
errstr);
|
|
(void)odm_terminate();
|
|
return (PCAP_ERROR);
|
|
}
|
|
|
|
return (0);
|
|
}
|
|
|
|
static int
|
|
bpf_odmcleanup(char *errbuf)
|
|
{
|
|
char *errstr;
|
|
|
|
if (odm_unlock(odmlockid) == -1) {
|
|
if (errbuf != NULL) {
|
|
if (odm_err_msg(odmerrno, &errstr) == -1)
|
|
errstr = "Unknown error";
|
|
pcap_snprintf(errbuf, PCAP_ERRBUF_SIZE,
|
|
"bpf_load: odm_unlock failed: %s",
|
|
errstr);
|
|
}
|
|
return (PCAP_ERROR);
|
|
}
|
|
|
|
if (odm_terminate() == -1) {
|
|
if (errbuf != NULL) {
|
|
if (odm_err_msg(odmerrno, &errstr) == -1)
|
|
errstr = "Unknown error";
|
|
pcap_snprintf(errbuf, PCAP_ERRBUF_SIZE,
|
|
"bpf_load: odm_terminate failed: %s",
|
|
errstr);
|
|
}
|
|
return (PCAP_ERROR);
|
|
}
|
|
|
|
return (0);
|
|
}
|
|
|
|
static int
|
|
bpf_load(char *errbuf)
|
|
{
|
|
long major;
|
|
int *minors;
|
|
int numminors, i, rc;
|
|
char buf[1024];
|
|
struct stat sbuf;
|
|
struct bpf_config cfg_bpf;
|
|
struct cfg_load cfg_ld;
|
|
struct cfg_kmod cfg_km;
|
|
|
|
/*
|
|
* This is very very close to what happens in the real implementation
|
|
* but I've fixed some (unlikely) bug situations.
|
|
*/
|
|
if (bpfloadedflag)
|
|
return (0);
|
|
|
|
if (bpf_odminit(errbuf) == PCAP_ERROR)
|
|
return (PCAP_ERROR);
|
|
|
|
major = genmajor(BPF_NAME);
|
|
if (major == -1) {
|
|
pcap_fmt_errmsg_for_errno(errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "bpf_load: genmajor failed");
|
|
(void)bpf_odmcleanup(NULL);
|
|
return (PCAP_ERROR);
|
|
}
|
|
|
|
minors = getminor(major, &numminors, BPF_NAME);
|
|
if (!minors) {
|
|
minors = genminor("bpf", major, 0, BPF_MINORS, 1, 1);
|
|
if (!minors) {
|
|
pcap_fmt_errmsg_for_errno(errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "bpf_load: genminor failed");
|
|
(void)bpf_odmcleanup(NULL);
|
|
return (PCAP_ERROR);
|
|
}
|
|
}
|
|
|
|
if (bpf_odmcleanup(errbuf) == PCAP_ERROR)
|
|
return (PCAP_ERROR);
|
|
|
|
rc = stat(BPF_NODE "0", &sbuf);
|
|
if (rc == -1 && errno != ENOENT) {
|
|
pcap_fmt_errmsg_for_errno(errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "bpf_load: can't stat %s", BPF_NODE "0");
|
|
return (PCAP_ERROR);
|
|
}
|
|
|
|
if (rc == -1 || getmajor(sbuf.st_rdev) != major) {
|
|
for (i = 0; i < BPF_MINORS; i++) {
|
|
pcap_snprintf(buf, sizeof(buf), "%s%d", BPF_NODE, i);
|
|
unlink(buf);
|
|
if (mknod(buf, S_IRUSR | S_IFCHR, domakedev(major, i)) == -1) {
|
|
pcap_fmt_errmsg_for_errno(errbuf,
|
|
PCAP_ERRBUF_SIZE, errno,
|
|
"bpf_load: can't mknod %s", buf);
|
|
return (PCAP_ERROR);
|
|
}
|
|
}
|
|
}
|
|
|
|
/* Check if the driver is loaded */
|
|
memset(&cfg_ld, 0x0, sizeof(cfg_ld));
|
|
pcap_snprintf(buf, sizeof(buf), "%s/%s", DRIVER_PATH, BPF_NAME);
|
|
cfg_ld.path = buf;
|
|
if ((sysconfig(SYS_QUERYLOAD, (void *)&cfg_ld, sizeof(cfg_ld)) == -1) ||
|
|
(cfg_ld.kmid == 0)) {
|
|
/* Driver isn't loaded, load it now */
|
|
if (sysconfig(SYS_SINGLELOAD, (void *)&cfg_ld, sizeof(cfg_ld)) == -1) {
|
|
pcap_fmt_errmsg_for_errno(errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "bpf_load: could not load driver");
|
|
return (PCAP_ERROR);
|
|
}
|
|
}
|
|
|
|
/* Configure the driver */
|
|
cfg_km.cmd = CFG_INIT;
|
|
cfg_km.kmid = cfg_ld.kmid;
|
|
cfg_km.mdilen = sizeof(cfg_bpf);
|
|
cfg_km.mdiptr = (void *)&cfg_bpf;
|
|
for (i = 0; i < BPF_MINORS; i++) {
|
|
cfg_bpf.devno = domakedev(major, i);
|
|
if (sysconfig(SYS_CFGKMOD, (void *)&cfg_km, sizeof(cfg_km)) == -1) {
|
|
pcap_fmt_errmsg_for_errno(errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "bpf_load: could not configure driver");
|
|
return (PCAP_ERROR);
|
|
}
|
|
}
|
|
|
|
bpfloadedflag = 1;
|
|
|
|
return (0);
|
|
}
|
|
#endif
|
|
|
|
/*
|
|
* Undo any operations done when opening the device when necessary.
|
|
*/
|
|
static void
|
|
pcap_cleanup_bpf(pcap_t *p)
|
|
{
|
|
struct pcap_bpf *pb = p->priv;
|
|
#ifdef HAVE_BSD_IEEE80211
|
|
int sock;
|
|
struct ifmediareq req;
|
|
struct ifreq ifr;
|
|
#endif
|
|
|
|
if (pb->must_do_on_close != 0) {
|
|
/*
|
|
* There's something we have to do when closing this
|
|
* pcap_t.
|
|
*/
|
|
#ifdef HAVE_BSD_IEEE80211
|
|
if (pb->must_do_on_close & MUST_CLEAR_RFMON) {
|
|
/*
|
|
* We put the interface into rfmon mode;
|
|
* take it out of rfmon mode.
|
|
*
|
|
* XXX - if somebody else wants it in rfmon
|
|
* mode, this code cannot know that, so it'll take
|
|
* it out of rfmon mode.
|
|
*/
|
|
sock = socket(AF_INET, SOCK_DGRAM, 0);
|
|
if (sock == -1) {
|
|
fprintf(stderr,
|
|
"Can't restore interface flags (socket() failed: %s).\n"
|
|
"Please adjust manually.\n",
|
|
strerror(errno));
|
|
} else {
|
|
memset(&req, 0, sizeof(req));
|
|
strncpy(req.ifm_name, pb->device,
|
|
sizeof(req.ifm_name));
|
|
if (ioctl(sock, SIOCGIFMEDIA, &req) < 0) {
|
|
fprintf(stderr,
|
|
"Can't restore interface flags (SIOCGIFMEDIA failed: %s).\n"
|
|
"Please adjust manually.\n",
|
|
strerror(errno));
|
|
} else {
|
|
if (req.ifm_current & IFM_IEEE80211_MONITOR) {
|
|
/*
|
|
* Rfmon mode is currently on;
|
|
* turn it off.
|
|
*/
|
|
memset(&ifr, 0, sizeof(ifr));
|
|
(void)strncpy(ifr.ifr_name,
|
|
pb->device,
|
|
sizeof(ifr.ifr_name));
|
|
ifr.ifr_media =
|
|
req.ifm_current & ~IFM_IEEE80211_MONITOR;
|
|
if (ioctl(sock, SIOCSIFMEDIA,
|
|
&ifr) == -1) {
|
|
fprintf(stderr,
|
|
"Can't restore interface flags (SIOCSIFMEDIA failed: %s).\n"
|
|
"Please adjust manually.\n",
|
|
strerror(errno));
|
|
}
|
|
}
|
|
}
|
|
close(sock);
|
|
}
|
|
}
|
|
#endif /* HAVE_BSD_IEEE80211 */
|
|
|
|
#if defined(__FreeBSD__) && defined(SIOCIFCREATE2)
|
|
/*
|
|
* Attempt to destroy the usbusN interface that we created.
|
|
*/
|
|
if (pb->must_do_on_close & MUST_DESTROY_USBUS) {
|
|
if (if_nametoindex(pb->device) > 0) {
|
|
int s;
|
|
|
|
s = socket(AF_LOCAL, SOCK_DGRAM, 0);
|
|
if (s >= 0) {
|
|
pcap_strlcpy(ifr.ifr_name, pb->device,
|
|
sizeof(ifr.ifr_name));
|
|
ioctl(s, SIOCIFDESTROY, &ifr);
|
|
close(s);
|
|
}
|
|
}
|
|
}
|
|
#endif /* defined(__FreeBSD__) && defined(SIOCIFCREATE2) */
|
|
/*
|
|
* Take this pcap out of the list of pcaps for which we
|
|
* have to take the interface out of some mode.
|
|
*/
|
|
pcap_remove_from_pcaps_to_close(p);
|
|
pb->must_do_on_close = 0;
|
|
}
|
|
|
|
#ifdef HAVE_ZEROCOPY_BPF
|
|
if (pb->zerocopy) {
|
|
/*
|
|
* Delete the mappings. Note that p->buffer gets
|
|
* initialized to one of the mmapped regions in
|
|
* this case, so do not try and free it directly;
|
|
* null it out so that pcap_cleanup_live_common()
|
|
* doesn't try to free it.
|
|
*/
|
|
if (pb->zbuf1 != MAP_FAILED && pb->zbuf1 != NULL)
|
|
(void) munmap(pb->zbuf1, pb->zbufsize);
|
|
if (pb->zbuf2 != MAP_FAILED && pb->zbuf2 != NULL)
|
|
(void) munmap(pb->zbuf2, pb->zbufsize);
|
|
p->buffer = NULL;
|
|
}
|
|
#endif
|
|
if (pb->device != NULL) {
|
|
free(pb->device);
|
|
pb->device = NULL;
|
|
}
|
|
pcap_cleanup_live_common(p);
|
|
}
|
|
|
|
static int
|
|
check_setif_failure(pcap_t *p, int error)
|
|
{
|
|
#ifdef __APPLE__
|
|
int fd;
|
|
struct ifreq ifr;
|
|
int err;
|
|
#endif
|
|
|
|
if (error == ENXIO) {
|
|
/*
|
|
* No such device exists.
|
|
*/
|
|
#ifdef __APPLE__
|
|
if (p->opt.rfmon && strncmp(p->opt.device, "wlt", 3) == 0) {
|
|
/*
|
|
* Monitor mode was requested, and we're trying
|
|
* to open a "wltN" device. Assume that this
|
|
* is 10.4 and that we were asked to open an
|
|
* "enN" device; if that device exists, return
|
|
* "monitor mode not supported on the device".
|
|
*/
|
|
fd = socket(AF_INET, SOCK_DGRAM, 0);
|
|
if (fd != -1) {
|
|
pcap_strlcpy(ifr.ifr_name, "en",
|
|
sizeof(ifr.ifr_name));
|
|
pcap_strlcat(ifr.ifr_name, p->opt.device + 3,
|
|
sizeof(ifr.ifr_name));
|
|
if (ioctl(fd, SIOCGIFFLAGS, (char *)&ifr) < 0) {
|
|
/*
|
|
* We assume this failed because
|
|
* the underlying device doesn't
|
|
* exist.
|
|
*/
|
|
err = PCAP_ERROR_NO_SUCH_DEVICE;
|
|
pcap_fmt_errmsg_for_errno(p->errbuf,
|
|
PCAP_ERRBUF_SIZE, errno,
|
|
"SIOCGIFFLAGS on %s failed",
|
|
ifr.ifr_name);
|
|
} else {
|
|
/*
|
|
* The underlying "enN" device
|
|
* exists, but there's no
|
|
* corresponding "wltN" device;
|
|
* that means that the "enN"
|
|
* device doesn't support
|
|
* monitor mode, probably because
|
|
* it's an Ethernet device rather
|
|
* than a wireless device.
|
|
*/
|
|
err = PCAP_ERROR_RFMON_NOTSUP;
|
|
}
|
|
close(fd);
|
|
} else {
|
|
/*
|
|
* We can't find out whether there's
|
|
* an underlying "enN" device, so
|
|
* just report "no such device".
|
|
*/
|
|
err = PCAP_ERROR_NO_SUCH_DEVICE;
|
|
pcap_fmt_errmsg_for_errno(p->errbuf,
|
|
errno, PCAP_ERRBUF_SIZE,
|
|
"socket() failed");
|
|
}
|
|
return (err);
|
|
}
|
|
#endif
|
|
/*
|
|
* No such device.
|
|
*/
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCSETIF failed");
|
|
return (PCAP_ERROR_NO_SUCH_DEVICE);
|
|
} else if (errno == ENETDOWN) {
|
|
/*
|
|
* Return a "network down" indication, so that
|
|
* the application can report that rather than
|
|
* saying we had a mysterious failure and
|
|
* suggest that they report a problem to the
|
|
* libpcap developers.
|
|
*/
|
|
return (PCAP_ERROR_IFACE_NOT_UP);
|
|
} else {
|
|
/*
|
|
* Some other error; fill in the error string, and
|
|
* return PCAP_ERROR.
|
|
*/
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCSETIF: %s", p->opt.device);
|
|
return (PCAP_ERROR);
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Default capture buffer size.
|
|
* 32K isn't very much for modern machines with fast networks; we
|
|
* pick .5M, as that's the maximum on at least some systems with BPF.
|
|
*
|
|
* However, on AIX 3.5, the larger buffer sized caused unrecoverable
|
|
* read failures under stress, so we leave it as 32K; yet another
|
|
* place where AIX's BPF is broken.
|
|
*/
|
|
#ifdef _AIX
|
|
#define DEFAULT_BUFSIZE 32768
|
|
#else
|
|
#define DEFAULT_BUFSIZE 524288
|
|
#endif
|
|
|
|
static int
|
|
pcap_activate_bpf(pcap_t *p)
|
|
{
|
|
struct pcap_bpf *pb = p->priv;
|
|
int status = 0;
|
|
#ifdef HAVE_BSD_IEEE80211
|
|
int retv;
|
|
#endif
|
|
int fd;
|
|
#ifdef LIFNAMSIZ
|
|
char *zonesep;
|
|
struct lifreq ifr;
|
|
char *ifrname = ifr.lifr_name;
|
|
const size_t ifnamsiz = sizeof(ifr.lifr_name);
|
|
#else
|
|
struct ifreq ifr;
|
|
char *ifrname = ifr.ifr_name;
|
|
const size_t ifnamsiz = sizeof(ifr.ifr_name);
|
|
#endif
|
|
struct bpf_version bv;
|
|
#ifdef __APPLE__
|
|
int sockfd;
|
|
char *wltdev = NULL;
|
|
#endif
|
|
#ifdef BIOCGDLTLIST
|
|
struct bpf_dltlist bdl;
|
|
#if defined(__APPLE__) || defined(HAVE_BSD_IEEE80211)
|
|
int new_dlt;
|
|
#endif
|
|
#endif /* BIOCGDLTLIST */
|
|
#if defined(BIOCGHDRCMPLT) && defined(BIOCSHDRCMPLT)
|
|
u_int spoof_eth_src = 1;
|
|
#endif
|
|
u_int v;
|
|
struct bpf_insn total_insn;
|
|
struct bpf_program total_prog;
|
|
struct utsname osinfo;
|
|
int have_osinfo = 0;
|
|
#ifdef HAVE_ZEROCOPY_BPF
|
|
struct bpf_zbuf bz;
|
|
u_int bufmode, zbufmax;
|
|
#endif
|
|
|
|
fd = bpf_open(p->errbuf);
|
|
if (fd < 0) {
|
|
status = fd;
|
|
goto bad;
|
|
}
|
|
|
|
p->fd = fd;
|
|
|
|
if (ioctl(fd, BIOCVERSION, (caddr_t)&bv) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCVERSION");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
if (bv.bv_major != BPF_MAJOR_VERSION ||
|
|
bv.bv_minor < BPF_MINOR_VERSION) {
|
|
pcap_snprintf(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
"kernel bpf filter out of date");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
|
|
/*
|
|
* Turn a negative snapshot value (invalid), a snapshot value of
|
|
* 0 (unspecified), or a value bigger than the normal maximum
|
|
* value, into the maximum allowed value.
|
|
*
|
|
* If some application really *needs* a bigger snapshot
|
|
* length, we should just increase MAXIMUM_SNAPLEN.
|
|
*/
|
|
if (p->snapshot <= 0 || p->snapshot > MAXIMUM_SNAPLEN)
|
|
p->snapshot = MAXIMUM_SNAPLEN;
|
|
|
|
#if defined(LIFNAMSIZ) && defined(ZONENAME_MAX) && defined(lifr_zoneid)
|
|
/*
|
|
* Retrieve the zoneid of the zone we are currently executing in.
|
|
*/
|
|
if ((ifr.lifr_zoneid = getzoneid()) == -1) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "getzoneid()");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
/*
|
|
* Check if the given source datalink name has a '/' separated
|
|
* zonename prefix string. The zonename prefixed source datalink can
|
|
* be used by pcap consumers in the Solaris global zone to capture
|
|
* traffic on datalinks in non-global zones. Non-global zones
|
|
* do not have access to datalinks outside of their own namespace.
|
|
*/
|
|
if ((zonesep = strchr(p->opt.device, '/')) != NULL) {
|
|
char path_zname[ZONENAME_MAX];
|
|
int znamelen;
|
|
char *lnamep;
|
|
|
|
if (ifr.lifr_zoneid != GLOBAL_ZONEID) {
|
|
pcap_snprintf(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
"zonename/linkname only valid in global zone.");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
znamelen = zonesep - p->opt.device;
|
|
(void) pcap_strlcpy(path_zname, p->opt.device, znamelen + 1);
|
|
ifr.lifr_zoneid = getzoneidbyname(path_zname);
|
|
if (ifr.lifr_zoneid == -1) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "getzoneidbyname(%s)", path_zname);
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
lnamep = strdup(zonesep + 1);
|
|
if (lnamep == NULL) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "strdup");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
free(p->opt.device);
|
|
p->opt.device = lnamep;
|
|
}
|
|
#endif
|
|
|
|
pb->device = strdup(p->opt.device);
|
|
if (pb->device == NULL) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "strdup");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
|
|
/*
|
|
* Attempt to find out the version of the OS on which we're running.
|
|
*/
|
|
if (uname(&osinfo) == 0)
|
|
have_osinfo = 1;
|
|
|
|
#ifdef __APPLE__
|
|
/*
|
|
* See comment in pcap_can_set_rfmon_bpf() for an explanation
|
|
* of why we check the version number.
|
|
*/
|
|
if (p->opt.rfmon) {
|
|
if (have_osinfo) {
|
|
/*
|
|
* We assume osinfo.sysname is "Darwin", because
|
|
* __APPLE__ is defined. We just check the version.
|
|
*/
|
|
if (osinfo.release[0] < '8' &&
|
|
osinfo.release[1] == '.') {
|
|
/*
|
|
* 10.3 (Darwin 7.x) or earlier.
|
|
*/
|
|
status = PCAP_ERROR_RFMON_NOTSUP;
|
|
goto bad;
|
|
}
|
|
if (osinfo.release[0] == '8' &&
|
|
osinfo.release[1] == '.') {
|
|
/*
|
|
* 10.4 (Darwin 8.x). s/en/wlt/
|
|
*/
|
|
if (strncmp(p->opt.device, "en", 2) != 0) {
|
|
/*
|
|
* Not an enN device; check
|
|
* whether the device even exists.
|
|
*/
|
|
sockfd = socket(AF_INET, SOCK_DGRAM, 0);
|
|
if (sockfd != -1) {
|
|
pcap_strlcpy(ifrname,
|
|
p->opt.device, ifnamsiz);
|
|
if (ioctl(sockfd, SIOCGIFFLAGS,
|
|
(char *)&ifr) < 0) {
|
|
/*
|
|
* We assume this
|
|
* failed because
|
|
* the underlying
|
|
* device doesn't
|
|
* exist.
|
|
*/
|
|
status = PCAP_ERROR_NO_SUCH_DEVICE;
|
|
pcap_fmt_errmsg_for_errno(p->errbuf,
|
|
PCAP_ERRBUF_SIZE,
|
|
errno,
|
|
"SIOCGIFFLAGS failed");
|
|
} else
|
|
status = PCAP_ERROR_RFMON_NOTSUP;
|
|
close(sockfd);
|
|
} else {
|
|
/*
|
|
* We can't find out whether
|
|
* the device exists, so just
|
|
* report "no such device".
|
|
*/
|
|
status = PCAP_ERROR_NO_SUCH_DEVICE;
|
|
pcap_fmt_errmsg_for_errno(p->errbuf,
|
|
PCAP_ERRBUF_SIZE, errno,
|
|
"socket() failed");
|
|
}
|
|
goto bad;
|
|
}
|
|
wltdev = malloc(strlen(p->opt.device) + 2);
|
|
if (wltdev == NULL) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf,
|
|
PCAP_ERRBUF_SIZE, errno,
|
|
"malloc");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
strcpy(wltdev, "wlt");
|
|
strcat(wltdev, p->opt.device + 2);
|
|
free(p->opt.device);
|
|
p->opt.device = wltdev;
|
|
}
|
|
/*
|
|
* Everything else is 10.5 or later; for those,
|
|
* we just open the enN device, and set the DLT.
|
|
*/
|
|
}
|
|
}
|
|
#endif /* __APPLE__ */
|
|
|
|
/*
|
|
* If this is FreeBSD, and the device name begins with "usbus",
|
|
* try to create the interface if it's not available.
|
|
*/
|
|
#if defined(__FreeBSD__) && defined(SIOCIFCREATE2)
|
|
if (strncmp(p->opt.device, usbus_prefix, USBUS_PREFIX_LEN) == 0) {
|
|
/*
|
|
* Do we already have an interface with that name?
|
|
*/
|
|
if (if_nametoindex(p->opt.device) == 0) {
|
|
/*
|
|
* No. We need to create it, and, if we
|
|
* succeed, remember that we should destroy
|
|
* it when the pcap_t is closed.
|
|
*/
|
|
int s;
|
|
|
|
/*
|
|
* Open a socket to use for ioctls to
|
|
* create the interface.
|
|
*/
|
|
s = socket(AF_LOCAL, SOCK_DGRAM, 0);
|
|
if (s < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf,
|
|
PCAP_ERRBUF_SIZE, errno,
|
|
"Can't open socket");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
|
|
/*
|
|
* If we haven't already done so, arrange to have
|
|
* "pcap_close_all()" called when we exit.
|
|
*/
|
|
if (!pcap_do_addexit(p)) {
|
|
/*
|
|
* "atexit()" failed; don't create the
|
|
* interface, just give up.
|
|
*/
|
|
pcap_snprintf(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
"atexit failed");
|
|
close(s);
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
|
|
/*
|
|
* Create the interface.
|
|
*/
|
|
pcap_strlcpy(ifr.ifr_name, p->opt.device, sizeof(ifr.ifr_name));
|
|
if (ioctl(s, SIOCIFCREATE2, &ifr) < 0) {
|
|
if (errno == EINVAL) {
|
|
pcap_snprintf(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
"Invalid USB bus interface %s",
|
|
p->opt.device);
|
|
} else {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf,
|
|
PCAP_ERRBUF_SIZE, errno,
|
|
"Can't create interface for %s",
|
|
p->opt.device);
|
|
}
|
|
close(s);
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
|
|
/*
|
|
* Make sure we clean this up when we close.
|
|
*/
|
|
pb->must_do_on_close |= MUST_DESTROY_USBUS;
|
|
|
|
/*
|
|
* Add this to the list of pcaps to close when we exit.
|
|
*/
|
|
pcap_add_to_pcaps_to_close(p);
|
|
}
|
|
}
|
|
#endif /* defined(__FreeBSD__) && defined(SIOCIFCREATE2) */
|
|
|
|
#ifdef HAVE_ZEROCOPY_BPF
|
|
/*
|
|
* If the BPF extension to set buffer mode is present, try setting
|
|
* the mode to zero-copy. If that fails, use regular buffering. If
|
|
* it succeeds but other setup fails, return an error to the user.
|
|
*/
|
|
bufmode = BPF_BUFMODE_ZBUF;
|
|
if (ioctl(fd, BIOCSETBUFMODE, (caddr_t)&bufmode) == 0) {
|
|
/*
|
|
* We have zerocopy BPF; use it.
|
|
*/
|
|
pb->zerocopy = 1;
|
|
|
|
/*
|
|
* How to pick a buffer size: first, query the maximum buffer
|
|
* size supported by zero-copy. This also lets us quickly
|
|
* determine whether the kernel generally supports zero-copy.
|
|
* Then, if a buffer size was specified, use that, otherwise
|
|
* query the default buffer size, which reflects kernel
|
|
* policy for a desired default. Round to the nearest page
|
|
* size.
|
|
*/
|
|
if (ioctl(fd, BIOCGETZMAX, (caddr_t)&zbufmax) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCGETZMAX");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
|
|
if (p->opt.buffer_size != 0) {
|
|
/*
|
|
* A buffer size was explicitly specified; use it.
|
|
*/
|
|
v = p->opt.buffer_size;
|
|
} else {
|
|
if ((ioctl(fd, BIOCGBLEN, (caddr_t)&v) < 0) ||
|
|
v < DEFAULT_BUFSIZE)
|
|
v = DEFAULT_BUFSIZE;
|
|
}
|
|
#ifndef roundup
|
|
#define roundup(x, y) ((((x)+((y)-1))/(y))*(y)) /* to any y */
|
|
#endif
|
|
pb->zbufsize = roundup(v, getpagesize());
|
|
if (pb->zbufsize > zbufmax)
|
|
pb->zbufsize = zbufmax;
|
|
pb->zbuf1 = mmap(NULL, pb->zbufsize, PROT_READ | PROT_WRITE,
|
|
MAP_ANON, -1, 0);
|
|
pb->zbuf2 = mmap(NULL, pb->zbufsize, PROT_READ | PROT_WRITE,
|
|
MAP_ANON, -1, 0);
|
|
if (pb->zbuf1 == MAP_FAILED || pb->zbuf2 == MAP_FAILED) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "mmap");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
memset(&bz, 0, sizeof(bz)); /* bzero() deprecated, replaced with memset() */
|
|
bz.bz_bufa = pb->zbuf1;
|
|
bz.bz_bufb = pb->zbuf2;
|
|
bz.bz_buflen = pb->zbufsize;
|
|
if (ioctl(fd, BIOCSETZBUF, (caddr_t)&bz) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCSETZBUF");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
(void)strncpy(ifrname, p->opt.device, ifnamsiz);
|
|
if (ioctl(fd, BIOCSETIF, (caddr_t)&ifr) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCSETIF: %s", p->opt.device);
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
v = pb->zbufsize - sizeof(struct bpf_zbuf_header);
|
|
} else
|
|
#endif
|
|
{
|
|
/*
|
|
* We don't have zerocopy BPF.
|
|
* Set the buffer size.
|
|
*/
|
|
if (p->opt.buffer_size != 0) {
|
|
/*
|
|
* A buffer size was explicitly specified; use it.
|
|
*/
|
|
if (ioctl(fd, BIOCSBLEN,
|
|
(caddr_t)&p->opt.buffer_size) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf,
|
|
PCAP_ERRBUF_SIZE, errno,
|
|
"BIOCSBLEN: %s", p->opt.device);
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
|
|
/*
|
|
* Now bind to the device.
|
|
*/
|
|
(void)strncpy(ifrname, p->opt.device, ifnamsiz);
|
|
#ifdef BIOCSETLIF
|
|
if (ioctl(fd, BIOCSETLIF, (caddr_t)&ifr) < 0)
|
|
#else
|
|
if (ioctl(fd, BIOCSETIF, (caddr_t)&ifr) < 0)
|
|
#endif
|
|
{
|
|
status = check_setif_failure(p, errno);
|
|
goto bad;
|
|
}
|
|
} else {
|
|
/*
|
|
* No buffer size was explicitly specified.
|
|
*
|
|
* Try finding a good size for the buffer;
|
|
* DEFAULT_BUFSIZE may be too big, so keep
|
|
* cutting it in half until we find a size
|
|
* that works, or run out of sizes to try.
|
|
* If the default is larger, don't make it smaller.
|
|
*/
|
|
if ((ioctl(fd, BIOCGBLEN, (caddr_t)&v) < 0) ||
|
|
v < DEFAULT_BUFSIZE)
|
|
v = DEFAULT_BUFSIZE;
|
|
for ( ; v != 0; v >>= 1) {
|
|
/*
|
|
* Ignore the return value - this is because the
|
|
* call fails on BPF systems that don't have
|
|
* kernel malloc. And if the call fails, it's
|
|
* no big deal, we just continue to use the
|
|
* standard buffer size.
|
|
*/
|
|
(void) ioctl(fd, BIOCSBLEN, (caddr_t)&v);
|
|
|
|
(void)strncpy(ifrname, p->opt.device, ifnamsiz);
|
|
#ifdef BIOCSETLIF
|
|
if (ioctl(fd, BIOCSETLIF, (caddr_t)&ifr) >= 0)
|
|
#else
|
|
if (ioctl(fd, BIOCSETIF, (caddr_t)&ifr) >= 0)
|
|
#endif
|
|
break; /* that size worked; we're done */
|
|
|
|
if (errno != ENOBUFS) {
|
|
status = check_setif_failure(p, errno);
|
|
goto bad;
|
|
}
|
|
}
|
|
|
|
if (v == 0) {
|
|
pcap_snprintf(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
"BIOCSBLEN: %s: No buffer size worked",
|
|
p->opt.device);
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
}
|
|
}
|
|
|
|
/* Get the data link layer type. */
|
|
if (ioctl(fd, BIOCGDLT, (caddr_t)&v) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCGDLT");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
|
|
#ifdef _AIX
|
|
/*
|
|
* AIX's BPF returns IFF_ types, not DLT_ types, in BIOCGDLT.
|
|
*/
|
|
switch (v) {
|
|
|
|
case IFT_ETHER:
|
|
case IFT_ISO88023:
|
|
v = DLT_EN10MB;
|
|
break;
|
|
|
|
case IFT_FDDI:
|
|
v = DLT_FDDI;
|
|
break;
|
|
|
|
case IFT_ISO88025:
|
|
v = DLT_IEEE802;
|
|
break;
|
|
|
|
case IFT_LOOP:
|
|
v = DLT_NULL;
|
|
break;
|
|
|
|
default:
|
|
/*
|
|
* We don't know what to map this to yet.
|
|
*/
|
|
pcap_snprintf(p->errbuf, PCAP_ERRBUF_SIZE, "unknown interface type %u",
|
|
v);
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
#endif
|
|
#if _BSDI_VERSION - 0 >= 199510
|
|
/* The SLIP and PPP link layer header changed in BSD/OS 2.1 */
|
|
switch (v) {
|
|
|
|
case DLT_SLIP:
|
|
v = DLT_SLIP_BSDOS;
|
|
break;
|
|
|
|
case DLT_PPP:
|
|
v = DLT_PPP_BSDOS;
|
|
break;
|
|
|
|
case 11: /*DLT_FR*/
|
|
v = DLT_FRELAY;
|
|
break;
|
|
|
|
case 12: /*DLT_C_HDLC*/
|
|
v = DLT_CHDLC;
|
|
break;
|
|
}
|
|
#endif
|
|
|
|
#ifdef BIOCGDLTLIST
|
|
/*
|
|
* We know the default link type -- now determine all the DLTs
|
|
* this interface supports. If this fails with EINVAL, it's
|
|
* not fatal; we just don't get to use the feature later.
|
|
*/
|
|
if (get_dlt_list(fd, v, &bdl, p->errbuf) == -1) {
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
p->dlt_count = bdl.bfl_len;
|
|
p->dlt_list = bdl.bfl_list;
|
|
|
|
#ifdef __APPLE__
|
|
/*
|
|
* Monitor mode fun, continued.
|
|
*
|
|
* For 10.5 and, we're assuming, later releases, as noted above,
|
|
* 802.1 adapters that support monitor mode offer both DLT_EN10MB,
|
|
* DLT_IEEE802_11, and possibly some 802.11-plus-radio-information
|
|
* DLT_ value. Choosing one of the 802.11 DLT_ values will turn
|
|
* monitor mode on.
|
|
*
|
|
* Therefore, if the user asked for monitor mode, we filter out
|
|
* the DLT_EN10MB value, as you can't get that in monitor mode,
|
|
* and, if the user didn't ask for monitor mode, we filter out
|
|
* the 802.11 DLT_ values, because selecting those will turn
|
|
* monitor mode on. Then, for monitor mode, if an 802.11-plus-
|
|
* radio DLT_ value is offered, we try to select that, otherwise
|
|
* we try to select DLT_IEEE802_11.
|
|
*/
|
|
if (have_osinfo) {
|
|
if (isdigit((unsigned)osinfo.release[0]) &&
|
|
(osinfo.release[0] == '9' ||
|
|
isdigit((unsigned)osinfo.release[1]))) {
|
|
/*
|
|
* 10.5 (Darwin 9.x), or later.
|
|
*/
|
|
new_dlt = find_802_11(&bdl);
|
|
if (new_dlt != -1) {
|
|
/*
|
|
* We have at least one 802.11 DLT_ value,
|
|
* so this is an 802.11 interface.
|
|
* new_dlt is the best of the 802.11
|
|
* DLT_ values in the list.
|
|
*/
|
|
if (p->opt.rfmon) {
|
|
/*
|
|
* Our caller wants monitor mode.
|
|
* Purge DLT_EN10MB from the list
|
|
* of link-layer types, as selecting
|
|
* it will keep monitor mode off.
|
|
*/
|
|
remove_non_802_11(p);
|
|
|
|
/*
|
|
* If the new mode we want isn't
|
|
* the default mode, attempt to
|
|
* select the new mode.
|
|
*/
|
|
if ((u_int)new_dlt != v) {
|
|
if (ioctl(p->fd, BIOCSDLT,
|
|
&new_dlt) != -1) {
|
|
/*
|
|
* We succeeded;
|
|
* make this the
|
|
* new DLT_ value.
|
|
*/
|
|
v = new_dlt;
|
|
}
|
|
}
|
|
} else {
|
|
/*
|
|
* Our caller doesn't want
|
|
* monitor mode. Unless this
|
|
* is being done by pcap_open_live(),
|
|
* purge the 802.11 link-layer types
|
|
* from the list, as selecting
|
|
* one of them will turn monitor
|
|
* mode on.
|
|
*/
|
|
if (!p->oldstyle)
|
|
remove_802_11(p);
|
|
}
|
|
} else {
|
|
if (p->opt.rfmon) {
|
|
/*
|
|
* The caller requested monitor
|
|
* mode, but we have no 802.11
|
|
* link-layer types, so they
|
|
* can't have it.
|
|
*/
|
|
status = PCAP_ERROR_RFMON_NOTSUP;
|
|
goto bad;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
#elif defined(HAVE_BSD_IEEE80211)
|
|
/*
|
|
* *BSD with the new 802.11 ioctls.
|
|
* Do we want monitor mode?
|
|
*/
|
|
if (p->opt.rfmon) {
|
|
/*
|
|
* Try to put the interface into monitor mode.
|
|
*/
|
|
retv = monitor_mode(p, 1);
|
|
if (retv != 0) {
|
|
/*
|
|
* We failed.
|
|
*/
|
|
status = retv;
|
|
goto bad;
|
|
}
|
|
|
|
/*
|
|
* We're in monitor mode.
|
|
* Try to find the best 802.11 DLT_ value and, if we
|
|
* succeed, try to switch to that mode if we're not
|
|
* already in that mode.
|
|
*/
|
|
new_dlt = find_802_11(&bdl);
|
|
if (new_dlt != -1) {
|
|
/*
|
|
* We have at least one 802.11 DLT_ value.
|
|
* new_dlt is the best of the 802.11
|
|
* DLT_ values in the list.
|
|
*
|
|
* If the new mode we want isn't the default mode,
|
|
* attempt to select the new mode.
|
|
*/
|
|
if ((u_int)new_dlt != v) {
|
|
if (ioctl(p->fd, BIOCSDLT, &new_dlt) != -1) {
|
|
/*
|
|
* We succeeded; make this the
|
|
* new DLT_ value.
|
|
*/
|
|
v = new_dlt;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
#endif /* various platforms */
|
|
#endif /* BIOCGDLTLIST */
|
|
|
|
/*
|
|
* If this is an Ethernet device, and we don't have a DLT_ list,
|
|
* give it a list with DLT_EN10MB and DLT_DOCSIS. (That'd give
|
|
* 802.11 interfaces DLT_DOCSIS, which isn't the right thing to
|
|
* do, but there's not much we can do about that without finding
|
|
* some other way of determining whether it's an Ethernet or 802.11
|
|
* device.)
|
|
*/
|
|
if (v == DLT_EN10MB && p->dlt_count == 0) {
|
|
p->dlt_list = (u_int *) malloc(sizeof(u_int) * 2);
|
|
/*
|
|
* If that fails, just leave the list empty.
|
|
*/
|
|
if (p->dlt_list != NULL) {
|
|
p->dlt_list[0] = DLT_EN10MB;
|
|
p->dlt_list[1] = DLT_DOCSIS;
|
|
p->dlt_count = 2;
|
|
}
|
|
}
|
|
#ifdef PCAP_FDDIPAD
|
|
if (v == DLT_FDDI)
|
|
p->fddipad = PCAP_FDDIPAD;
|
|
else
|
|
#endif
|
|
p->fddipad = 0;
|
|
p->linktype = v;
|
|
|
|
#if defined(BIOCGHDRCMPLT) && defined(BIOCSHDRCMPLT)
|
|
/*
|
|
* Do a BIOCSHDRCMPLT, if defined, to turn that flag on, so
|
|
* the link-layer source address isn't forcibly overwritten.
|
|
* (Should we ignore errors? Should we do this only if
|
|
* we're open for writing?)
|
|
*
|
|
* XXX - I seem to remember some packet-sending bug in some
|
|
* BSDs - check CVS log for "bpf.c"?
|
|
*/
|
|
if (ioctl(fd, BIOCSHDRCMPLT, &spoof_eth_src) == -1) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCSHDRCMPLT");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
#endif
|
|
/* set timeout */
|
|
#ifdef HAVE_ZEROCOPY_BPF
|
|
/*
|
|
* In zero-copy mode, we just use the timeout in select().
|
|
* XXX - what if we're in non-blocking mode and the *application*
|
|
* is using select() or poll() or kqueues or....?
|
|
*/
|
|
if (p->opt.timeout && !pb->zerocopy) {
|
|
#else
|
|
if (p->opt.timeout) {
|
|
#endif
|
|
/*
|
|
* XXX - is this seconds/nanoseconds in AIX?
|
|
* (Treating it as such doesn't fix the timeout
|
|
* problem described below.)
|
|
*
|
|
* XXX - Mac OS X 10.6 mishandles BIOCSRTIMEOUT in
|
|
* 64-bit userland - it takes, as an argument, a
|
|
* "struct BPF_TIMEVAL", which has 32-bit tv_sec
|
|
* and tv_usec, rather than a "struct timeval".
|
|
*
|
|
* If this platform defines "struct BPF_TIMEVAL",
|
|
* we check whether the structure size in BIOCSRTIMEOUT
|
|
* is that of a "struct timeval" and, if not, we use
|
|
* a "struct BPF_TIMEVAL" rather than a "struct timeval".
|
|
* (That way, if the bug is fixed in a future release,
|
|
* we will still do the right thing.)
|
|
*/
|
|
struct timeval to;
|
|
#ifdef HAVE_STRUCT_BPF_TIMEVAL
|
|
struct BPF_TIMEVAL bpf_to;
|
|
|
|
if (IOCPARM_LEN(BIOCSRTIMEOUT) != sizeof(struct timeval)) {
|
|
bpf_to.tv_sec = p->opt.timeout / 1000;
|
|
bpf_to.tv_usec = (p->opt.timeout * 1000) % 1000000;
|
|
if (ioctl(p->fd, BIOCSRTIMEOUT, (caddr_t)&bpf_to) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf,
|
|
errno, PCAP_ERRBUF_SIZE, "BIOCSRTIMEOUT");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
} else {
|
|
#endif
|
|
to.tv_sec = p->opt.timeout / 1000;
|
|
to.tv_usec = (p->opt.timeout * 1000) % 1000000;
|
|
if (ioctl(p->fd, BIOCSRTIMEOUT, (caddr_t)&to) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf,
|
|
errno, PCAP_ERRBUF_SIZE, "BIOCSRTIMEOUT");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
#ifdef HAVE_STRUCT_BPF_TIMEVAL
|
|
}
|
|
#endif
|
|
}
|
|
|
|
#ifdef BIOCIMMEDIATE
|
|
/*
|
|
* Darren Reed notes that
|
|
*
|
|
* On AIX (4.2 at least), if BIOCIMMEDIATE is not set, the
|
|
* timeout appears to be ignored and it waits until the buffer
|
|
* is filled before returning. The result of not having it
|
|
* set is almost worse than useless if your BPF filter
|
|
* is reducing things to only a few packets (i.e. one every
|
|
* second or so).
|
|
*
|
|
* so we always turn BIOCIMMEDIATE mode on if this is AIX.
|
|
*
|
|
* For other platforms, we don't turn immediate mode on by default,
|
|
* as that would mean we get woken up for every packet, which
|
|
* probably isn't what you want for a packet sniffer.
|
|
*
|
|
* We set immediate mode if the caller requested it by calling
|
|
* pcap_set_immediate() before calling pcap_activate().
|
|
*/
|
|
#ifndef _AIX
|
|
if (p->opt.immediate) {
|
|
#endif /* _AIX */
|
|
v = 1;
|
|
if (ioctl(p->fd, BIOCIMMEDIATE, &v) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCIMMEDIATE");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
#ifndef _AIX
|
|
}
|
|
#endif /* _AIX */
|
|
#else /* BIOCIMMEDIATE */
|
|
if (p->opt.immediate) {
|
|
/*
|
|
* We don't support immediate mode. Fail.
|
|
*/
|
|
pcap_snprintf(p->errbuf, PCAP_ERRBUF_SIZE, "Immediate mode not supported");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
#endif /* BIOCIMMEDIATE */
|
|
|
|
if (p->opt.promisc) {
|
|
/* set promiscuous mode, just warn if it fails */
|
|
if (ioctl(p->fd, BIOCPROMISC, NULL) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCPROMISC");
|
|
status = PCAP_WARNING_PROMISC_NOTSUP;
|
|
}
|
|
}
|
|
|
|
#ifdef BIOCSTSTAMP
|
|
v = BPF_T_BINTIME;
|
|
if (ioctl(p->fd, BIOCSTSTAMP, &v) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCSTSTAMP");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
#endif /* BIOCSTSTAMP */
|
|
|
|
if (ioctl(fd, BIOCGBLEN, (caddr_t)&v) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCGBLEN");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
p->bufsize = v;
|
|
#ifdef HAVE_ZEROCOPY_BPF
|
|
if (!pb->zerocopy) {
|
|
#endif
|
|
p->buffer = malloc(p->bufsize);
|
|
if (p->buffer == NULL) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "malloc");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
#ifdef _AIX
|
|
/* For some strange reason this seems to prevent the EFAULT
|
|
* problems we have experienced from AIX BPF. */
|
|
memset(p->buffer, 0x0, p->bufsize);
|
|
#endif
|
|
#ifdef HAVE_ZEROCOPY_BPF
|
|
}
|
|
#endif
|
|
|
|
/*
|
|
* If there's no filter program installed, there's
|
|
* no indication to the kernel of what the snapshot
|
|
* length should be, so no snapshotting is done.
|
|
*
|
|
* Therefore, when we open the device, we install
|
|
* an "accept everything" filter with the specified
|
|
* snapshot length.
|
|
*/
|
|
total_insn.code = (u_short)(BPF_RET | BPF_K);
|
|
total_insn.jt = 0;
|
|
total_insn.jf = 0;
|
|
total_insn.k = p->snapshot;
|
|
|
|
total_prog.bf_len = 1;
|
|
total_prog.bf_insns = &total_insn;
|
|
if (ioctl(p->fd, BIOCSETF, (caddr_t)&total_prog) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCSETF");
|
|
status = PCAP_ERROR;
|
|
goto bad;
|
|
}
|
|
|
|
/*
|
|
* On most BPF platforms, either you can do a "select()" or
|
|
* "poll()" on a BPF file descriptor and it works correctly,
|
|
* or you can do it and it will return "readable" if the
|
|
* hold buffer is full but not if the timeout expires *and*
|
|
* a non-blocking read will, if the hold buffer is empty
|
|
* but the store buffer isn't empty, rotate the buffers
|
|
* and return what packets are available.
|
|
*
|
|
* In the latter case, the fact that a non-blocking read
|
|
* will give you the available packets means you can work
|
|
* around the failure of "select()" and "poll()" to wake up
|
|
* and return "readable" when the timeout expires by using
|
|
* the timeout as the "select()" or "poll()" timeout, putting
|
|
* the BPF descriptor into non-blocking mode, and read from
|
|
* it regardless of whether "select()" reports it as readable
|
|
* or not.
|
|
*
|
|
* However, in FreeBSD 4.3 and 4.4, "select()" and "poll()"
|
|
* won't wake up and return "readable" if the timer expires
|
|
* and non-blocking reads return EWOULDBLOCK if the hold
|
|
* buffer is empty, even if the store buffer is non-empty.
|
|
*
|
|
* This means the workaround in question won't work.
|
|
*
|
|
* Therefore, on FreeBSD 4.3 and 4.4, we set "p->selectable_fd"
|
|
* to -1, which means "sorry, you can't use 'select()' or 'poll()'
|
|
* here". On all other BPF platforms, we set it to the FD for
|
|
* the BPF device; in NetBSD, OpenBSD, and Darwin, a non-blocking
|
|
* read will, if the hold buffer is empty and the store buffer
|
|
* isn't empty, rotate the buffers and return what packets are
|
|
* there (and in sufficiently recent versions of OpenBSD
|
|
* "select()" and "poll()" should work correctly).
|
|
*
|
|
* XXX - what about AIX?
|
|
*/
|
|
p->selectable_fd = p->fd; /* assume select() works until we know otherwise */
|
|
if (have_osinfo) {
|
|
/*
|
|
* We can check what OS this is.
|
|
*/
|
|
if (strcmp(osinfo.sysname, "FreeBSD") == 0) {
|
|
if (strncmp(osinfo.release, "4.3-", 4) == 0 ||
|
|
strncmp(osinfo.release, "4.4-", 4) == 0)
|
|
p->selectable_fd = -1;
|
|
}
|
|
}
|
|
|
|
p->read_op = pcap_read_bpf;
|
|
p->inject_op = pcap_inject_bpf;
|
|
p->setfilter_op = pcap_setfilter_bpf;
|
|
p->setdirection_op = pcap_setdirection_bpf;
|
|
p->set_datalink_op = pcap_set_datalink_bpf;
|
|
p->getnonblock_op = pcap_getnonblock_bpf;
|
|
p->setnonblock_op = pcap_setnonblock_bpf;
|
|
p->stats_op = pcap_stats_bpf;
|
|
p->cleanup_op = pcap_cleanup_bpf;
|
|
|
|
return (status);
|
|
bad:
|
|
pcap_cleanup_bpf(p);
|
|
return (status);
|
|
}
|
|
|
|
/*
|
|
* Not all interfaces can be bound to by BPF, so try to bind to
|
|
* the specified interface; return 0 if we fail with
|
|
* PCAP_ERROR_NO_SUCH_DEVICE (which means we got an ENXIO when we tried
|
|
* to bind, which means this interface isn't in the list of interfaces
|
|
* attached to BPF) and 1 otherwise.
|
|
*/
|
|
static int
|
|
check_bpf_bindable(const char *name)
|
|
{
|
|
int fd;
|
|
char errbuf[PCAP_ERRBUF_SIZE];
|
|
|
|
/*
|
|
* On macOS, we don't do this check if the device name begins
|
|
* with "wlt"; at least some versions of macOS (actually, it
|
|
* was called "Mac OS X" then...) offer monitor mode capturing
|
|
* by having a separate "monitor mode" device for each wireless
|
|
* adapter, rather than by implementing the ioctls that
|
|
* {Free,Net,Open,DragonFly}BSD provide. Opening that device
|
|
* puts the adapter into monitor mode, which, at least for
|
|
* some adapters, causes them to deassociate from the network
|
|
* with which they're associated.
|
|
*
|
|
* Instead, we try to open the corresponding "en" device (so
|
|
* that we don't end up with, for users without sufficient
|
|
* privilege to open capture devices, a list of adapters that
|
|
* only includes the wlt devices).
|
|
*/
|
|
#ifdef __APPLE__
|
|
if (strncmp(name, "wlt", 3) == 0) {
|
|
char *en_name;
|
|
size_t en_name_len;
|
|
|
|
/*
|
|
* Try to allocate a buffer for the "en"
|
|
* device's name.
|
|
*/
|
|
en_name_len = strlen(name) - 1;
|
|
en_name = malloc(en_name_len + 1);
|
|
if (en_name == NULL) {
|
|
pcap_fmt_errmsg_for_errno(errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "malloc");
|
|
return (-1);
|
|
}
|
|
strcpy(en_name, "en");
|
|
strcat(en_name, name + 3);
|
|
fd = bpf_open_and_bind(en_name, errbuf);
|
|
free(en_name);
|
|
} else
|
|
#endif /* __APPLE */
|
|
fd = bpf_open_and_bind(name, errbuf);
|
|
if (fd < 0) {
|
|
/*
|
|
* Error - was it PCAP_ERROR_NO_SUCH_DEVICE?
|
|
*/
|
|
if (fd == PCAP_ERROR_NO_SUCH_DEVICE) {
|
|
/*
|
|
* Yes, so we can't bind to this because it's
|
|
* not something supported by BPF.
|
|
*/
|
|
return (0);
|
|
}
|
|
/*
|
|
* No, so we don't know whether it's supported or not;
|
|
* say it is, so that the user can at least try to
|
|
* open it and report the error (which is probably
|
|
* "you don't have permission to open BPF devices";
|
|
* reporting those interfaces means users will ask
|
|
* "why am I getting a permissions error when I try
|
|
* to capture" rather than "why am I not seeing any
|
|
* interfaces", making the underlying problem clearer).
|
|
*/
|
|
return (1);
|
|
}
|
|
|
|
/*
|
|
* Success.
|
|
*/
|
|
close(fd);
|
|
return (1);
|
|
}
|
|
|
|
#if defined(__FreeBSD__) && defined(SIOCIFCREATE2)
|
|
static int
|
|
get_usb_if_flags(const char *name _U_, bpf_u_int32 *flags _U_, char *errbuf _U_)
|
|
{
|
|
/*
|
|
* XXX - if there's a way to determine whether there's something
|
|
* plugged into a given USB bus, use that to determine whether
|
|
* this device is "connected" or not.
|
|
*/
|
|
return (0);
|
|
}
|
|
|
|
static int
|
|
finddevs_usb(pcap_if_list_t *devlistp, char *errbuf)
|
|
{
|
|
DIR *usbdir;
|
|
struct dirent *usbitem;
|
|
size_t name_max;
|
|
char *name;
|
|
|
|
/*
|
|
* We might have USB sniffing support, so try looking for USB
|
|
* interfaces.
|
|
*
|
|
* We want to report a usbusN device for each USB bus, but
|
|
* usbusN interfaces might, or might not, exist for them -
|
|
* we create one if there isn't already one.
|
|
*
|
|
* So, instead, we look in /dev/usb for all buses and create
|
|
* a "usbusN" device for each one.
|
|
*/
|
|
usbdir = opendir("/dev/usb");
|
|
if (usbdir == NULL) {
|
|
/*
|
|
* Just punt.
|
|
*/
|
|
return (0);
|
|
}
|
|
|
|
/*
|
|
* Leave enough room for a 32-bit (10-digit) bus number.
|
|
* Yes, that's overkill, but we won't be using
|
|
* the buffer very long.
|
|
*/
|
|
name_max = USBUS_PREFIX_LEN + 10 + 1;
|
|
name = malloc(name_max);
|
|
if (name == NULL) {
|
|
closedir(usbdir);
|
|
return (0);
|
|
}
|
|
while ((usbitem = readdir(usbdir)) != NULL) {
|
|
char *p;
|
|
size_t busnumlen;
|
|
|
|
if (strcmp(usbitem->d_name, ".") == 0 ||
|
|
strcmp(usbitem->d_name, "..") == 0) {
|
|
/*
|
|
* Ignore these.
|
|
*/
|
|
continue;
|
|
}
|
|
p = strchr(usbitem->d_name, '.');
|
|
if (p == NULL)
|
|
continue;
|
|
busnumlen = p - usbitem->d_name;
|
|
memcpy(name, usbus_prefix, USBUS_PREFIX_LEN);
|
|
memcpy(name + USBUS_PREFIX_LEN, usbitem->d_name, busnumlen);
|
|
*(name + USBUS_PREFIX_LEN + busnumlen) = '\0';
|
|
/*
|
|
* There's an entry in this directory for every USB device,
|
|
* not for every bus; if there's more than one device on
|
|
* the bus, there'll be more than one entry for that bus,
|
|
* so we need to avoid adding multiple capture devices
|
|
* for each bus.
|
|
*/
|
|
if (find_or_add_dev(devlistp, name, PCAP_IF_UP,
|
|
get_usb_if_flags, NULL, errbuf) == NULL) {
|
|
free(name);
|
|
closedir(usbdir);
|
|
return (PCAP_ERROR);
|
|
}
|
|
}
|
|
free(name);
|
|
closedir(usbdir);
|
|
return (0);
|
|
}
|
|
#endif
|
|
|
|
/*
|
|
* Get additional flags for a device, using SIOCGIFMEDIA.
|
|
*/
|
|
#ifdef SIOCGIFMEDIA
|
|
static int
|
|
get_if_flags(const char *name, bpf_u_int32 *flags, char *errbuf)
|
|
{
|
|
int sock;
|
|
struct ifmediareq req;
|
|
|
|
sock = socket(AF_INET, SOCK_DGRAM, 0);
|
|
if (sock == -1) {
|
|
pcap_fmt_errmsg_for_errno(errbuf, PCAP_ERRBUF_SIZE, errno,
|
|
"Can't create socket to get media information for %s",
|
|
name);
|
|
return (-1);
|
|
}
|
|
memset(&req, 0, sizeof(req));
|
|
strncpy(req.ifm_name, name, sizeof(req.ifm_name));
|
|
if (ioctl(sock, SIOCGIFMEDIA, &req) < 0) {
|
|
if (errno == EOPNOTSUPP || errno == EINVAL || errno == ENOTTY ||
|
|
errno == ENODEV || errno == EPERM) {
|
|
/*
|
|
* Not supported, so we can't provide any
|
|
* additional information. Assume that
|
|
* this means that "connected" vs.
|
|
* "disconnected" doesn't apply.
|
|
*
|
|
* The ioctl routine for Apple's pktap devices,
|
|
* annoyingly, checks for "are you root?" before
|
|
* checking whether the ioctl is valid, so it
|
|
* returns EPERM, rather than ENOTSUP, for the
|
|
* invalid SIOCGIFMEDIA, unless you're root.
|
|
* So, just as we do for some ethtool ioctls
|
|
* on Linux, which makes the same mistake, we
|
|
* also treat EPERM as meaning "not supported".
|
|
*/
|
|
*flags |= PCAP_IF_CONNECTION_STATUS_NOT_APPLICABLE;
|
|
close(sock);
|
|
return (0);
|
|
}
|
|
pcap_fmt_errmsg_for_errno(errbuf, PCAP_ERRBUF_SIZE, errno,
|
|
"SIOCGIFMEDIA on %s failed", name);
|
|
close(sock);
|
|
return (-1);
|
|
}
|
|
close(sock);
|
|
|
|
/*
|
|
* OK, what type of network is this?
|
|
*/
|
|
switch (IFM_TYPE(req.ifm_active)) {
|
|
|
|
case IFM_IEEE80211:
|
|
/*
|
|
* Wireless.
|
|
*/
|
|
*flags |= PCAP_IF_WIRELESS;
|
|
break;
|
|
}
|
|
|
|
/*
|
|
* Do we know whether it's connected?
|
|
*/
|
|
if (req.ifm_status & IFM_AVALID) {
|
|
/*
|
|
* Yes.
|
|
*/
|
|
if (req.ifm_status & IFM_ACTIVE) {
|
|
/*
|
|
* It's connected.
|
|
*/
|
|
*flags |= PCAP_IF_CONNECTION_STATUS_CONNECTED;
|
|
} else {
|
|
/*
|
|
* It's disconnected.
|
|
*/
|
|
*flags |= PCAP_IF_CONNECTION_STATUS_DISCONNECTED;
|
|
}
|
|
}
|
|
return (0);
|
|
}
|
|
#else
|
|
static int
|
|
get_if_flags(const char *name _U_, bpf_u_int32 *flags _U_, char *errbuf _U_)
|
|
{
|
|
/*
|
|
* Nothing we can do other than mark loopback devices as "the
|
|
* connected/disconnected status doesn't apply".
|
|
*
|
|
* XXX - on Solaris, can we do what the dladm command does,
|
|
* i.e. get a connected/disconnected indication from a kstat?
|
|
* (Note that you can also get the link speed, and possibly
|
|
* other information, from a kstat as well.)
|
|
*/
|
|
if (*flags & PCAP_IF_LOOPBACK) {
|
|
/*
|
|
* Loopback devices aren't wireless, and "connected"/
|
|
* "disconnected" doesn't apply to them.
|
|
*/
|
|
*flags |= PCAP_IF_CONNECTION_STATUS_NOT_APPLICABLE;
|
|
return (0);
|
|
}
|
|
return (0);
|
|
}
|
|
#endif
|
|
|
|
int
|
|
pcap_platform_finddevs(pcap_if_list_t *devlistp, char *errbuf)
|
|
{
|
|
/*
|
|
* Get the list of regular interfaces first.
|
|
*/
|
|
if (pcap_findalldevs_interfaces(devlistp, errbuf, check_bpf_bindable,
|
|
get_if_flags) == -1)
|
|
return (-1); /* failure */
|
|
|
|
#if defined(__FreeBSD__) && defined(SIOCIFCREATE2)
|
|
if (finddevs_usb(devlistp, errbuf) == -1)
|
|
return (-1);
|
|
#endif
|
|
|
|
return (0);
|
|
}
|
|
|
|
#ifdef HAVE_BSD_IEEE80211
|
|
static int
|
|
monitor_mode(pcap_t *p, int set)
|
|
{
|
|
struct pcap_bpf *pb = p->priv;
|
|
int sock;
|
|
struct ifmediareq req;
|
|
IFM_ULIST_TYPE *media_list;
|
|
int i;
|
|
int can_do;
|
|
struct ifreq ifr;
|
|
|
|
sock = socket(AF_INET, SOCK_DGRAM, 0);
|
|
if (sock == -1) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "can't open socket");
|
|
return (PCAP_ERROR);
|
|
}
|
|
|
|
memset(&req, 0, sizeof req);
|
|
strncpy(req.ifm_name, p->opt.device, sizeof req.ifm_name);
|
|
|
|
/*
|
|
* Find out how many media types we have.
|
|
*/
|
|
if (ioctl(sock, SIOCGIFMEDIA, &req) < 0) {
|
|
/*
|
|
* Can't get the media types.
|
|
*/
|
|
switch (errno) {
|
|
|
|
case ENXIO:
|
|
/*
|
|
* There's no such device.
|
|
*/
|
|
close(sock);
|
|
return (PCAP_ERROR_NO_SUCH_DEVICE);
|
|
|
|
case EINVAL:
|
|
/*
|
|
* Interface doesn't support SIOC{G,S}IFMEDIA.
|
|
*/
|
|
close(sock);
|
|
return (PCAP_ERROR_RFMON_NOTSUP);
|
|
|
|
default:
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "SIOCGIFMEDIA");
|
|
close(sock);
|
|
return (PCAP_ERROR);
|
|
}
|
|
}
|
|
if (req.ifm_count == 0) {
|
|
/*
|
|
* No media types.
|
|
*/
|
|
close(sock);
|
|
return (PCAP_ERROR_RFMON_NOTSUP);
|
|
}
|
|
|
|
/*
|
|
* Allocate a buffer to hold all the media types, and
|
|
* get the media types.
|
|
*/
|
|
media_list = malloc(req.ifm_count * sizeof(*media_list));
|
|
if (media_list == NULL) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "malloc");
|
|
close(sock);
|
|
return (PCAP_ERROR);
|
|
}
|
|
req.ifm_ulist = media_list;
|
|
if (ioctl(sock, SIOCGIFMEDIA, &req) < 0) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "SIOCGIFMEDIA");
|
|
free(media_list);
|
|
close(sock);
|
|
return (PCAP_ERROR);
|
|
}
|
|
|
|
/*
|
|
* Look for an 802.11 "automatic" media type.
|
|
* We assume that all 802.11 adapters have that media type,
|
|
* and that it will carry the monitor mode supported flag.
|
|
*/
|
|
can_do = 0;
|
|
for (i = 0; i < req.ifm_count; i++) {
|
|
if (IFM_TYPE(media_list[i]) == IFM_IEEE80211
|
|
&& IFM_SUBTYPE(media_list[i]) == IFM_AUTO) {
|
|
/* OK, does it do monitor mode? */
|
|
if (media_list[i] & IFM_IEEE80211_MONITOR) {
|
|
can_do = 1;
|
|
break;
|
|
}
|
|
}
|
|
}
|
|
free(media_list);
|
|
if (!can_do) {
|
|
/*
|
|
* This adapter doesn't support monitor mode.
|
|
*/
|
|
close(sock);
|
|
return (PCAP_ERROR_RFMON_NOTSUP);
|
|
}
|
|
|
|
if (set) {
|
|
/*
|
|
* Don't just check whether we can enable monitor mode,
|
|
* do so, if it's not already enabled.
|
|
*/
|
|
if ((req.ifm_current & IFM_IEEE80211_MONITOR) == 0) {
|
|
/*
|
|
* Monitor mode isn't currently on, so turn it on,
|
|
* and remember that we should turn it off when the
|
|
* pcap_t is closed.
|
|
*/
|
|
|
|
/*
|
|
* If we haven't already done so, arrange to have
|
|
* "pcap_close_all()" called when we exit.
|
|
*/
|
|
if (!pcap_do_addexit(p)) {
|
|
/*
|
|
* "atexit()" failed; don't put the interface
|
|
* in monitor mode, just give up.
|
|
*/
|
|
close(sock);
|
|
return (PCAP_ERROR);
|
|
}
|
|
memset(&ifr, 0, sizeof(ifr));
|
|
(void)strncpy(ifr.ifr_name, p->opt.device,
|
|
sizeof(ifr.ifr_name));
|
|
ifr.ifr_media = req.ifm_current | IFM_IEEE80211_MONITOR;
|
|
if (ioctl(sock, SIOCSIFMEDIA, &ifr) == -1) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf,
|
|
PCAP_ERRBUF_SIZE, errno, "SIOCSIFMEDIA");
|
|
close(sock);
|
|
return (PCAP_ERROR);
|
|
}
|
|
|
|
pb->must_do_on_close |= MUST_CLEAR_RFMON;
|
|
|
|
/*
|
|
* Add this to the list of pcaps to close when we exit.
|
|
*/
|
|
pcap_add_to_pcaps_to_close(p);
|
|
}
|
|
}
|
|
return (0);
|
|
}
|
|
#endif /* HAVE_BSD_IEEE80211 */
|
|
|
|
#if defined(BIOCGDLTLIST) && (defined(__APPLE__) || defined(HAVE_BSD_IEEE80211))
|
|
/*
|
|
* Check whether we have any 802.11 link-layer types; return the best
|
|
* of the 802.11 link-layer types if we find one, and return -1
|
|
* otherwise.
|
|
*
|
|
* DLT_IEEE802_11_RADIO, with the radiotap header, is considered the
|
|
* best 802.11 link-layer type; any of the other 802.11-plus-radio
|
|
* headers are second-best; 802.11 with no radio information is
|
|
* the least good.
|
|
*/
|
|
static int
|
|
find_802_11(struct bpf_dltlist *bdlp)
|
|
{
|
|
int new_dlt;
|
|
u_int i;
|
|
|
|
/*
|
|
* Scan the list of DLT_ values, looking for 802.11 values,
|
|
* and, if we find any, choose the best of them.
|
|
*/
|
|
new_dlt = -1;
|
|
for (i = 0; i < bdlp->bfl_len; i++) {
|
|
switch (bdlp->bfl_list[i]) {
|
|
|
|
case DLT_IEEE802_11:
|
|
/*
|
|
* 802.11, but no radio.
|
|
*
|
|
* Offer this, and select it as the new mode
|
|
* unless we've already found an 802.11
|
|
* header with radio information.
|
|
*/
|
|
if (new_dlt == -1)
|
|
new_dlt = bdlp->bfl_list[i];
|
|
break;
|
|
|
|
#ifdef DLT_PRISM_HEADER
|
|
case DLT_PRISM_HEADER:
|
|
#endif
|
|
#ifdef DLT_AIRONET_HEADER
|
|
case DLT_AIRONET_HEADER:
|
|
#endif
|
|
case DLT_IEEE802_11_RADIO_AVS:
|
|
/*
|
|
* 802.11 with radio, but not radiotap.
|
|
*
|
|
* Offer this, and select it as the new mode
|
|
* unless we've already found the radiotap DLT_.
|
|
*/
|
|
if (new_dlt != DLT_IEEE802_11_RADIO)
|
|
new_dlt = bdlp->bfl_list[i];
|
|
break;
|
|
|
|
case DLT_IEEE802_11_RADIO:
|
|
/*
|
|
* 802.11 with radiotap.
|
|
*
|
|
* Offer this, and select it as the new mode.
|
|
*/
|
|
new_dlt = bdlp->bfl_list[i];
|
|
break;
|
|
|
|
default:
|
|
/*
|
|
* Not 802.11.
|
|
*/
|
|
break;
|
|
}
|
|
}
|
|
|
|
return (new_dlt);
|
|
}
|
|
#endif /* defined(BIOCGDLTLIST) && (defined(__APPLE__) || defined(HAVE_BSD_IEEE80211)) */
|
|
|
|
#if defined(__APPLE__) && defined(BIOCGDLTLIST)
|
|
/*
|
|
* Remove non-802.11 header types from the list of DLT_ values, as we're in
|
|
* monitor mode, and those header types aren't supported in monitor mode.
|
|
*/
|
|
static void
|
|
remove_non_802_11(pcap_t *p)
|
|
{
|
|
int i, j;
|
|
|
|
/*
|
|
* Scan the list of DLT_ values and discard non-802.11 ones.
|
|
*/
|
|
j = 0;
|
|
for (i = 0; i < p->dlt_count; i++) {
|
|
switch (p->dlt_list[i]) {
|
|
|
|
case DLT_EN10MB:
|
|
case DLT_RAW:
|
|
/*
|
|
* Not 802.11. Don't offer this one.
|
|
*/
|
|
continue;
|
|
|
|
default:
|
|
/*
|
|
* Just copy this mode over.
|
|
*/
|
|
break;
|
|
}
|
|
|
|
/*
|
|
* Copy this DLT_ value to its new position.
|
|
*/
|
|
p->dlt_list[j] = p->dlt_list[i];
|
|
j++;
|
|
}
|
|
|
|
/*
|
|
* Set the DLT_ count to the number of entries we copied.
|
|
*/
|
|
p->dlt_count = j;
|
|
}
|
|
|
|
/*
|
|
* Remove 802.11 link-layer types from the list of DLT_ values, as
|
|
* we're not in monitor mode, and those DLT_ values will switch us
|
|
* to monitor mode.
|
|
*/
|
|
static void
|
|
remove_802_11(pcap_t *p)
|
|
{
|
|
int i, j;
|
|
|
|
/*
|
|
* Scan the list of DLT_ values and discard 802.11 values.
|
|
*/
|
|
j = 0;
|
|
for (i = 0; i < p->dlt_count; i++) {
|
|
switch (p->dlt_list[i]) {
|
|
|
|
case DLT_IEEE802_11:
|
|
#ifdef DLT_PRISM_HEADER
|
|
case DLT_PRISM_HEADER:
|
|
#endif
|
|
#ifdef DLT_AIRONET_HEADER
|
|
case DLT_AIRONET_HEADER:
|
|
#endif
|
|
case DLT_IEEE802_11_RADIO:
|
|
case DLT_IEEE802_11_RADIO_AVS:
|
|
#ifdef DLT_PPI
|
|
case DLT_PPI:
|
|
#endif
|
|
/*
|
|
* 802.11. Don't offer this one.
|
|
*/
|
|
continue;
|
|
|
|
default:
|
|
/*
|
|
* Just copy this mode over.
|
|
*/
|
|
break;
|
|
}
|
|
|
|
/*
|
|
* Copy this DLT_ value to its new position.
|
|
*/
|
|
p->dlt_list[j] = p->dlt_list[i];
|
|
j++;
|
|
}
|
|
|
|
/*
|
|
* Set the DLT_ count to the number of entries we copied.
|
|
*/
|
|
p->dlt_count = j;
|
|
}
|
|
#endif /* defined(__APPLE__) && defined(BIOCGDLTLIST) */
|
|
|
|
static int
|
|
pcap_setfilter_bpf(pcap_t *p, struct bpf_program *fp)
|
|
{
|
|
struct pcap_bpf *pb = p->priv;
|
|
|
|
/*
|
|
* Free any user-mode filter we might happen to have installed.
|
|
*/
|
|
pcap_freecode(&p->fcode);
|
|
|
|
/*
|
|
* Try to install the kernel filter.
|
|
*/
|
|
if (ioctl(p->fd, BIOCSETF, (caddr_t)fp) == 0) {
|
|
/*
|
|
* It worked.
|
|
*/
|
|
pb->filtering_in_kernel = 1; /* filtering in the kernel */
|
|
|
|
/*
|
|
* Discard any previously-received packets, as they might
|
|
* have passed whatever filter was formerly in effect, but
|
|
* might not pass this filter (BIOCSETF discards packets
|
|
* buffered in the kernel, so you can lose packets in any
|
|
* case).
|
|
*/
|
|
p->cc = 0;
|
|
return (0);
|
|
}
|
|
|
|
/*
|
|
* We failed.
|
|
*
|
|
* If it failed with EINVAL, that's probably because the program
|
|
* is invalid or too big. Validate it ourselves; if we like it
|
|
* (we currently allow backward branches, to support protochain),
|
|
* run it in userland. (There's no notion of "too big" for
|
|
* userland.)
|
|
*
|
|
* Otherwise, just give up.
|
|
* XXX - if the copy of the program into the kernel failed,
|
|
* we will get EINVAL rather than, say, EFAULT on at least
|
|
* some kernels.
|
|
*/
|
|
if (errno != EINVAL) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, PCAP_ERRBUF_SIZE,
|
|
errno, "BIOCSETF");
|
|
return (-1);
|
|
}
|
|
|
|
/*
|
|
* install_bpf_program() validates the program.
|
|
*
|
|
* XXX - what if we already have a filter in the kernel?
|
|
*/
|
|
if (install_bpf_program(p, fp) < 0)
|
|
return (-1);
|
|
pb->filtering_in_kernel = 0; /* filtering in userland */
|
|
return (0);
|
|
}
|
|
|
|
/*
|
|
* Set direction flag: Which packets do we accept on a forwarding
|
|
* single device? IN, OUT or both?
|
|
*/
|
|
#if defined(BIOCSDIRECTION)
|
|
static int
|
|
pcap_setdirection_bpf(pcap_t *p, pcap_direction_t d)
|
|
{
|
|
u_int direction;
|
|
|
|
direction = (d == PCAP_D_IN) ? BPF_D_IN :
|
|
((d == PCAP_D_OUT) ? BPF_D_OUT : BPF_D_INOUT);
|
|
if (ioctl(p->fd, BIOCSDIRECTION, &direction) == -1) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, sizeof(p->errbuf),
|
|
errno, "Cannot set direction to %s",
|
|
(d == PCAP_D_IN) ? "PCAP_D_IN" :
|
|
((d == PCAP_D_OUT) ? "PCAP_D_OUT" : "PCAP_D_INOUT"));
|
|
return (-1);
|
|
}
|
|
return (0);
|
|
}
|
|
#elif defined(BIOCSSEESENT)
|
|
static int
|
|
pcap_setdirection_bpf(pcap_t *p, pcap_direction_t d)
|
|
{
|
|
u_int seesent;
|
|
|
|
/*
|
|
* We don't support PCAP_D_OUT.
|
|
*/
|
|
if (d == PCAP_D_OUT) {
|
|
pcap_snprintf(p->errbuf, sizeof(p->errbuf),
|
|
"Setting direction to PCAP_D_OUT is not supported on BPF");
|
|
return -1;
|
|
}
|
|
|
|
seesent = (d == PCAP_D_INOUT);
|
|
if (ioctl(p->fd, BIOCSSEESENT, &seesent) == -1) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, sizeof(p->errbuf),
|
|
errno, "Cannot set direction to %s",
|
|
(d == PCAP_D_INOUT) ? "PCAP_D_INOUT" : "PCAP_D_IN");
|
|
return (-1);
|
|
}
|
|
return (0);
|
|
}
|
|
#else
|
|
static int
|
|
pcap_setdirection_bpf(pcap_t *p, pcap_direction_t d _U_)
|
|
{
|
|
(void) pcap_snprintf(p->errbuf, sizeof(p->errbuf),
|
|
"This system doesn't support BIOCSSEESENT, so the direction can't be set");
|
|
return (-1);
|
|
}
|
|
#endif
|
|
|
|
#ifdef BIOCSDLT
|
|
static int
|
|
pcap_set_datalink_bpf(pcap_t *p, int dlt)
|
|
{
|
|
if (ioctl(p->fd, BIOCSDLT, &dlt) == -1) {
|
|
pcap_fmt_errmsg_for_errno(p->errbuf, sizeof(p->errbuf),
|
|
errno, "Cannot set DLT %d", dlt);
|
|
return (-1);
|
|
}
|
|
return (0);
|
|
}
|
|
#else
|
|
static int
|
|
pcap_set_datalink_bpf(pcap_t *p _U_, int dlt _U_)
|
|
{
|
|
return (0);
|
|
}
|
|
#endif
|
|
|
|
/*
|
|
* Platform-specific information.
|
|
*/
|
|
const char *
|
|
pcap_lib_version(void)
|
|
{
|
|
#ifdef HAVE_ZEROCOPY_BPF
|
|
return (PCAP_VERSION_STRING " (with zerocopy support)");
|
|
#else
|
|
return (PCAP_VERSION_STRING);
|
|
#endif
|
|
}
|