Move interface.c to networking directory since it's only used by ifconfig.

2 files changed, 2094 insertions, 1 deletions

diff --git a/networking/Makefile.in b/networking/Makefile.in
index ecfd823..376ecea 100644
--- a/networking/Makefile.in
+++ b/networking/Makefile.in
@@ -17,7 +17,7 @@ NETWORKING-$(CONFIG_FTPGET)       += ftpgetput.o
 NETWORKING-$(CONFIG_FTPPUT)       += ftpgetput.o
 NETWORKING-$(CONFIG_HOSTNAME)     += hostname.o
 NETWORKING-$(CONFIG_HTTPD)        += httpd.o
-NETWORKING-$(CONFIG_IFCONFIG)     += ifconfig.o
+NETWORKING-$(CONFIG_IFCONFIG)     += ifconfig.o interface.o
 NETWORKING-$(CONFIG_IFUPDOWN)     += ifupdown.o
 NETWORKING-$(CONFIG_INETD)        += inetd.o
 NETWORKING-$(CONFIG_IP)           += ip.o
diff --git a/networking/interface.c b/networking/interface.c
new file mode 100644
index 0000000..ce7d67d
--- /dev/null
+++ b/networking/interface.c
@@ -0,0 +1,2093 @@
+/*
+ * stolen from net-tools-1.59 and stripped down for busybox by
+ *			Erik Andersen <andersen@codepoet.org>
+ *
+ * Heavily modified by Manuel Novoa III       Mar 12, 2001
+ *
+ * Pruned unused code using KEEP_UNUSED define.
+ * Added print_bytes_scaled function to reduce code size.
+ * Added some (potentially) missing defines.
+ * Improved display support for -a and for a named interface.
+ *
+ * -----------------------------------------------------------
+ *
+ * ifconfig   This file contains an implementation of the command
+ *              that either displays or sets the characteristics of
+ *              one or more of the system's networking interfaces.
+ *
+ * Version:     $Id: interface.c,v 1.25 2004/08/26 21:45:21 andersen Exp $
+ *
+ * Author:      Fred N. van Kempen, <waltje@uwalt.nl.mugnet.org>
+ *              and others.  Copyright 1993 MicroWalt Corporation
+ *
+ *              This program is free software; you can redistribute it
+ *              and/or  modify it under  the terms of  the GNU General
+ *              Public  License as  published  by  the  Free  Software
+ *              Foundation;  either  version 2 of the License, or  (at
+ *              your option) any later version.
+ *
+ * Patched to support 'add' and 'del' keywords for INET(4) addresses
+ * by Mrs. Brisby <mrs.brisby@nimh.org>
+ *
+ * {1.34} - 19980630 - Arnaldo Carvalho de Melo <acme@conectiva.com.br>
+ *                     - gettext instead of catgets for i18n
+ *          10/1998  - Andi Kleen. Use interface list primitives.
+ *	    20001008 - Bernd Eckenfels, Patch from RH for setting mtu
+ *			(default AF was wrong)
+ */
+
+/* #define KEEP_UNUSED */
+
+/*
+ *
+ * Protocol Families.
+ *
+ */
+#define HAVE_AFINET 1
+#undef HAVE_AFIPX
+#undef HAVE_AFATALK
+#undef HAVE_AFNETROM
+#undef HAVE_AFX25
+#undef HAVE_AFECONET
+#undef HAVE_AFASH
+
+/*
+ *
+ * Device Hardware types.
+ *
+ */
+#define HAVE_HWETHER	1
+#define HAVE_HWPPP	1
+#undef HAVE_HWSLIP
+
+
+#include "inet_common.h"
+#include <stdio.h>
+#include <errno.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <fcntl.h>
+#include <ctype.h>
+#include <sys/ioctl.h>
+#include <sys/types.h>
+#include <net/if.h>
+#include <net/if_arp.h>
+#include "libbb.h"
+
+#ifdef CONFIG_FEATURE_IPV6
+# define HAVE_AFINET6 1
+#else
+# undef HAVE_AFINET6
+#endif
+
+#define _(x) x
+#define _PATH_PROCNET_DEV               "/proc/net/dev"
+#define _PATH_PROCNET_IFINET6           "/proc/net/if_inet6"
+#define new(p) ((p) = xcalloc(1,sizeof(*(p))))
+#define KRELEASE(maj,min,patch) ((maj) * 65536 + (min)*256 + (patch))
+
+#ifdef HAVE_HWSLIP
+#include <net/if_slip.h>
+#endif
+
+#if HAVE_AFINET6
+
+#ifndef _LINUX_IN6_H
+/*
+ *    This is in linux/include/net/ipv6.h.
+ */
+
+struct in6_ifreq {
+	struct in6_addr ifr6_addr;
+	uint32_t ifr6_prefixlen;
+	unsigned int ifr6_ifindex;
+};
+
+#endif
+
+#endif							/* HAVE_AFINET6 */
+
+#if HAVE_AFIPX
+#if (__GLIBC__ >= 2) && (__GLIBC_MINOR__ >= 1)
+#include <netipx/ipx.h>
+#endif
+#endif
+
+/* Defines for glibc2.0 users. */
+#ifndef SIOCSIFTXQLEN
+#define SIOCSIFTXQLEN      0x8943
+#define SIOCGIFTXQLEN      0x8942
+#endif
+
+/* ifr_qlen is ifru_ivalue, but it isn't present in 2.0 kernel headers */
+#ifndef ifr_qlen
+#define ifr_qlen        ifr_ifru.ifru_mtu
+#endif
+
+#ifndef HAVE_TXQUEUELEN
+#define HAVE_TXQUEUELEN 1
+#endif
+
+#ifndef IFF_DYNAMIC
+#define IFF_DYNAMIC     0x8000	/* dialup device with changing addresses */
+#endif
+
+/* This structure defines protocol families and their handlers. */
+struct aftype {
+	const char *name;
+	const char *title;
+	int af;
+	int alen;
+	char *(*print) (unsigned char *);
+	char *(*sprint) (struct sockaddr *, int numeric);
+	int (*input) (int type, char *bufp, struct sockaddr *);
+	void (*herror) (char *text);
+	int (*rprint) (int options);
+	int (*rinput) (int typ, int ext, char **argv);
+
+	/* may modify src */
+	int (*getmask) (char *src, struct sockaddr * mask, char *name);
+
+	int fd;
+	char *flag_file;
+};
+
+#ifdef KEEP_UNUSED
+
+static int flag_unx;
+
+#ifdef HAVE_AFIPX
+static int flag_ipx;
+#endif
+#ifdef HAVE_AFX25
+static int flag_ax25;
+#endif
+#ifdef HAVE_AFATALK
+static int flag_ddp;
+#endif
+#ifdef HAVE_AFNETROM
+static int flag_netrom;
+#endif
+static int flag_inet;
+
+#ifdef HAVE_AFINET6
+static int flag_inet6;
+#endif
+#ifdef HAVE_AFECONET
+static int flag_econet;
+#endif
+#ifdef HAVE_AFX25
+static int flag_x25 = 0;
+#endif
+#ifdef HAVE_AFASH
+static int flag_ash;
+#endif
+
+
+static struct aftrans_t {
+	char *alias;
+	char *name;
+	int *flag;
+} aftrans[] = {
+
+#ifdef HAVE_AFX25
+	{
+	"ax25", "ax25", &flag_ax25},
+#endif
+	{
+	"ip", "inet", &flag_inet},
+#ifdef HAVE_AFINET6
+	{
+	"ip6", "inet6", &flag_inet6},
+#endif
+#ifdef HAVE_AFIPX
+	{
+	"ipx", "ipx", &flag_ipx},
+#endif
+#ifdef HAVE_AFATALK
+	{
+	"appletalk", "ddp", &flag_ddp},
+#endif
+#ifdef HAVE_AFNETROM
+	{
+	"netrom", "netrom", &flag_netrom},
+#endif
+	{
+	"inet", "inet", &flag_inet},
+#ifdef HAVE_AFINET6
+	{
+	"inet6", "inet6", &flag_inet6},
+#endif
+#ifdef HAVE_AFATALK
+	{
+	"ddp", "ddp", &flag_ddp},
+#endif
+	{
+	"unix", "unix", &flag_unx}, {
+	"tcpip", "inet", &flag_inet},
+#ifdef HAVE_AFECONET
+	{
+	"econet", "ec", &flag_econet},
+#endif
+#ifdef HAVE_AFX25
+	{
+	"x25", "x25", &flag_x25},
+#endif
+#ifdef HAVE_AFASH
+	{
+	"ash", "ash", &flag_ash},
+#endif
+	{
+	0, 0, 0}
+};
+
+static char afname[256] = "";
+#endif							/* KEEP_UNUSED */
+
+#if HAVE_AFUNIX
+
+/* Display a UNIX domain address. */
+static char *UNIX_print(unsigned char *ptr)
+{
+	return (ptr);
+}
+
+
+/* Display a UNIX domain address. */
+static char *UNIX_sprint(struct sockaddr *sap, int numeric)
+{
+	static char buf[64];
+
+	if (sap->sa_family == 0xFFFF || sap->sa_family == 0)
+		return safe_strncpy(buf, _("[NONE SET]"), sizeof(buf));
+	return (UNIX_print(sap->sa_data));
+}
+
+
+static struct aftype unix_aftype = {
+	"unix", "UNIX Domain", AF_UNIX, 0,
+	UNIX_print, UNIX_sprint, NULL, NULL,
+	NULL, NULL, NULL,
+	-1,
+	"/proc/net/unix"
+};
+#endif							/* HAVE_AFUNIX */
+
+#if HAVE_AFINET
+
+#ifdef KEEP_UNUSED
+static void INET_reserror(char *text)
+{
+	herror(text);
+}
+
+/* Display an Internet socket address. */
+static char *INET_print(unsigned char *ptr)
+{
+	return (inet_ntoa((*(struct in_addr *) ptr)));
+}
+#endif							/* KEEP_UNUSED */
+
+/* Display an Internet socket address. */
+static char *INET_sprint(struct sockaddr *sap, int numeric)
+{
+	static char buff[128];
+
+	if (sap->sa_family == 0xFFFF || sap->sa_family == 0)
+		return safe_strncpy(buff, _("[NONE SET]"), sizeof(buff));
+
+	if (INET_rresolve(buff, sizeof(buff), (struct sockaddr_in *) sap,
+					  numeric, 0xffffff00) != 0)
+		return (NULL);
+
+	return (buff);
+}
+
+#ifdef KEEP_UNUSED
+static char *INET_sprintmask(struct sockaddr *sap, int numeric,
+							 unsigned int netmask)
+{
+	static char buff[128];
+
+	if (sap->sa_family == 0xFFFF || sap->sa_family == 0)
+		return safe_strncpy(buff, _("[NONE SET]"), sizeof(buff));
+	if (INET_rresolve(buff, sizeof(buff), (struct sockaddr_in *) sap,
+					  numeric, netmask) != 0)
+		return (NULL);
+	return (buff);
+}
+
+static int INET_getsock(char *bufp, struct sockaddr *sap)
+{
+	char *sp = bufp, *bp;
+	unsigned int i;
+	unsigned val;
+	struct sockaddr_in *sin;
+
+	sin = (struct sockaddr_in *) sap;
+	sin->sin_family = AF_INET;
+	sin->sin_port = 0;
+
+	val = 0;
+	bp = (char *) &val;
+	for (i = 0; i < sizeof(sin->sin_addr.s_addr); i++) {
+		*sp = toupper(*sp);
+
+		if ((*sp >= 'A') && (*sp <= 'F'))
+			bp[i] |= (int) (*sp - 'A') + 10;
+		else if ((*sp >= '0') && (*sp <= '9'))
+			bp[i] |= (int) (*sp - '0');
+		else
+			return (-1);
+
+		bp[i] <<= 4;
+		sp++;
+		*sp = toupper(*sp);
+
+		if ((*sp >= 'A') && (*sp <= 'F'))
+			bp[i] |= (int) (*sp - 'A') + 10;
+		else if ((*sp >= '0') && (*sp <= '9'))
+			bp[i] |= (int) (*sp - '0');
+		else
+			return (-1);
+
+		sp++;
+	}
+	sin->sin_addr.s_addr = htonl(val);
+
+	return (sp - bufp);
+}
+
+static int INET_input(int type, char *bufp, struct sockaddr *sap)
+{
+	switch (type) {
+	case 1:
+		return (INET_getsock(bufp, sap));
+	case 256:
+		return (INET_resolve(bufp, (struct sockaddr_in *) sap, 1));
+	default:
+		return (INET_resolve(bufp, (struct sockaddr_in *) sap, 0));
+	}
+}
+
+static int INET_getnetmask(char *adr, struct sockaddr *m, char *name)
+{
+	struct sockaddr_in *mask = (struct sockaddr_in *) m;
+	char *slash, *end;
+	int prefix;
+
+	if ((slash = strchr(adr, '/')) == NULL)
+		return 0;
+
+	*slash++ = '\0';
+	prefix = strtoul(slash, &end, 0);
+	if (*end != '\0')
+		return -1;
+
+	if (name) {
+		sprintf(name, "/%d", prefix);
+	}
+	mask->sin_family = AF_INET;
+	mask->sin_addr.s_addr = htonl(~(0xffffffffU >> prefix));
+	return 1;
+}
+#endif							/* KEEP_UNUSED */
+
+static struct aftype inet_aftype = {
+	"inet", "DARPA Internet", AF_INET, sizeof(unsigned long),
+	NULL /* UNUSED INET_print */ , INET_sprint,
+	NULL /* UNUSED INET_input */ , NULL /* UNUSED INET_reserror */ ,
+	NULL /*INET_rprint */ , NULL /*INET_rinput */ ,
+	NULL /* UNUSED INET_getnetmask */ ,
+	-1,
+	NULL
+};
+
+#endif							/* HAVE_AFINET */
+
+#if HAVE_AFINET6
+
+#ifdef KEEP_UNUSED
+static void INET6_reserror(char *text)
+{
+	herror(text);
+}
+
+/* Display an Internet socket address. */
+static char *INET6_print(unsigned char *ptr)
+{
+	static char name[80];
+
+	inet_ntop(AF_INET6, (struct in6_addr *) ptr, name, 80);
+	return name;
+}
+#endif							/* KEEP_UNUSED */
+
+/* Display an Internet socket address. */
+/* dirty! struct sockaddr usually doesn't suffer for inet6 addresses, fst. */
+static char *INET6_sprint(struct sockaddr *sap, int numeric)
+{
+	static char buff[128];
+
+	if (sap->sa_family == 0xFFFF || sap->sa_family == 0)
+		return safe_strncpy(buff, _("[NONE SET]"), sizeof(buff));
+	if (INET6_rresolve
+		(buff, sizeof(buff), (struct sockaddr_in6 *) sap, numeric) != 0)
+		return safe_strncpy(buff, _("[UNKNOWN]"), sizeof(buff));
+	return (buff);
+}
+
+#ifdef KEEP_UNUSED
+static int INET6_getsock(char *bufp, struct sockaddr *sap)
+{
+	struct sockaddr_in6 *sin6;
+
+	sin6 = (struct sockaddr_in6 *) sap;
+	sin6->sin6_family = AF_INET6;
+	sin6->sin6_port = 0;
+
+	if (inet_pton(AF_INET6, bufp, sin6->sin6_addr.s6_addr) <= 0)
+		return (-1);
+
+	return 16;			/* ?;) */
+}
+
+static int INET6_input(int type, char *bufp, struct sockaddr *sap)
+{
+	switch (type) {
+	case 1:
+		return (INET6_getsock(bufp, sap));
+	default:
+		return (INET6_resolve(bufp, (struct sockaddr_in6 *) sap));
+	}
+}
+#endif							/* KEEP_UNUSED */
+
+static struct aftype inet6_aftype = {
+	"inet6", "IPv6", AF_INET6, sizeof(struct in6_addr),
+	NULL /* UNUSED INET6_print */ , INET6_sprint,
+	NULL /* UNUSED INET6_input */ , NULL /* UNUSED INET6_reserror */ ,
+	NULL /*INET6_rprint */ , NULL /*INET6_rinput */ ,
+	NULL /* UNUSED INET6_getnetmask */ ,
+	-1,
+	NULL
+};
+
+#endif							/* HAVE_AFINET6 */
+
+/* Display an UNSPEC address. */
+static char *UNSPEC_print(unsigned char *ptr)
+{
+	static char buff[sizeof(struct sockaddr) * 3 + 1];
+	char *pos;
+	unsigned int i;
+
+	pos = buff;
+	for (i = 0; i < sizeof(struct sockaddr); i++) {
+		/* careful -- not every libc's sprintf returns # bytes written */
+		sprintf(pos, "%02X-", (*ptr++ & 0377));
+		pos += 3;
+	}
+	/* Erase trailing "-".  Works as long as sizeof(struct sockaddr) != 0 */
+	*--pos = '\0';
+	return (buff);
+}
+
+/* Display an UNSPEC socket address. */
+static char *UNSPEC_sprint(struct sockaddr *sap, int numeric)
+{
+	static char buf[64];
+
+	if (sap->sa_family == 0xFFFF || sap->sa_family == 0)
+		return safe_strncpy(buf, _("[NONE SET]"), sizeof(buf));
+	return (UNSPEC_print(sap->sa_data));
+}
+
+static struct aftype unspec_aftype = {
+	"unspec", "UNSPEC", AF_UNSPEC, 0,
+	UNSPEC_print, UNSPEC_sprint, NULL, NULL,
+	NULL,
+};
+
+static struct aftype * const aftypes[] = {
+#if HAVE_AFUNIX
+	&unix_aftype,
+#endif
+#if HAVE_AFINET
+	&inet_aftype,
+#endif
+#if HAVE_AFINET6
+	&inet6_aftype,
+#endif
+#if HAVE_AFAX25
+	&ax25_aftype,
+#endif
+#if HAVE_AFNETROM
+	&netrom_aftype,
+#endif
+#if HAVE_AFROSE
+	&rose_aftype,
+#endif
+#if HAVE_AFIPX
+	&ipx_aftype,
+#endif
+#if HAVE_AFATALK
+	&ddp_aftype,
+#endif
+#if HAVE_AFECONET
+	&ec_aftype,
+#endif
+#if HAVE_AFASH
+	&ash_aftype,
+#endif
+#if HAVE_AFX25
+	&x25_aftype,
+#endif
+	&unspec_aftype,
+	NULL
+};
+
+#ifdef KEEP_UNUSED
+static short sVafinit = 0;
+
+static void afinit()
+{
+	unspec_aftype.title = _("UNSPEC");
+#if HAVE_AFUNIX
+	unix_aftype.title = _("UNIX Domain");
+#endif
+#if HAVE_AFINET
+	inet_aftype.title = _("DARPA Internet");
+#endif
+#if HAVE_AFINET6
+	inet6_aftype.title = _("IPv6");
+#endif
+#if HAVE_AFAX25
+	ax25_aftype.title = _("AMPR AX.25");
+#endif
+#if HAVE_AFNETROM
+	netrom_aftype.title = _("AMPR NET/ROM");
+#endif
+#if HAVE_AFIPX
+	ipx_aftype.title = _("Novell IPX");
+#endif
+#if HAVE_AFATALK
+	ddp_aftype.title = _("Appletalk DDP");
+#endif
+#if HAVE_AFECONET
+	ec_aftype.title = _("Econet");
+#endif
+#if HAVE_AFX25
+	x25_aftype.title = _("CCITT X.25");
+#endif
+#if HAVE_AFROSE
+	rose_aftype.title = _("AMPR ROSE");
+#endif
+#if HAVE_AFASH
+	ash_aftype.title = _("Ash");
+#endif
+	sVafinit = 1;
+}
+
+static int aftrans_opt(const char *arg)
+{
+	struct aftrans_t *paft;
+	char *tmp1, *tmp2;
+	char buf[256];
+
+	safe_strncpy(buf, arg, sizeof(buf));
+
+	tmp1 = buf;
+
+	while (tmp1) {
+
+		tmp2 = strchr(tmp1, ',');
+
+		if (tmp2)
+			*(tmp2++) = '\0';
+
+		paft = aftrans;
+		for (paft = aftrans; paft->alias; paft++) {
+			if (strcmp(tmp1, paft->alias))
+				continue;
+			if (strlen(paft->name) + strlen(afname) + 1 >= sizeof(afname)) {
+				bb_error_msg(_("Too many address family arguments."));
+				return (0);
+			}
+			if (paft->flag)
+				(*paft->flag)++;
+			if (afname[0])
+				strcat(afname, ",");
+			strcat(afname, paft->name);
+			break;
+		}
+		if (!paft->alias) {
+			bb_error_msg(_("Unknown address family `%s'."), tmp1);
+			return (1);
+		}
+		tmp1 = tmp2;
+	}
+
+	return (0);
+}
+
+/* set the default AF list from the program name or a constant value    */
+static void aftrans_def(char *tool, char *argv0, char *dflt)
+{
+	char *tmp;
+	char *buf;
+
+	strcpy(afname, dflt);
+
+	if (!(tmp = strrchr(argv0, '/')))
+		tmp = argv0;	/* no slash?! */
+	else
+		tmp++;
+
+	if (!(buf = strdup(tmp)))
+		return;
+
+	if (strlen(tool) >= strlen(tmp)) {
+		free(buf);
+		return;
+	}
+	tmp = buf + (strlen(tmp) - strlen(tool));
+
+	if (strcmp(tmp, tool) != 0) {
+		free(buf);
+		return;
+	}
+	*tmp = '\0';
+	if ((tmp = strchr(buf, '_')))
+		*tmp = '\0';
+
+	afname[0] = '\0';
+	if (aftrans_opt(buf))
+		strcpy(afname, buf);
+
+	free(buf);
+}
+
+/* Check our protocol family table for this family. */
+static struct aftype *get_aftype(const char *name)
+{
+	struct aftype * const *afp;
+
+#ifdef KEEP_UNUSED
+	if (!sVafinit)
+		afinit();
+#endif							/* KEEP_UNUSED */
+
+	afp = aftypes;
+	while (*afp != NULL) {
+		if (!strcmp((*afp)->name, name))
+			return (*afp);
+		afp++;
+	}
+	if (strchr(name, ','))
+		bb_error_msg(_("Please don't supply more than one address family."));
+	return (NULL);
+}
+#endif							/* KEEP_UNUSED */
+
+/* Check our protocol family table for this family. */
+static struct aftype *get_afntype(int af)
+{
+	struct aftype * const *afp;
+
+#ifdef KEEP_UNUSED
+	if (!sVafinit)
+		afinit();
+#endif							/* KEEP_UNUSED */
+
+	afp = aftypes;
+	while (*afp != NULL) {
+		if ((*afp)->af == af)
+			return (*afp);
+		afp++;
+	}
+	return (NULL);
+}
+
+/* Check our protocol family table for this family and return its socket */
+static int get_socket_for_af(int af)
+{
+	struct aftype * const *afp;
+
+#ifdef KEEP_UNUSED
+	if (!sVafinit)
+		afinit();
+#endif							/* KEEP_UNUSED */
+
+	afp = aftypes;
+	while (*afp != NULL) {
+		if ((*afp)->af == af)
+			return (*afp)->fd;
+		afp++;
+	}
+	return -1;
+}
+
+#ifdef KEEP_UNUSED
+/* type: 0=all, 1=getroute */
+static void print_aflist(int type)
+{
+	int count = 0;
+	char *txt;
+	struct aftype * const *afp;
+
+#ifdef KEEP_UNUSED
+	if (!sVafinit)
+		afinit();
+#endif							/* KEEP_UNUSED */
+
+	afp = aftypes;
+	while (*afp != NULL) {
+		if ((type == 1 && ((*afp)->rprint == NULL)) || ((*afp)->af == 0)) {
+			afp++;
+			continue;
+		}
+		if ((count % 3) == 0)
+			fprintf(stderr, count ? "\n    " : "    ");
+		txt = (*afp)->name;
+		if (!txt)
+			txt = "..";
+		fprintf(stderr, "%s (%s) ", txt, _((*afp)->title));
+		count++;
+		afp++;
+	}
+	fprintf(stderr, "\n");
+}
+#endif							/* KEEP_UNUSED */
+
+struct user_net_device_stats {
+	unsigned long long rx_packets;	/* total packets received       */
+	unsigned long long tx_packets;	/* total packets transmitted    */
+	unsigned long long rx_bytes;	/* total bytes received         */
+	unsigned long long tx_bytes;	/* total bytes transmitted      */
+	unsigned long rx_errors;	/* bad packets received         */
+	unsigned long tx_errors;	/* packet transmit problems     */
+	unsigned long rx_dropped;	/* no space in linux buffers    */
+	unsigned long tx_dropped;	/* no space available in linux  */
+	unsigned long rx_multicast;	/* multicast packets received   */
+	unsigned long rx_compressed;
+	unsigned long tx_compressed;
+	unsigned long collisions;
+
+	/* detailed rx_errors: */
+	unsigned long rx_length_errors;
+	unsigned long rx_over_errors;	/* receiver ring buff overflow  */
+	unsigned long rx_crc_errors;	/* recved pkt with crc error    */
+	unsigned long rx_frame_errors;	/* recv'd frame alignment error */
+	unsigned long rx_fifo_errors;	/* recv'r fifo overrun          */
+	unsigned long rx_missed_errors;	/* receiver missed packet     */
+	/* detailed tx_errors */
+	unsigned long tx_aborted_errors;
+	unsigned long tx_carrier_errors;
+	unsigned long tx_fifo_errors;
+	unsigned long tx_heartbeat_errors;
+	unsigned long tx_window_errors;
+};
+
+struct interface {
+	struct interface *next, *prev;
+	char name[IFNAMSIZ];	/* interface name        */
+	short type;			/* if type               */
+	short flags;		/* various flags         */
+	int metric;			/* routing metric        */
+	int mtu;			/* MTU value             */
+	int tx_queue_len;	/* transmit queue length */
+	struct ifmap map;	/* hardware setup        */
+	struct sockaddr addr;	/* IP address            */
+	struct sockaddr dstaddr;	/* P-P IP address        */
+	struct sockaddr broadaddr;	/* IP broadcast address  */
+	struct sockaddr netmask;	/* IP network mask       */
+	struct sockaddr ipxaddr_bb;	/* IPX network address   */
+	struct sockaddr ipxaddr_sn;	/* IPX network address   */
+	struct sockaddr ipxaddr_e3;	/* IPX network address   */
+	struct sockaddr ipxaddr_e2;	/* IPX network address   */
+	struct sockaddr ddpaddr;	/* Appletalk DDP address */
+	struct sockaddr ecaddr;	/* Econet address        */
+	int has_ip;
+	int has_ipx_bb;
+	int has_ipx_sn;
+	int has_ipx_e3;
+	int has_ipx_e2;
+	int has_ax25;
+	int has_ddp;
+	int has_econet;
+	char hwaddr[32];	/* HW address            */
+	int statistics_valid;
+	struct user_net_device_stats stats;	/* statistics            */
+	int keepalive;		/* keepalive value for SLIP */
+	int outfill;		/* outfill value for SLIP */
+};
+
+
+int interface_opt_a = 0;	/* show all interfaces          */
+
+#ifdef KEEP_UNUSED
+static int opt_i = 0;	/* show the statistics          */
+static int opt_v = 0;	/* debugging output flag        */
+
+static int addr_family = 0;	/* currently selected AF        */
+#endif							/* KEEP_UNUSED */
+
+static struct interface *int_list, *int_last;
+static int skfd = -1;	/* generic raw socket desc.     */
+
+
+static int sockets_open(int family)
+{
+	struct aftype * const *aft;
+	int sfd = -1;
+	static int force = -1;
+
+	if (force < 0) {
+		force = 0;
+		if (get_kernel_revision() < KRELEASE(2, 1, 0))
+			force = 1;
+		if (access("/proc/net", R_OK))
+			force = 1;
+	}
+	for (aft = aftypes; *aft; aft++) {
+		struct aftype *af = *aft;
+		int type = SOCK_DGRAM;
+
+		if (af->af == AF_UNSPEC)
+			continue;
+		if (family && family != af->af)
+			continue;
+		if (af->fd != -1) {
+			sfd = af->fd;
+			continue;
+		}
+		/* Check some /proc file first to not stress kmod */
+		if (!family && !force && af->flag_file) {
+			if (access(af->flag_file, R_OK))
+				continue;
+		}
+#if HAVE_AFNETROM
+		if (af->af == AF_NETROM)
+			type = SOCK_SEQPACKET;
+#endif
+#if HAVE_AFX25
+		if (af->af == AF_X25)
+			type = SOCK_SEQPACKET;
+#endif
+		af->fd = socket(af->af, type, 0);
+		if (af->fd >= 0)
+			sfd = af->fd;
+	}
+	if (sfd < 0) {
+		bb_error_msg(_("No usable address families found."));
+	}
+	return sfd;
+}
+
+#ifdef CONFIG_FEATURE_CLEAN_UP
+static void sockets_close(void)
+{
+	struct aftype * const *aft;
+	for (aft = aftypes; *aft != NULL; aft++) {
+		struct aftype *af = *aft;
+		if( af->fd != -1 ) {
+			close(af->fd);
+			af->fd = -1;
+		}
+	}
+}
+#endif
+
+/* like strcmp(), but knows about numbers */
+static int nstrcmp(const char *a, const char *b)
+{
+	const char *a_ptr = a;
+	const char *b_ptr = b;
+
+	while (*a == *b) {
+		if (*a == '\0') {
+			return 0;
+		}
+		if (!isdigit(*a) && isdigit(*(a+1))) {
+			a_ptr = a+1;
+			b_ptr = b+1;
+		}
+		a++;
+		b++;
+	}
+
+	if (isdigit(*a) && isdigit(*b)) {
+		return atoi(a_ptr) > atoi(b_ptr) ? 1 : -1;
+	}
+	return *a - *b;
+}
+
+static struct interface *add_interface(char *name)
+{
+	struct interface *ife, **nextp, *new;
+
+	for (ife = int_last; ife; ife = ife->prev) {
+		int n = nstrcmp(ife->name, name);
+
+		if (n == 0)
+			return ife;
+		if (n < 0)
+			break;
+	}
+	new(new);
+	safe_strncpy(new->name, name, IFNAMSIZ);
+	nextp = ife ? &ife->next : &int_list;
+	new->prev = ife;
+	new->next = *nextp;
+	if (new->next)
+		new->next->prev = new;
+	else
+		int_last = new;
+	*nextp = new;
+	return new;
+}
+
+
+static int if_readconf(void)
+{
+	int numreqs = 30;
+	struct ifconf ifc;
+	struct ifreq *ifr;
+	int n, err = -1;
+	int skfd2;
+
+	/* SIOCGIFCONF currently seems to only work properly on AF_INET sockets
+	   (as of 2.1.128) */
+	skfd2 = get_socket_for_af(AF_INET);
+	if (skfd2 < 0) {
+		bb_perror_msg(("warning: no inet socket available"));
+		/* Try to soldier on with whatever socket we can get hold of.  */
+		skfd2 = sockets_open(0);
+		if (skfd2 < 0)
+			return -1;
+	}
+
+	ifc.ifc_buf = NULL;
+	for (;;) {
+		ifc.ifc_len = sizeof(struct ifreq) * numreqs;
+		ifc.ifc_buf = xrealloc(ifc.ifc_buf, ifc.ifc_len);
+
+		if (ioctl(skfd2, SIOCGIFCONF, &ifc) < 0) {
+			perror("SIOCGIFCONF");
+			goto out;
+		}
+		if (ifc.ifc_len == sizeof(struct ifreq) * numreqs) {
+			/* assume it overflowed and try again */
+			numreqs += 10;
+			continue;
+		}
+		break;
+	}
+
+	ifr = ifc.ifc_req;
+	for (n = 0; n < ifc.ifc_len; n += sizeof(struct ifreq)) {
+		add_interface(ifr->ifr_name);
+		ifr++;
+	}
+	err = 0;
+
+  out:
+	free(ifc.ifc_buf);
+	return err;
+}
+
+static char *get_name(char *name, char *p)
+{
+	/* Extract <name>[:<alias>] from nul-terminated p where p matches
+	   <name>[:<alias>]: after leading whitespace.
+	   If match is not made, set name empty and return unchanged p */
+	int namestart=0, nameend=0, aliasend;
+	while (isspace(p[namestart]))
+		namestart++;
+	nameend=namestart;
+	while (p[nameend] && p[nameend]!=':' && !isspace(p[nameend]))
+		nameend++;
+	if (p[nameend]==':') {
+		aliasend=nameend+1;
+		while (p[aliasend] && isdigit(p[aliasend]))
+			aliasend++;
+		if (p[aliasend]==':') {
+			nameend=aliasend;
+		}
+		if ((nameend-namestart)<IFNAMSIZ) {
+			memcpy(name,&p[namestart],nameend-namestart);
+			name[nameend-namestart]='\0';
+			p=&p[nameend];
+		} else {
+			/* Interface name too large */
+			name[0]='\0';
+		}
+	} else {
+		/* first ':' not found - return empty */
+		name[0]='\0';
+	}
+	return p + 1;
+}
+
+/* If scanf supports size qualifiers for %n conversions, then we can
+ * use a modified fmt that simply stores the position in the fields
+ * having no associated fields in the proc string.  Of course, we need
+ * to zero them again when we're done.  But that is smaller than the
+ * old approach of multiple scanf occurrences with large numbers of
+ * args. */
+
+/* static const char *ss_fmt[] = { */
+/* 	"%Ln%Lu%lu%lu%lu%lu%ln%ln%Ln%Lu%lu%lu%lu%lu%lu", */
+/* 	"%Lu%Lu%lu%lu%lu%lu%ln%ln%Lu%Lu%lu%lu%lu%lu%lu", */
+/* 	"%Lu%Lu%lu%lu%lu%lu%lu%lu%Lu%Lu%lu%lu%lu%lu%lu%lu" */
+/* }; */
+
+	/* Lie about the size of the int pointed to for %n. */
+#if INT_MAX == LONG_MAX
+static const char * const ss_fmt[] = {
+	"%n%Lu%u%u%u%u%n%n%n%Lu%u%u%u%u%u",
+	"%Lu%Lu%u%u%u%u%n%n%Lu%Lu%u%u%u%u%u",
+	"%Lu%Lu%u%u%u%u%u%u%Lu%Lu%u%u%u%u%u%u"
+};
+#else
+static const char * const ss_fmt[] = {
+	"%n%Lu%lu%lu%lu%lu%n%n%n%Lu%lu%lu%lu%lu%lu",
+	"%Lu%Lu%lu%lu%lu%lu%n%n%Lu%Lu%lu%lu%lu%lu%lu",
+	"%Lu%Lu%lu%lu%lu%lu%lu%lu%Lu%Lu%lu%lu%lu%lu%lu%lu"
+};
+
+#endif
+
+static void get_dev_fields(char *bp, struct interface *ife, int procnetdev_vsn)
+{
+	memset(&ife->stats, 0, sizeof(struct user_net_device_stats));
+
+	sscanf(bp, ss_fmt[procnetdev_vsn],
+		   &ife->stats.rx_bytes, /* missing for 0 */
+		   &ife->stats.rx_packets,
+		   &ife->stats.rx_errors,
+		   &ife->stats.rx_dropped,
+		   &ife->stats.rx_fifo_errors,
+		   &ife->stats.rx_frame_errors,
+		   &ife->stats.rx_compressed, /* missing for <= 1 */
+		   &ife->stats.rx_multicast, /* missing for <= 1 */
+		   &ife->stats.tx_bytes, /* missing for 0 */
+		   &ife->stats.tx_packets,
+		   &ife->stats.tx_errors,
+		   &ife->stats.tx_dropped,
+		   &ife->stats.tx_fifo_errors,
+		   &ife->stats.collisions,
+		   &ife->stats.tx_carrier_errors,
+		   &ife->stats.tx_compressed /* missing for <= 1 */
+		   );
+
+	if (procnetdev_vsn <= 1) {
+		if (procnetdev_vsn == 0) {
+			ife->stats.rx_bytes = 0;
+			ife->stats.tx_bytes = 0;
+		}
+		ife->stats.rx_multicast = 0;
+		ife->stats.rx_compressed = 0;
+		ife->stats.tx_compressed = 0;
+	}
+}
+
+static inline int procnetdev_version(char *buf)
+{
+	if (strstr(buf, "compressed"))
+		return 2;
+	if (strstr(buf, "bytes"))
+		return 1;
+	return 0;
+}
+
+static int if_readlist_proc(char *target)
+{
+	static int proc_read;
+	FILE *fh;
+	char buf[512];
+	struct interface *ife;
+	int err, procnetdev_vsn;
+
+	if (proc_read)
+		return 0;
+	if (!target)
+		proc_read = 1;
+
+	fh = fopen(_PATH_PROCNET_DEV, "r");
+	if (!fh) {
+		bb_perror_msg(_("Warning: cannot open %s. Limited output."), _PATH_PROCNET_DEV);
+		return if_readconf();
+	}
+	fgets(buf, sizeof buf, fh);	/* eat line */
+	fgets(buf, sizeof buf, fh);
+
+	procnetdev_vsn = procnetdev_version(buf);
+
+	err = 0;
+	while (fgets(buf, sizeof buf, fh)) {
+		char *s, name[128];
+
+		s = get_name(name, buf);
+		ife = add_interface(name);
+		get_dev_fields(s, ife, procnetdev_vsn);
+		ife->statistics_valid = 1;
+		if (target && !strcmp(target, name))
+			break;
+	}
+	if (ferror(fh)) {
+		perror(_PATH_PROCNET_DEV);
+		err = -1;
+		proc_read = 0;
+	}
+	fclose(fh);
+	return err;
+}
+
+static int if_readlist(void)
+{
+	int err = if_readlist_proc(NULL);
+
+	if (!err)
+		err = if_readconf();
+	return err;
+}
+
+static int for_all_interfaces(int (*doit) (struct interface *, void *),
+							  void *cookie)
+{
+	struct interface *ife;
+
+	if (!int_list && (if_readlist() < 0))
+		return -1;
+	for (ife = int_list; ife; ife = ife->next) {
+		int err = doit(ife, cookie);
+
+		if (err)
+			return err;
+	}
+	return 0;
+}
+
+/* Support for fetching an IPX address */
+
+#if HAVE_AFIPX
+static int ipx_getaddr(int sock, int ft, struct ifreq *ifr)
+{
+	((struct sockaddr_ipx *) &ifr->ifr_addr)->sipx_type = ft;
+	return ioctl(sock, SIOCGIFADDR, ifr);
+}
+#endif
+
+
+/* Fetch the interface configuration from the kernel. */
+static int if_fetch(struct interface *ife)
+{
+	struct ifreq ifr;
+	int fd;
+	char *ifname = ife->name;
+
+	strcpy(ifr.ifr_name, ifname);
+	if (ioctl(skfd, SIOCGIFFLAGS, &ifr) < 0)
+		return (-1);
+	ife->flags = ifr.ifr_flags;
+
+	strcpy(ifr.ifr_name, ifname);
+	if (ioctl(skfd, SIOCGIFHWADDR, &ifr) < 0)
+		memset(ife->hwaddr, 0, 32);
+	else
+		memcpy(ife->hwaddr, ifr.ifr_hwaddr.sa_data, 8);
+
+	ife->type = ifr.ifr_hwaddr.sa_family;
+
+	strcpy(ifr.ifr_name, ifname);
+	if (ioctl(skfd, SIOCGIFMETRIC, &ifr) < 0)
+		ife->metric = 0;
+	else
+		ife->metric = ifr.ifr_metric;
+
+	strcpy(ifr.ifr_name, ifname);
+	if (ioctl(skfd, SIOCGIFMTU, &ifr) < 0)
+		ife->mtu = 0;
+	else
+		ife->mtu = ifr.ifr_mtu;
+
+#ifdef HAVE_HWSLIP
+	if (ife->type == ARPHRD_SLIP || ife->type == ARPHRD_CSLIP ||
+		ife->type == ARPHRD_SLIP6 || ife->type == ARPHRD_CSLIP6 ||
+		ife->type == ARPHRD_ADAPT) {
+#ifdef SIOCGOUTFILL
+		strcpy(ifr.ifr_name, ifname);
+		if (ioctl(skfd, SIOCGOUTFILL, &ifr) < 0)
+			ife->outfill = 0;
+		else
+			ife->outfill = (unsigned int) ifr.ifr_data;
+#endif
+#ifdef SIOCGKEEPALIVE
+		strcpy(ifr.ifr_name, ifname);
+		if (ioctl(skfd, SIOCGKEEPALIVE, &ifr) < 0)
+			ife->keepalive = 0;
+		else
+			ife->keepalive = (unsigned int) ifr.ifr_data;
+#endif
+	}
+#endif
+
+#ifdef SIOCGIFMAP
+	strcpy(ifr.ifr_name, ifname);
+	if (ioctl(skfd, SIOCGIFMAP, &ifr) == 0)
+		ife->map = ifr.ifr_map;
+	else
+#endif
+		memset(&ife->map, 0, sizeof(struct ifmap));
+
+#ifdef HAVE_TXQUEUELEN
+	strcpy(ifr.ifr_name, ifname);
+	if (ioctl(skfd, SIOCGIFTXQLEN, &ifr) < 0)
+		ife->tx_queue_len = -1;	/* unknown value */
+	else
+		ife->tx_queue_len = ifr.ifr_qlen;
+#else
+	ife->tx_queue_len = -1;	/* unknown value */
+#endif
+
+#if HAVE_AFINET
+	/* IPv4 address? */
+	fd = get_socket_for_af(AF_INET);
+	if (fd >= 0) {
+		strcpy(ifr.ifr_name, ifname);
+		ifr.ifr_addr.sa_family = AF_INET;
+		if (ioctl(fd, SIOCGIFADDR, &ifr) == 0) {
+			ife->has_ip = 1;
+			ife->addr = ifr.ifr_addr;
+			strcpy(ifr.ifr_name, ifname);
+			if (ioctl(fd, SIOCGIFDSTADDR, &ifr) < 0)
+				memset(&ife->dstaddr, 0, sizeof(struct sockaddr));
+			else
+				ife->dstaddr = ifr.ifr_dstaddr;
+
+			strcpy(ifr.ifr_name, ifname);
+			if (ioctl(fd, SIOCGIFBRDADDR, &ifr) < 0)
+				memset(&ife->broadaddr, 0, sizeof(struct sockaddr));
+			else
+				ife->broadaddr = ifr.ifr_broadaddr;
+
+			strcpy(ifr.ifr_name, ifname);
+			if (ioctl(fd, SIOCGIFNETMASK, &ifr) < 0)
+				memset(&ife->netmask, 0, sizeof(struct sockaddr));
+			else
+				ife->netmask = ifr.ifr_netmask;
+		} else
+			memset(&ife->addr, 0, sizeof(struct sockaddr));
+	}
+#endif
+
+#if HAVE_AFATALK
+	/* DDP address maybe ? */
+	fd = get_socket_for_af(AF_APPLETALK);
+	if (fd >= 0) {
+		strcpy(ifr.ifr_name, ifname);
+		if (ioctl(fd, SIOCGIFADDR, &ifr) == 0) {
+			ife->ddpaddr = ifr.ifr_addr;
+			ife->has_ddp = 1;
+		}
+	}
+#endif
+
+#if HAVE_AFIPX
+	/* Look for IPX addresses with all framing types */
+	fd = get_socket_for_af(AF_IPX);
+	if (fd >= 0) {
+		strcpy(ifr.ifr_name, ifname);
+		if (!ipx_getaddr(fd, IPX_FRAME_ETHERII, &ifr)) {
+			ife->has_ipx_bb = 1;
+			ife->ipxaddr_bb = ifr.ifr_addr;
+		}
+		strcpy(ifr.ifr_name, ifname);
+		if (!ipx_getaddr(fd, IPX_FRAME_SNAP, &ifr)) {
+			ife->has_ipx_sn = 1;
+			ife->ipxaddr_sn = ifr.ifr_addr;
+		}
+		strcpy(ifr.ifr_name, ifname);
+		if (!ipx_getaddr(fd, IPX_FRAME_8023, &ifr)) {
+			ife->has_ipx_e3 = 1;
+			ife->ipxaddr_e3 = ifr.ifr_addr;
+		}
+		strcpy(ifr.ifr_name, ifname);
+		if (!ipx_getaddr(fd, IPX_FRAME_8022, &ifr)) {
+			ife->has_ipx_e2 = 1;
+			ife->ipxaddr_e2 = ifr.ifr_addr;
+		}
+	}
+#endif
+
+#if HAVE_AFECONET
+	/* Econet address maybe? */
+	fd = get_socket_for_af(AF_ECONET);
+	if (fd >= 0) {
+		strcpy(ifr.ifr_name, ifname);
+		if (ioctl(fd, SIOCGIFADDR, &ifr) == 0) {
+			ife->ecaddr = ifr.ifr_addr;
+			ife->has_econet = 1;
+		}
+	}
+#endif
+
+	return 0;
+}
+
+
+static int do_if_fetch(struct interface *ife)
+{
+	if (if_fetch(ife) < 0) {
+		char *errmsg;
+
+		if (errno == ENODEV) {
+			/* Give better error message for this case. */
+			errmsg = _("Device not found");
+		} else {
+			errmsg = strerror(errno);
+		}
+		bb_error_msg(_("%s: error fetching interface information: %s"),
+				ife->name, errmsg);
+		return -1;
+	}
+	return 0;
+}
+
+/* This structure defines hardware protocols and their handlers. */
+struct hwtype {
+	const char *name;
+	const char *title;
+	int type;
+	int alen;
+	char *(*print) (unsigned char *);
+	int (*input) (char *, struct sockaddr *);
+	int (*activate) (int fd);
+	int suppress_null_addr;
+};
+
+static const struct hwtype unspec_hwtype = {
+	"unspec", "UNSPEC", -1, 0,
+	UNSPEC_print, NULL, NULL
+};
+
+static const struct hwtype loop_hwtype = {
+	"loop", "Local Loopback", ARPHRD_LOOPBACK, 0,
+	NULL, NULL, NULL
+};
+
+#if HAVE_HWETHER
+#include <net/if_arp.h>
+
+#if (__GLIBC__ >=2 && __GLIBC_MINOR >= 1) || defined(_NEWLIB_VERSION)
+#include <net/ethernet.h>
+#else
+#include <linux/if_ether.h>
+#endif
+
+/* Display an Ethernet address in readable format. */
+static char *pr_ether(unsigned char *ptr)
+{
+	static char buff[64];
+
+	snprintf(buff, sizeof(buff), "%02X:%02X:%02X:%02X:%02X:%02X",
+			 (ptr[0] & 0377), (ptr[1] & 0377), (ptr[2] & 0377),
+			 (ptr[3] & 0377), (ptr[4] & 0377), (ptr[5] & 0377)
+		);
+	return (buff);
+}
+
+#ifdef KEEP_UNUSED
+/* Input an Ethernet address and convert to binary. */
+static int in_ether(char *bufp, struct sockaddr *sap)
+{
+	unsigned char *ptr;
+	char c, *orig;
+	int i;
+	unsigned val;
+
+	sap->sa_family = ether_hwtype.type;
+	ptr = sap->sa_data;
+
+	i = 0;
+	orig = bufp;
+	while ((*bufp != '\0') && (i < ETH_ALEN)) {
+		val = 0;
+		c = *bufp++;
+		if (isdigit(c))
+			val = c - '0';
+		else if (c >= 'a' && c <= 'f')
+			val = c - 'a' + 10;
+		else if (c >= 'A' && c <= 'F')
+			val = c - 'A' + 10;
+		else {
+#ifdef DEBUG
+			bb_error_msg(_("in_ether(%s): invalid ether address!\n"), orig);
+#endif
+			errno = EINVAL;
+			return (-1);
+		}
+		val <<= 4;
+		c = *bufp;
+		if (isdigit(c))
+			val |= c - '0';
+		else if (c >= 'a' && c <= 'f')
+			val |= c - 'a' + 10;
+		else if (c >= 'A' && c <= 'F')
+			val |= c - 'A' + 10;
+		else if (c == ':' || c == 0)
+			val >>= 4;
+		else {
+#ifdef DEBUG
+			bb_error_msg(_("in_ether(%s): invalid ether address!"), orig);
+#endif
+			errno = EINVAL;
+			return (-1);
+		}
+		if (c != 0)
+			bufp++;
+		*ptr++ = (unsigned char) (val & 0377);
+		i++;
+
+		/* We might get a semicolon here - not required. */
+		if (*bufp == ':') {
+#ifdef DEBUG
+			if (i == ETH_ALEN) {
+				bb_error_msg(_("in_ether(%s): trailing : ignored!"), orig);
+			}
+#endif
+			bufp++;
+		}
+	}
+
+#ifdef DEBUG
+	/* That's it.  Any trailing junk? */
+	if ((i == ETH_ALEN) && (*bufp != '\0')) {
+		bb_error_msg(_("in_ether(%s): trailing junk!"), orig);
+		errno = EINVAL;
+		return (-1);
+	}
+	bb_error_msg("in_ether(%s): %s", orig, pr_ether(sap->sa_data));
+#endif
+
+	return (0);
+}
+#endif							/* KEEP_UNUSED */
+
+
+static const struct hwtype ether_hwtype = {
+	"ether", "Ethernet", ARPHRD_ETHER, ETH_ALEN,
+	pr_ether, NULL /* UNUSED in_ether */ , NULL
+};
+
+
+#endif							/* HAVE_HWETHER */
+
+
+#if HAVE_HWPPP
+
+#include <net/if_arp.h>
+
+#ifdef KEEP_UNUSED
+/* Start the PPP encapsulation on the file descriptor. */
+static int do_ppp(int fd)
+{
+	bb_error_msg(_("You cannot start PPP with this program."));
+	return -1;
+}
+#endif							/* KEEP_UNUSED */
+
+static const struct hwtype ppp_hwtype = {
+	"ppp", "Point-Point Protocol", ARPHRD_PPP, 0,
+	NULL, NULL, NULL /* UNUSED do_ppp */ , 0
+};
+
+
+#endif							/* HAVE_PPP */
+
+static const struct hwtype * const hwtypes[] = {
+
+	&loop_hwtype,
+
+#if HAVE_HWSLIP
+	&slip_hwtype,
+	&cslip_hwtype,
+	&slip6_hwtype,
+	&cslip6_hwtype,
+	&adaptive_hwtype,
+#endif
+#if HAVE_HWSTRIP
+	&strip_hwtype,
+#endif
+#if HAVE_HWASH
+	&ash_hwtype,
+#endif
+#if HAVE_HWETHER
+	&ether_hwtype,
+#endif
+#if HAVE_HWTR
+	&tr_hwtype,
+#ifdef ARPHRD_IEEE802_TR
+	&tr_hwtype1,
+#endif
+#endif
+#if HAVE_HWAX25
+	&ax25_hwtype,
+#endif
+#if HAVE_HWNETROM
+	&netrom_hwtype,
+#endif
+#if HAVE_HWROSE
+	&rose_hwtype,
+#endif
+#if HAVE_HWTUNNEL
+	&tunnel_hwtype,
+#endif
+#if HAVE_HWPPP
+	&ppp_hwtype,
+#endif
+#if HAVE_HWHDLCLAPB
+	&hdlc_hwtype,
+	&lapb_hwtype,
+#endif
+#if HAVE_HWARC
+	&arcnet_hwtype,
+#endif
+#if HAVE_HWFR
+	&dlci_hwtype,
+	&frad_hwtype,
+#endif
+#if HAVE_HWSIT
+	&sit_hwtype,
+#endif
+#if HAVE_HWFDDI
+	&fddi_hwtype,
+#endif
+#if HAVE_HWHIPPI
+	&hippi_hwtype,
+#endif
+#if HAVE_HWIRDA
+	&irda_hwtype,
+#endif
+#if HAVE_HWEC
+	&ec_hwtype,
+#endif
+#if HAVE_HWX25
+	&x25_hwtype,
+#endif
+	&unspec_hwtype,
+	NULL
+};
+
+#ifdef KEEP_UNUSED
+static short sVhwinit = 0;
+
+static void hwinit()
+{
+	loop_hwtype.title = _("Local Loopback");
+	unspec_hwtype.title = _("UNSPEC");
+#if HAVE_HWSLIP
+	slip_hwtype.title = _("Serial Line IP");
+	cslip_hwtype.title = _("VJ Serial Line IP");
+	slip6_hwtype.title = _("6-bit Serial Line IP");
+	cslip6_hwtype.title = _("VJ 6-bit Serial Line IP");
+	adaptive_hwtype.title = _("Adaptive Serial Line IP");
+#endif
+#if HAVE_HWETHER
+	ether_hwtype.title = _("Ethernet");
+#endif
+#if HAVE_HWASH
+	ash_hwtype.title = _("Ash");
+#endif
+#if HAVE_HWFDDI
+	fddi_hwtype.title = _("Fiber Distributed Data Interface");
+#endif
+#if HAVE_HWHIPPI
+	hippi_hwtype.title = _("HIPPI");
+#endif
+#if HAVE_HWAX25
+	ax25_hwtype.title = _("AMPR AX.25");
+#endif
+#if HAVE_HWROSE
+	rose_hwtype.title = _("AMPR ROSE");
+#endif
+#if HAVE_HWNETROM
+	netrom_hwtype.title = _("AMPR NET/ROM");
+#endif
+#if HAVE_HWX25
+	x25_hwtype.title = _("generic X.25");
+#endif
+#if HAVE_HWTUNNEL
+	tunnel_hwtype.title = _("IPIP Tunnel");
+#endif
+#if HAVE_HWPPP
+	ppp_hwtype.title = _("Point-to-Point Protocol");
+#endif
+#if HAVE_HWHDLCLAPB
+	hdlc_hwtype.title = _("(Cisco)-HDLC");
+	lapb_hwtype.title = _("LAPB");
+#endif
+#if HAVE_HWARC
+	arcnet_hwtype.title = _("ARCnet");
+#endif
+#if HAVE_HWFR
+	dlci_hwtype.title = _("Frame Relay DLCI");
+	frad_hwtype.title = _("Frame Relay Access Device");
+#endif
+#if HAVE_HWSIT
+	sit_hwtype.title = _("IPv6-in-IPv4");
+#endif
+#if HAVE_HWIRDA
+	irda_hwtype.title = _("IrLAP");
+#endif
+#if HAVE_HWTR
+	tr_hwtype.title = _("16/4 Mbps Token Ring");
+#ifdef ARPHRD_IEEE802_TR
+	tr_hwtype1.title = _("16/4 Mbps Token Ring (New)");
+#endif
+#endif
+#if HAVE_HWEC
+	ec_hwtype.title = _("Econet");
+#endif
+	sVhwinit = 1;
+}
+#endif							/* KEEP_UNUSED */
+
+#ifdef IFF_PORTSEL
+#if 0
+static const char * const if_port_text[][4] = {
+	/* Keep in step with <linux/netdevice.h> */
+	{"unknown", NULL, NULL, NULL},
+	{"10base2", "bnc", "coax", NULL},
+	{"10baseT", "utp", "tpe", NULL},
+	{"AUI", "thick", "db15", NULL},
+	{"100baseT", NULL, NULL, NULL},
+	{"100baseTX", NULL, NULL, NULL},
+	{"100baseFX", NULL, NULL, NULL},
+	{NULL, NULL, NULL, NULL},
+};
+#else
+static const char * const if_port_text[] = {
+	/* Keep in step with <linux/netdevice.h> */
+	"unknown",
+	"10base2",
+	"10baseT",
+	"AUI",
+	"100baseT",
+	"100baseTX",
+	"100baseFX",
+	NULL
+};
+#endif
+#endif
+
+/* Check our hardware type table for this type. */
+static const struct hwtype *get_hwntype(int type)
+{
+	const struct hwtype * const *hwp;
+
+#ifdef KEEP_UNUSED
+	if (!sVhwinit)
+		hwinit();
+#endif							/* KEEP_UNUSED */
+
+	hwp = hwtypes;
+	while (*hwp != NULL) {
+		if ((*hwp)->type == type)
+			return (*hwp);
+		hwp++;
+	}
+	return (NULL);
+}
+
+/* return 1 if address is all zeros */
+static int hw_null_address(const struct hwtype *hw, void *ap)
+{
+	unsigned int i;
+	unsigned char *address = (unsigned char *) ap;
+
+	for (i = 0; i < hw->alen; i++)
+		if (address[i])
+			return 0;
+	return 1;
+}
+
+static const char TRext[] = "\0\0\0Ki\0Mi\0Gi\0Ti";
+
+static void print_bytes_scaled(unsigned long long ull, const char *end)
+{
+	unsigned long long int_part;
+	const char *ext;
+	unsigned int frac_part;
+	int i;
+
+	frac_part = 0;
+	ext = TRext;
+	int_part = ull;
+	i = 4;
+	do {
+#if 0
+		/* This does correct rounding and is a little larger.  But it
+		 * uses KiB as the smallest displayed unit. */
+		if ((int_part < (1024*1024 - 51)) || !--i) {
+			i = 0;
+			int_part += 51;		/* 1024*.05 = 51.2 */
+			frac_part = ((((unsigned int) int_part) & (1024-1)) * 10) / 1024;
+		}
+		int_part /= 1024;
+		ext += 3;	/* KiB, MiB, GiB, TiB */
+#else
+		if (int_part >= 1024) {
+			frac_part = ((((unsigned int) int_part) & (1024-1)) * 10) / 1024;
+			int_part /= 1024;
+			ext += 3;	/* KiB, MiB, GiB, TiB */
+		}
+		--i;
+#endif
+	} while (i);
+
+	printf("X bytes:%Lu (%Lu.%u %sB)%s", ull, int_part, frac_part, ext, end);
+}
+
+static const char * const ife_print_flags_strs[] = {
+	"UP ",
+	"BROADCAST ",
+	"DEBUG ",
+	"LOOPBACK ",
+	"POINTOPOINT ",
+	"NOTRAILERS ",
+	"RUNNING ",
+	"NOARP ",
+	"PROMISC ",
+	"ALLMULTI ",
+	"SLAVE ",
+	"MASTER ",
+	"MULTICAST ",
+#ifdef HAVE_DYNAMIC
+	"DYNAMIC "
+#endif
+};
+
+static const unsigned short ife_print_flags_mask[] = {
+	IFF_UP,
+	IFF_BROADCAST,
+	IFF_DEBUG,
+	IFF_LOOPBACK,
+	IFF_POINTOPOINT,
+	IFF_NOTRAILERS,
+	IFF_RUNNING,
+	IFF_NOARP,
+	IFF_PROMISC,
+	IFF_ALLMULTI,
+	IFF_SLAVE,
+	IFF_MASTER,
+	IFF_MULTICAST,
+#ifdef HAVE_DYNAMIC
+	IFF_DYNAMIC
+#endif
+	0
+};
+
+static void ife_print(struct interface *ptr)
+{
+	struct aftype *ap;
+	const struct hwtype *hw;
+	int hf;
+	int can_compress = 0;
+
+#if HAVE_AFIPX
+	static struct aftype *ipxtype = NULL;
+#endif
+#if HAVE_AFECONET
+	static struct aftype *ectype = NULL;
+#endif
+#if HAVE_AFATALK
+	static struct aftype *ddptype = NULL;
+#endif
+#if HAVE_AFINET6
+	FILE *f;
+	char addr6[40], devname[20];
+	struct sockaddr_in6 sap;
+	int plen, scope, dad_status, if_idx;
+	char addr6p[8][5];
+#endif
+
+	ap = get_afntype(ptr->addr.sa_family);
+	if (ap == NULL)
+		ap = get_afntype(0);
+
+	hf = ptr->type;
+
+	if (hf == ARPHRD_CSLIP || hf == ARPHRD_CSLIP6)
+		can_compress = 1;
+
+	hw = get_hwntype(hf);
+	if (hw == NULL)
+		hw = get_hwntype(-1);
+
+	printf(_("%-9.9s Link encap:%s  "), ptr->name, _(hw->title));
+	/* For some hardware types (eg Ash, ATM) we don't print the
+	   hardware address if it's null.  */
+	if (hw->print != NULL && (!(hw_null_address(hw, ptr->hwaddr) &&
+								hw->suppress_null_addr)))
+		printf(_("HWaddr %s  "), hw->print(ptr->hwaddr));
+#ifdef IFF_PORTSEL
+	if (ptr->flags & IFF_PORTSEL) {
+		printf(_("Media:%s"), if_port_text[ptr->map.port] /* [0] */);
+		if (ptr->flags & IFF_AUTOMEDIA)
+			printf(_("(auto)"));
+	}
+#endif
+	printf("\n");
+
+#if HAVE_AFINET
+	if (ptr->has_ip) {
+		printf(_("          %s addr:%s "), ap->name,
+			   ap->sprint(&ptr->addr, 1));
+		if (ptr->flags & IFF_POINTOPOINT) {
+			printf(_(" P-t-P:%s "), ap->sprint(&ptr->dstaddr, 1));
+		}
+		if (ptr->flags & IFF_BROADCAST) {
+			printf(_(" Bcast:%s "), ap->sprint(&ptr->broadaddr, 1));
+		}
+		printf(_(" Mask:%s\n"), ap->sprint(&ptr->netmask, 1));
+	}
+#endif
+
+#if HAVE_AFINET6
+
+#define IPV6_ADDR_ANY           0x0000U
+
+#define IPV6_ADDR_UNICAST       0x0001U
+#define IPV6_ADDR_MULTICAST     0x0002U
+#define IPV6_ADDR_ANYCAST       0x0004U
+
+#define IPV6_ADDR_LOOPBACK      0x0010U
+#define IPV6_ADDR_LINKLOCAL     0x0020U
+#define IPV6_ADDR_SITELOCAL     0x0040U
+
+#define IPV6_ADDR_COMPATv4      0x0080U
+
+#define IPV6_ADDR_SCOPE_MASK    0x00f0U
+
+#define IPV6_ADDR_MAPPED        0x1000U
+#define IPV6_ADDR_RESERVED      0x2000U	/* reserved address space */
+
+	if ((f = fopen(_PATH_PROCNET_IFINET6, "r")) != NULL) {
+		while (fscanf
+			   (f, "%4s%4s%4s%4s%4s%4s%4s%4s %02x %02x %02x %02x %20s\n",
+				addr6p[0], addr6p[1], addr6p[2], addr6p[3], addr6p[4],
+				addr6p[5], addr6p[6], addr6p[7], &if_idx, &plen, &scope,
+				&dad_status, devname) != EOF) {
+			if (!strcmp(devname, ptr->name)) {
+				sprintf(addr6, "%s:%s:%s:%s:%s:%s:%s:%s",
+						addr6p[0], addr6p[1], addr6p[2], addr6p[3],
+						addr6p[4], addr6p[5], addr6p[6], addr6p[7]);
+				inet_pton(AF_INET6, addr6,
+						  (struct sockaddr *) &sap.sin6_addr);
+				sap.sin6_family = AF_INET6;
+				printf(_("          inet6 addr: %s/%d"),
+					   inet6_aftype.sprint((struct sockaddr *) &sap, 1),
+					   plen);
+				printf(_(" Scope:"));
+				switch (scope & IPV6_ADDR_SCOPE_MASK) {
+				case 0:
+					printf(_("Global"));
+					break;
+				case IPV6_ADDR_LINKLOCAL:
+					printf(_("Link"));
+					break;
+				case IPV6_ADDR_SITELOCAL:
+					printf(_("Site"));
+					break;
+				case IPV6_ADDR_COMPATv4:
+					printf(_("Compat"));
+					break;
+				case IPV6_ADDR_LOOPBACK:
+					printf(_("Host"));
+					break;
+				default:
+					printf(_("Unknown"));
+				}
+				printf("\n");
+			}
+		}
+		fclose(f);
+	}
+#endif
+
+#if HAVE_AFIPX
+	if (ipxtype == NULL)
+		ipxtype = get_afntype(AF_IPX);
+
+	if (ipxtype != NULL) {
+		if (ptr->has_ipx_bb)
+			printf(_("          IPX/Ethernet II addr:%s\n"),
+				   ipxtype->sprint(&ptr->ipxaddr_bb, 1));
+		if (ptr->has_ipx_sn)
+			printf(_("          IPX/Ethernet SNAP addr:%s\n"),
+				   ipxtype->sprint(&ptr->ipxaddr_sn, 1));
+		if (ptr->has_ipx_e2)
+			printf(_("          IPX/Ethernet 802.2 addr:%s\n"),
+				   ipxtype->sprint(&ptr->ipxaddr_e2, 1));
+		if (ptr->has_ipx_e3)
+			printf(_("          IPX/Ethernet 802.3 addr:%s\n"),
+				   ipxtype->sprint(&ptr->ipxaddr_e3, 1));
+	}
+#endif
+
+#if HAVE_AFATALK
+	if (ddptype == NULL)
+		ddptype = get_afntype(AF_APPLETALK);
+	if (ddptype != NULL) {
+		if (ptr->has_ddp)
+			printf(_("          EtherTalk Phase 2 addr:%s\n"),
+				   ddptype->sprint(&ptr->ddpaddr, 1));
+	}
+#endif
+
+#if HAVE_AFECONET
+	if (ectype == NULL)
+		ectype = get_afntype(AF_ECONET);
+	if (ectype != NULL) {
+		if (ptr->has_econet)
+			printf(_("          econet addr:%s\n"),
+				   ectype->sprint(&ptr->ecaddr, 1));
+	}
+#endif
+
+	printf("          ");
+	/* DONT FORGET TO ADD THE FLAGS IN ife_print_short, too */
+
+	if (ptr->flags == 0) {
+		printf(_("[NO FLAGS] "));
+	} else {
+		int i = 0;
+		do {
+			if (ptr->flags & ife_print_flags_mask[i]) {
+				printf(_(ife_print_flags_strs[i]));
+			}
+		} while (ife_print_flags_mask[++i]);
+	}
+
+	/* DONT FORGET TO ADD THE FLAGS IN ife_print_short */
+	printf(_(" MTU:%d  Metric:%d"), ptr->mtu, ptr->metric ? ptr->metric : 1);
+#ifdef SIOCSKEEPALIVE
+	if (ptr->outfill || ptr->keepalive)
+		printf(_("  Outfill:%d  Keepalive:%d"), ptr->outfill, ptr->keepalive);
+#endif
+	printf("\n");
+
+	/* If needed, display the interface statistics. */
+
+	if (ptr->statistics_valid) {
+		/* XXX: statistics are currently only printed for the primary address,
+		 *      not for the aliases, although strictly speaking they're shared
+		 *      by all addresses.
+		 */
+		printf("          ");
+
+		printf(_("RX packets:%Lu errors:%lu dropped:%lu overruns:%lu frame:%lu\n"),
+			   ptr->stats.rx_packets, ptr->stats.rx_errors,
+			   ptr->stats.rx_dropped, ptr->stats.rx_fifo_errors,
+			   ptr->stats.rx_frame_errors);
+		if (can_compress)
+			printf(_("             compressed:%lu\n"),
+				   ptr->stats.rx_compressed);
+		printf("          ");
+		printf(_("TX packets:%Lu errors:%lu dropped:%lu overruns:%lu carrier:%lu\n"),
+			   ptr->stats.tx_packets, ptr->stats.tx_errors,
+			   ptr->stats.tx_dropped, ptr->stats.tx_fifo_errors,
+			   ptr->stats.tx_carrier_errors);
+		printf(_("          collisions:%lu "), ptr->stats.collisions);
+		if (can_compress)
+			printf(_("compressed:%lu "), ptr->stats.tx_compressed);
+		if (ptr->tx_queue_len != -1)
+			printf(_("txqueuelen:%d "), ptr->tx_queue_len);
+		printf("\n          R");
+		print_bytes_scaled(ptr->stats.rx_bytes, "  T");
+		print_bytes_scaled(ptr->stats.tx_bytes, "\n");
+
+	}
+
+	if ((ptr->map.irq || ptr->map.mem_start || ptr->map.dma ||
+		 ptr->map.base_addr)) {
+		printf("          ");
+		if (ptr->map.irq)
+			printf(_("Interrupt:%d "), ptr->map.irq);
+		if (ptr->map.base_addr >= 0x100)	/* Only print devices using it for
+											   I/O maps */
+			printf(_("Base address:0x%lx "),
+				   (unsigned long) ptr->map.base_addr);
+		if (ptr->map.mem_start) {
+			printf(_("Memory:%lx-%lx "), ptr->map.mem_start,
+				   ptr->map.mem_end);
+		}
+		if (ptr->map.dma)
+			printf(_("DMA chan:%x "), ptr->map.dma);
+		printf("\n");
+	}
+	printf("\n");
+}
+
+
+static int do_if_print(struct interface *ife, void *cookie)
+{
+	int *opt_a = (int *) cookie;
+	int res;
+
+	res = do_if_fetch(ife);
+	if (res >= 0) {
+		if ((ife->flags & IFF_UP) || *opt_a)
+			ife_print(ife);
+	}
+	return res;
+}
+
+static struct interface *lookup_interface(char *name)
+{
+	struct interface *ife = NULL;
+
+	if (if_readlist_proc(name) < 0)
+		return NULL;
+	ife = add_interface(name);
+	return ife;
+}
+
+/* for ipv4 add/del modes */
+static int if_print(char *ifname)
+{
+	int res;
+
+	if (!ifname) {
+		res = for_all_interfaces(do_if_print, &interface_opt_a);
+	} else {
+		struct interface *ife;
+
+		ife = lookup_interface(ifname);
+		res = do_if_fetch(ife);
+		if (res >= 0)
+			ife_print(ife);
+	}
+	return res;
+}
+
+int display_interfaces(char *ifname)
+{
+	int status;
+
+	/* Create a channel to the NET kernel. */
+	if ((skfd = sockets_open(0)) < 0) {
+		bb_perror_msg_and_die("socket");
+	}
+
+	/* Do we have to show the current setup? */
+	status = if_print(ifname);
+#ifdef CONFIG_FEATURE_CLEAN_UP
+	sockets_close();
+#endif
+	exit(status < 0);
+}