summaryrefslogtreecommitdiffstats
path: root/freebsd-userspace/commands/usr.bin/netstat/main.c
diff options
context:
space:
mode:
Diffstat (limited to 'freebsd-userspace/commands/usr.bin/netstat/main.c')
-rw-r--r--freebsd-userspace/commands/usr.bin/netstat/main.c805
1 files changed, 805 insertions, 0 deletions
diff --git a/freebsd-userspace/commands/usr.bin/netstat/main.c b/freebsd-userspace/commands/usr.bin/netstat/main.c
new file mode 100644
index 00000000..09b26a26
--- /dev/null
+++ b/freebsd-userspace/commands/usr.bin/netstat/main.c
@@ -0,0 +1,805 @@
+/*-
+ * Copyright (c) 1983, 1988, 1993
+ * 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 the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * 3. All advertising materials mentioning features or use of this software
+ * must display the following acknowledgement:
+ * This product includes software developed by the University of
+ * California, Berkeley and its contributors.
+ * 4. 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 BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#ifndef lint
+char const copyright[] =
+"@(#) Copyright (c) 1983, 1988, 1993\n\
+ Regents of the University of California. All rights reserved.\n";
+#endif /* not lint */
+
+#if 0
+#ifndef lint
+static char sccsid[] = "@(#)main.c 8.4 (Berkeley) 3/1/94";
+#endif /* not lint */
+#endif
+
+#include <sys/cdefs.h>
+__FBSDID("$FreeBSD$");
+
+#include <sys/param.h>
+#include <sys/file.h>
+#ifdef __rtems__
+#include <freebsd/sys/protosw.h>
+#else
+#include <sys/protosw.h>
+#endif
+#include <sys/socket.h>
+#include <sys/socketvar.h>
+
+#include <netinet/in.h>
+
+#ifdef NETGRAPH
+#include <netgraph/ng_socket.h>
+#endif
+
+#include <ctype.h>
+#include <err.h>
+#include <errno.h>
+#include <kvm.h>
+#include <limits.h>
+#include <netdb.h>
+#include <nlist.h>
+#include <paths.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include "netstat.h"
+
+static struct nlist nl[] = {
+#define N_IFNET 0
+ { .n_name = "_ifnet" },
+#define N_RTSTAT 1
+ { .n_name = "_rtstat" },
+#define N_RTREE 2
+ { .n_name = "_rt_tables"},
+#define N_MRTSTAT 3
+ { .n_name = "_mrtstat" },
+#define N_MFCHASHTBL 4
+ { .n_name = "_mfchashtbl" },
+#define N_VIFTABLE 5
+ { .n_name = "_viftable" },
+#define N_IPX 6
+ { .n_name = "_ipxpcb_list"},
+#define N_IPXSTAT 7
+ { .n_name = "_ipxstat"},
+#define N_SPXSTAT 8
+ { .n_name = "_spx_istat"},
+#define N_DDPSTAT 9
+ { .n_name = "_ddpstat"},
+#define N_DDPCB 10
+ { .n_name = "_ddpcb"},
+#define N_NGSOCKS 11
+ { .n_name = "_ngsocklist"},
+#define N_IP6STAT 12
+ { .n_name = "_ip6stat" },
+#define N_ICMP6STAT 13
+ { .n_name = "_icmp6stat" },
+#define N_IPSECSTAT 14
+ { .n_name = "_ipsec4stat" },
+#define N_IPSEC6STAT 15
+ { .n_name = "_ipsec6stat" },
+#define N_PIM6STAT 16
+ { .n_name = "_pim6stat" },
+#define N_MRT6STAT 17
+ { .n_name = "_mrt6stat" },
+#define N_MF6CTABLE 18
+ { .n_name = "_mf6ctable" },
+#define N_MIF6TABLE 19
+ { .n_name = "_mif6table" },
+#define N_PFKEYSTAT 20
+ { .n_name = "_pfkeystat" },
+#define N_MBSTAT 21
+ { .n_name = "_mbstat" },
+#define N_MBTYPES 22
+ { .n_name = "_mbtypes" },
+#define N_NMBCLUSTERS 23
+ { .n_name = "_nmbclusters" },
+#define N_NMBUFS 24
+ { .n_name = "_nmbufs" },
+#define N_MBHI 25
+ { .n_name = "_mbuf_hiwm" },
+#define N_CLHI 26
+ { .n_name = "_clust_hiwm" },
+#define N_NCPUS 27
+ { .n_name = "_smp_cpus" },
+#define N_PAGESZ 28
+ { .n_name = "_pagesize" },
+#define N_MBPSTAT 29
+ { .n_name = "_mb_statpcpu" },
+#define N_RTTRASH 30
+ { .n_name = "_rttrash" },
+#define N_MBLO 31
+ { .n_name = "_mbuf_lowm" },
+#define N_CLLO 32
+ { .n_name = "_clust_lowm" },
+#define N_CARPSTAT 33
+ { .n_name = "_carpstats" },
+#define N_PFSYNCSTAT 34
+ { .n_name = "_pfsyncstats" },
+#define N_AHSTAT 35
+ { .n_name = "_ahstat" },
+#define N_ESPSTAT 36
+ { .n_name = "_espstat" },
+#define N_IPCOMPSTAT 37
+ { .n_name = "_ipcompstat" },
+#define N_TCPSTAT 38
+ { .n_name = "_tcpstat" },
+#define N_UDPSTAT 39
+ { .n_name = "_udpstat" },
+#define N_IPSTAT 40
+ { .n_name = "_ipstat" },
+#define N_ICMPSTAT 41
+ { .n_name = "_icmpstat" },
+#define N_IGMPSTAT 42
+ { .n_name = "_igmpstat" },
+#define N_PIMSTAT 43
+ { .n_name = "_pimstat" },
+#define N_TCBINFO 44
+ { .n_name = "_tcbinfo" },
+#define N_UDBINFO 45
+ { .n_name = "_udbinfo" },
+#define N_DIVCBINFO 46
+ { .n_name = "_divcbinfo" },
+#define N_RIPCBINFO 47
+ { .n_name = "_ripcbinfo" },
+#define N_UNP_COUNT 48
+ { .n_name = "_unp_count" },
+#define N_UNP_GENCNT 49
+ { .n_name = "_unp_gencnt" },
+#define N_UNP_DHEAD 50
+ { .n_name = "_unp_dhead" },
+#define N_UNP_SHEAD 51
+ { .n_name = "_unp_shead" },
+#define N_RIP6STAT 52
+ { .n_name = "_rip6stat" },
+#define N_SCTPSTAT 53
+ { .n_name = "_sctpstat" },
+#define N_MFCTABLESIZE 54
+ { .n_name = "_mfctablesize" },
+#define N_ARPSTAT 55
+ { .n_name = "_arpstat" },
+ { .n_name = NULL },
+};
+
+struct protox {
+ int pr_index; /* index into nlist of cb head */
+ int pr_sindex; /* index into nlist of stat block */
+ u_char pr_wanted; /* 1 if wanted, 0 otherwise */
+ void (*pr_cblocks)(u_long, const char *, int, int);
+ /* control blocks printing routine */
+ void (*pr_stats)(u_long, const char *, int, int);
+ /* statistics printing routine */
+ void (*pr_istats)(char *); /* per/if statistics printing routine */
+ const char *pr_name; /* well-known name */
+ int pr_usesysctl; /* non-zero if we use sysctl, not kvm */
+ int pr_protocol;
+} protox[] = {
+ { N_TCBINFO, N_TCPSTAT, 1, protopr,
+ tcp_stats, NULL, "tcp", 1, IPPROTO_TCP },
+ { N_UDBINFO, N_UDPSTAT, 1, protopr,
+ udp_stats, NULL, "udp", 1, IPPROTO_UDP },
+#ifdef SCTP
+ { -1, N_SCTPSTAT, 1, sctp_protopr,
+ sctp_stats, NULL, "sctp", 1, IPPROTO_SCTP },
+#endif
+ { N_DIVCBINFO, -1, 1, protopr,
+ NULL, NULL, "divert", 1, IPPROTO_DIVERT },
+ { N_RIPCBINFO, N_IPSTAT, 1, protopr,
+ ip_stats, NULL, "ip", 1, IPPROTO_RAW },
+ { N_RIPCBINFO, N_ICMPSTAT, 1, protopr,
+ icmp_stats, NULL, "icmp", 1, IPPROTO_ICMP },
+ { N_RIPCBINFO, N_IGMPSTAT, 1, protopr,
+ igmp_stats, NULL, "igmp", 1, IPPROTO_IGMP },
+#ifdef IPSEC
+ { -1, N_IPSECSTAT, 1, NULL, /* keep as compat */
+ ipsec_stats, NULL, "ipsec", 0, 0},
+ { -1, N_AHSTAT, 1, NULL,
+ ah_stats, NULL, "ah", 0, 0},
+ { -1, N_ESPSTAT, 1, NULL,
+ esp_stats, NULL, "esp", 0, 0},
+ { -1, N_IPCOMPSTAT, 1, NULL,
+ ipcomp_stats, NULL, "ipcomp", 0, 0},
+#endif
+ { N_RIPCBINFO, N_PIMSTAT, 1, protopr,
+ pim_stats, NULL, "pim", 1, IPPROTO_PIM },
+ { -1, N_CARPSTAT, 1, NULL,
+ carp_stats, NULL, "carp", 1, 0 },
+ { -1, N_PFSYNCSTAT, 1, NULL,
+ pfsync_stats, NULL, "pfsync", 1, 0 },
+ { -1, N_ARPSTAT, 1, NULL,
+ arp_stats, NULL, "arp", 1, 0 },
+ { -1, -1, 0, NULL,
+ NULL, NULL, NULL, 0, 0 }
+};
+
+#ifdef INET6
+struct protox ip6protox[] = {
+ { N_TCBINFO, N_TCPSTAT, 1, protopr,
+ tcp_stats, NULL, "tcp", 1, IPPROTO_TCP },
+ { N_UDBINFO, N_UDPSTAT, 1, protopr,
+ udp_stats, NULL, "udp", 1, IPPROTO_UDP },
+ { N_RIPCBINFO, N_IP6STAT, 1, protopr,
+ ip6_stats, ip6_ifstats, "ip6", 1, IPPROTO_RAW },
+ { N_RIPCBINFO, N_ICMP6STAT, 1, protopr,
+ icmp6_stats, icmp6_ifstats, "icmp6", 1, IPPROTO_ICMPV6 },
+#ifdef IPSEC
+ { -1, N_IPSEC6STAT, 1, NULL,
+ ipsec_stats, NULL, "ipsec6", 0, 0 },
+#endif
+#ifdef notyet
+ { -1, N_PIM6STAT, 1, NULL,
+ pim6_stats, NULL, "pim6", 1, 0 },
+#endif
+ { -1, N_RIP6STAT, 1, NULL,
+ rip6_stats, NULL, "rip6", 1, 0 },
+ { -1, -1, 0, NULL,
+ NULL, NULL, NULL, 0, 0 }
+};
+#endif /*INET6*/
+
+#ifdef IPSEC
+struct protox pfkeyprotox[] = {
+ { -1, N_PFKEYSTAT, 1, NULL,
+ pfkey_stats, NULL, "pfkey", 0, 0 },
+ { -1, -1, 0, NULL,
+ NULL, NULL, NULL, 0, 0 }
+};
+#endif
+
+struct protox atalkprotox[] = {
+ { N_DDPCB, N_DDPSTAT, 1, atalkprotopr,
+ ddp_stats, NULL, "ddp", 0, 0 },
+ { -1, -1, 0, NULL,
+ NULL, NULL, NULL, 0, 0 }
+};
+#ifdef NETGRAPH
+struct protox netgraphprotox[] = {
+ { N_NGSOCKS, -1, 1, netgraphprotopr,
+ NULL, NULL, "ctrl", 0, 0 },
+ { N_NGSOCKS, -1, 1, netgraphprotopr,
+ NULL, NULL, "data", 0, 0 },
+ { -1, -1, 0, NULL,
+ NULL, NULL, NULL, 0, 0 }
+};
+#endif
+#ifdef IPX
+struct protox ipxprotox[] = {
+ { N_IPX, N_IPXSTAT, 1, ipxprotopr,
+ ipx_stats, NULL, "ipx", 0, 0 },
+ { N_IPX, N_SPXSTAT, 1, ipxprotopr,
+ spx_stats, NULL, "spx", 0, 0 },
+ { -1, -1, 0, NULL,
+ NULL, NULL, 0, 0, 0 }
+};
+#endif
+
+struct protox *protoprotox[] = {
+ protox,
+#ifdef INET6
+ ip6protox,
+#endif
+#ifdef IPSEC
+ pfkeyprotox,
+#endif
+#ifdef IPX
+ ipxprotox,
+#endif
+ atalkprotox, NULL };
+
+static void printproto(struct protox *, const char *);
+static void usage(void);
+static struct protox *name2protox(const char *);
+static struct protox *knownname(const char *);
+
+static kvm_t *kvmd;
+static char *nlistf = NULL, *memf = NULL;
+
+int Aflag; /* show addresses of protocol control block */
+int aflag; /* show all sockets (including servers) */
+int Bflag; /* show information about bpf consumers */
+int bflag; /* show i/f total bytes in/out */
+int dflag; /* show i/f dropped packets */
+int gflag; /* show group (multicast) routing or stats */
+int hflag; /* show counters in human readable format */
+int iflag; /* show interfaces */
+int Lflag; /* show size of listen queues */
+int mflag; /* show memory stats */
+int noutputs = 0; /* how much outputs before we exit */
+int numeric_addr; /* show addresses numerically */
+int numeric_port; /* show ports numerically */
+static int pflag; /* show given protocol */
+int rflag; /* show routing tables (or routing stats) */
+int sflag; /* show protocol statistics */
+int tflag; /* show i/f watchdog timers */
+int Wflag; /* wide display */
+int xflag; /* extra information, includes all socket buffer info */
+int zflag; /* zero stats */
+
+int interval; /* repeat interval for i/f stats */
+
+char *interface; /* desired i/f for stats, or NULL for all i/fs */
+int unit; /* unit number for above */
+
+int af; /* address family */
+int live; /* true if we are examining a live system */
+
+int
+main(int argc, char *argv[])
+{
+ struct protox *tp = NULL; /* for printing cblocks & stats */
+ int ch;
+
+ af = AF_UNSPEC;
+
+ while ((ch = getopt(argc, argv, "AaBbdf:ghI:iLlM:mN:np:q:rSstuWw:xz")) != -1)
+ switch(ch) {
+ case 'A':
+ Aflag = 1;
+ break;
+ case 'a':
+ aflag = 1;
+ break;
+ case 'B':
+ Bflag = 1;
+ break;
+ case 'b':
+ bflag = 1;
+ break;
+ case 'd':
+ dflag = 1;
+ break;
+ case 'f':
+ if (strcmp(optarg, "ipx") == 0)
+ af = AF_IPX;
+ else if (strcmp(optarg, "inet") == 0)
+ af = AF_INET;
+#ifdef INET6
+ else if (strcmp(optarg, "inet6") == 0)
+ af = AF_INET6;
+#endif
+#ifdef IPSEC
+ else if (strcmp(optarg, "pfkey") == 0)
+ af = PF_KEY;
+#endif
+ else if (strcmp(optarg, "unix") == 0)
+ af = AF_UNIX;
+ else if (strcmp(optarg, "atalk") == 0)
+ af = AF_APPLETALK;
+#ifdef NETGRAPH
+ else if (strcmp(optarg, "ng") == 0
+ || strcmp(optarg, "netgraph") == 0)
+ af = AF_NETGRAPH;
+#endif
+ else if (strcmp(optarg, "link") == 0)
+ af = AF_LINK;
+ else {
+ errx(1, "%s: unknown address family", optarg);
+ }
+ break;
+ case 'g':
+ gflag = 1;
+ break;
+ case 'h':
+ hflag = 1;
+ break;
+ case 'I': {
+ char *cp;
+
+ iflag = 1;
+ for (cp = interface = optarg; isalpha(*cp); cp++)
+ continue;
+ unit = atoi(cp);
+ break;
+ }
+ case 'i':
+ iflag = 1;
+ break;
+ case 'L':
+ Lflag = 1;
+ break;
+ case 'M':
+ memf = optarg;
+ break;
+ case 'm':
+ mflag = 1;
+ break;
+ case 'N':
+ nlistf = optarg;
+ break;
+ case 'n':
+ numeric_addr = numeric_port = 1;
+ break;
+ case 'p':
+ if ((tp = name2protox(optarg)) == NULL) {
+ errx(1,
+ "%s: unknown or uninstrumented protocol",
+ optarg);
+ }
+ pflag = 1;
+ break;
+ case 'q':
+ noutputs = atoi(optarg);
+ if (noutputs != 0)
+ noutputs++;
+ break;
+ case 'r':
+ rflag = 1;
+ break;
+ case 's':
+ ++sflag;
+ break;
+ case 'S':
+ numeric_addr = 1;
+ break;
+ case 't':
+ tflag = 1;
+ break;
+ case 'u':
+ af = AF_UNIX;
+ break;
+ case 'W':
+ case 'l':
+ Wflag = 1;
+ break;
+ case 'w':
+ interval = atoi(optarg);
+ iflag = 1;
+ break;
+ case 'x':
+ xflag = 1;
+ break;
+ case 'z':
+ zflag = 1;
+ break;
+ case '?':
+ default:
+ usage();
+ }
+ argv += optind;
+ argc -= optind;
+
+#define BACKWARD_COMPATIBILITY
+#ifdef BACKWARD_COMPATIBILITY
+ if (*argv) {
+ if (isdigit(**argv)) {
+ interval = atoi(*argv);
+ if (interval <= 0)
+ usage();
+ ++argv;
+ iflag = 1;
+ }
+ if (*argv) {
+ nlistf = *argv;
+ if (*++argv)
+ memf = *argv;
+ }
+ }
+#endif
+
+ /*
+ * Discard setgid privileges if not the running kernel so that bad
+ * guys can't print interesting stuff from kernel memory.
+ */
+ live = (nlistf == NULL && memf == NULL);
+ if (!live)
+ setgid(getgid());
+
+ if (Bflag) {
+ if (!live)
+ usage();
+ bpf_stats(interface);
+ exit(0);
+ }
+ if (mflag) {
+ if (!live) {
+ if (kread(0, NULL, 0) == 0)
+ mbpr(kvmd, nl[N_MBSTAT].n_value);
+ } else
+ mbpr(NULL, 0);
+ exit(0);
+ }
+#if 0
+ /*
+ * Keep file descriptors open to avoid overhead
+ * of open/close on each call to get* routines.
+ */
+ sethostent(1);
+ setnetent(1);
+#else
+ /*
+ * This does not make sense any more with DNS being default over
+ * the files. Doing a setXXXXent(1) causes a tcp connection to be
+ * used for the queries, which is slower.
+ */
+#endif
+ kread(0, NULL, 0);
+ if (iflag && !sflag) {
+ intpr(interval, nl[N_IFNET].n_value, NULL);
+ exit(0);
+ }
+ if (rflag) {
+ if (sflag)
+ rt_stats(nl[N_RTSTAT].n_value, nl[N_RTTRASH].n_value);
+ else
+ routepr(nl[N_RTREE].n_value);
+ exit(0);
+ }
+ if (gflag) {
+ if (sflag) {
+ if (af == AF_INET || af == AF_UNSPEC)
+ mrt_stats(nl[N_MRTSTAT].n_value);
+#ifdef INET6
+ if (af == AF_INET6 || af == AF_UNSPEC)
+ mrt6_stats(nl[N_MRT6STAT].n_value);
+#endif
+ } else {
+ if (af == AF_INET || af == AF_UNSPEC)
+ mroutepr(nl[N_MFCHASHTBL].n_value,
+ nl[N_MFCTABLESIZE].n_value,
+ nl[N_VIFTABLE].n_value);
+#ifdef INET6
+ if (af == AF_INET6 || af == AF_UNSPEC)
+ mroute6pr(nl[N_MF6CTABLE].n_value,
+ nl[N_MIF6TABLE].n_value);
+#endif
+ }
+ exit(0);
+ }
+
+ if (tp) {
+ printproto(tp, tp->pr_name);
+ exit(0);
+ }
+ if (af == AF_INET || af == AF_UNSPEC)
+ for (tp = protox; tp->pr_name; tp++)
+ printproto(tp, tp->pr_name);
+#ifdef INET6
+ if (af == AF_INET6 || af == AF_UNSPEC)
+ for (tp = ip6protox; tp->pr_name; tp++)
+ printproto(tp, tp->pr_name);
+#endif /*INET6*/
+#ifdef IPSEC
+ if (af == PF_KEY || af == AF_UNSPEC)
+ for (tp = pfkeyprotox; tp->pr_name; tp++)
+ printproto(tp, tp->pr_name);
+#endif /*IPSEC*/
+#ifdef IPX
+ if (af == AF_IPX || af == AF_UNSPEC) {
+ for (tp = ipxprotox; tp->pr_name; tp++)
+ printproto(tp, tp->pr_name);
+ }
+#endif /* IPX */
+ if (af == AF_APPLETALK || af == AF_UNSPEC)
+ for (tp = atalkprotox; tp->pr_name; tp++)
+ printproto(tp, tp->pr_name);
+#ifdef NETGRAPH
+ if (af == AF_NETGRAPH || af == AF_UNSPEC)
+ for (tp = netgraphprotox; tp->pr_name; tp++)
+ printproto(tp, tp->pr_name);
+#endif /* NETGRAPH */
+ if ((af == AF_UNIX || af == AF_UNSPEC) && !sflag)
+ unixpr(nl[N_UNP_COUNT].n_value, nl[N_UNP_GENCNT].n_value,
+ nl[N_UNP_DHEAD].n_value, nl[N_UNP_SHEAD].n_value);
+ exit(0);
+}
+
+/*
+ * Print out protocol statistics or control blocks (per sflag).
+ * If the interface was not specifically requested, and the symbol
+ * is not in the namelist, ignore this one.
+ */
+static void
+printproto(tp, name)
+ struct protox *tp;
+ const char *name;
+{
+ void (*pr)(u_long, const char *, int, int);
+ u_long off;
+
+ if (sflag) {
+ if (iflag) {
+ if (tp->pr_istats)
+ intpr(interval, nl[N_IFNET].n_value,
+ tp->pr_istats);
+ else if (pflag)
+ printf("%s: no per-interface stats routine\n",
+ tp->pr_name);
+ return;
+ } else {
+ pr = tp->pr_stats;
+ if (!pr) {
+ if (pflag)
+ printf("%s: no stats routine\n",
+ tp->pr_name);
+ return;
+ }
+ if (tp->pr_usesysctl && live)
+ off = 0;
+ else if (tp->pr_sindex < 0) {
+ if (pflag)
+ printf(
+ "%s: stats routine doesn't work on cores\n",
+ tp->pr_name);
+ return;
+ } else
+ off = nl[tp->pr_sindex].n_value;
+ }
+ } else {
+ pr = tp->pr_cblocks;
+ if (!pr) {
+ if (pflag)
+ printf("%s: no PCB routine\n", tp->pr_name);
+ return;
+ }
+ if (tp->pr_usesysctl && live)
+ off = 0;
+ else if (tp->pr_index < 0) {
+ if (pflag)
+ printf(
+ "%s: PCB routine doesn't work on cores\n",
+ tp->pr_name);
+ return;
+ } else
+ off = nl[tp->pr_index].n_value;
+ }
+ if (pr != NULL && (off || (live && tp->pr_usesysctl) ||
+ af != AF_UNSPEC))
+ (*pr)(off, name, af, tp->pr_protocol);
+}
+
+/*
+ * Read kernel memory, return 0 on success.
+ */
+int
+kread(u_long addr, void *buf, size_t size)
+{
+ char errbuf[_POSIX2_LINE_MAX];
+
+ if (kvmd == NULL) {
+ kvmd = kvm_openfiles(nlistf, memf, NULL, O_RDONLY, errbuf);
+ setgid(getgid());
+ if (kvmd != NULL) {
+ if (kvm_nlist(kvmd, nl) < 0) {
+ if (nlistf)
+ errx(1, "%s: kvm_nlist: %s", nlistf,
+ kvm_geterr(kvmd));
+ else
+ errx(1, "kvm_nlist: %s", kvm_geterr(kvmd));
+ }
+
+ if (nl[0].n_type == 0) {
+ if (nlistf)
+ errx(1, "%s: no namelist", nlistf);
+ else
+ errx(1, "no namelist");
+ }
+ } else {
+ warnx("kvm not available: %s", errbuf);
+ return(-1);
+ }
+ }
+ if (!buf)
+ return (0);
+ if (kvm_read(kvmd, addr, buf, size) != (ssize_t)size) {
+ warnx("%s", kvm_geterr(kvmd));
+ return (-1);
+ }
+ return (0);
+}
+
+const char *
+plural(uintmax_t n)
+{
+ return (n != 1 ? "s" : "");
+}
+
+const char *
+plurales(uintmax_t n)
+{
+ return (n != 1 ? "es" : "");
+}
+
+const char *
+pluralies(uintmax_t n)
+{
+ return (n != 1 ? "ies" : "y");
+}
+
+/*
+ * Find the protox for the given "well-known" name.
+ */
+static struct protox *
+knownname(const char *name)
+{
+ struct protox **tpp, *tp;
+
+ for (tpp = protoprotox; *tpp; tpp++)
+ for (tp = *tpp; tp->pr_name; tp++)
+ if (strcmp(tp->pr_name, name) == 0)
+ return (tp);
+ return (NULL);
+}
+
+/*
+ * Find the protox corresponding to name.
+ */
+static struct protox *
+name2protox(const char *name)
+{
+ struct protox *tp;
+ char **alias; /* alias from p->aliases */
+ struct protoent *p;
+
+ /*
+ * Try to find the name in the list of "well-known" names. If that
+ * fails, check if name is an alias for an Internet protocol.
+ */
+ if ((tp = knownname(name)) != NULL)
+ return (tp);
+
+ setprotoent(1); /* make protocol lookup cheaper */
+ while ((p = getprotoent()) != NULL) {
+ /* assert: name not same as p->name */
+ for (alias = p->p_aliases; *alias; alias++)
+ if (strcmp(name, *alias) == 0) {
+ endprotoent();
+ return (knownname(p->p_name));
+ }
+ }
+ endprotoent();
+ return (NULL);
+}
+
+static void
+usage(void)
+{
+ (void)fprintf(stderr, "%s\n%s\n%s\n%s\n%s\n%s\n%s\n%s\n%s\n%s\n%s\n",
+"usage: netstat [-AaLnSWx] [-f protocol_family | -p protocol]\n"
+" [-M core] [-N system]",
+" netstat -i | -I interface [-abdhntW] [-f address_family]\n"
+" [-M core] [-N system]",
+" netstat -w wait [-I interface] [-d] [-M core] [-N system] [-q howmany]",
+" netstat -s [-s] [-z] [-f protocol_family | -p protocol]\n"
+" [-M core] [-N system]",
+" netstat -i | -I interface -s [-f protocol_family | -p protocol]\n"
+" [-M core] [-N system]",
+" netstat -m [-M core] [-N system]",
+" netstat -B [-I interface]",
+" netstat -r [-AanW] [-f address_family] [-M core] [-N system]",
+" netstat -rs [-s] [-M core] [-N system]",
+" netstat -g [-W] [-f address_family] [-M core] [-N system]",
+" netstat -gs [-s] [-f address_family] [-M core] [-N system]");
+ exit(1);
+}