Index: head/sys/netinet6/ip6_mroute.c =================================================================== --- head/sys/netinet6/ip6_mroute.c (revision 281171) +++ head/sys/netinet6/ip6_mroute.c (revision 281172) @@ -1,1951 +1,1976 @@ /*- * Copyright (C) 1998 WIDE Project. * 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. Neither the name of the project 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 PROJECT 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 PROJECT 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. * * $KAME: ip6_mroute.c,v 1.58 2001/12/18 02:36:31 itojun Exp $ */ /*- * Copyright (c) 1989 Stephen Deering * Copyright (c) 1992, 1993 * The Regents of the University of California. All rights reserved. * * This code is derived from software contributed to Berkeley by * Stephen Deering of Stanford University. * * 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. * 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. * * @(#)ip_mroute.c 8.2 (Berkeley) 11/15/93 * BSDI ip_mroute.c,v 2.10 1996/11/14 00:29:52 jch Exp */ /* * IP multicast forwarding procedures * * Written by David Waitzman, BBN Labs, August 1988. * Modified by Steve Deering, Stanford, February 1989. * Modified by Mark J. Steiglitz, Stanford, May, 1991 * Modified by Van Jacobson, LBL, January 1993 * Modified by Ajit Thyagarajan, PARC, August 1993 * Modified by Bill Fenner, PARC, April 1994 * * MROUTING Revision: 3.5.1.2 + PIM-SMv2 (pimd) Support */ #include __FBSDID("$FreeBSD$"); #include "opt_inet6.h" #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include static MALLOC_DEFINE(M_MRTABLE6, "mf6c", "multicast forwarding cache entry"); static int ip6_mdq(struct mbuf *, struct ifnet *, struct mf6c *); static void phyint_send(struct ip6_hdr *, struct mif6 *, struct mbuf *); static int register_send(struct ip6_hdr *, struct mif6 *, struct mbuf *); static int set_pim6(int *); static int socket_send(struct socket *, struct mbuf *, struct sockaddr_in6 *); extern int in6_mcast_loop; extern struct domain inet6domain; static const struct encaptab *pim6_encap_cookie; static const struct protosw in6_pim_protosw = { .pr_type = SOCK_RAW, .pr_domain = &inet6domain, .pr_protocol = IPPROTO_PIM, .pr_flags = PR_ATOMIC|PR_ADDR|PR_LASTHDR, .pr_input = pim6_input, .pr_output = rip6_output, .pr_ctloutput = rip6_ctloutput, .pr_usrreqs = &rip6_usrreqs }; static int pim6_encapcheck(const struct mbuf *, int, int, void *); static VNET_DEFINE(int, ip6_mrouter_ver) = 0; #define V_ip6_mrouter_ver VNET(ip6_mrouter_ver) SYSCTL_DECL(_net_inet6); SYSCTL_DECL(_net_inet6_ip6); static SYSCTL_NODE(_net_inet6, IPPROTO_PIM, pim, CTLFLAG_RW, 0, "PIM"); static struct mrt6stat mrt6stat; SYSCTL_STRUCT(_net_inet6_ip6, OID_AUTO, mrt6stat, CTLFLAG_RW, &mrt6stat, mrt6stat, "Multicast Routing Statistics (struct mrt6stat, netinet6/ip6_mroute.h)"); #define MRT6STAT_INC(name) mrt6stat.name += 1 #define NO_RTE_FOUND 0x1 #define RTE_FOUND 0x2 static struct mtx mrouter6_mtx; #define MROUTER6_LOCK() mtx_lock(&mrouter6_mtx) #define MROUTER6_UNLOCK() mtx_unlock(&mrouter6_mtx) #define MROUTER6_LOCK_ASSERT() do { \ mtx_assert(&mrouter6_mtx, MA_OWNED); \ NET_ASSERT_GIANT(); \ } while (0) #define MROUTER6_LOCK_INIT() \ mtx_init(&mrouter6_mtx, "IPv6 multicast forwarding", NULL, MTX_DEF) #define MROUTER6_LOCK_DESTROY() mtx_destroy(&mrouter6_mtx) static struct mf6c *mf6ctable[MF6CTBLSIZ]; SYSCTL_OPAQUE(_net_inet6_ip6, OID_AUTO, mf6ctable, CTLFLAG_RD, &mf6ctable, sizeof(mf6ctable), "S,*mf6ctable[MF6CTBLSIZ]", "IPv6 Multicast Forwarding Table (struct *mf6ctable[MF6CTBLSIZ], " "netinet6/ip6_mroute.h)"); static struct mtx mfc6_mtx; #define MFC6_LOCK() mtx_lock(&mfc6_mtx) #define MFC6_UNLOCK() mtx_unlock(&mfc6_mtx) #define MFC6_LOCK_ASSERT() do { \ mtx_assert(&mfc6_mtx, MA_OWNED); \ NET_ASSERT_GIANT(); \ } while (0) #define MFC6_LOCK_INIT() \ mtx_init(&mfc6_mtx, "IPv6 multicast forwarding cache", NULL, MTX_DEF) #define MFC6_LOCK_DESTROY() mtx_destroy(&mfc6_mtx) static u_char n6expire[MF6CTBLSIZ]; static struct mif6 mif6table[MAXMIFS]; -SYSCTL_OPAQUE(_net_inet6_ip6, OID_AUTO, mif6table, CTLFLAG_RD, - &mif6table, sizeof(mif6table), "S,mif6[MAXMIFS]", - "IPv6 Multicast Interfaces (struct mif6[MAXMIFS], netinet6/ip6_mroute.h)"); +static int +sysctl_mif6table(SYSCTL_HANDLER_ARGS) +{ + struct mif6_sctl *out; + int error; + + out = malloc(sizeof(struct mif6_sctl) * MAXMIFS, M_TEMP, M_WAITOK); + for (int i = 0; i < MAXMIFS; i++) { + out[i].m6_flags = mif6table[i].m6_flags; + out[i].m6_rate_limit = mif6table[i].m6_rate_limit; + out[i].m6_lcl_addr = mif6table[i].m6_lcl_addr; + if (mif6table[i].m6_ifp != NULL) + out[i].m6_ifp = mif6table[i].m6_ifp->if_index; + else + out[i].m6_ifp = 0; + out[i].m6_pkt_in = mif6table[i].m6_pkt_in; + out[i].m6_pkt_out = mif6table[i].m6_pkt_out; + out[i].m6_bytes_in = mif6table[i].m6_bytes_in; + out[i].m6_bytes_out = mif6table[i].m6_bytes_out; + } + error = SYSCTL_OUT(req, out, sizeof(struct mif6_sctl) * MAXMIFS); + free(out, M_TEMP); + return (error); +} +SYSCTL_PROC(_net_inet6_ip6, OID_AUTO, mif6table, CTLTYPE_OPAQUE | CTLFLAG_RD, + NULL, 0, sysctl_mif6table, "S,mif6_sctl[MAXMIFS]", + "IPv6 Multicast Interfaces (struct mif6_sctl[MAXMIFS], " + "netinet6/ip6_mroute.h)"); static struct mtx mif6_mtx; #define MIF6_LOCK() mtx_lock(&mif6_mtx) #define MIF6_UNLOCK() mtx_unlock(&mif6_mtx) #define MIF6_LOCK_ASSERT() mtx_assert(&mif6_mtx, MA_OWNED) #define MIF6_LOCK_INIT() \ mtx_init(&mif6_mtx, "IPv6 multicast interfaces", NULL, MTX_DEF) #define MIF6_LOCK_DESTROY() mtx_destroy(&mif6_mtx) #ifdef MRT6DEBUG static VNET_DEFINE(u_int, mrt6debug) = 0; /* debug level */ #define V_mrt6debug VNET(mrt6debug) #define DEBUG_MFC 0x02 #define DEBUG_FORWARD 0x04 #define DEBUG_EXPIRE 0x08 #define DEBUG_XMIT 0x10 #define DEBUG_REG 0x20 #define DEBUG_PIM 0x40 #define DEBUG_ERR 0x80 #define DEBUG_ANY 0x7f #define MRT6_DLOG(m, fmt, ...) \ if (V_mrt6debug & (m)) \ log(((m) & DEBUG_ERR) ? LOG_ERR: LOG_DEBUG, \ "%s: " fmt "\n", __func__, ##__VA_ARGS__) #else #define MRT6_DLOG(m, fmt, ...) #endif static void expire_upcalls(void *); #define EXPIRE_TIMEOUT (hz / 4) /* 4x / second */ #define UPCALL_EXPIRE 6 /* number of timeouts */ /* * XXX TODO: maintain a count to if_allmulti() calls in struct ifnet. */ /* * 'Interfaces' associated with decapsulator (so we can tell * packets that went through it from ones that get reflected * by a broken gateway). Different from IPv4 register_if, * these interfaces are linked into the system ifnet list, * because per-interface IPv6 statistics are maintained in * ifp->if_afdata. But it does not have any routes point * to them. I.e., packets can't be sent this way. They * only exist as a placeholder for multicast source * verification. */ static struct ifnet *multicast_register_if6; #define ENCAP_HOPS 64 /* * Private variables. */ static mifi_t nummifs = 0; static mifi_t reg_mif_num = (mifi_t)-1; static struct pim6stat pim6stat; SYSCTL_STRUCT(_net_inet6_pim, PIM6CTL_STATS, stats, CTLFLAG_RW, &pim6stat, pim6stat, "PIM Statistics (struct pim6stat, netinet6/pim6_var.h)"); #define PIM6STAT_INC(name) pim6stat.name += 1 static VNET_DEFINE(int, pim6); #define V_pim6 VNET(pim6) /* * Hash function for a source, group entry */ #define MF6CHASH(a, g) MF6CHASHMOD((a).s6_addr32[0] ^ (a).s6_addr32[1] ^ \ (a).s6_addr32[2] ^ (a).s6_addr32[3] ^ \ (g).s6_addr32[0] ^ (g).s6_addr32[1] ^ \ (g).s6_addr32[2] ^ (g).s6_addr32[3]) /* * Find a route for a given origin IPv6 address and Multicast group address. */ #define MF6CFIND(o, g, rt) do { \ struct mf6c *_rt = mf6ctable[MF6CHASH(o,g)]; \ rt = NULL; \ while (_rt) { \ if (IN6_ARE_ADDR_EQUAL(&_rt->mf6c_origin.sin6_addr, &(o)) && \ IN6_ARE_ADDR_EQUAL(&_rt->mf6c_mcastgrp.sin6_addr, &(g)) && \ (_rt->mf6c_stall == NULL)) { \ rt = _rt; \ break; \ } \ _rt = _rt->mf6c_next; \ } \ if (rt == NULL) { \ MRT6STAT_INC(mrt6s_mfc_misses); \ } \ } while (/*CONSTCOND*/ 0) /* * Macros to compute elapsed time efficiently * Borrowed from Van Jacobson's scheduling code * XXX: replace with timersub() ? */ #define TV_DELTA(a, b, delta) do { \ int xxs; \ \ delta = (a).tv_usec - (b).tv_usec; \ if ((xxs = (a).tv_sec - (b).tv_sec)) { \ switch (xxs) { \ case 2: \ delta += 1000000; \ /* FALLTHROUGH */ \ case 1: \ delta += 1000000; \ break; \ default: \ delta += (1000000 * xxs); \ } \ } \ } while (/*CONSTCOND*/ 0) /* XXX: replace with timercmp(a, b, <) ? */ #define TV_LT(a, b) (((a).tv_usec < (b).tv_usec && \ (a).tv_sec <= (b).tv_sec) || (a).tv_sec < (b).tv_sec) #ifdef UPCALL_TIMING #define UPCALL_MAX 50 static u_long upcall_data[UPCALL_MAX + 1]; static void collate(); #endif /* UPCALL_TIMING */ static int ip6_mrouter_init(struct socket *, int, int); static int add_m6fc(struct mf6cctl *); static int add_m6if(struct mif6ctl *); static int del_m6fc(struct mf6cctl *); static int del_m6if(mifi_t *); static int del_m6if_locked(mifi_t *); static int get_mif6_cnt(struct sioc_mif_req6 *); static int get_sg_cnt(struct sioc_sg_req6 *); static struct callout expire_upcalls_ch; int X_ip6_mforward(struct ip6_hdr *, struct ifnet *, struct mbuf *); int X_ip6_mrouter_done(void); int X_ip6_mrouter_set(struct socket *, struct sockopt *); int X_ip6_mrouter_get(struct socket *, struct sockopt *); int X_mrt6_ioctl(u_long, caddr_t); /* * Handle MRT setsockopt commands to modify the multicast routing tables. */ int X_ip6_mrouter_set(struct socket *so, struct sockopt *sopt) { int error = 0; int optval; struct mif6ctl mifc; struct mf6cctl mfcc; mifi_t mifi; if (so != V_ip6_mrouter && sopt->sopt_name != MRT6_INIT) return (EPERM); switch (sopt->sopt_name) { case MRT6_INIT: #ifdef MRT6_OINIT case MRT6_OINIT: #endif error = sooptcopyin(sopt, &optval, sizeof(optval), sizeof(optval)); if (error) break; error = ip6_mrouter_init(so, optval, sopt->sopt_name); break; case MRT6_DONE: error = X_ip6_mrouter_done(); break; case MRT6_ADD_MIF: error = sooptcopyin(sopt, &mifc, sizeof(mifc), sizeof(mifc)); if (error) break; error = add_m6if(&mifc); break; case MRT6_ADD_MFC: error = sooptcopyin(sopt, &mfcc, sizeof(mfcc), sizeof(mfcc)); if (error) break; error = add_m6fc(&mfcc); break; case MRT6_DEL_MFC: error = sooptcopyin(sopt, &mfcc, sizeof(mfcc), sizeof(mfcc)); if (error) break; error = del_m6fc(&mfcc); break; case MRT6_DEL_MIF: error = sooptcopyin(sopt, &mifi, sizeof(mifi), sizeof(mifi)); if (error) break; error = del_m6if(&mifi); break; case MRT6_PIM: error = sooptcopyin(sopt, &optval, sizeof(optval), sizeof(optval)); if (error) break; error = set_pim6(&optval); break; default: error = EOPNOTSUPP; break; } return (error); } /* * Handle MRT getsockopt commands */ int X_ip6_mrouter_get(struct socket *so, struct sockopt *sopt) { int error = 0; if (so != V_ip6_mrouter) return (EACCES); switch (sopt->sopt_name) { case MRT6_PIM: error = sooptcopyout(sopt, &V_pim6, sizeof(V_pim6)); break; } return (error); } /* * Handle ioctl commands to obtain information from the cache */ int X_mrt6_ioctl(u_long cmd, caddr_t data) { int ret; ret = EINVAL; switch (cmd) { case SIOCGETSGCNT_IN6: ret = get_sg_cnt((struct sioc_sg_req6 *)data); break; case SIOCGETMIFCNT_IN6: ret = get_mif6_cnt((struct sioc_mif_req6 *)data); break; default: break; } return (ret); } /* * returns the packet, byte, rpf-failure count for the source group provided */ static int get_sg_cnt(struct sioc_sg_req6 *req) { struct mf6c *rt; int ret; ret = 0; MFC6_LOCK(); MF6CFIND(req->src.sin6_addr, req->grp.sin6_addr, rt); if (rt == NULL) { ret = ESRCH; } else { req->pktcnt = rt->mf6c_pkt_cnt; req->bytecnt = rt->mf6c_byte_cnt; req->wrong_if = rt->mf6c_wrong_if; } MFC6_UNLOCK(); return (ret); } /* * returns the input and output packet and byte counts on the mif provided */ static int get_mif6_cnt(struct sioc_mif_req6 *req) { mifi_t mifi; int ret; ret = 0; mifi = req->mifi; MIF6_LOCK(); if (mifi >= nummifs) { ret = EINVAL; } else { req->icount = mif6table[mifi].m6_pkt_in; req->ocount = mif6table[mifi].m6_pkt_out; req->ibytes = mif6table[mifi].m6_bytes_in; req->obytes = mif6table[mifi].m6_bytes_out; } MIF6_UNLOCK(); return (ret); } static int set_pim6(int *i) { if ((*i != 1) && (*i != 0)) return (EINVAL); V_pim6 = *i; return (0); } /* * Enable multicast routing */ static int ip6_mrouter_init(struct socket *so, int v, int cmd) { MRT6_DLOG(DEBUG_ANY, "so_type = %d, pr_protocol = %d", so->so_type, so->so_proto->pr_protocol); if (so->so_type != SOCK_RAW || so->so_proto->pr_protocol != IPPROTO_ICMPV6) return (EOPNOTSUPP); if (v != 1) return (ENOPROTOOPT); MROUTER6_LOCK(); if (V_ip6_mrouter != NULL) { MROUTER6_UNLOCK(); return (EADDRINUSE); } V_ip6_mrouter = so; V_ip6_mrouter_ver = cmd; bzero((caddr_t)mf6ctable, sizeof(mf6ctable)); bzero((caddr_t)n6expire, sizeof(n6expire)); V_pim6 = 0;/* used for stubbing out/in pim stuff */ callout_init(&expire_upcalls_ch, 0); callout_reset(&expire_upcalls_ch, EXPIRE_TIMEOUT, expire_upcalls, NULL); MROUTER6_UNLOCK(); MRT6_DLOG(DEBUG_ANY, "finished"); return (0); } /* * Disable IPv6 multicast forwarding. */ int X_ip6_mrouter_done(void) { mifi_t mifi; u_long i; struct mf6c *rt; struct rtdetq *rte; MROUTER6_LOCK(); if (V_ip6_mrouter == NULL) { MROUTER6_UNLOCK(); return (EINVAL); } /* * For each phyint in use, disable promiscuous reception of all IPv6 * multicasts. */ for (mifi = 0; mifi < nummifs; mifi++) { if (mif6table[mifi].m6_ifp && !(mif6table[mifi].m6_flags & MIFF_REGISTER)) { if_allmulti(mif6table[mifi].m6_ifp, 0); } } bzero((caddr_t)mif6table, sizeof(mif6table)); nummifs = 0; V_pim6 = 0; /* used to stub out/in pim specific code */ callout_stop(&expire_upcalls_ch); /* * Free all multicast forwarding cache entries. */ MFC6_LOCK(); for (i = 0; i < MF6CTBLSIZ; i++) { rt = mf6ctable[i]; while (rt) { struct mf6c *frt; for (rte = rt->mf6c_stall; rte != NULL; ) { struct rtdetq *n = rte->next; m_freem(rte->m); free(rte, M_MRTABLE6); rte = n; } frt = rt; rt = rt->mf6c_next; free(frt, M_MRTABLE6); } } bzero((caddr_t)mf6ctable, sizeof(mf6ctable)); MFC6_UNLOCK(); /* * Reset register interface */ if (reg_mif_num != (mifi_t)-1 && multicast_register_if6 != NULL) { if_detach(multicast_register_if6); if_free(multicast_register_if6); reg_mif_num = (mifi_t)-1; multicast_register_if6 = NULL; } V_ip6_mrouter = NULL; V_ip6_mrouter_ver = 0; MROUTER6_UNLOCK(); MRT6_DLOG(DEBUG_ANY, "finished"); return (0); } static struct sockaddr_in6 sin6 = { sizeof(sin6), AF_INET6 }; /* * Add a mif to the mif table */ static int add_m6if(struct mif6ctl *mifcp) { struct mif6 *mifp; struct ifnet *ifp; int error; MIF6_LOCK(); if (mifcp->mif6c_mifi >= MAXMIFS) { MIF6_UNLOCK(); return (EINVAL); } mifp = mif6table + mifcp->mif6c_mifi; if (mifp->m6_ifp != NULL) { MIF6_UNLOCK(); return (EADDRINUSE); /* XXX: is it appropriate? */ } if (mifcp->mif6c_pifi == 0 || mifcp->mif6c_pifi > V_if_index) { MIF6_UNLOCK(); return (ENXIO); } ifp = ifnet_byindex(mifcp->mif6c_pifi); if (mifcp->mif6c_flags & MIFF_REGISTER) { if (reg_mif_num == (mifi_t)-1) { ifp = if_alloc(IFT_OTHER); if_initname(ifp, "register_mif", 0); ifp->if_flags |= IFF_LOOPBACK; if_attach(ifp); multicast_register_if6 = ifp; reg_mif_num = mifcp->mif6c_mifi; /* * it is impossible to guess the ifindex of the * register interface. So mif6c_pifi is automatically * calculated. */ mifcp->mif6c_pifi = ifp->if_index; } else { ifp = multicast_register_if6; } } else { /* Make sure the interface supports multicast */ if ((ifp->if_flags & IFF_MULTICAST) == 0) { MIF6_UNLOCK(); return (EOPNOTSUPP); } error = if_allmulti(ifp, 1); if (error) { MIF6_UNLOCK(); return (error); } } mifp->m6_flags = mifcp->mif6c_flags; mifp->m6_ifp = ifp; /* initialize per mif pkt counters */ mifp->m6_pkt_in = 0; mifp->m6_pkt_out = 0; mifp->m6_bytes_in = 0; mifp->m6_bytes_out = 0; /* Adjust nummifs up if the mifi is higher than nummifs */ if (nummifs <= mifcp->mif6c_mifi) nummifs = mifcp->mif6c_mifi + 1; MIF6_UNLOCK(); MRT6_DLOG(DEBUG_ANY, "mif #%d, phyint %s", mifcp->mif6c_mifi, if_name(ifp)); return (0); } /* * Delete a mif from the mif table */ static int del_m6if_locked(mifi_t *mifip) { struct mif6 *mifp = mif6table + *mifip; mifi_t mifi; struct ifnet *ifp; MIF6_LOCK_ASSERT(); if (*mifip >= nummifs) return (EINVAL); if (mifp->m6_ifp == NULL) return (EINVAL); if (!(mifp->m6_flags & MIFF_REGISTER)) { /* XXX: TODO: Maintain an ALLMULTI refcount in struct ifnet. */ ifp = mifp->m6_ifp; if_allmulti(ifp, 0); } else { if (reg_mif_num != (mifi_t)-1 && multicast_register_if6 != NULL) { if_detach(multicast_register_if6); if_free(multicast_register_if6); reg_mif_num = (mifi_t)-1; multicast_register_if6 = NULL; } } bzero((caddr_t)mifp, sizeof(*mifp)); /* Adjust nummifs down */ for (mifi = nummifs; mifi > 0; mifi--) if (mif6table[mifi - 1].m6_ifp) break; nummifs = mifi; MRT6_DLOG(DEBUG_ANY, "mif %d, nummifs %d", *mifip, nummifs); return (0); } static int del_m6if(mifi_t *mifip) { int cc; MIF6_LOCK(); cc = del_m6if_locked(mifip); MIF6_UNLOCK(); return (cc); } /* * Add an mfc entry */ static int add_m6fc(struct mf6cctl *mfccp) { struct mf6c *rt; u_long hash; struct rtdetq *rte; u_short nstl; char ip6bufo[INET6_ADDRSTRLEN], ip6bufg[INET6_ADDRSTRLEN]; MFC6_LOCK(); MF6CFIND(mfccp->mf6cc_origin.sin6_addr, mfccp->mf6cc_mcastgrp.sin6_addr, rt); /* If an entry already exists, just update the fields */ if (rt) { MRT6_DLOG(DEBUG_MFC, "no upcall o %s g %s p %x", ip6_sprintf(ip6bufo, &mfccp->mf6cc_origin.sin6_addr), ip6_sprintf(ip6bufg, &mfccp->mf6cc_mcastgrp.sin6_addr), mfccp->mf6cc_parent); rt->mf6c_parent = mfccp->mf6cc_parent; rt->mf6c_ifset = mfccp->mf6cc_ifset; MFC6_UNLOCK(); return (0); } /* * Find the entry for which the upcall was made and update */ hash = MF6CHASH(mfccp->mf6cc_origin.sin6_addr, mfccp->mf6cc_mcastgrp.sin6_addr); for (rt = mf6ctable[hash], nstl = 0; rt; rt = rt->mf6c_next) { if (IN6_ARE_ADDR_EQUAL(&rt->mf6c_origin.sin6_addr, &mfccp->mf6cc_origin.sin6_addr) && IN6_ARE_ADDR_EQUAL(&rt->mf6c_mcastgrp.sin6_addr, &mfccp->mf6cc_mcastgrp.sin6_addr) && (rt->mf6c_stall != NULL)) { if (nstl++) log(LOG_ERR, "add_m6fc: %s o %s g %s p %x dbx %p\n", "multiple kernel entries", ip6_sprintf(ip6bufo, &mfccp->mf6cc_origin.sin6_addr), ip6_sprintf(ip6bufg, &mfccp->mf6cc_mcastgrp.sin6_addr), mfccp->mf6cc_parent, rt->mf6c_stall); MRT6_DLOG(DEBUG_MFC, "o %s g %s p %x dbg %p", ip6_sprintf(ip6bufo, &mfccp->mf6cc_origin.sin6_addr), ip6_sprintf(ip6bufg, &mfccp->mf6cc_mcastgrp.sin6_addr), mfccp->mf6cc_parent, rt->mf6c_stall); rt->mf6c_origin = mfccp->mf6cc_origin; rt->mf6c_mcastgrp = mfccp->mf6cc_mcastgrp; rt->mf6c_parent = mfccp->mf6cc_parent; rt->mf6c_ifset = mfccp->mf6cc_ifset; /* initialize pkt counters per src-grp */ rt->mf6c_pkt_cnt = 0; rt->mf6c_byte_cnt = 0; rt->mf6c_wrong_if = 0; rt->mf6c_expire = 0; /* Don't clean this guy up */ n6expire[hash]--; /* free packets Qed at the end of this entry */ for (rte = rt->mf6c_stall; rte != NULL; ) { struct rtdetq *n = rte->next; ip6_mdq(rte->m, rte->ifp, rt); m_freem(rte->m); #ifdef UPCALL_TIMING collate(&(rte->t)); #endif /* UPCALL_TIMING */ free(rte, M_MRTABLE6); rte = n; } rt->mf6c_stall = NULL; } } /* * It is possible that an entry is being inserted without an upcall */ if (nstl == 0) { MRT6_DLOG(DEBUG_MFC, "no upcall h %lu o %s g %s p %x", hash, ip6_sprintf(ip6bufo, &mfccp->mf6cc_origin.sin6_addr), ip6_sprintf(ip6bufg, &mfccp->mf6cc_mcastgrp.sin6_addr), mfccp->mf6cc_parent); for (rt = mf6ctable[hash]; rt; rt = rt->mf6c_next) { if (IN6_ARE_ADDR_EQUAL(&rt->mf6c_origin.sin6_addr, &mfccp->mf6cc_origin.sin6_addr)&& IN6_ARE_ADDR_EQUAL(&rt->mf6c_mcastgrp.sin6_addr, &mfccp->mf6cc_mcastgrp.sin6_addr)) { rt->mf6c_origin = mfccp->mf6cc_origin; rt->mf6c_mcastgrp = mfccp->mf6cc_mcastgrp; rt->mf6c_parent = mfccp->mf6cc_parent; rt->mf6c_ifset = mfccp->mf6cc_ifset; /* initialize pkt counters per src-grp */ rt->mf6c_pkt_cnt = 0; rt->mf6c_byte_cnt = 0; rt->mf6c_wrong_if = 0; if (rt->mf6c_expire) n6expire[hash]--; rt->mf6c_expire = 0; } } if (rt == NULL) { /* no upcall, so make a new entry */ rt = (struct mf6c *)malloc(sizeof(*rt), M_MRTABLE6, M_NOWAIT); if (rt == NULL) { MFC6_UNLOCK(); return (ENOBUFS); } /* insert new entry at head of hash chain */ rt->mf6c_origin = mfccp->mf6cc_origin; rt->mf6c_mcastgrp = mfccp->mf6cc_mcastgrp; rt->mf6c_parent = mfccp->mf6cc_parent; rt->mf6c_ifset = mfccp->mf6cc_ifset; /* initialize pkt counters per src-grp */ rt->mf6c_pkt_cnt = 0; rt->mf6c_byte_cnt = 0; rt->mf6c_wrong_if = 0; rt->mf6c_expire = 0; rt->mf6c_stall = NULL; /* link into table */ rt->mf6c_next = mf6ctable[hash]; mf6ctable[hash] = rt; } } MFC6_UNLOCK(); return (0); } #ifdef UPCALL_TIMING /* * collect delay statistics on the upcalls */ static void collate(struct timeval *t) { u_long d; struct timeval tp; u_long delta; GET_TIME(tp); if (TV_LT(*t, tp)) { TV_DELTA(tp, *t, delta); d = delta >> 10; if (d > UPCALL_MAX) d = UPCALL_MAX; ++upcall_data[d]; } } #endif /* UPCALL_TIMING */ /* * Delete an mfc entry */ static int del_m6fc(struct mf6cctl *mfccp) { #ifdef MRT6DEBUG char ip6bufo[INET6_ADDRSTRLEN], ip6bufg[INET6_ADDRSTRLEN]; #endif struct sockaddr_in6 origin; struct sockaddr_in6 mcastgrp; struct mf6c *rt; struct mf6c **nptr; u_long hash; origin = mfccp->mf6cc_origin; mcastgrp = mfccp->mf6cc_mcastgrp; hash = MF6CHASH(origin.sin6_addr, mcastgrp.sin6_addr); MRT6_DLOG(DEBUG_MFC, "orig %s mcastgrp %s", ip6_sprintf(ip6bufo, &origin.sin6_addr), ip6_sprintf(ip6bufg, &mcastgrp.sin6_addr)); MFC6_LOCK(); nptr = &mf6ctable[hash]; while ((rt = *nptr) != NULL) { if (IN6_ARE_ADDR_EQUAL(&origin.sin6_addr, &rt->mf6c_origin.sin6_addr) && IN6_ARE_ADDR_EQUAL(&mcastgrp.sin6_addr, &rt->mf6c_mcastgrp.sin6_addr) && rt->mf6c_stall == NULL) break; nptr = &rt->mf6c_next; } if (rt == NULL) { MFC6_UNLOCK(); return (EADDRNOTAVAIL); } *nptr = rt->mf6c_next; free(rt, M_MRTABLE6); MFC6_UNLOCK(); return (0); } static int socket_send(struct socket *s, struct mbuf *mm, struct sockaddr_in6 *src) { if (s) { if (sbappendaddr(&s->so_rcv, (struct sockaddr *)src, mm, (struct mbuf *)0) != 0) { sorwakeup(s); return (0); } } m_freem(mm); return (-1); } /* * IPv6 multicast forwarding function. This function assumes that the packet * pointed to by "ip6" has arrived on (or is about to be sent to) the interface * pointed to by "ifp", and the packet is to be relayed to other networks * that have members of the packet's destination IPv6 multicast group. * * The packet is returned unscathed to the caller, unless it is * erroneous, in which case a non-zero return value tells the caller to * discard it. * * NOTE: this implementation assumes that m->m_pkthdr.rcvif is NULL iff * this function is called in the originating context (i.e., not when * forwarding a packet from other node). ip6_output(), which is currently the * only function that calls this function is called in the originating context, * explicitly ensures this condition. It is caller's responsibility to ensure * that if this function is called from somewhere else in the originating * context in the future. */ int X_ip6_mforward(struct ip6_hdr *ip6, struct ifnet *ifp, struct mbuf *m) { struct rtdetq *rte; struct mbuf *mb0; struct mf6c *rt; struct mif6 *mifp; struct mbuf *mm; u_long hash; mifi_t mifi; char ip6bufs[INET6_ADDRSTRLEN], ip6bufd[INET6_ADDRSTRLEN]; #ifdef UPCALL_TIMING struct timeval tp; GET_TIME(tp); #endif /* UPCALL_TIMING */ MRT6_DLOG(DEBUG_FORWARD, "src %s, dst %s, ifindex %d", ip6_sprintf(ip6bufs, &ip6->ip6_src), ip6_sprintf(ip6bufd, &ip6->ip6_dst), ifp->if_index); /* * Don't forward a packet with Hop limit of zero or one, * or a packet destined to a local-only group. */ if (ip6->ip6_hlim <= 1 || IN6_IS_ADDR_MC_INTFACELOCAL(&ip6->ip6_dst) || IN6_IS_ADDR_MC_LINKLOCAL(&ip6->ip6_dst)) return (0); ip6->ip6_hlim--; /* * Source address check: do not forward packets with unspecified * source. It was discussed in July 2000, on ipngwg mailing list. * This is rather more serious than unicast cases, because some * MLD packets can be sent with the unspecified source address * (although such packets must normally set 1 to the hop limit field). */ if (IN6_IS_ADDR_UNSPECIFIED(&ip6->ip6_src)) { IP6STAT_INC(ip6s_cantforward); if (V_ip6_log_time + V_ip6_log_interval < time_uptime) { V_ip6_log_time = time_uptime; log(LOG_DEBUG, "cannot forward " "from %s to %s nxt %d received on %s\n", ip6_sprintf(ip6bufs, &ip6->ip6_src), ip6_sprintf(ip6bufd, &ip6->ip6_dst), ip6->ip6_nxt, if_name(m->m_pkthdr.rcvif)); } return (0); } MFC6_LOCK(); /* * Determine forwarding mifs from the forwarding cache table */ MF6CFIND(ip6->ip6_src, ip6->ip6_dst, rt); MRT6STAT_INC(mrt6s_mfc_lookups); /* Entry exists, so forward if necessary */ if (rt) { MFC6_UNLOCK(); return (ip6_mdq(m, ifp, rt)); } /* * If we don't have a route for packet's origin, * Make a copy of the packet & send message to routing daemon. */ MRT6STAT_INC(mrt6s_no_route); MRT6_DLOG(DEBUG_FORWARD | DEBUG_MFC, "no rte s %s g %s", ip6_sprintf(ip6bufs, &ip6->ip6_src), ip6_sprintf(ip6bufd, &ip6->ip6_dst)); /* * Allocate mbufs early so that we don't do extra work if we * are just going to fail anyway. */ rte = (struct rtdetq *)malloc(sizeof(*rte), M_MRTABLE6, M_NOWAIT); if (rte == NULL) { MFC6_UNLOCK(); return (ENOBUFS); } mb0 = m_copy(m, 0, M_COPYALL); /* * Pullup packet header if needed before storing it, * as other references may modify it in the meantime. */ if (mb0 && (!M_WRITABLE(mb0) || mb0->m_len < sizeof(struct ip6_hdr))) mb0 = m_pullup(mb0, sizeof(struct ip6_hdr)); if (mb0 == NULL) { free(rte, M_MRTABLE6); MFC6_UNLOCK(); return (ENOBUFS); } /* is there an upcall waiting for this packet? */ hash = MF6CHASH(ip6->ip6_src, ip6->ip6_dst); for (rt = mf6ctable[hash]; rt; rt = rt->mf6c_next) { if (IN6_ARE_ADDR_EQUAL(&ip6->ip6_src, &rt->mf6c_origin.sin6_addr) && IN6_ARE_ADDR_EQUAL(&ip6->ip6_dst, &rt->mf6c_mcastgrp.sin6_addr) && (rt->mf6c_stall != NULL)) break; } if (rt == NULL) { struct mrt6msg *im; #ifdef MRT6_OINIT struct omrt6msg *oim; #endif /* no upcall, so make a new entry */ rt = (struct mf6c *)malloc(sizeof(*rt), M_MRTABLE6, M_NOWAIT); if (rt == NULL) { free(rte, M_MRTABLE6); m_freem(mb0); MFC6_UNLOCK(); return (ENOBUFS); } /* * Make a copy of the header to send to the user * level process */ mm = m_copy(mb0, 0, sizeof(struct ip6_hdr)); if (mm == NULL) { free(rte, M_MRTABLE6); m_freem(mb0); free(rt, M_MRTABLE6); MFC6_UNLOCK(); return (ENOBUFS); } /* * Send message to routing daemon */ sin6.sin6_addr = ip6->ip6_src; im = NULL; #ifdef MRT6_OINIT oim = NULL; #endif switch (V_ip6_mrouter_ver) { #ifdef MRT6_OINIT case MRT6_OINIT: oim = mtod(mm, struct omrt6msg *); oim->im6_msgtype = MRT6MSG_NOCACHE; oim->im6_mbz = 0; break; #endif case MRT6_INIT: im = mtod(mm, struct mrt6msg *); im->im6_msgtype = MRT6MSG_NOCACHE; im->im6_mbz = 0; break; default: free(rte, M_MRTABLE6); m_freem(mb0); free(rt, M_MRTABLE6); MFC6_UNLOCK(); return (EINVAL); } MRT6_DLOG(DEBUG_FORWARD, "getting the iif info in the kernel"); for (mifp = mif6table, mifi = 0; mifi < nummifs && mifp->m6_ifp != ifp; mifp++, mifi++) ; switch (V_ip6_mrouter_ver) { #ifdef MRT6_OINIT case MRT6_OINIT: oim->im6_mif = mifi; break; #endif case MRT6_INIT: im->im6_mif = mifi; break; } if (socket_send(V_ip6_mrouter, mm, &sin6) < 0) { log(LOG_WARNING, "ip6_mforward: ip6_mrouter " "socket queue full\n"); MRT6STAT_INC(mrt6s_upq_sockfull); free(rte, M_MRTABLE6); m_freem(mb0); free(rt, M_MRTABLE6); MFC6_UNLOCK(); return (ENOBUFS); } MRT6STAT_INC(mrt6s_upcalls); /* insert new entry at head of hash chain */ bzero(rt, sizeof(*rt)); rt->mf6c_origin.sin6_family = AF_INET6; rt->mf6c_origin.sin6_len = sizeof(struct sockaddr_in6); rt->mf6c_origin.sin6_addr = ip6->ip6_src; rt->mf6c_mcastgrp.sin6_family = AF_INET6; rt->mf6c_mcastgrp.sin6_len = sizeof(struct sockaddr_in6); rt->mf6c_mcastgrp.sin6_addr = ip6->ip6_dst; rt->mf6c_expire = UPCALL_EXPIRE; n6expire[hash]++; rt->mf6c_parent = MF6C_INCOMPLETE_PARENT; /* link into table */ rt->mf6c_next = mf6ctable[hash]; mf6ctable[hash] = rt; /* Add this entry to the end of the queue */ rt->mf6c_stall = rte; } else { /* determine if q has overflowed */ struct rtdetq **p; int npkts = 0; for (p = &rt->mf6c_stall; *p != NULL; p = &(*p)->next) if (++npkts > MAX_UPQ6) { MRT6STAT_INC(mrt6s_upq_ovflw); free(rte, M_MRTABLE6); m_freem(mb0); MFC6_UNLOCK(); return (0); } /* Add this entry to the end of the queue */ *p = rte; } rte->next = NULL; rte->m = mb0; rte->ifp = ifp; #ifdef UPCALL_TIMING rte->t = tp; #endif /* UPCALL_TIMING */ MFC6_UNLOCK(); return (0); } /* * Clean up cache entries if upcalls are not serviced * Call from the Slow Timeout mechanism, every half second. */ static void expire_upcalls(void *unused) { #ifdef MRT6DEBUG char ip6bufo[INET6_ADDRSTRLEN], ip6bufg[INET6_ADDRSTRLEN]; #endif struct rtdetq *rte; struct mf6c *mfc, **nptr; u_long i; MFC6_LOCK(); for (i = 0; i < MF6CTBLSIZ; i++) { if (n6expire[i] == 0) continue; nptr = &mf6ctable[i]; while ((mfc = *nptr) != NULL) { rte = mfc->mf6c_stall; /* * Skip real cache entries * Make sure it wasn't marked to not expire (shouldn't happen) * If it expires now */ if (rte != NULL && mfc->mf6c_expire != 0 && --mfc->mf6c_expire == 0) { MRT6_DLOG(DEBUG_EXPIRE, "expiring (%s %s)", ip6_sprintf(ip6bufo, &mfc->mf6c_origin.sin6_addr), ip6_sprintf(ip6bufg, &mfc->mf6c_mcastgrp.sin6_addr)); /* * drop all the packets * free the mbuf with the pkt, if, timing info */ do { struct rtdetq *n = rte->next; m_freem(rte->m); free(rte, M_MRTABLE6); rte = n; } while (rte != NULL); MRT6STAT_INC(mrt6s_cache_cleanups); n6expire[i]--; *nptr = mfc->mf6c_next; free(mfc, M_MRTABLE6); } else { nptr = &mfc->mf6c_next; } } } MFC6_UNLOCK(); callout_reset(&expire_upcalls_ch, EXPIRE_TIMEOUT, expire_upcalls, NULL); } /* * Packet forwarding routine once entry in the cache is made */ static int ip6_mdq(struct mbuf *m, struct ifnet *ifp, struct mf6c *rt) { struct ip6_hdr *ip6 = mtod(m, struct ip6_hdr *); mifi_t mifi, iif; struct mif6 *mifp; int plen = m->m_pkthdr.len; struct in6_addr src0, dst0; /* copies for local work */ u_int32_t iszone, idzone, oszone, odzone; int error = 0; /* * Macro to send packet on mif. Since RSVP packets don't get counted on * input, they shouldn't get counted on output, so statistics keeping is * separate. */ #define MC6_SEND(ip6, mifp, m) do { \ if ((mifp)->m6_flags & MIFF_REGISTER) \ register_send((ip6), (mifp), (m)); \ else \ phyint_send((ip6), (mifp), (m)); \ } while (/*CONSTCOND*/ 0) /* * Don't forward if it didn't arrive from the parent mif * for its origin. */ mifi = rt->mf6c_parent; if ((mifi >= nummifs) || (mif6table[mifi].m6_ifp != ifp)) { /* came in the wrong interface */ MRT6_DLOG(DEBUG_FORWARD, "wrong if: ifid %d mifi %d mififid %x", ifp->if_index, mifi, mif6table[mifi].m6_ifp->if_index); MRT6STAT_INC(mrt6s_wrong_if); rt->mf6c_wrong_if++; /* * If we are doing PIM processing, and we are forwarding * packets on this interface, send a message to the * routing daemon. */ /* have to make sure this is a valid mif */ if (mifi < nummifs && mif6table[mifi].m6_ifp) if (V_pim6 && (m->m_flags & M_LOOP) == 0) { /* * Check the M_LOOP flag to avoid an * unnecessary PIM assert. * XXX: M_LOOP is an ad-hoc hack... */ static struct sockaddr_in6 sin6 = { sizeof(sin6), AF_INET6 }; struct mbuf *mm; struct mrt6msg *im; #ifdef MRT6_OINIT struct omrt6msg *oim; #endif mm = m_copy(m, 0, sizeof(struct ip6_hdr)); if (mm && (!M_WRITABLE(mm) || mm->m_len < sizeof(struct ip6_hdr))) mm = m_pullup(mm, sizeof(struct ip6_hdr)); if (mm == NULL) return (ENOBUFS); #ifdef MRT6_OINIT oim = NULL; #endif im = NULL; switch (V_ip6_mrouter_ver) { #ifdef MRT6_OINIT case MRT6_OINIT: oim = mtod(mm, struct omrt6msg *); oim->im6_msgtype = MRT6MSG_WRONGMIF; oim->im6_mbz = 0; break; #endif case MRT6_INIT: im = mtod(mm, struct mrt6msg *); im->im6_msgtype = MRT6MSG_WRONGMIF; im->im6_mbz = 0; break; default: m_freem(mm); return (EINVAL); } for (mifp = mif6table, iif = 0; iif < nummifs && mifp && mifp->m6_ifp != ifp; mifp++, iif++) ; switch (V_ip6_mrouter_ver) { #ifdef MRT6_OINIT case MRT6_OINIT: oim->im6_mif = iif; sin6.sin6_addr = oim->im6_src; break; #endif case MRT6_INIT: im->im6_mif = iif; sin6.sin6_addr = im->im6_src; break; } MRT6STAT_INC(mrt6s_upcalls); if (socket_send(V_ip6_mrouter, mm, &sin6) < 0) { MRT6_DLOG(DEBUG_ANY, "ip6_mrouter socket queue full"); MRT6STAT_INC(mrt6s_upq_sockfull); return (ENOBUFS); } /* if socket Q full */ } /* if PIM */ return (0); } /* if wrong iif */ /* If I sourced this packet, it counts as output, else it was input. */ if (m->m_pkthdr.rcvif == NULL) { /* XXX: is rcvif really NULL when output?? */ mif6table[mifi].m6_pkt_out++; mif6table[mifi].m6_bytes_out += plen; } else { mif6table[mifi].m6_pkt_in++; mif6table[mifi].m6_bytes_in += plen; } rt->mf6c_pkt_cnt++; rt->mf6c_byte_cnt += plen; /* * For each mif, forward a copy of the packet if there are group * members downstream on the interface. */ src0 = ip6->ip6_src; dst0 = ip6->ip6_dst; if ((error = in6_setscope(&src0, ifp, &iszone)) != 0 || (error = in6_setscope(&dst0, ifp, &idzone)) != 0) { IP6STAT_INC(ip6s_badscope); return (error); } for (mifp = mif6table, mifi = 0; mifi < nummifs; mifp++, mifi++) { if (IF_ISSET(mifi, &rt->mf6c_ifset)) { /* * check if the outgoing packet is going to break * a scope boundary. * XXX For packets through PIM register tunnel * interface, we believe a routing daemon. */ if (!(mif6table[rt->mf6c_parent].m6_flags & MIFF_REGISTER) && !(mif6table[mifi].m6_flags & MIFF_REGISTER)) { if (in6_setscope(&src0, mif6table[mifi].m6_ifp, &oszone) || in6_setscope(&dst0, mif6table[mifi].m6_ifp, &odzone) || iszone != oszone || idzone != odzone) { IP6STAT_INC(ip6s_badscope); continue; } } mifp->m6_pkt_out++; mifp->m6_bytes_out += plen; MC6_SEND(ip6, mifp, m); } } return (0); } static void phyint_send(struct ip6_hdr *ip6, struct mif6 *mifp, struct mbuf *m) { #ifdef MRT6DEBUG char ip6bufs[INET6_ADDRSTRLEN], ip6bufd[INET6_ADDRSTRLEN]; #endif struct mbuf *mb_copy; struct ifnet *ifp = mifp->m6_ifp; int error = 0; u_long linkmtu; /* * Make a new reference to the packet; make sure that * the IPv6 header is actually copied, not just referenced, * so that ip6_output() only scribbles on the copy. */ mb_copy = m_copy(m, 0, M_COPYALL); if (mb_copy && (!M_WRITABLE(mb_copy) || mb_copy->m_len < sizeof(struct ip6_hdr))) mb_copy = m_pullup(mb_copy, sizeof(struct ip6_hdr)); if (mb_copy == NULL) { return; } /* set MCAST flag to the outgoing packet */ mb_copy->m_flags |= M_MCAST; /* * If we sourced the packet, call ip6_output since we may devide * the packet into fragments when the packet is too big for the * outgoing interface. * Otherwise, we can simply send the packet to the interface * sending queue. */ if (m->m_pkthdr.rcvif == NULL) { struct ip6_moptions im6o; im6o.im6o_multicast_ifp = ifp; /* XXX: ip6_output will override ip6->ip6_hlim */ im6o.im6o_multicast_hlim = ip6->ip6_hlim; im6o.im6o_multicast_loop = 1; error = ip6_output(mb_copy, NULL, NULL, IPV6_FORWARDING, &im6o, NULL, NULL); MRT6_DLOG(DEBUG_XMIT, "mif %u err %d", (uint16_t)(mifp - mif6table), error); return; } /* * If configured to loop back multicasts by default, * loop back a copy now. */ if (in6_mcast_loop) { struct sockaddr_in6 dst6; bzero(&dst6, sizeof(dst6)); dst6.sin6_len = sizeof(struct sockaddr_in6); dst6.sin6_family = AF_INET6; dst6.sin6_addr = ip6->ip6_dst; ip6_mloopback(ifp, m, &dst6); } /* * Put the packet into the sending queue of the outgoing interface * if it would fit in the MTU of the interface. */ linkmtu = IN6_LINKMTU(ifp); if (mb_copy->m_pkthdr.len <= linkmtu || linkmtu < IPV6_MMTU) { struct sockaddr_in6 dst6; bzero(&dst6, sizeof(dst6)); dst6.sin6_len = sizeof(struct sockaddr_in6); dst6.sin6_family = AF_INET6; dst6.sin6_addr = ip6->ip6_dst; IP_PROBE(send, NULL, NULL, ip6, ifp, NULL, ip6); /* * We just call if_output instead of nd6_output here, since * we need no ND for a multicast forwarded packet...right? */ m_clrprotoflags(m); /* Avoid confusing lower layers. */ error = (*ifp->if_output)(ifp, mb_copy, (struct sockaddr *)&dst6, NULL); MRT6_DLOG(DEBUG_XMIT, "mif %u err %d", (uint16_t)(mifp - mif6table), error); } else { /* * pMTU discovery is intentionally disabled by default, since * various router may notify pMTU in multicast, which can be * a DDoS to a router */ if (V_ip6_mcast_pmtu) icmp6_error(mb_copy, ICMP6_PACKET_TOO_BIG, 0, linkmtu); else { MRT6_DLOG(DEBUG_XMIT, " packet too big on %s o %s " "g %s size %d (discarded)", if_name(ifp), ip6_sprintf(ip6bufs, &ip6->ip6_src), ip6_sprintf(ip6bufd, &ip6->ip6_dst), mb_copy->m_pkthdr.len); m_freem(mb_copy); /* simply discard the packet */ } } } static int register_send(struct ip6_hdr *ip6, struct mif6 *mif, struct mbuf *m) { #ifdef MRT6DEBUG char ip6bufs[INET6_ADDRSTRLEN], ip6bufd[INET6_ADDRSTRLEN]; #endif struct mbuf *mm; int i, len = m->m_pkthdr.len; static struct sockaddr_in6 sin6 = { sizeof(sin6), AF_INET6 }; struct mrt6msg *im6; MRT6_DLOG(DEBUG_ANY, "src %s dst %s", ip6_sprintf(ip6bufs, &ip6->ip6_src), ip6_sprintf(ip6bufd, &ip6->ip6_dst)); PIM6STAT_INC(pim6s_snd_registers); /* Make a copy of the packet to send to the user level process. */ mm = m_gethdr(M_NOWAIT, MT_DATA); if (mm == NULL) return (ENOBUFS); mm->m_data += max_linkhdr; mm->m_len = sizeof(struct ip6_hdr); if ((mm->m_next = m_copy(m, 0, M_COPYALL)) == NULL) { m_freem(mm); return (ENOBUFS); } i = MHLEN - M_LEADINGSPACE(mm); if (i > len) i = len; mm = m_pullup(mm, i); if (mm == NULL) return (ENOBUFS); /* TODO: check it! */ mm->m_pkthdr.len = len + sizeof(struct ip6_hdr); /* * Send message to routing daemon */ sin6.sin6_addr = ip6->ip6_src; im6 = mtod(mm, struct mrt6msg *); im6->im6_msgtype = MRT6MSG_WHOLEPKT; im6->im6_mbz = 0; im6->im6_mif = mif - mif6table; /* iif info is not given for reg. encap.n */ MRT6STAT_INC(mrt6s_upcalls); if (socket_send(V_ip6_mrouter, mm, &sin6) < 0) { MRT6_DLOG(DEBUG_ANY, "ip6_mrouter socket queue full"); MRT6STAT_INC(mrt6s_upq_sockfull); return (ENOBUFS); } return (0); } /* * pim6_encapcheck() is called by the encap6_input() path at runtime to * determine if a packet is for PIM; allowing PIM to be dynamically loaded * into the kernel. */ static int pim6_encapcheck(const struct mbuf *m, int off, int proto, void *arg) { #ifdef DIAGNOSTIC KASSERT(proto == IPPROTO_PIM, ("not for IPPROTO_PIM")); #endif if (proto != IPPROTO_PIM) return 0; /* not for us; reject the datagram. */ return 64; /* claim the datagram. */ } /* * PIM sparse mode hook * Receives the pim control messages, and passes them up to the listening * socket, using rip6_input. * The only message processed is the REGISTER pim message; the pim header * is stripped off, and the inner packet is passed to register_mforward. */ int pim6_input(struct mbuf **mp, int *offp, int proto) { struct pim *pim; /* pointer to a pim struct */ struct ip6_hdr *ip6; int pimlen; struct mbuf *m = *mp; int minlen; int off = *offp; PIM6STAT_INC(pim6s_rcv_total); ip6 = mtod(m, struct ip6_hdr *); pimlen = m->m_pkthdr.len - *offp; /* * Validate lengths */ if (pimlen < PIM_MINLEN) { PIM6STAT_INC(pim6s_rcv_tooshort); MRT6_DLOG(DEBUG_PIM, "PIM packet too short"); m_freem(m); return (IPPROTO_DONE); } /* * if the packet is at least as big as a REGISTER, go ahead * and grab the PIM REGISTER header size, to avoid another * possible m_pullup() later. * * PIM_MINLEN == pimhdr + u_int32 == 8 * PIM6_REG_MINLEN == pimhdr + reghdr + eip6hdr == 4 + 4 + 40 */ minlen = (pimlen >= PIM6_REG_MINLEN) ? PIM6_REG_MINLEN : PIM_MINLEN; /* * Make sure that the IP6 and PIM headers in contiguous memory, and * possibly the PIM REGISTER header */ #ifndef PULLDOWN_TEST IP6_EXTHDR_CHECK(m, off, minlen, IPPROTO_DONE); /* adjust pointer */ ip6 = mtod(m, struct ip6_hdr *); /* adjust mbuf to point to the PIM header */ pim = (struct pim *)((caddr_t)ip6 + off); #else IP6_EXTHDR_GET(pim, struct pim *, m, off, minlen); if (pim == NULL) { PIM6STAT_INC(pim6s_rcv_tooshort); return (IPPROTO_DONE); } #endif #define PIM6_CHECKSUM #ifdef PIM6_CHECKSUM { int cksumlen; /* * Validate checksum. * If PIM REGISTER, exclude the data packet */ if (pim->pim_type == PIM_REGISTER) cksumlen = PIM_MINLEN; else cksumlen = pimlen; if (in6_cksum(m, IPPROTO_PIM, off, cksumlen)) { PIM6STAT_INC(pim6s_rcv_badsum); MRT6_DLOG(DEBUG_PIM, "invalid checksum"); m_freem(m); return (IPPROTO_DONE); } } #endif /* PIM_CHECKSUM */ /* PIM version check */ if (pim->pim_ver != PIM_VERSION) { PIM6STAT_INC(pim6s_rcv_badversion); MRT6_DLOG(DEBUG_ANY | DEBUG_ERR, "incorrect version %d, expecting %d", pim->pim_ver, PIM_VERSION); m_freem(m); return (IPPROTO_DONE); } if (pim->pim_type == PIM_REGISTER) { /* * since this is a REGISTER, we'll make a copy of the register * headers ip6+pim+u_int32_t+encap_ip6, to be passed up to the * routing daemon. */ static struct sockaddr_in6 dst = { sizeof(dst), AF_INET6 }; struct mbuf *mcp; struct ip6_hdr *eip6; u_int32_t *reghdr; int rc; #ifdef MRT6DEBUG char ip6bufs[INET6_ADDRSTRLEN], ip6bufd[INET6_ADDRSTRLEN]; #endif PIM6STAT_INC(pim6s_rcv_registers); if ((reg_mif_num >= nummifs) || (reg_mif_num == (mifi_t) -1)) { MRT6_DLOG(DEBUG_PIM, "register mif not set: %d", reg_mif_num); m_freem(m); return (IPPROTO_DONE); } reghdr = (u_int32_t *)(pim + 1); if ((ntohl(*reghdr) & PIM_NULL_REGISTER)) goto pim6_input_to_daemon; /* * Validate length */ if (pimlen < PIM6_REG_MINLEN) { PIM6STAT_INC(pim6s_rcv_tooshort); PIM6STAT_INC(pim6s_rcv_badregisters); MRT6_DLOG(DEBUG_ANY | DEBUG_ERR, "register packet " "size too small %d from %s", pimlen, ip6_sprintf(ip6bufs, &ip6->ip6_src)); m_freem(m); return (IPPROTO_DONE); } eip6 = (struct ip6_hdr *) (reghdr + 1); MRT6_DLOG(DEBUG_PIM, "eip6: %s -> %s, eip6 plen %d", ip6_sprintf(ip6bufs, &eip6->ip6_src), ip6_sprintf(ip6bufd, &eip6->ip6_dst), ntohs(eip6->ip6_plen)); /* verify the version number of the inner packet */ if ((eip6->ip6_vfc & IPV6_VERSION_MASK) != IPV6_VERSION) { PIM6STAT_INC(pim6s_rcv_badregisters); MRT6_DLOG(DEBUG_ANY, "invalid IP version (%d) " "of the inner packet", (eip6->ip6_vfc & IPV6_VERSION)); m_freem(m); return (IPPROTO_NONE); } /* verify the inner packet is destined to a mcast group */ if (!IN6_IS_ADDR_MULTICAST(&eip6->ip6_dst)) { PIM6STAT_INC(pim6s_rcv_badregisters); MRT6_DLOG(DEBUG_PIM, "inner packet of register " "is not multicast %s", ip6_sprintf(ip6bufd, &eip6->ip6_dst)); m_freem(m); return (IPPROTO_DONE); } /* * make a copy of the whole header to pass to the daemon later. */ mcp = m_copy(m, 0, off + PIM6_REG_MINLEN); if (mcp == NULL) { MRT6_DLOG(DEBUG_ANY | DEBUG_ERR, "pim register: " "could not copy register head"); m_freem(m); return (IPPROTO_DONE); } /* * forward the inner ip6 packet; point m_data at the inner ip6. */ m_adj(m, off + PIM_MINLEN); MRT6_DLOG(DEBUG_PIM, "forwarding decapsulated register: " "src %s, dst %s, mif %d", ip6_sprintf(ip6bufs, &eip6->ip6_src), ip6_sprintf(ip6bufd, &eip6->ip6_dst), reg_mif_num); rc = if_simloop(mif6table[reg_mif_num].m6_ifp, m, dst.sin6_family, 0); /* prepare the register head to send to the mrouting daemon */ m = mcp; } /* * Pass the PIM message up to the daemon; if it is a register message * pass the 'head' only up to the daemon. This includes the * encapsulator ip6 header, pim header, register header and the * encapsulated ip6 header. */ pim6_input_to_daemon: rip6_input(&m, offp, proto); return (IPPROTO_DONE); } static int ip6_mroute_modevent(module_t mod, int type, void *unused) { switch (type) { case MOD_LOAD: MROUTER6_LOCK_INIT(); MFC6_LOCK_INIT(); MIF6_LOCK_INIT(); pim6_encap_cookie = encap_attach_func(AF_INET6, IPPROTO_PIM, pim6_encapcheck, (const struct protosw *)&in6_pim_protosw, NULL); if (pim6_encap_cookie == NULL) { printf("ip6_mroute: unable to attach pim6 encap\n"); MIF6_LOCK_DESTROY(); MFC6_LOCK_DESTROY(); MROUTER6_LOCK_DESTROY(); return (EINVAL); } ip6_mforward = X_ip6_mforward; ip6_mrouter_done = X_ip6_mrouter_done; ip6_mrouter_get = X_ip6_mrouter_get; ip6_mrouter_set = X_ip6_mrouter_set; mrt6_ioctl = X_mrt6_ioctl; break; case MOD_UNLOAD: if (V_ip6_mrouter != NULL) return EINVAL; if (pim6_encap_cookie) { encap_detach(pim6_encap_cookie); pim6_encap_cookie = NULL; } X_ip6_mrouter_done(); ip6_mforward = NULL; ip6_mrouter_done = NULL; ip6_mrouter_get = NULL; ip6_mrouter_set = NULL; mrt6_ioctl = NULL; MIF6_LOCK_DESTROY(); MFC6_LOCK_DESTROY(); MROUTER6_LOCK_DESTROY(); break; default: return (EOPNOTSUPP); } return (0); } static moduledata_t ip6_mroutemod = { "ip6_mroute", ip6_mroute_modevent, 0 }; DECLARE_MODULE(ip6_mroute, ip6_mroutemod, SI_SUB_PSEUDO, SI_ORDER_ANY); Index: head/sys/netinet6/ip6_mroute.h =================================================================== --- head/sys/netinet6/ip6_mroute.h (revision 281171) +++ head/sys/netinet6/ip6_mroute.h (revision 281172) @@ -1,265 +1,279 @@ /*- * Copyright (C) 1998 WIDE Project. * 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. Neither the name of the project 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 PROJECT 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 PROJECT 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. * * $KAME: ip6_mroute.h,v 1.19 2001/06/14 06:12:55 suz Exp $ * $FreeBSD$ */ /* BSDI ip_mroute.h,v 2.5 1996/10/11 16:01:48 pjd Exp */ /* * Definitions for IP multicast forwarding. * * Written by David Waitzman, BBN Labs, August 1988. * Modified by Steve Deering, Stanford, February 1989. * Modified by Ajit Thyagarajan, PARC, August 1993. * Modified by Ajit Thyagarajan, PARC, August 1994. * Modified by Ahmed Helmy, USC, September 1996. * * MROUTING Revision: 1.2 */ #ifndef _NETINET6_IP6_MROUTE_H_ #define _NETINET6_IP6_MROUTE_H_ /* * Multicast Routing set/getsockopt commands. */ #ifdef _KERNEL #define MRT6_OINIT 100 /* initialize forwarder (omrt6msg) */ #endif #define MRT6_DONE 101 /* shut down forwarder */ #define MRT6_ADD_MIF 102 /* add multicast interface */ #define MRT6_DEL_MIF 103 /* delete multicast interface */ #define MRT6_ADD_MFC 104 /* insert forwarding cache entry */ #define MRT6_DEL_MFC 105 /* delete forwarding cache entry */ #define MRT6_PIM 107 /* enable pim code */ #define MRT6_INIT 108 /* initialize forwarder (mrt6msg) */ #if BSD >= 199103 #define GET_TIME(t) microtime(&t) #elif defined(sun) #define GET_TIME(t) uniqtime(&t) #else #define GET_TIME(t) ((t) = time) #endif /* * Types and macros for handling bitmaps with one bit per multicast interface. */ typedef u_short mifi_t; /* type of a mif index */ #define MAXMIFS 64 #ifndef IF_SETSIZE #define IF_SETSIZE 256 #endif typedef u_int32_t if_mask; #define NIFBITS (sizeof(if_mask) * NBBY) /* bits per mask */ #ifndef howmany #define howmany(x, y) (((x) + ((y) - 1)) / (y)) #endif typedef struct if_set { if_mask ifs_bits[howmany(IF_SETSIZE, NIFBITS)]; } if_set; #define IF_SET(n, p) ((p)->ifs_bits[(n)/NIFBITS] |= (1 << ((n) % NIFBITS))) #define IF_CLR(n, p) ((p)->ifs_bits[(n)/NIFBITS] &= ~(1 << ((n) % NIFBITS))) #define IF_ISSET(n, p) ((p)->ifs_bits[(n)/NIFBITS] & (1 << ((n) % NIFBITS))) #define IF_COPY(f, t) bcopy(f, t, sizeof(*(f))) #define IF_ZERO(p) bzero(p, sizeof(*(p))) /* * Argument structure for MRT6_ADD_IF. */ struct mif6ctl { mifi_t mif6c_mifi; /* the index of the mif to be added */ u_char mif6c_flags; /* MIFF_ flags defined below */ u_short mif6c_pifi; /* the index of the physical IF */ }; #define MIFF_REGISTER 0x1 /* mif represents a register end-point */ /* * Argument structure for MRT6_ADD_MFC and MRT6_DEL_MFC */ struct mf6cctl { struct sockaddr_in6 mf6cc_origin; /* IPv6 origin of mcasts */ struct sockaddr_in6 mf6cc_mcastgrp; /* multicast group associated */ mifi_t mf6cc_parent; /* incoming ifindex */ struct if_set mf6cc_ifset; /* set of forwarding ifs */ }; /* * The kernel's multicast routing statistics. */ struct mrt6stat { uint64_t mrt6s_mfc_lookups; /* # forw. cache hash table hits */ uint64_t mrt6s_mfc_misses; /* # forw. cache hash table misses */ uint64_t mrt6s_upcalls; /* # calls to multicast routing daemon */ uint64_t mrt6s_no_route; /* no route for packet's origin */ uint64_t mrt6s_bad_tunnel; /* malformed tunnel options */ uint64_t mrt6s_cant_tunnel; /* no room for tunnel options */ uint64_t mrt6s_wrong_if; /* arrived on wrong interface */ uint64_t mrt6s_upq_ovflw; /* upcall Q overflow */ uint64_t mrt6s_cache_cleanups; /* # entries with no upcalls */ uint64_t mrt6s_drop_sel; /* pkts dropped selectively */ uint64_t mrt6s_q_overflow; /* pkts dropped - Q overflow */ uint64_t mrt6s_pkt2large; /* pkts dropped - size > BKT SIZE */ uint64_t mrt6s_upq_sockfull; /* upcalls dropped - socket full */ }; #ifdef MRT6_OINIT /* * Struct used to communicate from kernel to multicast router * note the convenient similarity to an IPv6 header. * XXX old version, superseded by mrt6msg. */ struct omrt6msg { u_long unused1; u_char im6_msgtype; /* what type of message */ u_char im6_mbz; /* must be zero */ u_char im6_mif; /* mif rec'd on */ u_char unused2; struct in6_addr im6_src, im6_dst; }; #endif /* * Structure used to communicate from kernel to multicast router. * We'll overlay the structure onto an MLD header (not an IPv6 header * like igmpmsg{} used for IPv4 implementation). This is because this * structure will be passed via an IPv6 raw socket, on which an application * will only receive the payload i.e. the data after the IPv6 header and all * the extension headers. (see Section 3 of RFC3542) */ struct mrt6msg { #define MRT6MSG_NOCACHE 1 #define MRT6MSG_WRONGMIF 2 #define MRT6MSG_WHOLEPKT 3 /* used for user level encap*/ u_char im6_mbz; /* must be zero */ u_char im6_msgtype; /* what type of message */ u_int16_t im6_mif; /* mif rec'd on */ u_int32_t im6_pad; /* padding for 64bit arch */ struct in6_addr im6_src, im6_dst; }; /* * Argument structure used by multicast routing daemon to get src-grp * packet counts */ struct sioc_sg_req6 { struct sockaddr_in6 src; struct sockaddr_in6 grp; u_quad_t pktcnt; u_quad_t bytecnt; u_quad_t wrong_if; }; /* * Argument structure used by mrouted to get mif pkt counts */ struct sioc_mif_req6 { mifi_t mifi; /* mif number */ u_quad_t icount; /* Input packet count on mif */ u_quad_t ocount; /* Output packet count on mif */ u_quad_t ibytes; /* Input byte count on mif */ u_quad_t obytes; /* Output byte count on mif */ }; +/* + * Structure to export 'struct mif6' to userland via sysctl. + */ +struct mif6_sctl { + u_char m6_flags; /* MIFF_ flags defined above */ + u_int m6_rate_limit; /* max rate */ + struct in6_addr m6_lcl_addr; /* local interface address */ + uint32_t m6_ifp; /* interface index */ + u_quad_t m6_pkt_in; /* # pkts in on interface */ + u_quad_t m6_pkt_out; /* # pkts out on interface */ + u_quad_t m6_bytes_in; /* # bytes in on interface */ + u_quad_t m6_bytes_out; /* # bytes out on interface */ +}; + #if defined(_KERNEL) || defined(KERNEL) /* * The kernel's multicast-interface structure. */ struct mif6 { u_char m6_flags; /* MIFF_ flags defined above */ u_int m6_rate_limit; /* max rate */ struct in6_addr m6_lcl_addr; /* local interface address */ struct ifnet *m6_ifp; /* pointer to interface */ u_quad_t m6_pkt_in; /* # pkts in on interface */ u_quad_t m6_pkt_out; /* # pkts out on interface */ u_quad_t m6_bytes_in; /* # bytes in on interface */ u_quad_t m6_bytes_out; /* # bytes out on interface */ #ifdef notyet u_int m6_rsvp_on; /* RSVP listening on this vif */ struct socket *m6_rsvpd; /* RSVP daemon socket */ #endif }; /* * The kernel's multicast forwarding cache entry structure */ struct mf6c { struct sockaddr_in6 mf6c_origin; /* IPv6 origin of mcasts */ struct sockaddr_in6 mf6c_mcastgrp; /* multicast group associated*/ mifi_t mf6c_parent; /* incoming IF */ struct if_set mf6c_ifset; /* set of outgoing IFs */ u_quad_t mf6c_pkt_cnt; /* pkt count for src-grp */ u_quad_t mf6c_byte_cnt; /* byte count for src-grp */ u_quad_t mf6c_wrong_if; /* wrong if for src-grp */ int mf6c_expire; /* time to clean entry up */ struct timeval mf6c_last_assert; /* last time I sent an assert*/ struct rtdetq *mf6c_stall; /* pkts waiting for route */ struct mf6c *mf6c_next; /* hash table linkage */ }; #define MF6C_INCOMPLETE_PARENT ((mifi_t)-1) /* * Argument structure used for pkt info. while upcall is made */ #ifndef _NETINET_IP_MROUTE_H_ struct rtdetq { /* XXX: rtdetq is also defined in ip_mroute.h */ struct mbuf *m; /* A copy of the packet */ struct ifnet *ifp; /* Interface pkt came in on */ #ifdef UPCALL_TIMING struct timeval t; /* Timestamp */ #endif /* UPCALL_TIMING */ struct rtdetq *next; }; #endif /* _NETINET_IP_MROUTE_H_ */ #define MF6CTBLSIZ 256 #if (MF6CTBLSIZ & (MF6CTBLSIZ - 1)) == 0 /* from sys:route.h */ #define MF6CHASHMOD(h) ((h) & (MF6CTBLSIZ - 1)) #else #define MF6CHASHMOD(h) ((h) % MF6CTBLSIZ) #endif #define MAX_UPQ6 4 /* max. no of pkts in upcall Q */ extern int (*ip6_mrouter_set)(struct socket *so, struct sockopt *sopt); extern int (*ip6_mrouter_get)(struct socket *so, struct sockopt *sopt); extern int (*ip6_mrouter_done)(void); extern int (*mrt6_ioctl)(u_long, caddr_t); #endif /* _KERNEL */ #endif /* !_NETINET6_IP6_MROUTE_H_ */ Index: head/sys/sys/param.h =================================================================== --- head/sys/sys/param.h (revision 281171) +++ head/sys/sys/param.h (revision 281172) @@ -1,348 +1,348 @@ /*- * Copyright (c) 1982, 1986, 1989, 1993 * The Regents of the University of California. All rights reserved. * (c) UNIX System Laboratories, Inc. * All or some portions of this file are derived from material licensed * to the University of California by American Telephone and Telegraph * Co. or Unix System Laboratories, Inc. and are reproduced herein with * the permission of UNIX System Laboratories, Inc. * * 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. * 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. * * @(#)param.h 8.3 (Berkeley) 4/4/95 * $FreeBSD$ */ #ifndef _SYS_PARAM_H_ #define _SYS_PARAM_H_ #include #define BSD 199506 /* System version (year & month). */ #define BSD4_3 1 #define BSD4_4 1 /* * __FreeBSD_version numbers are documented in the Porter's Handbook. * If you bump the version for any reason, you should update the documentation * there. * Currently this lives here in the doc/ repository: * * head/en_US.ISO8859-1/books/porters-handbook/book.xml * * scheme is: Rxx * 'R' is in the range 0 to 4 if this is a release branch or * x.0-CURRENT before RELENG_*_0 is created, otherwise 'R' is * in the range 5 to 9. */ #undef __FreeBSD_version -#define __FreeBSD_version 1100067 /* Master, propagated to newvers */ +#define __FreeBSD_version 1100068 /* Master, propagated to newvers */ /* * __FreeBSD_kernel__ indicates that this system uses the kernel of FreeBSD, * which by definition is always true on FreeBSD. This macro is also defined * on other systems that use the kernel of FreeBSD, such as GNU/kFreeBSD. * * It is tempting to use this macro in userland code when we want to enable * kernel-specific routines, and in fact it's fine to do this in code that * is part of FreeBSD itself. However, be aware that as presence of this * macro is still not widespread (e.g. older FreeBSD versions, 3rd party * compilers, etc), it is STRONGLY DISCOURAGED to check for this macro in * external applications without also checking for __FreeBSD__ as an * alternative. */ #undef __FreeBSD_kernel__ #define __FreeBSD_kernel__ #ifdef _KERNEL #define P_OSREL_SIGWAIT 700000 #define P_OSREL_SIGSEGV 700004 #define P_OSREL_MAP_ANON 800104 #define P_OSREL_MAP_FSTRICT 1100036 #define P_OSREL_MAJOR(x) ((x) / 100000) #endif #ifndef LOCORE #include #endif /* * Machine-independent constants (some used in following include files). * Redefined constants are from POSIX 1003.1 limits file. * * MAXCOMLEN should be >= sizeof(ac_comm) (see ) */ #include #define MAXCOMLEN 19 /* max command name remembered */ #define MAXINTERP PATH_MAX /* max interpreter file name length */ #define MAXLOGNAME 33 /* max login name length (incl. NUL) */ #define MAXUPRC CHILD_MAX /* max simultaneous processes */ #define NCARGS ARG_MAX /* max bytes for an exec function */ #define NGROUPS (NGROUPS_MAX+1) /* max number groups */ #define NOFILE OPEN_MAX /* max open files per process */ #define NOGROUP 65535 /* marker for empty group set member */ #define MAXHOSTNAMELEN 256 /* max hostname size */ #define SPECNAMELEN 63 /* max length of devicename */ /* More types and definitions used throughout the kernel. */ #ifdef _KERNEL #include #include #ifndef LOCORE #include #include #endif #ifndef FALSE #define FALSE 0 #endif #ifndef TRUE #define TRUE 1 #endif #endif #ifndef _KERNEL /* Signals. */ #include #endif /* Machine type dependent parameters. */ #include #ifndef _KERNEL #include #endif #ifndef DEV_BSHIFT #define DEV_BSHIFT 9 /* log2(DEV_BSIZE) */ #endif #define DEV_BSIZE (1<>PAGE_SHIFT) #endif /* * btodb() is messy and perhaps slow because `bytes' may be an off_t. We * want to shift an unsigned type to avoid sign extension and we don't * want to widen `bytes' unnecessarily. Assume that the result fits in * a daddr_t. */ #ifndef btodb #define btodb(bytes) /* calculates (bytes / DEV_BSIZE) */ \ (sizeof (bytes) > sizeof(long) \ ? (daddr_t)((unsigned long long)(bytes) >> DEV_BSHIFT) \ : (daddr_t)((unsigned long)(bytes) >> DEV_BSHIFT)) #endif #ifndef dbtob #define dbtob(db) /* calculates (db * DEV_BSIZE) */ \ ((off_t)(db) << DEV_BSHIFT) #endif #define PRIMASK 0x0ff #define PCATCH 0x100 /* OR'd with pri for tsleep to check signals */ #define PDROP 0x200 /* OR'd with pri to stop re-entry of interlock mutex */ #define NZERO 0 /* default "nice" */ #define NBBY 8 /* number of bits in a byte */ #define NBPW sizeof(int) /* number of bytes per word (integer) */ #define CMASK 022 /* default file mask: S_IWGRP|S_IWOTH */ #define NODEV (dev_t)(-1) /* non-existent device */ /* * File system parameters and macros. * * MAXBSIZE - Filesystems are made out of blocks of at most MAXBSIZE bytes * per block. MAXBSIZE may be made larger without effecting * any existing filesystems as long as it does not exceed MAXPHYS, * and may be made smaller at the risk of not being able to use * filesystems which require a block size exceeding MAXBSIZE. * * BKVASIZE - Nominal buffer space per buffer, in bytes. BKVASIZE is the * minimum KVM memory reservation the kernel is willing to make. * Filesystems can of course request smaller chunks. Actual * backing memory uses a chunk size of a page (PAGE_SIZE). * * If you make BKVASIZE too small you risk seriously fragmenting * the buffer KVM map which may slow things down a bit. If you * make it too big the kernel will not be able to optimally use * the KVM memory reserved for the buffer cache and will wind * up with too-few buffers. * * The default is 16384, roughly 2x the block size used by a * normal UFS filesystem. */ #define MAXBSIZE 65536 /* must be power of 2 */ #define BKVASIZE 16384 /* must be power of 2 */ #define BKVAMASK (BKVASIZE-1) /* * MAXPATHLEN defines the longest permissible path length after expanding * symbolic links. It is used to allocate a temporary buffer from the buffer * pool in which to do the name expansion, hence should be a power of two, * and must be less than or equal to MAXBSIZE. MAXSYMLINKS defines the * maximum number of symbolic links that may be expanded in a path name. * It should be set high enough to allow all legitimate uses, but halt * infinite loops reasonably quickly. */ #define MAXPATHLEN PATH_MAX #define MAXSYMLINKS 32 /* Bit map related macros. */ #define setbit(a,i) (((unsigned char *)(a))[(i)/NBBY] |= 1<<((i)%NBBY)) #define clrbit(a,i) (((unsigned char *)(a))[(i)/NBBY] &= ~(1<<((i)%NBBY))) #define isset(a,i) \ (((const unsigned char *)(a))[(i)/NBBY] & (1<<((i)%NBBY))) #define isclr(a,i) \ ((((const unsigned char *)(a))[(i)/NBBY] & (1<<((i)%NBBY))) == 0) /* Macros for counting and rounding. */ #ifndef howmany #define howmany(x, y) (((x)+((y)-1))/(y)) #endif #define nitems(x) (sizeof((x)) / sizeof((x)[0])) #define rounddown(x, y) (((x)/(y))*(y)) #define rounddown2(x, y) ((x)&(~((y)-1))) /* if y is power of two */ #define roundup(x, y) ((((x)+((y)-1))/(y))*(y)) /* to any y */ #define roundup2(x, y) (((x)+((y)-1))&(~((y)-1))) /* if y is powers of two */ #define powerof2(x) ((((x)-1)&(x))==0) /* Macros for min/max. */ #define MIN(a,b) (((a)<(b))?(a):(b)) #define MAX(a,b) (((a)>(b))?(a):(b)) #ifdef _KERNEL /* * Basic byte order function prototypes for non-inline functions. */ #ifndef LOCORE #ifndef _BYTEORDER_PROTOTYPED #define _BYTEORDER_PROTOTYPED __BEGIN_DECLS __uint32_t htonl(__uint32_t); __uint16_t htons(__uint16_t); __uint32_t ntohl(__uint32_t); __uint16_t ntohs(__uint16_t); __END_DECLS #endif #endif #ifndef lint #ifndef _BYTEORDER_FUNC_DEFINED #define _BYTEORDER_FUNC_DEFINED #define htonl(x) __htonl(x) #define htons(x) __htons(x) #define ntohl(x) __ntohl(x) #define ntohs(x) __ntohs(x) #endif /* !_BYTEORDER_FUNC_DEFINED */ #endif /* lint */ #endif /* _KERNEL */ /* * Scale factor for scaled integers used to count %cpu time and load avgs. * * The number of CPU `tick's that map to a unique `%age' can be expressed * by the formula (1 / (2 ^ (FSHIFT - 11))). The maximum load average that * can be calculated (assuming 32 bits) can be closely approximated using * the formula (2 ^ (2 * (16 - FSHIFT))) for (FSHIFT < 15). * * For the scheduler to maintain a 1:1 mapping of CPU `tick' to `%age', * FSHIFT must be at least 11; this gives us a maximum load avg of ~1024. */ #define FSHIFT 11 /* bits to right of fixed binary point */ #define FSCALE (1<> (PAGE_SHIFT - DEV_BSHIFT)) #define ctodb(db) /* calculates pages to devblks */ \ ((db) << (PAGE_SHIFT - DEV_BSHIFT)) /* * Old spelling of __containerof(). */ #define member2struct(s, m, x) \ ((struct s *)(void *)((char *)(x) - offsetof(struct s, m))) /* * Access a variable length array that has been declared as a fixed * length array. */ #define __PAST_END(array, offset) (((__typeof__(*(array)) *)(array))[offset]) #endif /* _SYS_PARAM_H_ */ Index: head/usr.bin/netstat/mroute6.c =================================================================== --- head/usr.bin/netstat/mroute6.c (revision 281171) +++ head/usr.bin/netstat/mroute6.c (revision 281172) @@ -1,324 +1,279 @@ /*- * Copyright (C) 1998 WIDE Project. * 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. Neither the name of the project 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 PROJECT 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 PROJECT 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. */ /*- * Copyright (c) 1989 Stephen Deering * Copyright (c) 1992, 1993 * The Regents of the University of California. All rights reserved. * * This code is derived from software contributed to Berkeley by * Stephen Deering of Stanford University. * * 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. * * @(#)mroute.c 8.2 (Berkeley) 4/28/95 */ #include __FBSDID("$FreeBSD$"); #ifdef INET6 #include #include #include #include #include #include #include #include #include -#include #include #include #include -#include #include #include #include #include #include #define KERNEL 1 #include #undef KERNEL #include "netstat.h" -/* - * kvm(3) bindings for every needed symbol - */ -static struct nlist mrl[] = { -#define N_MF6CTABLE 0 - { .n_name = "_mf6ctable" }, -#define N_MIF6TABLE 1 - { .n_name = "_mif6table" }, -#define N_MRT6STAT 2 - { .n_name = "_mrt6stat" }, - { .n_name = NULL }, -}; - - #define WID_ORG (Wflag ? 39 : (numeric_addr ? 29 : 18)) /* width of origin column */ #define WID_GRP (Wflag ? 18 : (numeric_addr ? 16 : 18)) /* width of group column */ void mroute6pr() { struct mf6c *mf6ctable[MF6CTBLSIZ], *mfcp; - struct mif6 mif6table[MAXMIFS]; + struct mif6_sctl mif6table[MAXMIFS]; struct mf6c mfc; struct rtdetq rte, *rtep; - struct mif6 *mifp; - u_long mfcaddr, mifaddr; + struct mif6_sctl *mifp; mifi_t mifi; int i; int banner_printed; int saved_numeric_addr; mifi_t maxmif = 0; long int waitings; size_t len; - kresolve_list(mrl); - mfcaddr = mrl[N_MF6CTABLE].n_value; - mifaddr = mrl[N_MIF6TABLE].n_value; + if (live == 0) + return; - if (mfcaddr == 0 || mifaddr == 0) { - fprintf(stderr, "No IPv6 MROUTING kernel support.\n"); + len = sizeof(mif6table); + if (sysctlbyname("net.inet6.ip6.mif6table", mif6table, &len, NULL, 0) < + 0) { + xo_warn("sysctl: net.inet6.ip6.mif6table"); return; } - len = sizeof(mif6table); - if (live) { - if (sysctlbyname("net.inet6.ip6.mif6table", mif6table, &len, - NULL, 0) < 0) { - xo_warn("sysctl: net.inet6.ip6.mif6table"); - return; - } - } else - kread(mifaddr, (char *)mif6table, sizeof(mif6table)); - saved_numeric_addr = numeric_addr; numeric_addr = 1; banner_printed = 0; for (mifi = 0, mifp = mif6table; mifi < MAXMIFS; ++mifi, ++mifp) { - struct ifnet ifnet; char ifname[IFNAMSIZ]; - if (mifp->m6_ifp == NULL) + if (mifp->m6_ifp == 0) continue; - /* XXX KVM */ - kread((u_long)mifp->m6_ifp, (char *)&ifnet, sizeof(ifnet)); - maxmif = mifi; if (!banner_printed) { xo_open_list("multicast-interface"); xo_emit("\n{T:IPv6 Multicast Interface Table}\n" "{T: Mif Rate PhyIF Pkts-In Pkts-Out}\n"); banner_printed = 1; } xo_open_instance("multicast-interface"); xo_emit(" {:mif/%2u} {:rate-limit/%4d}", mifi, mifp->m6_rate_limit); xo_emit(" {:ifname/%5s}", (mifp->m6_flags & MIFF_REGISTER) ? - "reg0" : if_indextoname(ifnet.if_index, ifname)); + "reg0" : if_indextoname(mifp->m6_ifp, ifname)); xo_emit(" {:received-packets/%9ju} {:sent-packets/%9ju}\n", (uintmax_t)mifp->m6_pkt_in, (uintmax_t)mifp->m6_pkt_out); xo_close_instance("multicast-interface"); } if (banner_printed) xo_open_list("multicast-interface"); else xo_emit("\n{T:IPv6 Multicast Interface Table is empty}\n"); len = sizeof(mf6ctable); - if (live) { - if (sysctlbyname("net.inet6.ip6.mf6ctable", mf6ctable, &len, - NULL, 0) < 0) { - xo_warn("sysctl: net.inet6.ip6.mf6ctable"); - return; - } - } else - kread(mfcaddr, (char *)mf6ctable, sizeof(mf6ctable)); + if (sysctlbyname("net.inet6.ip6.mf6ctable", mf6ctable, &len, NULL, 0) < + 0) { + xo_warn("sysctl: net.inet6.ip6.mf6ctable"); + return; + } banner_printed = 0; for (i = 0; i < MF6CTBLSIZ; ++i) { mfcp = mf6ctable[i]; while(mfcp) { kread((u_long)mfcp, (char *)&mfc, sizeof(mfc)); if (!banner_printed) { xo_open_list("multicast-forwarding-cache"); xo_emit("\n" "{T:IPv6 Multicast Forwarding Cache}\n"); xo_emit(" {T:%-*.*s} {T:%-*.*s} {T:%s}", WID_ORG, WID_ORG, "Origin", WID_GRP, WID_GRP, "Group", " Packets Waits In-Mif Out-Mifs\n"); banner_printed = 1; } xo_open_instance("multicast-forwarding-cache"); xo_emit(" {:origin/%-*.*s}", WID_ORG, WID_ORG, routename6(&mfc.mf6c_origin)); xo_emit(" {:group/%-*.*s}", WID_GRP, WID_GRP, routename6(&mfc.mf6c_mcastgrp)); xo_emit(" {:total-packets/%9ju}", (uintmax_t)mfc.mf6c_pkt_cnt); for (waitings = 0, rtep = mfc.mf6c_stall; rtep; ) { waitings++; /* XXX KVM */ kread((u_long)rtep, (char *)&rte, sizeof(rte)); rtep = rte.next; } xo_emit(" {:waitings/%3ld}", waitings); if (mfc.mf6c_parent == MF6C_INCOMPLETE_PARENT) xo_emit(" --- "); else xo_emit(" {:parent/%3d} ", mfc.mf6c_parent); xo_open_list("mif"); for (mifi = 0; mifi <= maxmif; mifi++) { if (IF_ISSET(mifi, &mfc.mf6c_ifset)) xo_emit(" {l:%u}", mifi); } xo_close_list("mif"); xo_emit("\n"); mfcp = mfc.mf6c_next; xo_close_instance("multicast-forwarding-cache"); } } if (banner_printed) xo_close_list("multicast-forwarding-cache"); else xo_emit("\n{T:IPv6 Multicast Forwarding Table is empty}\n"); xo_emit("\n"); numeric_addr = saved_numeric_addr; } void mrt6_stats() { struct mrt6stat mrtstat; - u_long mstaddr; size_t len = sizeof mrtstat; - kresolve_list(mrl); - mstaddr = mrl[N_MRT6STAT].n_value; - - if (mstaddr == 0) { - fprintf(stderr, "No IPv6 MROUTING kernel support.\n"); + if (sysctlbyname("net.inet6.ip6.mrt6stat", &mrtstat, &len, NULL, 0) < + 0) { + xo_warn("sysctl: net.inet6.ip6.mrt6stat"); return; } - - if (live) { - if (sysctlbyname("net.inet6.ip6.mrt6stat", &mrtstat, &len, - NULL, 0) < 0) { - xo_warn("sysctl: net.inet6.ip6.mrt6stat"); - return; - } - } else - kread(mstaddr, (char *)&mrtstat, sizeof(mrtstat)); xo_open_container("multicast-statistics"); xo_emit("{T:IPv6 multicast forwarding}:\n"); #define p(f, m) if (mrtstat.f || sflag <= 1) \ xo_emit(m, (uintmax_t)mrtstat.f, plural(mrtstat.f)) #define p2(f, m) if (mrtstat.f || sflag <= 1) \ xo_emit(m, (uintmax_t)mrtstat.f, plurales(mrtstat.f)) p(mrt6s_mfc_lookups, "\t{:cache-lookups/%ju} " "{N:/multicast forwarding cache lookup%s}\n"); p2(mrt6s_mfc_misses, "\t{:cache-misses/%ju} " "{N:/multicast forwarding cache miss%s}\n"); p(mrt6s_upcalls, "\t{:upcalls/%ju} " "{N:/upcall%s to multicast routing daemon}\n"); p(mrt6s_upq_ovflw, "\t{:upcall-overflows/%ju} " "{N:/upcall queue overflow%s}\n"); p(mrt6s_upq_sockfull, "\t{:upcalls-dropped-full-buffer/%ju} " "{N:/upcall%s dropped due to full socket buffer}\n"); p(mrt6s_cache_cleanups, "\t{:cache-cleanups/%ju} " "{N:/cache cleanup%s}\n"); p(mrt6s_no_route, "\t{:dropped-no-origin/%ju} " "{N:/datagram%s with no route for origin}\n"); p(mrt6s_bad_tunnel, "\t{:dropped-bad-tunnel/%ju} " "{N:/datagram%s arrived with bad tunneling}\n"); p(mrt6s_cant_tunnel, "\t{:dropped-could-not-tunnel/%ju} " "{N:/datagram%s could not be tunneled}\n"); p(mrt6s_wrong_if, "\t{:dropped-wrong-incoming-interface/%ju} " "{N:/datagram%s arrived on wrong interface}\n"); p(mrt6s_drop_sel, "\t{:dropped-selectively/%ju} " "{N:/datagram%s selectively dropped}\n"); p(mrt6s_q_overflow, "\t{:dropped-queue-overflow/%ju} " "{N:/datagram%s dropped due to queue overflow}\n"); p(mrt6s_pkt2large, "\t{:dropped-too-large/%ju} " "{N:/datagram%s dropped for being too large}\n"); #undef p2 #undef p xo_close_container("multicast-statistics"); } #endif /*INET6*/ Index: head/usr.bin/netstat/netstat.1 =================================================================== --- head/usr.bin/netstat/netstat.1 (revision 281171) +++ head/usr.bin/netstat/netstat.1 (revision 281172) @@ -1,822 +1,820 @@ .\" Copyright (c) 1983, 1990, 1992, 1993 .\" 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 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. .\" 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. .\" .\" @(#)netstat.1 8.8 (Berkeley) 4/18/94 .\" $FreeBSD$ .\" -.Dd February 21, 2015 +.Dd April 7, 2015 .Dt NETSTAT 1 .Os .Sh NAME .Nm netstat .Nd show network status and statistics .Sh SYNOPSIS .Bk -words .Bl -tag -width "netstat" .It Nm .Op Fl -libxo .Op Fl 46AaLnRSTWx .Op Fl f Ar protocol_family | Fl p Ar protocol .Op Fl M Ar core .Op Fl N Ar system .It Nm Fl i | I Ar interface .Op Fl -libxo .Op Fl 46abdhnW .Op Fl f Ar address_family .Op Fl M Ar core .Op Fl N Ar system .It Nm Fl w Ar wait .Op Fl -libxo .Op Fl I Ar interface .Op Fl 46d .Op Fl M Ar core .Op Fl N Ar system .Op Fl q Ar howmany .It Nm Fl s .Op Fl -libxo .Op Fl 46sz .Op Fl f Ar protocol_family | Fl p Ar protocol .Op Fl M Ar core .Op Fl N Ar system .It Nm Fl i | I Ar interface Fl s .Op Fl -libxo .Op Fl 46s .Op Fl f Ar protocol_family | Fl p Ar protocol .Op Fl M Ar core .Op Fl N Ar system .It Nm Fl m .Op Fl -libxo .Op Fl M Ar core .Op Fl N Ar system .It Nm Fl B .Op Fl -libxo .Op Fl z .Op Fl I Ar interface .It Nm Fl r .Op Fl -libxo .Op Fl 46AnW .Op Fl F Ar fibnum .Op Fl f Ar address_family .Op Fl M Ar core .Op Fl N Ar system .It Nm Fl rs .Op Fl -libxo .Op Fl s .Op Fl M Ar core .Op Fl N Ar system .It Nm Fl g .Op Fl -libxo .Op Fl 46W .Op Fl f Ar address_family -.Op Fl M Ar core -.Op Fl N Ar system .It Nm Fl gs .Op Fl -libxo .Op Fl 46s .Op Fl f Ar address_family .Op Fl M Ar core .Op Fl N Ar system .It Nm Fl Q .Op Fl -libxo .El .Ek .Sh DESCRIPTION The .Nm command symbolically displays the contents of various network-related data structures. There are a number of output formats, depending on the options for the information presented. .Bl -tag -width indent .It Xo .Bk -words .Nm .Op Fl 46AaLnRSTWx .Op Fl f Ar protocol_family | Fl p Ar protocol .Op Fl M Ar core .Op Fl N Ar system .Ek .Xc Display a list of active sockets (protocol control blocks) for each network protocol. .Pp The default display for active sockets shows the local and remote addresses, send and receive queue sizes (in bytes), protocol, and the internal state of the protocol. Address formats are of the form .Dq host.port or .Dq network.port if a socket's address specifies a network but no specific host address. When known, the host and network addresses are displayed symbolically according to the databases .Xr hosts 5 and .Xr networks 5 , respectively. If a symbolic name for an address is unknown, or if the .Fl n option is specified, the address is printed numerically, according to the address family. For more information regarding the Internet IPv4 .Dq dot format , refer to .Xr inet 3 . Unspecified, or .Dq wildcard , addresses and ports appear as .Dq Li * . .Bl -tag -width indent .It Fl 4 Show IPv4 only. See .Sx GENERAL OPTIONS . .It Fl 6 Show IPv6 only. See .Sx GENERAL OPTIONS . .It Fl A Show the address of a protocol control block (PCB) associated with a socket; used for debugging. .It Fl a Show the state of all sockets; normally sockets used by server processes are not shown. .It Fl L Show the size of the various listen queues. The first count shows the number of unaccepted connections, the second count shows the amount of unaccepted incomplete connections, and the third count is the maximum number of queued connections. .It Fl n Do not resolve numeric addresses and port numbers to names. See .Sx GENERAL OPTIONS . .It Fl R Display the flowid and flowtype for each socket. flowid is a 32 bit hardware specific identifier for each flow. flowtype defines which protocol fields are hashed to produce the id. A complete listing is available in .Pa sys/mbuf.h under .Dv M_HASHTYPE_* . .It Fl S Show network addresses as numbers (as with .Fl n ) but show ports symbolically. .It Fl T Display diagnostic information from the TCP control block. Fields include the number of packets requiring retransmission, received out-of-order, and those advertising a zero-sized window. .It Fl W Avoid truncating addresses even if this causes some fields to overflow. .It Fl x Display socket buffer and TCP timer statistics for each internet socket. .Pp The .Fl x flag causes .Nm to output all the information recorded about data stored in the socket buffers. The fields are: .Bl -column ".Li R-MBUF" .It Li R-MBUF Ta Number of mbufs in the receive queue. .It Li S-MBUF Ta Number of mbufs in the send queue. .It Li R-CLUS Ta Number of clusters, of any type, in the receive queue. .It Li S-CLUS Ta Number of clusters, of any type, in the send queue. .It Li R-HIWA Ta Receive buffer high water mark, in bytes. .It Li S-HIWA Ta Send buffer high water mark, in bytes. .It Li R-LOWA Ta Receive buffer low water mark, in bytes. .It Li S-LOWA Ta Send buffer low water mark, in bytes. .It Li R-BCNT Ta Receive buffer byte count. .It Li S-BCNT Ta Send buffer byte count. .It Li R-BMAX Ta Maximum bytes that can be used in the receive buffer. .It Li S-BMAX Ta Maximum bytes that can be used in the send buffer. .El .It Fl f Ar protocol_family Filter by .Ar protocol_family . See .Sx GENERAL OPTIONS . .It Fl p Ar protocol Filter by .Ar protocol . See .Sx GENERAL OPTIONS . .It Fl M Use an alternative core. See .Sx GENERAL OPTIONS . .It Fl N Use an alternative kernel image. See .Sx GENERAL OPTIONS . .El .It Xo .Bk -words .Nm .Fl i | I Ar interface .Op Fl 46abdhnW .Op Fl f Ar address_family .Op Fl M Ar core .Op Fl N Ar system .Ek .Xc Show the state of all network interfaces or a single .Ar interface which have been auto-configured (interfaces statically configured into a system, but not located at boot time are not shown). An asterisk .Pq Dq Li * after an interface name indicates that the interface is .Dq down . .Pp When .Nm is invoked with .Fl i .Pq all interfaces or .Fl I Ar interface , it provides a table of cumulative statistics regarding packets transferred, errors, and collisions. The network addresses of the interface and the maximum transmission unit .Pq Dq mtu are also displayed. .Bl -tag -width indent .It Fl 4 Show IPv4 only. See .Sx GENERAL OPTIONS . .It Fl 6 Show IPv6 only. See .Sx GENERAL OPTIONS . .It Fl a Multicast addresses currently in use are shown for each Ethernet interface and for each IP interface address. Multicast addresses are shown on separate lines following the interface address with which they are associated. .It Fl b Show the number of bytes in and out. .It Fl d Show the number of dropped packets. .It Fl h Print all counters in human readable form. .It Fl n Do not resolve numeric addresses and port numbers to names. See .Sx GENERAL OPTIONS . .It Fl W Avoid truncating interface names even if this causes some fields to overflow. .Sx GENERAL OPTIONS . .It Fl f Ar protocol_family Filter by .Ar protocol_family . See .Sx GENERAL OPTIONS . .El .It Xo .Bk -words .Nm .Fl w Ar wait .Op Fl I Ar interface .Op Fl 46d .Op Fl M Ar core .Op Fl N Ar system .Op Fl q Ar howmany .Ek .Xc At intervals of .Ar wait seconds, display the information regarding packet traffic on all configured network interfaces or a single .Ar interface . .Pp When .Nm is invoked with the .Fl w option and a .Ar wait interval argument, it displays a running count of statistics related to network interfaces. An obsolescent version of this option used a numeric parameter with no option, and is currently supported for backward compatibility. By default, this display summarizes information for all interfaces. Information for a specific interface may be displayed with the .Fl I Ar interface option. .Bl -tag -width indent .It Fl I Ar interface Only show information regarding .Ar interface .It Fl 4 Show IPv4 only. See .Sx GENERAL OPTIONS . .It Fl 6 Show IPv6 only. See .Sx GENERAL OPTIONS . .It Fl d Show the number of dropped packets. .It Fl M Use an alternative core. See .Sx GENERAL OPTIONS . .It Fl N Use an alternative kernel image. See .Sx GENERAL OPTIONS . .It Fl q Exit after .Ar howmany outputs. .El .It Xo .Bk -words .Nm .Fl s .Op Fl 46sz .Op Fl f Ar protocol_family | Fl p Ar protocol .Op Fl M Ar core .Op Fl N Ar system .Ek .Xc Display system-wide statistics for each network protocol. .Bl -tag -width indent .It Fl 4 Show IPv4 only. See .Sx GENERAL OPTIONS . .It Fl 6 Show IPv6 only. See .Sx GENERAL OPTIONS . .It Fl s If .Fl s is repeated, counters with a value of zero are suppressed. .It Fl z Reset statistic counters after displaying them. .It Fl f Ar protocol_family Filter by .Ar protocol_family . See .Sx GENERAL OPTIONS . .It Fl p Ar protocol Filter by .Ar protocol . See .Sx GENERAL OPTIONS . .It Fl M Use an alternative core. See .Sx GENERAL OPTIONS . .It Fl N Use an alternative kernel image See .Sx GENERAL OPTIONS . .El .It Xo .Bk -words .Nm .Fl i | I Ar interface Fl s .Op Fl 46s .Op Fl f Ar protocol_family | Fl p Ar protocol .Op Fl M Ar core .Op Fl N Ar system .Ek .Xc Display per-interface statistics for each network protocol. .Bl -tag -width indent .It Fl 4 Show IPv4 only See .Sx GENERAL OPTIONS . .It Fl 6 Show IPv6 only See .Sx GENERAL OPTIONS . .It Fl s If .Fl s is repeated, counters with a value of zero are suppressed. .It Fl f Ar protocol_family Filter by .Ar protocol_family . See .Sx GENERAL OPTIONS . .It Fl p Ar protocol Filter by .Ar protocol . See .Sx GENERAL OPTIONS . .It Fl M Use an alternative core See .Sx GENERAL OPTIONS . .It Fl N Use an alternative kernel image See .Sx GENERAL OPTIONS . .El .It Xo .Bk -words .Nm .Fl m .Op Fl M Ar core .Op Fl N Ar system .Ek .Xc Show statistics recorded by the memory management routines .Pq Xr mbuf 9 . The network manages a private pool of memory buffers. .Bl -tag -width indent .It Fl M Use an alternative core See .Sx GENERAL OPTIONS . .It Fl N Use an alternative kernel image See .Sx GENERAL OPTIONS . .El .It Xo .Bk -words .Nm .Fl B .Op Fl z .Op Fl I Ar interface .Ek .Xc Show statistics about .Xr bpf 4 peers. This includes information like how many packets have been matched, dropped and received by the bpf device, also information about current buffer sizes and device states. .Pp The .Xr bpf 4 flags displayed when .Nm is invoked with the .Fl B option represent the underlying parameters of the bpf peer. Each flag is represented as a single lower case letter. The mapping between the letters and flags in order of appearance are: .Bl -column ".Li i" .It Li p Ta Set if listening promiscuously .It Li i Ta Dv BIOCIMMEDIATE No has been set on the device .It Li f Ta Dv BIOCGHDRCMPLT No status: source link addresses are being filled automatically .It Li s Ta Dv BIOCGSEESENT No status: see packets originating locally and remotely on the interface. .It Li a Ta Packet reception generates a signal .It Li l Ta Dv BIOCLOCK No status: descriptor has been locked .El .Pp For more information about these flags, please refer to .Xr bpf 4 . .Bl -tag -width indent .It Fl z Reset statistic counters after displaying them. .El .It Xo .Bk -words .Nm .Fl r .Op Fl 46AnW .Op Fl F Ar fibnum .Op Fl f Ar address_family .Op Fl M Ar core .Op Fl N Ar system .Ek .Xc Display the contents of routing tables. .Pp When .Nm is invoked with the routing table option .Fl r , it lists the available routes and their status. Each route consists of a destination host or network, and a gateway to use in forwarding packets. The flags field shows a collection of information about the route stored as binary choices. The individual flags are discussed in more detail in the .Xr route 8 and .Xr route 4 manual pages. The mapping between letters and flags is: .Bl -column ".Li W" ".Dv RTF_WASCLONED" .It Li 1 Ta Dv RTF_PROTO1 Ta "Protocol specific routing flag #1" .It Li 2 Ta Dv RTF_PROTO2 Ta "Protocol specific routing flag #2" .It Li 3 Ta Dv RTF_PROTO3 Ta "Protocol specific routing flag #3" .It Li B Ta Dv RTF_BLACKHOLE Ta "Just discard pkts (during updates)" .It Li b Ta Dv RTF_BROADCAST Ta "The route represents a broadcast address" .It Li D Ta Dv RTF_DYNAMIC Ta "Created dynamically (by redirect)" .It Li G Ta Dv RTF_GATEWAY Ta "Destination requires forwarding by intermediary" .It Li H Ta Dv RTF_HOST Ta "Host entry (net otherwise)" .It Li L Ta Dv RTF_LLINFO Ta "Valid protocol to link address translation" .It Li M Ta Dv RTF_MODIFIED Ta "Modified dynamically (by redirect)" .It Li R Ta Dv RTF_REJECT Ta "Host or net unreachable" .It Li S Ta Dv RTF_STATIC Ta "Manually added" .It Li U Ta Dv RTF_UP Ta "Route usable" .It Li X Ta Dv RTF_XRESOLVE Ta "External daemon translates proto to link address" .El .Pp Direct routes are created for each interface attached to the local host; the gateway field for such entries shows the address of the outgoing interface. The refcnt field gives the current number of active uses of the route. Connection oriented protocols normally hold on to a single route for the duration of a connection while connectionless protocols obtain a route while sending to the same destination. The use field provides a count of the number of packets sent using that route. The interface entry indicates the network interface utilized for the route. .Bl -tag -width indent .It Fl 4 Show IPv4 only. See .Sx GENERAL OPTIONS . .It Fl 6 Show IPv6 only. See .Sx GENERAL OPTIONS . .It Fl A Show the contents of the internal Patricia tree structures; used for debugging. .It Fl n Do not resolve numeric addresses and port numbers to names. See .Sx GENERAL OPTIONS . .It Fl W Show the path MTU for each route, and print interface names with a wider field size. .It Fl F Display the routing table with the number .Ar fibnum . If the specified .Ar fibnum is -1 or .Fl F is not specified, the default routing table is displayed. .It Fl f Display the routing table for a particular .Ar address_family . .It Fl M Use an alternative core See .Sx GENERAL OPTIONS . .It Fl N Use an alternative kernel image See .Sx GENERAL OPTIONS . .El .It Xo .Bk -words .Nm .Fl rs .Op Fl s .Op Fl M Ar core .Op Fl N Ar system .Ek .Xc Display routing statistics. .Bl -tag -width indent .It Fl s If .Fl s is repeated, counters with a value of zero are suppressed. .It Fl M Use an alternative core See .Sx GENERAL OPTIONS . .It Fl N Use an alternative kernel image See .Sx GENERAL OPTIONS . .El .It Xo .Bk -words .Nm .Fl g .Op Fl 46W .Op Fl f Ar address_family .Op Fl M Ar core .Op Fl N Ar system .Ek .Xc Display the contents of the multicast virtual interface tables, and multicast forwarding caches. Entries in these tables will appear only when the kernel is actively forwarding multicast sessions. This option is applicable only to the .Cm inet and .Cm inet6 address families. .Bl -tag -width indent .It Fl 4 Show IPv4 only See .Sx GENERAL OPTIONS . .It Fl 6 Show IPv6 only See .Sx GENERAL OPTIONS . .It Fl W Avoid truncating addresses even if this causes some fields to overflow. .It Fl f Ar protocol_family Filter by .Ar protocol_family . See .Sx GENERAL OPTIONS . .It Fl M Use an alternative core See .Sx GENERAL OPTIONS . .It Fl N Use an alternative kernel image See .Sx GENERAL OPTIONS . .El .It Xo .Bk -words .Nm .Fl gs .Op Fl 46s .Op Fl f Ar address_family .Op Fl M Ar core .Op Fl N Ar system .Ek .Xc Show multicast routing statistics. .Bl -tag -width indent .It Fl 4 Show IPv4 only See .Sx GENERAL OPTIONS . .It Fl 6 Show IPv6 only See .Sx GENERAL OPTIONS . .It Fl s If .Fl s is repeated, counters with a value of zero are suppressed. .It Fl f Ar protocol_family Filter by .Ar protocol_family . See .Sx GENERAL OPTIONS . .It Fl M Use an alternative core See .Sx GENERAL OPTIONS . .It Fl N Use an alternative kernel image See .Sx GENERAL OPTIONS . .El .It Xo .Bk -words .Nm .Fl Q .Ek .Xc Show .Xr netisr 9 statistics. The flags field shows available ISR handlers: .Bl -column ".Li W" ".Dv NETISR_SNP_FLAGS_DRAINEDCPU" .It Li C Ta Dv NETISR_SNP_FLAGS_M2CPUID Ta "Able to map mbuf to cpu id" .It Li D Ta Dv NETISR_SNP_FLAGS_DRAINEDCPU Ta "Has queue drain handler" .It Li F Ta Dv NETISR_SNP_FLAGS_M2FLOW Ta "Able to map mbuf to flow id" .El .El .Pp .Ss GENERAL OPTIONS Some options have the general meaning: .Bl -tag -width flag .It Fl 4 Is shorthand for .Fl f .Ar inet .Pq Show only IPv4 .It Fl 6 Is shorthand for .Fl f .Ar inet6 .Pq Show only IPv6 .It Fl f Ar address_family , Fl p Ar protocol Limit display to those records of the specified .Ar address_family or a single .Ar protocol . The following address families and protocols are recognized: .Pp .Bl -tag -width ".Cm netgraph , ng Pq Dv AF_NETGRAPH" -compact .It Em Family .Em Protocols .It Cm inet Pq Dv AF_INET .Cm divert , icmp , igmp , ip , ipsec , pim, sctp , tcp , udp .It Cm inet6 Pq Dv AF_INET6 .Cm icmp6 , ip6 , ipsec6 , rip6 , tcp , udp .It Cm pfkey Pq Dv PF_KEY .Cm pfkey .It Cm netgraph , ng Pq Dv AF_NETGRAPH .Cm ctrl , data .It Cm unix Pq Dv AF_UNIX .It Cm link Pq Dv AF_LINK .El .Pp The program will complain if .Ar protocol is unknown or if there is no statistics routine for it. .It Fl M Extract values associated with the name list from the specified core instead of the default .Pa /dev/kmem . .It Fl N Extract the name list from the specified system instead of the default, which is the kernel image the system has booted from. .It Fl n Show network addresses and ports as numbers. Normally .Nm attempts to resolve addresses and ports, and display them symbolically. .El .Sh SEE ALSO .Xr fstat 1 , .Xr nfsstat 1 , .Xr procstat 1 , .Xr ps 1 , .Xr sockstat 1 , .Xr libxo 3 , .Xr xo_parse_args 3 , .Xr bpf 4 , .Xr inet 4 , .Xr route 4 , .Xr unix 4 , .Xr hosts 5 , .Xr networks 5 , .Xr protocols 5 , .Xr services 5 , .Xr iostat 8 , .Xr route 8 , .Xr trpt 8 , .Xr vmstat 8 , .Xr mbuf 9 .Sh HISTORY The .Nm command appeared in .Bx 4.2 . .Pp IPv6 support was added by WIDE/KAME project. .Sh BUGS The notion of errors is ill-defined.