sys/net/ip_mroute/ip_mroute.h

   1 /*
   2  * Copyright (c) 1989 Stephen Deering.
   3  * Copyright (c) 1992, 1993
   4  *      The Regents of the University of California.  All rights reserved.
   5  *
   6  * This code is derived from software contributed to Berkeley by
   7  * Stephen Deering of Stanford University.
   8  *
   9  * Redistribution and use in source and binary forms, with or without
  10  * modification, are permitted provided that the following conditions
  11  * are met:
  12  * 1. Redistributions of source code must retain the above copyright
  13  *    notice, this list of conditions and the following disclaimer.
  14  * 2. Redistributions in binary form must reproduce the above copyright
  15  *    notice, this list of conditions and the following disclaimer in the
  16  *    documentation and/or other materials provided with the distribution.
  17  * 3. All advertising materials mentioning features or use of this software
  18  *    must display the following acknowledgement:
  19  *      This product includes software developed by the University of
  20  *      California, Berkeley and its contributors.
  21  * 4. Neither the name of the University nor the names of its contributors
  22  *    may be used to endorse or promote products derived from this software
  23  *    without specific prior written permission.
  24  *
  25  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  26  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  27  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  28  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  29  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  30  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  31  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  32  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  33  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  34  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  35  * SUCH DAMAGE.
  36  *
  37  *      @(#)ip_mroute.h 8.1 (Berkeley) 6/10/93
  38  * $FreeBSD: src/sys/netinet/ip_mroute.h,v 1.17.2.3 2003/08/24 08:24:38 hsu Exp $
  39  * $DragonFly: src/sys/net/ip_mroute/ip_mroute.h,v 1.3 2003/08/24 23:07:07 hsu Exp $
  40  */
  41
  42 #ifndef _NETINET_IP_MROUTE_H_
  43 #define _NETINET_IP_MROUTE_H_
  44
  45 /*
  46  * Definitions for IP multicast forwarding.
  47  *
  48  * Written by David Waitzman, BBN Labs, August 1988.
  49  * Modified by Steve Deering, Stanford, February 1989.
  50  * Modified by Ajit Thyagarajan, PARC, August 1993.
  51  * Modified by Ajit Thyagarajan, PARC, August 1994.
  52  * Modified by Ahmed Helmy, SGI, June 1996.
  53  * Modified by Pavlin Radoslavov, ICSI, October 2002.
  54  *
  55  * MROUTING Revision: 3.3.1.3
  56  * and PIM-SMv2 and PIM-DM support, advanced API support,
  57  * bandwidth metering and signaling.
  58  */
  59
  60
  61 /*
  62  * Multicast Routing set/getsockopt commands.
  63  */
  64 #define MRT_INIT        100     /* initialize forwarder */
  65 #define MRT_DONE        101     /* shut down forwarder */
  66 #define MRT_ADD_VIF     102     /* create virtual interface */
  67 #define MRT_DEL_VIF     103     /* delete virtual interface */
  68 #define MRT_ADD_MFC     104     /* insert forwarding cache entry */
  69 #define MRT_DEL_MFC     105     /* delete forwarding cache entry */
  70 #define MRT_VERSION     106     /* get kernel version number */
  71 #define MRT_ASSERT      107     /* enable assert processing */
  72 #define MRT_PIM         MRT_ASSERT /* enable PIM processing */
  73 #define MRT_API_SUPPORT 109     /* supported MRT API */
  74 #define MRT_API_CONFIG  110     /* config MRT API */
  75 #define MRT_ADD_BW_UPCALL 111   /* create bandwidth monitor */
  76 #define MRT_DEL_BW_UPCALL 112   /* delete bandwidth monitor */
  77
  78
  79 #define GET_TIME(t)     microtime(&t)
  80
  81 /*
  82  * Types and macros for handling bitmaps with one bit per virtual interface.
  83  */
  84 #define MAXVIFS 32
  85 typedef u_long vifbitmap_t;
  86 typedef u_short vifi_t;         /* type of a vif index */
  87 #define ALL_VIFS (vifi_t)-1
  88
  89 #define VIFM_SET(n, m)          ((m) |= (1 << (n)))
  90 #define VIFM_CLR(n, m)          ((m) &= ~(1 << (n)))
  91 #define VIFM_ISSET(n, m)        ((m) & (1 << (n)))
  92 #define VIFM_CLRALL(m)          ((m) = 0x00000000)
  93 #define VIFM_COPY(mfrom, mto)   ((mto) = (mfrom))
  94 #define VIFM_SAME(m1, m2)       ((m1) == (m2))
  95
  96
  97 /*
  98  * Argument structure for MRT_ADD_VIF.
  99  * (MRT_DEL_VIF takes a single vifi_t argument.)
 100  */
 101 struct vifctl {
 102         vifi_t  vifc_vifi;              /* the index of the vif to be added */
 103         u_char  vifc_flags;             /* VIFF_ flags defined below */
 104         u_char  vifc_threshold;         /* min ttl required to forward on vif */
 105         u_int   vifc_rate_limit;        /* max rate */
 106         struct  in_addr vifc_lcl_addr;  /* local interface address */
 107         struct  in_addr vifc_rmt_addr;  /* remote address (tunnels only) */
 108 };
 109
 110 #define VIFF_TUNNEL     0x1             /* vif represents a tunnel end-point */
 111 #define VIFF_SRCRT      0x2             /* tunnel uses IP source routing */
 112 #define VIFF_REGISTER   0x4             /* used for PIM Register encap/decap */
 113
 114 /*
 115  * Argument structure for MRT_ADD_MFC and MRT_DEL_MFC
 116  * XXX if you change this, make sure to change struct mfcctl2 as well.
 117  */
 118 struct mfcctl {
 119     struct in_addr  mfcc_origin;                /* ip origin of mcasts       */
 120     struct in_addr  mfcc_mcastgrp;              /* multicast group associated*/
 121     vifi_t          mfcc_parent;                /* incoming vif              */
 122     u_char          mfcc_ttls[MAXVIFS];         /* forwarding ttls on vifs   */
 123 };
 124
 125 /*
 126  * The new argument structure for MRT_ADD_MFC and MRT_DEL_MFC overlays
 127  * and extends the old struct mfcctl.
 128  */
 129 struct mfcctl2 {
 130         /* the mfcctl fields */
 131         struct in_addr  mfcc_origin;            /* ip origin of mcasts       */
 132         struct in_addr  mfcc_mcastgrp;          /* multicast group associated*/
 133         vifi_t          mfcc_parent;            /* incoming vif              */
 134         u_char          mfcc_ttls[MAXVIFS];     /* forwarding ttls on vifs   */
 135
 136         /* extension fields */
 137         uint8_t         mfcc_flags[MAXVIFS];    /* the MRT_MFC_FLAGS_* flags */
 138         struct in_addr  mfcc_rp;                /* the RP address            */
 139 };
 140 /*
 141  * The advanced-API flags.
 142  *
 143  * The MRT_MFC_FLAGS_XXX API flags are also used as flags
 144  * for the mfcc_flags field.
 145  */
 146 #define MRT_MFC_FLAGS_DISABLE_WRONGVIF  (1 << 0) /* disable WRONGVIF signals */
 147 #define MRT_MFC_FLAGS_BORDER_VIF        (1 << 1) /* border vif               */
 148 #define MRT_MFC_RP                      (1 << 8) /* enable RP address        */
 149 #define MRT_MFC_BW_UPCALL               (1 << 9) /* enable bw upcalls        */
 150 #define MRT_MFC_FLAGS_ALL               (MRT_MFC_FLAGS_DISABLE_WRONGVIF |    \
 151                                          MRT_MFC_FLAGS_BORDER_VIF)
 152 #define MRT_API_FLAGS_ALL               (MRT_MFC_FLAGS_ALL |                 \
 153                                          MRT_MFC_RP |                        \
 154                                          MRT_MFC_BW_UPCALL)
 155
 156 /*
 157  * Structure for installing or delivering an upcall if the
 158  * measured bandwidth is above or below a threshold.
 159  *
 160  * User programs (e.g. daemons) may have a need to know when the
 161  * bandwidth used by some data flow is above or below some threshold.
 162  * This interface allows the userland to specify the threshold (in
 163  * bytes and/or packets) and the measurement interval. Flows are
 164  * all packet with the same source and destination IP address.
 165  * At the moment the code is only used for multicast destinations
 166  * but there is nothing that prevents its use for unicast.
 167  *
 168  * The measurement interval cannot be shorter than some Tmin (currently, 3s).
 169  * The threshold is set in packets and/or bytes per_interval.
 170  *
 171  * Measurement works as follows:
 172  *
 173  * For >= measurements:
 174  * The first packet marks the start of a measurement interval.
 175  * During an interval we count packets and bytes, and when we
 176  * pass the threshold we deliver an upcall and we are done.
 177  * The first packet after the end of the interval resets the
 178  * count and restarts the measurement.
 179  *
 180  * For <= measurement:
 181  * We start a timer to fire at the end of the interval, and
 182  * then for each incoming packet we count packets and bytes.
 183  * When the timer fires, we compare the value with the threshold,
 184  * schedule an upcall if we are below, and restart the measurement
 185  * (reschedule timer and zero counters).
 186  */
 187
 188 struct bw_data {
 189         struct timeval  b_time;
 190         uint64_t        b_packets;
 191         uint64_t        b_bytes;
 192 };
 193
 194 struct bw_upcall {
 195         struct in_addr  bu_src;                 /* source address            */
 196         struct in_addr  bu_dst;                 /* destination address       */
 197         uint32_t        bu_flags;               /* misc flags (see below)    */
 198 #define BW_UPCALL_UNIT_PACKETS   (1 << 0)       /* threshold (in packets)    */
 199 #define BW_UPCALL_UNIT_BYTES     (1 << 1)       /* threshold (in bytes)      */
 200 #define BW_UPCALL_GEQ            (1 << 2)       /* upcall if bw >= threshold */
 201 #define BW_UPCALL_LEQ            (1 << 3)       /* upcall if bw <= threshold */
 202 #define BW_UPCALL_DELETE_ALL     (1 << 4)       /* delete all upcalls for s,d*/
 203         struct bw_data  bu_threshold;           /* the bw threshold          */
 204         struct bw_data  bu_measured;            /* the measured bw           */
 205 };
 206
 207 /* max. number of upcalls to deliver together */
 208 #define BW_UPCALLS_MAX                          128
 209 /* min. threshold time interval for bandwidth measurement */
 210 #define BW_UPCALL_THRESHOLD_INTERVAL_MIN_SEC    3
 211 #define BW_UPCALL_THRESHOLD_INTERVAL_MIN_USEC   0
 212
 213 /*
 214  * The kernel's multicast routing statistics.
 215  */
 216 struct mrtstat {
 217     u_long      mrts_mfc_lookups;       /* # forw. cache hash table hits   */
 218     u_long      mrts_mfc_misses;        /* # forw. cache hash table misses */
 219     u_long      mrts_upcalls;           /* # calls to mrouted              */
 220     u_long      mrts_no_route;          /* no route for packet's origin    */
 221     u_long      mrts_bad_tunnel;        /* malformed tunnel options        */
 222     u_long      mrts_cant_tunnel;       /* no room for tunnel options      */
 223     u_long      mrts_wrong_if;          /* arrived on wrong interface      */
 224     u_long      mrts_upq_ovflw;         /* upcall Q overflow               */
 225     u_long      mrts_cache_cleanups;    /* # entries with no upcalls       */
 226     u_long      mrts_drop_sel;          /* pkts dropped selectively        */
 227     u_long      mrts_q_overflow;        /* pkts dropped - Q overflow       */
 228     u_long      mrts_pkt2large;         /* pkts dropped - size > BKT SIZE  */
 229     u_long      mrts_upq_sockfull;      /* upcalls dropped - socket full */
 230 };
 231
 232 /*
 233  * Argument structure used by mrouted to get src-grp pkt counts
 234  */
 235 struct sioc_sg_req {
 236     struct in_addr src;
 237     struct in_addr grp;
 238     u_long pktcnt;
 239     u_long bytecnt;
 240     u_long wrong_if;
 241 };
 242
 243 /*
 244  * Argument structure used by mrouted to get vif pkt counts
 245  */
 246 struct sioc_vif_req {
 247     vifi_t vifi;                /* vif number                           */
 248     u_long icount;              /* Input packet count on vif            */
 249     u_long ocount;              /* Output packet count on vif           */
 250     u_long ibytes;              /* Input byte count on vif              */
 251     u_long obytes;              /* Output byte count on vif             */
 252 };
 253
 254
 255 /*
 256  * The kernel's virtual-interface structure.
 257  */
 258 struct vif {
 259     u_char              v_flags;        /* VIFF_ flags defined above         */
 260     u_char              v_threshold;    /* min ttl required to forward on vif*/
 261     u_int               v_rate_limit;   /* max rate                          */
 262     struct tbf         *v_tbf;          /* token bucket structure at intf.   */
 263     struct in_addr      v_lcl_addr;     /* local interface address           */
 264     struct in_addr      v_rmt_addr;     /* remote address (tunnels only)     */
 265     struct ifnet       *v_ifp;          /* pointer to interface              */
 266     u_long              v_pkt_in;       /* # pkts in on interface            */
 267     u_long              v_pkt_out;      /* # pkts out on interface           */
 268     u_long              v_bytes_in;     /* # bytes in on interface           */
 269     u_long              v_bytes_out;    /* # bytes out on interface          */
 270     struct route        v_route;        /* cached route if this is a tunnel */
 271     u_int               v_rsvp_on;      /* RSVP listening on this vif */
 272     struct socket      *v_rsvpd;        /* RSVP daemon socket */
 273 };
 274
 275 /*
 276  * The kernel's multicast forwarding cache entry structure
 277  * (A field for the type of service (mfc_tos) is to be added
 278  * at a future point)
 279  */
 280 struct mfc {
 281         struct in_addr  mfc_origin;             /* IP origin of mcasts       */
 282         struct in_addr  mfc_mcastgrp;           /* multicast group associated*/
 283         vifi_t          mfc_parent;             /* incoming vif              */
 284         u_char          mfc_ttls[MAXVIFS];      /* forwarding ttls on vifs   */
 285         u_long          mfc_pkt_cnt;            /* pkt count for src-grp     */
 286         u_long          mfc_byte_cnt;           /* byte count for src-grp    */
 287         u_long          mfc_wrong_if;           /* wrong if for src-grp      */
 288         int             mfc_expire;             /* time to clean entry up    */
 289         struct timeval  mfc_last_assert;        /* last time I sent an assert*/
 290         struct rtdetq   *mfc_stall;             /* q of packets awaiting mfc */
 291         struct mfc      *mfc_next;              /* next mfc entry            */
 292         uint8_t         mfc_flags[MAXVIFS];     /* the MRT_MFC_FLAGS_* flags */
 293         struct in_addr  mfc_rp;                 /* the RP address            */
 294         struct bw_meter *mfc_bw_meter;          /* list of bandwidth meters  */
 295 };
 296
 297 /*
 298  * Struct used to communicate from kernel to multicast router
 299  * note the convenient similarity to an IP packet
 300  */
 301 struct igmpmsg {
 302     u_long          unused1;
 303     u_long          unused2;
 304     u_char          im_msgtype;                 /* what type of message     */
 305 #define IGMPMSG_NOCACHE         1       /* no MFC in the kernel             */
 306 #define IGMPMSG_WRONGVIF        2       /* packet came from wrong interface */
 307 #define IGMPMSG_WHOLEPKT        3       /* PIM pkt for user level encap.    */
 308 #define IGMPMSG_BW_UPCALL       4       /* BW monitoring upcall             */
 309     u_char          im_mbz;                     /* must be zero             */
 310     u_char          im_vif;                     /* vif rec'd on             */
 311     u_char          unused3;
 312     struct in_addr  im_src, im_dst;
 313 };
 314
 315 /*
 316  * Argument structure used for pkt info. while upcall is made
 317  */
 318 struct rtdetq {
 319     struct mbuf         *m;             /* A copy of the packet             */
 320     struct ifnet        *ifp;           /* Interface pkt came in on         */
 321     vifi_t              xmt_vif;        /* Saved copy of imo_multicast_vif  */
 322     struct rtdetq       *next;          /* Next in list of packets          */
 323 };
 324
 325 #define MFCTBLSIZ       256
 326 #if (MFCTBLSIZ & (MFCTBLSIZ - 1)) == 0    /* from sys:route.h */
 327 #define MFCHASHMOD(h)   ((h) & (MFCTBLSIZ - 1))
 328 #else
 329 #define MFCHASHMOD(h)   ((h) % MFCTBLSIZ)
 330 #endif
 331
 332 #define MAX_UPQ 4               /* max. no of pkts in upcall Q */
 333
 334 /*
 335  * Token Bucket filter code
 336  */
 337 #define MAX_BKT_SIZE    10000             /* 10K bytes size             */
 338 #define MAXQSIZE        10                /* max # of pkts in queue     */
 339
 340 /*
 341  * the token bucket filter at each vif
 342  */
 343 struct tbf
 344 {
 345     struct timeval tbf_last_pkt_t; /* arr. time of last pkt     */
 346     u_long tbf_n_tok;           /* no of tokens in bucket       */
 347     u_long tbf_q_len;           /* length of queue at this vif  */
 348     u_long tbf_max_q_len;       /* max. queue length            */
 349     struct mbuf *tbf_q;         /* Packet queue                 */
 350     struct mbuf *tbf_t;         /* tail-insertion pointer       */
 351 };
 352
 353 /*
 354  * Structure for measuring the bandwidth and sending an upcall if the
 355  * measured bandwidth is above or below a threshold.
 356  */
 357 struct bw_meter {
 358         struct bw_meter *bm_mfc_next;           /* next bw meter (same mfc)  */
 359         struct bw_meter *bm_time_next;          /* next bw meter (same time) */
 360         uint32_t        bm_time_hash;           /* the time hash value       */
 361         struct mfc      *bm_mfc;                /* the corresponding mfc     */
 362         uint32_t        bm_flags;               /* misc flags (see below)    */
 363 #define BW_METER_UNIT_PACKETS   (1 << 0)        /* threshold (in packets)    */
 364 #define BW_METER_UNIT_BYTES     (1 << 1)        /* threshold (in bytes)      */
 365 #define BW_METER_GEQ            (1 << 2)        /* upcall if bw >= threshold */
 366 #define BW_METER_LEQ            (1 << 3)        /* upcall if bw <= threshold */
 367 #define BW_METER_USER_FLAGS     (BW_METER_UNIT_PACKETS |                \
 368                                  BW_METER_UNIT_BYTES |                  \
 369                                  BW_METER_GEQ |                         \
 370                                  BW_METER_LEQ)
 371
 372 #define BW_METER_UPCALL_DELIVERED (1 << 24)     /* upcall was delivered      */
 373
 374         struct bw_data  bm_threshold;           /* the upcall threshold      */
 375         struct bw_data  bm_measured;            /* the measured bw           */
 376         struct timeval  bm_start_time;          /* abs. time                 */
 377 };
 378
 379 #ifdef _KERNEL
 380
 381 struct sockopt;
 382
 383 extern int      (*ip_mrouter_set)(struct socket *, struct sockopt *);
 384 extern int      (*ip_mrouter_get)(struct socket *, struct sockopt *);
 385 extern int      (*ip_mrouter_done)(void);
 386 extern int      (*mrt_ioctl)(int, caddr_t);
 387
 388 #endif /* _KERNEL */
 389
 390 #endif /* _NETINET_IP_MROUTE_H_ */