lwip_mld6.c

00001 /**
00002  * @file
00003  * Multicast listener discovery
00004  *
00005  * @defgroup mld6 MLD6
00006  * @ingroup ip6
00007  * Multicast listener discovery for IPv6. Aims to be compliant with RFC 2710.
00008  * No support for MLDv2.\n
00009  * Note: The allnodes (ff01::1, ff02::1) group is assumed be received by your 
00010  * netif since it must always be received for correct IPv6 operation (e.g. SLAAC).
00011  * Ensure the netif filters are configured accordingly!\n
00012  * The netif flags also need NETIF_FLAG_MLD6 flag set to enable MLD6 on a
00013  * netif ("netif->flags |= NETIF_FLAG_MLD6;").\n
00014  * To be called from TCPIP thread.
00015  */
00016 
00017 /*
00018  * Copyright (c) 2010 Inico Technologies Ltd.
00019  * All rights reserved.
00020  *
00021  * Redistribution and use in source and binary forms, with or without modification,
00022  * are permitted provided that the following conditions are met:
00023  *
00024  * 1. Redistributions of source code must retain the above copyright notice,
00025  *    this list of conditions and the following disclaimer.
00026  * 2. Redistributions in binary form must reproduce the above copyright notice,
00027  *    this list of conditions and the following disclaimer in the documentation
00028  *    and/or other materials provided with the distribution.
00029  * 3. The name of the author may not be used to endorse or promote products
00030  *    derived from this software without specific prior written permission.
00031  *
00032  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED
00033  * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
00034  * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT
00035  * SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
00036  * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT
00037  * OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
00038  * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
00039  * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
00040  * IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY
00041  * OF SUCH DAMAGE.
00042  *
00043  * This file is part of the lwIP TCP/IP stack.
00044  *
00045  * Author: Ivan Delamer <delamer@inicotech.com>
00046  *
00047  *
00048  * Please coordinate changes and requests with Ivan Delamer
00049  * <delamer@inicotech.com>
00050  */
00051 
00052 /* Based on igmp.c implementation of igmp v2 protocol */
00053 
00054 #include "lwip/opt.h"
00055 
00056 #if LWIP_IPV6 && LWIP_IPV6_MLD  /* don't build if not configured for use in lwipopts.h */
00057 
00058 #include "lwip/mld6.h"
00059 #include "lwip/prot/mld6.h"
00060 #include "lwip/icmp6.h"
00061 #include "lwip/ip6.h"
00062 #include "lwip/ip6_addr.h"
00063 #include "lwip/ip.h"
00064 #include "lwip/inet_chksum.h"
00065 #include "lwip/pbuf.h"
00066 #include "lwip/netif.h"
00067 #include "lwip/memp.h"
00068 #include "lwip/stats.h"
00069 
00070 #include <string.h>
00071 
00072 
00073 /*
00074  * MLD constants
00075  */
00076 #define MLD6_HL                           1
00077 #define MLD6_JOIN_DELAYING_MEMBER_TMR_MS  (500)
00078 
00079 #define MLD6_GROUP_NON_MEMBER             0
00080 #define MLD6_GROUP_DELAYING_MEMBER        1
00081 #define MLD6_GROUP_IDLE_MEMBER            2
00082 
00083 /* Forward declarations. */
00084 static struct mld_group *mld6_new_group(struct netif *ifp, const ip6_addr_t *addr);
00085 static err_t mld6_remove_group(struct netif *netif, struct mld_group *group);
00086 static void mld6_delayed_report(struct mld_group *group, u16_t maxresp);
00087 static void mld6_send(struct netif *netif, struct mld_group *group, u8_t type);
00088 
00089 
00090 /**
00091  * Stop MLD processing on interface
00092  *
00093  * @param netif network interface on which stop MLD processing
00094  */
00095 err_t
00096 mld6_stop(struct netif *netif)
00097 {
00098   struct mld_group *group = netif_mld6_data(netif);
00099 
00100   netif_set_client_data(netif, LWIP_NETIF_CLIENT_DATA_INDEX_MLD6, NULL);
00101 
00102   while (group != NULL) {
00103     struct mld_group *next = group->next; /* avoid use-after-free below */
00104 
00105     /* disable the group at the MAC level */
00106     if (netif->mld_mac_filter != NULL) {
00107       netif->mld_mac_filter(netif, &(group->group_address), NETIF_DEL_MAC_FILTER);
00108     }
00109 
00110     /* free group */
00111     memp_free(MEMP_MLD6_GROUP, group);
00112 
00113     /* move to "next" */
00114     group = next;
00115   }
00116   return ERR_OK;
00117 }
00118 
00119 /**
00120  * Report MLD memberships for this interface
00121  *
00122  * @param netif network interface on which report MLD memberships
00123  */
00124 void
00125 mld6_report_groups(struct netif *netif)
00126 {
00127   struct mld_group *group = netif_mld6_data(netif);
00128 
00129   while (group != NULL) {
00130     mld6_delayed_report(group, MLD6_JOIN_DELAYING_MEMBER_TMR_MS);
00131     group = group->next;
00132   }
00133 }
00134 
00135 /**
00136  * Search for a group that is joined on a netif
00137  *
00138  * @param ifp the network interface for which to look
00139  * @param addr the group ipv6 address to search for
00140  * @return a struct mld_group* if the group has been found,
00141  *         NULL if the group wasn't found.
00142  */
00143 struct mld_group *
00144 mld6_lookfor_group(struct netif *ifp, const ip6_addr_t *addr)
00145 {
00146   struct mld_group *group = netif_mld6_data(ifp);
00147 
00148   while (group != NULL) {
00149     if (ip6_addr_cmp(&(group->group_address), addr)) {
00150       return group;
00151     }
00152     group = group->next;
00153   }
00154 
00155   return NULL;
00156 }
00157 
00158 
00159 /**
00160  * create a new group
00161  *
00162  * @param ifp the network interface for which to create
00163  * @param addr the new group ipv6
00164  * @return a struct mld_group*,
00165  *         NULL on memory error.
00166  */
00167 static struct mld_group *
00168 mld6_new_group(struct netif *ifp, const ip6_addr_t *addr)
00169 {
00170   struct mld_group *group;
00171 
00172   group = (struct mld_group *)memp_malloc(MEMP_MLD6_GROUP);
00173   if (group != NULL) {
00174     ip6_addr_set(&(group->group_address), addr);
00175     group->timer              = 0; /* Not running */
00176     group->group_state        = MLD6_GROUP_IDLE_MEMBER;
00177     group->last_reporter_flag = 0;
00178     group->use                = 0;
00179     group->next               = netif_mld6_data(ifp);
00180 
00181     netif_set_client_data(ifp, LWIP_NETIF_CLIENT_DATA_INDEX_MLD6, group);
00182   }
00183 
00184   return group;
00185 }
00186 
00187 /**
00188  * Remove a group from the mld_group_list, but do not free it yet
00189  *
00190  * @param group the group to remove
00191  * @return ERR_OK if group was removed from the list, an err_t otherwise
00192  */
00193 static err_t
00194 mld6_remove_group(struct netif *netif, struct mld_group *group)
00195 {
00196   err_t err = ERR_OK;
00197 
00198   /* Is it the first group? */
00199   if (netif_mld6_data(netif) == group) {
00200     netif_set_client_data(netif, LWIP_NETIF_CLIENT_DATA_INDEX_MLD6, group->next);
00201   } else {
00202     /* look for group further down the list */
00203     struct mld_group *tmpGroup;
00204     for (tmpGroup = netif_mld6_data(netif); tmpGroup != NULL; tmpGroup = tmpGroup->next) {
00205       if (tmpGroup->next == group) {
00206         tmpGroup->next = group->next;
00207         break;
00208       }
00209     }
00210     /* Group not find group */
00211     if (tmpGroup == NULL) {
00212       err = ERR_ARG;
00213     }
00214   }
00215 
00216   return err;
00217 }
00218 
00219 
00220 /**
00221  * Process an input MLD message. Called by icmp6_input.
00222  *
00223  * @param p the mld packet, p->payload pointing to the icmpv6 header
00224  * @param inp the netif on which this packet was received
00225  */
00226 void
00227 mld6_input(struct pbuf *p, struct netif *inp)
00228 {
00229   struct mld_header *mld_hdr;
00230   struct mld_group *group;
00231 
00232   MLD6_STATS_INC(mld6.recv);
00233 
00234   /* Check that mld header fits in packet. */
00235   if (p->len < sizeof(struct mld_header)) {
00236     /* @todo debug message */
00237     pbuf_free(p);
00238     MLD6_STATS_INC(mld6.lenerr);
00239     MLD6_STATS_INC(mld6.drop);
00240     return;
00241   }
00242 
00243   mld_hdr = (struct mld_header *)p->payload;
00244 
00245   switch (mld_hdr->type) {
00246   case ICMP6_TYPE_MLQ: /* Multicast listener query. */
00247     /* Is it a general query? */
00248     if (ip6_addr_isallnodes_linklocal(ip6_current_dest_addr()) &&
00249         ip6_addr_isany(&(mld_hdr->multicast_address))) {
00250       MLD6_STATS_INC(mld6.rx_general);
00251       /* Report all groups, except all nodes group, and if-local groups. */
00252       group = netif_mld6_data(inp);
00253       while (group != NULL) {
00254         if ((!(ip6_addr_ismulticast_iflocal(&(group->group_address)))) &&
00255             (!(ip6_addr_isallnodes_linklocal(&(group->group_address))))) {
00256           mld6_delayed_report(group, mld_hdr->max_resp_delay);
00257         }
00258         group = group->next;
00259       }
00260     } else {
00261       /* Have we joined this group?
00262        * We use IP6 destination address to have a memory aligned copy.
00263        * mld_hdr->multicast_address should be the same. */
00264       MLD6_STATS_INC(mld6.rx_group);
00265       group = mld6_lookfor_group(inp, ip6_current_dest_addr());
00266       if (group != NULL) {
00267         /* Schedule a report. */
00268         mld6_delayed_report(group, mld_hdr->max_resp_delay);
00269       }
00270     }
00271     break; /* ICMP6_TYPE_MLQ */
00272   case ICMP6_TYPE_MLR: /* Multicast listener report. */
00273     /* Have we joined this group?
00274      * We use IP6 destination address to have a memory aligned copy.
00275      * mld_hdr->multicast_address should be the same. */
00276     MLD6_STATS_INC(mld6.rx_report);
00277     group = mld6_lookfor_group(inp, ip6_current_dest_addr());
00278     if (group != NULL) {
00279       /* If we are waiting to report, cancel it. */
00280       if (group->group_state == MLD6_GROUP_DELAYING_MEMBER) {
00281         group->timer = 0; /* stopped */
00282         group->group_state = MLD6_GROUP_IDLE_MEMBER;
00283         group->last_reporter_flag = 0;
00284       }
00285     }
00286     break; /* ICMP6_TYPE_MLR */
00287   case ICMP6_TYPE_MLD: /* Multicast listener done. */
00288     /* Do nothing, router will query us. */
00289     break; /* ICMP6_TYPE_MLD */
00290   default:
00291     MLD6_STATS_INC(mld6.proterr);
00292     MLD6_STATS_INC(mld6.drop);
00293     break;
00294   }
00295 
00296   pbuf_free(p);
00297 }
00298 
00299 /**
00300  * @ingroup mld6
00301  * Join a group on one or all network interfaces.
00302  *
00303  * If the group is to be joined on all interfaces, the given group address must
00304  * not have a zone set (i.e., it must have its zone index set to IP6_NO_ZONE).
00305  * If the group is to be joined on one particular interface, the given group
00306  * address may or may not have a zone set.
00307  *
00308  * @param srcaddr ipv6 address (zoned) of the network interface which should
00309  *                join a new group. If IP6_ADDR_ANY6, join on all netifs
00310  * @param groupaddr the ipv6 address of the group to join (possibly but not
00311  *                  necessarily zoned)
00312  * @return ERR_OK if group was joined on the netif(s), an err_t otherwise
00313  */
00314 err_t
00315 mld6_joingroup(const ip6_addr_t *srcaddr, const ip6_addr_t *groupaddr)
00316 {
00317   err_t         err = ERR_VAL; /* no matching interface */
00318   struct netif *netif;
00319 
00320   LWIP_ASSERT_CORE_LOCKED();
00321 
00322   /* loop through netif's */
00323   NETIF_FOREACH(netif) {
00324     /* Should we join this interface ? */
00325     if (ip6_addr_isany(srcaddr) ||
00326         netif_get_ip6_addr_match(netif, srcaddr) >= 0) {
00327       err = mld6_joingroup_netif(netif, groupaddr);
00328       if (err != ERR_OK) {
00329         return err;
00330       }
00331     }
00332   }
00333 
00334   return err;
00335 }
00336 
00337 /**
00338  * @ingroup mld6
00339  * Join a group on a network interface.
00340  *
00341  * @param netif the network interface which should join a new group.
00342  * @param groupaddr the ipv6 address of the group to join (possibly but not
00343  *                  necessarily zoned)
00344  * @return ERR_OK if group was joined on the netif, an err_t otherwise
00345  */
00346 err_t
00347 mld6_joingroup_netif(struct netif *netif, const ip6_addr_t *groupaddr)
00348 {
00349   struct mld_group *group;
00350 #if LWIP_IPV6_SCOPES
00351   ip6_addr_t ip6addr;
00352 
00353   /* If the address has a particular scope but no zone set, use the netif to
00354    * set one now. Within the mld6 module, all addresses are properly zoned. */
00355   if (ip6_addr_lacks_zone(groupaddr, IP6_MULTICAST)) {
00356     ip6_addr_set(&ip6addr, groupaddr);
00357     ip6_addr_assign_zone(&ip6addr, IP6_MULTICAST, netif);
00358     groupaddr = &ip6addr;
00359   }
00360   IP6_ADDR_ZONECHECK_NETIF(groupaddr, netif);
00361 #endif /* LWIP_IPV6_SCOPES */
00362 
00363   LWIP_ASSERT_CORE_LOCKED();
00364 
00365   /* find group or create a new one if not found */
00366   group = mld6_lookfor_group(netif, groupaddr);
00367 
00368   if (group == NULL) {
00369     /* Joining a new group. Create a new group entry. */
00370     group = mld6_new_group(netif, groupaddr);
00371     if (group == NULL) {
00372       return ERR_MEM;
00373     }
00374 
00375     /* Activate this address on the MAC layer. */
00376     if (netif->mld_mac_filter != NULL) {
00377       netif->mld_mac_filter(netif, groupaddr, NETIF_ADD_MAC_FILTER);
00378     }
00379 
00380     /* Report our membership. */
00381     MLD6_STATS_INC(mld6.tx_report);
00382     mld6_send(netif, group, ICMP6_TYPE_MLR);
00383     mld6_delayed_report(group, MLD6_JOIN_DELAYING_MEMBER_TMR_MS);
00384   }
00385 
00386   /* Increment group use */
00387   group->use++;
00388   return ERR_OK;
00389 }
00390 
00391 /**
00392  * @ingroup mld6
00393  * Leave a group on a network interface.
00394  *
00395  * Zoning of address follows the same rules as @ref mld6_joingroup.
00396  *
00397  * @param srcaddr ipv6 address (zoned) of the network interface which should
00398  *                leave the group. If IP6_ADDR_ANY6, leave on all netifs
00399  * @param groupaddr the ipv6 address of the group to leave (possibly, but not
00400  *                  necessarily zoned)
00401  * @return ERR_OK if group was left on the netif(s), an err_t otherwise
00402  */
00403 err_t
00404 mld6_leavegroup(const ip6_addr_t *srcaddr, const ip6_addr_t *groupaddr)
00405 {
00406   err_t         err = ERR_VAL; /* no matching interface */
00407   struct netif *netif;
00408 
00409   LWIP_ASSERT_CORE_LOCKED();
00410 
00411   /* loop through netif's */
00412   NETIF_FOREACH(netif) {
00413     /* Should we leave this interface ? */
00414     if (ip6_addr_isany(srcaddr) ||
00415         netif_get_ip6_addr_match(netif, srcaddr) >= 0) {
00416       err_t res = mld6_leavegroup_netif(netif, groupaddr);
00417       if (err != ERR_OK) {
00418         /* Store this result if we have not yet gotten a success */
00419         err = res;
00420       }
00421     }
00422   }
00423 
00424   return err;
00425 }
00426 
00427 /**
00428  * @ingroup mld6
00429  * Leave a group on a network interface.
00430  *
00431  * @param netif the network interface which should leave the group.
00432  * @param groupaddr the ipv6 address of the group to leave (possibly, but not
00433  *                  necessarily zoned)
00434  * @return ERR_OK if group was left on the netif, an err_t otherwise
00435  */
00436 err_t
00437 mld6_leavegroup_netif(struct netif *netif, const ip6_addr_t *groupaddr)
00438 {
00439   struct mld_group *group;
00440 #if LWIP_IPV6_SCOPES
00441   ip6_addr_t ip6addr;
00442 
00443   if (ip6_addr_lacks_zone(groupaddr, IP6_MULTICAST)) {
00444     ip6_addr_set(&ip6addr, groupaddr);
00445     ip6_addr_assign_zone(&ip6addr, IP6_MULTICAST, netif);
00446     groupaddr = &ip6addr;
00447   }
00448   IP6_ADDR_ZONECHECK_NETIF(groupaddr, netif);
00449 #endif /* LWIP_IPV6_SCOPES */
00450 
00451   LWIP_ASSERT_CORE_LOCKED();
00452 
00453   /* find group */
00454   group = mld6_lookfor_group(netif, groupaddr);
00455 
00456   if (group != NULL) {
00457     /* Leave if there is no other use of the group */
00458     if (group->use <= 1) {
00459       /* Remove the group from the list */
00460       mld6_remove_group(netif, group);
00461 
00462       /* If we are the last reporter for this group */
00463       if (group->last_reporter_flag) {
00464         MLD6_STATS_INC(mld6.tx_leave);
00465         mld6_send(netif, group, ICMP6_TYPE_MLD);
00466       }
00467 
00468       /* Disable the group at the MAC level */
00469       if (netif->mld_mac_filter != NULL) {
00470         netif->mld_mac_filter(netif, groupaddr, NETIF_DEL_MAC_FILTER);
00471       }
00472 
00473       /* free group struct */
00474       memp_free(MEMP_MLD6_GROUP, group);
00475     } else {
00476       /* Decrement group use */
00477       group->use--;
00478     }
00479 
00480     /* Left group */
00481     return ERR_OK;
00482   }
00483 
00484   /* Group not found */
00485   return ERR_VAL;
00486 }
00487 
00488 
00489 /**
00490  * Periodic timer for mld processing. Must be called every
00491  * MLD6_TMR_INTERVAL milliseconds (100).
00492  *
00493  * When a delaying member expires, a membership report is sent.
00494  */
00495 void
00496 mld6_tmr(void)
00497 {
00498   struct netif *netif;
00499 
00500   NETIF_FOREACH(netif) {
00501     struct mld_group *group = netif_mld6_data(netif);
00502 
00503     while (group != NULL) {
00504       if (group->timer > 0) {
00505         group->timer--;
00506         if (group->timer == 0) {
00507           /* If the state is MLD6_GROUP_DELAYING_MEMBER then we send a report for this group */
00508           if (group->group_state == MLD6_GROUP_DELAYING_MEMBER) {
00509             MLD6_STATS_INC(mld6.tx_report);
00510             mld6_send(netif, group, ICMP6_TYPE_MLR);
00511             group->group_state = MLD6_GROUP_IDLE_MEMBER;
00512           }
00513         }
00514       }
00515       group = group->next;
00516     }
00517   }
00518 }
00519 
00520 /**
00521  * Schedule a delayed membership report for a group
00522  *
00523  * @param group the mld_group for which "delaying" membership report
00524  *              should be sent
00525  * @param maxresp_in the max resp delay provided in the query
00526  */
00527 static void
00528 mld6_delayed_report(struct mld_group *group, u16_t maxresp_in)
00529 {
00530   /* Convert maxresp from milliseconds to tmr ticks */
00531   u16_t maxresp = maxresp_in / MLD6_TMR_INTERVAL;
00532   if (maxresp == 0) {
00533     maxresp = 1;
00534   }
00535 
00536 #ifdef LWIP_RAND
00537   /* Randomize maxresp. (if LWIP_RAND is supported) */
00538   maxresp = (u16_t)(LWIP_RAND() % maxresp);
00539   if (maxresp == 0) {
00540     maxresp = 1;
00541   }
00542 #endif /* LWIP_RAND */
00543 
00544   /* Apply timer value if no report has been scheduled already. */
00545   if ((group->group_state == MLD6_GROUP_IDLE_MEMBER) ||
00546      ((group->group_state == MLD6_GROUP_DELAYING_MEMBER) &&
00547       ((group->timer == 0) || (maxresp < group->timer)))) {
00548     group->timer = maxresp;
00549     group->group_state = MLD6_GROUP_DELAYING_MEMBER;
00550   }
00551 }
00552 
00553 /**
00554  * Send a MLD message (report or done).
00555  *
00556  * An IPv6 hop-by-hop options header with a router alert option
00557  * is prepended.
00558  *
00559  * @param group the group to report or quit
00560  * @param type ICMP6_TYPE_MLR (report) or ICMP6_TYPE_MLD (done)
00561  */
00562 static void
00563 mld6_send(struct netif *netif, struct mld_group *group, u8_t type)
00564 {
00565   struct mld_header *mld_hdr;
00566   struct pbuf *p;
00567   const ip6_addr_t *src_addr;
00568 
00569   /* Allocate a packet. Size is MLD header + IPv6 Hop-by-hop options header. */
00570   p = pbuf_alloc(PBUF_IP, sizeof(struct mld_header) + MLD6_HBH_HLEN, PBUF_RAM);
00571   if (p == NULL) {
00572     MLD6_STATS_INC(mld6.memerr);
00573     return;
00574   }
00575 
00576   /* Move to make room for Hop-by-hop options header. */
00577   if (pbuf_remove_header(p, MLD6_HBH_HLEN)) {
00578     pbuf_free(p);
00579     MLD6_STATS_INC(mld6.lenerr);
00580     return;
00581   }
00582 
00583   /* Select our source address. */
00584   if (!ip6_addr_isvalid(netif_ip6_addr_state(netif, 0))) {
00585     /* This is a special case, when we are performing duplicate address detection.
00586      * We must join the multicast group, but we don't have a valid address yet. */
00587     src_addr = IP6_ADDR_ANY6;
00588   } else {
00589     /* Use link-local address as source address. */
00590     src_addr = netif_ip6_addr(netif, 0);
00591   }
00592 
00593   /* MLD message header pointer. */
00594   mld_hdr = (struct mld_header *)p->payload;
00595 
00596   /* Set fields. */
00597   mld_hdr->type = type;
00598   mld_hdr->code = 0;
00599   mld_hdr->chksum = 0;
00600   mld_hdr->max_resp_delay = 0;
00601   mld_hdr->reserved = 0;
00602   ip6_addr_copy_to_packed(mld_hdr->multicast_address, group->group_address);
00603 
00604 #if CHECKSUM_GEN_ICMP6
00605   IF__NETIF_CHECKSUM_ENABLED(netif, NETIF_CHECKSUM_GEN_ICMP6) {
00606     mld_hdr->chksum = ip6_chksum_pseudo(p, IP6_NEXTH_ICMP6, p->len,
00607       src_addr, &(group->group_address));
00608   }
00609 #endif /* CHECKSUM_GEN_ICMP6 */
00610 
00611   /* Add hop-by-hop headers options: router alert with MLD value. */
00612   ip6_options_add_hbh_ra(p, IP6_NEXTH_ICMP6, IP6_ROUTER_ALERT_VALUE_MLD);
00613 
00614   if (type == ICMP6_TYPE_MLR) {
00615     /* Remember we were the last to report */
00616     group->last_reporter_flag = 1;
00617   }
00618 
00619   /* Send the packet out. */
00620   MLD6_STATS_INC(mld6.xmit);
00621   ip6_output_if(p, (ip6_addr_isany(src_addr)) ? NULL : src_addr, &(group->group_address),
00622       MLD6_HL, 0, IP6_NEXTH_HOPBYHOP, netif);
00623   pbuf_free(p);
00624 }
00625 
00626 #endif /* LWIP_IPV6 */