Hi everyone,
I am working on a program that requires putting the node in
all-multicast mode. The program needs to see all the packets going to
solicited-node multicast addresses. I did this by modifying the
interface flags to add IFF_ALLMULTI.
Unfortunately recvmsg() was only returning for the packets destined
for multicast groups I had explicitly joined and not all of them.
After this change, I can see the packets as expected. A similar change
will probably be required for the ip4 variant of this function.
I also noticed that for this to work, IFF_PROMISC needs to be set as well.
Shouldn't it be possible to have IFF_MULTI without IFF_PROMISC? I am using
the em(4) driver and I can see the following code in there:
ifp->if_flags &= ~IFF_ALLMULTI;
if (ifp->if_flags & IFF_PROMISC || ac->ac_multirangecnt > 0 ||
ac->ac_multicnt > MAX_NUM_MULTICAST_ADDRESSES) {
ifp->if_flags |= IFF_ALLMULTI;
reg_rctl |= E1000_RCTL_MPE;
if (ifp->if_flags & IFF_PROMISC)
reg_rctl |= E1000_RCTL_UPE;
} else {
ETHER_FIRST_MULTI(step, ac, enm);
while (enm != NULL) {
bcopy(enm->enm_addrlo, mta + i, ETH_LENGTH_OF_ADDRESS);
i += ETH_LENGTH_OF_ADDRESS;
ETHER_NEXT_MULTI(step, enm);
}
em_mc_addr_list_update(&sc->hw, mta, ac->ac_multicnt,
0, 1);
}
I am not sure how to control ac->ac_multirangecnt from userspace or if
this behaviour is intended in the first place.
diff --git a/sys/netinet6/in6.c b/sys/netinet6/in6.c
index 8b9ff6d..5630fb9 100644
--- a/sys/netinet6/in6.c
+++ b/sys/netinet6/in6.c
@@ -1342,6 +1342,9 @@ in6_hasmulti(struct in6_addr *maddr6, struct ifnet *ifp)
struct in6_multi *in6m;
int joined;
+ if ((ifp->if_flags & IFF_ALLMULTI) != 0)
+ return 1;
+
IN6_LOOKUP_MULTI(*maddr6, ifp, in6m);
joined = (in6m != NULL);
