Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
/*-
|
|
|
|
|
* Copyright (c) 2007 Bruce M. Simpson.
|
|
|
|
|
* Copyright (c) 2005 Robert N. M. Watson.
|
|
|
|
|
* 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. The name of the author may not be used to endorse or promote
|
|
|
|
|
* products derived from this software without specific prior written
|
|
|
|
|
* permission.
|
|
|
|
|
*
|
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR 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.
|
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* IPv4 multicast socket, group, and socket option processing module.
|
|
|
|
|
* Until further notice, this file requires INET to compile.
|
|
|
|
|
* TODO: Make this infrastructure independent of address family.
|
|
|
|
|
* TODO: Teach netinet6 to use this code.
|
|
|
|
|
* TODO: Hook up SSM logic to IGMPv3/MLDv2.
|
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
#include <sys/cdefs.h>
|
|
|
|
|
__FBSDID("$FreeBSD$");
|
|
|
|
|
|
|
|
|
|
#include <sys/param.h>
|
|
|
|
|
#include <sys/systm.h>
|
|
|
|
|
#include <sys/kernel.h>
|
|
|
|
|
#include <sys/malloc.h>
|
|
|
|
|
#include <sys/mbuf.h>
|
Over the past couple of years, there have been a number of reports relating
the use of divert sockets to dead locks. A number of LORs have been reported
between divert and a number of other network subsystems including: IPSEC, Pfil,
multicast, ipfw and others. Other dead locks could occur because of recursive
entry into the IP stack. This change should take care of most if not all of
these issues.
A summary of the changes follow:
- We disallow multicast operations on divert sockets. It really doesn't make
semantic sense to allow this, since typically you would set multicast
parameters on multicast end points.
NOTE: As a part of this change, we actually dis-allow multicast options on
any socket that IS a divert socket OR IS NOT a SOCK_RAW or SOCK_DGRAM family
- We check to see if there are any socket options that have been specified on
the socket, and if there was (which is very un-common and also probably
doesnt make sense to support) we duplicate the mbuf carrying the options.
- We then drop the INP/INFO locks over the call to ip_output(). It should be
noted that since we no longer support multicast operations on divert sockets
and we have duplicated any socket options, we no longer need the reference
to the pcb to be coherent.
- Finally, we replaced the call to ip_input() to use netisr queuing. This
should remove the recursive entry into the IP stack from divert.
By dropping the locks over the call to ip_output() we eliminate all the lock
ordering issues above. By switching over to netisr on the inbound path,
we can no longer recursively enter the ip_input() code via divert.
I have tested this change by using the following command:
ipfwpcap -r 8000 - | tcpdump -r - -nn -v
This should exercise the input and re-injection (outbound) path, which is
very similar to the work load performed by natd(8). Additionally, I have
run some ospf daemons which have a heavy reliance on raw sockets and
multicast.
Approved by: re@ (kensmith)
MFC after: 1 month
LOR: 163
LOR: 181
LOR: 202
LOR: 203
Discussed with: julian, andre et al (on freebsd-net)
In collaboration with: bms [1], rwatson [2]
[1] bms helped out with the multicast decisions
[2] rwatson submitted the original netisr patches and came up with some
of the original ideas on how to combat this issue.
2007-08-06 22:06:36 +00:00
|
|
|
#include <sys/protosw.h>
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
#include <sys/socket.h>
|
|
|
|
|
#include <sys/socketvar.h>
|
|
|
|
|
#include <sys/sysctl.h>
|
Commit step 1 of the vimage project, (network stack)
virtualization work done by Marko Zec (zec@).
This is the first in a series of commits over the course
of the next few weeks.
Mark all uses of global variables to be virtualized
with a V_ prefix.
Use macros to map them back to their global names for
now, so this is a NOP change only.
We hope to have caught at least 85-90% of what is needed
so we do not invalidate a lot of outstanding patches again.
Obtained from: //depot/projects/vimage-commit2/...
Reviewed by: brooks, des, ed, mav, julian,
jamie, kris, rwatson, zec, ...
(various people I forgot, different versions)
md5 (with a bit of help)
Sponsored by: NLnet Foundation, The FreeBSD Foundation
X-MFC after: never
V_Commit_Message_Reviewed_By: more people than the patch
2008-08-17 23:27:27 +00:00
|
|
|
#include <sys/vimage.h>
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
|
|
|
|
|
#include <net/if.h>
|
|
|
|
|
#include <net/if_dl.h>
|
|
|
|
|
#include <net/route.h>
|
2008-12-02 21:37:28 +00:00
|
|
|
#include <net/vnet.h>
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
|
|
|
|
|
#include <netinet/in.h>
|
|
|
|
|
#include <netinet/in_systm.h>
|
|
|
|
|
#include <netinet/in_pcb.h>
|
|
|
|
|
#include <netinet/in_var.h>
|
|
|
|
|
#include <netinet/ip_var.h>
|
|
|
|
|
#include <netinet/igmp_var.h>
|
2008-12-02 21:37:28 +00:00
|
|
|
#include <netinet/vinet.h>
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
|
|
|
|
|
#ifndef __SOCKUNION_DECLARED
|
|
|
|
|
union sockunion {
|
|
|
|
|
struct sockaddr_storage ss;
|
|
|
|
|
struct sockaddr sa;
|
|
|
|
|
struct sockaddr_dl sdl;
|
|
|
|
|
struct sockaddr_in sin;
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
struct sockaddr_in6 sin6;
|
|
|
|
|
#endif
|
|
|
|
|
};
|
|
|
|
|
typedef union sockunion sockunion_t;
|
|
|
|
|
#define __SOCKUNION_DECLARED
|
|
|
|
|
#endif /* __SOCKUNION_DECLARED */
|
|
|
|
|
|
|
|
|
|
static MALLOC_DEFINE(M_IPMADDR, "in_multi", "IPv4 multicast group");
|
|
|
|
|
static MALLOC_DEFINE(M_IPMOPTS, "ip_moptions", "IPv4 multicast options");
|
|
|
|
|
static MALLOC_DEFINE(M_IPMSOURCE, "in_msource", "IPv4 multicast source filter");
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* The IPv4 multicast list (in_multihead and associated structures) are
|
|
|
|
|
* protected by the global in_multi_mtx. See in_var.h for more details. For
|
|
|
|
|
* now, in_multi_mtx is marked as recursible due to IGMP's calling back into
|
|
|
|
|
* ip_output() to send IGMP packets while holding the lock; this probably is
|
|
|
|
|
* not quite desirable.
|
|
|
|
|
*/
|
2008-11-19 09:39:34 +00:00
|
|
|
#ifdef VIMAGE_GLOBALS
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
struct in_multihead in_multihead; /* XXX BSS initialization */
|
2008-11-19 09:39:34 +00:00
|
|
|
#endif
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
struct mtx in_multi_mtx;
|
|
|
|
|
MTX_SYSINIT(in_multi_mtx, &in_multi_mtx, "in_multi_mtx", MTX_DEF | MTX_RECURSE);
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Functions with non-static linkage defined in this file should be
|
|
|
|
|
* declared in in_var.h:
|
|
|
|
|
* imo_match_group()
|
|
|
|
|
* imo_match_source()
|
|
|
|
|
* in_addmulti()
|
|
|
|
|
* in_delmulti()
|
|
|
|
|
* in_delmulti_locked()
|
|
|
|
|
* and ip_var.h:
|
|
|
|
|
* inp_freemoptions()
|
|
|
|
|
* inp_getmoptions()
|
|
|
|
|
* inp_setmoptions()
|
|
|
|
|
*/
|
|
|
|
|
static int imo_grow(struct ip_moptions *);
|
|
|
|
|
static int imo_join_source(struct ip_moptions *, size_t, sockunion_t *);
|
|
|
|
|
static int imo_leave_source(struct ip_moptions *, size_t, sockunion_t *);
|
|
|
|
|
static int inp_change_source_filter(struct inpcb *, struct sockopt *);
|
|
|
|
|
static struct ip_moptions *
|
|
|
|
|
inp_findmoptions(struct inpcb *);
|
|
|
|
|
static int inp_get_source_filters(struct inpcb *, struct sockopt *);
|
|
|
|
|
static int inp_join_group(struct inpcb *, struct sockopt *);
|
|
|
|
|
static int inp_leave_group(struct inpcb *, struct sockopt *);
|
|
|
|
|
static int inp_set_multicast_if(struct inpcb *, struct sockopt *);
|
|
|
|
|
static int inp_set_source_filters(struct inpcb *, struct sockopt *);
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Resize the ip_moptions vector to the next power-of-two minus 1.
|
|
|
|
|
* May be called with locks held; do not sleep.
|
|
|
|
|
*/
|
|
|
|
|
static int
|
|
|
|
|
imo_grow(struct ip_moptions *imo)
|
|
|
|
|
{
|
|
|
|
|
struct in_multi **nmships;
|
|
|
|
|
struct in_multi **omships;
|
|
|
|
|
struct in_mfilter *nmfilters;
|
|
|
|
|
struct in_mfilter *omfilters;
|
|
|
|
|
size_t idx;
|
|
|
|
|
size_t newmax;
|
|
|
|
|
size_t oldmax;
|
|
|
|
|
|
|
|
|
|
nmships = NULL;
|
|
|
|
|
nmfilters = NULL;
|
|
|
|
|
omships = imo->imo_membership;
|
|
|
|
|
omfilters = imo->imo_mfilters;
|
|
|
|
|
oldmax = imo->imo_max_memberships;
|
|
|
|
|
newmax = ((oldmax + 1) * 2) - 1;
|
|
|
|
|
|
|
|
|
|
if (newmax <= IP_MAX_MEMBERSHIPS) {
|
|
|
|
|
nmships = (struct in_multi **)realloc(omships,
|
|
|
|
|
sizeof(struct in_multi *) * newmax, M_IPMOPTS, M_NOWAIT);
|
|
|
|
|
nmfilters = (struct in_mfilter *)realloc(omfilters,
|
|
|
|
|
sizeof(struct in_mfilter) * newmax, M_IPMSOURCE, M_NOWAIT);
|
|
|
|
|
if (nmships != NULL && nmfilters != NULL) {
|
|
|
|
|
/* Initialize newly allocated source filter heads. */
|
|
|
|
|
for (idx = oldmax; idx < newmax; idx++) {
|
|
|
|
|
nmfilters[idx].imf_fmode = MCAST_EXCLUDE;
|
|
|
|
|
nmfilters[idx].imf_nsources = 0;
|
|
|
|
|
TAILQ_INIT(&nmfilters[idx].imf_sources);
|
|
|
|
|
}
|
|
|
|
|
imo->imo_max_memberships = newmax;
|
|
|
|
|
imo->imo_membership = nmships;
|
|
|
|
|
imo->imo_mfilters = nmfilters;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (nmships == NULL || nmfilters == NULL) {
|
|
|
|
|
if (nmships != NULL)
|
|
|
|
|
free(nmships, M_IPMOPTS);
|
|
|
|
|
if (nmfilters != NULL)
|
|
|
|
|
free(nmfilters, M_IPMSOURCE);
|
|
|
|
|
return (ETOOMANYREFS);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Add a source to a multicast filter list.
|
|
|
|
|
* Assumes the associated inpcb is locked.
|
|
|
|
|
*/
|
|
|
|
|
static int
|
|
|
|
|
imo_join_source(struct ip_moptions *imo, size_t gidx, sockunion_t *src)
|
|
|
|
|
{
|
|
|
|
|
struct in_msource *ims, *nims;
|
|
|
|
|
struct in_mfilter *imf;
|
|
|
|
|
|
|
|
|
|
KASSERT(src->ss.ss_family == AF_INET, ("%s: !AF_INET", __func__));
|
|
|
|
|
KASSERT(imo->imo_mfilters != NULL,
|
|
|
|
|
("%s: imo_mfilters vector not allocated", __func__));
|
|
|
|
|
|
|
|
|
|
imf = &imo->imo_mfilters[gidx];
|
|
|
|
|
if (imf->imf_nsources == IP_MAX_SOURCE_FILTER)
|
|
|
|
|
return (ENOBUFS);
|
|
|
|
|
|
|
|
|
|
ims = imo_match_source(imo, gidx, &src->sa);
|
|
|
|
|
if (ims != NULL)
|
|
|
|
|
return (EADDRNOTAVAIL);
|
|
|
|
|
|
|
|
|
|
/* Do not sleep with inp lock held. */
|
2008-10-23 15:53:51 +00:00
|
|
|
nims = malloc(sizeof(struct in_msource),
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
M_IPMSOURCE, M_NOWAIT | M_ZERO);
|
|
|
|
|
if (nims == NULL)
|
|
|
|
|
return (ENOBUFS);
|
|
|
|
|
|
|
|
|
|
nims->ims_addr = src->ss;
|
|
|
|
|
TAILQ_INSERT_TAIL(&imf->imf_sources, nims, ims_next);
|
|
|
|
|
imf->imf_nsources++;
|
|
|
|
|
|
|
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int
|
|
|
|
|
imo_leave_source(struct ip_moptions *imo, size_t gidx, sockunion_t *src)
|
|
|
|
|
{
|
|
|
|
|
struct in_msource *ims;
|
|
|
|
|
struct in_mfilter *imf;
|
|
|
|
|
|
|
|
|
|
KASSERT(src->ss.ss_family == AF_INET, ("%s: !AF_INET", __func__));
|
|
|
|
|
KASSERT(imo->imo_mfilters != NULL,
|
|
|
|
|
("%s: imo_mfilters vector not allocated", __func__));
|
|
|
|
|
|
|
|
|
|
imf = &imo->imo_mfilters[gidx];
|
|
|
|
|
if (imf->imf_nsources == IP_MAX_SOURCE_FILTER)
|
|
|
|
|
return (ENOBUFS);
|
|
|
|
|
|
|
|
|
|
ims = imo_match_source(imo, gidx, &src->sa);
|
|
|
|
|
if (ims == NULL)
|
|
|
|
|
return (EADDRNOTAVAIL);
|
|
|
|
|
|
|
|
|
|
TAILQ_REMOVE(&imf->imf_sources, ims, ims_next);
|
2008-10-23 15:53:51 +00:00
|
|
|
free(ims, M_IPMSOURCE);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
imf->imf_nsources--;
|
|
|
|
|
|
|
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Find an IPv4 multicast group entry for this ip_moptions instance
|
|
|
|
|
* which matches the specified group, and optionally an interface.
|
|
|
|
|
* Return its index into the array, or -1 if not found.
|
|
|
|
|
*/
|
|
|
|
|
size_t
|
|
|
|
|
imo_match_group(struct ip_moptions *imo, struct ifnet *ifp,
|
|
|
|
|
struct sockaddr *group)
|
|
|
|
|
{
|
|
|
|
|
sockunion_t *gsa;
|
|
|
|
|
struct in_multi **pinm;
|
|
|
|
|
int idx;
|
|
|
|
|
int nmships;
|
|
|
|
|
|
|
|
|
|
gsa = (sockunion_t *)group;
|
|
|
|
|
|
|
|
|
|
/* The imo_membership array may be lazy allocated. */
|
|
|
|
|
if (imo->imo_membership == NULL || imo->imo_num_memberships == 0)
|
|
|
|
|
return (-1);
|
|
|
|
|
|
|
|
|
|
nmships = imo->imo_num_memberships;
|
|
|
|
|
pinm = &imo->imo_membership[0];
|
|
|
|
|
for (idx = 0; idx < nmships; idx++, pinm++) {
|
|
|
|
|
if (*pinm == NULL)
|
|
|
|
|
continue;
|
|
|
|
|
#if 0
|
|
|
|
|
printf("%s: trying ifp = %p, inaddr = %s ", __func__,
|
|
|
|
|
ifp, inet_ntoa(gsa->sin.sin_addr));
|
|
|
|
|
printf("against %p, %s\n",
|
|
|
|
|
(*pinm)->inm_ifp, inet_ntoa((*pinm)->inm_addr));
|
|
|
|
|
#endif
|
|
|
|
|
if ((ifp == NULL || ((*pinm)->inm_ifp == ifp)) &&
|
|
|
|
|
(*pinm)->inm_addr.s_addr == gsa->sin.sin_addr.s_addr) {
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
if (idx >= nmships)
|
|
|
|
|
idx = -1;
|
|
|
|
|
|
|
|
|
|
return (idx);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Find a multicast source entry for this imo which matches
|
|
|
|
|
* the given group index for this socket, and source address.
|
|
|
|
|
*/
|
|
|
|
|
struct in_msource *
|
|
|
|
|
imo_match_source(struct ip_moptions *imo, size_t gidx, struct sockaddr *src)
|
|
|
|
|
{
|
|
|
|
|
struct in_mfilter *imf;
|
|
|
|
|
struct in_msource *ims, *pims;
|
|
|
|
|
|
|
|
|
|
KASSERT(src->sa_family == AF_INET, ("%s: !AF_INET", __func__));
|
|
|
|
|
KASSERT(gidx != -1 && gidx < imo->imo_num_memberships,
|
|
|
|
|
("%s: invalid index %d\n", __func__, (int)gidx));
|
|
|
|
|
|
|
|
|
|
/* The imo_mfilters array may be lazy allocated. */
|
|
|
|
|
if (imo->imo_mfilters == NULL)
|
|
|
|
|
return (NULL);
|
|
|
|
|
|
|
|
|
|
pims = NULL;
|
|
|
|
|
imf = &imo->imo_mfilters[gidx];
|
|
|
|
|
TAILQ_FOREACH(ims, &imf->imf_sources, ims_next) {
|
|
|
|
|
/*
|
|
|
|
|
* Perform bitwise comparison of two IPv4 addresses.
|
|
|
|
|
* TODO: Do the same for IPv6.
|
|
|
|
|
* Do not use sa_equal() for this as it is not aware of
|
|
|
|
|
* deeper structure in sockaddr_in or sockaddr_in6.
|
|
|
|
|
*/
|
|
|
|
|
if (((struct sockaddr_in *)&ims->ims_addr)->sin_addr.s_addr ==
|
|
|
|
|
((struct sockaddr_in *)src)->sin_addr.s_addr) {
|
|
|
|
|
pims = ims;
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return (pims);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Join an IPv4 multicast group.
|
|
|
|
|
*/
|
|
|
|
|
struct in_multi *
|
|
|
|
|
in_addmulti(struct in_addr *ap, struct ifnet *ifp)
|
|
|
|
|
{
|
Step 1.5 of importing the network stack virtualization infrastructure
from the vimage project, as per plan established at devsummit 08/08:
http://wiki.freebsd.org/Image/Notes200808DevSummit
Introduce INIT_VNET_*() initializer macros, VNET_FOREACH() iterator
macros, and CURVNET_SET() context setting macros, all currently
resolving to NOPs.
Prepare for virtualization of selected SYSCTL objects by introducing a
family of SYSCTL_V_*() macros, currently resolving to their global
counterparts, i.e. SYSCTL_V_INT() == SYSCTL_INT().
Move selected #defines from sys/sys/vimage.h to newly introduced header
files specific to virtualized subsystems (sys/net/vnet.h,
sys/netinet/vinet.h etc.).
All the changes are verified to have zero functional impact at this
point in time by doing MD5 comparision between pre- and post-change
object files(*).
(*) netipsec/keysock.c did not validate depending on compile time options.
Implemented by: julian, bz, brooks, zec
Reviewed by: julian, bz, brooks, kris, rwatson, ...
Approved by: julian (mentor)
Obtained from: //depot/projects/vimage-commit2/...
X-MFC after: never
Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-10-02 15:37:58 +00:00
|
|
|
INIT_VNET_INET(ifp->if_vnet);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
struct in_multi *inm;
|
|
|
|
|
|
|
|
|
|
inm = NULL;
|
|
|
|
|
|
|
|
|
|
IFF_LOCKGIANT(ifp);
|
|
|
|
|
IN_MULTI_LOCK();
|
|
|
|
|
|
|
|
|
|
IN_LOOKUP_MULTI(*ap, ifp, inm);
|
|
|
|
|
if (inm != NULL) {
|
|
|
|
|
/*
|
|
|
|
|
* If we already joined this group, just bump the
|
|
|
|
|
* refcount and return it.
|
|
|
|
|
*/
|
|
|
|
|
KASSERT(inm->inm_refcount >= 1,
|
|
|
|
|
("%s: bad refcount %d", __func__, inm->inm_refcount));
|
|
|
|
|
++inm->inm_refcount;
|
|
|
|
|
} else do {
|
|
|
|
|
sockunion_t gsa;
|
|
|
|
|
struct ifmultiaddr *ifma;
|
|
|
|
|
struct in_multi *ninm;
|
|
|
|
|
int error;
|
|
|
|
|
|
|
|
|
|
memset(&gsa, 0, sizeof(gsa));
|
|
|
|
|
gsa.sin.sin_family = AF_INET;
|
|
|
|
|
gsa.sin.sin_len = sizeof(struct sockaddr_in);
|
|
|
|
|
gsa.sin.sin_addr = *ap;
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Check if a link-layer group is already associated
|
|
|
|
|
* with this network-layer group on the given ifnet.
|
|
|
|
|
* If so, bump the refcount on the existing network-layer
|
|
|
|
|
* group association and return it.
|
|
|
|
|
*/
|
|
|
|
|
error = if_addmulti(ifp, &gsa.sa, &ifma);
|
|
|
|
|
if (error)
|
|
|
|
|
break;
|
|
|
|
|
if (ifma->ifma_protospec != NULL) {
|
|
|
|
|
inm = (struct in_multi *)ifma->ifma_protospec;
|
|
|
|
|
#ifdef INVARIANTS
|
|
|
|
|
if (inm->inm_ifma != ifma || inm->inm_ifp != ifp ||
|
|
|
|
|
inm->inm_addr.s_addr != ap->s_addr)
|
|
|
|
|
panic("%s: ifma is inconsistent", __func__);
|
|
|
|
|
#endif
|
|
|
|
|
++inm->inm_refcount;
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* A new membership is needed; construct it and
|
|
|
|
|
* perform the IGMP join.
|
|
|
|
|
*/
|
|
|
|
|
ninm = malloc(sizeof(*ninm), M_IPMADDR, M_NOWAIT | M_ZERO);
|
|
|
|
|
if (ninm == NULL) {
|
|
|
|
|
if_delmulti_ifma(ifma);
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
ninm->inm_addr = *ap;
|
|
|
|
|
ninm->inm_ifp = ifp;
|
|
|
|
|
ninm->inm_ifma = ifma;
|
|
|
|
|
ninm->inm_refcount = 1;
|
|
|
|
|
ifma->ifma_protospec = ninm;
|
Commit step 1 of the vimage project, (network stack)
virtualization work done by Marko Zec (zec@).
This is the first in a series of commits over the course
of the next few weeks.
Mark all uses of global variables to be virtualized
with a V_ prefix.
Use macros to map them back to their global names for
now, so this is a NOP change only.
We hope to have caught at least 85-90% of what is needed
so we do not invalidate a lot of outstanding patches again.
Obtained from: //depot/projects/vimage-commit2/...
Reviewed by: brooks, des, ed, mav, julian,
jamie, kris, rwatson, zec, ...
(various people I forgot, different versions)
md5 (with a bit of help)
Sponsored by: NLnet Foundation, The FreeBSD Foundation
X-MFC after: never
V_Commit_Message_Reviewed_By: more people than the patch
2008-08-17 23:27:27 +00:00
|
|
|
LIST_INSERT_HEAD(&V_in_multihead, ninm, inm_link);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
|
|
|
|
|
igmp_joingroup(ninm);
|
|
|
|
|
|
|
|
|
|
inm = ninm;
|
|
|
|
|
} while (0);
|
|
|
|
|
|
|
|
|
|
IN_MULTI_UNLOCK();
|
|
|
|
|
IFF_UNLOCKGIANT(ifp);
|
|
|
|
|
|
|
|
|
|
return (inm);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Leave an IPv4 multicast group.
|
|
|
|
|
* It is OK to call this routine if the underlying ifnet went away.
|
|
|
|
|
*
|
|
|
|
|
* XXX: To deal with the ifp going away, we cheat; the link-layer code in net
|
|
|
|
|
* will set ifma_ifp to NULL when the associated ifnet instance is detached
|
|
|
|
|
* from the system.
|
|
|
|
|
*
|
|
|
|
|
* The only reason we need to violate layers and check ifma_ifp here at all
|
|
|
|
|
* is because certain hardware drivers still require Giant to be held,
|
|
|
|
|
* and it must always be taken before other locks.
|
|
|
|
|
*/
|
|
|
|
|
void
|
|
|
|
|
in_delmulti(struct in_multi *inm)
|
|
|
|
|
{
|
|
|
|
|
struct ifnet *ifp;
|
|
|
|
|
|
|
|
|
|
KASSERT(inm != NULL, ("%s: inm is NULL", __func__));
|
|
|
|
|
KASSERT(inm->inm_ifma != NULL, ("%s: no ifma", __func__));
|
|
|
|
|
ifp = inm->inm_ifma->ifma_ifp;
|
|
|
|
|
|
|
|
|
|
if (ifp != NULL) {
|
|
|
|
|
/*
|
|
|
|
|
* Sanity check that netinet's notion of ifp is the
|
|
|
|
|
* same as net's.
|
|
|
|
|
*/
|
|
|
|
|
KASSERT(inm->inm_ifp == ifp, ("%s: bad ifp", __func__));
|
|
|
|
|
IFF_LOCKGIANT(ifp);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
IN_MULTI_LOCK();
|
|
|
|
|
in_delmulti_locked(inm);
|
|
|
|
|
IN_MULTI_UNLOCK();
|
|
|
|
|
|
|
|
|
|
if (ifp != NULL)
|
|
|
|
|
IFF_UNLOCKGIANT(ifp);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Delete a multicast address record, with locks held.
|
|
|
|
|
*
|
|
|
|
|
* It is OK to call this routine if the ifp went away.
|
|
|
|
|
* Assumes that caller holds the IN_MULTI lock, and that
|
|
|
|
|
* Giant was taken before other locks if required by the hardware.
|
|
|
|
|
*/
|
|
|
|
|
void
|
|
|
|
|
in_delmulti_locked(struct in_multi *inm)
|
|
|
|
|
{
|
|
|
|
|
struct ifmultiaddr *ifma;
|
|
|
|
|
|
|
|
|
|
IN_MULTI_LOCK_ASSERT();
|
|
|
|
|
KASSERT(inm->inm_refcount >= 1, ("%s: freeing freed inm", __func__));
|
|
|
|
|
|
|
|
|
|
if (--inm->inm_refcount == 0) {
|
|
|
|
|
igmp_leavegroup(inm);
|
|
|
|
|
|
|
|
|
|
ifma = inm->inm_ifma;
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (bootverbose)
|
|
|
|
|
printf("%s: purging ifma %p\n", __func__, ifma);
|
|
|
|
|
#endif
|
|
|
|
|
KASSERT(ifma->ifma_protospec == inm,
|
|
|
|
|
("%s: ifma_protospec != inm", __func__));
|
|
|
|
|
ifma->ifma_protospec = NULL;
|
|
|
|
|
|
|
|
|
|
LIST_REMOVE(inm, inm_link);
|
|
|
|
|
free(inm, M_IPMADDR);
|
|
|
|
|
|
|
|
|
|
if_delmulti_ifma(ifma);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Block or unblock an ASM/SSM multicast source on an inpcb.
|
|
|
|
|
*/
|
|
|
|
|
static int
|
|
|
|
|
inp_change_source_filter(struct inpcb *inp, struct sockopt *sopt)
|
|
|
|
|
{
|
Step 1.5 of importing the network stack virtualization infrastructure
from the vimage project, as per plan established at devsummit 08/08:
http://wiki.freebsd.org/Image/Notes200808DevSummit
Introduce INIT_VNET_*() initializer macros, VNET_FOREACH() iterator
macros, and CURVNET_SET() context setting macros, all currently
resolving to NOPs.
Prepare for virtualization of selected SYSCTL objects by introducing a
family of SYSCTL_V_*() macros, currently resolving to their global
counterparts, i.e. SYSCTL_V_INT() == SYSCTL_INT().
Move selected #defines from sys/sys/vimage.h to newly introduced header
files specific to virtualized subsystems (sys/net/vnet.h,
sys/netinet/vinet.h etc.).
All the changes are verified to have zero functional impact at this
point in time by doing MD5 comparision between pre- and post-change
object files(*).
(*) netipsec/keysock.c did not validate depending on compile time options.
Implemented by: julian, bz, brooks, zec
Reviewed by: julian, bz, brooks, kris, rwatson, ...
Approved by: julian (mentor)
Obtained from: //depot/projects/vimage-commit2/...
X-MFC after: never
Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-10-02 15:37:58 +00:00
|
|
|
INIT_VNET_NET(curvnet);
|
|
|
|
|
INIT_VNET_INET(curvnet);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
struct group_source_req gsr;
|
|
|
|
|
sockunion_t *gsa, *ssa;
|
|
|
|
|
struct ifnet *ifp;
|
|
|
|
|
struct in_mfilter *imf;
|
|
|
|
|
struct ip_moptions *imo;
|
|
|
|
|
struct in_msource *ims;
|
|
|
|
|
size_t idx;
|
|
|
|
|
int error;
|
|
|
|
|
int block;
|
|
|
|
|
|
|
|
|
|
ifp = NULL;
|
|
|
|
|
error = 0;
|
|
|
|
|
block = 0;
|
|
|
|
|
|
|
|
|
|
memset(&gsr, 0, sizeof(struct group_source_req));
|
|
|
|
|
gsa = (sockunion_t *)&gsr.gsr_group;
|
|
|
|
|
ssa = (sockunion_t *)&gsr.gsr_source;
|
|
|
|
|
|
|
|
|
|
switch (sopt->sopt_name) {
|
|
|
|
|
case IP_BLOCK_SOURCE:
|
|
|
|
|
case IP_UNBLOCK_SOURCE: {
|
|
|
|
|
struct ip_mreq_source mreqs;
|
|
|
|
|
|
|
|
|
|
error = sooptcopyin(sopt, &mreqs,
|
|
|
|
|
sizeof(struct ip_mreq_source),
|
|
|
|
|
sizeof(struct ip_mreq_source));
|
|
|
|
|
if (error)
|
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
|
|
gsa->sin.sin_family = AF_INET;
|
|
|
|
|
gsa->sin.sin_len = sizeof(struct sockaddr_in);
|
|
|
|
|
gsa->sin.sin_addr = mreqs.imr_multiaddr;
|
|
|
|
|
|
|
|
|
|
ssa->sin.sin_family = AF_INET;
|
|
|
|
|
ssa->sin.sin_len = sizeof(struct sockaddr_in);
|
|
|
|
|
ssa->sin.sin_addr = mreqs.imr_sourceaddr;
|
|
|
|
|
|
|
|
|
|
if (mreqs.imr_interface.s_addr != INADDR_ANY)
|
|
|
|
|
INADDR_TO_IFP(mreqs.imr_interface, ifp);
|
|
|
|
|
|
|
|
|
|
if (sopt->sopt_name == IP_BLOCK_SOURCE)
|
|
|
|
|
block = 1;
|
|
|
|
|
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (bootverbose) {
|
|
|
|
|
printf("%s: imr_interface = %s, ifp = %p\n",
|
|
|
|
|
__func__, inet_ntoa(mreqs.imr_interface), ifp);
|
|
|
|
|
}
|
|
|
|
|
#endif
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
case MCAST_BLOCK_SOURCE:
|
|
|
|
|
case MCAST_UNBLOCK_SOURCE:
|
|
|
|
|
error = sooptcopyin(sopt, &gsr,
|
|
|
|
|
sizeof(struct group_source_req),
|
|
|
|
|
sizeof(struct group_source_req));
|
|
|
|
|
if (error)
|
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
|
|
if (gsa->sin.sin_family != AF_INET ||
|
|
|
|
|
gsa->sin.sin_len != sizeof(struct sockaddr_in))
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
|
|
|
|
|
if (ssa->sin.sin_family != AF_INET ||
|
|
|
|
|
ssa->sin.sin_len != sizeof(struct sockaddr_in))
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
|
Commit step 1 of the vimage project, (network stack)
virtualization work done by Marko Zec (zec@).
This is the first in a series of commits over the course
of the next few weeks.
Mark all uses of global variables to be virtualized
with a V_ prefix.
Use macros to map them back to their global names for
now, so this is a NOP change only.
We hope to have caught at least 85-90% of what is needed
so we do not invalidate a lot of outstanding patches again.
Obtained from: //depot/projects/vimage-commit2/...
Reviewed by: brooks, des, ed, mav, julian,
jamie, kris, rwatson, zec, ...
(various people I forgot, different versions)
md5 (with a bit of help)
Sponsored by: NLnet Foundation, The FreeBSD Foundation
X-MFC after: never
V_Commit_Message_Reviewed_By: more people than the patch
2008-08-17 23:27:27 +00:00
|
|
|
if (gsr.gsr_interface == 0 || V_if_index < gsr.gsr_interface)
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
return (EADDRNOTAVAIL);
|
|
|
|
|
|
|
|
|
|
ifp = ifnet_byindex(gsr.gsr_interface);
|
|
|
|
|
|
|
|
|
|
if (sopt->sopt_name == MCAST_BLOCK_SOURCE)
|
|
|
|
|
block = 1;
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
default:
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (bootverbose) {
|
|
|
|
|
printf("%s: unknown sopt_name %d\n", __func__,
|
|
|
|
|
sopt->sopt_name);
|
|
|
|
|
}
|
|
|
|
|
#endif
|
|
|
|
|
return (EOPNOTSUPP);
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* XXX INET6 */
|
|
|
|
|
if (!IN_MULTICAST(ntohl(gsa->sin.sin_addr.s_addr)))
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Check if we are actually a member of this group.
|
|
|
|
|
*/
|
|
|
|
|
imo = inp_findmoptions(inp);
|
|
|
|
|
idx = imo_match_group(imo, ifp, &gsa->sa);
|
|
|
|
|
if (idx == -1 || imo->imo_mfilters == NULL) {
|
|
|
|
|
error = EADDRNOTAVAIL;
|
|
|
|
|
goto out_locked;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
KASSERT(imo->imo_mfilters != NULL,
|
|
|
|
|
("%s: imo_mfilters not allocated", __func__));
|
|
|
|
|
imf = &imo->imo_mfilters[idx];
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* SSM multicast truth table for block/unblock operations.
|
|
|
|
|
*
|
|
|
|
|
* Operation Filter Mode Entry exists? Action
|
|
|
|
|
*
|
|
|
|
|
* block exclude no add source to filter
|
|
|
|
|
* unblock include no add source to filter
|
|
|
|
|
* block include no EINVAL
|
|
|
|
|
* unblock exclude no EINVAL
|
|
|
|
|
* block exclude yes EADDRNOTAVAIL
|
|
|
|
|
* unblock include yes EADDRNOTAVAIL
|
|
|
|
|
* block include yes remove source from filter
|
|
|
|
|
* unblock exclude yes remove source from filter
|
|
|
|
|
*
|
|
|
|
|
* FreeBSD does not explicitly distinguish between ASM and SSM
|
|
|
|
|
* mode sockets; all sockets are assumed to have a filter list.
|
|
|
|
|
*/
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (bootverbose) {
|
|
|
|
|
printf("%s: imf_fmode is %s\n", __func__,
|
|
|
|
|
imf->imf_fmode == MCAST_INCLUDE ? "include" : "exclude");
|
|
|
|
|
}
|
|
|
|
|
#endif
|
|
|
|
|
ims = imo_match_source(imo, idx, &ssa->sa);
|
|
|
|
|
if (ims == NULL) {
|
|
|
|
|
if ((block == 1 && imf->imf_fmode == MCAST_EXCLUDE) ||
|
|
|
|
|
(block == 0 && imf->imf_fmode == MCAST_INCLUDE)) {
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (bootverbose) {
|
|
|
|
|
printf("%s: adding %s to filter list\n",
|
|
|
|
|
__func__, inet_ntoa(ssa->sin.sin_addr));
|
|
|
|
|
}
|
|
|
|
|
#endif
|
|
|
|
|
error = imo_join_source(imo, idx, ssa);
|
|
|
|
|
}
|
|
|
|
|
if ((block == 1 && imf->imf_fmode == MCAST_INCLUDE) ||
|
|
|
|
|
(block == 0 && imf->imf_fmode == MCAST_EXCLUDE)) {
|
|
|
|
|
/*
|
|
|
|
|
* If the socket is in inclusive mode:
|
|
|
|
|
* the source is already blocked as it has no entry.
|
|
|
|
|
* If the socket is in exclusive mode:
|
|
|
|
|
* the source is already unblocked as it has no entry.
|
|
|
|
|
*/
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (bootverbose) {
|
|
|
|
|
printf("%s: ims %p; %s already [un]blocked\n",
|
|
|
|
|
__func__, ims,
|
|
|
|
|
inet_ntoa(ssa->sin.sin_addr));
|
|
|
|
|
}
|
|
|
|
|
#endif
|
|
|
|
|
error = EINVAL;
|
|
|
|
|
}
|
|
|
|
|
} else {
|
|
|
|
|
if ((block == 1 && imf->imf_fmode == MCAST_EXCLUDE) ||
|
|
|
|
|
(block == 0 && imf->imf_fmode == MCAST_INCLUDE)) {
|
|
|
|
|
/*
|
|
|
|
|
* If the socket is in exclusive mode:
|
|
|
|
|
* the source is already blocked as it has an entry.
|
|
|
|
|
* If the socket is in inclusive mode:
|
|
|
|
|
* the source is already unblocked as it has an entry.
|
|
|
|
|
*/
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (bootverbose) {
|
|
|
|
|
printf("%s: ims %p; %s already [un]blocked\n",
|
|
|
|
|
__func__, ims,
|
|
|
|
|
inet_ntoa(ssa->sin.sin_addr));
|
|
|
|
|
}
|
|
|
|
|
#endif
|
|
|
|
|
error = EADDRNOTAVAIL;
|
|
|
|
|
}
|
|
|
|
|
if ((block == 1 && imf->imf_fmode == MCAST_INCLUDE) ||
|
|
|
|
|
(block == 0 && imf->imf_fmode == MCAST_EXCLUDE)) {
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (bootverbose) {
|
|
|
|
|
printf("%s: removing %s from filter list\n",
|
|
|
|
|
__func__, inet_ntoa(ssa->sin.sin_addr));
|
|
|
|
|
}
|
|
|
|
|
#endif
|
|
|
|
|
error = imo_leave_source(imo, idx, ssa);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
out_locked:
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Given an inpcb, return its multicast options structure pointer. Accepts
|
|
|
|
|
* an unlocked inpcb pointer, but will return it locked. May sleep.
|
|
|
|
|
*/
|
|
|
|
|
static struct ip_moptions *
|
|
|
|
|
inp_findmoptions(struct inpcb *inp)
|
|
|
|
|
{
|
|
|
|
|
struct ip_moptions *imo;
|
|
|
|
|
struct in_multi **immp;
|
|
|
|
|
struct in_mfilter *imfp;
|
|
|
|
|
size_t idx;
|
|
|
|
|
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
if (inp->inp_moptions != NULL)
|
|
|
|
|
return (inp->inp_moptions);
|
|
|
|
|
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
|
|
|
|
|
imo = (struct ip_moptions *)malloc(sizeof(*imo), M_IPMOPTS,
|
|
|
|
|
M_WAITOK);
|
|
|
|
|
immp = (struct in_multi **)malloc(sizeof(*immp) * IP_MIN_MEMBERSHIPS,
|
|
|
|
|
M_IPMOPTS, M_WAITOK | M_ZERO);
|
|
|
|
|
imfp = (struct in_mfilter *)malloc(
|
|
|
|
|
sizeof(struct in_mfilter) * IP_MIN_MEMBERSHIPS,
|
|
|
|
|
M_IPMSOURCE, M_WAITOK);
|
|
|
|
|
|
|
|
|
|
imo->imo_multicast_ifp = NULL;
|
|
|
|
|
imo->imo_multicast_addr.s_addr = INADDR_ANY;
|
|
|
|
|
imo->imo_multicast_vif = -1;
|
|
|
|
|
imo->imo_multicast_ttl = IP_DEFAULT_MULTICAST_TTL;
|
|
|
|
|
imo->imo_multicast_loop = IP_DEFAULT_MULTICAST_LOOP;
|
|
|
|
|
imo->imo_num_memberships = 0;
|
|
|
|
|
imo->imo_max_memberships = IP_MIN_MEMBERSHIPS;
|
|
|
|
|
imo->imo_membership = immp;
|
|
|
|
|
|
|
|
|
|
/* Initialize per-group source filters. */
|
|
|
|
|
for (idx = 0; idx < IP_MIN_MEMBERSHIPS; idx++) {
|
|
|
|
|
imfp[idx].imf_fmode = MCAST_EXCLUDE;
|
|
|
|
|
imfp[idx].imf_nsources = 0;
|
|
|
|
|
TAILQ_INIT(&imfp[idx].imf_sources);
|
|
|
|
|
}
|
|
|
|
|
imo->imo_mfilters = imfp;
|
|
|
|
|
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
if (inp->inp_moptions != NULL) {
|
|
|
|
|
free(imfp, M_IPMSOURCE);
|
|
|
|
|
free(immp, M_IPMOPTS);
|
|
|
|
|
free(imo, M_IPMOPTS);
|
|
|
|
|
return (inp->inp_moptions);
|
|
|
|
|
}
|
|
|
|
|
inp->inp_moptions = imo;
|
|
|
|
|
return (imo);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Discard the IP multicast options (and source filters).
|
|
|
|
|
*/
|
|
|
|
|
void
|
|
|
|
|
inp_freemoptions(struct ip_moptions *imo)
|
|
|
|
|
{
|
|
|
|
|
struct in_mfilter *imf;
|
|
|
|
|
struct in_msource *ims, *tims;
|
|
|
|
|
size_t idx, nmships;
|
|
|
|
|
|
|
|
|
|
KASSERT(imo != NULL, ("%s: ip_moptions is NULL", __func__));
|
|
|
|
|
|
|
|
|
|
nmships = imo->imo_num_memberships;
|
|
|
|
|
for (idx = 0; idx < nmships; ++idx) {
|
|
|
|
|
in_delmulti(imo->imo_membership[idx]);
|
|
|
|
|
|
|
|
|
|
if (imo->imo_mfilters != NULL) {
|
|
|
|
|
imf = &imo->imo_mfilters[idx];
|
|
|
|
|
TAILQ_FOREACH_SAFE(ims, &imf->imf_sources,
|
|
|
|
|
ims_next, tims) {
|
|
|
|
|
TAILQ_REMOVE(&imf->imf_sources, ims, ims_next);
|
2008-10-23 15:53:51 +00:00
|
|
|
free(ims, M_IPMSOURCE);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
imf->imf_nsources--;
|
|
|
|
|
}
|
|
|
|
|
KASSERT(imf->imf_nsources == 0,
|
|
|
|
|
("%s: did not free all imf_nsources", __func__));
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (imo->imo_mfilters != NULL)
|
|
|
|
|
free(imo->imo_mfilters, M_IPMSOURCE);
|
|
|
|
|
free(imo->imo_membership, M_IPMOPTS);
|
|
|
|
|
free(imo, M_IPMOPTS);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Atomically get source filters on a socket for an IPv4 multicast group.
|
|
|
|
|
* Called with INP lock held; returns with lock released.
|
|
|
|
|
*/
|
|
|
|
|
static int
|
|
|
|
|
inp_get_source_filters(struct inpcb *inp, struct sockopt *sopt)
|
|
|
|
|
{
|
Step 1.5 of importing the network stack virtualization infrastructure
from the vimage project, as per plan established at devsummit 08/08:
http://wiki.freebsd.org/Image/Notes200808DevSummit
Introduce INIT_VNET_*() initializer macros, VNET_FOREACH() iterator
macros, and CURVNET_SET() context setting macros, all currently
resolving to NOPs.
Prepare for virtualization of selected SYSCTL objects by introducing a
family of SYSCTL_V_*() macros, currently resolving to their global
counterparts, i.e. SYSCTL_V_INT() == SYSCTL_INT().
Move selected #defines from sys/sys/vimage.h to newly introduced header
files specific to virtualized subsystems (sys/net/vnet.h,
sys/netinet/vinet.h etc.).
All the changes are verified to have zero functional impact at this
point in time by doing MD5 comparision between pre- and post-change
object files(*).
(*) netipsec/keysock.c did not validate depending on compile time options.
Implemented by: julian, bz, brooks, zec
Reviewed by: julian, bz, brooks, kris, rwatson, ...
Approved by: julian (mentor)
Obtained from: //depot/projects/vimage-commit2/...
X-MFC after: never
Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-10-02 15:37:58 +00:00
|
|
|
INIT_VNET_NET(curvnet);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
struct __msfilterreq msfr;
|
|
|
|
|
sockunion_t *gsa;
|
|
|
|
|
struct ifnet *ifp;
|
|
|
|
|
struct ip_moptions *imo;
|
|
|
|
|
struct in_mfilter *imf;
|
|
|
|
|
struct in_msource *ims;
|
|
|
|
|
struct sockaddr_storage *ptss;
|
|
|
|
|
struct sockaddr_storage *tss;
|
|
|
|
|
int error;
|
|
|
|
|
size_t idx;
|
|
|
|
|
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK_ASSERT(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
|
|
|
|
|
imo = inp->inp_moptions;
|
|
|
|
|
KASSERT(imo != NULL, ("%s: null ip_moptions", __func__));
|
|
|
|
|
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
|
|
|
|
|
error = sooptcopyin(sopt, &msfr, sizeof(struct __msfilterreq),
|
|
|
|
|
sizeof(struct __msfilterreq));
|
|
|
|
|
if (error)
|
|
|
|
|
return (error);
|
|
|
|
|
|
Commit step 1 of the vimage project, (network stack)
virtualization work done by Marko Zec (zec@).
This is the first in a series of commits over the course
of the next few weeks.
Mark all uses of global variables to be virtualized
with a V_ prefix.
Use macros to map them back to their global names for
now, so this is a NOP change only.
We hope to have caught at least 85-90% of what is needed
so we do not invalidate a lot of outstanding patches again.
Obtained from: //depot/projects/vimage-commit2/...
Reviewed by: brooks, des, ed, mav, julian,
jamie, kris, rwatson, zec, ...
(various people I forgot, different versions)
md5 (with a bit of help)
Sponsored by: NLnet Foundation, The FreeBSD Foundation
X-MFC after: never
V_Commit_Message_Reviewed_By: more people than the patch
2008-08-17 23:27:27 +00:00
|
|
|
if (msfr.msfr_ifindex == 0 || V_if_index < msfr.msfr_ifindex)
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
return (EINVAL);
|
|
|
|
|
|
|
|
|
|
ifp = ifnet_byindex(msfr.msfr_ifindex);
|
|
|
|
|
if (ifp == NULL)
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Lookup group on the socket.
|
|
|
|
|
*/
|
|
|
|
|
gsa = (sockunion_t *)&msfr.msfr_group;
|
|
|
|
|
idx = imo_match_group(imo, ifp, &gsa->sa);
|
|
|
|
|
if (idx == -1 || imo->imo_mfilters == NULL) {
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
return (EADDRNOTAVAIL);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
imf = &imo->imo_mfilters[idx];
|
|
|
|
|
msfr.msfr_fmode = imf->imf_fmode;
|
|
|
|
|
msfr.msfr_nsrcs = imf->imf_nsources;
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* If the user specified a buffer, copy out the source filter
|
|
|
|
|
* entries to userland gracefully.
|
|
|
|
|
* msfr.msfr_nsrcs is always set to the total number of filter
|
|
|
|
|
* entries which the kernel currently has for this group.
|
|
|
|
|
*/
|
|
|
|
|
tss = NULL;
|
|
|
|
|
if (msfr.msfr_srcs != NULL && msfr.msfr_nsrcs > 0) {
|
|
|
|
|
/*
|
|
|
|
|
* Make a copy of the source vector so that we do not
|
|
|
|
|
* thrash the inpcb lock whilst copying it out.
|
|
|
|
|
* We only copy out the number of entries which userland
|
|
|
|
|
* has asked for, but we always tell userland how big the
|
|
|
|
|
* buffer really needs to be.
|
|
|
|
|
*/
|
2008-10-23 20:26:15 +00:00
|
|
|
tss = malloc(sizeof(struct sockaddr_storage) * msfr.msfr_nsrcs,
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
M_TEMP, M_NOWAIT);
|
|
|
|
|
if (tss == NULL) {
|
|
|
|
|
error = ENOBUFS;
|
|
|
|
|
} else {
|
|
|
|
|
ptss = tss;
|
|
|
|
|
TAILQ_FOREACH(ims, &imf->imf_sources, ims_next) {
|
|
|
|
|
memcpy(ptss++, &ims->ims_addr,
|
|
|
|
|
sizeof(struct sockaddr_storage));
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
|
|
|
|
|
if (tss != NULL) {
|
|
|
|
|
error = copyout(tss, msfr.msfr_srcs,
|
|
|
|
|
sizeof(struct sockaddr_storage) * msfr.msfr_nsrcs);
|
2008-10-23 15:53:51 +00:00
|
|
|
free(tss, M_TEMP);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (error)
|
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
|
|
error = sooptcopyout(sopt, &msfr, sizeof(struct __msfilterreq));
|
|
|
|
|
|
|
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Return the IP multicast options in response to user getsockopt().
|
|
|
|
|
*/
|
|
|
|
|
int
|
|
|
|
|
inp_getmoptions(struct inpcb *inp, struct sockopt *sopt)
|
|
|
|
|
{
|
Step 1.5 of importing the network stack virtualization infrastructure
from the vimage project, as per plan established at devsummit 08/08:
http://wiki.freebsd.org/Image/Notes200808DevSummit
Introduce INIT_VNET_*() initializer macros, VNET_FOREACH() iterator
macros, and CURVNET_SET() context setting macros, all currently
resolving to NOPs.
Prepare for virtualization of selected SYSCTL objects by introducing a
family of SYSCTL_V_*() macros, currently resolving to their global
counterparts, i.e. SYSCTL_V_INT() == SYSCTL_INT().
Move selected #defines from sys/sys/vimage.h to newly introduced header
files specific to virtualized subsystems (sys/net/vnet.h,
sys/netinet/vinet.h etc.).
All the changes are verified to have zero functional impact at this
point in time by doing MD5 comparision between pre- and post-change
object files(*).
(*) netipsec/keysock.c did not validate depending on compile time options.
Implemented by: julian, bz, brooks, zec
Reviewed by: julian, bz, brooks, kris, rwatson, ...
Approved by: julian (mentor)
Obtained from: //depot/projects/vimage-commit2/...
X-MFC after: never
Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-10-02 15:37:58 +00:00
|
|
|
INIT_VNET_INET(curvnet);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
struct ip_mreqn mreqn;
|
|
|
|
|
struct ip_moptions *imo;
|
|
|
|
|
struct ifnet *ifp;
|
|
|
|
|
struct in_ifaddr *ia;
|
|
|
|
|
int error, optval;
|
|
|
|
|
u_char coptval;
|
|
|
|
|
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
imo = inp->inp_moptions;
|
Over the past couple of years, there have been a number of reports relating
the use of divert sockets to dead locks. A number of LORs have been reported
between divert and a number of other network subsystems including: IPSEC, Pfil,
multicast, ipfw and others. Other dead locks could occur because of recursive
entry into the IP stack. This change should take care of most if not all of
these issues.
A summary of the changes follow:
- We disallow multicast operations on divert sockets. It really doesn't make
semantic sense to allow this, since typically you would set multicast
parameters on multicast end points.
NOTE: As a part of this change, we actually dis-allow multicast options on
any socket that IS a divert socket OR IS NOT a SOCK_RAW or SOCK_DGRAM family
- We check to see if there are any socket options that have been specified on
the socket, and if there was (which is very un-common and also probably
doesnt make sense to support) we duplicate the mbuf carrying the options.
- We then drop the INP/INFO locks over the call to ip_output(). It should be
noted that since we no longer support multicast operations on divert sockets
and we have duplicated any socket options, we no longer need the reference
to the pcb to be coherent.
- Finally, we replaced the call to ip_input() to use netisr queuing. This
should remove the recursive entry into the IP stack from divert.
By dropping the locks over the call to ip_output() we eliminate all the lock
ordering issues above. By switching over to netisr on the inbound path,
we can no longer recursively enter the ip_input() code via divert.
I have tested this change by using the following command:
ipfwpcap -r 8000 - | tcpdump -r - -nn -v
This should exercise the input and re-injection (outbound) path, which is
very similar to the work load performed by natd(8). Additionally, I have
run some ospf daemons which have a heavy reliance on raw sockets and
multicast.
Approved by: re@ (kensmith)
MFC after: 1 month
LOR: 163
LOR: 181
LOR: 202
LOR: 203
Discussed with: julian, andre et al (on freebsd-net)
In collaboration with: bms [1], rwatson [2]
[1] bms helped out with the multicast decisions
[2] rwatson submitted the original netisr patches and came up with some
of the original ideas on how to combat this issue.
2007-08-06 22:06:36 +00:00
|
|
|
/*
|
|
|
|
|
* If socket is neither of type SOCK_RAW or SOCK_DGRAM,
|
|
|
|
|
* or is a divert socket, reject it.
|
|
|
|
|
*/
|
|
|
|
|
if (inp->inp_socket->so_proto->pr_protocol == IPPROTO_DIVERT ||
|
|
|
|
|
(inp->inp_socket->so_proto->pr_type != SOCK_RAW &&
|
|
|
|
|
inp->inp_socket->so_proto->pr_type != SOCK_DGRAM)) {
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Over the past couple of years, there have been a number of reports relating
the use of divert sockets to dead locks. A number of LORs have been reported
between divert and a number of other network subsystems including: IPSEC, Pfil,
multicast, ipfw and others. Other dead locks could occur because of recursive
entry into the IP stack. This change should take care of most if not all of
these issues.
A summary of the changes follow:
- We disallow multicast operations on divert sockets. It really doesn't make
semantic sense to allow this, since typically you would set multicast
parameters on multicast end points.
NOTE: As a part of this change, we actually dis-allow multicast options on
any socket that IS a divert socket OR IS NOT a SOCK_RAW or SOCK_DGRAM family
- We check to see if there are any socket options that have been specified on
the socket, and if there was (which is very un-common and also probably
doesnt make sense to support) we duplicate the mbuf carrying the options.
- We then drop the INP/INFO locks over the call to ip_output(). It should be
noted that since we no longer support multicast operations on divert sockets
and we have duplicated any socket options, we no longer need the reference
to the pcb to be coherent.
- Finally, we replaced the call to ip_input() to use netisr queuing. This
should remove the recursive entry into the IP stack from divert.
By dropping the locks over the call to ip_output() we eliminate all the lock
ordering issues above. By switching over to netisr on the inbound path,
we can no longer recursively enter the ip_input() code via divert.
I have tested this change by using the following command:
ipfwpcap -r 8000 - | tcpdump -r - -nn -v
This should exercise the input and re-injection (outbound) path, which is
very similar to the work load performed by natd(8). Additionally, I have
run some ospf daemons which have a heavy reliance on raw sockets and
multicast.
Approved by: re@ (kensmith)
MFC after: 1 month
LOR: 163
LOR: 181
LOR: 202
LOR: 203
Discussed with: julian, andre et al (on freebsd-net)
In collaboration with: bms [1], rwatson [2]
[1] bms helped out with the multicast decisions
[2] rwatson submitted the original netisr patches and came up with some
of the original ideas on how to combat this issue.
2007-08-06 22:06:36 +00:00
|
|
|
return (EOPNOTSUPP);
|
|
|
|
|
}
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
|
|
|
|
|
error = 0;
|
|
|
|
|
switch (sopt->sopt_name) {
|
|
|
|
|
case IP_MULTICAST_VIF:
|
|
|
|
|
if (imo != NULL)
|
|
|
|
|
optval = imo->imo_multicast_vif;
|
|
|
|
|
else
|
|
|
|
|
optval = -1;
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
error = sooptcopyout(sopt, &optval, sizeof(int));
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case IP_MULTICAST_IF:
|
|
|
|
|
memset(&mreqn, 0, sizeof(struct ip_mreqn));
|
|
|
|
|
if (imo != NULL) {
|
|
|
|
|
ifp = imo->imo_multicast_ifp;
|
|
|
|
|
if (imo->imo_multicast_addr.s_addr != INADDR_ANY) {
|
|
|
|
|
mreqn.imr_address = imo->imo_multicast_addr;
|
|
|
|
|
} else if (ifp != NULL) {
|
|
|
|
|
mreqn.imr_ifindex = ifp->if_index;
|
|
|
|
|
IFP_TO_IA(ifp, ia);
|
|
|
|
|
if (ia != NULL) {
|
|
|
|
|
mreqn.imr_address =
|
|
|
|
|
IA_SIN(ia)->sin_addr;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
if (sopt->sopt_valsize == sizeof(struct ip_mreqn)) {
|
|
|
|
|
error = sooptcopyout(sopt, &mreqn,
|
|
|
|
|
sizeof(struct ip_mreqn));
|
|
|
|
|
} else {
|
|
|
|
|
error = sooptcopyout(sopt, &mreqn.imr_address,
|
|
|
|
|
sizeof(struct in_addr));
|
|
|
|
|
}
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case IP_MULTICAST_TTL:
|
|
|
|
|
if (imo == 0)
|
|
|
|
|
optval = coptval = IP_DEFAULT_MULTICAST_TTL;
|
|
|
|
|
else
|
|
|
|
|
optval = coptval = imo->imo_multicast_ttl;
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
if (sopt->sopt_valsize == sizeof(u_char))
|
|
|
|
|
error = sooptcopyout(sopt, &coptval, sizeof(u_char));
|
|
|
|
|
else
|
|
|
|
|
error = sooptcopyout(sopt, &optval, sizeof(int));
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case IP_MULTICAST_LOOP:
|
|
|
|
|
if (imo == 0)
|
|
|
|
|
optval = coptval = IP_DEFAULT_MULTICAST_LOOP;
|
|
|
|
|
else
|
|
|
|
|
optval = coptval = imo->imo_multicast_loop;
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
if (sopt->sopt_valsize == sizeof(u_char))
|
|
|
|
|
error = sooptcopyout(sopt, &coptval, sizeof(u_char));
|
|
|
|
|
else
|
|
|
|
|
error = sooptcopyout(sopt, &optval, sizeof(int));
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case IP_MSFILTER:
|
|
|
|
|
if (imo == NULL) {
|
|
|
|
|
error = EADDRNOTAVAIL;
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
} else {
|
|
|
|
|
error = inp_get_source_filters(inp, sopt);
|
|
|
|
|
}
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
default:
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
error = ENOPROTOOPT;
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
INP_UNLOCK_ASSERT(inp);
|
|
|
|
|
|
|
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Join an IPv4 multicast group, possibly with a source.
|
|
|
|
|
*/
|
|
|
|
|
static int
|
|
|
|
|
inp_join_group(struct inpcb *inp, struct sockopt *sopt)
|
|
|
|
|
{
|
Step 1.5 of importing the network stack virtualization infrastructure
from the vimage project, as per plan established at devsummit 08/08:
http://wiki.freebsd.org/Image/Notes200808DevSummit
Introduce INIT_VNET_*() initializer macros, VNET_FOREACH() iterator
macros, and CURVNET_SET() context setting macros, all currently
resolving to NOPs.
Prepare for virtualization of selected SYSCTL objects by introducing a
family of SYSCTL_V_*() macros, currently resolving to their global
counterparts, i.e. SYSCTL_V_INT() == SYSCTL_INT().
Move selected #defines from sys/sys/vimage.h to newly introduced header
files specific to virtualized subsystems (sys/net/vnet.h,
sys/netinet/vinet.h etc.).
All the changes are verified to have zero functional impact at this
point in time by doing MD5 comparision between pre- and post-change
object files(*).
(*) netipsec/keysock.c did not validate depending on compile time options.
Implemented by: julian, bz, brooks, zec
Reviewed by: julian, bz, brooks, kris, rwatson, ...
Approved by: julian (mentor)
Obtained from: //depot/projects/vimage-commit2/...
X-MFC after: never
Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-10-02 15:37:58 +00:00
|
|
|
INIT_VNET_NET(curvnet);
|
|
|
|
|
INIT_VNET_INET(curvnet);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
struct group_source_req gsr;
|
|
|
|
|
sockunion_t *gsa, *ssa;
|
|
|
|
|
struct ifnet *ifp;
|
|
|
|
|
struct in_mfilter *imf;
|
|
|
|
|
struct ip_moptions *imo;
|
|
|
|
|
struct in_multi *inm;
|
|
|
|
|
size_t idx;
|
|
|
|
|
int error;
|
|
|
|
|
|
|
|
|
|
ifp = NULL;
|
|
|
|
|
error = 0;
|
|
|
|
|
|
|
|
|
|
memset(&gsr, 0, sizeof(struct group_source_req));
|
|
|
|
|
gsa = (sockunion_t *)&gsr.gsr_group;
|
|
|
|
|
gsa->ss.ss_family = AF_UNSPEC;
|
|
|
|
|
ssa = (sockunion_t *)&gsr.gsr_source;
|
|
|
|
|
ssa->ss.ss_family = AF_UNSPEC;
|
|
|
|
|
|
|
|
|
|
switch (sopt->sopt_name) {
|
|
|
|
|
case IP_ADD_MEMBERSHIP:
|
|
|
|
|
case IP_ADD_SOURCE_MEMBERSHIP: {
|
|
|
|
|
struct ip_mreq_source mreqs;
|
|
|
|
|
|
|
|
|
|
if (sopt->sopt_name == IP_ADD_MEMBERSHIP) {
|
|
|
|
|
error = sooptcopyin(sopt, &mreqs,
|
|
|
|
|
sizeof(struct ip_mreq),
|
|
|
|
|
sizeof(struct ip_mreq));
|
|
|
|
|
/*
|
|
|
|
|
* Do argument switcharoo from ip_mreq into
|
|
|
|
|
* ip_mreq_source to avoid using two instances.
|
|
|
|
|
*/
|
|
|
|
|
mreqs.imr_interface = mreqs.imr_sourceaddr;
|
|
|
|
|
mreqs.imr_sourceaddr.s_addr = INADDR_ANY;
|
|
|
|
|
} else if (sopt->sopt_name == IP_ADD_SOURCE_MEMBERSHIP) {
|
|
|
|
|
error = sooptcopyin(sopt, &mreqs,
|
|
|
|
|
sizeof(struct ip_mreq_source),
|
|
|
|
|
sizeof(struct ip_mreq_source));
|
|
|
|
|
}
|
|
|
|
|
if (error)
|
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
|
|
gsa->sin.sin_family = AF_INET;
|
|
|
|
|
gsa->sin.sin_len = sizeof(struct sockaddr_in);
|
|
|
|
|
gsa->sin.sin_addr = mreqs.imr_multiaddr;
|
|
|
|
|
|
|
|
|
|
if (sopt->sopt_name == IP_ADD_SOURCE_MEMBERSHIP) {
|
|
|
|
|
ssa->sin.sin_family = AF_INET;
|
|
|
|
|
ssa->sin.sin_len = sizeof(struct sockaddr_in);
|
|
|
|
|
ssa->sin.sin_addr = mreqs.imr_sourceaddr;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Obtain ifp. If no interface address was provided,
|
2007-07-09 10:36:47 +00:00
|
|
|
* use the interface of the route in the unicast FIB for
|
|
|
|
|
* the given multicast destination; usually, this is the
|
|
|
|
|
* default route.
|
|
|
|
|
* If this lookup fails, attempt to use the first non-loopback
|
|
|
|
|
* interface with multicast capability in the system as a
|
|
|
|
|
* last resort. The legacy IPv4 ASM API requires that we do
|
|
|
|
|
* this in order to allow groups to be joined when the routing
|
|
|
|
|
* table has not yet been populated during boot.
|
|
|
|
|
* If all of these conditions fail, return EADDRNOTAVAIL, and
|
|
|
|
|
* reject the IPv4 multicast join.
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
*/
|
|
|
|
|
if (mreqs.imr_interface.s_addr != INADDR_ANY) {
|
|
|
|
|
INADDR_TO_IFP(mreqs.imr_interface, ifp);
|
|
|
|
|
} else {
|
|
|
|
|
struct route ro;
|
|
|
|
|
|
|
|
|
|
ro.ro_rt = NULL;
|
|
|
|
|
*(struct sockaddr_in *)&ro.ro_dst = gsa->sin;
|
This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,
The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.
Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:
- Kip Macy revised the locking code completely, thus completing
the last piece of the puzzle, Kip has also been conducting
active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
|
|
|
in_rtalloc_ign(&ro, 0,
|
Add code to allow the system to handle multiple routing tables.
This particular implementation is designed to be fully backwards compatible
and to be MFC-able to 7.x (and 6.x)
Currently the only protocol that can make use of the multiple tables is IPv4
Similar functionality exists in OpenBSD and Linux.
From my notes:
-----
One thing where FreeBSD has been falling behind, and which by chance I
have some time to work on is "policy based routing", which allows
different
packet streams to be routed by more than just the destination address.
Constraints:
------------
I want to make some form of this available in the 6.x tree
(and by extension 7.x) , but FreeBSD in general needs it so I might as
well do it in -current and back port the portions I need.
One of the ways that this can be done is to have the ability to
instantiate multiple kernel routing tables (which I will now
refer to as "Forwarding Information Bases" or "FIBs" for political
correctness reasons). Which FIB a particular packet uses to make
the next hop decision can be decided by a number of mechanisms.
The policies these mechanisms implement are the "Policies" referred
to in "Policy based routing".
One of the constraints I have if I try to back port this work to
6.x is that it must be implemented as a EXTENSION to the existing
ABIs in 6.x so that third party applications do not need to be
recompiled in timespan of the branch.
This first version will not have some of the bells and whistles that
will come with later versions. It will, for example, be limited to 16
tables in the first commit.
Implementation method, Compatible version. (part 1)
-------------------------------
For this reason I have implemented a "sufficient subset" of a
multiple routing table solution in Perforce, and back-ported it
to 6.x. (also in Perforce though not always caught up with what I
have done in -current/P4). The subset allows a number of FIBs
to be defined at compile time (8 is sufficient for my purposes in 6.x)
and implements the changes needed to allow IPV4 to use them. I have not
done the changes for ipv6 simply because I do not need it, and I do not
have enough knowledge of ipv6 (e.g. neighbor discovery) needed to do it.
Other protocol families are left untouched and should there be
users with proprietary protocol families, they should continue to work
and be oblivious to the existence of the extra FIBs.
To understand how this is done, one must know that the current FIB
code starts everything off with a single dimensional array of
pointers to FIB head structures (One per protocol family), each of
which in turn points to the trie of routes available to that family.
The basic change in the ABI compatible version of the change is to
extent that array to be a 2 dimensional array, so that
instead of protocol family X looking at rt_tables[X] for the
table it needs, it looks at rt_tables[Y][X] when for all
protocol families except ipv4 Y is always 0.
Code that is unaware of the change always just sees the first row
of the table, which of course looks just like the one dimensional
array that existed before.
The entry points rtrequest(), rtalloc(), rtalloc1(), rtalloc_ign()
are all maintained, but refer only to the first row of the array,
so that existing callers in proprietary protocols can continue to
do the "right thing".
Some new entry points are added, for the exclusive use of ipv4 code
called in_rtrequest(), in_rtalloc(), in_rtalloc1() and in_rtalloc_ign(),
which have an extra argument which refers the code to the correct row.
In addition, there are some new entry points (currently called
rtalloc_fib() and friends) that check the Address family being
looked up and call either rtalloc() (and friends) if the protocol
is not IPv4 forcing the action to row 0 or to the appropriate row
if it IS IPv4 (and that info is available). These are for calling
from code that is not specific to any particular protocol. The way
these are implemented would change in the non ABI preserving code
to be added later.
One feature of the first version of the code is that for ipv4,
the interface routes show up automatically on all the FIBs, so
that no matter what FIB you select you always have the basic
direct attached hosts available to you. (rtinit() does this
automatically).
You CAN delete an interface route from one FIB should you want
to but by default it's there. ARP information is also available
in each FIB. It's assumed that the same machine would have the
same MAC address, regardless of which FIB you are using to get
to it.
This brings us as to how the correct FIB is selected for an outgoing
IPV4 packet.
Firstly, all packets have a FIB associated with them. if nothing
has been done to change it, it will be FIB 0. The FIB is changed
in the following ways.
Packets fall into one of a number of classes.
1/ locally generated packets, coming from a socket/PCB.
Such packets select a FIB from a number associated with the
socket/PCB. This in turn is inherited from the process,
but can be changed by a socket option. The process in turn
inherits it on fork. I have written a utility call setfib
that acts a bit like nice..
setfib -3 ping target.example.com # will use fib 3 for ping.
It is an obvious extension to make it a property of a jail
but I have not done so. It can be achieved by combining the setfib and
jail commands.
2/ packets received on an interface for forwarding.
By default these packets would use table 0,
(or possibly a number settable in a sysctl(not yet)).
but prior to routing the firewall can inspect them (see below).
(possibly in the future you may be able to associate a FIB
with packets received on an interface.. An ifconfig arg, but not yet.)
3/ packets inspected by a packet classifier, which can arbitrarily
associate a fib with it on a packet by packet basis.
A fib assigned to a packet by a packet classifier
(such as ipfw) would over-ride a fib associated by
a more default source. (such as cases 1 or 2).
4/ a tcp listen socket associated with a fib will generate
accept sockets that are associated with that same fib.
5/ Packets generated in response to some other packet (e.g. reset
or icmp packets). These should use the FIB associated with the
packet being reponded to.
6/ Packets generated during encapsulation.
gif, tun and other tunnel interfaces will encapsulate using the FIB
that was in effect withthe proces that set up the tunnel.
thus setfib 1 ifconfig gif0 [tunnel instructions]
will set the fib for the tunnel to use to be fib 1.
Routing messages would be associated with their
process, and thus select one FIB or another.
messages from the kernel would be associated with the fib they
refer to and would only be received by a routing socket associated
with that fib. (not yet implemented)
In addition Netstat has been edited to be able to cope with the
fact that the array is now 2 dimensional. (It looks in system
memory using libkvm (!)). Old versions of netstat see only the first FIB.
In addition two sysctls are added to give:
a) the number of FIBs compiled in (active)
b) the default FIB of the calling process.
Early testing experience:
-------------------------
Basically our (IronPort's) appliance does this functionality already
using ipfw fwd but that method has some drawbacks.
For example,
It can't fully simulate a routing table because it can't influence the
socket's choice of local address when a connect() is done.
Testing during the generating of these changes has been
remarkably smooth so far. Multiple tables have co-existed
with no notable side effects, and packets have been routes
accordingly.
ipfw has grown 2 new keywords:
setfib N ip from anay to any
count ip from any to any fib N
In pf there seems to be a requirement to be able to give symbolic names to the
fibs but I do not have that capacity. I am not sure if it is required.
SCTP has interestingly enough built in support for this, called VRFs
in Cisco parlance. it will be interesting to see how that handles it
when it suddenly actually does something.
Where to next:
--------------------
After committing the ABI compatible version and MFCing it, I'd
like to proceed in a forward direction in -current. this will
result in some roto-tilling in the routing code.
Firstly: the current code's idea of having a separate tree per
protocol family, all of the same format, and pointed to by the
1 dimensional array is a bit silly. Especially when one considers that
there is code that makes assumptions about every protocol having the
same internal structures there. Some protocols don't WANT that
sort of structure. (for example the whole idea of a netmask is foreign
to appletalk). This needs to be made opaque to the external code.
My suggested first change is to add routing method pointers to the
'domain' structure, along with information pointing the data.
instead of having an array of pointers to uniform structures,
there would be an array pointing to the 'domain' structures
for each protocol address domain (protocol family),
and the methods this reached would be called. The methods would have
an argument that gives FIB number, but the protocol would be free
to ignore it.
When the ABI can be changed it raises the possibilty of the
addition of a fib entry into the "struct route". Currently,
the structure contains the sockaddr of the desination, and the resulting
fib entry. To make this work fully, one could add a fib number
so that given an address and a fib, one can find the third element, the
fib entry.
Interaction with the ARP layer/ LL layer would need to be
revisited as well. Qing Li has been working on this already.
This work was sponsored by Ironport Systems/Cisco
Reviewed by: several including rwatson, bz and mlair (parts each)
Obtained from: Ironport systems/Cisco
2008-05-09 23:03:00 +00:00
|
|
|
inp->inp_inc.inc_fibnum);
|
2007-07-09 10:36:47 +00:00
|
|
|
if (ro.ro_rt != NULL) {
|
|
|
|
|
ifp = ro.ro_rt->rt_ifp;
|
|
|
|
|
KASSERT(ifp != NULL, ("%s: null ifp",
|
|
|
|
|
__func__));
|
|
|
|
|
RTFREE(ro.ro_rt);
|
|
|
|
|
} else {
|
|
|
|
|
struct in_ifaddr *ia;
|
|
|
|
|
struct ifnet *mfp = NULL;
|
Commit step 1 of the vimage project, (network stack)
virtualization work done by Marko Zec (zec@).
This is the first in a series of commits over the course
of the next few weeks.
Mark all uses of global variables to be virtualized
with a V_ prefix.
Use macros to map them back to their global names for
now, so this is a NOP change only.
We hope to have caught at least 85-90% of what is needed
so we do not invalidate a lot of outstanding patches again.
Obtained from: //depot/projects/vimage-commit2/...
Reviewed by: brooks, des, ed, mav, julian,
jamie, kris, rwatson, zec, ...
(various people I forgot, different versions)
md5 (with a bit of help)
Sponsored by: NLnet Foundation, The FreeBSD Foundation
X-MFC after: never
V_Commit_Message_Reviewed_By: more people than the patch
2008-08-17 23:27:27 +00:00
|
|
|
TAILQ_FOREACH(ia, &V_in_ifaddrhead, ia_link) {
|
2007-07-09 10:36:47 +00:00
|
|
|
mfp = ia->ia_ifp;
|
|
|
|
|
if (!(mfp->if_flags & IFF_LOOPBACK) &&
|
|
|
|
|
(mfp->if_flags & IFF_MULTICAST)) {
|
|
|
|
|
ifp = mfp;
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
}
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (bootverbose) {
|
|
|
|
|
printf("%s: imr_interface = %s, ifp = %p\n",
|
|
|
|
|
__func__, inet_ntoa(mreqs.imr_interface), ifp);
|
|
|
|
|
}
|
|
|
|
|
#endif
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
case MCAST_JOIN_GROUP:
|
|
|
|
|
case MCAST_JOIN_SOURCE_GROUP:
|
|
|
|
|
if (sopt->sopt_name == MCAST_JOIN_GROUP) {
|
|
|
|
|
error = sooptcopyin(sopt, &gsr,
|
|
|
|
|
sizeof(struct group_req),
|
|
|
|
|
sizeof(struct group_req));
|
|
|
|
|
} else if (sopt->sopt_name == MCAST_JOIN_SOURCE_GROUP) {
|
|
|
|
|
error = sooptcopyin(sopt, &gsr,
|
|
|
|
|
sizeof(struct group_source_req),
|
|
|
|
|
sizeof(struct group_source_req));
|
|
|
|
|
}
|
|
|
|
|
if (error)
|
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
|
|
if (gsa->sin.sin_family != AF_INET ||
|
|
|
|
|
gsa->sin.sin_len != sizeof(struct sockaddr_in))
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Overwrite the port field if present, as the sockaddr
|
|
|
|
|
* being copied in may be matched with a binary comparison.
|
|
|
|
|
* XXX INET6
|
|
|
|
|
*/
|
|
|
|
|
gsa->sin.sin_port = 0;
|
|
|
|
|
if (sopt->sopt_name == MCAST_JOIN_SOURCE_GROUP) {
|
|
|
|
|
if (ssa->sin.sin_family != AF_INET ||
|
|
|
|
|
ssa->sin.sin_len != sizeof(struct sockaddr_in))
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
ssa->sin.sin_port = 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Obtain the ifp.
|
|
|
|
|
*/
|
Commit step 1 of the vimage project, (network stack)
virtualization work done by Marko Zec (zec@).
This is the first in a series of commits over the course
of the next few weeks.
Mark all uses of global variables to be virtualized
with a V_ prefix.
Use macros to map them back to their global names for
now, so this is a NOP change only.
We hope to have caught at least 85-90% of what is needed
so we do not invalidate a lot of outstanding patches again.
Obtained from: //depot/projects/vimage-commit2/...
Reviewed by: brooks, des, ed, mav, julian,
jamie, kris, rwatson, zec, ...
(various people I forgot, different versions)
md5 (with a bit of help)
Sponsored by: NLnet Foundation, The FreeBSD Foundation
X-MFC after: never
V_Commit_Message_Reviewed_By: more people than the patch
2008-08-17 23:27:27 +00:00
|
|
|
if (gsr.gsr_interface == 0 || V_if_index < gsr.gsr_interface)
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
return (EADDRNOTAVAIL);
|
|
|
|
|
ifp = ifnet_byindex(gsr.gsr_interface);
|
|
|
|
|
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
default:
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (bootverbose) {
|
|
|
|
|
printf("%s: unknown sopt_name %d\n", __func__,
|
|
|
|
|
sopt->sopt_name);
|
|
|
|
|
}
|
|
|
|
|
#endif
|
|
|
|
|
return (EOPNOTSUPP);
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (!IN_MULTICAST(ntohl(gsa->sin.sin_addr.s_addr)))
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
|
|
|
|
|
if (ifp == NULL || (ifp->if_flags & IFF_MULTICAST) == 0)
|
|
|
|
|
return (EADDRNOTAVAIL);
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Check if we already hold membership of this group for this inpcb.
|
|
|
|
|
* If so, we do not need to perform the initial join.
|
|
|
|
|
*/
|
|
|
|
|
imo = inp_findmoptions(inp);
|
|
|
|
|
idx = imo_match_group(imo, ifp, &gsa->sa);
|
|
|
|
|
if (idx != -1) {
|
|
|
|
|
if (ssa->ss.ss_family != AF_UNSPEC) {
|
|
|
|
|
/*
|
|
|
|
|
* Attempting to join an ASM group (when already
|
|
|
|
|
* an ASM or SSM member) is an error.
|
|
|
|
|
*/
|
|
|
|
|
error = EADDRNOTAVAIL;
|
|
|
|
|
} else {
|
|
|
|
|
imf = &imo->imo_mfilters[idx];
|
|
|
|
|
if (imf->imf_nsources == 0) {
|
|
|
|
|
/*
|
|
|
|
|
* Attempting to join an SSM group (when
|
|
|
|
|
* already an ASM member) is an error.
|
|
|
|
|
*/
|
|
|
|
|
error = EINVAL;
|
|
|
|
|
} else {
|
|
|
|
|
/*
|
|
|
|
|
* Attempting to join an SSM group (when
|
|
|
|
|
* already an SSM member) means "add this
|
|
|
|
|
* source to the inclusive filter list".
|
|
|
|
|
*/
|
|
|
|
|
error = imo_join_source(imo, idx, ssa);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
goto out_locked;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Call imo_grow() to reallocate the membership and source filter
|
|
|
|
|
* vectors if they are full. If the size would exceed the hard limit,
|
|
|
|
|
* then we know we've really run out of entries. We keep the INP
|
|
|
|
|
* lock held to avoid introducing a race condition.
|
|
|
|
|
*/
|
|
|
|
|
if (imo->imo_num_memberships == imo->imo_max_memberships) {
|
|
|
|
|
error = imo_grow(imo);
|
|
|
|
|
if (error)
|
|
|
|
|
goto out_locked;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* So far, so good: perform the layer 3 join, layer 2 join,
|
|
|
|
|
* and make an IGMP announcement if needed.
|
|
|
|
|
*/
|
|
|
|
|
inm = in_addmulti(&gsa->sin.sin_addr, ifp);
|
|
|
|
|
if (inm == NULL) {
|
|
|
|
|
error = ENOBUFS;
|
|
|
|
|
goto out_locked;
|
|
|
|
|
}
|
|
|
|
|
idx = imo->imo_num_memberships;
|
|
|
|
|
imo->imo_membership[idx] = inm;
|
|
|
|
|
imo->imo_num_memberships++;
|
|
|
|
|
|
|
|
|
|
KASSERT(imo->imo_mfilters != NULL,
|
|
|
|
|
("%s: imf_mfilters vector was not allocated", __func__));
|
|
|
|
|
imf = &imo->imo_mfilters[idx];
|
|
|
|
|
KASSERT(TAILQ_EMPTY(&imf->imf_sources),
|
|
|
|
|
("%s: imf_sources not empty", __func__));
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* If this is a new SSM group join (i.e. a source was specified
|
|
|
|
|
* with this group), add this source to the filter list.
|
|
|
|
|
*/
|
|
|
|
|
if (ssa->ss.ss_family != AF_UNSPEC) {
|
|
|
|
|
/*
|
|
|
|
|
* An initial SSM join implies that this socket's membership
|
|
|
|
|
* of the multicast group is now in inclusive mode.
|
|
|
|
|
*/
|
|
|
|
|
imf->imf_fmode = MCAST_INCLUDE;
|
|
|
|
|
|
|
|
|
|
error = imo_join_source(imo, idx, ssa);
|
|
|
|
|
if (error) {
|
|
|
|
|
/*
|
|
|
|
|
* Drop inp lock before calling in_delmulti(),
|
|
|
|
|
* to prevent a lock order reversal.
|
|
|
|
|
*/
|
|
|
|
|
--imo->imo_num_memberships;
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
in_delmulti(inm);
|
|
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
out_locked:
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Leave an IPv4 multicast group on an inpcb, possibly with a source.
|
|
|
|
|
*/
|
|
|
|
|
static int
|
|
|
|
|
inp_leave_group(struct inpcb *inp, struct sockopt *sopt)
|
|
|
|
|
{
|
Step 1.5 of importing the network stack virtualization infrastructure
from the vimage project, as per plan established at devsummit 08/08:
http://wiki.freebsd.org/Image/Notes200808DevSummit
Introduce INIT_VNET_*() initializer macros, VNET_FOREACH() iterator
macros, and CURVNET_SET() context setting macros, all currently
resolving to NOPs.
Prepare for virtualization of selected SYSCTL objects by introducing a
family of SYSCTL_V_*() macros, currently resolving to their global
counterparts, i.e. SYSCTL_V_INT() == SYSCTL_INT().
Move selected #defines from sys/sys/vimage.h to newly introduced header
files specific to virtualized subsystems (sys/net/vnet.h,
sys/netinet/vinet.h etc.).
All the changes are verified to have zero functional impact at this
point in time by doing MD5 comparision between pre- and post-change
object files(*).
(*) netipsec/keysock.c did not validate depending on compile time options.
Implemented by: julian, bz, brooks, zec
Reviewed by: julian, bz, brooks, kris, rwatson, ...
Approved by: julian (mentor)
Obtained from: //depot/projects/vimage-commit2/...
X-MFC after: never
Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-10-02 15:37:58 +00:00
|
|
|
INIT_VNET_NET(curvnet);
|
|
|
|
|
INIT_VNET_INET(curvnet);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
struct group_source_req gsr;
|
|
|
|
|
struct ip_mreq_source mreqs;
|
|
|
|
|
sockunion_t *gsa, *ssa;
|
|
|
|
|
struct ifnet *ifp;
|
|
|
|
|
struct in_mfilter *imf;
|
|
|
|
|
struct ip_moptions *imo;
|
|
|
|
|
struct in_msource *ims, *tims;
|
|
|
|
|
struct in_multi *inm;
|
|
|
|
|
size_t idx;
|
|
|
|
|
int error;
|
|
|
|
|
|
|
|
|
|
ifp = NULL;
|
|
|
|
|
error = 0;
|
|
|
|
|
|
|
|
|
|
memset(&gsr, 0, sizeof(struct group_source_req));
|
|
|
|
|
gsa = (sockunion_t *)&gsr.gsr_group;
|
|
|
|
|
gsa->ss.ss_family = AF_UNSPEC;
|
|
|
|
|
ssa = (sockunion_t *)&gsr.gsr_source;
|
|
|
|
|
ssa->ss.ss_family = AF_UNSPEC;
|
|
|
|
|
|
|
|
|
|
switch (sopt->sopt_name) {
|
|
|
|
|
case IP_DROP_MEMBERSHIP:
|
|
|
|
|
case IP_DROP_SOURCE_MEMBERSHIP:
|
|
|
|
|
if (sopt->sopt_name == IP_DROP_MEMBERSHIP) {
|
|
|
|
|
error = sooptcopyin(sopt, &mreqs,
|
|
|
|
|
sizeof(struct ip_mreq),
|
|
|
|
|
sizeof(struct ip_mreq));
|
|
|
|
|
/*
|
|
|
|
|
* Swap interface and sourceaddr arguments,
|
|
|
|
|
* as ip_mreq and ip_mreq_source are laid
|
|
|
|
|
* out differently.
|
|
|
|
|
*/
|
|
|
|
|
mreqs.imr_interface = mreqs.imr_sourceaddr;
|
|
|
|
|
mreqs.imr_sourceaddr.s_addr = INADDR_ANY;
|
|
|
|
|
} else if (sopt->sopt_name == IP_DROP_SOURCE_MEMBERSHIP) {
|
|
|
|
|
error = sooptcopyin(sopt, &mreqs,
|
|
|
|
|
sizeof(struct ip_mreq_source),
|
|
|
|
|
sizeof(struct ip_mreq_source));
|
|
|
|
|
}
|
|
|
|
|
if (error)
|
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
|
|
gsa->sin.sin_family = AF_INET;
|
|
|
|
|
gsa->sin.sin_len = sizeof(struct sockaddr_in);
|
|
|
|
|
gsa->sin.sin_addr = mreqs.imr_multiaddr;
|
|
|
|
|
|
|
|
|
|
if (sopt->sopt_name == IP_DROP_SOURCE_MEMBERSHIP) {
|
|
|
|
|
ssa->sin.sin_family = AF_INET;
|
|
|
|
|
ssa->sin.sin_len = sizeof(struct sockaddr_in);
|
|
|
|
|
ssa->sin.sin_addr = mreqs.imr_sourceaddr;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (gsa->sin.sin_addr.s_addr != INADDR_ANY)
|
|
|
|
|
INADDR_TO_IFP(mreqs.imr_interface, ifp);
|
|
|
|
|
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (bootverbose) {
|
|
|
|
|
printf("%s: imr_interface = %s, ifp = %p\n",
|
|
|
|
|
__func__, inet_ntoa(mreqs.imr_interface), ifp);
|
|
|
|
|
}
|
|
|
|
|
#endif
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case MCAST_LEAVE_GROUP:
|
|
|
|
|
case MCAST_LEAVE_SOURCE_GROUP:
|
|
|
|
|
if (sopt->sopt_name == MCAST_LEAVE_GROUP) {
|
|
|
|
|
error = sooptcopyin(sopt, &gsr,
|
|
|
|
|
sizeof(struct group_req),
|
|
|
|
|
sizeof(struct group_req));
|
|
|
|
|
} else if (sopt->sopt_name == MCAST_LEAVE_SOURCE_GROUP) {
|
|
|
|
|
error = sooptcopyin(sopt, &gsr,
|
|
|
|
|
sizeof(struct group_source_req),
|
|
|
|
|
sizeof(struct group_source_req));
|
|
|
|
|
}
|
|
|
|
|
if (error)
|
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
|
|
if (gsa->sin.sin_family != AF_INET ||
|
|
|
|
|
gsa->sin.sin_len != sizeof(struct sockaddr_in))
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
|
|
|
|
|
if (sopt->sopt_name == MCAST_LEAVE_SOURCE_GROUP) {
|
|
|
|
|
if (ssa->sin.sin_family != AF_INET ||
|
|
|
|
|
ssa->sin.sin_len != sizeof(struct sockaddr_in))
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
}
|
|
|
|
|
|
Commit step 1 of the vimage project, (network stack)
virtualization work done by Marko Zec (zec@).
This is the first in a series of commits over the course
of the next few weeks.
Mark all uses of global variables to be virtualized
with a V_ prefix.
Use macros to map them back to their global names for
now, so this is a NOP change only.
We hope to have caught at least 85-90% of what is needed
so we do not invalidate a lot of outstanding patches again.
Obtained from: //depot/projects/vimage-commit2/...
Reviewed by: brooks, des, ed, mav, julian,
jamie, kris, rwatson, zec, ...
(various people I forgot, different versions)
md5 (with a bit of help)
Sponsored by: NLnet Foundation, The FreeBSD Foundation
X-MFC after: never
V_Commit_Message_Reviewed_By: more people than the patch
2008-08-17 23:27:27 +00:00
|
|
|
if (gsr.gsr_interface == 0 || V_if_index < gsr.gsr_interface)
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
return (EADDRNOTAVAIL);
|
|
|
|
|
|
|
|
|
|
ifp = ifnet_byindex(gsr.gsr_interface);
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
default:
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (bootverbose) {
|
|
|
|
|
printf("%s: unknown sopt_name %d\n", __func__,
|
|
|
|
|
sopt->sopt_name);
|
|
|
|
|
}
|
|
|
|
|
#endif
|
|
|
|
|
return (EOPNOTSUPP);
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (!IN_MULTICAST(ntohl(gsa->sin.sin_addr.s_addr)))
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Find the membership in the membership array.
|
|
|
|
|
*/
|
|
|
|
|
imo = inp_findmoptions(inp);
|
|
|
|
|
idx = imo_match_group(imo, ifp, &gsa->sa);
|
|
|
|
|
if (idx == -1) {
|
|
|
|
|
error = EADDRNOTAVAIL;
|
|
|
|
|
goto out_locked;
|
|
|
|
|
}
|
|
|
|
|
imf = &imo->imo_mfilters[idx];
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* If we were instructed only to leave a given source, do so.
|
|
|
|
|
*/
|
|
|
|
|
if (ssa->ss.ss_family != AF_UNSPEC) {
|
|
|
|
|
if (imf->imf_nsources == 0 ||
|
|
|
|
|
imf->imf_fmode == MCAST_EXCLUDE) {
|
|
|
|
|
/*
|
|
|
|
|
* Attempting to SSM leave an ASM group
|
|
|
|
|
* is an error; should use *_BLOCK_SOURCE instead.
|
|
|
|
|
* Attempting to SSM leave a source in a group when
|
|
|
|
|
* the socket is in 'exclude mode' is also an error.
|
|
|
|
|
*/
|
|
|
|
|
error = EINVAL;
|
|
|
|
|
} else {
|
|
|
|
|
error = imo_leave_source(imo, idx, ssa);
|
|
|
|
|
}
|
|
|
|
|
/*
|
|
|
|
|
* If an error occurred, or this source is not the last
|
|
|
|
|
* source in the group, do not leave the whole group.
|
|
|
|
|
*/
|
|
|
|
|
if (error || imf->imf_nsources > 0)
|
|
|
|
|
goto out_locked;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Give up the multicast address record to which the membership points.
|
|
|
|
|
*/
|
|
|
|
|
inm = imo->imo_membership[idx];
|
|
|
|
|
in_delmulti(inm);
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Free any source filters for this group if they exist.
|
|
|
|
|
* Revert inpcb to the default MCAST_EXCLUDE state.
|
|
|
|
|
*/
|
|
|
|
|
if (imo->imo_mfilters != NULL) {
|
|
|
|
|
TAILQ_FOREACH_SAFE(ims, &imf->imf_sources, ims_next, tims) {
|
|
|
|
|
TAILQ_REMOVE(&imf->imf_sources, ims, ims_next);
|
2008-10-23 15:53:51 +00:00
|
|
|
free(ims, M_IPMSOURCE);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
imf->imf_nsources--;
|
|
|
|
|
}
|
|
|
|
|
KASSERT(imf->imf_nsources == 0,
|
|
|
|
|
("%s: imf_nsources not 0", __func__));
|
|
|
|
|
KASSERT(TAILQ_EMPTY(&imf->imf_sources),
|
|
|
|
|
("%s: imf_sources not empty", __func__));
|
|
|
|
|
imf->imf_fmode = MCAST_EXCLUDE;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Remove the gap in the membership array.
|
|
|
|
|
*/
|
|
|
|
|
for (++idx; idx < imo->imo_num_memberships; ++idx)
|
|
|
|
|
imo->imo_membership[idx-1] = imo->imo_membership[idx];
|
|
|
|
|
imo->imo_num_memberships--;
|
|
|
|
|
|
|
|
|
|
out_locked:
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Select the interface for transmitting IPv4 multicast datagrams.
|
|
|
|
|
*
|
|
|
|
|
* Either an instance of struct in_addr or an instance of struct ip_mreqn
|
|
|
|
|
* may be passed to this socket option. An address of INADDR_ANY or an
|
|
|
|
|
* interface index of 0 is used to remove a previous selection.
|
|
|
|
|
* When no interface is selected, one is chosen for every send.
|
|
|
|
|
*/
|
|
|
|
|
static int
|
|
|
|
|
inp_set_multicast_if(struct inpcb *inp, struct sockopt *sopt)
|
|
|
|
|
{
|
Step 1.5 of importing the network stack virtualization infrastructure
from the vimage project, as per plan established at devsummit 08/08:
http://wiki.freebsd.org/Image/Notes200808DevSummit
Introduce INIT_VNET_*() initializer macros, VNET_FOREACH() iterator
macros, and CURVNET_SET() context setting macros, all currently
resolving to NOPs.
Prepare for virtualization of selected SYSCTL objects by introducing a
family of SYSCTL_V_*() macros, currently resolving to their global
counterparts, i.e. SYSCTL_V_INT() == SYSCTL_INT().
Move selected #defines from sys/sys/vimage.h to newly introduced header
files specific to virtualized subsystems (sys/net/vnet.h,
sys/netinet/vinet.h etc.).
All the changes are verified to have zero functional impact at this
point in time by doing MD5 comparision between pre- and post-change
object files(*).
(*) netipsec/keysock.c did not validate depending on compile time options.
Implemented by: julian, bz, brooks, zec
Reviewed by: julian, bz, brooks, kris, rwatson, ...
Approved by: julian (mentor)
Obtained from: //depot/projects/vimage-commit2/...
X-MFC after: never
Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-10-02 15:37:58 +00:00
|
|
|
INIT_VNET_NET(curvnet);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
struct in_addr addr;
|
|
|
|
|
struct ip_mreqn mreqn;
|
|
|
|
|
struct ifnet *ifp;
|
|
|
|
|
struct ip_moptions *imo;
|
|
|
|
|
int error;
|
|
|
|
|
|
|
|
|
|
if (sopt->sopt_valsize == sizeof(struct ip_mreqn)) {
|
|
|
|
|
/*
|
|
|
|
|
* An interface index was specified using the
|
|
|
|
|
* Linux-derived ip_mreqn structure.
|
|
|
|
|
*/
|
|
|
|
|
error = sooptcopyin(sopt, &mreqn, sizeof(struct ip_mreqn),
|
|
|
|
|
sizeof(struct ip_mreqn));
|
|
|
|
|
if (error)
|
|
|
|
|
return (error);
|
|
|
|
|
|
Commit step 1 of the vimage project, (network stack)
virtualization work done by Marko Zec (zec@).
This is the first in a series of commits over the course
of the next few weeks.
Mark all uses of global variables to be virtualized
with a V_ prefix.
Use macros to map them back to their global names for
now, so this is a NOP change only.
We hope to have caught at least 85-90% of what is needed
so we do not invalidate a lot of outstanding patches again.
Obtained from: //depot/projects/vimage-commit2/...
Reviewed by: brooks, des, ed, mav, julian,
jamie, kris, rwatson, zec, ...
(various people I forgot, different versions)
md5 (with a bit of help)
Sponsored by: NLnet Foundation, The FreeBSD Foundation
X-MFC after: never
V_Commit_Message_Reviewed_By: more people than the patch
2008-08-17 23:27:27 +00:00
|
|
|
if (mreqn.imr_ifindex < 0 || V_if_index < mreqn.imr_ifindex)
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
return (EINVAL);
|
|
|
|
|
|
|
|
|
|
if (mreqn.imr_ifindex == 0) {
|
|
|
|
|
ifp = NULL;
|
|
|
|
|
} else {
|
|
|
|
|
ifp = ifnet_byindex(mreqn.imr_ifindex);
|
|
|
|
|
if (ifp == NULL)
|
|
|
|
|
return (EADDRNOTAVAIL);
|
|
|
|
|
}
|
|
|
|
|
} else {
|
|
|
|
|
/*
|
|
|
|
|
* An interface was specified by IPv4 address.
|
|
|
|
|
* This is the traditional BSD usage.
|
|
|
|
|
*/
|
|
|
|
|
error = sooptcopyin(sopt, &addr, sizeof(struct in_addr),
|
|
|
|
|
sizeof(struct in_addr));
|
|
|
|
|
if (error)
|
|
|
|
|
return (error);
|
|
|
|
|
if (addr.s_addr == INADDR_ANY) {
|
|
|
|
|
ifp = NULL;
|
|
|
|
|
} else {
|
|
|
|
|
INADDR_TO_IFP(addr, ifp);
|
|
|
|
|
if (ifp == NULL)
|
|
|
|
|
return (EADDRNOTAVAIL);
|
|
|
|
|
}
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (bootverbose) {
|
|
|
|
|
printf("%s: ifp = %p, addr = %s\n",
|
|
|
|
|
__func__, ifp, inet_ntoa(addr)); /* XXX INET6 */
|
|
|
|
|
}
|
|
|
|
|
#endif
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* Reject interfaces which do not support multicast. */
|
|
|
|
|
if (ifp != NULL && (ifp->if_flags & IFF_MULTICAST) == 0)
|
|
|
|
|
return (EOPNOTSUPP);
|
|
|
|
|
|
|
|
|
|
imo = inp_findmoptions(inp);
|
|
|
|
|
imo->imo_multicast_ifp = ifp;
|
|
|
|
|
imo->imo_multicast_addr.s_addr = INADDR_ANY;
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
|
|
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Atomically set source filters on a socket for an IPv4 multicast group.
|
|
|
|
|
*/
|
|
|
|
|
static int
|
|
|
|
|
inp_set_source_filters(struct inpcb *inp, struct sockopt *sopt)
|
|
|
|
|
{
|
Step 1.5 of importing the network stack virtualization infrastructure
from the vimage project, as per plan established at devsummit 08/08:
http://wiki.freebsd.org/Image/Notes200808DevSummit
Introduce INIT_VNET_*() initializer macros, VNET_FOREACH() iterator
macros, and CURVNET_SET() context setting macros, all currently
resolving to NOPs.
Prepare for virtualization of selected SYSCTL objects by introducing a
family of SYSCTL_V_*() macros, currently resolving to their global
counterparts, i.e. SYSCTL_V_INT() == SYSCTL_INT().
Move selected #defines from sys/sys/vimage.h to newly introduced header
files specific to virtualized subsystems (sys/net/vnet.h,
sys/netinet/vinet.h etc.).
All the changes are verified to have zero functional impact at this
point in time by doing MD5 comparision between pre- and post-change
object files(*).
(*) netipsec/keysock.c did not validate depending on compile time options.
Implemented by: julian, bz, brooks, zec
Reviewed by: julian, bz, brooks, kris, rwatson, ...
Approved by: julian (mentor)
Obtained from: //depot/projects/vimage-commit2/...
X-MFC after: never
Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-10-02 15:37:58 +00:00
|
|
|
INIT_VNET_NET(curvnet);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
struct __msfilterreq msfr;
|
|
|
|
|
sockunion_t *gsa;
|
|
|
|
|
struct ifnet *ifp;
|
|
|
|
|
struct in_mfilter *imf;
|
|
|
|
|
struct ip_moptions *imo;
|
|
|
|
|
struct in_msource *ims, *tims;
|
|
|
|
|
size_t idx;
|
|
|
|
|
int error;
|
|
|
|
|
|
|
|
|
|
error = sooptcopyin(sopt, &msfr, sizeof(struct __msfilterreq),
|
|
|
|
|
sizeof(struct __msfilterreq));
|
|
|
|
|
if (error)
|
|
|
|
|
return (error);
|
|
|
|
|
|
|
|
|
|
if (msfr.msfr_nsrcs > IP_MAX_SOURCE_FILTER ||
|
|
|
|
|
(msfr.msfr_fmode != MCAST_EXCLUDE &&
|
|
|
|
|
msfr.msfr_fmode != MCAST_INCLUDE))
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
|
|
|
|
|
if (msfr.msfr_group.ss_family != AF_INET ||
|
|
|
|
|
msfr.msfr_group.ss_len != sizeof(struct sockaddr_in))
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
|
|
|
|
|
gsa = (sockunion_t *)&msfr.msfr_group;
|
|
|
|
|
if (!IN_MULTICAST(ntohl(gsa->sin.sin_addr.s_addr)))
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
|
|
|
|
|
gsa->sin.sin_port = 0; /* ignore port */
|
|
|
|
|
|
Commit step 1 of the vimage project, (network stack)
virtualization work done by Marko Zec (zec@).
This is the first in a series of commits over the course
of the next few weeks.
Mark all uses of global variables to be virtualized
with a V_ prefix.
Use macros to map them back to their global names for
now, so this is a NOP change only.
We hope to have caught at least 85-90% of what is needed
so we do not invalidate a lot of outstanding patches again.
Obtained from: //depot/projects/vimage-commit2/...
Reviewed by: brooks, des, ed, mav, julian,
jamie, kris, rwatson, zec, ...
(various people I forgot, different versions)
md5 (with a bit of help)
Sponsored by: NLnet Foundation, The FreeBSD Foundation
X-MFC after: never
V_Commit_Message_Reviewed_By: more people than the patch
2008-08-17 23:27:27 +00:00
|
|
|
if (msfr.msfr_ifindex == 0 || V_if_index < msfr.msfr_ifindex)
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
return (EADDRNOTAVAIL);
|
|
|
|
|
|
|
|
|
|
ifp = ifnet_byindex(msfr.msfr_ifindex);
|
|
|
|
|
if (ifp == NULL)
|
|
|
|
|
return (EADDRNOTAVAIL);
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Take the INP lock.
|
|
|
|
|
* Check if this socket is a member of this group.
|
|
|
|
|
*/
|
|
|
|
|
imo = inp_findmoptions(inp);
|
|
|
|
|
idx = imo_match_group(imo, ifp, &gsa->sa);
|
|
|
|
|
if (idx == -1 || imo->imo_mfilters == NULL) {
|
|
|
|
|
error = EADDRNOTAVAIL;
|
|
|
|
|
goto out_locked;
|
|
|
|
|
}
|
|
|
|
|
imf = &imo->imo_mfilters[idx];
|
|
|
|
|
|
|
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (bootverbose)
|
|
|
|
|
printf("%s: clearing source list\n", __func__);
|
|
|
|
|
#endif
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Remove any existing source filters.
|
|
|
|
|
*/
|
|
|
|
|
TAILQ_FOREACH_SAFE(ims, &imf->imf_sources, ims_next, tims) {
|
|
|
|
|
TAILQ_REMOVE(&imf->imf_sources, ims, ims_next);
|
2008-10-23 15:53:51 +00:00
|
|
|
free(ims, M_IPMSOURCE);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
imf->imf_nsources--;
|
|
|
|
|
}
|
|
|
|
|
KASSERT(imf->imf_nsources == 0,
|
|
|
|
|
("%s: source list not cleared", __func__));
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Apply any new source filters, if present.
|
|
|
|
|
*/
|
|
|
|
|
if (msfr.msfr_nsrcs > 0) {
|
|
|
|
|
struct in_msource **pnims;
|
|
|
|
|
struct in_msource *nims;
|
|
|
|
|
struct sockaddr_storage *kss;
|
|
|
|
|
struct sockaddr_storage *pkss;
|
|
|
|
|
sockunion_t *psu;
|
|
|
|
|
int i, j;
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Drop the inp lock so we may sleep if we need to
|
|
|
|
|
* in order to satisfy a malloc request.
|
|
|
|
|
* We will re-take it before changing socket state.
|
|
|
|
|
*/
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
#ifdef DIAGNOSTIC
|
|
|
|
|
if (bootverbose) {
|
|
|
|
|
printf("%s: loading %lu source list entries\n",
|
|
|
|
|
__func__, (unsigned long)msfr.msfr_nsrcs);
|
|
|
|
|
}
|
|
|
|
|
#endif
|
|
|
|
|
/*
|
|
|
|
|
* Make a copy of the user-space source vector so
|
|
|
|
|
* that we may copy them with a single copyin. This
|
|
|
|
|
* allows us to deal with page faults up-front.
|
|
|
|
|
*/
|
2008-10-23 20:26:15 +00:00
|
|
|
kss = malloc(sizeof(struct sockaddr_storage) * msfr.msfr_nsrcs,
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
M_TEMP, M_WAITOK);
|
|
|
|
|
error = copyin(msfr.msfr_srcs, kss,
|
|
|
|
|
sizeof(struct sockaddr_storage) * msfr.msfr_nsrcs);
|
|
|
|
|
if (error) {
|
2008-10-23 15:53:51 +00:00
|
|
|
free(kss, M_TEMP);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Perform argument checking on every sockaddr_storage
|
|
|
|
|
* structure in the vector provided to us. Overwrite
|
|
|
|
|
* fields which should not apply to source entries.
|
|
|
|
|
* TODO: Check for duplicate sources on this pass.
|
|
|
|
|
*/
|
|
|
|
|
psu = (sockunion_t *)kss;
|
|
|
|
|
for (i = 0; i < msfr.msfr_nsrcs; i++, psu++) {
|
|
|
|
|
switch (psu->ss.ss_family) {
|
|
|
|
|
case AF_INET:
|
|
|
|
|
if (psu->sin.sin_len !=
|
|
|
|
|
sizeof(struct sockaddr_in)) {
|
|
|
|
|
error = EINVAL;
|
|
|
|
|
} else {
|
|
|
|
|
psu->sin.sin_port = 0;
|
|
|
|
|
}
|
|
|
|
|
break;
|
|
|
|
|
#ifdef notyet
|
|
|
|
|
case AF_INET6;
|
|
|
|
|
if (psu->sin6.sin6_len !=
|
|
|
|
|
sizeof(struct sockaddr_in6)) {
|
|
|
|
|
error = EINVAL;
|
|
|
|
|
} else {
|
|
|
|
|
psu->sin6.sin6_port = 0;
|
|
|
|
|
psu->sin6.sin6_flowinfo = 0;
|
|
|
|
|
}
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
default:
|
|
|
|
|
error = EAFNOSUPPORT;
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
if (error)
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
if (error) {
|
2008-10-23 15:53:51 +00:00
|
|
|
free(kss, M_TEMP);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Allocate a block to track all the in_msource
|
|
|
|
|
* entries we are about to allocate, in case we
|
|
|
|
|
* abruptly need to free them.
|
|
|
|
|
*/
|
2008-10-23 20:26:15 +00:00
|
|
|
pnims = malloc(sizeof(struct in_msource *) * msfr.msfr_nsrcs,
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
M_TEMP, M_WAITOK | M_ZERO);
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Allocate up to nsrcs individual chunks.
|
|
|
|
|
* If we encounter an error, backtrack out of
|
|
|
|
|
* all allocations cleanly; updates must be atomic.
|
|
|
|
|
*/
|
|
|
|
|
pkss = kss;
|
|
|
|
|
nims = NULL;
|
|
|
|
|
for (i = 0; i < msfr.msfr_nsrcs; i++, pkss++) {
|
2008-10-23 20:26:15 +00:00
|
|
|
nims = malloc(sizeof(struct in_msource) *
|
|
|
|
|
msfr.msfr_nsrcs, M_IPMSOURCE, M_WAITOK | M_ZERO);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
pnims[i] = nims;
|
|
|
|
|
}
|
|
|
|
|
if (i < msfr.msfr_nsrcs) {
|
|
|
|
|
for (j = 0; j < i; j++) {
|
|
|
|
|
if (pnims[j] != NULL)
|
2008-10-23 15:53:51 +00:00
|
|
|
free(pnims[j], M_IPMSOURCE);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
}
|
2008-10-23 15:53:51 +00:00
|
|
|
free(pnims, M_TEMP);
|
|
|
|
|
free(kss, M_TEMP);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
return (ENOBUFS);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
INP_UNLOCK_ASSERT(inp);
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Finally, apply the filters to the socket.
|
|
|
|
|
* Re-take the inp lock; we are changing socket state.
|
|
|
|
|
*/
|
|
|
|
|
pkss = kss;
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
for (i = 0; i < msfr.msfr_nsrcs; i++, pkss++) {
|
|
|
|
|
memcpy(&(pnims[i]->ims_addr), pkss,
|
|
|
|
|
sizeof(struct sockaddr_storage));
|
|
|
|
|
TAILQ_INSERT_TAIL(&imf->imf_sources, pnims[i],
|
|
|
|
|
ims_next);
|
|
|
|
|
imf->imf_nsources++;
|
|
|
|
|
}
|
2008-10-23 15:53:51 +00:00
|
|
|
free(pnims, M_TEMP);
|
|
|
|
|
free(kss, M_TEMP);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Update the filter mode on the socket before releasing the inpcb.
|
|
|
|
|
*/
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK_ASSERT(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
imf->imf_fmode = msfr.msfr_fmode;
|
|
|
|
|
|
|
|
|
|
out_locked:
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Set the IP multicast options in response to user setsockopt().
|
|
|
|
|
*
|
|
|
|
|
* Many of the socket options handled in this function duplicate the
|
|
|
|
|
* functionality of socket options in the regular unicast API. However,
|
|
|
|
|
* it is not possible to merge the duplicate code, because the idempotence
|
|
|
|
|
* of the IPv4 multicast part of the BSD Sockets API must be preserved;
|
|
|
|
|
* the effects of these options must be treated as separate and distinct.
|
|
|
|
|
*/
|
|
|
|
|
int
|
|
|
|
|
inp_setmoptions(struct inpcb *inp, struct sockopt *sopt)
|
|
|
|
|
{
|
|
|
|
|
struct ip_moptions *imo;
|
|
|
|
|
int error;
|
|
|
|
|
|
|
|
|
|
error = 0;
|
|
|
|
|
|
Over the past couple of years, there have been a number of reports relating
the use of divert sockets to dead locks. A number of LORs have been reported
between divert and a number of other network subsystems including: IPSEC, Pfil,
multicast, ipfw and others. Other dead locks could occur because of recursive
entry into the IP stack. This change should take care of most if not all of
these issues.
A summary of the changes follow:
- We disallow multicast operations on divert sockets. It really doesn't make
semantic sense to allow this, since typically you would set multicast
parameters on multicast end points.
NOTE: As a part of this change, we actually dis-allow multicast options on
any socket that IS a divert socket OR IS NOT a SOCK_RAW or SOCK_DGRAM family
- We check to see if there are any socket options that have been specified on
the socket, and if there was (which is very un-common and also probably
doesnt make sense to support) we duplicate the mbuf carrying the options.
- We then drop the INP/INFO locks over the call to ip_output(). It should be
noted that since we no longer support multicast operations on divert sockets
and we have duplicated any socket options, we no longer need the reference
to the pcb to be coherent.
- Finally, we replaced the call to ip_input() to use netisr queuing. This
should remove the recursive entry into the IP stack from divert.
By dropping the locks over the call to ip_output() we eliminate all the lock
ordering issues above. By switching over to netisr on the inbound path,
we can no longer recursively enter the ip_input() code via divert.
I have tested this change by using the following command:
ipfwpcap -r 8000 - | tcpdump -r - -nn -v
This should exercise the input and re-injection (outbound) path, which is
very similar to the work load performed by natd(8). Additionally, I have
run some ospf daemons which have a heavy reliance on raw sockets and
multicast.
Approved by: re@ (kensmith)
MFC after: 1 month
LOR: 163
LOR: 181
LOR: 202
LOR: 203
Discussed with: julian, andre et al (on freebsd-net)
In collaboration with: bms [1], rwatson [2]
[1] bms helped out with the multicast decisions
[2] rwatson submitted the original netisr patches and came up with some
of the original ideas on how to combat this issue.
2007-08-06 22:06:36 +00:00
|
|
|
/*
|
|
|
|
|
* If socket is neither of type SOCK_RAW or SOCK_DGRAM,
|
|
|
|
|
* or is a divert socket, reject it.
|
|
|
|
|
* XXX Unlocked read of inp_socket believed OK.
|
|
|
|
|
*/
|
|
|
|
|
if (inp->inp_socket->so_proto->pr_protocol == IPPROTO_DIVERT ||
|
|
|
|
|
(inp->inp_socket->so_proto->pr_type != SOCK_RAW &&
|
|
|
|
|
inp->inp_socket->so_proto->pr_type != SOCK_DGRAM))
|
|
|
|
|
return (EOPNOTSUPP);
|
|
|
|
|
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
switch (sopt->sopt_name) {
|
|
|
|
|
case IP_MULTICAST_VIF: {
|
|
|
|
|
int vifi;
|
|
|
|
|
/*
|
|
|
|
|
* Select a multicast VIF for transmission.
|
|
|
|
|
* Only useful if multicast forwarding is active.
|
|
|
|
|
*/
|
|
|
|
|
if (legal_vif_num == NULL) {
|
|
|
|
|
error = EOPNOTSUPP;
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
error = sooptcopyin(sopt, &vifi, sizeof(int), sizeof(int));
|
|
|
|
|
if (error)
|
|
|
|
|
break;
|
|
|
|
|
if (!legal_vif_num(vifi) && (vifi != -1)) {
|
|
|
|
|
error = EINVAL;
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
imo = inp_findmoptions(inp);
|
|
|
|
|
imo->imo_multicast_vif = vifi;
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
case IP_MULTICAST_IF:
|
|
|
|
|
error = inp_set_multicast_if(inp, sopt);
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case IP_MULTICAST_TTL: {
|
|
|
|
|
u_char ttl;
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Set the IP time-to-live for outgoing multicast packets.
|
|
|
|
|
* The original multicast API required a char argument,
|
|
|
|
|
* which is inconsistent with the rest of the socket API.
|
|
|
|
|
* We allow either a char or an int.
|
|
|
|
|
*/
|
|
|
|
|
if (sopt->sopt_valsize == sizeof(u_char)) {
|
|
|
|
|
error = sooptcopyin(sopt, &ttl, sizeof(u_char),
|
|
|
|
|
sizeof(u_char));
|
|
|
|
|
if (error)
|
|
|
|
|
break;
|
|
|
|
|
} else {
|
|
|
|
|
u_int ittl;
|
|
|
|
|
|
|
|
|
|
error = sooptcopyin(sopt, &ittl, sizeof(u_int),
|
|
|
|
|
sizeof(u_int));
|
|
|
|
|
if (error)
|
|
|
|
|
break;
|
|
|
|
|
if (ittl > 255) {
|
|
|
|
|
error = EINVAL;
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
ttl = (u_char)ittl;
|
|
|
|
|
}
|
|
|
|
|
imo = inp_findmoptions(inp);
|
|
|
|
|
imo->imo_multicast_ttl = ttl;
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
case IP_MULTICAST_LOOP: {
|
|
|
|
|
u_char loop;
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Set the loopback flag for outgoing multicast packets.
|
|
|
|
|
* Must be zero or one. The original multicast API required a
|
|
|
|
|
* char argument, which is inconsistent with the rest
|
|
|
|
|
* of the socket API. We allow either a char or an int.
|
|
|
|
|
*/
|
|
|
|
|
if (sopt->sopt_valsize == sizeof(u_char)) {
|
|
|
|
|
error = sooptcopyin(sopt, &loop, sizeof(u_char),
|
|
|
|
|
sizeof(u_char));
|
|
|
|
|
if (error)
|
|
|
|
|
break;
|
|
|
|
|
} else {
|
|
|
|
|
u_int iloop;
|
|
|
|
|
|
|
|
|
|
error = sooptcopyin(sopt, &iloop, sizeof(u_int),
|
|
|
|
|
sizeof(u_int));
|
|
|
|
|
if (error)
|
|
|
|
|
break;
|
|
|
|
|
loop = (u_char)iloop;
|
|
|
|
|
}
|
|
|
|
|
imo = inp_findmoptions(inp);
|
|
|
|
|
imo->imo_multicast_loop = !!loop;
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
Import rewrite of IPv4 socket multicast layer to support source-specific
and protocol-independent host mode multicast. The code is written to
accomodate IPv6, IGMPv3 and MLDv2 with only a little additional work.
This change only pertains to FreeBSD's use as a multicast end-station and
does not concern multicast routing; for an IGMPv3/MLDv2 router
implementation, consider the XORP project.
The work is based on Wilbert de Graaf's IGMPv3 code drop for FreeBSD 4.6,
which is available at: http://www.kloosterhof.com/wilbert/igmpv3.html
Summary
* IPv4 multicast socket processing is now moved out of ip_output.c
into a new module, in_mcast.c.
* The in_mcast.c module implements the IPv4 legacy any-source API in
terms of the protocol-independent source-specific API.
* Source filters are lazy allocated as the common case does not use them.
They are part of per inpcb state and are covered by the inpcb lock.
* struct ip_mreqn is now supported to allow applications to specify
multicast joins by interface index in the legacy IPv4 any-source API.
* In UDP, an incoming multicast datagram only requires that the source
port matches the 4-tuple if the socket was already bound by source port.
An unbound socket SHOULD be able to receive multicasts sent from an
ephemeral source port.
* The UDP socket multicast filter mode defaults to exclusive, that is,
sources present in the per-socket list will be blocked from delivery.
* The RFC 3678 userland functions have been added to libc: setsourcefilter,
getsourcefilter, setipv4sourcefilter, getipv4sourcefilter.
* Definitions for IGMPv3 are merged but not yet used.
* struct sockaddr_storage is now referenced from <netinet/in.h>. It
is therefore defined there if not already declared in the same way
as for the C99 types.
* The RFC 1724 hack (specify 0.0.0.0/8 addresses to IP_MULTICAST_IF
which are then interpreted as interface indexes) is now deprecated.
* A patch for the Rhyolite.com routed in the FreeBSD base system
is available in the -net archives. This only affects individuals
running RIPv1 or RIPv2 via point-to-point and/or unnumbered interfaces.
* Make IPv6 detach path similar to IPv4's in code flow; functionally same.
* Bump __FreeBSD_version to 700048; see UPDATING.
This work was financially supported by another FreeBSD committer.
Obtained from: p4://bms_netdev
Submitted by: Wilbert de Graaf (original work)
Reviewed by: rwatson (locking), silence from fenner,
net@ (but with encouragement)
2007-06-12 16:24:56 +00:00
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
case IP_ADD_MEMBERSHIP:
|
|
|
|
|
case IP_ADD_SOURCE_MEMBERSHIP:
|
|
|
|
|
case MCAST_JOIN_GROUP:
|
|
|
|
|
case MCAST_JOIN_SOURCE_GROUP:
|
|
|
|
|
error = inp_join_group(inp, sopt);
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case IP_DROP_MEMBERSHIP:
|
|
|
|
|
case IP_DROP_SOURCE_MEMBERSHIP:
|
|
|
|
|
case MCAST_LEAVE_GROUP:
|
|
|
|
|
case MCAST_LEAVE_SOURCE_GROUP:
|
|
|
|
|
error = inp_leave_group(inp, sopt);
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case IP_BLOCK_SOURCE:
|
|
|
|
|
case IP_UNBLOCK_SOURCE:
|
|
|
|
|
case MCAST_BLOCK_SOURCE:
|
|
|
|
|
case MCAST_UNBLOCK_SOURCE:
|
|
|
|
|
error = inp_change_source_filter(inp, sopt);
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case IP_MSFILTER:
|
|
|
|
|
error = inp_set_source_filters(inp, sopt);
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
default:
|
|
|
|
|
error = EOPNOTSUPP;
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
INP_UNLOCK_ASSERT(inp);
|
|
|
|
|
|
|
|
|
|
return (error);
|
|
|
|
|
}
|