2005-01-07 02:30:35 +00:00
|
|
|
/*-
|
2017-11-20 19:43:44 +00:00
|
|
|
* SPDX-License-Identifier: BSD-3-Clause
|
|
|
|
*
|
1999-11-22 02:45:11 +00:00
|
|
|
* Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project.
|
|
|
|
* All rights reserved.
|
|
|
|
*
|
|
|
|
* Redistribution and use in source and binary forms, with or without
|
|
|
|
* modification, are permitted provided that the following conditions
|
|
|
|
* are met:
|
|
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
|
|
* notice, this list of conditions and the following disclaimer.
|
|
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
|
|
* documentation and/or other materials provided with the distribution.
|
|
|
|
* 3. Neither the name of the project nor the names of its contributors
|
|
|
|
* may be used to endorse or promote products derived from this software
|
|
|
|
* without specific prior written permission.
|
|
|
|
*
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE PROJECT AND CONTRIBUTORS ``AS IS'' AND
|
|
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE PROJECT OR CONTRIBUTORS BE LIABLE
|
|
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
|
|
* SUCH DAMAGE.
|
|
|
|
*/
|
|
|
|
|
2005-01-07 02:30:35 +00:00
|
|
|
/*-
|
1999-11-22 02:45:11 +00:00
|
|
|
* Copyright (c) 1982, 1986, 1988, 1993
|
2008-07-05 18:03:39 +00:00
|
|
|
* The Regents of the University of California.
|
|
|
|
* All rights reserved.
|
1999-11-22 02:45:11 +00:00
|
|
|
*
|
|
|
|
* 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.
|
2017-02-28 23:42:47 +00:00
|
|
|
* 3. Neither the name of the University nor the names of its contributors
|
1999-11-22 02:45:11 +00:00
|
|
|
* may be used to endorse or promote products derived from this software
|
|
|
|
* without specific prior written permission.
|
|
|
|
*
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
|
|
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
|
|
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
|
|
* SUCH DAMAGE.
|
|
|
|
*
|
|
|
|
* @(#)raw_ip.c 8.2 (Berkeley) 1/4/94
|
|
|
|
*/
|
|
|
|
|
2007-12-10 16:03:40 +00:00
|
|
|
#include <sys/cdefs.h>
|
|
|
|
__FBSDID("$FreeBSD$");
|
|
|
|
|
1999-12-22 19:13:38 +00:00
|
|
|
#include "opt_ipsec.h"
|
2001-06-11 12:39:29 +00:00
|
|
|
#include "opt_inet6.h"
|
1999-12-22 19:13:38 +00:00
|
|
|
|
1999-11-22 02:45:11 +00:00
|
|
|
#include <sys/param.h>
|
2002-04-30 01:54:54 +00:00
|
|
|
#include <sys/errno.h>
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
#include <sys/jail.h>
|
2013-07-09 09:54:54 +00:00
|
|
|
#include <sys/kernel.h>
|
2002-04-30 01:54:54 +00:00
|
|
|
#include <sys/lock.h>
|
1999-11-22 02:45:11 +00:00
|
|
|
#include <sys/malloc.h>
|
|
|
|
#include <sys/mbuf.h>
|
2007-06-13 22:42:43 +00:00
|
|
|
#include <sys/priv.h>
|
2002-04-30 01:54:54 +00:00
|
|
|
#include <sys/proc.h>
|
1999-11-22 02:45:11 +00:00
|
|
|
#include <sys/protosw.h>
|
2002-04-30 01:54:54 +00:00
|
|
|
#include <sys/signalvar.h>
|
|
|
|
#include <sys/socket.h>
|
1999-11-22 02:45:11 +00:00
|
|
|
#include <sys/socketvar.h>
|
2002-04-30 01:54:54 +00:00
|
|
|
#include <sys/sx.h>
|
2005-07-25 12:31:43 +00:00
|
|
|
#include <sys/syslog.h>
|
1999-11-22 02:45:11 +00:00
|
|
|
|
|
|
|
#include <net/if.h>
|
2013-10-26 17:58:36 +00:00
|
|
|
#include <net/if_var.h>
|
1999-11-22 02:45:11 +00:00
|
|
|
#include <net/if_types.h>
|
2002-04-30 01:54:54 +00:00
|
|
|
#include <net/route.h>
|
2008-12-02 21:37:28 +00:00
|
|
|
#include <net/vnet.h>
|
1999-11-22 02:45:11 +00:00
|
|
|
|
|
|
|
#include <netinet/in.h>
|
|
|
|
#include <netinet/in_var.h>
|
|
|
|
#include <netinet/in_systm.h>
|
|
|
|
#include <netinet/in_pcb.h>
|
2008-12-02 21:37:28 +00:00
|
|
|
|
|
|
|
#include <netinet/icmp6.h>
|
2002-04-30 01:54:54 +00:00
|
|
|
#include <netinet/ip6.h>
|
2010-08-19 11:31:03 +00:00
|
|
|
#include <netinet/ip_var.h>
|
2002-04-30 01:54:54 +00:00
|
|
|
#include <netinet6/ip6protosw.h>
|
|
|
|
#include <netinet6/ip6_mroute.h>
|
1999-11-22 02:45:11 +00:00
|
|
|
#include <netinet6/in6_pcb.h>
|
2002-04-30 01:54:54 +00:00
|
|
|
#include <netinet6/ip6_var.h>
|
1999-11-22 02:45:11 +00:00
|
|
|
#include <netinet6/nd6.h>
|
2002-04-30 01:54:54 +00:00
|
|
|
#include <netinet6/raw_ip6.h>
|
2000-07-04 16:35:15 +00:00
|
|
|
#include <netinet6/scope6_var.h>
|
2010-08-19 11:31:03 +00:00
|
|
|
#include <netinet6/send.h>
|
1999-11-22 02:45:11 +00:00
|
|
|
|
2017-02-06 08:49:57 +00:00
|
|
|
#include <netipsec/ipsec_support.h>
|
2002-10-16 02:25:05 +00:00
|
|
|
|
1999-11-22 02:45:11 +00:00
|
|
|
#include <machine/stdarg.h>
|
|
|
|
|
|
|
|
#define satosin6(sa) ((struct sockaddr_in6 *)(sa))
|
|
|
|
#define ifatoia6(ifa) ((struct in6_ifaddr *)(ifa))
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Raw interface to IP6 protocol.
|
|
|
|
*/
|
|
|
|
|
Build on Jeff Roberson's linker-set based dynamic per-CPU allocator
(DPCPU), as suggested by Peter Wemm, and implement a new per-virtual
network stack memory allocator. Modify vnet to use the allocator
instead of monolithic global container structures (vinet, ...). This
change solves many binary compatibility problems associated with
VIMAGE, and restores ELF symbols for virtualized global variables.
Each virtualized global variable exists as a "reference copy", and also
once per virtual network stack. Virtualized global variables are
tagged at compile-time, placing the in a special linker set, which is
loaded into a contiguous region of kernel memory. Virtualized global
variables in the base kernel are linked as normal, but those in modules
are copied and relocated to a reserved portion of the kernel's vnet
region with the help of a the kernel linker.
Virtualized global variables exist in per-vnet memory set up when the
network stack instance is created, and are initialized statically from
the reference copy. Run-time access occurs via an accessor macro, which
converts from the current vnet and requested symbol to a per-vnet
address. When "options VIMAGE" is not compiled into the kernel, normal
global ELF symbols will be used instead and indirection is avoided.
This change restores static initialization for network stack global
variables, restores support for non-global symbols and types, eliminates
the need for many subsystem constructors, eliminates large per-subsystem
structures that caused many binary compatibility issues both for
monitoring applications (netstat) and kernel modules, removes the
per-function INIT_VNET_*() macros throughout the stack, eliminates the
need for vnet_symmap ksym(2) munging, and eliminates duplicate
definitions of virtualized globals under VIMAGE_GLOBALS.
Bump __FreeBSD_version and update UPDATING.
Portions submitted by: bz
Reviewed by: bz, zec
Discussed with: gnn, jamie, jeff, jhb, julian, sam
Suggested by: peter
Approved by: re (kensmith)
2009-07-14 22:48:30 +00:00
|
|
|
VNET_DECLARE(struct inpcbhead, ripcb);
|
|
|
|
VNET_DECLARE(struct inpcbinfo, ripcbinfo);
|
2009-07-16 21:13:04 +00:00
|
|
|
#define V_ripcb VNET(ripcb)
|
|
|
|
#define V_ripcbinfo VNET(ripcbinfo)
|
Build on Jeff Roberson's linker-set based dynamic per-CPU allocator
(DPCPU), as suggested by Peter Wemm, and implement a new per-virtual
network stack memory allocator. Modify vnet to use the allocator
instead of monolithic global container structures (vinet, ...). This
change solves many binary compatibility problems associated with
VIMAGE, and restores ELF symbols for virtualized global variables.
Each virtualized global variable exists as a "reference copy", and also
once per virtual network stack. Virtualized global variables are
tagged at compile-time, placing the in a special linker set, which is
loaded into a contiguous region of kernel memory. Virtualized global
variables in the base kernel are linked as normal, but those in modules
are copied and relocated to a reserved portion of the kernel's vnet
region with the help of a the kernel linker.
Virtualized global variables exist in per-vnet memory set up when the
network stack instance is created, and are initialized statically from
the reference copy. Run-time access occurs via an accessor macro, which
converts from the current vnet and requested symbol to a per-vnet
address. When "options VIMAGE" is not compiled into the kernel, normal
global ELF symbols will be used instead and indirection is avoided.
This change restores static initialization for network stack global
variables, restores support for non-global symbols and types, eliminates
the need for many subsystem constructors, eliminates large per-subsystem
structures that caused many binary compatibility issues both for
monitoring applications (netstat) and kernel modules, removes the
per-function INIT_VNET_*() macros throughout the stack, eliminates the
need for vnet_symmap ksym(2) munging, and eliminates duplicate
definitions of virtualized globals under VIMAGE_GLOBALS.
Bump __FreeBSD_version and update UPDATING.
Portions submitted by: bz
Reviewed by: bz, zec
Discussed with: gnn, jamie, jeff, jhb, julian, sam
Suggested by: peter
Approved by: re (kensmith)
2009-07-14 22:48:30 +00:00
|
|
|
|
2008-11-26 22:32:07 +00:00
|
|
|
extern u_long rip_sendspace;
|
|
|
|
extern u_long rip_recvspace;
|
|
|
|
|
2013-07-09 09:54:54 +00:00
|
|
|
VNET_PCPUSTAT_DEFINE(struct rip6stat, rip6stat);
|
|
|
|
VNET_PCPUSTAT_SYSINIT(rip6stat);
|
|
|
|
|
|
|
|
#ifdef VIMAGE
|
|
|
|
VNET_PCPUSTAT_SYSUNINIT(rip6stat);
|
|
|
|
#endif /* VIMAGE */
|
2010-04-29 11:52:42 +00:00
|
|
|
|
2007-02-24 11:38:47 +00:00
|
|
|
/*
|
Bite the bullet, and make the IPv6 SSM and MLDv2 mega-commit:
import from p4 bms_netdev. Summary of changes:
* Connect netinet6/in6_mcast.c to build.
The legacy KAME KPIs are mostly preserved.
* Eliminate now dead code from ip6_output.c.
Don't do mbuf bingo, we are not going to do RFC 2292 style
CMSG tricks for multicast options as they are not required
by any current IPv6 normative reference.
* Refactor transports (UDP, raw_ip6) to do own mcast filtering.
SCTP, TCP unaffected by this change.
* Add ip6_msource, in6_msource structs to in6_var.h.
* Hookup mld_ifinfo state to in6_ifextra, allocate from
domifattach path.
* Eliminate IN6_LOOKUP_MULTI(), it is no longer referenced.
Kernel consumers which need this should use in6m_lookup().
* Refactor IPv6 socket group memberships to use a vector (like IPv4).
* Update ifmcstat(8) for IPv6 SSM.
* Add witness lock order for IN6_MULTI_LOCK.
* Move IN6_MULTI_LOCK out of lower ip6_output()/ip6_input() paths.
* Introduce IP6STAT_ADD/SUB/INC/DEC as per rwatson's IPv4 cleanup.
* Update carp(4) for new IPv6 SSM KPIs.
* Virtualize ip6_mrouter socket.
Changes mostly localized to IPv6 MROUTING.
* Don't do a local group lookup in MROUTING.
* Kill unused KAME prototypes in6_purgemkludge(), in6_restoremkludge().
* Preserve KAME DAD timer jitter behaviour in MLDv1 compatibility mode.
* Bump __FreeBSD_version to 800084.
* Update UPDATING.
NOTE WELL:
* This code hasn't been tested against real MLDv2 queriers
(yet), although the on-wire protocol has been verified in Wireshark.
* There are a few unresolved issues in the socket layer APIs to
do with scope ID propagation.
* There is a LOR present in ip6_output()'s use of
in6_setscope() which needs to be resolved. See comments in mld6.c.
This is believed to be benign and can't be avoided for the moment
without re-introducing an indirect netisr.
This work was mostly derived from the IGMPv3 implementation, and
has been sponsored by a third party.
2009-04-29 19:19:13 +00:00
|
|
|
* Hooks for multicast routing. They all default to NULL, so leave them not
|
|
|
|
* initialized and rely on BSS being set to 0.
|
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The socket used to communicate with the multicast routing daemon.
|
|
|
|
*/
|
Build on Jeff Roberson's linker-set based dynamic per-CPU allocator
(DPCPU), as suggested by Peter Wemm, and implement a new per-virtual
network stack memory allocator. Modify vnet to use the allocator
instead of monolithic global container structures (vinet, ...). This
change solves many binary compatibility problems associated with
VIMAGE, and restores ELF symbols for virtualized global variables.
Each virtualized global variable exists as a "reference copy", and also
once per virtual network stack. Virtualized global variables are
tagged at compile-time, placing the in a special linker set, which is
loaded into a contiguous region of kernel memory. Virtualized global
variables in the base kernel are linked as normal, but those in modules
are copied and relocated to a reserved portion of the kernel's vnet
region with the help of a the kernel linker.
Virtualized global variables exist in per-vnet memory set up when the
network stack instance is created, and are initialized statically from
the reference copy. Run-time access occurs via an accessor macro, which
converts from the current vnet and requested symbol to a per-vnet
address. When "options VIMAGE" is not compiled into the kernel, normal
global ELF symbols will be used instead and indirection is avoided.
This change restores static initialization for network stack global
variables, restores support for non-global symbols and types, eliminates
the need for many subsystem constructors, eliminates large per-subsystem
structures that caused many binary compatibility issues both for
monitoring applications (netstat) and kernel modules, removes the
per-function INIT_VNET_*() macros throughout the stack, eliminates the
need for vnet_symmap ksym(2) munging, and eliminates duplicate
definitions of virtualized globals under VIMAGE_GLOBALS.
Bump __FreeBSD_version and update UPDATING.
Portions submitted by: bz
Reviewed by: bz, zec
Discussed with: gnn, jamie, jeff, jhb, julian, sam
Suggested by: peter
Approved by: re (kensmith)
2009-07-14 22:48:30 +00:00
|
|
|
VNET_DEFINE(struct socket *, ip6_mrouter);
|
Bite the bullet, and make the IPv6 SSM and MLDv2 mega-commit:
import from p4 bms_netdev. Summary of changes:
* Connect netinet6/in6_mcast.c to build.
The legacy KAME KPIs are mostly preserved.
* Eliminate now dead code from ip6_output.c.
Don't do mbuf bingo, we are not going to do RFC 2292 style
CMSG tricks for multicast options as they are not required
by any current IPv6 normative reference.
* Refactor transports (UDP, raw_ip6) to do own mcast filtering.
SCTP, TCP unaffected by this change.
* Add ip6_msource, in6_msource structs to in6_var.h.
* Hookup mld_ifinfo state to in6_ifextra, allocate from
domifattach path.
* Eliminate IN6_LOOKUP_MULTI(), it is no longer referenced.
Kernel consumers which need this should use in6m_lookup().
* Refactor IPv6 socket group memberships to use a vector (like IPv4).
* Update ifmcstat(8) for IPv6 SSM.
* Add witness lock order for IN6_MULTI_LOCK.
* Move IN6_MULTI_LOCK out of lower ip6_output()/ip6_input() paths.
* Introduce IP6STAT_ADD/SUB/INC/DEC as per rwatson's IPv4 cleanup.
* Update carp(4) for new IPv6 SSM KPIs.
* Virtualize ip6_mrouter socket.
Changes mostly localized to IPv6 MROUTING.
* Don't do a local group lookup in MROUTING.
* Kill unused KAME prototypes in6_purgemkludge(), in6_restoremkludge().
* Preserve KAME DAD timer jitter behaviour in MLDv1 compatibility mode.
* Bump __FreeBSD_version to 800084.
* Update UPDATING.
NOTE WELL:
* This code hasn't been tested against real MLDv2 queriers
(yet), although the on-wire protocol has been verified in Wireshark.
* There are a few unresolved issues in the socket layer APIs to
do with scope ID propagation.
* There is a LOR present in ip6_output()'s use of
in6_setscope() which needs to be resolved. See comments in mld6.c.
This is believed to be benign and can't be avoided for the moment
without re-introducing an indirect netisr.
This work was mostly derived from the IGMPv3 implementation, and
has been sponsored by a third party.
2009-04-29 19:19:13 +00:00
|
|
|
|
|
|
|
/*
|
|
|
|
* The various mrouter functions.
|
2007-02-24 11:38:47 +00:00
|
|
|
*/
|
|
|
|
int (*ip6_mrouter_set)(struct socket *, struct sockopt *);
|
|
|
|
int (*ip6_mrouter_get)(struct socket *, struct sockopt *);
|
|
|
|
int (*ip6_mrouter_done)(void);
|
|
|
|
int (*ip6_mforward)(struct ip6_hdr *, struct ifnet *, struct mbuf *);
|
2009-06-21 10:29:31 +00:00
|
|
|
int (*mrt6_ioctl)(u_long, caddr_t);
|
2007-02-24 11:38:47 +00:00
|
|
|
|
1999-11-22 02:45:11 +00:00
|
|
|
/*
|
2008-07-05 18:03:39 +00:00
|
|
|
* Setup generic address and protocol structures for raw_input routine, then
|
|
|
|
* pass them along with mbuf chain.
|
1999-11-22 02:45:11 +00:00
|
|
|
*/
|
|
|
|
int
|
2007-07-05 16:23:49 +00:00
|
|
|
rip6_input(struct mbuf **mp, int *offp, int proto)
|
1999-11-22 02:45:11 +00:00
|
|
|
{
|
Bite the bullet, and make the IPv6 SSM and MLDv2 mega-commit:
import from p4 bms_netdev. Summary of changes:
* Connect netinet6/in6_mcast.c to build.
The legacy KAME KPIs are mostly preserved.
* Eliminate now dead code from ip6_output.c.
Don't do mbuf bingo, we are not going to do RFC 2292 style
CMSG tricks for multicast options as they are not required
by any current IPv6 normative reference.
* Refactor transports (UDP, raw_ip6) to do own mcast filtering.
SCTP, TCP unaffected by this change.
* Add ip6_msource, in6_msource structs to in6_var.h.
* Hookup mld_ifinfo state to in6_ifextra, allocate from
domifattach path.
* Eliminate IN6_LOOKUP_MULTI(), it is no longer referenced.
Kernel consumers which need this should use in6m_lookup().
* Refactor IPv6 socket group memberships to use a vector (like IPv4).
* Update ifmcstat(8) for IPv6 SSM.
* Add witness lock order for IN6_MULTI_LOCK.
* Move IN6_MULTI_LOCK out of lower ip6_output()/ip6_input() paths.
* Introduce IP6STAT_ADD/SUB/INC/DEC as per rwatson's IPv4 cleanup.
* Update carp(4) for new IPv6 SSM KPIs.
* Virtualize ip6_mrouter socket.
Changes mostly localized to IPv6 MROUTING.
* Don't do a local group lookup in MROUTING.
* Kill unused KAME prototypes in6_purgemkludge(), in6_restoremkludge().
* Preserve KAME DAD timer jitter behaviour in MLDv1 compatibility mode.
* Bump __FreeBSD_version to 800084.
* Update UPDATING.
NOTE WELL:
* This code hasn't been tested against real MLDv2 queriers
(yet), although the on-wire protocol has been verified in Wireshark.
* There are a few unresolved issues in the socket layer APIs to
do with scope ID propagation.
* There is a LOR present in ip6_output()'s use of
in6_setscope() which needs to be resolved. See comments in mld6.c.
This is believed to be benign and can't be avoided for the moment
without re-introducing an indirect netisr.
This work was mostly derived from the IGMPv3 implementation, and
has been sponsored by a third party.
2009-04-29 19:19:13 +00:00
|
|
|
struct ifnet *ifp;
|
1999-11-22 02:45:11 +00:00
|
|
|
struct mbuf *m = *mp;
|
2017-05-17 00:34:34 +00:00
|
|
|
struct ip6_hdr *ip6 = mtod(m, struct ip6_hdr *);
|
|
|
|
struct inpcb *in6p;
|
2016-04-15 17:30:33 +00:00
|
|
|
struct inpcb *last = NULL;
|
2001-06-11 12:39:29 +00:00
|
|
|
struct mbuf *opts = NULL;
|
2003-11-02 19:09:29 +00:00
|
|
|
struct sockaddr_in6 fromsa;
|
2018-07-04 02:47:16 +00:00
|
|
|
struct epoch_tracker et;
|
1999-11-22 02:45:11 +00:00
|
|
|
|
2013-06-19 20:48:34 +00:00
|
|
|
RIP6STAT_INC(rip6s_ipackets);
|
2001-06-11 12:39:29 +00:00
|
|
|
|
2017-03-06 04:01:58 +00:00
|
|
|
init_sin6(&fromsa, m, 0); /* general init */
|
1999-11-22 02:45:11 +00:00
|
|
|
|
Bite the bullet, and make the IPv6 SSM and MLDv2 mega-commit:
import from p4 bms_netdev. Summary of changes:
* Connect netinet6/in6_mcast.c to build.
The legacy KAME KPIs are mostly preserved.
* Eliminate now dead code from ip6_output.c.
Don't do mbuf bingo, we are not going to do RFC 2292 style
CMSG tricks for multicast options as they are not required
by any current IPv6 normative reference.
* Refactor transports (UDP, raw_ip6) to do own mcast filtering.
SCTP, TCP unaffected by this change.
* Add ip6_msource, in6_msource structs to in6_var.h.
* Hookup mld_ifinfo state to in6_ifextra, allocate from
domifattach path.
* Eliminate IN6_LOOKUP_MULTI(), it is no longer referenced.
Kernel consumers which need this should use in6m_lookup().
* Refactor IPv6 socket group memberships to use a vector (like IPv4).
* Update ifmcstat(8) for IPv6 SSM.
* Add witness lock order for IN6_MULTI_LOCK.
* Move IN6_MULTI_LOCK out of lower ip6_output()/ip6_input() paths.
* Introduce IP6STAT_ADD/SUB/INC/DEC as per rwatson's IPv4 cleanup.
* Update carp(4) for new IPv6 SSM KPIs.
* Virtualize ip6_mrouter socket.
Changes mostly localized to IPv6 MROUTING.
* Don't do a local group lookup in MROUTING.
* Kill unused KAME prototypes in6_purgemkludge(), in6_restoremkludge().
* Preserve KAME DAD timer jitter behaviour in MLDv1 compatibility mode.
* Bump __FreeBSD_version to 800084.
* Update UPDATING.
NOTE WELL:
* This code hasn't been tested against real MLDv2 queriers
(yet), although the on-wire protocol has been verified in Wireshark.
* There are a few unresolved issues in the socket layer APIs to
do with scope ID propagation.
* There is a LOR present in ip6_output()'s use of
in6_setscope() which needs to be resolved. See comments in mld6.c.
This is believed to be benign and can't be avoided for the moment
without re-introducing an indirect netisr.
This work was mostly derived from the IGMPv3 implementation, and
has been sponsored by a third party.
2009-04-29 19:19:13 +00:00
|
|
|
ifp = m->m_pkthdr.rcvif;
|
|
|
|
|
2018-07-04 02:47:16 +00:00
|
|
|
INP_INFO_RLOCK_ET(&V_ripcbinfo, et);
|
2018-06-12 22:18:20 +00:00
|
|
|
CK_LIST_FOREACH(in6p, &V_ripcb, inp_list) {
|
MFp4:
Bring in updated jail support from bz_jail branch.
This enhances the current jail implementation to permit multiple
addresses per jail. In addtion to IPv4, IPv6 is supported as well.
Due to updated checks it is even possible to have jails without
an IP address at all, which basically gives one a chroot with
restricted process view, no networking,..
SCTP support was updated and supports IPv6 in jails as well.
Cpuset support permits jails to be bound to specific processor
sets after creation.
Jails can have an unrestricted (no duplicate protection, etc.) name
in addition to the hostname. The jail name cannot be changed from
within a jail and is considered to be used for management purposes
or as audit-token in the future.
DDB 'show jails' command was added to aid debugging.
Proper compat support permits 32bit jail binaries to be used on 64bit
systems to manage jails. Also backward compatibility was preserved where
possible: for jail v1 syscalls, as well as with user space management
utilities.
Both jail as well as prison version were updated for the new features.
A gap was intentionally left as the intermediate versions had been
used by various patches floating around the last years.
Bump __FreeBSD_version for the afore mentioned and in kernel changes.
Special thanks to:
- Pawel Jakub Dawidek (pjd) for his multi-IPv4 patches
and Olivier Houchard (cognet) for initial single-IPv6 patches.
- Jeff Roberson (jeff) and Randall Stewart (rrs) for their
help, ideas and review on cpuset and SCTP support.
- Robert Watson (rwatson) for lots and lots of help, discussions,
suggestions and review of most of the patch at various stages.
- John Baldwin (jhb) for his help.
- Simon L. Nielsen (simon) as early adopter testing changes
on cluster machines as well as all the testers and people
who provided feedback the last months on freebsd-jail and
other channels.
- My employer, CK Software GmbH, for the support so I could work on this.
Reviewed by: (see above)
MFC after: 3 months (this is just so that I get the mail)
X-MFC Before: 7.2-RELEASE if possible
2008-11-29 14:32:14 +00:00
|
|
|
/* XXX inp locking */
|
Another step assimilating IPv[46] PCB code - directly use
the inpcb names rather than the following IPv6 compat macros:
in6pcb,in6p_sp, in6p_ip6_nxt,in6p_flowinfo,in6p_vflag,
in6p_flags,in6p_socket,in6p_lport,in6p_fport,in6p_ppcb and
sotoin6pcb().
Apart from removing duplicate code in netipsec, this is a pure
whitespace, not a functional change.
Discussed with: rwatson
Reviewed by: rwatson (version before review requested changes)
MFC after: 4 weeks (set the timer and see then)
2008-12-15 21:50:54 +00:00
|
|
|
if ((in6p->inp_vflag & INP_IPV6) == 0)
|
1999-11-22 02:45:11 +00:00
|
|
|
continue;
|
Another step assimilating IPv[46] PCB code - directly use
the inpcb names rather than the following IPv6 compat macros:
in6pcb,in6p_sp, in6p_ip6_nxt,in6p_flowinfo,in6p_vflag,
in6p_flags,in6p_socket,in6p_lport,in6p_fport,in6p_ppcb and
sotoin6pcb().
Apart from removing duplicate code in netipsec, this is a pure
whitespace, not a functional change.
Discussed with: rwatson
Reviewed by: rwatson (version before review requested changes)
MFC after: 4 weeks (set the timer and see then)
2008-12-15 21:50:54 +00:00
|
|
|
if (in6p->inp_ip_p &&
|
|
|
|
in6p->inp_ip_p != proto)
|
2008-07-27 20:30:34 +00:00
|
|
|
continue;
|
1999-11-22 02:45:11 +00:00
|
|
|
if (!IN6_IS_ADDR_UNSPECIFIED(&in6p->in6p_laddr) &&
|
|
|
|
!IN6_ARE_ADDR_EQUAL(&in6p->in6p_laddr, &ip6->ip6_dst))
|
2008-07-27 20:30:34 +00:00
|
|
|
continue;
|
1999-11-22 02:45:11 +00:00
|
|
|
if (!IN6_IS_ADDR_UNSPECIFIED(&in6p->in6p_faddr) &&
|
|
|
|
!IN6_ARE_ADDR_EQUAL(&in6p->in6p_faddr, &ip6->ip6_src))
|
2008-07-27 20:30:34 +00:00
|
|
|
continue;
|
2018-10-24 10:42:35 +00:00
|
|
|
if (last != NULL) {
|
|
|
|
struct mbuf *n = m_copym(m, 0, M_COPYALL, M_NOWAIT);
|
|
|
|
|
|
|
|
#if defined(IPSEC) || defined(IPSEC_SUPPORT)
|
|
|
|
/*
|
|
|
|
* Check AH/ESP integrity.
|
|
|
|
*/
|
|
|
|
if (IPSEC_ENABLED(ipv6)) {
|
|
|
|
if (n != NULL &&
|
|
|
|
IPSEC_CHECK_POLICY(ipv6, n, last) != 0) {
|
|
|
|
m_freem(n);
|
|
|
|
/* Do not inject data into pcb. */
|
|
|
|
n = NULL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
#endif /* IPSEC */
|
|
|
|
if (n) {
|
|
|
|
if (last->inp_flags & INP_CONTROLOPTS ||
|
|
|
|
last->inp_socket->so_options & SO_TIMESTAMP)
|
|
|
|
ip6_savecontrol(last, n, &opts);
|
|
|
|
/* strip intermediate headers */
|
|
|
|
m_adj(n, *offp);
|
|
|
|
if (sbappendaddr(&last->inp_socket->so_rcv,
|
|
|
|
(struct sockaddr *)&fromsa,
|
|
|
|
n, opts) == 0) {
|
|
|
|
m_freem(n);
|
|
|
|
if (opts)
|
|
|
|
m_freem(opts);
|
|
|
|
RIP6STAT_INC(rip6s_fullsock);
|
|
|
|
} else
|
|
|
|
sorwakeup(last->inp_socket);
|
|
|
|
opts = NULL;
|
|
|
|
}
|
|
|
|
INP_RUNLOCK(last);
|
|
|
|
last = NULL;
|
|
|
|
}
|
|
|
|
INP_RLOCK(in6p);
|
|
|
|
if (__predict_false(in6p->inp_flags2 & INP_FREED))
|
|
|
|
goto skip_2;
|
2009-12-13 13:57:32 +00:00
|
|
|
if (jailed_without_vnet(in6p->inp_cred)) {
|
Bite the bullet, and make the IPv6 SSM and MLDv2 mega-commit:
import from p4 bms_netdev. Summary of changes:
* Connect netinet6/in6_mcast.c to build.
The legacy KAME KPIs are mostly preserved.
* Eliminate now dead code from ip6_output.c.
Don't do mbuf bingo, we are not going to do RFC 2292 style
CMSG tricks for multicast options as they are not required
by any current IPv6 normative reference.
* Refactor transports (UDP, raw_ip6) to do own mcast filtering.
SCTP, TCP unaffected by this change.
* Add ip6_msource, in6_msource structs to in6_var.h.
* Hookup mld_ifinfo state to in6_ifextra, allocate from
domifattach path.
* Eliminate IN6_LOOKUP_MULTI(), it is no longer referenced.
Kernel consumers which need this should use in6m_lookup().
* Refactor IPv6 socket group memberships to use a vector (like IPv4).
* Update ifmcstat(8) for IPv6 SSM.
* Add witness lock order for IN6_MULTI_LOCK.
* Move IN6_MULTI_LOCK out of lower ip6_output()/ip6_input() paths.
* Introduce IP6STAT_ADD/SUB/INC/DEC as per rwatson's IPv4 cleanup.
* Update carp(4) for new IPv6 SSM KPIs.
* Virtualize ip6_mrouter socket.
Changes mostly localized to IPv6 MROUTING.
* Don't do a local group lookup in MROUTING.
* Kill unused KAME prototypes in6_purgemkludge(), in6_restoremkludge().
* Preserve KAME DAD timer jitter behaviour in MLDv1 compatibility mode.
* Bump __FreeBSD_version to 800084.
* Update UPDATING.
NOTE WELL:
* This code hasn't been tested against real MLDv2 queriers
(yet), although the on-wire protocol has been verified in Wireshark.
* There are a few unresolved issues in the socket layer APIs to
do with scope ID propagation.
* There is a LOR present in ip6_output()'s use of
in6_setscope() which needs to be resolved. See comments in mld6.c.
This is believed to be benign and can't be avoided for the moment
without re-introducing an indirect netisr.
This work was mostly derived from the IGMPv3 implementation, and
has been sponsored by a third party.
2009-04-29 19:19:13 +00:00
|
|
|
/*
|
|
|
|
* Allow raw socket in jail to receive multicast;
|
|
|
|
* assume process had PRIV_NETINET_RAW at attach,
|
|
|
|
* and fall through into normal filter path if so.
|
|
|
|
*/
|
|
|
|
if (!IN6_IS_ADDR_MULTICAST(&ip6->ip6_dst) &&
|
|
|
|
prison_check_ip6(in6p->inp_cred,
|
|
|
|
&ip6->ip6_dst) != 0)
|
2018-10-24 10:42:35 +00:00
|
|
|
goto skip_2;
|
Bite the bullet, and make the IPv6 SSM and MLDv2 mega-commit:
import from p4 bms_netdev. Summary of changes:
* Connect netinet6/in6_mcast.c to build.
The legacy KAME KPIs are mostly preserved.
* Eliminate now dead code from ip6_output.c.
Don't do mbuf bingo, we are not going to do RFC 2292 style
CMSG tricks for multicast options as they are not required
by any current IPv6 normative reference.
* Refactor transports (UDP, raw_ip6) to do own mcast filtering.
SCTP, TCP unaffected by this change.
* Add ip6_msource, in6_msource structs to in6_var.h.
* Hookup mld_ifinfo state to in6_ifextra, allocate from
domifattach path.
* Eliminate IN6_LOOKUP_MULTI(), it is no longer referenced.
Kernel consumers which need this should use in6m_lookup().
* Refactor IPv6 socket group memberships to use a vector (like IPv4).
* Update ifmcstat(8) for IPv6 SSM.
* Add witness lock order for IN6_MULTI_LOCK.
* Move IN6_MULTI_LOCK out of lower ip6_output()/ip6_input() paths.
* Introduce IP6STAT_ADD/SUB/INC/DEC as per rwatson's IPv4 cleanup.
* Update carp(4) for new IPv6 SSM KPIs.
* Virtualize ip6_mrouter socket.
Changes mostly localized to IPv6 MROUTING.
* Don't do a local group lookup in MROUTING.
* Kill unused KAME prototypes in6_purgemkludge(), in6_restoremkludge().
* Preserve KAME DAD timer jitter behaviour in MLDv1 compatibility mode.
* Bump __FreeBSD_version to 800084.
* Update UPDATING.
NOTE WELL:
* This code hasn't been tested against real MLDv2 queriers
(yet), although the on-wire protocol has been verified in Wireshark.
* There are a few unresolved issues in the socket layer APIs to
do with scope ID propagation.
* There is a LOR present in ip6_output()'s use of
in6_setscope() which needs to be resolved. See comments in mld6.c.
This is believed to be benign and can't be avoided for the moment
without re-introducing an indirect netisr.
This work was mostly derived from the IGMPv3 implementation, and
has been sponsored by a third party.
2009-04-29 19:19:13 +00:00
|
|
|
}
|
2001-06-11 12:39:29 +00:00
|
|
|
if (in6p->in6p_cksum != -1) {
|
2013-06-19 20:48:34 +00:00
|
|
|
RIP6STAT_INC(rip6s_isum);
|
2005-10-19 01:21:49 +00:00
|
|
|
if (in6_cksum(m, proto, *offp,
|
2001-06-11 12:39:29 +00:00
|
|
|
m->m_pkthdr.len - *offp)) {
|
2013-06-19 20:48:34 +00:00
|
|
|
RIP6STAT_INC(rip6s_badsum);
|
2018-10-24 10:42:35 +00:00
|
|
|
goto skip_2;
|
2001-06-11 12:39:29 +00:00
|
|
|
}
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
Bite the bullet, and make the IPv6 SSM and MLDv2 mega-commit:
import from p4 bms_netdev. Summary of changes:
* Connect netinet6/in6_mcast.c to build.
The legacy KAME KPIs are mostly preserved.
* Eliminate now dead code from ip6_output.c.
Don't do mbuf bingo, we are not going to do RFC 2292 style
CMSG tricks for multicast options as they are not required
by any current IPv6 normative reference.
* Refactor transports (UDP, raw_ip6) to do own mcast filtering.
SCTP, TCP unaffected by this change.
* Add ip6_msource, in6_msource structs to in6_var.h.
* Hookup mld_ifinfo state to in6_ifextra, allocate from
domifattach path.
* Eliminate IN6_LOOKUP_MULTI(), it is no longer referenced.
Kernel consumers which need this should use in6m_lookup().
* Refactor IPv6 socket group memberships to use a vector (like IPv4).
* Update ifmcstat(8) for IPv6 SSM.
* Add witness lock order for IN6_MULTI_LOCK.
* Move IN6_MULTI_LOCK out of lower ip6_output()/ip6_input() paths.
* Introduce IP6STAT_ADD/SUB/INC/DEC as per rwatson's IPv4 cleanup.
* Update carp(4) for new IPv6 SSM KPIs.
* Virtualize ip6_mrouter socket.
Changes mostly localized to IPv6 MROUTING.
* Don't do a local group lookup in MROUTING.
* Kill unused KAME prototypes in6_purgemkludge(), in6_restoremkludge().
* Preserve KAME DAD timer jitter behaviour in MLDv1 compatibility mode.
* Bump __FreeBSD_version to 800084.
* Update UPDATING.
NOTE WELL:
* This code hasn't been tested against real MLDv2 queriers
(yet), although the on-wire protocol has been verified in Wireshark.
* There are a few unresolved issues in the socket layer APIs to
do with scope ID propagation.
* There is a LOR present in ip6_output()'s use of
in6_setscope() which needs to be resolved. See comments in mld6.c.
This is believed to be benign and can't be avoided for the moment
without re-introducing an indirect netisr.
This work was mostly derived from the IGMPv3 implementation, and
has been sponsored by a third party.
2009-04-29 19:19:13 +00:00
|
|
|
/*
|
|
|
|
* If this raw socket has multicast state, and we
|
|
|
|
* have received a multicast, check if this socket
|
|
|
|
* should receive it, as multicast filtering is now
|
|
|
|
* the responsibility of the transport layer.
|
|
|
|
*/
|
|
|
|
if (in6p->in6p_moptions &&
|
|
|
|
IN6_IS_ADDR_MULTICAST(&ip6->ip6_dst)) {
|
2009-11-19 11:55:19 +00:00
|
|
|
/*
|
|
|
|
* If the incoming datagram is for MLD, allow it
|
|
|
|
* through unconditionally to the raw socket.
|
|
|
|
*
|
|
|
|
* Use the M_RTALERT_MLD flag to check for MLD
|
|
|
|
* traffic without having to inspect the mbuf chain
|
|
|
|
* more deeply, as all MLDv1/v2 host messages MUST
|
|
|
|
* contain the Router Alert option.
|
|
|
|
*
|
|
|
|
* In the case of MLDv1, we may not have explicitly
|
|
|
|
* joined the group, and may have set IFF_ALLMULTI
|
|
|
|
* on the interface. im6o_mc_filter() may discard
|
|
|
|
* control traffic we actually need to see.
|
|
|
|
*
|
|
|
|
* Userland multicast routing daemons should continue
|
|
|
|
* filter the control traffic appropriately.
|
|
|
|
*/
|
Bite the bullet, and make the IPv6 SSM and MLDv2 mega-commit:
import from p4 bms_netdev. Summary of changes:
* Connect netinet6/in6_mcast.c to build.
The legacy KAME KPIs are mostly preserved.
* Eliminate now dead code from ip6_output.c.
Don't do mbuf bingo, we are not going to do RFC 2292 style
CMSG tricks for multicast options as they are not required
by any current IPv6 normative reference.
* Refactor transports (UDP, raw_ip6) to do own mcast filtering.
SCTP, TCP unaffected by this change.
* Add ip6_msource, in6_msource structs to in6_var.h.
* Hookup mld_ifinfo state to in6_ifextra, allocate from
domifattach path.
* Eliminate IN6_LOOKUP_MULTI(), it is no longer referenced.
Kernel consumers which need this should use in6m_lookup().
* Refactor IPv6 socket group memberships to use a vector (like IPv4).
* Update ifmcstat(8) for IPv6 SSM.
* Add witness lock order for IN6_MULTI_LOCK.
* Move IN6_MULTI_LOCK out of lower ip6_output()/ip6_input() paths.
* Introduce IP6STAT_ADD/SUB/INC/DEC as per rwatson's IPv4 cleanup.
* Update carp(4) for new IPv6 SSM KPIs.
* Virtualize ip6_mrouter socket.
Changes mostly localized to IPv6 MROUTING.
* Don't do a local group lookup in MROUTING.
* Kill unused KAME prototypes in6_purgemkludge(), in6_restoremkludge().
* Preserve KAME DAD timer jitter behaviour in MLDv1 compatibility mode.
* Bump __FreeBSD_version to 800084.
* Update UPDATING.
NOTE WELL:
* This code hasn't been tested against real MLDv2 queriers
(yet), although the on-wire protocol has been verified in Wireshark.
* There are a few unresolved issues in the socket layer APIs to
do with scope ID propagation.
* There is a LOR present in ip6_output()'s use of
in6_setscope() which needs to be resolved. See comments in mld6.c.
This is believed to be benign and can't be avoided for the moment
without re-introducing an indirect netisr.
This work was mostly derived from the IGMPv3 implementation, and
has been sponsored by a third party.
2009-04-29 19:19:13 +00:00
|
|
|
int blocked;
|
|
|
|
|
2009-11-19 11:55:19 +00:00
|
|
|
blocked = MCAST_PASS;
|
|
|
|
if ((m->m_flags & M_RTALERT_MLD) == 0) {
|
|
|
|
struct sockaddr_in6 mcaddr;
|
Bite the bullet, and make the IPv6 SSM and MLDv2 mega-commit:
import from p4 bms_netdev. Summary of changes:
* Connect netinet6/in6_mcast.c to build.
The legacy KAME KPIs are mostly preserved.
* Eliminate now dead code from ip6_output.c.
Don't do mbuf bingo, we are not going to do RFC 2292 style
CMSG tricks for multicast options as they are not required
by any current IPv6 normative reference.
* Refactor transports (UDP, raw_ip6) to do own mcast filtering.
SCTP, TCP unaffected by this change.
* Add ip6_msource, in6_msource structs to in6_var.h.
* Hookup mld_ifinfo state to in6_ifextra, allocate from
domifattach path.
* Eliminate IN6_LOOKUP_MULTI(), it is no longer referenced.
Kernel consumers which need this should use in6m_lookup().
* Refactor IPv6 socket group memberships to use a vector (like IPv4).
* Update ifmcstat(8) for IPv6 SSM.
* Add witness lock order for IN6_MULTI_LOCK.
* Move IN6_MULTI_LOCK out of lower ip6_output()/ip6_input() paths.
* Introduce IP6STAT_ADD/SUB/INC/DEC as per rwatson's IPv4 cleanup.
* Update carp(4) for new IPv6 SSM KPIs.
* Virtualize ip6_mrouter socket.
Changes mostly localized to IPv6 MROUTING.
* Don't do a local group lookup in MROUTING.
* Kill unused KAME prototypes in6_purgemkludge(), in6_restoremkludge().
* Preserve KAME DAD timer jitter behaviour in MLDv1 compatibility mode.
* Bump __FreeBSD_version to 800084.
* Update UPDATING.
NOTE WELL:
* This code hasn't been tested against real MLDv2 queriers
(yet), although the on-wire protocol has been verified in Wireshark.
* There are a few unresolved issues in the socket layer APIs to
do with scope ID propagation.
* There is a LOR present in ip6_output()'s use of
in6_setscope() which needs to be resolved. See comments in mld6.c.
This is believed to be benign and can't be avoided for the moment
without re-introducing an indirect netisr.
This work was mostly derived from the IGMPv3 implementation, and
has been sponsored by a third party.
2009-04-29 19:19:13 +00:00
|
|
|
|
2009-11-19 11:55:19 +00:00
|
|
|
bzero(&mcaddr, sizeof(struct sockaddr_in6));
|
|
|
|
mcaddr.sin6_len = sizeof(struct sockaddr_in6);
|
|
|
|
mcaddr.sin6_family = AF_INET6;
|
|
|
|
mcaddr.sin6_addr = ip6->ip6_dst;
|
|
|
|
|
|
|
|
blocked = im6o_mc_filter(in6p->in6p_moptions,
|
|
|
|
ifp,
|
|
|
|
(struct sockaddr *)&mcaddr,
|
|
|
|
(struct sockaddr *)&fromsa);
|
|
|
|
}
|
Bite the bullet, and make the IPv6 SSM and MLDv2 mega-commit:
import from p4 bms_netdev. Summary of changes:
* Connect netinet6/in6_mcast.c to build.
The legacy KAME KPIs are mostly preserved.
* Eliminate now dead code from ip6_output.c.
Don't do mbuf bingo, we are not going to do RFC 2292 style
CMSG tricks for multicast options as they are not required
by any current IPv6 normative reference.
* Refactor transports (UDP, raw_ip6) to do own mcast filtering.
SCTP, TCP unaffected by this change.
* Add ip6_msource, in6_msource structs to in6_var.h.
* Hookup mld_ifinfo state to in6_ifextra, allocate from
domifattach path.
* Eliminate IN6_LOOKUP_MULTI(), it is no longer referenced.
Kernel consumers which need this should use in6m_lookup().
* Refactor IPv6 socket group memberships to use a vector (like IPv4).
* Update ifmcstat(8) for IPv6 SSM.
* Add witness lock order for IN6_MULTI_LOCK.
* Move IN6_MULTI_LOCK out of lower ip6_output()/ip6_input() paths.
* Introduce IP6STAT_ADD/SUB/INC/DEC as per rwatson's IPv4 cleanup.
* Update carp(4) for new IPv6 SSM KPIs.
* Virtualize ip6_mrouter socket.
Changes mostly localized to IPv6 MROUTING.
* Don't do a local group lookup in MROUTING.
* Kill unused KAME prototypes in6_purgemkludge(), in6_restoremkludge().
* Preserve KAME DAD timer jitter behaviour in MLDv1 compatibility mode.
* Bump __FreeBSD_version to 800084.
* Update UPDATING.
NOTE WELL:
* This code hasn't been tested against real MLDv2 queriers
(yet), although the on-wire protocol has been verified in Wireshark.
* There are a few unresolved issues in the socket layer APIs to
do with scope ID propagation.
* There is a LOR present in ip6_output()'s use of
in6_setscope() which needs to be resolved. See comments in mld6.c.
This is believed to be benign and can't be avoided for the moment
without re-introducing an indirect netisr.
This work was mostly derived from the IGMPv3 implementation, and
has been sponsored by a third party.
2009-04-29 19:19:13 +00:00
|
|
|
if (blocked != MCAST_PASS) {
|
|
|
|
IP6STAT_INC(ip6s_notmember);
|
2018-10-24 10:42:35 +00:00
|
|
|
goto skip_2;
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
last = in6p;
|
2018-10-24 10:42:35 +00:00
|
|
|
continue;
|
|
|
|
skip_2:
|
|
|
|
INP_RUNLOCK(in6p);
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
2018-07-04 02:47:16 +00:00
|
|
|
INP_INFO_RUNLOCK_ET(&V_ripcbinfo, et);
|
2017-02-06 08:49:57 +00:00
|
|
|
#if defined(IPSEC) || defined(IPSEC_SUPPORT)
|
2001-06-11 12:39:29 +00:00
|
|
|
/*
|
|
|
|
* Check AH/ESP integrity.
|
|
|
|
*/
|
2017-02-06 08:49:57 +00:00
|
|
|
if (IPSEC_ENABLED(ipv6) && last != NULL &&
|
|
|
|
IPSEC_CHECK_POLICY(ipv6, m, last) != 0) {
|
2001-06-11 12:39:29 +00:00
|
|
|
m_freem(m);
|
2013-04-09 07:11:22 +00:00
|
|
|
IP6STAT_DEC(ip6s_delivered);
|
2008-07-05 18:03:39 +00:00
|
|
|
/* Do not inject data into pcb. */
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RUNLOCK(last);
|
2002-10-16 02:25:05 +00:00
|
|
|
} else
|
2007-07-03 12:13:45 +00:00
|
|
|
#endif /* IPSEC */
|
2008-12-16 03:12:44 +00:00
|
|
|
if (last != NULL) {
|
2008-12-17 13:00:18 +00:00
|
|
|
if (last->inp_flags & INP_CONTROLOPTS ||
|
Another step assimilating IPv[46] PCB code - directly use
the inpcb names rather than the following IPv6 compat macros:
in6pcb,in6p_sp, in6p_ip6_nxt,in6p_flowinfo,in6p_vflag,
in6p_flags,in6p_socket,in6p_lport,in6p_fport,in6p_ppcb and
sotoin6pcb().
Apart from removing duplicate code in netipsec, this is a pure
whitespace, not a functional change.
Discussed with: rwatson
Reviewed by: rwatson (version before review requested changes)
MFC after: 4 weeks (set the timer and see then)
2008-12-15 21:50:54 +00:00
|
|
|
last->inp_socket->so_options & SO_TIMESTAMP)
|
2003-10-29 12:52:28 +00:00
|
|
|
ip6_savecontrol(last, m, &opts);
|
2008-07-05 18:03:39 +00:00
|
|
|
/* Strip intermediate headers. */
|
1999-11-22 02:45:11 +00:00
|
|
|
m_adj(m, *offp);
|
Another step assimilating IPv[46] PCB code - directly use
the inpcb names rather than the following IPv6 compat macros:
in6pcb,in6p_sp, in6p_ip6_nxt,in6p_flowinfo,in6p_vflag,
in6p_flags,in6p_socket,in6p_lport,in6p_fport,in6p_ppcb and
sotoin6pcb().
Apart from removing duplicate code in netipsec, this is a pure
whitespace, not a functional change.
Discussed with: rwatson
Reviewed by: rwatson (version before review requested changes)
MFC after: 4 weeks (set the timer and see then)
2008-12-15 21:50:54 +00:00
|
|
|
if (sbappendaddr(&last->inp_socket->so_rcv,
|
2008-07-05 18:03:39 +00:00
|
|
|
(struct sockaddr *)&fromsa, m, opts) == 0) {
|
1999-11-22 02:45:11 +00:00
|
|
|
m_freem(m);
|
|
|
|
if (opts)
|
|
|
|
m_freem(opts);
|
2013-06-19 20:48:34 +00:00
|
|
|
RIP6STAT_INC(rip6s_fullsock);
|
2002-05-31 11:52:35 +00:00
|
|
|
} else
|
Another step assimilating IPv[46] PCB code - directly use
the inpcb names rather than the following IPv6 compat macros:
in6pcb,in6p_sp, in6p_ip6_nxt,in6p_flowinfo,in6p_vflag,
in6p_flags,in6p_socket,in6p_lport,in6p_fport,in6p_ppcb and
sotoin6pcb().
Apart from removing duplicate code in netipsec, this is a pure
whitespace, not a functional change.
Discussed with: rwatson
Reviewed by: rwatson (version before review requested changes)
MFC after: 4 weeks (set the timer and see then)
2008-12-15 21:50:54 +00:00
|
|
|
sorwakeup(last->inp_socket);
|
2008-04-21 12:06:41 +00:00
|
|
|
INP_RUNLOCK(last);
|
1999-11-22 02:45:11 +00:00
|
|
|
} else {
|
2013-06-19 20:48:34 +00:00
|
|
|
RIP6STAT_INC(rip6s_nosock);
|
2001-06-11 12:39:29 +00:00
|
|
|
if (m->m_flags & M_MCAST)
|
2013-06-19 20:48:34 +00:00
|
|
|
RIP6STAT_INC(rip6s_nosockmcast);
|
1999-11-22 02:45:11 +00:00
|
|
|
if (proto == IPPROTO_NONE)
|
|
|
|
m_freem(m);
|
2018-02-05 09:22:07 +00:00
|
|
|
else
|
1999-11-22 02:45:11 +00:00
|
|
|
icmp6_error(m, ICMP6_PARAM_PROB,
|
2008-07-05 18:03:39 +00:00
|
|
|
ICMP6_PARAMPROB_NEXTHEADER,
|
2018-02-05 09:22:07 +00:00
|
|
|
ip6_get_prevhdr(m, *offp));
|
2013-04-09 07:11:22 +00:00
|
|
|
IP6STAT_DEC(ip6s_delivered);
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
2008-07-05 18:03:39 +00:00
|
|
|
return (IPPROTO_DONE);
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
|
|
|
|
2000-07-04 16:35:15 +00:00
|
|
|
void
|
2007-07-05 16:23:49 +00:00
|
|
|
rip6_ctlinput(int cmd, struct sockaddr *sa, void *d)
|
2000-07-04 16:35:15 +00:00
|
|
|
{
|
2001-06-11 12:39:29 +00:00
|
|
|
struct ip6ctlparam *ip6cp = NULL;
|
|
|
|
const struct sockaddr_in6 *sa6_src = NULL;
|
2004-02-13 14:50:01 +00:00
|
|
|
void *cmdarg;
|
2008-01-08 19:08:58 +00:00
|
|
|
struct inpcb *(*notify)(struct inpcb *, int) = in6_rtchange;
|
2000-07-04 16:35:15 +00:00
|
|
|
|
|
|
|
if (sa->sa_family != AF_INET6 ||
|
|
|
|
sa->sa_len != sizeof(struct sockaddr_in6))
|
|
|
|
return;
|
|
|
|
|
|
|
|
if ((unsigned)cmd >= PRC_NCMDS)
|
|
|
|
return;
|
|
|
|
if (PRC_IS_REDIRECT(cmd))
|
|
|
|
notify = in6_rtchange, d = NULL;
|
|
|
|
else if (cmd == PRC_HOSTDEAD)
|
|
|
|
d = NULL;
|
|
|
|
else if (inet6ctlerrmap[cmd] == 0)
|
|
|
|
return;
|
|
|
|
|
2008-07-05 18:03:39 +00:00
|
|
|
/*
|
|
|
|
* If the parameter is from icmp6, decode it.
|
|
|
|
*/
|
2000-07-04 16:35:15 +00:00
|
|
|
if (d != NULL) {
|
2001-06-11 12:39:29 +00:00
|
|
|
ip6cp = (struct ip6ctlparam *)d;
|
2004-02-13 14:50:01 +00:00
|
|
|
cmdarg = ip6cp->ip6c_cmdarg;
|
2001-06-11 12:39:29 +00:00
|
|
|
sa6_src = ip6cp->ip6c_src;
|
2000-07-04 16:35:15 +00:00
|
|
|
} else {
|
2004-02-13 14:50:01 +00:00
|
|
|
cmdarg = NULL;
|
2001-06-11 12:39:29 +00:00
|
|
|
sa6_src = &sa6_any;
|
2000-07-04 16:35:15 +00:00
|
|
|
}
|
|
|
|
|
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
|
|
|
(void) in6_pcbnotify(&V_ripcbinfo, sa, 0,
|
2008-07-05 18:03:39 +00:00
|
|
|
(const struct sockaddr *)sa6_src, 0, cmd, cmdarg, notify);
|
2000-07-04 16:35:15 +00:00
|
|
|
}
|
|
|
|
|
1999-11-22 02:45:11 +00:00
|
|
|
/*
|
2008-07-05 18:03:39 +00:00
|
|
|
* Generate IPv6 header and pass packet to ip6_output. Tack on options user
|
|
|
|
* may have setup with control call.
|
1999-11-22 02:45:11 +00:00
|
|
|
*/
|
|
|
|
int
|
2014-08-08 01:57:15 +00:00
|
|
|
rip6_output(struct mbuf *m, struct socket *so, ...)
|
1999-11-22 02:45:11 +00:00
|
|
|
{
|
2003-10-09 16:13:47 +00:00
|
|
|
struct mbuf *control;
|
2010-08-19 11:31:03 +00:00
|
|
|
struct m_tag *mtag;
|
1999-11-22 02:45:11 +00:00
|
|
|
struct sockaddr_in6 *dstsock;
|
|
|
|
struct ip6_hdr *ip6;
|
|
|
|
struct inpcb *in6p;
|
|
|
|
u_int plen = m->m_pkthdr.len;
|
|
|
|
int error = 0;
|
2005-07-21 16:32:50 +00:00
|
|
|
struct ip6_pktopts opt, *optp;
|
1999-11-22 02:45:11 +00:00
|
|
|
struct ifnet *oifp = NULL;
|
|
|
|
int type = 0, code = 0; /* for ICMPv6 output statistics only */
|
2005-07-25 12:31:43 +00:00
|
|
|
int scope_ambiguous = 0;
|
2010-08-17 17:30:56 +00:00
|
|
|
int use_defzone = 0;
|
2016-01-10 13:40:29 +00:00
|
|
|
int hlim = 0;
|
2009-06-23 22:08:55 +00:00
|
|
|
struct in6_addr in6a;
|
1999-11-22 02:45:11 +00:00
|
|
|
va_list ap;
|
|
|
|
|
2014-08-08 01:57:15 +00:00
|
|
|
va_start(ap, so);
|
1999-11-22 02:45:11 +00:00
|
|
|
dstsock = va_arg(ap, struct sockaddr_in6 *);
|
|
|
|
control = va_arg(ap, struct mbuf *);
|
|
|
|
va_end(ap);
|
|
|
|
|
Another step assimilating IPv[46] PCB code - directly use
the inpcb names rather than the following IPv6 compat macros:
in6pcb,in6p_sp, in6p_ip6_nxt,in6p_flowinfo,in6p_vflag,
in6p_flags,in6p_socket,in6p_lport,in6p_fport,in6p_ppcb and
sotoin6pcb().
Apart from removing duplicate code in netipsec, this is a pure
whitespace, not a functional change.
Discussed with: rwatson
Reviewed by: rwatson (version before review requested changes)
MFC after: 4 weeks (set the timer and see then)
2008-12-15 21:50:54 +00:00
|
|
|
in6p = sotoinpcb(so);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(in6p);
|
1999-11-22 02:45:11 +00:00
|
|
|
|
2008-12-16 06:01:08 +00:00
|
|
|
if (control != NULL) {
|
2005-07-21 15:06:32 +00:00
|
|
|
if ((error = ip6_setpktopts(control, &opt,
|
2008-01-24 08:25:59 +00:00
|
|
|
in6p->in6p_outputopts, so->so_cred,
|
|
|
|
so->so_proto->pr_protocol)) != 0) {
|
1999-11-22 02:45:11 +00:00
|
|
|
goto bad;
|
2003-10-24 18:26:30 +00:00
|
|
|
}
|
2005-07-21 16:32:50 +00:00
|
|
|
optp = &opt;
|
|
|
|
} else
|
|
|
|
optp = in6p->in6p_outputopts;
|
1999-11-22 02:45:11 +00:00
|
|
|
|
2005-07-25 12:31:43 +00:00
|
|
|
/*
|
|
|
|
* Check and convert scope zone ID into internal form.
|
2008-07-05 18:03:39 +00:00
|
|
|
*
|
2005-07-25 12:31:43 +00:00
|
|
|
* XXX: we may still need to determine the zone later.
|
|
|
|
*/
|
|
|
|
if (!(so->so_state & SS_ISCONNECTED)) {
|
2010-08-20 17:52:49 +00:00
|
|
|
if (!optp || !optp->ip6po_pktinfo ||
|
|
|
|
!optp->ip6po_pktinfo->ipi6_ifindex)
|
2010-08-17 17:30:56 +00:00
|
|
|
use_defzone = V_ip6_use_defzone;
|
|
|
|
if (dstsock->sin6_scope_id == 0 && !use_defzone)
|
2005-07-25 12:31:43 +00:00
|
|
|
scope_ambiguous = 1;
|
2010-08-17 17:30:56 +00:00
|
|
|
if ((error = sa6_embedscope(dstsock, use_defzone)) != 0)
|
2005-07-25 12:31:43 +00:00
|
|
|
goto bad;
|
|
|
|
}
|
|
|
|
|
1999-11-22 02:45:11 +00:00
|
|
|
/*
|
2008-07-05 18:03:39 +00:00
|
|
|
* For an ICMPv6 packet, we should know its type and code to update
|
|
|
|
* statistics.
|
1999-11-22 02:45:11 +00:00
|
|
|
*/
|
|
|
|
if (so->so_proto->pr_protocol == IPPROTO_ICMPV6) {
|
|
|
|
struct icmp6_hdr *icmp6;
|
|
|
|
if (m->m_len < sizeof(struct icmp6_hdr) &&
|
|
|
|
(m = m_pullup(m, sizeof(struct icmp6_hdr))) == NULL) {
|
|
|
|
error = ENOBUFS;
|
|
|
|
goto bad;
|
|
|
|
}
|
|
|
|
icmp6 = mtod(m, struct icmp6_hdr *);
|
|
|
|
type = icmp6->icmp6_type;
|
|
|
|
code = icmp6->icmp6_code;
|
|
|
|
}
|
|
|
|
|
2012-12-05 08:04:20 +00:00
|
|
|
M_PREPEND(m, sizeof(*ip6), M_NOWAIT);
|
2004-08-12 18:31:36 +00:00
|
|
|
if (m == NULL) {
|
|
|
|
error = ENOBUFS;
|
|
|
|
goto bad;
|
|
|
|
}
|
1999-11-22 02:45:11 +00:00
|
|
|
ip6 = mtod(m, struct ip6_hdr *);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Source address selection.
|
|
|
|
*/
|
2016-01-10 13:40:29 +00:00
|
|
|
error = in6_selectsrc_socket(dstsock, optp, in6p, so->so_cred,
|
|
|
|
scope_ambiguous, &in6a, &hlim);
|
|
|
|
|
2009-06-23 22:08:55 +00:00
|
|
|
if (error)
|
2003-10-24 18:26:30 +00:00
|
|
|
goto bad;
|
2010-04-27 15:07:08 +00:00
|
|
|
error = prison_check_ip6(in6p->inp_cred, &in6a);
|
2009-02-05 14:06:09 +00:00
|
|
|
if (error != 0)
|
|
|
|
goto bad;
|
2009-06-23 22:08:55 +00:00
|
|
|
ip6->ip6_src = in6a;
|
2005-07-25 12:31:43 +00:00
|
|
|
|
|
|
|
ip6->ip6_dst = dstsock->sin6_addr;
|
|
|
|
|
2008-07-05 18:03:39 +00:00
|
|
|
/*
|
|
|
|
* Fill in the rest of the IPv6 header fields.
|
|
|
|
*/
|
1999-12-22 19:13:38 +00:00
|
|
|
ip6->ip6_flow = (ip6->ip6_flow & ~IPV6_FLOWINFO_MASK) |
|
Another step assimilating IPv[46] PCB code - directly use
the inpcb names rather than the following IPv6 compat macros:
in6pcb,in6p_sp, in6p_ip6_nxt,in6p_flowinfo,in6p_vflag,
in6p_flags,in6p_socket,in6p_lport,in6p_fport,in6p_ppcb and
sotoin6pcb().
Apart from removing duplicate code in netipsec, this is a pure
whitespace, not a functional change.
Discussed with: rwatson
Reviewed by: rwatson (version before review requested changes)
MFC after: 4 weeks (set the timer and see then)
2008-12-15 21:50:54 +00:00
|
|
|
(in6p->inp_flow & IPV6_FLOWINFO_MASK);
|
1999-12-22 19:13:38 +00:00
|
|
|
ip6->ip6_vfc = (ip6->ip6_vfc & ~IPV6_VERSION_MASK) |
|
2008-07-05 18:03:39 +00:00
|
|
|
(IPV6_VERSION & IPV6_VERSION_MASK);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* ip6_plen will be filled in ip6_output, so not fill it here.
|
|
|
|
*/
|
Another step assimilating IPv[46] PCB code - directly use
the inpcb names rather than the following IPv6 compat macros:
in6pcb,in6p_sp, in6p_ip6_nxt,in6p_flowinfo,in6p_vflag,
in6p_flags,in6p_socket,in6p_lport,in6p_fport,in6p_ppcb and
sotoin6pcb().
Apart from removing duplicate code in netipsec, this is a pure
whitespace, not a functional change.
Discussed with: rwatson
Reviewed by: rwatson (version before review requested changes)
MFC after: 4 weeks (set the timer and see then)
2008-12-15 21:50:54 +00:00
|
|
|
ip6->ip6_nxt = in6p->inp_ip_p;
|
2016-01-10 13:40:29 +00:00
|
|
|
ip6->ip6_hlim = hlim;
|
1999-11-22 02:45:11 +00:00
|
|
|
|
|
|
|
if (so->so_proto->pr_protocol == IPPROTO_ICMPV6 ||
|
|
|
|
in6p->in6p_cksum != -1) {
|
|
|
|
struct mbuf *n;
|
|
|
|
int off;
|
|
|
|
u_int16_t *p;
|
|
|
|
|
2008-07-05 18:03:39 +00:00
|
|
|
/* Compute checksum. */
|
1999-11-22 02:45:11 +00:00
|
|
|
if (so->so_proto->pr_protocol == IPPROTO_ICMPV6)
|
|
|
|
off = offsetof(struct icmp6_hdr, icmp6_cksum);
|
|
|
|
else
|
|
|
|
off = in6p->in6p_cksum;
|
2019-04-19 17:21:35 +00:00
|
|
|
if (plen < off + 2) {
|
1999-11-22 02:45:11 +00:00
|
|
|
error = EINVAL;
|
|
|
|
goto bad;
|
|
|
|
}
|
|
|
|
off += sizeof(struct ip6_hdr);
|
|
|
|
|
|
|
|
n = m;
|
|
|
|
while (n && n->m_len <= off) {
|
|
|
|
off -= n->m_len;
|
|
|
|
n = n->m_next;
|
|
|
|
}
|
|
|
|
if (!n)
|
|
|
|
goto bad;
|
|
|
|
p = (u_int16_t *)(mtod(n, caddr_t) + off);
|
|
|
|
*p = 0;
|
|
|
|
*p = in6_cksum(m, ip6->ip6_nxt, sizeof(*ip6), plen);
|
|
|
|
}
|
|
|
|
|
2010-08-19 11:31:03 +00:00
|
|
|
/*
|
|
|
|
* Send RA/RS messages to user land for protection, before sending
|
|
|
|
* them to rtadvd/rtsol.
|
|
|
|
*/
|
|
|
|
if ((send_sendso_input_hook != NULL) &&
|
|
|
|
so->so_proto->pr_protocol == IPPROTO_ICMPV6) {
|
|
|
|
switch (type) {
|
|
|
|
case ND_ROUTER_ADVERT:
|
|
|
|
case ND_ROUTER_SOLICIT:
|
|
|
|
mtag = m_tag_get(PACKET_TAG_ND_OUTGOING,
|
|
|
|
sizeof(unsigned short), M_NOWAIT);
|
|
|
|
if (mtag == NULL)
|
|
|
|
goto bad;
|
|
|
|
m_tag_prepend(m, mtag);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2005-07-21 16:32:50 +00:00
|
|
|
error = ip6_output(m, optp, NULL, 0, in6p->in6p_moptions, &oifp, in6p);
|
1999-11-22 02:45:11 +00:00
|
|
|
if (so->so_proto->pr_protocol == IPPROTO_ICMPV6) {
|
|
|
|
if (oifp)
|
|
|
|
icmp6_ifoutstat_inc(oifp, type, code);
|
2009-04-12 13:22:33 +00:00
|
|
|
ICMP6STAT_INC(icp6s_outhist[type]);
|
2001-06-11 12:39:29 +00:00
|
|
|
} else
|
2013-06-19 20:48:34 +00:00
|
|
|
RIP6STAT_INC(rip6s_opackets);
|
1999-11-22 02:45:11 +00:00
|
|
|
|
|
|
|
goto freectl;
|
|
|
|
|
|
|
|
bad:
|
|
|
|
if (m)
|
|
|
|
m_freem(m);
|
|
|
|
|
|
|
|
freectl:
|
2008-12-16 06:01:08 +00:00
|
|
|
if (control != NULL) {
|
2005-07-21 16:32:50 +00:00
|
|
|
ip6_clearpktopts(&opt, -1);
|
1999-11-22 02:45:11 +00:00
|
|
|
m_freem(control);
|
2001-06-11 12:39:29 +00:00
|
|
|
}
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(in6p);
|
2003-10-06 14:02:09 +00:00
|
|
|
return (error);
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Raw IPv6 socket option processing.
|
|
|
|
*/
|
|
|
|
int
|
2007-07-05 16:23:49 +00:00
|
|
|
rip6_ctloutput(struct socket *so, struct sockopt *sopt)
|
1999-11-22 02:45:11 +00:00
|
|
|
{
|
2012-02-03 11:00:53 +00:00
|
|
|
struct inpcb *inp;
|
1999-11-22 02:45:11 +00:00
|
|
|
int error;
|
|
|
|
|
|
|
|
if (sopt->sopt_level == IPPROTO_ICMPV6)
|
|
|
|
/*
|
|
|
|
* XXX: is it better to call icmp6_ctloutput() directly
|
|
|
|
* from protosw?
|
|
|
|
*/
|
2003-10-06 14:02:09 +00:00
|
|
|
return (icmp6_ctloutput(so, sopt));
|
2012-02-03 11:00:53 +00:00
|
|
|
else if (sopt->sopt_level != IPPROTO_IPV6) {
|
|
|
|
if (sopt->sopt_level == SOL_SOCKET &&
|
|
|
|
sopt->sopt_name == SO_SETFIB) {
|
|
|
|
inp = sotoinpcb(so);
|
|
|
|
INP_WLOCK(inp);
|
|
|
|
inp->inp_inc.inc_fibnum = so->so_fibnum;
|
|
|
|
INP_WUNLOCK(inp);
|
|
|
|
return (0);
|
|
|
|
}
|
1999-11-22 02:45:11 +00:00
|
|
|
return (EINVAL);
|
2012-02-03 11:00:53 +00:00
|
|
|
}
|
1999-11-22 02:45:11 +00:00
|
|
|
|
|
|
|
error = 0;
|
|
|
|
|
|
|
|
switch (sopt->sopt_dir) {
|
|
|
|
case SOPT_GET:
|
|
|
|
switch (sopt->sopt_name) {
|
2000-01-28 05:27:14 +00:00
|
|
|
case MRT6_INIT:
|
|
|
|
case MRT6_DONE:
|
|
|
|
case MRT6_ADD_MIF:
|
|
|
|
case MRT6_DEL_MIF:
|
|
|
|
case MRT6_ADD_MFC:
|
|
|
|
case MRT6_DEL_MFC:
|
|
|
|
case MRT6_PIM:
|
2007-02-24 11:38:47 +00:00
|
|
|
error = ip6_mrouter_get ? ip6_mrouter_get(so, sopt) :
|
|
|
|
EOPNOTSUPP;
|
2000-01-28 05:27:14 +00:00
|
|
|
break;
|
2003-10-26 18:17:01 +00:00
|
|
|
case IPV6_CHECKSUM:
|
|
|
|
error = ip6_raw_ctloutput(so, sopt);
|
|
|
|
break;
|
1999-11-22 02:45:11 +00:00
|
|
|
default:
|
|
|
|
error = ip6_ctloutput(so, sopt);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
case SOPT_SET:
|
|
|
|
switch (sopt->sopt_name) {
|
2000-01-28 05:27:14 +00:00
|
|
|
case MRT6_INIT:
|
|
|
|
case MRT6_DONE:
|
|
|
|
case MRT6_ADD_MIF:
|
|
|
|
case MRT6_DEL_MIF:
|
|
|
|
case MRT6_ADD_MFC:
|
|
|
|
case MRT6_DEL_MFC:
|
|
|
|
case MRT6_PIM:
|
2007-02-24 11:38:47 +00:00
|
|
|
error = ip6_mrouter_set ? ip6_mrouter_set(so, sopt) :
|
|
|
|
EOPNOTSUPP;
|
2000-01-28 05:27:14 +00:00
|
|
|
break;
|
2003-10-26 18:17:01 +00:00
|
|
|
case IPV6_CHECKSUM:
|
|
|
|
error = ip6_raw_ctloutput(so, sopt);
|
|
|
|
break;
|
1999-11-22 02:45:11 +00:00
|
|
|
default:
|
|
|
|
error = ip6_ctloutput(so, sopt);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
return (error);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
2001-09-12 08:38:13 +00:00
|
|
|
rip6_attach(struct socket *so, int proto, struct thread *td)
|
1999-11-22 02:45:11 +00:00
|
|
|
{
|
|
|
|
struct inpcb *inp;
|
2005-03-29 01:26:27 +00:00
|
|
|
struct icmp6_filter *filter;
|
2006-04-12 03:07:22 +00:00
|
|
|
int error;
|
1999-11-22 02:45:11 +00:00
|
|
|
|
|
|
|
inp = sotoinpcb(so);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
KASSERT(inp == NULL, ("rip6_attach: inp != NULL"));
|
2008-07-05 18:03:39 +00:00
|
|
|
|
2008-01-24 08:25:59 +00:00
|
|
|
error = priv_check(td, PRIV_NETINET_RAW);
|
|
|
|
if (error)
|
2008-07-05 18:03:39 +00:00
|
|
|
return (error);
|
1999-12-22 19:13:38 +00:00
|
|
|
error = soreserve(so, rip_sendspace, rip_recvspace);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
if (error)
|
2008-07-05 18:03:39 +00:00
|
|
|
return (error);
|
2008-10-23 20:26:15 +00:00
|
|
|
filter = malloc(sizeof(struct icmp6_filter), M_PCB, M_NOWAIT);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
if (filter == NULL)
|
2008-07-05 18:03:39 +00:00
|
|
|
return (ENOMEM);
|
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
|
|
|
INP_INFO_WLOCK(&V_ripcbinfo);
|
|
|
|
error = in_pcballoc(so, &V_ripcbinfo);
|
2004-07-27 23:44:03 +00:00
|
|
|
if (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
|
|
|
INP_INFO_WUNLOCK(&V_ripcbinfo);
|
2008-10-23 15:53:51 +00:00
|
|
|
free(filter, M_PCB);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (error);
|
2004-07-27 23:44:03 +00:00
|
|
|
}
|
1999-11-22 02:45:11 +00:00
|
|
|
inp = (struct inpcb *)so->so_pcb;
|
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
|
|
|
INP_INFO_WUNLOCK(&V_ripcbinfo);
|
1999-11-22 02:45:11 +00:00
|
|
|
inp->inp_vflag |= INP_IPV6;
|
Another step assimilating IPv[46] PCB code - directly use
the inpcb names rather than the following IPv6 compat macros:
in6pcb,in6p_sp, in6p_ip6_nxt,in6p_flowinfo,in6p_vflag,
in6p_flags,in6p_socket,in6p_lport,in6p_fport,in6p_ppcb and
sotoin6pcb().
Apart from removing duplicate code in netipsec, this is a pure
whitespace, not a functional change.
Discussed with: rwatson
Reviewed by: rwatson (version before review requested changes)
MFC after: 4 weeks (set the timer and see then)
2008-12-15 21:50:54 +00:00
|
|
|
inp->inp_ip_p = (long)proto;
|
1999-11-22 02:45:11 +00:00
|
|
|
inp->in6p_hops = -1; /* use kernel default */
|
|
|
|
inp->in6p_cksum = -1;
|
2005-03-29 01:26:27 +00:00
|
|
|
inp->in6p_icmp6filt = filter;
|
1999-11-22 02:45:11 +00:00
|
|
|
ICMP6_FILTER_SETPASSALL(inp->in6p_icmp6filt);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (0);
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
|
|
|
|
Chance protocol switch method pru_detach() so that it returns void
rather than an error. Detaches do not "fail", they other occur or
the protocol flags SS_PROTOREF to take ownership of the socket.
soclose() no longer looks at so_pcb to see if it's NULL, relying
entirely on the protocol to decide whether it's time to free the
socket or not using SS_PROTOREF. so_pcb is now entirely owned and
managed by the protocol code. Likewise, no longer test so_pcb in
other socket functions, such as soreceive(), which have no business
digging into protocol internals.
Protocol detach routines no longer try to free the socket on detach,
this is performed in the socket code if the protocol permits it.
In rts_detach(), no longer test for rp != NULL in detach, and
likewise in other protocols that don't permit a NULL so_pcb, reduce
the incidence of testing for it during detach.
netinet and netinet6 are not fully updated to this change, which
will be in an upcoming commit. In their current state they may leak
memory or panic.
MFC after: 3 months
2006-04-01 15:42:02 +00:00
|
|
|
static void
|
1999-11-22 02:45:11 +00:00
|
|
|
rip6_detach(struct socket *so)
|
|
|
|
{
|
|
|
|
struct inpcb *inp;
|
|
|
|
|
|
|
|
inp = sotoinpcb(so);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
KASSERT(inp != NULL, ("rip6_detach: inp == NULL"));
|
2006-07-21 17:11:15 +00:00
|
|
|
|
Bite the bullet, and make the IPv6 SSM and MLDv2 mega-commit:
import from p4 bms_netdev. Summary of changes:
* Connect netinet6/in6_mcast.c to build.
The legacy KAME KPIs are mostly preserved.
* Eliminate now dead code from ip6_output.c.
Don't do mbuf bingo, we are not going to do RFC 2292 style
CMSG tricks for multicast options as they are not required
by any current IPv6 normative reference.
* Refactor transports (UDP, raw_ip6) to do own mcast filtering.
SCTP, TCP unaffected by this change.
* Add ip6_msource, in6_msource structs to in6_var.h.
* Hookup mld_ifinfo state to in6_ifextra, allocate from
domifattach path.
* Eliminate IN6_LOOKUP_MULTI(), it is no longer referenced.
Kernel consumers which need this should use in6m_lookup().
* Refactor IPv6 socket group memberships to use a vector (like IPv4).
* Update ifmcstat(8) for IPv6 SSM.
* Add witness lock order for IN6_MULTI_LOCK.
* Move IN6_MULTI_LOCK out of lower ip6_output()/ip6_input() paths.
* Introduce IP6STAT_ADD/SUB/INC/DEC as per rwatson's IPv4 cleanup.
* Update carp(4) for new IPv6 SSM KPIs.
* Virtualize ip6_mrouter socket.
Changes mostly localized to IPv6 MROUTING.
* Don't do a local group lookup in MROUTING.
* Kill unused KAME prototypes in6_purgemkludge(), in6_restoremkludge().
* Preserve KAME DAD timer jitter behaviour in MLDv1 compatibility mode.
* Bump __FreeBSD_version to 800084.
* Update UPDATING.
NOTE WELL:
* This code hasn't been tested against real MLDv2 queriers
(yet), although the on-wire protocol has been verified in Wireshark.
* There are a few unresolved issues in the socket layer APIs to
do with scope ID propagation.
* There is a LOR present in ip6_output()'s use of
in6_setscope() which needs to be resolved. See comments in mld6.c.
This is believed to be benign and can't be avoided for the moment
without re-introducing an indirect netisr.
This work was mostly derived from the IGMPv3 implementation, and
has been sponsored by a third party.
2009-04-29 19:19:13 +00:00
|
|
|
if (so == V_ip6_mrouter && ip6_mrouter_done)
|
2000-02-27 18:35:10 +00:00
|
|
|
ip6_mrouter_done();
|
2007-02-24 11:38:47 +00:00
|
|
|
/* xxx: RSVP */
|
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
|
|
|
INP_INFO_WLOCK(&V_ripcbinfo);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
2008-10-23 15:53:51 +00:00
|
|
|
free(inp->in6p_icmp6filt, M_PCB);
|
2008-11-26 20:52:26 +00:00
|
|
|
in_pcbdetach(inp);
|
2008-11-27 12:04:35 +00:00
|
|
|
in_pcbfree(inp);
|
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
|
|
|
INP_INFO_WUNLOCK(&V_ripcbinfo);
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
|
|
|
|
2006-07-21 17:11:15 +00:00
|
|
|
/* XXXRW: This can't ever be called. */
|
2006-04-01 15:15:05 +00:00
|
|
|
static void
|
1999-11-22 02:45:11 +00:00
|
|
|
rip6_abort(struct socket *so)
|
|
|
|
{
|
2006-07-21 17:11:15 +00:00
|
|
|
struct inpcb *inp;
|
|
|
|
|
|
|
|
inp = sotoinpcb(so);
|
|
|
|
KASSERT(inp != NULL, ("rip6_abort: inp == NULL"));
|
|
|
|
|
|
|
|
soisdisconnected(so);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
rip6_close(struct socket *so)
|
|
|
|
{
|
|
|
|
struct inpcb *inp;
|
|
|
|
|
|
|
|
inp = sotoinpcb(so);
|
|
|
|
KASSERT(inp != NULL, ("rip6_close: inp == NULL"));
|
|
|
|
|
1999-11-22 02:45:11 +00:00
|
|
|
soisdisconnected(so);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
|
|
|
rip6_disconnect(struct socket *so)
|
|
|
|
{
|
2008-07-05 18:03:39 +00:00
|
|
|
struct inpcb *inp;
|
|
|
|
|
|
|
|
inp = sotoinpcb(so);
|
|
|
|
KASSERT(inp != NULL, ("rip6_disconnect: inp == NULL"));
|
1999-11-22 02:45:11 +00:00
|
|
|
|
2002-05-31 11:52:35 +00:00
|
|
|
if ((so->so_state & SS_ISCONNECTED) == 0)
|
2008-07-05 18:03:39 +00:00
|
|
|
return (ENOTCONN);
|
1999-11-22 02:45:11 +00:00
|
|
|
inp->in6p_faddr = in6addr_any;
|
2006-04-01 15:15:05 +00:00
|
|
|
rip6_abort(so);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
return (0);
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
2001-09-12 08:38:13 +00:00
|
|
|
rip6_bind(struct socket *so, struct sockaddr *nam, struct thread *td)
|
1999-11-22 02:45:11 +00:00
|
|
|
{
|
2019-01-09 01:11:19 +00:00
|
|
|
struct epoch_tracker et;
|
2008-07-05 18:03:39 +00:00
|
|
|
struct inpcb *inp;
|
1999-11-22 02:45:11 +00:00
|
|
|
struct sockaddr_in6 *addr = (struct sockaddr_in6 *)nam;
|
2009-06-23 20:19:09 +00:00
|
|
|
struct ifaddr *ifa = NULL;
|
2005-07-25 12:31:43 +00:00
|
|
|
int error = 0;
|
1999-11-22 02:45:11 +00:00
|
|
|
|
2008-07-05 18:03:39 +00:00
|
|
|
inp = sotoinpcb(so);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
KASSERT(inp != NULL, ("rip6_bind: inp == NULL"));
|
2008-07-05 18:03:39 +00:00
|
|
|
|
1999-11-22 02:45:11 +00:00
|
|
|
if (nam->sa_len != sizeof(*addr))
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EINVAL);
|
2009-02-05 14:06:09 +00:00
|
|
|
if ((error = prison_check_ip6(td->td_ucred, &addr->sin6_addr)) != 0)
|
|
|
|
return (error);
|
2018-05-23 21:02:14 +00:00
|
|
|
if (CK_STAILQ_EMPTY(&V_ifnet) || addr->sin6_family != AF_INET6)
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EADDRNOTAVAIL);
|
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 ((error = sa6_embedscope(addr, V_ip6_use_defzone)) != 0)
|
2008-07-05 18:03:39 +00:00
|
|
|
return (error);
|
2005-07-25 12:31:43 +00:00
|
|
|
|
2019-01-09 01:11:19 +00:00
|
|
|
NET_EPOCH_ENTER(et);
|
1999-11-22 02:45:11 +00:00
|
|
|
if (!IN6_IS_ADDR_UNSPECIFIED(&addr->sin6_addr) &&
|
2018-05-23 21:02:14 +00:00
|
|
|
(ifa = ifa_ifwithaddr((struct sockaddr *)addr)) == NULL) {
|
2019-01-09 01:11:19 +00:00
|
|
|
NET_EPOCH_EXIT(et);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EADDRNOTAVAIL);
|
2018-05-23 21:02:14 +00:00
|
|
|
}
|
2009-06-23 20:19:09 +00:00
|
|
|
if (ifa != NULL &&
|
|
|
|
((struct in6_ifaddr *)ifa)->ia6_flags &
|
1999-11-22 02:45:11 +00:00
|
|
|
(IN6_IFF_ANYCAST|IN6_IFF_NOTREADY|
|
|
|
|
IN6_IFF_DETACHED|IN6_IFF_DEPRECATED)) {
|
2019-01-09 01:11:19 +00:00
|
|
|
NET_EPOCH_EXIT(et);
|
2003-10-06 14:02:09 +00:00
|
|
|
return (EADDRNOTAVAIL);
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
2019-01-09 01:11:19 +00:00
|
|
|
NET_EPOCH_EXIT(et);
|
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
|
|
|
INP_INFO_WLOCK(&V_ripcbinfo);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
1999-11-22 02:45:11 +00:00
|
|
|
inp->in6p_laddr = addr->sin6_addr;
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
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
|
|
|
INP_INFO_WUNLOCK(&V_ripcbinfo);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (0);
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
2001-09-12 08:38:13 +00:00
|
|
|
rip6_connect(struct socket *so, struct sockaddr *nam, struct thread *td)
|
1999-11-22 02:45:11 +00:00
|
|
|
{
|
2008-07-05 18:03:39 +00:00
|
|
|
struct inpcb *inp;
|
1999-11-22 02:45:11 +00:00
|
|
|
struct sockaddr_in6 *addr = (struct sockaddr_in6 *)nam;
|
2009-06-23 22:08:55 +00:00
|
|
|
struct in6_addr in6a;
|
2005-07-25 12:31:43 +00:00
|
|
|
int error = 0, scope_ambiguous = 0;
|
1999-11-22 02:45:11 +00:00
|
|
|
|
2008-07-05 18:03:39 +00:00
|
|
|
inp = sotoinpcb(so);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
KASSERT(inp != NULL, ("rip6_connect: inp == NULL"));
|
2008-07-05 18:03:39 +00:00
|
|
|
|
1999-11-22 02:45:11 +00:00
|
|
|
if (nam->sa_len != sizeof(*addr))
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EINVAL);
|
2018-05-23 21:02:14 +00:00
|
|
|
if (CK_STAILQ_EMPTY(&V_ifnet))
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EADDRNOTAVAIL);
|
1999-11-22 02:45:11 +00:00
|
|
|
if (addr->sin6_family != AF_INET6)
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EAFNOSUPPORT);
|
2005-07-25 12:31:43 +00:00
|
|
|
|
|
|
|
/*
|
2008-07-05 18:03:39 +00:00
|
|
|
* Application should provide a proper zone ID or the use of default
|
|
|
|
* zone IDs should be enabled. Unfortunately, some applications do
|
|
|
|
* not behave as it should, so we need a workaround. Even if an
|
|
|
|
* appropriate ID is not determined, we'll see if we can determine
|
|
|
|
* the outgoing interface. If we can, determine the zone ID based on
|
|
|
|
* the interface below.
|
2005-07-25 12:31:43 +00:00
|
|
|
*/
|
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 (addr->sin6_scope_id == 0 && !V_ip6_use_defzone)
|
2005-07-25 12:31:43 +00:00
|
|
|
scope_ambiguous = 1;
|
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 ((error = sa6_embedscope(addr, V_ip6_use_defzone)) != 0)
|
2008-07-05 18:03:39 +00:00
|
|
|
return (error);
|
2005-07-25 12:31:43 +00:00
|
|
|
|
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
|
|
|
INP_INFO_WLOCK(&V_ripcbinfo);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
1999-11-22 02:45:11 +00:00
|
|
|
/* Source address selection. XXX: need pcblookup? */
|
2016-01-10 13:40:29 +00:00
|
|
|
error = in6_selectsrc_socket(addr, inp->in6p_outputopts,
|
|
|
|
inp, so->so_cred, scope_ambiguous, &in6a, NULL);
|
2009-06-23 22:08:55 +00:00
|
|
|
if (error) {
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
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
|
|
|
INP_INFO_WUNLOCK(&V_ripcbinfo);
|
2009-06-23 22:08:55 +00:00
|
|
|
return (error);
|
2004-07-27 23:44:03 +00:00
|
|
|
}
|
2005-07-25 12:31:43 +00:00
|
|
|
|
1999-11-22 02:45:11 +00:00
|
|
|
inp->in6p_faddr = addr->sin6_addr;
|
2009-06-23 22:08:55 +00:00
|
|
|
inp->in6p_laddr = in6a;
|
1999-11-22 02:45:11 +00:00
|
|
|
soisconnected(so);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
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
|
|
|
INP_INFO_WUNLOCK(&V_ripcbinfo);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (0);
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
|
|
|
rip6_shutdown(struct socket *so)
|
|
|
|
{
|
2004-07-27 23:44:03 +00:00
|
|
|
struct inpcb *inp;
|
|
|
|
|
|
|
|
inp = sotoinpcb(so);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
KASSERT(inp != NULL, ("rip6_shutdown: inp == NULL"));
|
2008-07-05 18:03:39 +00:00
|
|
|
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WLOCK(inp);
|
1999-11-22 02:45:11 +00:00
|
|
|
socantsendmore(so);
|
2008-04-17 21:38:18 +00:00
|
|
|
INP_WUNLOCK(inp);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (0);
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
|
|
|
rip6_send(struct socket *so, int flags, struct mbuf *m, struct sockaddr *nam,
|
2007-07-05 16:29:40 +00:00
|
|
|
struct mbuf *control, struct thread *td)
|
1999-11-22 02:45:11 +00:00
|
|
|
{
|
2008-07-05 18:03:39 +00:00
|
|
|
struct inpcb *inp;
|
1999-11-22 02:45:11 +00:00
|
|
|
struct sockaddr_in6 tmp;
|
|
|
|
struct sockaddr_in6 *dst;
|
2004-07-27 23:44:03 +00:00
|
|
|
int ret;
|
1999-11-22 02:45:11 +00:00
|
|
|
|
2008-07-05 18:03:39 +00:00
|
|
|
inp = sotoinpcb(so);
|
Update in_pcb-derived basic socket types following changes to
pru_abort(), pru_detach(), and in_pcbdetach():
- Universally support and enforce the invariant that so_pcb is
never NULL, converting dozens of unnecessary NULL checks into
assertions, and eliminating dozens of unnecessary error handling
cases in protocol code.
- In some cases, eliminate unnecessary pcbinfo locking, as it is no
longer required to ensure so_pcb != NULL. For example, in protocol
shutdown methods, and in raw IP send.
- Abort and detach protocol switch methods no longer return failures,
nor attempt to free sockets, as the socket layer does this.
- Invoke in_pcbfree() after in_pcbdetach() in order to free the
detached in_pcb structure for a socket.
MFC after: 3 months
2006-04-01 16:20:54 +00:00
|
|
|
KASSERT(inp != NULL, ("rip6_send: inp == NULL"));
|
2008-07-05 18:03:39 +00:00
|
|
|
|
|
|
|
/* Always copy sockaddr to avoid overwrites. */
|
2004-07-27 23:44:03 +00:00
|
|
|
/* Unlocked read. */
|
1999-11-22 02:45:11 +00:00
|
|
|
if (so->so_state & SS_ISCONNECTED) {
|
|
|
|
if (nam) {
|
|
|
|
m_freem(m);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EISCONN);
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
|
|
|
/* XXX */
|
|
|
|
bzero(&tmp, sizeof(tmp));
|
|
|
|
tmp.sin6_family = AF_INET6;
|
|
|
|
tmp.sin6_len = sizeof(struct sockaddr_in6);
|
2008-07-30 09:26:27 +00:00
|
|
|
INP_RLOCK(inp);
|
1999-11-22 02:45:11 +00:00
|
|
|
bcopy(&inp->in6p_faddr, &tmp.sin6_addr,
|
2008-07-30 09:26:27 +00:00
|
|
|
sizeof(struct in6_addr));
|
|
|
|
INP_RUNLOCK(inp);
|
1999-11-22 02:45:11 +00:00
|
|
|
dst = &tmp;
|
|
|
|
} else {
|
|
|
|
if (nam == NULL) {
|
|
|
|
m_freem(m);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (ENOTCONN);
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
2005-07-25 12:31:43 +00:00
|
|
|
if (nam->sa_len != sizeof(struct sockaddr_in6)) {
|
|
|
|
m_freem(m);
|
2008-07-05 18:03:39 +00:00
|
|
|
return (EINVAL);
|
2005-07-25 12:31:43 +00:00
|
|
|
}
|
2000-07-04 16:35:15 +00:00
|
|
|
tmp = *(struct sockaddr_in6 *)nam;
|
|
|
|
dst = &tmp;
|
2005-07-25 12:31:43 +00:00
|
|
|
|
|
|
|
if (dst->sin6_family == AF_UNSPEC) {
|
|
|
|
/*
|
|
|
|
* XXX: we allow this case for backward
|
|
|
|
* compatibility to buggy applications that
|
|
|
|
* rely on old (and wrong) kernel behavior.
|
|
|
|
*/
|
|
|
|
log(LOG_INFO, "rip6 SEND: address family is "
|
|
|
|
"unspec. Assume AF_INET6\n");
|
|
|
|
dst->sin6_family = AF_INET6;
|
|
|
|
} else if (dst->sin6_family != AF_INET6) {
|
|
|
|
m_freem(m);
|
|
|
|
return(EAFNOSUPPORT);
|
|
|
|
}
|
2000-07-04 16:35:15 +00:00
|
|
|
}
|
2004-07-27 23:44:03 +00:00
|
|
|
ret = rip6_output(m, so, dst, control);
|
|
|
|
return (ret);
|
1999-11-22 02:45:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
struct pr_usrreqs rip6_usrreqs = {
|
2004-11-08 14:44:54 +00:00
|
|
|
.pru_abort = rip6_abort,
|
|
|
|
.pru_attach = rip6_attach,
|
|
|
|
.pru_bind = rip6_bind,
|
|
|
|
.pru_connect = rip6_connect,
|
|
|
|
.pru_control = in6_control,
|
|
|
|
.pru_detach = rip6_detach,
|
|
|
|
.pru_disconnect = rip6_disconnect,
|
2007-05-11 10:20:51 +00:00
|
|
|
.pru_peeraddr = in6_getpeeraddr,
|
2004-11-08 14:44:54 +00:00
|
|
|
.pru_send = rip6_send,
|
|
|
|
.pru_shutdown = rip6_shutdown,
|
2007-05-11 10:20:51 +00:00
|
|
|
.pru_sockaddr = in6_getsockaddr,
|
2006-07-21 17:11:15 +00:00
|
|
|
.pru_close = rip6_close,
|
1999-11-22 02:45:11 +00:00
|
|
|
};
|