2000-07-04 16:35:15 +00:00
|
|
|
/* $FreeBSD$ */
|
2002-04-19 04:46:24 +00:00
|
|
|
/* $KAME: if_gif.c,v 1.87 2001/10/19 08:50:27 itojun Exp $ */
|
2000-07-04 16:35:15 +00:00
|
|
|
|
2005-01-07 01:45:51 +00:00
|
|
|
/*-
|
1999-12-07 17:39:16 +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.
|
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
#include "opt_inet.h"
|
|
|
|
|
#include "opt_inet6.h"
|
2002-08-12 16:08:23 +00:00
|
|
|
#include "opt_mac.h"
|
1999-12-07 17:39:16 +00:00
|
|
|
|
|
|
|
|
#include <sys/param.h>
|
|
|
|
|
#include <sys/systm.h>
|
|
|
|
|
#include <sys/kernel.h>
|
|
|
|
|
#include <sys/malloc.h>
|
|
|
|
|
#include <sys/mbuf.h>
|
2004-05-30 20:27:19 +00:00
|
|
|
#include <sys/module.h>
|
1999-12-07 17:39:16 +00:00
|
|
|
#include <sys/socket.h>
|
|
|
|
|
#include <sys/sockio.h>
|
|
|
|
|
#include <sys/errno.h>
|
|
|
|
|
#include <sys/time.h>
|
2002-02-26 01:56:56 +00:00
|
|
|
#include <sys/sysctl.h>
|
1999-12-07 17:39:16 +00:00
|
|
|
#include <sys/syslog.h>
|
Add code to allow the system to handle multiple routing tables.
This particular implementation is designed to be fully backwards compatible
and to be MFC-able to 7.x (and 6.x)
Currently the only protocol that can make use of the multiple tables is IPv4
Similar functionality exists in OpenBSD and Linux.
From my notes:
-----
One thing where FreeBSD has been falling behind, and which by chance I
have some time to work on is "policy based routing", which allows
different
packet streams to be routed by more than just the destination address.
Constraints:
------------
I want to make some form of this available in the 6.x tree
(and by extension 7.x) , but FreeBSD in general needs it so I might as
well do it in -current and back port the portions I need.
One of the ways that this can be done is to have the ability to
instantiate multiple kernel routing tables (which I will now
refer to as "Forwarding Information Bases" or "FIBs" for political
correctness reasons). Which FIB a particular packet uses to make
the next hop decision can be decided by a number of mechanisms.
The policies these mechanisms implement are the "Policies" referred
to in "Policy based routing".
One of the constraints I have if I try to back port this work to
6.x is that it must be implemented as a EXTENSION to the existing
ABIs in 6.x so that third party applications do not need to be
recompiled in timespan of the branch.
This first version will not have some of the bells and whistles that
will come with later versions. It will, for example, be limited to 16
tables in the first commit.
Implementation method, Compatible version. (part 1)
-------------------------------
For this reason I have implemented a "sufficient subset" of a
multiple routing table solution in Perforce, and back-ported it
to 6.x. (also in Perforce though not always caught up with what I
have done in -current/P4). The subset allows a number of FIBs
to be defined at compile time (8 is sufficient for my purposes in 6.x)
and implements the changes needed to allow IPV4 to use them. I have not
done the changes for ipv6 simply because I do not need it, and I do not
have enough knowledge of ipv6 (e.g. neighbor discovery) needed to do it.
Other protocol families are left untouched and should there be
users with proprietary protocol families, they should continue to work
and be oblivious to the existence of the extra FIBs.
To understand how this is done, one must know that the current FIB
code starts everything off with a single dimensional array of
pointers to FIB head structures (One per protocol family), each of
which in turn points to the trie of routes available to that family.
The basic change in the ABI compatible version of the change is to
extent that array to be a 2 dimensional array, so that
instead of protocol family X looking at rt_tables[X] for the
table it needs, it looks at rt_tables[Y][X] when for all
protocol families except ipv4 Y is always 0.
Code that is unaware of the change always just sees the first row
of the table, which of course looks just like the one dimensional
array that existed before.
The entry points rtrequest(), rtalloc(), rtalloc1(), rtalloc_ign()
are all maintained, but refer only to the first row of the array,
so that existing callers in proprietary protocols can continue to
do the "right thing".
Some new entry points are added, for the exclusive use of ipv4 code
called in_rtrequest(), in_rtalloc(), in_rtalloc1() and in_rtalloc_ign(),
which have an extra argument which refers the code to the correct row.
In addition, there are some new entry points (currently called
rtalloc_fib() and friends) that check the Address family being
looked up and call either rtalloc() (and friends) if the protocol
is not IPv4 forcing the action to row 0 or to the appropriate row
if it IS IPv4 (and that info is available). These are for calling
from code that is not specific to any particular protocol. The way
these are implemented would change in the non ABI preserving code
to be added later.
One feature of the first version of the code is that for ipv4,
the interface routes show up automatically on all the FIBs, so
that no matter what FIB you select you always have the basic
direct attached hosts available to you. (rtinit() does this
automatically).
You CAN delete an interface route from one FIB should you want
to but by default it's there. ARP information is also available
in each FIB. It's assumed that the same machine would have the
same MAC address, regardless of which FIB you are using to get
to it.
This brings us as to how the correct FIB is selected for an outgoing
IPV4 packet.
Firstly, all packets have a FIB associated with them. if nothing
has been done to change it, it will be FIB 0. The FIB is changed
in the following ways.
Packets fall into one of a number of classes.
1/ locally generated packets, coming from a socket/PCB.
Such packets select a FIB from a number associated with the
socket/PCB. This in turn is inherited from the process,
but can be changed by a socket option. The process in turn
inherits it on fork. I have written a utility call setfib
that acts a bit like nice..
setfib -3 ping target.example.com # will use fib 3 for ping.
It is an obvious extension to make it a property of a jail
but I have not done so. It can be achieved by combining the setfib and
jail commands.
2/ packets received on an interface for forwarding.
By default these packets would use table 0,
(or possibly a number settable in a sysctl(not yet)).
but prior to routing the firewall can inspect them (see below).
(possibly in the future you may be able to associate a FIB
with packets received on an interface.. An ifconfig arg, but not yet.)
3/ packets inspected by a packet classifier, which can arbitrarily
associate a fib with it on a packet by packet basis.
A fib assigned to a packet by a packet classifier
(such as ipfw) would over-ride a fib associated by
a more default source. (such as cases 1 or 2).
4/ a tcp listen socket associated with a fib will generate
accept sockets that are associated with that same fib.
5/ Packets generated in response to some other packet (e.g. reset
or icmp packets). These should use the FIB associated with the
packet being reponded to.
6/ Packets generated during encapsulation.
gif, tun and other tunnel interfaces will encapsulate using the FIB
that was in effect withthe proces that set up the tunnel.
thus setfib 1 ifconfig gif0 [tunnel instructions]
will set the fib for the tunnel to use to be fib 1.
Routing messages would be associated with their
process, and thus select one FIB or another.
messages from the kernel would be associated with the fib they
refer to and would only be received by a routing socket associated
with that fib. (not yet implemented)
In addition Netstat has been edited to be able to cope with the
fact that the array is now 2 dimensional. (It looks in system
memory using libkvm (!)). Old versions of netstat see only the first FIB.
In addition two sysctls are added to give:
a) the number of FIBs compiled in (active)
b) the default FIB of the calling process.
Early testing experience:
-------------------------
Basically our (IronPort's) appliance does this functionality already
using ipfw fwd but that method has some drawbacks.
For example,
It can't fully simulate a routing table because it can't influence the
socket's choice of local address when a connect() is done.
Testing during the generating of these changes has been
remarkably smooth so far. Multiple tables have co-existed
with no notable side effects, and packets have been routes
accordingly.
ipfw has grown 2 new keywords:
setfib N ip from anay to any
count ip from any to any fib N
In pf there seems to be a requirement to be able to give symbolic names to the
fibs but I do not have that capacity. I am not sure if it is required.
SCTP has interestingly enough built in support for this, called VRFs
in Cisco parlance. it will be interesting to see how that handles it
when it suddenly actually does something.
Where to next:
--------------------
After committing the ABI compatible version and MFCing it, I'd
like to proceed in a forward direction in -current. this will
result in some roto-tilling in the routing code.
Firstly: the current code's idea of having a separate tree per
protocol family, all of the same format, and pointed to by the
1 dimensional array is a bit silly. Especially when one considers that
there is code that makes assumptions about every protocol having the
same internal structures there. Some protocols don't WANT that
sort of structure. (for example the whole idea of a netmask is foreign
to appletalk). This needs to be made opaque to the external code.
My suggested first change is to add routing method pointers to the
'domain' structure, along with information pointing the data.
instead of having an array of pointers to uniform structures,
there would be an array pointing to the 'domain' structures
for each protocol address domain (protocol family),
and the methods this reached would be called. The methods would have
an argument that gives FIB number, but the protocol would be free
to ignore it.
When the ABI can be changed it raises the possibilty of the
addition of a fib entry into the "struct route". Currently,
the structure contains the sockaddr of the desination, and the resulting
fib entry. To make this work fully, one could add a fib number
so that given an address and a fib, one can find the third element, the
fib entry.
Interaction with the ARP layer/ LL layer would need to be
revisited as well. Qing Li has been working on this already.
This work was sponsored by Ironport Systems/Cisco
Reviewed by: several including rwatson, bz and mlair (parts each)
Obtained from: Ironport systems/Cisco
2008-05-09 23:03:00 +00:00
|
|
|
#include <sys/proc.h>
|
2000-07-04 16:35:15 +00:00
|
|
|
#include <sys/protosw.h>
|
2001-07-02 21:02:09 +00:00
|
|
|
#include <sys/conf.h>
|
Commit step 1 of the vimage project, (network stack)
virtualization work done by Marko Zec (zec@).
This is the first in a series of commits over the course
of the next few weeks.
Mark all uses of global variables to be virtualized
with a V_ prefix.
Use macros to map them back to their global names for
now, so this is a NOP change only.
We hope to have caught at least 85-90% of what is needed
so we do not invalidate a lot of outstanding patches again.
Obtained from: //depot/projects/vimage-commit2/...
Reviewed by: brooks, des, ed, mav, julian,
jamie, kris, rwatson, zec, ...
(various people I forgot, different versions)
md5 (with a bit of help)
Sponsored by: NLnet Foundation, The FreeBSD Foundation
X-MFC after: never
V_Commit_Message_Reviewed_By: more people than the patch
2008-08-17 23:27:27 +00:00
|
|
|
#include <sys/vimage.h>
|
1999-12-07 17:39:16 +00:00
|
|
|
#include <machine/cpu.h>
|
|
|
|
|
|
|
|
|
|
#include <net/if.h>
|
Major overhaul of pseudo-interface cloning. Highlights include:
- Split the code out into if_clone.[ch].
- Locked struct if_clone. [1]
- Add a per-cloner match function rather then simply matching names of
the form <name><unit> and <name>.
- Use the match function to allow creation of <interface>.<tag>
vlan interfaces. The old way is preserved unchanged!
- Also the match function to allow creation of stf(4) interfaces named
stf0, stf, or 6to4. This is the only major user visible change in
that "ifconfig stf" creates the interface stf rather then stf0 and
does not print "stf0" to stdout.
- Allow destroy functions to fail so they can refuse to delete
interfaces. Currently, we forbid the deletion of interfaces which
were created in the init function, particularly lo0, pflog0, and
pfsync0. In the case of lo0 this was a panic implementation so it
does not count as a user visiable change. :-)
- Since most interfaces do not need the new functionality, an family of
wrapper functions, ifc_simple_*(), were created to wrap old style
cloner functions.
- The IF_CLONE_INITIALIZER macro is replaced with a new incompatible
IFC_CLONE_INITIALIZER and ifc_simple consumers use IFC_SIMPLE_DECLARE
instead.
Submitted by: Maurycy Pawlowski-Wieronski <maurycy at fouk.org> [1]
Reviewed by: andre, mlaier
Discussed on: net
2004-06-22 20:13:25 +00:00
|
|
|
#include <net/if_clone.h>
|
1999-12-07 17:39:16 +00:00
|
|
|
#include <net/if_types.h>
|
|
|
|
|
#include <net/netisr.h>
|
|
|
|
|
#include <net/route.h>
|
|
|
|
|
#include <net/bpf.h>
|
|
|
|
|
|
|
|
|
|
#include <netinet/in.h>
|
|
|
|
|
#include <netinet/in_systm.h>
|
|
|
|
|
#include <netinet/ip.h>
|
2001-06-11 12:39:29 +00:00
|
|
|
#ifdef INET
|
|
|
|
|
#include <netinet/in_var.h>
|
1999-12-07 17:39:16 +00:00
|
|
|
#include <netinet/in_gif.h>
|
2001-07-02 21:02:09 +00:00
|
|
|
#include <netinet/ip_var.h>
|
1999-12-07 17:39:16 +00:00
|
|
|
#endif /* INET */
|
|
|
|
|
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
#ifndef INET
|
|
|
|
|
#include <netinet/in.h>
|
|
|
|
|
#endif
|
|
|
|
|
#include <netinet6/in6_var.h>
|
|
|
|
|
#include <netinet/ip6.h>
|
|
|
|
|
#include <netinet6/ip6_var.h>
|
2005-07-25 12:31:43 +00:00
|
|
|
#include <netinet6/scope6_var.h>
|
1999-12-07 17:39:16 +00:00
|
|
|
#include <netinet6/in6_gif.h>
|
2000-07-04 16:35:15 +00:00
|
|
|
#include <netinet6/ip6protosw.h>
|
1999-12-07 17:39:16 +00:00
|
|
|
#endif /* INET6 */
|
|
|
|
|
|
2000-07-04 16:35:15 +00:00
|
|
|
#include <netinet/ip_encap.h>
|
2005-12-21 21:29:45 +00:00
|
|
|
#include <net/ethernet.h>
|
|
|
|
|
#include <net/if_bridgevar.h>
|
1999-12-07 17:39:16 +00:00
|
|
|
#include <net/if_gif.h>
|
|
|
|
|
|
2006-10-22 11:52:19 +00:00
|
|
|
#include <security/mac/mac_framework.h>
|
|
|
|
|
|
2001-07-02 21:02:09 +00:00
|
|
|
#define GIFNAME "gif"
|
|
|
|
|
|
2004-03-22 15:43:14 +00:00
|
|
|
/*
|
2004-04-05 16:55:15 +00:00
|
|
|
* gif_mtx protects the global gif_softc_list.
|
2004-03-22 15:43:14 +00:00
|
|
|
*/
|
|
|
|
|
static struct mtx gif_mtx;
|
2001-07-02 21:02:09 +00:00
|
|
|
static MALLOC_DEFINE(M_GIF, "gif", "Generic Tunnel Interface");
|
2008-11-19 09:39:34 +00:00
|
|
|
|
Conditionally compile out V_ globals while instantiating the appropriate
container structures, depending on VIMAGE_GLOBALS compile time option.
Make VIMAGE_GLOBALS a new compile-time option, which by default will not
be defined, resulting in instatiations of global variables selected for
V_irtualization (enclosed in #ifdef VIMAGE_GLOBALS blocks) to be
effectively compiled out. Instantiate new global container structures
to hold V_irtualized variables: vnet_net_0, vnet_inet_0, vnet_inet6_0,
vnet_ipsec_0, vnet_netgraph_0, and vnet_gif_0.
Update the VSYM() macro so that depending on VIMAGE_GLOBALS the V_
macros resolve either to the original globals, or to fields inside
container structures, i.e. effectively
#ifdef VIMAGE_GLOBALS
#define V_rt_tables rt_tables
#else
#define V_rt_tables vnet_net_0._rt_tables
#endif
Update SYSCTL_V_*() macros to operate either on globals or on fields
inside container structs.
Extend the internal kldsym() lookups with the ability to resolve
selected fields inside the virtualization container structs. This
applies only to the fields which are explicitly registered for kldsym()
visibility via VNET_MOD_DECLARE() and vnet_mod_register(), currently
this is done only in sys/net/if.c.
Fix a few broken instances of MODULE_GLOBAL() macro use in SCTP code,
and modify the MODULE_GLOBAL() macro to resolve to V_ macros, which in
turn result in proper code being generated depending on VIMAGE_GLOBALS.
De-virtualize local static variables in sys/contrib/pf/net/pf_subr.c
which were prematurely V_irtualized by automated V_ prepending scripts
during earlier merging steps. PF virtualization will be done
separately, most probably after next PF import.
Convert a few variable initializations at instantiation to
initialization in init functions, most notably in ipfw. Also convert
TUNABLE_INT() initializers for V_ variables to TUNABLE_FETCH_INT() in
initializer functions.
Discussed at: devsummit Strassburg
Reviewed by: bz, julian
Approved by: julian (mentor)
Obtained from: //depot/projects/vimage-commit2/...
X-MFC after: never
Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-12-10 23:12:39 +00:00
|
|
|
#ifndef VIMAGE
|
|
|
|
|
#ifndef VIMAGE_GLOBALS
|
|
|
|
|
struct vnet_gif vnet_gif_0;
|
|
|
|
|
#endif
|
|
|
|
|
#endif
|
|
|
|
|
|
2008-11-19 09:39:34 +00:00
|
|
|
#ifdef VIMAGE_GLOBALS
|
2002-01-08 10:30:09 +00:00
|
|
|
static LIST_HEAD(, gif_softc) gif_softc_list;
|
2008-11-19 09:39:34 +00:00
|
|
|
static int max_gif_nesting;
|
|
|
|
|
static int parallel_tunnels;
|
|
|
|
|
#ifdef INET
|
|
|
|
|
int ip_gif_ttl;
|
|
|
|
|
#endif
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
int ip6_gif_hlim;
|
|
|
|
|
#endif
|
|
|
|
|
#endif
|
2001-07-02 21:02:09 +00:00
|
|
|
|
2001-09-26 23:50:17 +00:00
|
|
|
void (*ng_gif_input_p)(struct ifnet *ifp, struct mbuf **mp, int af);
|
|
|
|
|
void (*ng_gif_input_orphan_p)(struct ifnet *ifp, struct mbuf *m, int af);
|
|
|
|
|
void (*ng_gif_attach_p)(struct ifnet *ifp);
|
|
|
|
|
void (*ng_gif_detach_p)(struct ifnet *ifp);
|
|
|
|
|
|
2005-12-21 21:29:45 +00:00
|
|
|
static void gif_start(struct ifnet *);
|
2006-07-09 06:04:01 +00:00
|
|
|
static int gif_clone_create(struct if_clone *, int, caddr_t);
|
2004-04-14 00:57:49 +00:00
|
|
|
static void gif_clone_destroy(struct ifnet *);
|
First pass at separating per-vnet initializer functions
from existing functions for initializing global state.
At this stage, the new per-vnet initializer functions are
directly called from the existing global initialization code,
which should in most cases result in compiler inlining those
new functions, hence yielding a near-zero functional change.
Modify the existing initializer functions which are invoked via
protosw, like ip_init() et. al., to allow them to be invoked
multiple times, i.e. per each vnet. Global state, if any,
is initialized only if such functions are called within the
context of vnet0, which will be determined via the
IS_DEFAULT_VNET(curvnet) check (currently always true).
While here, V_irtualize a few remaining global UMA zones
used by net/netinet/netipsec networking code. While it is
not yet clear to me or anybody else whether this is the right
thing to do, at this stage this makes the code more readable,
and makes it easier to track uncollected UMA-zone-backed
objects on vnet removal. In the long run, it's quite possible
that some form of shared use of UMA zone pools among multiple
vnets should be considered.
Bump __FreeBSD_version due to changes in layout of structs
vnet_ipfw, vnet_inet and vnet_net.
Approved by: julian (mentor)
2009-04-06 22:29:41 +00:00
|
|
|
static int vnet_gif_iattach(const void *);
|
2001-07-02 21:02:09 +00:00
|
|
|
|
Introduce vnet module registration / initialization framework with
dependency tracking and ordering enforcement.
With this change, per-vnet initialization functions introduced with
r190787 are no longer directly called from traditional initialization
functions (which cc in most cases inlined to pre-r190787 code), but are
instead registered via the vnet framework first, and are invoked only
after all prerequisite modules have been initialized. In the long run,
this framework should allow us to both initialize and dismantle
multiple vnet instances in a correct order.
The problem this change aims to solve is how to replay the
initialization sequence of various network stack components, which
have been traditionally triggered via different mechanisms (SYSINIT,
protosw). Note that this initialization sequence was and still can be
subtly different depending on whether certain pieces of code have been
statically compiled into the kernel, loaded as modules by boot
loader, or kldloaded at run time.
The approach is simple - we record the initialization sequence
established by the traditional mechanisms whenever vnet_mod_register()
is called for a particular vnet module. The vnet_mod_register_multi()
variant allows a single initializer function to be registered multiple
times but with different arguments - currently this is only used in
kern/uipc_domain.c by net_add_domain() with different struct domain *
as arguments, which allows for protosw-registered initialization
routines to be invoked in a correct order by the new vnet
initialization framework.
For the purpose of identifying vnet modules, each vnet module has to
have a unique ID, which is statically assigned in sys/vimage.h.
Dynamic assignment of vnet module IDs is not supported yet.
A vnet module may specify a single prerequisite module at registration
time by filling in the vmi_dependson field of its vnet_modinfo struct
with the ID of the module it depends on. Unless specified otherwise,
all vnet modules depend on VNET_MOD_NET (container for ifnet list head,
rt_tables etc.), which thus has to and will always be initialized
first. The framework will panic if it detects any unresolved
dependencies before completing system initialization. Detection of
unresolved dependencies for vnet modules registered after boot
(kldloaded modules) is not provided.
Note that the fact that each module can specify only a single
prerequisite may become problematic in the long run. In particular,
INET6 depends on INET being already instantiated, due to TCP / UDP
structures residing in INET container. IPSEC also depends on INET,
which will in turn additionally complicate making INET6-only kernel
configs a reality.
The entire registration framework can be compiled out by turning on the
VIMAGE_GLOBALS kernel config option.
Reviewed by: bz
Approved by: julian (mentor)
2009-04-11 05:58:58 +00:00
|
|
|
#ifndef VIMAGE_GLOBALS
|
|
|
|
|
static const vnet_modinfo_t vnet_gif_modinfo = {
|
|
|
|
|
.vmi_id = VNET_MOD_GIF,
|
|
|
|
|
.vmi_name = "gif",
|
|
|
|
|
.vmi_dependson = VNET_MOD_NET,
|
|
|
|
|
.vmi_iattach = vnet_gif_iattach
|
|
|
|
|
};
|
|
|
|
|
#endif
|
|
|
|
|
|
Major overhaul of pseudo-interface cloning. Highlights include:
- Split the code out into if_clone.[ch].
- Locked struct if_clone. [1]
- Add a per-cloner match function rather then simply matching names of
the form <name><unit> and <name>.
- Use the match function to allow creation of <interface>.<tag>
vlan interfaces. The old way is preserved unchanged!
- Also the match function to allow creation of stf(4) interfaces named
stf0, stf, or 6to4. This is the only major user visible change in
that "ifconfig stf" creates the interface stf rather then stf0 and
does not print "stf0" to stdout.
- Allow destroy functions to fail so they can refuse to delete
interfaces. Currently, we forbid the deletion of interfaces which
were created in the init function, particularly lo0, pflog0, and
pfsync0. In the case of lo0 this was a panic implementation so it
does not count as a user visiable change. :-)
- Since most interfaces do not need the new functionality, an family of
wrapper functions, ifc_simple_*(), were created to wrap old style
cloner functions.
- The IF_CLONE_INITIALIZER macro is replaced with a new incompatible
IFC_CLONE_INITIALIZER and ifc_simple consumers use IFC_SIMPLE_DECLARE
instead.
Submitted by: Maurycy Pawlowski-Wieronski <maurycy at fouk.org> [1]
Reviewed by: andre, mlaier
Discussed on: net
2004-06-22 20:13:25 +00:00
|
|
|
IFC_SIMPLE_DECLARE(gif, 0);
|
2000-07-04 16:35:15 +00:00
|
|
|
|
2002-03-19 21:54:18 +00:00
|
|
|
static int gifmodevent(module_t, int, void *);
|
1999-12-07 17:39:16 +00:00
|
|
|
|
2002-02-26 01:56:56 +00:00
|
|
|
SYSCTL_DECL(_net_link);
|
|
|
|
|
SYSCTL_NODE(_net_link, IFT_GIF, gif, CTLFLAG_RW, 0,
|
|
|
|
|
"Generic Tunnel Interface");
|
2000-07-04 16:35:15 +00:00
|
|
|
#ifndef MAX_GIF_NEST
|
|
|
|
|
/*
|
2002-02-26 01:56:56 +00:00
|
|
|
* This macro controls the default upper limitation on nesting of gif tunnels.
|
2000-07-04 16:35:15 +00:00
|
|
|
* Since, setting a large value to this macro with a careless configuration
|
|
|
|
|
* may introduce system crash, we don't allow any nestings by default.
|
|
|
|
|
* If you need to configure nested gif tunnels, you can define this macro
|
2002-04-19 04:46:24 +00:00
|
|
|
* in your kernel configuration file. However, if you do so, please be
|
2000-07-04 16:35:15 +00:00
|
|
|
* careful to configure the tunnels so that it won't make a loop.
|
|
|
|
|
*/
|
|
|
|
|
#define MAX_GIF_NEST 1
|
|
|
|
|
#endif
|
Step 1.5 of importing the network stack virtualization infrastructure
from the vimage project, as per plan established at devsummit 08/08:
http://wiki.freebsd.org/Image/Notes200808DevSummit
Introduce INIT_VNET_*() initializer macros, VNET_FOREACH() iterator
macros, and CURVNET_SET() context setting macros, all currently
resolving to NOPs.
Prepare for virtualization of selected SYSCTL objects by introducing a
family of SYSCTL_V_*() macros, currently resolving to their global
counterparts, i.e. SYSCTL_V_INT() == SYSCTL_INT().
Move selected #defines from sys/sys/vimage.h to newly introduced header
files specific to virtualized subsystems (sys/net/vnet.h,
sys/netinet/vinet.h etc.).
All the changes are verified to have zero functional impact at this
point in time by doing MD5 comparision between pre- and post-change
object files(*).
(*) netipsec/keysock.c did not validate depending on compile time options.
Implemented by: julian, bz, brooks, zec
Reviewed by: julian, bz, brooks, kris, rwatson, ...
Approved by: julian (mentor)
Obtained from: //depot/projects/vimage-commit2/...
X-MFC after: never
Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-10-02 15:37:58 +00:00
|
|
|
SYSCTL_V_INT(V_NET, vnet_gif, _net_link_gif, OID_AUTO, max_nesting,
|
|
|
|
|
CTLFLAG_RW, max_gif_nesting, 0, "Max nested tunnels");
|
|
|
|
|
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
SYSCTL_DECL(_net_inet6_ip6);
|
|
|
|
|
SYSCTL_V_INT(V_NET, vnet_gif, _net_inet6_ip6, IPV6CTL_GIF_HLIM,
|
|
|
|
|
gifhlim, CTLFLAG_RW, ip6_gif_hlim, 0, "");
|
|
|
|
|
#endif
|
2002-02-26 01:56:56 +00:00
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* By default, we disallow creation of multiple tunnels between the same
|
|
|
|
|
* pair of addresses. Some applications require this functionality so
|
|
|
|
|
* we allow control over this check here.
|
|
|
|
|
*/
|
Step 1.5 of importing the network stack virtualization infrastructure
from the vimage project, as per plan established at devsummit 08/08:
http://wiki.freebsd.org/Image/Notes200808DevSummit
Introduce INIT_VNET_*() initializer macros, VNET_FOREACH() iterator
macros, and CURVNET_SET() context setting macros, all currently
resolving to NOPs.
Prepare for virtualization of selected SYSCTL objects by introducing a
family of SYSCTL_V_*() macros, currently resolving to their global
counterparts, i.e. SYSCTL_V_INT() == SYSCTL_INT().
Move selected #defines from sys/sys/vimage.h to newly introduced header
files specific to virtualized subsystems (sys/net/vnet.h,
sys/netinet/vinet.h etc.).
All the changes are verified to have zero functional impact at this
point in time by doing MD5 comparision between pre- and post-change
object files(*).
(*) netipsec/keysock.c did not validate depending on compile time options.
Implemented by: julian, bz, brooks, zec
Reviewed by: julian, bz, brooks, kris, rwatson, ...
Approved by: julian (mentor)
Obtained from: //depot/projects/vimage-commit2/...
X-MFC after: never
Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-10-02 15:37:58 +00:00
|
|
|
SYSCTL_V_INT(V_NET, vnet_gif, _net_link_gif, OID_AUTO, parallel_tunnels,
|
|
|
|
|
CTLFLAG_RW, parallel_tunnels, 0, "Allow parallel tunnels?");
|
1999-12-07 17:39:16 +00:00
|
|
|
|
2008-03-06 19:02:37 +00:00
|
|
|
/* copy from src/sys/net/if_ethersubr.c */
|
|
|
|
|
static const u_char etherbroadcastaddr[ETHER_ADDR_LEN] =
|
|
|
|
|
{ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
|
|
|
|
|
#ifndef ETHER_IS_BROADCAST
|
|
|
|
|
#define ETHER_IS_BROADCAST(addr) \
|
|
|
|
|
(bcmp(etherbroadcastaddr, (addr), ETHER_ADDR_LEN) == 0)
|
|
|
|
|
#endif
|
|
|
|
|
|
2004-04-14 00:57:49 +00:00
|
|
|
static int
|
2006-07-09 06:04:01 +00:00
|
|
|
gif_clone_create(ifc, unit, params)
|
2001-07-02 21:02:09 +00:00
|
|
|
struct if_clone *ifc;
|
2002-03-11 09:26:07 +00:00
|
|
|
int unit;
|
2006-07-09 06:04:01 +00:00
|
|
|
caddr_t params;
|
1999-12-07 17:39:16 +00:00
|
|
|
{
|
Step 1.5 of importing the network stack virtualization infrastructure
from the vimage project, as per plan established at devsummit 08/08:
http://wiki.freebsd.org/Image/Notes200808DevSummit
Introduce INIT_VNET_*() initializer macros, VNET_FOREACH() iterator
macros, and CURVNET_SET() context setting macros, all currently
resolving to NOPs.
Prepare for virtualization of selected SYSCTL objects by introducing a
family of SYSCTL_V_*() macros, currently resolving to their global
counterparts, i.e. SYSCTL_V_INT() == SYSCTL_INT().
Move selected #defines from sys/sys/vimage.h to newly introduced header
files specific to virtualized subsystems (sys/net/vnet.h,
sys/netinet/vinet.h etc.).
All the changes are verified to have zero functional impact at this
point in time by doing MD5 comparision between pre- and post-change
object files(*).
(*) netipsec/keysock.c did not validate depending on compile time options.
Implemented by: julian, bz, brooks, zec
Reviewed by: julian, bz, brooks, kris, rwatson, ...
Approved by: julian (mentor)
Obtained from: //depot/projects/vimage-commit2/...
X-MFC after: never
Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-10-02 15:37:58 +00:00
|
|
|
INIT_VNET_GIF(curvnet);
|
2001-06-11 12:39:29 +00:00
|
|
|
struct gif_softc *sc;
|
1999-12-07 17:39:16 +00:00
|
|
|
|
2004-07-06 03:26:26 +00:00
|
|
|
sc = malloc(sizeof(struct gif_softc), M_GIF, M_WAITOK | M_ZERO);
|
Add code to allow the system to handle multiple routing tables.
This particular implementation is designed to be fully backwards compatible
and to be MFC-able to 7.x (and 6.x)
Currently the only protocol that can make use of the multiple tables is IPv4
Similar functionality exists in OpenBSD and Linux.
From my notes:
-----
One thing where FreeBSD has been falling behind, and which by chance I
have some time to work on is "policy based routing", which allows
different
packet streams to be routed by more than just the destination address.
Constraints:
------------
I want to make some form of this available in the 6.x tree
(and by extension 7.x) , but FreeBSD in general needs it so I might as
well do it in -current and back port the portions I need.
One of the ways that this can be done is to have the ability to
instantiate multiple kernel routing tables (which I will now
refer to as "Forwarding Information Bases" or "FIBs" for political
correctness reasons). Which FIB a particular packet uses to make
the next hop decision can be decided by a number of mechanisms.
The policies these mechanisms implement are the "Policies" referred
to in "Policy based routing".
One of the constraints I have if I try to back port this work to
6.x is that it must be implemented as a EXTENSION to the existing
ABIs in 6.x so that third party applications do not need to be
recompiled in timespan of the branch.
This first version will not have some of the bells and whistles that
will come with later versions. It will, for example, be limited to 16
tables in the first commit.
Implementation method, Compatible version. (part 1)
-------------------------------
For this reason I have implemented a "sufficient subset" of a
multiple routing table solution in Perforce, and back-ported it
to 6.x. (also in Perforce though not always caught up with what I
have done in -current/P4). The subset allows a number of FIBs
to be defined at compile time (8 is sufficient for my purposes in 6.x)
and implements the changes needed to allow IPV4 to use them. I have not
done the changes for ipv6 simply because I do not need it, and I do not
have enough knowledge of ipv6 (e.g. neighbor discovery) needed to do it.
Other protocol families are left untouched and should there be
users with proprietary protocol families, they should continue to work
and be oblivious to the existence of the extra FIBs.
To understand how this is done, one must know that the current FIB
code starts everything off with a single dimensional array of
pointers to FIB head structures (One per protocol family), each of
which in turn points to the trie of routes available to that family.
The basic change in the ABI compatible version of the change is to
extent that array to be a 2 dimensional array, so that
instead of protocol family X looking at rt_tables[X] for the
table it needs, it looks at rt_tables[Y][X] when for all
protocol families except ipv4 Y is always 0.
Code that is unaware of the change always just sees the first row
of the table, which of course looks just like the one dimensional
array that existed before.
The entry points rtrequest(), rtalloc(), rtalloc1(), rtalloc_ign()
are all maintained, but refer only to the first row of the array,
so that existing callers in proprietary protocols can continue to
do the "right thing".
Some new entry points are added, for the exclusive use of ipv4 code
called in_rtrequest(), in_rtalloc(), in_rtalloc1() and in_rtalloc_ign(),
which have an extra argument which refers the code to the correct row.
In addition, there are some new entry points (currently called
rtalloc_fib() and friends) that check the Address family being
looked up and call either rtalloc() (and friends) if the protocol
is not IPv4 forcing the action to row 0 or to the appropriate row
if it IS IPv4 (and that info is available). These are for calling
from code that is not specific to any particular protocol. The way
these are implemented would change in the non ABI preserving code
to be added later.
One feature of the first version of the code is that for ipv4,
the interface routes show up automatically on all the FIBs, so
that no matter what FIB you select you always have the basic
direct attached hosts available to you. (rtinit() does this
automatically).
You CAN delete an interface route from one FIB should you want
to but by default it's there. ARP information is also available
in each FIB. It's assumed that the same machine would have the
same MAC address, regardless of which FIB you are using to get
to it.
This brings us as to how the correct FIB is selected for an outgoing
IPV4 packet.
Firstly, all packets have a FIB associated with them. if nothing
has been done to change it, it will be FIB 0. The FIB is changed
in the following ways.
Packets fall into one of a number of classes.
1/ locally generated packets, coming from a socket/PCB.
Such packets select a FIB from a number associated with the
socket/PCB. This in turn is inherited from the process,
but can be changed by a socket option. The process in turn
inherits it on fork. I have written a utility call setfib
that acts a bit like nice..
setfib -3 ping target.example.com # will use fib 3 for ping.
It is an obvious extension to make it a property of a jail
but I have not done so. It can be achieved by combining the setfib and
jail commands.
2/ packets received on an interface for forwarding.
By default these packets would use table 0,
(or possibly a number settable in a sysctl(not yet)).
but prior to routing the firewall can inspect them (see below).
(possibly in the future you may be able to associate a FIB
with packets received on an interface.. An ifconfig arg, but not yet.)
3/ packets inspected by a packet classifier, which can arbitrarily
associate a fib with it on a packet by packet basis.
A fib assigned to a packet by a packet classifier
(such as ipfw) would over-ride a fib associated by
a more default source. (such as cases 1 or 2).
4/ a tcp listen socket associated with a fib will generate
accept sockets that are associated with that same fib.
5/ Packets generated in response to some other packet (e.g. reset
or icmp packets). These should use the FIB associated with the
packet being reponded to.
6/ Packets generated during encapsulation.
gif, tun and other tunnel interfaces will encapsulate using the FIB
that was in effect withthe proces that set up the tunnel.
thus setfib 1 ifconfig gif0 [tunnel instructions]
will set the fib for the tunnel to use to be fib 1.
Routing messages would be associated with their
process, and thus select one FIB or another.
messages from the kernel would be associated with the fib they
refer to and would only be received by a routing socket associated
with that fib. (not yet implemented)
In addition Netstat has been edited to be able to cope with the
fact that the array is now 2 dimensional. (It looks in system
memory using libkvm (!)). Old versions of netstat see only the first FIB.
In addition two sysctls are added to give:
a) the number of FIBs compiled in (active)
b) the default FIB of the calling process.
Early testing experience:
-------------------------
Basically our (IronPort's) appliance does this functionality already
using ipfw fwd but that method has some drawbacks.
For example,
It can't fully simulate a routing table because it can't influence the
socket's choice of local address when a connect() is done.
Testing during the generating of these changes has been
remarkably smooth so far. Multiple tables have co-existed
with no notable side effects, and packets have been routes
accordingly.
ipfw has grown 2 new keywords:
setfib N ip from anay to any
count ip from any to any fib N
In pf there seems to be a requirement to be able to give symbolic names to the
fibs but I do not have that capacity. I am not sure if it is required.
SCTP has interestingly enough built in support for this, called VRFs
in Cisco parlance. it will be interesting to see how that handles it
when it suddenly actually does something.
Where to next:
--------------------
After committing the ABI compatible version and MFCing it, I'd
like to proceed in a forward direction in -current. this will
result in some roto-tilling in the routing code.
Firstly: the current code's idea of having a separate tree per
protocol family, all of the same format, and pointed to by the
1 dimensional array is a bit silly. Especially when one considers that
there is code that makes assumptions about every protocol having the
same internal structures there. Some protocols don't WANT that
sort of structure. (for example the whole idea of a netmask is foreign
to appletalk). This needs to be made opaque to the external code.
My suggested first change is to add routing method pointers to the
'domain' structure, along with information pointing the data.
instead of having an array of pointers to uniform structures,
there would be an array pointing to the 'domain' structures
for each protocol address domain (protocol family),
and the methods this reached would be called. The methods would have
an argument that gives FIB number, but the protocol would be free
to ignore it.
When the ABI can be changed it raises the possibilty of the
addition of a fib entry into the "struct route". Currently,
the structure contains the sockaddr of the desination, and the resulting
fib entry. To make this work fully, one could add a fib number
so that given an address and a fib, one can find the third element, the
fib entry.
Interaction with the ARP layer/ LL layer would need to be
revisited as well. Qing Li has been working on this already.
This work was sponsored by Ironport Systems/Cisco
Reviewed by: several including rwatson, bz and mlair (parts each)
Obtained from: Ironport systems/Cisco
2008-05-09 23:03:00 +00:00
|
|
|
sc->gif_fibnum = curthread->td_proc->p_fibnum;
|
2005-06-10 16:49:24 +00:00
|
|
|
GIF2IFP(sc) = if_alloc(IFT_GIF);
|
|
|
|
|
if (GIF2IFP(sc) == NULL) {
|
|
|
|
|
free(sc, M_GIF);
|
|
|
|
|
return (ENOSPC);
|
|
|
|
|
}
|
2001-07-02 21:02:09 +00:00
|
|
|
|
2006-01-30 08:39:09 +00:00
|
|
|
GIF_LOCK_INIT(sc);
|
|
|
|
|
|
2005-06-10 16:49:24 +00:00
|
|
|
GIF2IFP(sc)->if_softc = sc;
|
|
|
|
|
if_initname(GIF2IFP(sc), ifc->ifc_name, unit);
|
2000-07-04 16:35:15 +00:00
|
|
|
|
2001-07-02 21:02:09 +00:00
|
|
|
sc->encap_cookie4 = sc->encap_cookie6 = NULL;
|
2000-07-04 16:35:15 +00:00
|
|
|
|
2005-06-10 16:49:24 +00:00
|
|
|
GIF2IFP(sc)->if_addrlen = 0;
|
|
|
|
|
GIF2IFP(sc)->if_mtu = GIF_MTU;
|
|
|
|
|
GIF2IFP(sc)->if_flags = IFF_POINTOPOINT | IFF_MULTICAST;
|
2001-06-11 12:39:29 +00:00
|
|
|
#if 0
|
2001-07-02 21:02:09 +00:00
|
|
|
/* turn off ingress filter */
|
2005-06-10 16:49:24 +00:00
|
|
|
GIF2IFP(sc)->if_flags |= IFF_LINK2;
|
|
|
|
|
#endif
|
|
|
|
|
GIF2IFP(sc)->if_ioctl = gif_ioctl;
|
2005-12-21 21:29:45 +00:00
|
|
|
GIF2IFP(sc)->if_start = gif_start;
|
2005-06-10 16:49:24 +00:00
|
|
|
GIF2IFP(sc)->if_output = gif_output;
|
|
|
|
|
GIF2IFP(sc)->if_snd.ifq_maxlen = IFQ_MAXLEN;
|
|
|
|
|
if_attach(GIF2IFP(sc));
|
2005-06-26 18:11:11 +00:00
|
|
|
bpfattach(GIF2IFP(sc), DLT_NULL, sizeof(u_int32_t));
|
2001-09-26 23:50:17 +00:00
|
|
|
if (ng_gif_attach_p != NULL)
|
2005-06-10 16:49:24 +00:00
|
|
|
(*ng_gif_attach_p)(GIF2IFP(sc));
|
2006-01-30 08:39:09 +00:00
|
|
|
|
|
|
|
|
mtx_lock(&gif_mtx);
|
Commit step 1 of the vimage project, (network stack)
virtualization work done by Marko Zec (zec@).
This is the first in a series of commits over the course
of the next few weeks.
Mark all uses of global variables to be virtualized
with a V_ prefix.
Use macros to map them back to their global names for
now, so this is a NOP change only.
We hope to have caught at least 85-90% of what is needed
so we do not invalidate a lot of outstanding patches again.
Obtained from: //depot/projects/vimage-commit2/...
Reviewed by: brooks, des, ed, mav, julian,
jamie, kris, rwatson, zec, ...
(various people I forgot, different versions)
md5 (with a bit of help)
Sponsored by: NLnet Foundation, The FreeBSD Foundation
X-MFC after: never
V_Commit_Message_Reviewed_By: more people than the patch
2008-08-17 23:27:27 +00:00
|
|
|
LIST_INSERT_HEAD(&V_gif_softc_list, sc, gif_list);
|
2006-01-30 08:39:09 +00:00
|
|
|
mtx_unlock(&gif_mtx);
|
|
|
|
|
|
|
|
|
|
return (0);
|
2001-07-02 21:02:09 +00:00
|
|
|
}
|
|
|
|
|
|
2004-03-22 15:43:14 +00:00
|
|
|
static void
|
2005-10-12 19:52:16 +00:00
|
|
|
gif_clone_destroy(ifp)
|
|
|
|
|
struct ifnet *ifp;
|
2001-07-02 21:02:09 +00:00
|
|
|
{
|
2008-11-05 11:37:26 +00:00
|
|
|
#if defined(INET) || defined(INET6)
|
2001-07-02 21:02:09 +00:00
|
|
|
int err;
|
2008-11-05 11:37:26 +00:00
|
|
|
#endif
|
2005-10-12 19:52:16 +00:00
|
|
|
struct gif_softc *sc = ifp->if_softc;
|
|
|
|
|
|
|
|
|
|
mtx_lock(&gif_mtx);
|
|
|
|
|
LIST_REMOVE(sc, gif_list);
|
|
|
|
|
mtx_unlock(&gif_mtx);
|
2001-07-02 21:02:09 +00:00
|
|
|
|
2004-03-22 15:43:14 +00:00
|
|
|
gif_delete_tunnel(ifp);
|
2002-10-16 19:49:37 +00:00
|
|
|
#ifdef INET6
|
2001-07-02 21:02:09 +00:00
|
|
|
if (sc->encap_cookie6 != NULL) {
|
|
|
|
|
err = encap_detach(sc->encap_cookie6);
|
|
|
|
|
KASSERT(err == 0, ("Unexpected error detaching encap_cookie6"));
|
|
|
|
|
}
|
2002-10-16 19:49:37 +00:00
|
|
|
#endif
|
|
|
|
|
#ifdef INET
|
|
|
|
|
if (sc->encap_cookie4 != NULL) {
|
|
|
|
|
err = encap_detach(sc->encap_cookie4);
|
|
|
|
|
KASSERT(err == 0, ("Unexpected error detaching encap_cookie4"));
|
|
|
|
|
}
|
|
|
|
|
#endif
|
2001-07-02 21:02:09 +00:00
|
|
|
|
2001-09-26 23:50:17 +00:00
|
|
|
if (ng_gif_detach_p != NULL)
|
|
|
|
|
(*ng_gif_detach_p)(ifp);
|
2001-07-02 21:02:09 +00:00
|
|
|
bpfdetach(ifp);
|
|
|
|
|
if_detach(ifp);
|
2005-06-10 16:49:24 +00:00
|
|
|
if_free(ifp);
|
2001-07-02 21:02:09 +00:00
|
|
|
|
2006-01-30 08:39:09 +00:00
|
|
|
GIF_LOCK_DESTROY(sc);
|
|
|
|
|
|
2001-07-02 21:02:09 +00:00
|
|
|
free(sc, M_GIF);
|
|
|
|
|
}
|
|
|
|
|
|
First pass at separating per-vnet initializer functions
from existing functions for initializing global state.
At this stage, the new per-vnet initializer functions are
directly called from the existing global initialization code,
which should in most cases result in compiler inlining those
new functions, hence yielding a near-zero functional change.
Modify the existing initializer functions which are invoked via
protosw, like ip_init() et. al., to allow them to be invoked
multiple times, i.e. per each vnet. Global state, if any,
is initialized only if such functions are called within the
context of vnet0, which will be determined via the
IS_DEFAULT_VNET(curvnet) check (currently always true).
While here, V_irtualize a few remaining global UMA zones
used by net/netinet/netipsec networking code. While it is
not yet clear to me or anybody else whether this is the right
thing to do, at this stage this makes the code more readable,
and makes it easier to track uncollected UMA-zone-backed
objects on vnet removal. In the long run, it's quite possible
that some form of shared use of UMA zone pools among multiple
vnets should be considered.
Bump __FreeBSD_version due to changes in layout of structs
vnet_ipfw, vnet_inet and vnet_net.
Approved by: julian (mentor)
2009-04-06 22:29:41 +00:00
|
|
|
static int
|
|
|
|
|
vnet_gif_iattach(const void *unused __unused)
|
|
|
|
|
{
|
|
|
|
|
INIT_VNET_GIF(curvnet);
|
|
|
|
|
|
|
|
|
|
LIST_INIT(&V_gif_softc_list);
|
|
|
|
|
V_max_gif_nesting = MAX_GIF_NEST;
|
|
|
|
|
#ifdef XBONEHACK
|
|
|
|
|
V_parallel_tunnels = 1;
|
|
|
|
|
#else
|
|
|
|
|
V_parallel_tunnels = 0;
|
|
|
|
|
#endif
|
|
|
|
|
V_ip_gif_ttl = GIF_TTL;
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
V_ip6_gif_hlim = GIF_HLIM;
|
|
|
|
|
#endif
|
|
|
|
|
|
|
|
|
|
return (0);
|
|
|
|
|
}
|
|
|
|
|
|
2001-07-02 21:02:09 +00:00
|
|
|
static int
|
|
|
|
|
gifmodevent(mod, type, data)
|
|
|
|
|
module_t mod;
|
|
|
|
|
int type;
|
|
|
|
|
void *data;
|
|
|
|
|
{
|
|
|
|
|
|
|
|
|
|
switch (type) {
|
|
|
|
|
case MOD_LOAD:
|
2004-03-22 15:43:14 +00:00
|
|
|
mtx_init(&gif_mtx, "gif_mtx", NULL, MTX_DEF);
|
2001-07-02 21:02:09 +00:00
|
|
|
|
Introduce vnet module registration / initialization framework with
dependency tracking and ordering enforcement.
With this change, per-vnet initialization functions introduced with
r190787 are no longer directly called from traditional initialization
functions (which cc in most cases inlined to pre-r190787 code), but are
instead registered via the vnet framework first, and are invoked only
after all prerequisite modules have been initialized. In the long run,
this framework should allow us to both initialize and dismantle
multiple vnet instances in a correct order.
The problem this change aims to solve is how to replay the
initialization sequence of various network stack components, which
have been traditionally triggered via different mechanisms (SYSINIT,
protosw). Note that this initialization sequence was and still can be
subtly different depending on whether certain pieces of code have been
statically compiled into the kernel, loaded as modules by boot
loader, or kldloaded at run time.
The approach is simple - we record the initialization sequence
established by the traditional mechanisms whenever vnet_mod_register()
is called for a particular vnet module. The vnet_mod_register_multi()
variant allows a single initializer function to be registered multiple
times but with different arguments - currently this is only used in
kern/uipc_domain.c by net_add_domain() with different struct domain *
as arguments, which allows for protosw-registered initialization
routines to be invoked in a correct order by the new vnet
initialization framework.
For the purpose of identifying vnet modules, each vnet module has to
have a unique ID, which is statically assigned in sys/vimage.h.
Dynamic assignment of vnet module IDs is not supported yet.
A vnet module may specify a single prerequisite module at registration
time by filling in the vmi_dependson field of its vnet_modinfo struct
with the ID of the module it depends on. Unless specified otherwise,
all vnet modules depend on VNET_MOD_NET (container for ifnet list head,
rt_tables etc.), which thus has to and will always be initialized
first. The framework will panic if it detects any unresolved
dependencies before completing system initialization. Detection of
unresolved dependencies for vnet modules registered after boot
(kldloaded modules) is not provided.
Note that the fact that each module can specify only a single
prerequisite may become problematic in the long run. In particular,
INET6 depends on INET being already instantiated, due to TCP / UDP
structures residing in INET container. IPSEC also depends on INET,
which will in turn additionally complicate making INET6-only kernel
configs a reality.
The entire registration framework can be compiled out by turning on the
VIMAGE_GLOBALS kernel config option.
Reviewed by: bz
Approved by: julian (mentor)
2009-04-11 05:58:58 +00:00
|
|
|
#ifndef VIMAGE_GLOBALS
|
|
|
|
|
vnet_mod_register(&vnet_gif_modinfo);
|
|
|
|
|
#else
|
First pass at separating per-vnet initializer functions
from existing functions for initializing global state.
At this stage, the new per-vnet initializer functions are
directly called from the existing global initialization code,
which should in most cases result in compiler inlining those
new functions, hence yielding a near-zero functional change.
Modify the existing initializer functions which are invoked via
protosw, like ip_init() et. al., to allow them to be invoked
multiple times, i.e. per each vnet. Global state, if any,
is initialized only if such functions are called within the
context of vnet0, which will be determined via the
IS_DEFAULT_VNET(curvnet) check (currently always true).
While here, V_irtualize a few remaining global UMA zones
used by net/netinet/netipsec networking code. While it is
not yet clear to me or anybody else whether this is the right
thing to do, at this stage this makes the code more readable,
and makes it easier to track uncollected UMA-zone-backed
objects on vnet removal. In the long run, it's quite possible
that some form of shared use of UMA zone pools among multiple
vnets should be considered.
Bump __FreeBSD_version due to changes in layout of structs
vnet_ipfw, vnet_inet and vnet_net.
Approved by: julian (mentor)
2009-04-06 22:29:41 +00:00
|
|
|
vnet_gif_iattach(NULL);
|
Introduce vnet module registration / initialization framework with
dependency tracking and ordering enforcement.
With this change, per-vnet initialization functions introduced with
r190787 are no longer directly called from traditional initialization
functions (which cc in most cases inlined to pre-r190787 code), but are
instead registered via the vnet framework first, and are invoked only
after all prerequisite modules have been initialized. In the long run,
this framework should allow us to both initialize and dismantle
multiple vnet instances in a correct order.
The problem this change aims to solve is how to replay the
initialization sequence of various network stack components, which
have been traditionally triggered via different mechanisms (SYSINIT,
protosw). Note that this initialization sequence was and still can be
subtly different depending on whether certain pieces of code have been
statically compiled into the kernel, loaded as modules by boot
loader, or kldloaded at run time.
The approach is simple - we record the initialization sequence
established by the traditional mechanisms whenever vnet_mod_register()
is called for a particular vnet module. The vnet_mod_register_multi()
variant allows a single initializer function to be registered multiple
times but with different arguments - currently this is only used in
kern/uipc_domain.c by net_add_domain() with different struct domain *
as arguments, which allows for protosw-registered initialization
routines to be invoked in a correct order by the new vnet
initialization framework.
For the purpose of identifying vnet modules, each vnet module has to
have a unique ID, which is statically assigned in sys/vimage.h.
Dynamic assignment of vnet module IDs is not supported yet.
A vnet module may specify a single prerequisite module at registration
time by filling in the vmi_dependson field of its vnet_modinfo struct
with the ID of the module it depends on. Unless specified otherwise,
all vnet modules depend on VNET_MOD_NET (container for ifnet list head,
rt_tables etc.), which thus has to and will always be initialized
first. The framework will panic if it detects any unresolved
dependencies before completing system initialization. Detection of
unresolved dependencies for vnet modules registered after boot
(kldloaded modules) is not provided.
Note that the fact that each module can specify only a single
prerequisite may become problematic in the long run. In particular,
INET6 depends on INET being already instantiated, due to TCP / UDP
structures residing in INET container. IPSEC also depends on INET,
which will in turn additionally complicate making INET6-only kernel
configs a reality.
The entire registration framework can be compiled out by turning on the
VIMAGE_GLOBALS kernel config option.
Reviewed by: bz
Approved by: julian (mentor)
2009-04-11 05:58:58 +00:00
|
|
|
#endif
|
2008-11-19 09:39:34 +00:00
|
|
|
if_clone_attach(&gif_cloner);
|
2001-07-02 21:02:09 +00:00
|
|
|
|
|
|
|
|
break;
|
|
|
|
|
case MOD_UNLOAD:
|
|
|
|
|
if_clone_detach(&gif_cloner);
|
2004-03-22 15:43:14 +00:00
|
|
|
mtx_destroy(&gif_mtx);
|
2001-07-02 21:02:09 +00:00
|
|
|
#ifdef INET6
|
First pass at separating per-vnet initializer functions
from existing functions for initializing global state.
At this stage, the new per-vnet initializer functions are
directly called from the existing global initialization code,
which should in most cases result in compiler inlining those
new functions, hence yielding a near-zero functional change.
Modify the existing initializer functions which are invoked via
protosw, like ip_init() et. al., to allow them to be invoked
multiple times, i.e. per each vnet. Global state, if any,
is initialized only if such functions are called within the
context of vnet0, which will be determined via the
IS_DEFAULT_VNET(curvnet) check (currently always true).
While here, V_irtualize a few remaining global UMA zones
used by net/netinet/netipsec networking code. While it is
not yet clear to me or anybody else whether this is the right
thing to do, at this stage this makes the code more readable,
and makes it easier to track uncollected UMA-zone-backed
objects on vnet removal. In the long run, it's quite possible
that some form of shared use of UMA zone pools among multiple
vnets should be considered.
Bump __FreeBSD_version due to changes in layout of structs
vnet_ipfw, vnet_inet and vnet_net.
Approved by: julian (mentor)
2009-04-06 22:29:41 +00:00
|
|
|
V_ip6_gif_hlim = 0; /* XXX -> vnet_gif_idetach() */
|
2000-07-04 16:35:15 +00:00
|
|
|
#endif
|
2001-07-02 21:02:09 +00:00
|
|
|
break;
|
2004-07-15 08:26:07 +00:00
|
|
|
default:
|
|
|
|
|
return EOPNOTSUPP;
|
1999-12-07 17:39:16 +00:00
|
|
|
}
|
2001-07-02 21:02:09 +00:00
|
|
|
return 0;
|
1999-12-07 17:39:16 +00:00
|
|
|
}
|
|
|
|
|
|
2001-07-02 21:02:09 +00:00
|
|
|
static moduledata_t gif_mod = {
|
|
|
|
|
"if_gif",
|
|
|
|
|
gifmodevent,
|
|
|
|
|
0
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
DECLARE_MODULE(if_gif, gif_mod, SI_SUB_PSEUDO, SI_ORDER_ANY);
|
2001-09-26 23:37:15 +00:00
|
|
|
MODULE_VERSION(if_gif, 1);
|
1999-12-07 17:39:16 +00:00
|
|
|
|
2002-10-16 19:49:37 +00:00
|
|
|
int
|
2000-07-04 16:35:15 +00:00
|
|
|
gif_encapcheck(m, off, proto, arg)
|
|
|
|
|
const struct mbuf *m;
|
|
|
|
|
int off;
|
|
|
|
|
int proto;
|
|
|
|
|
void *arg;
|
|
|
|
|
{
|
|
|
|
|
struct ip ip;
|
|
|
|
|
struct gif_softc *sc;
|
|
|
|
|
|
|
|
|
|
sc = (struct gif_softc *)arg;
|
|
|
|
|
if (sc == NULL)
|
|
|
|
|
return 0;
|
|
|
|
|
|
2005-06-10 16:49:24 +00:00
|
|
|
if ((GIF2IFP(sc)->if_flags & IFF_UP) == 0)
|
2000-07-04 16:35:15 +00:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
|
|
/* no physical address */
|
|
|
|
|
if (!sc->gif_psrc || !sc->gif_pdst)
|
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
|
|
switch (proto) {
|
|
|
|
|
#ifdef INET
|
|
|
|
|
case IPPROTO_IPV4:
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
case IPPROTO_IPV6:
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
2005-12-21 21:29:45 +00:00
|
|
|
case IPPROTO_ETHERIP:
|
|
|
|
|
break;
|
|
|
|
|
|
2000-07-04 16:35:15 +00:00
|
|
|
default:
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
2002-10-17 17:42:46 +00:00
|
|
|
/* Bail on short packets */
|
|
|
|
|
if (m->m_pkthdr.len < sizeof(ip))
|
|
|
|
|
return 0;
|
|
|
|
|
|
2002-02-26 20:11:33 +00:00
|
|
|
m_copydata(m, 0, sizeof(ip), (caddr_t)&ip);
|
2000-07-04 16:35:15 +00:00
|
|
|
|
|
|
|
|
switch (ip.ip_v) {
|
|
|
|
|
#ifdef INET
|
|
|
|
|
case 4:
|
|
|
|
|
if (sc->gif_psrc->sa_family != AF_INET ||
|
|
|
|
|
sc->gif_pdst->sa_family != AF_INET)
|
|
|
|
|
return 0;
|
|
|
|
|
return gif_encapcheck4(m, off, proto, arg);
|
|
|
|
|
#endif
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
case 6:
|
2002-10-16 19:49:37 +00:00
|
|
|
if (m->m_pkthdr.len < sizeof(struct ip6_hdr))
|
|
|
|
|
return 0;
|
2000-07-04 16:35:15 +00:00
|
|
|
if (sc->gif_psrc->sa_family != AF_INET6 ||
|
|
|
|
|
sc->gif_pdst->sa_family != AF_INET6)
|
|
|
|
|
return 0;
|
|
|
|
|
return gif_encapcheck6(m, off, proto, arg);
|
|
|
|
|
#endif
|
|
|
|
|
default:
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2005-12-21 21:29:45 +00:00
|
|
|
static void
|
|
|
|
|
gif_start(struct ifnet *ifp)
|
|
|
|
|
{
|
|
|
|
|
struct gif_softc *sc;
|
|
|
|
|
struct mbuf *m;
|
|
|
|
|
|
|
|
|
|
sc = ifp->if_softc;
|
|
|
|
|
|
|
|
|
|
ifp->if_drv_flags |= IFF_DRV_OACTIVE;
|
|
|
|
|
for (;;) {
|
|
|
|
|
IFQ_DEQUEUE(&ifp->if_snd, m);
|
|
|
|
|
if (m == 0)
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
gif_output(ifp, m, sc->gif_pdst, NULL);
|
|
|
|
|
|
|
|
|
|
}
|
|
|
|
|
ifp->if_drv_flags &= ~IFF_DRV_OACTIVE;
|
|
|
|
|
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
|
1999-12-07 17:39:16 +00:00
|
|
|
int
|
2009-04-16 20:30:28 +00:00
|
|
|
gif_output(ifp, m, dst, ro)
|
1999-12-07 17:39:16 +00:00
|
|
|
struct ifnet *ifp;
|
|
|
|
|
struct mbuf *m;
|
|
|
|
|
struct sockaddr *dst;
|
2009-04-16 20:30:28 +00:00
|
|
|
struct route *ro;
|
1999-12-07 17:39:16 +00:00
|
|
|
{
|
Step 1.5 of importing the network stack virtualization infrastructure
from the vimage project, as per plan established at devsummit 08/08:
http://wiki.freebsd.org/Image/Notes200808DevSummit
Introduce INIT_VNET_*() initializer macros, VNET_FOREACH() iterator
macros, and CURVNET_SET() context setting macros, all currently
resolving to NOPs.
Prepare for virtualization of selected SYSCTL objects by introducing a
family of SYSCTL_V_*() macros, currently resolving to their global
counterparts, i.e. SYSCTL_V_INT() == SYSCTL_INT().
Move selected #defines from sys/sys/vimage.h to newly introduced header
files specific to virtualized subsystems (sys/net/vnet.h,
sys/netinet/vinet.h etc.).
All the changes are verified to have zero functional impact at this
point in time by doing MD5 comparision between pre- and post-change
object files(*).
(*) netipsec/keysock.c did not validate depending on compile time options.
Implemented by: julian, bz, brooks, zec
Reviewed by: julian, bz, brooks, kris, rwatson, ...
Approved by: julian (mentor)
Obtained from: //depot/projects/vimage-commit2/...
X-MFC after: never
Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-10-02 15:37:58 +00:00
|
|
|
INIT_VNET_GIF(ifp->if_vnet);
|
2005-06-10 16:49:24 +00:00
|
|
|
struct gif_softc *sc = ifp->if_softc;
|
2004-04-05 16:55:15 +00:00
|
|
|
struct m_tag *mtag;
|
1999-12-07 17:39:16 +00:00
|
|
|
int error = 0;
|
2004-04-05 16:55:15 +00:00
|
|
|
int gif_called;
|
2005-06-26 18:11:11 +00:00
|
|
|
u_int32_t af;
|
1999-12-07 17:39:16 +00:00
|
|
|
|
2002-08-01 21:00:05 +00:00
|
|
|
#ifdef MAC
|
2007-10-24 19:04:04 +00:00
|
|
|
error = mac_ifnet_check_transmit(ifp, m);
|
2002-08-12 16:08:23 +00:00
|
|
|
if (error) {
|
|
|
|
|
m_freem(m);
|
|
|
|
|
goto end;
|
|
|
|
|
}
|
2002-08-01 21:00:05 +00:00
|
|
|
#endif
|
|
|
|
|
|
1999-12-07 17:39:16 +00:00
|
|
|
/*
|
|
|
|
|
* gif may cause infinite recursion calls when misconfigured.
|
2004-04-05 16:55:15 +00:00
|
|
|
* We'll prevent this by detecting loops.
|
|
|
|
|
*
|
|
|
|
|
* High nesting level may cause stack exhaustion.
|
1999-12-07 17:39:16 +00:00
|
|
|
* We'll prevent this by introducing upper limit.
|
|
|
|
|
*/
|
2004-04-05 16:55:15 +00:00
|
|
|
gif_called = 1;
|
|
|
|
|
mtag = m_tag_locate(m, MTAG_GIF, MTAG_GIF_CALLED, NULL);
|
|
|
|
|
while (mtag != NULL) {
|
|
|
|
|
if (*(struct ifnet **)(mtag + 1) == ifp) {
|
|
|
|
|
log(LOG_NOTICE,
|
|
|
|
|
"gif_output: loop detected on %s\n",
|
|
|
|
|
(*(struct ifnet **)(mtag + 1))->if_xname);
|
|
|
|
|
m_freem(m);
|
|
|
|
|
error = EIO; /* is there better errno? */
|
|
|
|
|
goto end;
|
|
|
|
|
}
|
|
|
|
|
mtag = m_tag_locate(m, MTAG_GIF, MTAG_GIF_CALLED, mtag);
|
|
|
|
|
gif_called++;
|
|
|
|
|
}
|
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 (gif_called > V_max_gif_nesting) {
|
1999-12-07 17:39:16 +00:00
|
|
|
log(LOG_NOTICE,
|
|
|
|
|
"gif_output: recursively called too many times(%d)\n",
|
2004-03-22 14:24:26 +00:00
|
|
|
gif_called);
|
1999-12-07 17:39:16 +00:00
|
|
|
m_freem(m);
|
|
|
|
|
error = EIO; /* is there better errno? */
|
|
|
|
|
goto end;
|
|
|
|
|
}
|
2004-04-05 16:55:15 +00:00
|
|
|
mtag = m_tag_alloc(MTAG_GIF, MTAG_GIF_CALLED, sizeof(struct ifnet *),
|
|
|
|
|
M_NOWAIT);
|
|
|
|
|
if (mtag == NULL) {
|
|
|
|
|
m_freem(m);
|
|
|
|
|
error = ENOMEM;
|
|
|
|
|
goto end;
|
|
|
|
|
}
|
|
|
|
|
*(struct ifnet **)(mtag + 1) = ifp;
|
|
|
|
|
m_tag_prepend(m, mtag);
|
2000-07-04 16:35:15 +00:00
|
|
|
|
1999-12-07 17:39:16 +00:00
|
|
|
m->m_flags &= ~(M_BCAST|M_MCAST);
|
2006-01-30 08:39:09 +00:00
|
|
|
|
|
|
|
|
GIF_LOCK(sc);
|
|
|
|
|
|
1999-12-07 17:39:16 +00:00
|
|
|
if (!(ifp->if_flags & IFF_UP) ||
|
|
|
|
|
sc->gif_psrc == NULL || sc->gif_pdst == NULL) {
|
2006-06-02 14:10:52 +00:00
|
|
|
GIF_UNLOCK(sc);
|
1999-12-07 17:39:16 +00:00
|
|
|
m_freem(m);
|
|
|
|
|
error = ENETDOWN;
|
|
|
|
|
goto end;
|
|
|
|
|
}
|
|
|
|
|
|
2005-06-26 18:11:11 +00:00
|
|
|
/* BPF writes need to be handled specially. */
|
|
|
|
|
if (dst->sa_family == AF_UNSPEC) {
|
|
|
|
|
bcopy(dst->sa_data, &af, sizeof(af));
|
|
|
|
|
dst->sa_family = af;
|
|
|
|
|
}
|
|
|
|
|
|
2005-12-21 21:29:45 +00:00
|
|
|
af = dst->sa_family;
|
2006-06-02 19:59:33 +00:00
|
|
|
BPF_MTAP2(ifp, &af, sizeof(af), m);
|
2000-07-04 16:35:15 +00:00
|
|
|
ifp->if_opackets++;
|
1999-12-07 17:39:16 +00:00
|
|
|
ifp->if_obytes += m->m_pkthdr.len;
|
|
|
|
|
|
2005-12-21 21:29:45 +00:00
|
|
|
/* override to IPPROTO_ETHERIP for bridged traffic */
|
|
|
|
|
if (ifp->if_bridge)
|
|
|
|
|
af = AF_LINK;
|
|
|
|
|
|
Add code to allow the system to handle multiple routing tables.
This particular implementation is designed to be fully backwards compatible
and to be MFC-able to 7.x (and 6.x)
Currently the only protocol that can make use of the multiple tables is IPv4
Similar functionality exists in OpenBSD and Linux.
From my notes:
-----
One thing where FreeBSD has been falling behind, and which by chance I
have some time to work on is "policy based routing", which allows
different
packet streams to be routed by more than just the destination address.
Constraints:
------------
I want to make some form of this available in the 6.x tree
(and by extension 7.x) , but FreeBSD in general needs it so I might as
well do it in -current and back port the portions I need.
One of the ways that this can be done is to have the ability to
instantiate multiple kernel routing tables (which I will now
refer to as "Forwarding Information Bases" or "FIBs" for political
correctness reasons). Which FIB a particular packet uses to make
the next hop decision can be decided by a number of mechanisms.
The policies these mechanisms implement are the "Policies" referred
to in "Policy based routing".
One of the constraints I have if I try to back port this work to
6.x is that it must be implemented as a EXTENSION to the existing
ABIs in 6.x so that third party applications do not need to be
recompiled in timespan of the branch.
This first version will not have some of the bells and whistles that
will come with later versions. It will, for example, be limited to 16
tables in the first commit.
Implementation method, Compatible version. (part 1)
-------------------------------
For this reason I have implemented a "sufficient subset" of a
multiple routing table solution in Perforce, and back-ported it
to 6.x. (also in Perforce though not always caught up with what I
have done in -current/P4). The subset allows a number of FIBs
to be defined at compile time (8 is sufficient for my purposes in 6.x)
and implements the changes needed to allow IPV4 to use them. I have not
done the changes for ipv6 simply because I do not need it, and I do not
have enough knowledge of ipv6 (e.g. neighbor discovery) needed to do it.
Other protocol families are left untouched and should there be
users with proprietary protocol families, they should continue to work
and be oblivious to the existence of the extra FIBs.
To understand how this is done, one must know that the current FIB
code starts everything off with a single dimensional array of
pointers to FIB head structures (One per protocol family), each of
which in turn points to the trie of routes available to that family.
The basic change in the ABI compatible version of the change is to
extent that array to be a 2 dimensional array, so that
instead of protocol family X looking at rt_tables[X] for the
table it needs, it looks at rt_tables[Y][X] when for all
protocol families except ipv4 Y is always 0.
Code that is unaware of the change always just sees the first row
of the table, which of course looks just like the one dimensional
array that existed before.
The entry points rtrequest(), rtalloc(), rtalloc1(), rtalloc_ign()
are all maintained, but refer only to the first row of the array,
so that existing callers in proprietary protocols can continue to
do the "right thing".
Some new entry points are added, for the exclusive use of ipv4 code
called in_rtrequest(), in_rtalloc(), in_rtalloc1() and in_rtalloc_ign(),
which have an extra argument which refers the code to the correct row.
In addition, there are some new entry points (currently called
rtalloc_fib() and friends) that check the Address family being
looked up and call either rtalloc() (and friends) if the protocol
is not IPv4 forcing the action to row 0 or to the appropriate row
if it IS IPv4 (and that info is available). These are for calling
from code that is not specific to any particular protocol. The way
these are implemented would change in the non ABI preserving code
to be added later.
One feature of the first version of the code is that for ipv4,
the interface routes show up automatically on all the FIBs, so
that no matter what FIB you select you always have the basic
direct attached hosts available to you. (rtinit() does this
automatically).
You CAN delete an interface route from one FIB should you want
to but by default it's there. ARP information is also available
in each FIB. It's assumed that the same machine would have the
same MAC address, regardless of which FIB you are using to get
to it.
This brings us as to how the correct FIB is selected for an outgoing
IPV4 packet.
Firstly, all packets have a FIB associated with them. if nothing
has been done to change it, it will be FIB 0. The FIB is changed
in the following ways.
Packets fall into one of a number of classes.
1/ locally generated packets, coming from a socket/PCB.
Such packets select a FIB from a number associated with the
socket/PCB. This in turn is inherited from the process,
but can be changed by a socket option. The process in turn
inherits it on fork. I have written a utility call setfib
that acts a bit like nice..
setfib -3 ping target.example.com # will use fib 3 for ping.
It is an obvious extension to make it a property of a jail
but I have not done so. It can be achieved by combining the setfib and
jail commands.
2/ packets received on an interface for forwarding.
By default these packets would use table 0,
(or possibly a number settable in a sysctl(not yet)).
but prior to routing the firewall can inspect them (see below).
(possibly in the future you may be able to associate a FIB
with packets received on an interface.. An ifconfig arg, but not yet.)
3/ packets inspected by a packet classifier, which can arbitrarily
associate a fib with it on a packet by packet basis.
A fib assigned to a packet by a packet classifier
(such as ipfw) would over-ride a fib associated by
a more default source. (such as cases 1 or 2).
4/ a tcp listen socket associated with a fib will generate
accept sockets that are associated with that same fib.
5/ Packets generated in response to some other packet (e.g. reset
or icmp packets). These should use the FIB associated with the
packet being reponded to.
6/ Packets generated during encapsulation.
gif, tun and other tunnel interfaces will encapsulate using the FIB
that was in effect withthe proces that set up the tunnel.
thus setfib 1 ifconfig gif0 [tunnel instructions]
will set the fib for the tunnel to use to be fib 1.
Routing messages would be associated with their
process, and thus select one FIB or another.
messages from the kernel would be associated with the fib they
refer to and would only be received by a routing socket associated
with that fib. (not yet implemented)
In addition Netstat has been edited to be able to cope with the
fact that the array is now 2 dimensional. (It looks in system
memory using libkvm (!)). Old versions of netstat see only the first FIB.
In addition two sysctls are added to give:
a) the number of FIBs compiled in (active)
b) the default FIB of the calling process.
Early testing experience:
-------------------------
Basically our (IronPort's) appliance does this functionality already
using ipfw fwd but that method has some drawbacks.
For example,
It can't fully simulate a routing table because it can't influence the
socket's choice of local address when a connect() is done.
Testing during the generating of these changes has been
remarkably smooth so far. Multiple tables have co-existed
with no notable side effects, and packets have been routes
accordingly.
ipfw has grown 2 new keywords:
setfib N ip from anay to any
count ip from any to any fib N
In pf there seems to be a requirement to be able to give symbolic names to the
fibs but I do not have that capacity. I am not sure if it is required.
SCTP has interestingly enough built in support for this, called VRFs
in Cisco parlance. it will be interesting to see how that handles it
when it suddenly actually does something.
Where to next:
--------------------
After committing the ABI compatible version and MFCing it, I'd
like to proceed in a forward direction in -current. this will
result in some roto-tilling in the routing code.
Firstly: the current code's idea of having a separate tree per
protocol family, all of the same format, and pointed to by the
1 dimensional array is a bit silly. Especially when one considers that
there is code that makes assumptions about every protocol having the
same internal structures there. Some protocols don't WANT that
sort of structure. (for example the whole idea of a netmask is foreign
to appletalk). This needs to be made opaque to the external code.
My suggested first change is to add routing method pointers to the
'domain' structure, along with information pointing the data.
instead of having an array of pointers to uniform structures,
there would be an array pointing to the 'domain' structures
for each protocol address domain (protocol family),
and the methods this reached would be called. The methods would have
an argument that gives FIB number, but the protocol would be free
to ignore it.
When the ABI can be changed it raises the possibilty of the
addition of a fib entry into the "struct route". Currently,
the structure contains the sockaddr of the desination, and the resulting
fib entry. To make this work fully, one could add a fib number
so that given an address and a fib, one can find the third element, the
fib entry.
Interaction with the ARP layer/ LL layer would need to be
revisited as well. Qing Li has been working on this already.
This work was sponsored by Ironport Systems/Cisco
Reviewed by: several including rwatson, bz and mlair (parts each)
Obtained from: Ironport systems/Cisco
2008-05-09 23:03:00 +00:00
|
|
|
M_SETFIB(m, sc->gif_fibnum);
|
2001-06-11 12:39:29 +00:00
|
|
|
/* inner AF-specific encapsulation */
|
|
|
|
|
|
2000-07-04 16:35:15 +00:00
|
|
|
/* XXX should we check if our outer source is legal? */
|
|
|
|
|
|
2001-06-11 12:39:29 +00:00
|
|
|
/* dispatch to output logic based on outer AF */
|
1999-12-07 17:39:16 +00:00
|
|
|
switch (sc->gif_psrc->sa_family) {
|
|
|
|
|
#ifdef INET
|
|
|
|
|
case AF_INET:
|
2005-12-21 21:29:45 +00:00
|
|
|
error = in_gif_output(ifp, af, m);
|
1999-12-07 17:39:16 +00:00
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
case AF_INET6:
|
2005-12-21 21:29:45 +00:00
|
|
|
error = in6_gif_output(ifp, af, m);
|
1999-12-07 17:39:16 +00:00
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
default:
|
2000-07-04 16:35:15 +00:00
|
|
|
m_freem(m);
|
1999-12-07 17:39:16 +00:00
|
|
|
error = ENETDOWN;
|
|
|
|
|
}
|
|
|
|
|
|
2006-06-02 14:10:52 +00:00
|
|
|
GIF_UNLOCK(sc);
|
1999-12-07 17:39:16 +00:00
|
|
|
end:
|
2001-06-11 12:39:29 +00:00
|
|
|
if (error)
|
|
|
|
|
ifp->if_oerrors++;
|
2006-01-30 08:39:09 +00:00
|
|
|
return (error);
|
1999-12-07 17:39:16 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void
|
2002-10-17 17:39:56 +00:00
|
|
|
gif_input(m, af, ifp)
|
1999-12-07 17:39:16 +00:00
|
|
|
struct mbuf *m;
|
|
|
|
|
int af;
|
2002-10-17 17:39:56 +00:00
|
|
|
struct ifnet *ifp;
|
1999-12-07 17:39:16 +00:00
|
|
|
{
|
2005-12-21 21:29:45 +00:00
|
|
|
int isr, n;
|
|
|
|
|
struct etherip_header *eip;
|
2008-03-06 19:02:37 +00:00
|
|
|
struct ether_header *eh;
|
|
|
|
|
struct ifnet *oldifp;
|
1999-12-07 17:39:16 +00:00
|
|
|
|
2002-10-17 17:39:56 +00:00
|
|
|
if (ifp == NULL) {
|
1999-12-07 17:39:16 +00:00
|
|
|
/* just in case */
|
|
|
|
|
m_freem(m);
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
|
2002-10-17 17:39:56 +00:00
|
|
|
m->m_pkthdr.rcvif = ifp;
|
2002-08-01 21:00:05 +00:00
|
|
|
|
|
|
|
|
#ifdef MAC
|
2007-10-24 19:04:04 +00:00
|
|
|
mac_ifnet_create_mbuf(ifp, m);
|
2002-08-01 21:00:05 +00:00
|
|
|
#endif
|
|
|
|
|
|
2006-06-02 19:59:33 +00:00
|
|
|
if (bpf_peers_present(ifp->if_bpf)) {
|
2001-06-11 12:39:29 +00:00
|
|
|
u_int32_t af1 = af;
|
2003-12-28 03:56:00 +00:00
|
|
|
bpf_mtap2(ifp->if_bpf, &af1, sizeof(af1), m);
|
1999-12-07 17:39:16 +00:00
|
|
|
}
|
|
|
|
|
|
2001-09-26 23:50:17 +00:00
|
|
|
if (ng_gif_input_p != NULL) {
|
2002-10-17 17:39:56 +00:00
|
|
|
(*ng_gif_input_p)(ifp, &m, af);
|
2001-09-26 23:50:17 +00:00
|
|
|
if (m == NULL)
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
|
1999-12-07 17:39:16 +00:00
|
|
|
/*
|
|
|
|
|
* Put the packet to the network layer input queue according to the
|
|
|
|
|
* specified address family.
|
|
|
|
|
* Note: older versions of gif_input directly called network layer
|
2002-04-19 04:46:24 +00:00
|
|
|
* input functions, e.g. ip6_input, here. We changed the policy to
|
1999-12-07 17:39:16 +00:00
|
|
|
* prevent too many recursive calls of such input functions, which
|
2002-04-19 04:46:24 +00:00
|
|
|
* might cause kernel panic. But the change may introduce another
|
1999-12-07 17:39:16 +00:00
|
|
|
* problem; if the input queue is full, packets are discarded.
|
2002-04-19 04:46:24 +00:00
|
|
|
* The kernel stack overflow really happened, and we believed
|
|
|
|
|
* queue-full rarely occurs, so we changed the policy.
|
1999-12-07 17:39:16 +00:00
|
|
|
*/
|
|
|
|
|
switch (af) {
|
|
|
|
|
#ifdef INET
|
|
|
|
|
case AF_INET:
|
|
|
|
|
isr = NETISR_IP;
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
case AF_INET6:
|
|
|
|
|
isr = NETISR_IPV6;
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
2005-12-21 21:29:45 +00:00
|
|
|
case AF_LINK:
|
|
|
|
|
n = sizeof(struct etherip_header) + sizeof(struct ether_header);
|
|
|
|
|
if (n > m->m_len) {
|
|
|
|
|
m = m_pullup(m, n);
|
|
|
|
|
if (m == NULL) {
|
|
|
|
|
ifp->if_ierrors++;
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
eip = mtod(m, struct etherip_header *);
|
|
|
|
|
if (eip->eip_ver !=
|
|
|
|
|
(ETHERIP_VERSION & ETHERIP_VER_VERS_MASK)) {
|
|
|
|
|
/* discard unknown versions */
|
|
|
|
|
m_freem(m);
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
m_adj(m, sizeof(struct etherip_header));
|
|
|
|
|
|
|
|
|
|
m->m_flags &= ~(M_BCAST|M_MCAST);
|
|
|
|
|
m->m_pkthdr.rcvif = ifp;
|
|
|
|
|
|
2008-03-06 19:02:37 +00:00
|
|
|
if (ifp->if_bridge) {
|
|
|
|
|
oldifp = ifp;
|
|
|
|
|
eh = mtod(m, struct ether_header *);
|
|
|
|
|
if (ETHER_IS_MULTICAST(eh->ether_dhost)) {
|
|
|
|
|
if (ETHER_IS_BROADCAST(eh->ether_dhost))
|
|
|
|
|
m->m_flags |= M_BCAST;
|
|
|
|
|
else
|
|
|
|
|
m->m_flags |= M_MCAST;
|
|
|
|
|
ifp->if_imcasts++;
|
|
|
|
|
}
|
2005-12-21 21:29:45 +00:00
|
|
|
BRIDGE_INPUT(ifp, m);
|
2008-03-06 19:02:37 +00:00
|
|
|
|
|
|
|
|
if (m != NULL && ifp != oldifp) {
|
|
|
|
|
/*
|
|
|
|
|
* The bridge gave us back itself or one of the
|
|
|
|
|
* members for which the frame is addressed.
|
|
|
|
|
*/
|
|
|
|
|
ether_demux(ifp, m);
|
|
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
}
|
2005-12-21 21:29:45 +00:00
|
|
|
if (m != NULL)
|
|
|
|
|
m_freem(m);
|
|
|
|
|
return;
|
|
|
|
|
|
1999-12-07 17:39:16 +00:00
|
|
|
default:
|
2001-09-26 23:50:17 +00:00
|
|
|
if (ng_gif_input_orphan_p != NULL)
|
2002-10-17 17:39:56 +00:00
|
|
|
(*ng_gif_input_orphan_p)(ifp, m, af);
|
2001-09-26 23:50:17 +00:00
|
|
|
else
|
|
|
|
|
m_freem(m);
|
1999-12-07 17:39:16 +00:00
|
|
|
return;
|
|
|
|
|
}
|
|
|
|
|
|
2002-10-17 17:39:56 +00:00
|
|
|
ifp->if_ipackets++;
|
|
|
|
|
ifp->if_ibytes += m->m_pkthdr.len;
|
2003-03-04 23:19:55 +00:00
|
|
|
netisr_dispatch(isr, m);
|
1999-12-07 17:39:16 +00:00
|
|
|
}
|
|
|
|
|
|
2000-07-04 16:35:15 +00:00
|
|
|
/* XXX how should we handle IPv6 scope on SIOC[GS]IFPHYADDR? */
|
1999-12-07 17:39:16 +00:00
|
|
|
int
|
|
|
|
|
gif_ioctl(ifp, cmd, data)
|
|
|
|
|
struct ifnet *ifp;
|
|
|
|
|
u_long cmd;
|
|
|
|
|
caddr_t data;
|
|
|
|
|
{
|
2005-06-10 16:49:24 +00:00
|
|
|
struct gif_softc *sc = ifp->if_softc;
|
1999-12-07 17:39:16 +00:00
|
|
|
struct ifreq *ifr = (struct ifreq*)data;
|
|
|
|
|
int error = 0, size;
|
2000-07-04 16:35:15 +00:00
|
|
|
struct sockaddr *dst, *src;
|
2002-10-17 17:42:46 +00:00
|
|
|
#ifdef SIOCSIFMTU /* xxx */
|
|
|
|
|
u_long mtu;
|
|
|
|
|
#endif
|
|
|
|
|
|
1999-12-07 17:39:16 +00:00
|
|
|
switch (cmd) {
|
|
|
|
|
case SIOCSIFADDR:
|
2002-10-16 19:49:37 +00:00
|
|
|
ifp->if_flags |= IFF_UP;
|
1999-12-07 17:39:16 +00:00
|
|
|
break;
|
2000-07-04 16:35:15 +00:00
|
|
|
|
1999-12-07 17:39:16 +00:00
|
|
|
case SIOCSIFDSTADDR:
|
|
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case SIOCADDMULTI:
|
|
|
|
|
case SIOCDELMULTI:
|
|
|
|
|
break;
|
|
|
|
|
|
2000-07-04 16:35:15 +00:00
|
|
|
#ifdef SIOCSIFMTU /* xxx */
|
1999-12-07 17:39:16 +00:00
|
|
|
case SIOCGIFMTU:
|
|
|
|
|
break;
|
2000-07-04 16:35:15 +00:00
|
|
|
|
1999-12-07 17:39:16 +00:00
|
|
|
case SIOCSIFMTU:
|
2002-10-17 17:42:46 +00:00
|
|
|
mtu = ifr->ifr_mtu;
|
|
|
|
|
if (mtu < GIF_MTU_MIN || mtu > GIF_MTU_MAX)
|
|
|
|
|
return (EINVAL);
|
|
|
|
|
ifp->if_mtu = mtu;
|
1999-12-07 17:39:16 +00:00
|
|
|
break;
|
2000-07-04 16:35:15 +00:00
|
|
|
#endif /* SIOCSIFMTU */
|
1999-12-07 17:39:16 +00:00
|
|
|
|
2002-10-17 17:42:46 +00:00
|
|
|
#ifdef INET
|
1999-12-07 17:39:16 +00:00
|
|
|
case SIOCSIFPHYADDR:
|
2002-10-17 17:42:46 +00:00
|
|
|
#endif
|
1999-12-07 17:39:16 +00:00
|
|
|
#ifdef INET6
|
|
|
|
|
case SIOCSIFPHYADDR_IN6:
|
|
|
|
|
#endif /* INET6 */
|
2001-06-11 12:39:29 +00:00
|
|
|
case SIOCSLIFPHYADDR:
|
2000-07-04 16:35:15 +00:00
|
|
|
switch (cmd) {
|
2001-06-11 12:39:29 +00:00
|
|
|
#ifdef INET
|
2000-07-04 16:35:15 +00:00
|
|
|
case SIOCSIFPHYADDR:
|
1999-12-07 17:39:16 +00:00
|
|
|
src = (struct sockaddr *)
|
|
|
|
|
&(((struct in_aliasreq *)data)->ifra_addr);
|
|
|
|
|
dst = (struct sockaddr *)
|
|
|
|
|
&(((struct in_aliasreq *)data)->ifra_dstaddr);
|
2000-07-04 16:35:15 +00:00
|
|
|
break;
|
2001-06-11 12:39:29 +00:00
|
|
|
#endif
|
2000-07-04 16:35:15 +00:00
|
|
|
#ifdef INET6
|
|
|
|
|
case SIOCSIFPHYADDR_IN6:
|
|
|
|
|
src = (struct sockaddr *)
|
|
|
|
|
&(((struct in6_aliasreq *)data)->ifra_addr);
|
|
|
|
|
dst = (struct sockaddr *)
|
|
|
|
|
&(((struct in6_aliasreq *)data)->ifra_dstaddr);
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
2001-06-11 12:39:29 +00:00
|
|
|
case SIOCSLIFPHYADDR:
|
|
|
|
|
src = (struct sockaddr *)
|
|
|
|
|
&(((struct if_laddrreq *)data)->addr);
|
|
|
|
|
dst = (struct sockaddr *)
|
|
|
|
|
&(((struct if_laddrreq *)data)->dstaddr);
|
2002-10-16 19:49:37 +00:00
|
|
|
break;
|
2002-02-26 20:11:33 +00:00
|
|
|
default:
|
2002-10-16 19:49:37 +00:00
|
|
|
return EINVAL;
|
2001-06-11 12:39:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* sa_family must be equal */
|
|
|
|
|
if (src->sa_family != dst->sa_family)
|
|
|
|
|
return EINVAL;
|
|
|
|
|
|
|
|
|
|
/* validate sa_len */
|
|
|
|
|
switch (src->sa_family) {
|
|
|
|
|
#ifdef INET
|
|
|
|
|
case AF_INET:
|
|
|
|
|
if (src->sa_len != sizeof(struct sockaddr_in))
|
|
|
|
|
return EINVAL;
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
case AF_INET6:
|
|
|
|
|
if (src->sa_len != sizeof(struct sockaddr_in6))
|
|
|
|
|
return EINVAL;
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
default:
|
|
|
|
|
return EAFNOSUPPORT;
|
|
|
|
|
}
|
|
|
|
|
switch (dst->sa_family) {
|
|
|
|
|
#ifdef INET
|
|
|
|
|
case AF_INET:
|
|
|
|
|
if (dst->sa_len != sizeof(struct sockaddr_in))
|
|
|
|
|
return EINVAL;
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
case AF_INET6:
|
|
|
|
|
if (dst->sa_len != sizeof(struct sockaddr_in6))
|
|
|
|
|
return EINVAL;
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
default:
|
|
|
|
|
return EAFNOSUPPORT;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* check sa_family looks sane for the cmd */
|
|
|
|
|
switch (cmd) {
|
|
|
|
|
case SIOCSIFPHYADDR:
|
|
|
|
|
if (src->sa_family == AF_INET)
|
|
|
|
|
break;
|
|
|
|
|
return EAFNOSUPPORT;
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
case SIOCSIFPHYADDR_IN6:
|
|
|
|
|
if (src->sa_family == AF_INET6)
|
|
|
|
|
break;
|
|
|
|
|
return EAFNOSUPPORT;
|
|
|
|
|
#endif /* INET6 */
|
|
|
|
|
case SIOCSLIFPHYADDR:
|
|
|
|
|
/* checks done in the above */
|
|
|
|
|
break;
|
2000-07-04 16:35:15 +00:00
|
|
|
}
|
1999-12-07 17:39:16 +00:00
|
|
|
|
2005-06-10 16:49:24 +00:00
|
|
|
error = gif_set_tunnel(GIF2IFP(sc), src, dst);
|
1999-12-07 17:39:16 +00:00
|
|
|
break;
|
|
|
|
|
|
2000-07-04 16:35:15 +00:00
|
|
|
#ifdef SIOCDIFPHYADDR
|
|
|
|
|
case SIOCDIFPHYADDR:
|
2005-06-10 16:49:24 +00:00
|
|
|
gif_delete_tunnel(GIF2IFP(sc));
|
2000-07-04 16:35:15 +00:00
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
|
1999-12-07 17:39:16 +00:00
|
|
|
case SIOCGIFPSRCADDR:
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
case SIOCGIFPSRCADDR_IN6:
|
|
|
|
|
#endif /* INET6 */
|
|
|
|
|
if (sc->gif_psrc == NULL) {
|
|
|
|
|
error = EADDRNOTAVAIL;
|
|
|
|
|
goto bad;
|
|
|
|
|
}
|
|
|
|
|
src = sc->gif_psrc;
|
2001-06-11 12:39:29 +00:00
|
|
|
switch (cmd) {
|
1999-12-07 17:39:16 +00:00
|
|
|
#ifdef INET
|
2001-06-11 12:39:29 +00:00
|
|
|
case SIOCGIFPSRCADDR:
|
1999-12-07 17:39:16 +00:00
|
|
|
dst = &ifr->ifr_addr;
|
2001-06-11 12:39:29 +00:00
|
|
|
size = sizeof(ifr->ifr_addr);
|
1999-12-07 17:39:16 +00:00
|
|
|
break;
|
|
|
|
|
#endif /* INET */
|
|
|
|
|
#ifdef INET6
|
2001-06-11 12:39:29 +00:00
|
|
|
case SIOCGIFPSRCADDR_IN6:
|
1999-12-07 17:39:16 +00:00
|
|
|
dst = (struct sockaddr *)
|
|
|
|
|
&(((struct in6_ifreq *)data)->ifr_addr);
|
2001-06-11 12:39:29 +00:00
|
|
|
size = sizeof(((struct in6_ifreq *)data)->ifr_addr);
|
1999-12-07 17:39:16 +00:00
|
|
|
break;
|
|
|
|
|
#endif /* INET6 */
|
|
|
|
|
default:
|
|
|
|
|
error = EADDRNOTAVAIL;
|
|
|
|
|
goto bad;
|
|
|
|
|
}
|
2001-06-11 12:39:29 +00:00
|
|
|
if (src->sa_len > size)
|
|
|
|
|
return EINVAL;
|
|
|
|
|
bcopy((caddr_t)src, (caddr_t)dst, src->sa_len);
|
2005-07-25 12:31:43 +00:00
|
|
|
#ifdef INET6
|
|
|
|
|
if (dst->sa_family == AF_INET6) {
|
|
|
|
|
error = sa6_recoverscope((struct sockaddr_in6 *)dst);
|
|
|
|
|
if (error != 0)
|
|
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
#endif
|
1999-12-07 17:39:16 +00:00
|
|
|
break;
|
2000-07-04 16:35:15 +00:00
|
|
|
|
1999-12-07 17:39:16 +00:00
|
|
|
case SIOCGIFPDSTADDR:
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
case SIOCGIFPDSTADDR_IN6:
|
|
|
|
|
#endif /* INET6 */
|
|
|
|
|
if (sc->gif_pdst == NULL) {
|
|
|
|
|
error = EADDRNOTAVAIL;
|
|
|
|
|
goto bad;
|
|
|
|
|
}
|
|
|
|
|
src = sc->gif_pdst;
|
2001-06-11 12:39:29 +00:00
|
|
|
switch (cmd) {
|
1999-12-07 17:39:16 +00:00
|
|
|
#ifdef INET
|
2001-06-11 12:39:29 +00:00
|
|
|
case SIOCGIFPDSTADDR:
|
1999-12-07 17:39:16 +00:00
|
|
|
dst = &ifr->ifr_addr;
|
2001-06-11 12:39:29 +00:00
|
|
|
size = sizeof(ifr->ifr_addr);
|
1999-12-07 17:39:16 +00:00
|
|
|
break;
|
|
|
|
|
#endif /* INET */
|
|
|
|
|
#ifdef INET6
|
2001-06-11 12:39:29 +00:00
|
|
|
case SIOCGIFPDSTADDR_IN6:
|
1999-12-07 17:39:16 +00:00
|
|
|
dst = (struct sockaddr *)
|
|
|
|
|
&(((struct in6_ifreq *)data)->ifr_addr);
|
2001-06-11 12:39:29 +00:00
|
|
|
size = sizeof(((struct in6_ifreq *)data)->ifr_addr);
|
1999-12-07 17:39:16 +00:00
|
|
|
break;
|
|
|
|
|
#endif /* INET6 */
|
|
|
|
|
default:
|
|
|
|
|
error = EADDRNOTAVAIL;
|
|
|
|
|
goto bad;
|
|
|
|
|
}
|
2001-06-11 12:39:29 +00:00
|
|
|
if (src->sa_len > size)
|
|
|
|
|
return EINVAL;
|
|
|
|
|
bcopy((caddr_t)src, (caddr_t)dst, src->sa_len);
|
2005-07-25 12:31:43 +00:00
|
|
|
#ifdef INET6
|
|
|
|
|
if (dst->sa_family == AF_INET6) {
|
|
|
|
|
error = sa6_recoverscope((struct sockaddr_in6 *)dst);
|
|
|
|
|
if (error != 0)
|
|
|
|
|
return (error);
|
|
|
|
|
}
|
|
|
|
|
#endif
|
2001-06-11 12:39:29 +00:00
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case SIOCGLIFPHYADDR:
|
|
|
|
|
if (sc->gif_psrc == NULL || sc->gif_pdst == NULL) {
|
|
|
|
|
error = EADDRNOTAVAIL;
|
|
|
|
|
goto bad;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* copy src */
|
|
|
|
|
src = sc->gif_psrc;
|
|
|
|
|
dst = (struct sockaddr *)
|
|
|
|
|
&(((struct if_laddrreq *)data)->addr);
|
|
|
|
|
size = sizeof(((struct if_laddrreq *)data)->addr);
|
|
|
|
|
if (src->sa_len > size)
|
|
|
|
|
return EINVAL;
|
|
|
|
|
bcopy((caddr_t)src, (caddr_t)dst, src->sa_len);
|
|
|
|
|
|
|
|
|
|
/* copy dst */
|
|
|
|
|
src = sc->gif_pdst;
|
|
|
|
|
dst = (struct sockaddr *)
|
|
|
|
|
&(((struct if_laddrreq *)data)->dstaddr);
|
|
|
|
|
size = sizeof(((struct if_laddrreq *)data)->dstaddr);
|
|
|
|
|
if (src->sa_len > size)
|
|
|
|
|
return EINVAL;
|
|
|
|
|
bcopy((caddr_t)src, (caddr_t)dst, src->sa_len);
|
1999-12-07 17:39:16 +00:00
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
case SIOCSIFFLAGS:
|
2000-07-04 16:35:15 +00:00
|
|
|
/* if_ioctl() takes care of it */
|
1999-12-07 17:39:16 +00:00
|
|
|
break;
|
|
|
|
|
|
|
|
|
|
default:
|
|
|
|
|
error = EINVAL;
|
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
bad:
|
|
|
|
|
return error;
|
|
|
|
|
}
|
2001-07-02 21:02:09 +00:00
|
|
|
|
2004-03-22 15:43:14 +00:00
|
|
|
/*
|
|
|
|
|
* XXXRW: There's a general event-ordering issue here: the code to check
|
|
|
|
|
* if a given tunnel is already present happens before we perform a
|
|
|
|
|
* potentially blocking setup of the tunnel. This code needs to be
|
|
|
|
|
* re-ordered so that the check and replacement can be atomic using
|
|
|
|
|
* a mutex.
|
|
|
|
|
*/
|
2002-10-16 19:49:37 +00:00
|
|
|
int
|
|
|
|
|
gif_set_tunnel(ifp, src, dst)
|
|
|
|
|
struct ifnet *ifp;
|
|
|
|
|
struct sockaddr *src;
|
|
|
|
|
struct sockaddr *dst;
|
|
|
|
|
{
|
Step 1.5 of importing the network stack virtualization infrastructure
from the vimage project, as per plan established at devsummit 08/08:
http://wiki.freebsd.org/Image/Notes200808DevSummit
Introduce INIT_VNET_*() initializer macros, VNET_FOREACH() iterator
macros, and CURVNET_SET() context setting macros, all currently
resolving to NOPs.
Prepare for virtualization of selected SYSCTL objects by introducing a
family of SYSCTL_V_*() macros, currently resolving to their global
counterparts, i.e. SYSCTL_V_INT() == SYSCTL_INT().
Move selected #defines from sys/sys/vimage.h to newly introduced header
files specific to virtualized subsystems (sys/net/vnet.h,
sys/netinet/vinet.h etc.).
All the changes are verified to have zero functional impact at this
point in time by doing MD5 comparision between pre- and post-change
object files(*).
(*) netipsec/keysock.c did not validate depending on compile time options.
Implemented by: julian, bz, brooks, zec
Reviewed by: julian, bz, brooks, kris, rwatson, ...
Approved by: julian (mentor)
Obtained from: //depot/projects/vimage-commit2/...
X-MFC after: never
Sponsored by: NLnet Foundation, The FreeBSD Foundation
2008-10-02 15:37:58 +00:00
|
|
|
INIT_VNET_GIF(ifp->if_vnet);
|
2005-06-10 16:49:24 +00:00
|
|
|
struct gif_softc *sc = ifp->if_softc;
|
2002-10-16 19:49:37 +00:00
|
|
|
struct gif_softc *sc2;
|
|
|
|
|
struct sockaddr *osrc, *odst, *sa;
|
|
|
|
|
int error = 0;
|
|
|
|
|
|
2004-03-22 15:43:14 +00:00
|
|
|
mtx_lock(&gif_mtx);
|
Commit step 1 of the vimage project, (network stack)
virtualization work done by Marko Zec (zec@).
This is the first in a series of commits over the course
of the next few weeks.
Mark all uses of global variables to be virtualized
with a V_ prefix.
Use macros to map them back to their global names for
now, so this is a NOP change only.
We hope to have caught at least 85-90% of what is needed
so we do not invalidate a lot of outstanding patches again.
Obtained from: //depot/projects/vimage-commit2/...
Reviewed by: brooks, des, ed, mav, julian,
jamie, kris, rwatson, zec, ...
(various people I forgot, different versions)
md5 (with a bit of help)
Sponsored by: NLnet Foundation, The FreeBSD Foundation
X-MFC after: never
V_Commit_Message_Reviewed_By: more people than the patch
2008-08-17 23:27:27 +00:00
|
|
|
LIST_FOREACH(sc2, &V_gif_softc_list, gif_list) {
|
2002-10-16 19:49:37 +00:00
|
|
|
if (sc2 == sc)
|
|
|
|
|
continue;
|
|
|
|
|
if (!sc2->gif_pdst || !sc2->gif_psrc)
|
|
|
|
|
continue;
|
|
|
|
|
if (sc2->gif_pdst->sa_family != dst->sa_family ||
|
|
|
|
|
sc2->gif_pdst->sa_len != dst->sa_len ||
|
|
|
|
|
sc2->gif_psrc->sa_family != src->sa_family ||
|
|
|
|
|
sc2->gif_psrc->sa_len != src->sa_len)
|
|
|
|
|
continue;
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Disallow parallel tunnels unless instructed
|
|
|
|
|
* otherwise.
|
|
|
|
|
*/
|
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 (!V_parallel_tunnels &&
|
2002-10-16 19:49:37 +00:00
|
|
|
bcmp(sc2->gif_pdst, dst, dst->sa_len) == 0 &&
|
|
|
|
|
bcmp(sc2->gif_psrc, src, src->sa_len) == 0) {
|
|
|
|
|
error = EADDRNOTAVAIL;
|
2004-03-22 15:43:14 +00:00
|
|
|
mtx_unlock(&gif_mtx);
|
2002-10-16 19:49:37 +00:00
|
|
|
goto bad;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* XXX both end must be valid? (I mean, not 0.0.0.0) */
|
|
|
|
|
}
|
2004-03-22 15:43:14 +00:00
|
|
|
mtx_unlock(&gif_mtx);
|
2002-10-16 19:49:37 +00:00
|
|
|
|
|
|
|
|
/* XXX we can detach from both, but be polite just in case */
|
|
|
|
|
if (sc->gif_psrc)
|
|
|
|
|
switch (sc->gif_psrc->sa_family) {
|
|
|
|
|
#ifdef INET
|
|
|
|
|
case AF_INET:
|
|
|
|
|
(void)in_gif_detach(sc);
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
case AF_INET6:
|
|
|
|
|
(void)in6_gif_detach(sc);
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
osrc = sc->gif_psrc;
|
2003-02-19 05:47:46 +00:00
|
|
|
sa = (struct sockaddr *)malloc(src->sa_len, M_IFADDR, M_WAITOK);
|
2002-10-16 19:49:37 +00:00
|
|
|
bcopy((caddr_t)src, (caddr_t)sa, src->sa_len);
|
|
|
|
|
sc->gif_psrc = sa;
|
|
|
|
|
|
|
|
|
|
odst = sc->gif_pdst;
|
2003-02-19 05:47:46 +00:00
|
|
|
sa = (struct sockaddr *)malloc(dst->sa_len, M_IFADDR, M_WAITOK);
|
2002-10-16 19:49:37 +00:00
|
|
|
bcopy((caddr_t)dst, (caddr_t)sa, dst->sa_len);
|
|
|
|
|
sc->gif_pdst = sa;
|
|
|
|
|
|
|
|
|
|
switch (sc->gif_psrc->sa_family) {
|
|
|
|
|
#ifdef INET
|
|
|
|
|
case AF_INET:
|
|
|
|
|
error = in_gif_attach(sc);
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
case AF_INET6:
|
2005-07-25 12:31:43 +00:00
|
|
|
/*
|
|
|
|
|
* Check validity of the scope zone ID of the addresses, and
|
|
|
|
|
* convert it into the kernel internal form if necessary.
|
|
|
|
|
*/
|
|
|
|
|
error = sa6_embedscope((struct sockaddr_in6 *)sc->gif_psrc, 0);
|
|
|
|
|
if (error != 0)
|
|
|
|
|
break;
|
|
|
|
|
error = sa6_embedscope((struct sockaddr_in6 *)sc->gif_pdst, 0);
|
|
|
|
|
if (error != 0)
|
|
|
|
|
break;
|
2002-10-16 19:49:37 +00:00
|
|
|
error = in6_gif_attach(sc);
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
}
|
|
|
|
|
if (error) {
|
|
|
|
|
/* rollback */
|
|
|
|
|
free((caddr_t)sc->gif_psrc, M_IFADDR);
|
|
|
|
|
free((caddr_t)sc->gif_pdst, M_IFADDR);
|
|
|
|
|
sc->gif_psrc = osrc;
|
|
|
|
|
sc->gif_pdst = odst;
|
|
|
|
|
goto bad;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (osrc)
|
|
|
|
|
free((caddr_t)osrc, M_IFADDR);
|
|
|
|
|
if (odst)
|
|
|
|
|
free((caddr_t)odst, M_IFADDR);
|
|
|
|
|
|
|
|
|
|
bad:
|
|
|
|
|
if (sc->gif_psrc && sc->gif_pdst)
|
2005-08-09 10:20:02 +00:00
|
|
|
ifp->if_drv_flags |= IFF_DRV_RUNNING;
|
2002-10-16 19:49:37 +00:00
|
|
|
else
|
2005-08-09 10:20:02 +00:00
|
|
|
ifp->if_drv_flags &= ~IFF_DRV_RUNNING;
|
2002-10-16 19:49:37 +00:00
|
|
|
|
|
|
|
|
return error;
|
|
|
|
|
}
|
|
|
|
|
|
2001-07-02 21:02:09 +00:00
|
|
|
void
|
2002-10-16 19:49:37 +00:00
|
|
|
gif_delete_tunnel(ifp)
|
|
|
|
|
struct ifnet *ifp;
|
2001-07-02 21:02:09 +00:00
|
|
|
{
|
2005-06-10 16:49:24 +00:00
|
|
|
struct gif_softc *sc = ifp->if_softc;
|
2001-07-02 21:02:09 +00:00
|
|
|
|
|
|
|
|
if (sc->gif_psrc) {
|
|
|
|
|
free((caddr_t)sc->gif_psrc, M_IFADDR);
|
|
|
|
|
sc->gif_psrc = NULL;
|
|
|
|
|
}
|
|
|
|
|
if (sc->gif_pdst) {
|
|
|
|
|
free((caddr_t)sc->gif_pdst, M_IFADDR);
|
|
|
|
|
sc->gif_pdst = NULL;
|
|
|
|
|
}
|
2002-10-16 19:49:37 +00:00
|
|
|
/* it is safe to detach from both */
|
|
|
|
|
#ifdef INET
|
|
|
|
|
(void)in_gif_detach(sc);
|
|
|
|
|
#endif
|
|
|
|
|
#ifdef INET6
|
|
|
|
|
(void)in6_gif_detach(sc);
|
|
|
|
|
#endif
|
2006-06-29 07:23:49 +00:00
|
|
|
ifp->if_drv_flags &= ~IFF_DRV_RUNNING;
|
2001-07-02 21:02:09 +00:00
|
|
|
}
|
