freebsd-dev/sys/net/if_gre.c

833 lines
19 KiB
C
Raw Normal View History

/*-
* SPDX-License-Identifier: BSD-2-Clause-FreeBSD
*
* Copyright (c) 1998 The NetBSD Foundation, Inc.
* Copyright (c) 2014, 2018 Andrey V. Elsukov <ae@FreeBSD.org>
* All rights reserved.
*
* This code is derived from software contributed to The NetBSD Foundation
* by Heiko W.Rupp <hwr@pilhuhn.de>
*
* IPv6-over-GRE contributed by Gert Doering <gert@greenie.muc.de>
*
* 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.
*
* THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. 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 FOUNDATION 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.
*
* $NetBSD: if_gre.c,v 1.49 2003/12/11 00:22:29 itojun Exp $
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include "opt_inet.h"
#include "opt_inet6.h"
#include "opt_rss.h"
#include <sys/param.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/malloc.h>
2004-05-30 20:27:19 +00:00
#include <sys/module.h>
#include <sys/mbuf.h>
#include <sys/priv.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>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/sockio.h>
#include <sys/sx.h>
#include <sys/sysctl.h>
#include <sys/syslog.h>
#include <sys/systm.h>
#include <net/ethernet.h>
#include <net/if.h>
#include <net/if_var.h>
#include <net/if_clone.h>
#include <net/if_types.h>
#include <net/netisr.h>
#include <net/vnet.h>
#include <net/route.h>
#include <netinet/in.h>
#include <netinet/in_pcb.h>
#ifdef INET
#include <netinet/in_var.h>
#include <netinet/ip.h>
#include <netinet/ip_var.h>
#ifdef RSS
#include <netinet/in_rss.h>
#endif
#endif
#ifdef INET6
#include <netinet/ip6.h>
#include <netinet6/in6_var.h>
#include <netinet6/ip6_var.h>
#ifdef RSS
#include <netinet6/in6_rss.h>
#endif
#endif
#include <netinet/ip_encap.h>
#include <netinet/udp.h>
#include <net/bpf.h>
#include <net/if_gre.h>
#include <machine/in_cksum.h>
#include <security/mac/mac_framework.h>
#define GREMTU 1476
static const char grename[] = "gre";
MALLOC_DEFINE(M_GRE, grename, "Generic Routing Encapsulation");
static struct sx gre_ioctl_sx;
SX_SYSINIT(gre_ioctl_sx, &gre_ioctl_sx, "gre_ioctl");
static int gre_clone_create(struct if_clone *, int, caddr_t);
2002-10-16 22:26:32 +00:00
static void gre_clone_destroy(struct ifnet *);
VNET_DEFINE_STATIC(struct if_clone *, gre_cloner);
#define V_gre_cloner VNET(gre_cloner)
#ifdef VIMAGE
static void gre_reassign(struct ifnet *, struct vnet *, char *);
#endif
static void gre_qflush(struct ifnet *);
static int gre_transmit(struct ifnet *, struct mbuf *);
static int gre_ioctl(struct ifnet *, u_long, caddr_t);
static int gre_output(struct ifnet *, struct mbuf *,
const struct sockaddr *, struct route *);
static void gre_delete_tunnel(struct gre_softc *);
SYSCTL_DECL(_net_link);
static SYSCTL_NODE(_net_link, IFT_TUNNEL, gre, CTLFLAG_RW | CTLFLAG_MPSAFE, 0,
"Generic Routing Encapsulation");
#ifndef MAX_GRE_NEST
/*
* This macro controls the default upper limitation on nesting of gre tunnels.
* 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 gre tunnels, you can define this macro
* in your kernel configuration file. However, if you do so, please be
* careful to configure the tunnels so that it won't make a loop.
*/
#define MAX_GRE_NEST 1
#endif
VNET_DEFINE_STATIC(int, max_gre_nesting) = MAX_GRE_NEST;
#define V_max_gre_nesting VNET(max_gre_nesting)
SYSCTL_INT(_net_link_gre, OID_AUTO, max_nesting, CTLFLAG_RW | CTLFLAG_VNET,
&VNET_NAME(max_gre_nesting), 0, "Max nested tunnels");
static void
vnet_gre_init(const void *unused __unused)
{
V_gre_cloner = if_clone_simple(grename, gre_clone_create,
gre_clone_destroy, 0);
#ifdef INET
in_gre_init();
#endif
#ifdef INET6
in6_gre_init();
#endif
}
VNET_SYSINIT(vnet_gre_init, SI_SUB_PROTO_IFATTACHDOMAIN, SI_ORDER_ANY,
vnet_gre_init, NULL);
static void
vnet_gre_uninit(const void *unused __unused)
{
if_clone_detach(V_gre_cloner);
#ifdef INET
in_gre_uninit();
#endif
#ifdef INET6
in6_gre_uninit();
#endif
/* XXX: epoch_call drain */
}
VNET_SYSUNINIT(vnet_gre_uninit, SI_SUB_PROTO_IFATTACHDOMAIN, SI_ORDER_ANY,
vnet_gre_uninit, NULL);
static int
gre_clone_create(struct if_clone *ifc, int unit, caddr_t params)
{
struct gre_softc *sc;
2004-07-06 03:28:24 +00:00
sc = malloc(sizeof(struct gre_softc), M_GRE, M_WAITOK | M_ZERO);
sc->gre_fibnum = curthread->td_proc->p_fibnum;
GRE2IFP(sc) = if_alloc(IFT_TUNNEL);
GRE2IFP(sc)->if_softc = sc;
if_initname(GRE2IFP(sc), grename, unit);
GRE2IFP(sc)->if_mtu = GREMTU;
GRE2IFP(sc)->if_flags = IFF_POINTOPOINT|IFF_MULTICAST;
GRE2IFP(sc)->if_output = gre_output;
GRE2IFP(sc)->if_ioctl = gre_ioctl;
GRE2IFP(sc)->if_transmit = gre_transmit;
GRE2IFP(sc)->if_qflush = gre_qflush;
#ifdef VIMAGE
GRE2IFP(sc)->if_reassign = gre_reassign;
#endif
2015-10-03 09:15:23 +00:00
GRE2IFP(sc)->if_capabilities |= IFCAP_LINKSTATE;
GRE2IFP(sc)->if_capenable |= IFCAP_LINKSTATE;
if_attach(GRE2IFP(sc));
bpfattach(GRE2IFP(sc), DLT_NULL, sizeof(u_int32_t));
return (0);
}
#ifdef VIMAGE
static void
gre_reassign(struct ifnet *ifp, struct vnet *new_vnet __unused,
char *unused __unused)
{
struct gre_softc *sc;
sx_xlock(&gre_ioctl_sx);
sc = ifp->if_softc;
if (sc != NULL)
gre_delete_tunnel(sc);
sx_xunlock(&gre_ioctl_sx);
}
#endif /* VIMAGE */
static void
gre_clone_destroy(struct ifnet *ifp)
{
struct gre_softc *sc;
sx_xlock(&gre_ioctl_sx);
sc = ifp->if_softc;
gre_delete_tunnel(sc);
bpfdetach(ifp);
if_detach(ifp);
ifp->if_softc = NULL;
sx_xunlock(&gre_ioctl_sx);
GRE_WAIT();
if_free(ifp);
free(sc, M_GRE);
}
static int
gre_ioctl(struct ifnet *ifp, u_long cmd, caddr_t data)
{
struct ifreq *ifr = (struct ifreq *)data;
struct gre_softc *sc;
uint32_t opt;
int error;
switch (cmd) {
case SIOCSIFMTU:
/* XXX: */
if (ifr->ifr_mtu < 576)
return (EINVAL);
ifp->if_mtu = ifr->ifr_mtu;
return (0);
case SIOCSIFADDR:
ifp->if_flags |= IFF_UP;
case SIOCSIFFLAGS:
case SIOCADDMULTI:
case SIOCDELMULTI:
return (0);
case GRESADDRS:
case GRESADDRD:
case GREGADDRS:
case GREGADDRD:
case GRESPROTO:
case GREGPROTO:
return (EOPNOTSUPP);
}
sx_xlock(&gre_ioctl_sx);
sc = ifp->if_softc;
if (sc == NULL) {
error = ENXIO;
goto end;
}
error = 0;
switch (cmd) {
case SIOCDIFPHYADDR:
if (sc->gre_family == 0)
break;
gre_delete_tunnel(sc);
break;
#ifdef INET
case SIOCSIFPHYADDR:
case SIOCGIFPSRCADDR:
case SIOCGIFPDSTADDR:
error = in_gre_ioctl(sc, cmd, data);
break;
#endif
#ifdef INET6
case SIOCSIFPHYADDR_IN6:
case SIOCGIFPSRCADDR_IN6:
case SIOCGIFPDSTADDR_IN6:
error = in6_gre_ioctl(sc, cmd, data);
break;
#endif
case SIOCGTUNFIB:
ifr->ifr_fib = sc->gre_fibnum;
break;
case SIOCSTUNFIB:
if ((error = priv_check(curthread, PRIV_NET_GRE)) != 0)
break;
if (ifr->ifr_fib >= rt_numfibs)
error = EINVAL;
else
sc->gre_fibnum = ifr->ifr_fib;
break;
case GRESKEY:
case GRESOPTS:
case GRESPORT:
if ((error = priv_check(curthread, PRIV_NET_GRE)) != 0)
break;
if ((error = copyin(ifr_data_get_ptr(ifr), &opt,
sizeof(opt))) != 0)
break;
if (cmd == GRESKEY) {
if (sc->gre_key == opt)
break;
} else if (cmd == GRESOPTS) {
if (opt & ~GRE_OPTMASK) {
error = EINVAL;
break;
}
if (sc->gre_options == opt)
break;
} else if (cmd == GRESPORT) {
if (opt != 0 && (opt < V_ipport_hifirstauto ||
opt > V_ipport_hilastauto)) {
error = EINVAL;
break;
}
if (sc->gre_port == opt)
break;
if ((sc->gre_options & GRE_UDPENCAP) == 0) {
/*
* UDP encapsulation is not enabled, thus
* there is no need to reattach softc.
*/
sc->gre_port = opt;
break;
}
}
switch (sc->gre_family) {
#ifdef INET
case AF_INET:
error = in_gre_setopts(sc, cmd, opt);
break;
#endif
#ifdef INET6
case AF_INET6:
error = in6_gre_setopts(sc, cmd, opt);
break;
#endif
default:
/*
* Tunnel is not yet configured.
* We can just change any parameters.
*/
if (cmd == GRESKEY)
sc->gre_key = opt;
if (cmd == GRESOPTS)
sc->gre_options = opt;
if (cmd == GRESPORT)
sc->gre_port = opt;
break;
}
/*
* XXX: Do we need to initiate change of interface
* state here?
*/
break;
case GREGKEY:
error = copyout(&sc->gre_key, ifr_data_get_ptr(ifr),
sizeof(sc->gre_key));
break;
case GREGOPTS:
error = copyout(&sc->gre_options, ifr_data_get_ptr(ifr),
sizeof(sc->gre_options));
break;
case GREGPORT:
error = copyout(&sc->gre_port, ifr_data_get_ptr(ifr),
sizeof(sc->gre_port));
break;
default:
error = EINVAL;
break;
}
if (error == 0 && sc->gre_family != 0) {
if (
#ifdef INET
cmd == SIOCSIFPHYADDR ||
#endif
#ifdef INET6
cmd == SIOCSIFPHYADDR_IN6 ||
#endif
0) {
if_link_state_change(ifp, LINK_STATE_UP);
}
}
end:
sx_xunlock(&gre_ioctl_sx);
return (error);
}
static void
gre_delete_tunnel(struct gre_softc *sc)
{
struct gre_socket *gs;
sx_assert(&gre_ioctl_sx, SA_XLOCKED);
if (sc->gre_family != 0) {
CK_LIST_REMOVE(sc, chain);
CK_LIST_REMOVE(sc, srchash);
GRE_WAIT();
free(sc->gre_hdr, M_GRE);
sc->gre_family = 0;
}
/*
* If this Tunnel was the last one that could use UDP socket,
* we should unlink socket from hash table and close it.
*/
if ((gs = sc->gre_so) != NULL && CK_LIST_EMPTY(&gs->list)) {
CK_LIST_REMOVE(gs, chain);
soclose(gs->so);
NET_EPOCH_CALL(gre_sofree, &gs->epoch_ctx);
sc->gre_so = NULL;
}
GRE2IFP(sc)->if_drv_flags &= ~IFF_DRV_RUNNING;
if_link_state_change(GRE2IFP(sc), LINK_STATE_DOWN);
}
struct gre_list *
gre_hashinit(void)
{
struct gre_list *hash;
int i;
hash = malloc(sizeof(struct gre_list) * GRE_HASH_SIZE,
M_GRE, M_WAITOK);
for (i = 0; i < GRE_HASH_SIZE; i++)
CK_LIST_INIT(&hash[i]);
return (hash);
}
void
gre_hashdestroy(struct gre_list *hash)
{
free(hash, M_GRE);
}
void
gre_sofree(epoch_context_t ctx)
{
struct gre_socket *gs;
gs = __containerof(ctx, struct gre_socket, epoch_ctx);
free(gs, M_GRE);
}
static __inline uint16_t
gre_cksum_add(uint16_t sum, uint16_t a)
{
uint16_t res;
res = sum + a;
return (res + (res < a));
}
void
gre_update_udphdr(struct gre_softc *sc, struct udphdr *udp, uint16_t csum)
{
sx_assert(&gre_ioctl_sx, SA_XLOCKED);
MPASS(sc->gre_options & GRE_UDPENCAP);
udp->uh_dport = htons(GRE_UDPPORT);
udp->uh_sport = htons(sc->gre_port);
udp->uh_sum = csum;
udp->uh_ulen = 0;
}
void
gre_update_hdr(struct gre_softc *sc, struct grehdr *gh)
{
uint32_t *opts;
uint16_t flags;
sx_assert(&gre_ioctl_sx, SA_XLOCKED);
flags = 0;
opts = gh->gre_opts;
if (sc->gre_options & GRE_ENABLE_CSUM) {
flags |= GRE_FLAGS_CP;
sc->gre_hlen += 2 * sizeof(uint16_t);
*opts++ = 0;
}
if (sc->gre_key != 0) {
flags |= GRE_FLAGS_KP;
sc->gre_hlen += sizeof(uint32_t);
*opts++ = htonl(sc->gre_key);
}
if (sc->gre_options & GRE_ENABLE_SEQ) {
flags |= GRE_FLAGS_SP;
sc->gre_hlen += sizeof(uint32_t);
*opts++ = 0;
} else
sc->gre_oseq = 0;
gh->gre_flags = htons(flags);
}
int
Rework IP encapsulation handling code. Currently it has several disadvantages: - it uses single mutex to protect internal structures. It is used by data- and control- path, thus there are no parallelism at all. - it uses single list to keep encap handlers for both INET and INET6 families. - struct encaptab keeps unneeded information (src, dst, masks, protosw), that isn't used by code in the source tree. - matches are prioritized and when many tunneling interfaces are registered, encapcheck handler of each interface is invoked for each packet. The search takes O(n) for n interfaces. All this work is done with exclusive lock held. What this patch includes: - the datapath is converted to be lockless using epoch(9) KPI. - struct encaptab now linked using CK_LIST. - all unused fields removed from struct encaptab. Several new fields addedr: min_length is the minimum packet length, that encapsulation handler expects to see; exact_match is maximum number of bits, that can return an encapsulation handler, when it wants to consume a packet. - IPv6 and IPv4 handlers are stored in separate lists; - added new "encap_lookup_t" method, that will be used later. It is targeted to speedup lookup of needed interface, when gif(4)/gre(4) have many interfaces. - the need to use protosw structure is eliminated. The only pr_input method was used from this structure, so I don't see the need to keep using it. - encap_input_t method changed to avoid using mbuf tags to store softc pointer. Now it is passed directly trough encap_input_t method. encap_getarg() funtions is removed. - all sockaddr structures and code that uses them removed. We don't have any code in the tree that uses them. All consumers use encap_attach_func() method, that relies on invoking of encapcheck() to determine the needed handler. - introduced struct encap_config, it contains parameters of encap handler that is going to be registered by encap_attach() function. - encap handlers are stored in lists ordered by exact_match value, thus handlers that need more bits to match will be checked first, and if encapcheck method returns exact_match value, the search will be stopped. - all current consumers changed to use new KPI. Reviewed by: mmacy Sponsored by: Yandex LLC Differential Revision: https://reviews.freebsd.org/D15617
2018-06-05 20:51:01 +00:00
gre_input(struct mbuf *m, int off, int proto, void *arg)
{
Rework IP encapsulation handling code. Currently it has several disadvantages: - it uses single mutex to protect internal structures. It is used by data- and control- path, thus there are no parallelism at all. - it uses single list to keep encap handlers for both INET and INET6 families. - struct encaptab keeps unneeded information (src, dst, masks, protosw), that isn't used by code in the source tree. - matches are prioritized and when many tunneling interfaces are registered, encapcheck handler of each interface is invoked for each packet. The search takes O(n) for n interfaces. All this work is done with exclusive lock held. What this patch includes: - the datapath is converted to be lockless using epoch(9) KPI. - struct encaptab now linked using CK_LIST. - all unused fields removed from struct encaptab. Several new fields addedr: min_length is the minimum packet length, that encapsulation handler expects to see; exact_match is maximum number of bits, that can return an encapsulation handler, when it wants to consume a packet. - IPv6 and IPv4 handlers are stored in separate lists; - added new "encap_lookup_t" method, that will be used later. It is targeted to speedup lookup of needed interface, when gif(4)/gre(4) have many interfaces. - the need to use protosw structure is eliminated. The only pr_input method was used from this structure, so I don't see the need to keep using it. - encap_input_t method changed to avoid using mbuf tags to store softc pointer. Now it is passed directly trough encap_input_t method. encap_getarg() funtions is removed. - all sockaddr structures and code that uses them removed. We don't have any code in the tree that uses them. All consumers use encap_attach_func() method, that relies on invoking of encapcheck() to determine the needed handler. - introduced struct encap_config, it contains parameters of encap handler that is going to be registered by encap_attach() function. - encap handlers are stored in lists ordered by exact_match value, thus handlers that need more bits to match will be checked first, and if encapcheck method returns exact_match value, the search will be stopped. - all current consumers changed to use new KPI. Reviewed by: mmacy Sponsored by: Yandex LLC Differential Revision: https://reviews.freebsd.org/D15617
2018-06-05 20:51:01 +00:00
struct gre_softc *sc = arg;
struct grehdr *gh;
struct ifnet *ifp;
uint32_t *opts;
#ifdef notyet
uint32_t key;
#endif
uint16_t flags;
int hlen, isr, af;
ifp = GRE2IFP(sc);
Rework IP encapsulation handling code. Currently it has several disadvantages: - it uses single mutex to protect internal structures. It is used by data- and control- path, thus there are no parallelism at all. - it uses single list to keep encap handlers for both INET and INET6 families. - struct encaptab keeps unneeded information (src, dst, masks, protosw), that isn't used by code in the source tree. - matches are prioritized and when many tunneling interfaces are registered, encapcheck handler of each interface is invoked for each packet. The search takes O(n) for n interfaces. All this work is done with exclusive lock held. What this patch includes: - the datapath is converted to be lockless using epoch(9) KPI. - struct encaptab now linked using CK_LIST. - all unused fields removed from struct encaptab. Several new fields addedr: min_length is the minimum packet length, that encapsulation handler expects to see; exact_match is maximum number of bits, that can return an encapsulation handler, when it wants to consume a packet. - IPv6 and IPv4 handlers are stored in separate lists; - added new "encap_lookup_t" method, that will be used later. It is targeted to speedup lookup of needed interface, when gif(4)/gre(4) have many interfaces. - the need to use protosw structure is eliminated. The only pr_input method was used from this structure, so I don't see the need to keep using it. - encap_input_t method changed to avoid using mbuf tags to store softc pointer. Now it is passed directly trough encap_input_t method. encap_getarg() funtions is removed. - all sockaddr structures and code that uses them removed. We don't have any code in the tree that uses them. All consumers use encap_attach_func() method, that relies on invoking of encapcheck() to determine the needed handler. - introduced struct encap_config, it contains parameters of encap handler that is going to be registered by encap_attach() function. - encap handlers are stored in lists ordered by exact_match value, thus handlers that need more bits to match will be checked first, and if encapcheck method returns exact_match value, the search will be stopped. - all current consumers changed to use new KPI. Reviewed by: mmacy Sponsored by: Yandex LLC Differential Revision: https://reviews.freebsd.org/D15617
2018-06-05 20:51:01 +00:00
hlen = off + sizeof(struct grehdr) + 4 * sizeof(uint32_t);
if (m->m_pkthdr.len < hlen)
goto drop;
if (m->m_len < hlen) {
m = m_pullup(m, hlen);
if (m == NULL)
goto drop;
}
Rework IP encapsulation handling code. Currently it has several disadvantages: - it uses single mutex to protect internal structures. It is used by data- and control- path, thus there are no parallelism at all. - it uses single list to keep encap handlers for both INET and INET6 families. - struct encaptab keeps unneeded information (src, dst, masks, protosw), that isn't used by code in the source tree. - matches are prioritized and when many tunneling interfaces are registered, encapcheck handler of each interface is invoked for each packet. The search takes O(n) for n interfaces. All this work is done with exclusive lock held. What this patch includes: - the datapath is converted to be lockless using epoch(9) KPI. - struct encaptab now linked using CK_LIST. - all unused fields removed from struct encaptab. Several new fields addedr: min_length is the minimum packet length, that encapsulation handler expects to see; exact_match is maximum number of bits, that can return an encapsulation handler, when it wants to consume a packet. - IPv6 and IPv4 handlers are stored in separate lists; - added new "encap_lookup_t" method, that will be used later. It is targeted to speedup lookup of needed interface, when gif(4)/gre(4) have many interfaces. - the need to use protosw structure is eliminated. The only pr_input method was used from this structure, so I don't see the need to keep using it. - encap_input_t method changed to avoid using mbuf tags to store softc pointer. Now it is passed directly trough encap_input_t method. encap_getarg() funtions is removed. - all sockaddr structures and code that uses them removed. We don't have any code in the tree that uses them. All consumers use encap_attach_func() method, that relies on invoking of encapcheck() to determine the needed handler. - introduced struct encap_config, it contains parameters of encap handler that is going to be registered by encap_attach() function. - encap handlers are stored in lists ordered by exact_match value, thus handlers that need more bits to match will be checked first, and if encapcheck method returns exact_match value, the search will be stopped. - all current consumers changed to use new KPI. Reviewed by: mmacy Sponsored by: Yandex LLC Differential Revision: https://reviews.freebsd.org/D15617
2018-06-05 20:51:01 +00:00
gh = (struct grehdr *)mtodo(m, off);
flags = ntohs(gh->gre_flags);
if (flags & ~GRE_FLAGS_MASK)
goto drop;
opts = gh->gre_opts;
hlen = 2 * sizeof(uint16_t);
if (flags & GRE_FLAGS_CP) {
/* reserved1 field must be zero */
if (((uint16_t *)opts)[1] != 0)
goto drop;
Rework IP encapsulation handling code. Currently it has several disadvantages: - it uses single mutex to protect internal structures. It is used by data- and control- path, thus there are no parallelism at all. - it uses single list to keep encap handlers for both INET and INET6 families. - struct encaptab keeps unneeded information (src, dst, masks, protosw), that isn't used by code in the source tree. - matches are prioritized and when many tunneling interfaces are registered, encapcheck handler of each interface is invoked for each packet. The search takes O(n) for n interfaces. All this work is done with exclusive lock held. What this patch includes: - the datapath is converted to be lockless using epoch(9) KPI. - struct encaptab now linked using CK_LIST. - all unused fields removed from struct encaptab. Several new fields addedr: min_length is the minimum packet length, that encapsulation handler expects to see; exact_match is maximum number of bits, that can return an encapsulation handler, when it wants to consume a packet. - IPv6 and IPv4 handlers are stored in separate lists; - added new "encap_lookup_t" method, that will be used later. It is targeted to speedup lookup of needed interface, when gif(4)/gre(4) have many interfaces. - the need to use protosw structure is eliminated. The only pr_input method was used from this structure, so I don't see the need to keep using it. - encap_input_t method changed to avoid using mbuf tags to store softc pointer. Now it is passed directly trough encap_input_t method. encap_getarg() funtions is removed. - all sockaddr structures and code that uses them removed. We don't have any code in the tree that uses them. All consumers use encap_attach_func() method, that relies on invoking of encapcheck() to determine the needed handler. - introduced struct encap_config, it contains parameters of encap handler that is going to be registered by encap_attach() function. - encap handlers are stored in lists ordered by exact_match value, thus handlers that need more bits to match will be checked first, and if encapcheck method returns exact_match value, the search will be stopped. - all current consumers changed to use new KPI. Reviewed by: mmacy Sponsored by: Yandex LLC Differential Revision: https://reviews.freebsd.org/D15617
2018-06-05 20:51:01 +00:00
if (in_cksum_skip(m, m->m_pkthdr.len, off) != 0)
goto drop;
hlen += 2 * sizeof(uint16_t);
opts++;
}
if (flags & GRE_FLAGS_KP) {
#ifdef notyet
/*
* XXX: The current implementation uses the key only for outgoing
* packets. But we can check the key value here, or even in the
* encapcheck function.
*/
key = ntohl(*opts);
#endif
hlen += sizeof(uint32_t);
opts++;
}
#ifdef notyet
} else
key = 0;
if (sc->gre_key != 0 && (key != sc->gre_key || key != 0))
goto drop;
#endif
if (flags & GRE_FLAGS_SP) {
#ifdef notyet
seq = ntohl(*opts);
#endif
hlen += sizeof(uint32_t);
}
switch (ntohs(gh->gre_proto)) {
case ETHERTYPE_WCCP:
/*
* For WCCP skip an additional 4 bytes if after GRE header
* doesn't follow an IP header.
*/
if (flags == 0 && (*(uint8_t *)gh->gre_opts & 0xF0) != 0x40)
hlen += sizeof(uint32_t);
/* FALLTHROUGH */
case ETHERTYPE_IP:
isr = NETISR_IP;
af = AF_INET;
break;
case ETHERTYPE_IPV6:
isr = NETISR_IPV6;
af = AF_INET6;
break;
default:
goto drop;
}
Rework IP encapsulation handling code. Currently it has several disadvantages: - it uses single mutex to protect internal structures. It is used by data- and control- path, thus there are no parallelism at all. - it uses single list to keep encap handlers for both INET and INET6 families. - struct encaptab keeps unneeded information (src, dst, masks, protosw), that isn't used by code in the source tree. - matches are prioritized and when many tunneling interfaces are registered, encapcheck handler of each interface is invoked for each packet. The search takes O(n) for n interfaces. All this work is done with exclusive lock held. What this patch includes: - the datapath is converted to be lockless using epoch(9) KPI. - struct encaptab now linked using CK_LIST. - all unused fields removed from struct encaptab. Several new fields addedr: min_length is the minimum packet length, that encapsulation handler expects to see; exact_match is maximum number of bits, that can return an encapsulation handler, when it wants to consume a packet. - IPv6 and IPv4 handlers are stored in separate lists; - added new "encap_lookup_t" method, that will be used later. It is targeted to speedup lookup of needed interface, when gif(4)/gre(4) have many interfaces. - the need to use protosw structure is eliminated. The only pr_input method was used from this structure, so I don't see the need to keep using it. - encap_input_t method changed to avoid using mbuf tags to store softc pointer. Now it is passed directly trough encap_input_t method. encap_getarg() funtions is removed. - all sockaddr structures and code that uses them removed. We don't have any code in the tree that uses them. All consumers use encap_attach_func() method, that relies on invoking of encapcheck() to determine the needed handler. - introduced struct encap_config, it contains parameters of encap handler that is going to be registered by encap_attach() function. - encap handlers are stored in lists ordered by exact_match value, thus handlers that need more bits to match will be checked first, and if encapcheck method returns exact_match value, the search will be stopped. - all current consumers changed to use new KPI. Reviewed by: mmacy Sponsored by: Yandex LLC Differential Revision: https://reviews.freebsd.org/D15617
2018-06-05 20:51:01 +00:00
m_adj(m, off + hlen);
m_clrprotoflags(m);
m->m_pkthdr.rcvif = ifp;
M_SETFIB(m, ifp->if_fib);
#ifdef MAC
mac_ifnet_create_mbuf(ifp, m);
#endif
BPF_MTAP2(ifp, &af, sizeof(af), m);
if_inc_counter(ifp, IFCOUNTER_IPACKETS, 1);
if_inc_counter(ifp, IFCOUNTER_IBYTES, m->m_pkthdr.len);
if ((ifp->if_flags & IFF_MONITOR) != 0)
m_freem(m);
else
netisr_dispatch(isr, m);
return (IPPROTO_DONE);
drop:
if_inc_counter(ifp, IFCOUNTER_IERRORS, 1);
m_freem(m);
return (IPPROTO_DONE);
}
static int
gre_output(struct ifnet *ifp, struct mbuf *m, const struct sockaddr *dst,
struct route *ro)
{
uint32_t af;
if (dst->sa_family == AF_UNSPEC)
bcopy(dst->sa_data, &af, sizeof(af));
else
af = RO_GET_FAMILY(ro, dst);
/*
* Now save the af in the inbound pkt csum data, this is a cheat since
* we are using the inbound csum_data field to carry the af over to
* the gre_transmit() routine, avoiding using yet another mtag.
*/
m->m_pkthdr.csum_data = af;
return (ifp->if_transmit(ifp, m));
}
static void
gre_setseqn(struct grehdr *gh, uint32_t seq)
{
uint32_t *opts;
uint16_t flags;
opts = gh->gre_opts;
flags = ntohs(gh->gre_flags);
KASSERT((flags & GRE_FLAGS_SP) != 0,
("gre_setseqn called, but GRE_FLAGS_SP isn't set "));
if (flags & GRE_FLAGS_CP)
opts++;
if (flags & GRE_FLAGS_KP)
opts++;
*opts = htonl(seq);
}
static uint32_t
gre_flowid(struct gre_softc *sc, struct mbuf *m, uint32_t af)
{
uint32_t flowid = 0;
if ((sc->gre_options & GRE_UDPENCAP) == 0 || sc->gre_port != 0)
return (flowid);
switch (af) {
#ifdef INET
case AF_INET:
#ifdef RSS
flowid = rss_hash_ip4_2tuple(mtod(m, struct ip *)->ip_src,
mtod(m, struct ip *)->ip_dst);
break;
#endif
flowid = mtod(m, struct ip *)->ip_src.s_addr ^
mtod(m, struct ip *)->ip_dst.s_addr;
break;
#endif
#ifdef INET6
case AF_INET6:
#ifdef RSS
flowid = rss_hash_ip6_2tuple(
&mtod(m, struct ip6_hdr *)->ip6_src,
&mtod(m, struct ip6_hdr *)->ip6_dst);
break;
#endif
flowid = mtod(m, struct ip6_hdr *)->ip6_src.s6_addr32[3] ^
mtod(m, struct ip6_hdr *)->ip6_dst.s6_addr32[3];
break;
#endif
default:
break;
}
return (flowid);
}
#define MTAG_GRE 1307983903
static int
gre_transmit(struct ifnet *ifp, struct mbuf *m)
{
GRE_RLOCK_TRACKER;
struct gre_softc *sc;
struct grehdr *gh;
struct udphdr *uh;
uint32_t af, flowid;
int error, len;
uint16_t proto;
len = 0;
GRE_RLOCK();
#ifdef MAC
error = mac_ifnet_check_transmit(ifp, m);
if (error) {
m_freem(m);
goto drop;
}
#endif
error = ENETDOWN;
sc = ifp->if_softc;
if ((ifp->if_flags & IFF_MONITOR) != 0 ||
(ifp->if_flags & IFF_UP) == 0 ||
(ifp->if_drv_flags & IFF_DRV_RUNNING) == 0 ||
sc->gre_family == 0 ||
(error = if_tunnel_check_nesting(ifp, m, MTAG_GRE,
V_max_gre_nesting)) != 0) {
m_freem(m);
goto drop;
}
af = m->m_pkthdr.csum_data;
BPF_MTAP2(ifp, &af, sizeof(af), m);
m->m_flags &= ~(M_BCAST|M_MCAST);
flowid = gre_flowid(sc, m, af);
M_SETFIB(m, sc->gre_fibnum);
M_PREPEND(m, sc->gre_hlen, M_NOWAIT);
if (m == NULL) {
error = ENOBUFS;
goto drop;
}
bcopy(sc->gre_hdr, mtod(m, void *), sc->gre_hlen);
/* Determine GRE proto */
switch (af) {
#ifdef INET
case AF_INET:
proto = htons(ETHERTYPE_IP);
break;
#endif
#ifdef INET6
case AF_INET6:
proto = htons(ETHERTYPE_IPV6);
break;
#endif
default:
m_freem(m);
error = ENETDOWN;
goto drop;
}
/* Determine offset of GRE header */
switch (sc->gre_family) {
#ifdef INET
case AF_INET:
len = sizeof(struct ip);
break;
#endif
#ifdef INET6
case AF_INET6:
len = sizeof(struct ip6_hdr);
break;
#endif
default:
m_freem(m);
error = ENETDOWN;
goto drop;
}
if (sc->gre_options & GRE_UDPENCAP) {
uh = (struct udphdr *)mtodo(m, len);
uh->uh_sport |= htons(V_ipport_hifirstauto) |
(flowid >> 16) | (flowid & 0xFFFF);
uh->uh_sport = htons(ntohs(uh->uh_sport) %
V_ipport_hilastauto);
uh->uh_ulen = htons(m->m_pkthdr.len - len);
uh->uh_sum = gre_cksum_add(uh->uh_sum,
htons(m->m_pkthdr.len - len + IPPROTO_UDP));
m->m_pkthdr.csum_flags = sc->gre_csumflags;
m->m_pkthdr.csum_data = offsetof(struct udphdr, uh_sum);
len += sizeof(struct udphdr);
}
gh = (struct grehdr *)mtodo(m, len);
gh->gre_proto = proto;
if (sc->gre_options & GRE_ENABLE_SEQ)
gre_setseqn(gh, sc->gre_oseq++);
if (sc->gre_options & GRE_ENABLE_CSUM) {
*(uint16_t *)gh->gre_opts = in_cksum_skip(m,
m->m_pkthdr.len, len);
}
len = m->m_pkthdr.len - len;
switch (sc->gre_family) {
#ifdef INET
case AF_INET:
error = in_gre_output(m, af, sc->gre_hlen);
break;
#endif
#ifdef INET6
case AF_INET6:
error = in6_gre_output(m, af, sc->gre_hlen, flowid);
break;
#endif
default:
m_freem(m);
error = ENETDOWN;
}
drop:
if (error)
if_inc_counter(ifp, IFCOUNTER_OERRORS, 1);
else {
if_inc_counter(ifp, IFCOUNTER_OPACKETS, 1);
if_inc_counter(ifp, IFCOUNTER_OBYTES, len);
}
GRE_RUNLOCK();
return (error);
}
static void
gre_qflush(struct ifnet *ifp __unused)
{
}
static int
gremodevent(module_t mod, int type, void *data)
{
switch (type) {
case MOD_LOAD:
case MOD_UNLOAD:
break;
default:
return (EOPNOTSUPP);
}
return (0);
}
static moduledata_t gre_mod = {
"if_gre",
gremodevent,
0
};
DECLARE_MODULE(if_gre, gre_mod, SI_SUB_PSEUDO, SI_ORDER_ANY);
MODULE_VERSION(if_gre, 1);