1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
|
/*
* Kernel routing table updates by routing socket.
* Copyright (C) 1997, 98 Kunihiro Ishiguro
*
* This file is part of GNU Zebra.
*
* GNU Zebra is free software; you can redistribute it and/or modify it
* under the terms of the GNU General Public License as published by the
* Free Software Foundation; either version 2, or (at your option) any
* later version.
*
* GNU Zebra is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
* You should have received a copy of the GNU General Public License along
* with this program; see the file COPYING; if not, write to the Free Software
* Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include <zebra.h>
#ifdef __OpenBSD__
#include <netmpls/mpls.h>
#endif
#include "if.h"
#include "prefix.h"
#include "sockunion.h"
#include "log.h"
#include "privs.h"
#include "vxlan.h"
#include "zebra/debug.h"
#include "zebra/rib.h"
#include "zebra/rt.h"
#include "zebra/kernel_socket.h"
#include "zebra/zebra_mpls.h"
extern struct zebra_privs_t zserv_privs;
/* kernel socket export */
extern int rtm_write(int message, union sockunion *dest, union sockunion *mask,
union sockunion *gate, union sockunion *mpls,
unsigned int index, int zebra_flags, int metric);
#ifdef HAVE_STRUCT_SOCKADDR_IN_SIN_LEN
/* Adjust netmask socket length. Return value is a adjusted sin_len
value. */
static int sin_masklen(struct in_addr mask)
{
char *p, *lim;
int len;
struct sockaddr_in sin;
if (mask.s_addr == 0)
return sizeof(long);
sin.sin_addr = mask;
len = sizeof(struct sockaddr_in);
lim = (char *)&sin.sin_addr;
p = lim + sizeof(sin.sin_addr);
while (*--p == 0 && p >= lim)
len--;
return len;
}
#endif /* HAVE_STRUCT_SOCKADDR_IN_SIN_LEN */
/* Interface between zebra message and rtm message. */
static int kernel_rtm_ipv4(int cmd, struct prefix *p, struct route_entry *re)
{
struct sockaddr_in *mask = NULL;
struct sockaddr_in sin_dest, sin_mask, sin_gate;
#ifdef __OpenBSD__
struct sockaddr_mpls smpls;
#endif
union sockunion *smplsp = NULL;
struct nexthop *nexthop;
int nexthop_num = 0;
ifindex_t ifindex = 0;
int gate = 0;
int error;
char prefix_buf[PREFIX_STRLEN];
if (IS_ZEBRA_DEBUG_RIB)
prefix2str(p, prefix_buf, sizeof(prefix_buf));
memset(&sin_dest, 0, sizeof(struct sockaddr_in));
sin_dest.sin_family = AF_INET;
#ifdef HAVE_STRUCT_SOCKADDR_IN_SIN_LEN
sin_dest.sin_len = sizeof(struct sockaddr_in);
#endif /* HAVE_STRUCT_SOCKADDR_IN_SIN_LEN */
sin_dest.sin_addr = p->u.prefix4;
memset(&sin_mask, 0, sizeof(struct sockaddr_in));
memset(&sin_gate, 0, sizeof(struct sockaddr_in));
sin_gate.sin_family = AF_INET;
#ifdef HAVE_STRUCT_SOCKADDR_IN_SIN_LEN
sin_gate.sin_len = sizeof(struct sockaddr_in);
#endif /* HAVE_STRUCT_SOCKADDR_IN_SIN_LEN */
/* Make gateway. */
for (ALL_NEXTHOPS(re->nexthop, nexthop)) {
if (CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_RECURSIVE))
continue;
gate = 0;
char gate_buf[INET_ADDRSTRLEN] = "NULL";
/*
* XXX We need to refrain from kernel operations in some cases,
* but this if statement seems overly cautious - what about
* other than ADD and DELETE?
*/
if ((cmd == RTM_ADD
&& CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_ACTIVE))
|| (cmd == RTM_DELETE
&& CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB))) {
if (nexthop->type == NEXTHOP_TYPE_IPV4
|| nexthop->type == NEXTHOP_TYPE_IPV4_IFINDEX) {
sin_gate.sin_addr = nexthop->gate.ipv4;
gate = 1;
}
if (nexthop->type == NEXTHOP_TYPE_IFINDEX
|| nexthop->type == NEXTHOP_TYPE_IPV4_IFINDEX)
ifindex = nexthop->ifindex;
if (nexthop->type == NEXTHOP_TYPE_BLACKHOLE) {
struct in_addr loopback;
loopback.s_addr = htonl(INADDR_LOOPBACK);
sin_gate.sin_addr = loopback;
gate = 1;
}
if (gate && p->prefixlen == 32)
mask = NULL;
else {
masklen2ip(p->prefixlen, &sin_mask.sin_addr);
sin_mask.sin_family = AF_INET;
#ifdef HAVE_STRUCT_SOCKADDR_IN_SIN_LEN
sin_mask.sin_len =
sin_masklen(sin_mask.sin_addr);
#endif /* HAVE_STRUCT_SOCKADDR_IN_SIN_LEN */
mask = &sin_mask;
}
#ifdef __OpenBSD__
if (nexthop->nh_label) {
memset(&smpls, 0, sizeof(smpls));
smpls.smpls_len = sizeof(smpls);
smpls.smpls_family = AF_MPLS;
smpls.smpls_label =
htonl(nexthop->nh_label->label[0]
<< MPLS_LABEL_OFFSET);
smplsp = (union sockunion *)&smpls;
}
#endif
error = rtm_write(
cmd, (union sockunion *)&sin_dest,
(union sockunion *)mask,
gate ? (union sockunion *)&sin_gate : NULL,
smplsp, ifindex, re->flags, re->metric);
if (IS_ZEBRA_DEBUG_RIB) {
if (!gate) {
zlog_debug(
"%s: %s: attention! gate not found for re %p",
__func__, prefix_buf, re);
route_entry_dump(p, NULL, re);
} else
inet_ntop(AF_INET, &sin_gate.sin_addr,
gate_buf, INET_ADDRSTRLEN);
}
switch (error) {
/* We only flag nexthops as being in FIB if rtm_write()
* did its work. */
case ZEBRA_ERR_NOERROR:
nexthop_num++;
if (IS_ZEBRA_DEBUG_RIB)
zlog_debug(
"%s: %s: successfully did NH %s",
__func__, prefix_buf, gate_buf);
if (cmd == RTM_ADD)
SET_FLAG(nexthop->flags,
NEXTHOP_FLAG_FIB);
break;
/* The only valid case for this error is kernel's
* failure to install
* a multipath route, which is common for FreeBSD. This
* should be
* ignored silently, but logged as an error otherwise.
*/
case ZEBRA_ERR_RTEXIST:
if (cmd != RTM_ADD)
zlog_err(
"%s: rtm_write() returned %d for command %d",
__func__, error, cmd);
continue;
break;
/* Given that our NEXTHOP_FLAG_FIB matches real kernel
* FIB, it isn't
* normal to get any other messages in ANY case.
*/
case ZEBRA_ERR_RTNOEXIST:
case ZEBRA_ERR_RTUNREACH:
default:
zlog_err(
"%s: %s: rtm_write() unexpectedly returned %d for command %s",
__func__,
prefix2str(p, prefix_buf,
sizeof(prefix_buf)),
error,
lookup_msg(rtm_type_str, cmd, NULL));
break;
}
} /* if (cmd and flags make sense) */
else if (IS_ZEBRA_DEBUG_RIB)
zlog_debug("%s: odd command %s for flags %d", __func__,
lookup_msg(rtm_type_str, cmd, NULL),
nexthop->flags);
} /* for (ALL_NEXTHOPS(...))*/
/* If there was no useful nexthop, then complain. */
if (nexthop_num == 0 && IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("%s: No useful nexthops were found in RIB entry %p",
__func__, re);
return 0; /*XXX*/
}
#ifdef SIN6_LEN
/* Calculate sin6_len value for netmask socket value. */
static int sin6_masklen(struct in6_addr mask)
{
struct sockaddr_in6 sin6;
char *p, *lim;
int len;
if (IN6_IS_ADDR_UNSPECIFIED(&mask))
return sizeof(long);
sin6.sin6_addr = mask;
len = sizeof(struct sockaddr_in6);
lim = (char *)&sin6.sin6_addr;
p = lim + sizeof(sin6.sin6_addr);
while (*--p == 0 && p >= lim)
len--;
return len;
}
#endif /* SIN6_LEN */
/* Interface between zebra message and rtm message. */
static int kernel_rtm_ipv6(int cmd, struct prefix *p, struct route_entry *re)
{
struct sockaddr_in6 *mask;
struct sockaddr_in6 sin_dest, sin_mask, sin_gate;
struct nexthop *nexthop;
int nexthop_num = 0;
ifindex_t ifindex = 0;
int gate = 0;
int error;
memset(&sin_dest, 0, sizeof(struct sockaddr_in6));
sin_dest.sin6_family = AF_INET6;
#ifdef SIN6_LEN
sin_dest.sin6_len = sizeof(struct sockaddr_in6);
#endif /* SIN6_LEN */
sin_dest.sin6_addr = p->u.prefix6;
memset(&sin_mask, 0, sizeof(struct sockaddr_in6));
memset(&sin_gate, 0, sizeof(struct sockaddr_in6));
sin_gate.sin6_family = AF_INET6;
#ifdef HAVE_STRUCT_SOCKADDR_IN_SIN_LEN
sin_gate.sin6_len = sizeof(struct sockaddr_in6);
#endif /* HAVE_STRUCT_SOCKADDR_IN_SIN_LEN */
/* Make gateway. */
for (ALL_NEXTHOPS(re->nexthop, nexthop)) {
if (CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_RECURSIVE))
continue;
gate = 0;
if ((cmd == RTM_ADD
&& CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_ACTIVE))
|| (cmd == RTM_DELETE
#if 0
&& CHECK_FLAG (nexthop->flags, NEXTHOP_FLAG_FIB)
#endif
)) {
if (nexthop->type == NEXTHOP_TYPE_IPV6
|| nexthop->type == NEXTHOP_TYPE_IPV6_IFINDEX) {
sin_gate.sin6_addr = nexthop->gate.ipv6;
gate = 1;
}
if (nexthop->type == NEXTHOP_TYPE_IFINDEX
|| nexthop->type == NEXTHOP_TYPE_IPV6_IFINDEX)
ifindex = nexthop->ifindex;
if (cmd == RTM_ADD)
SET_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB);
}
/* Under kame set interface index to link local address. */
#ifdef KAME
#define SET_IN6_LINKLOCAL_IFINDEX(a, i) \
do { \
(a).s6_addr[2] = ((i) >> 8) & 0xff; \
(a).s6_addr[3] = (i)&0xff; \
} while (0)
if (gate && IN6_IS_ADDR_LINKLOCAL(&sin_gate.sin6_addr))
SET_IN6_LINKLOCAL_IFINDEX(sin_gate.sin6_addr, ifindex);
#endif /* KAME */
if (gate && p->prefixlen == 128)
mask = NULL;
else {
masklen2ip6(p->prefixlen, &sin_mask.sin6_addr);
sin_mask.sin6_family = AF_INET6;
#ifdef SIN6_LEN
sin_mask.sin6_len = sin6_masklen(sin_mask.sin6_addr);
#endif /* SIN6_LEN */
mask = &sin_mask;
}
error = rtm_write(cmd, (union sockunion *)&sin_dest,
(union sockunion *)mask,
gate ? (union sockunion *)&sin_gate : NULL,
NULL, ifindex, re->flags, re->metric);
#if 0
if (error)
{
zlog_info ("kernel_rtm_ipv6(): nexthop %d add error=%d.",
nexthop_num, error);
}
#else
(void)error;
#endif
nexthop_num++;
}
/* If there is no useful nexthop then return. */
if (nexthop_num == 0) {
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("kernel_rtm_ipv6(): No useful nexthop.");
return 0;
}
return 0; /*XXX*/
}
static int kernel_rtm(int cmd, struct prefix *p, struct route_entry *re)
{
switch (PREFIX_FAMILY(p)) {
case AF_INET:
return kernel_rtm_ipv4(cmd, p, re);
case AF_INET6:
return kernel_rtm_ipv6(cmd, p, re);
}
return 0;
}
int kernel_route_rib(struct prefix *p, struct prefix *src_p,
struct route_entry *old, struct route_entry *new)
{
int route = 0;
if (src_p && src_p->prefixlen) {
zlog_err("route add: IPv6 sourcedest routes unsupported!");
return 1;
}
if (zserv_privs.change(ZPRIVS_RAISE))
zlog_err("Can't raise privileges");
if (old)
route |= kernel_rtm(RTM_DELETE, p, old);
if (new)
route |= kernel_rtm(RTM_ADD, p, new);
if (zserv_privs.change(ZPRIVS_LOWER))
zlog_err("Can't lower privileges");
return route;
}
int kernel_neigh_update(int add, int ifindex, uint32_t addr, char *lla,
int llalen)
{
/* TODO */
return 0;
}
extern int kernel_get_ipmr_sg_stats(struct zebra_vrf *zvrf, void *mroute)
{
return 0;
}
int kernel_add_vtep(vni_t vni, struct interface *ifp, struct in_addr *vtep_ip)
{
return 0;
}
int kernel_del_vtep(vni_t vni, struct interface *ifp, struct in_addr *vtep_ip)
{
return 0;
}
int kernel_add_mac(struct interface *ifp, vlanid_t vid, struct ethaddr *mac,
struct in_addr vtep_ip, u_char sticky)
{
return 0;
}
int kernel_del_mac(struct interface *ifp, vlanid_t vid, struct ethaddr *mac,
struct in_addr vtep_ip, int local)
{
return 0;
}
int kernel_add_neigh(struct interface *ifp, struct ipaddr *ip,
struct ethaddr *mac)
{
return 0;
}
int kernel_del_neigh(struct interface *ifp, struct ipaddr *ip)
{
return 0;
}
extern int kernel_interface_set_master(struct interface *master,
struct interface *slave)
{
return 0;
}
|