-
Notifications
You must be signed in to change notification settings - Fork 34
/
sctp_usrreq.c
6489 lines (6135 loc) · 186 KB
/
sctp_usrreq.c
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
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*-
* Copyright (c) 2001-2008, by Cisco Systems, Inc. All rights reserved.
* Copyright (c) 2008-2012, by Randall Stewart. All rights reserved.
* Copyright (c) 2008-2012, by Michael Tuexen. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* a) Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
*
* b) 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.
*
* c) Neither the name of Cisco Systems, Inc. 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 COPYRIGHT HOLDERS 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 COPYRIGHT OWNER 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 <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <netinet/sctp_os.h>
#include <sys/proc.h>
#include <netinet/sctp_pcb.h>
#include <netinet/sctp_header.h>
#include <netinet/sctp_var.h>
#ifdef INET6
#endif
#include <netinet/sctp_sysctl.h>
#include <netinet/sctp_output.h>
#include <netinet/sctp_uio.h>
#include <netinet/sctp_asconf.h>
#include <netinet/sctputil.h>
#include <netinet/sctp_indata.h>
#include <netinet/sctp_timer.h>
#include <netinet/sctp_auth.h>
#include <netinet/sctp_bsd_addr.h>
#include <netinet/udp.h>
extern struct sctp_cc_functions sctp_cc_functions[];
extern struct sctp_ss_functions sctp_ss_functions[];
void
sctp_init(void)
{
u_long sb_max_adj;
/* Initialize and modify the sysctled variables */
sctp_init_sysctls();
if ((nmbclusters / 8) > SCTP_ASOC_MAX_CHUNKS_ON_QUEUE)
SCTP_BASE_SYSCTL(sctp_max_chunks_on_queue) = (nmbclusters / 8);
/*
* Allow a user to take no more than 1/2 the number of clusters or
* the SB_MAX whichever is smaller for the send window.
*/
sb_max_adj = (u_long)((u_quad_t) (SB_MAX) * MCLBYTES / (MSIZE + MCLBYTES));
SCTP_BASE_SYSCTL(sctp_sendspace) = min(sb_max_adj,
(((uint32_t) nmbclusters / 2) * SCTP_DEFAULT_MAXSEGMENT));
/*
* Now for the recv window, should we take the same amount? or
* should I do 1/2 the SB_MAX instead in the SB_MAX min above. For
* now I will just copy.
*/
SCTP_BASE_SYSCTL(sctp_recvspace) = SCTP_BASE_SYSCTL(sctp_sendspace);
SCTP_BASE_VAR(first_time) = 0;
SCTP_BASE_VAR(sctp_pcb_initialized) = 0;
sctp_pcb_init();
#if defined(SCTP_PACKET_LOGGING)
SCTP_BASE_VAR(packet_log_writers) = 0;
SCTP_BASE_VAR(packet_log_end) = 0;
bzero(&SCTP_BASE_VAR(packet_log_buffer), SCTP_PACKET_LOG_SIZE);
#endif
}
void
sctp_finish(void)
{
sctp_pcb_finish();
}
void
sctp_pathmtu_adjustment(struct sctp_tcb *stcb, uint16_t nxtsz)
{
struct sctp_tmit_chunk *chk;
uint16_t overhead;
/* Adjust that too */
stcb->asoc.smallest_mtu = nxtsz;
/* now off to subtract IP_DF flag if needed */
overhead = IP_HDR_SIZE;
if (sctp_auth_is_required_chunk(SCTP_DATA, stcb->asoc.peer_auth_chunks)) {
overhead += sctp_get_auth_chunk_len(stcb->asoc.peer_hmac_id);
}
TAILQ_FOREACH(chk, &stcb->asoc.send_queue, sctp_next) {
if ((chk->send_size + overhead) > nxtsz) {
chk->flags |= CHUNK_FLAGS_FRAGMENT_OK;
}
}
TAILQ_FOREACH(chk, &stcb->asoc.sent_queue, sctp_next) {
if ((chk->send_size + overhead) > nxtsz) {
/*
* For this guy we also mark for immediate resend
* since we sent to big of chunk
*/
chk->flags |= CHUNK_FLAGS_FRAGMENT_OK;
if (chk->sent < SCTP_DATAGRAM_RESEND) {
sctp_flight_size_decrease(chk);
sctp_total_flight_decrease(stcb, chk);
}
if (chk->sent != SCTP_DATAGRAM_RESEND) {
sctp_ucount_incr(stcb->asoc.sent_queue_retran_cnt);
}
chk->sent = SCTP_DATAGRAM_RESEND;
chk->rec.data.doing_fast_retransmit = 0;
if (SCTP_BASE_SYSCTL(sctp_logging_level) & SCTP_FLIGHT_LOGGING_ENABLE) {
sctp_misc_ints(SCTP_FLIGHT_LOG_DOWN_PMTU,
chk->whoTo->flight_size,
chk->book_size,
(uintptr_t) chk->whoTo,
chk->rec.data.TSN_seq);
}
/* Clear any time so NO RTT is being done */
chk->do_rtt = 0;
}
}
}
#ifdef INET
static void
sctp_notify_mbuf(struct sctp_inpcb *inp,
struct sctp_tcb *stcb,
struct sctp_nets *net,
struct ip *ip,
struct sctphdr *sh)
{
struct icmp *icmph;
int totsz, tmr_stopped = 0;
uint16_t nxtsz;
/* protection */
if ((inp == NULL) || (stcb == NULL) || (net == NULL) ||
(ip == NULL) || (sh == NULL)) {
if (stcb != NULL) {
SCTP_TCB_UNLOCK(stcb);
}
return;
}
/* First job is to verify the vtag matches what I would send */
if (ntohl(sh->v_tag) != (stcb->asoc.peer_vtag)) {
SCTP_TCB_UNLOCK(stcb);
return;
}
icmph = (struct icmp *)((caddr_t)ip - (sizeof(struct icmp) -
sizeof(struct ip)));
if (icmph->icmp_type != ICMP_UNREACH) {
/* We only care about unreachable */
SCTP_TCB_UNLOCK(stcb);
return;
}
if (icmph->icmp_code != ICMP_UNREACH_NEEDFRAG) {
/* not a unreachable message due to frag. */
SCTP_TCB_UNLOCK(stcb);
return;
}
totsz = ntohs(ip->ip_len);
nxtsz = ntohs(icmph->icmp_nextmtu);
if (nxtsz == 0) {
/*
* old type router that does not tell us what the next size
* mtu is. Rats we will have to guess (in a educated fashion
* of course)
*/
nxtsz = sctp_get_prev_mtu(totsz);
}
/* Stop any PMTU timer */
if (SCTP_OS_TIMER_PENDING(&net->pmtu_timer.timer)) {
tmr_stopped = 1;
sctp_timer_stop(SCTP_TIMER_TYPE_PATHMTURAISE, inp, stcb, net,
SCTP_FROM_SCTP_USRREQ + SCTP_LOC_1);
}
/* Adjust destination size limit */
if (net->mtu > nxtsz) {
net->mtu = nxtsz;
if (net->port) {
net->mtu -= sizeof(struct udphdr);
}
}
/* now what about the ep? */
if (stcb->asoc.smallest_mtu > nxtsz) {
sctp_pathmtu_adjustment(stcb, nxtsz);
}
if (tmr_stopped)
sctp_timer_start(SCTP_TIMER_TYPE_PATHMTURAISE, inp, stcb, net);
SCTP_TCB_UNLOCK(stcb);
}
#endif
void
sctp_notify(struct sctp_inpcb *inp,
struct ip *ip,
struct sctphdr *sh,
struct sockaddr *to,
struct sctp_tcb *stcb,
struct sctp_nets *net)
{
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
struct socket *so;
#endif
struct icmp *icmph;
/* protection */
if ((inp == NULL) || (stcb == NULL) || (net == NULL) ||
(sh == NULL) || (to == NULL)) {
if (stcb)
SCTP_TCB_UNLOCK(stcb);
return;
}
/* First job is to verify the vtag matches what I would send */
if (ntohl(sh->v_tag) != (stcb->asoc.peer_vtag)) {
SCTP_TCB_UNLOCK(stcb);
return;
}
icmph = (struct icmp *)((caddr_t)ip - (sizeof(struct icmp) -
sizeof(struct ip)));
if (icmph->icmp_type != ICMP_UNREACH) {
/* We only care about unreachable */
SCTP_TCB_UNLOCK(stcb);
return;
}
if ((icmph->icmp_code == ICMP_UNREACH_NET) ||
(icmph->icmp_code == ICMP_UNREACH_HOST) ||
(icmph->icmp_code == ICMP_UNREACH_NET_UNKNOWN) ||
(icmph->icmp_code == ICMP_UNREACH_HOST_UNKNOWN) ||
(icmph->icmp_code == ICMP_UNREACH_ISOLATED) ||
(icmph->icmp_code == ICMP_UNREACH_NET_PROHIB) ||
(icmph->icmp_code == ICMP_UNREACH_HOST_PROHIB) ||
(icmph->icmp_code == ICMP_UNREACH_FILTER_PROHIB)) {
/*
* Hmm reachablity problems we must examine closely. If its
* not reachable, we may have lost a network. Or if there is
* NO protocol at the other end named SCTP. well we consider
* it a OOTB abort.
*/
if (net->dest_state & SCTP_ADDR_REACHABLE) {
/* Ok that destination is NOT reachable */
net->dest_state &= ~SCTP_ADDR_REACHABLE;
net->dest_state &= ~SCTP_ADDR_PF;
sctp_ulp_notify(SCTP_NOTIFY_INTERFACE_DOWN,
stcb, 0,
(void *)net, SCTP_SO_NOT_LOCKED);
}
SCTP_TCB_UNLOCK(stcb);
} else if ((icmph->icmp_code == ICMP_UNREACH_PROTOCOL) ||
(icmph->icmp_code == ICMP_UNREACH_PORT)) {
/*
* Here the peer is either playing tricks on us, including
* an address that belongs to someone who does not support
* SCTP OR was a userland implementation that shutdown and
* now is dead. In either case treat it like a OOTB abort
* with no TCB
*/
sctp_abort_notification(stcb, 1, 0, NULL, SCTP_SO_NOT_LOCKED);
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
so = SCTP_INP_SO(inp);
atomic_add_int(&stcb->asoc.refcnt, 1);
SCTP_TCB_UNLOCK(stcb);
SCTP_SOCKET_LOCK(so, 1);
SCTP_TCB_LOCK(stcb);
atomic_subtract_int(&stcb->asoc.refcnt, 1);
#endif
(void)sctp_free_assoc(inp, stcb, SCTP_NORMAL_PROC, SCTP_FROM_SCTP_USRREQ + SCTP_LOC_2);
#if defined(__APPLE__) || defined(SCTP_SO_LOCK_TESTING)
SCTP_SOCKET_UNLOCK(so, 1);
/* SCTP_TCB_UNLOCK(stcb); MT: I think this is not needed. */
#endif
/* no need to unlock here, since the TCB is gone */
} else {
SCTP_TCB_UNLOCK(stcb);
}
}
#ifdef INET
void
sctp_ctlinput(cmd, sa, vip)
int cmd;
struct sockaddr *sa;
void *vip;
{
struct ip *ip = vip;
struct sctphdr *sh;
uint32_t vrf_id;
/* FIX, for non-bsd is this right? */
vrf_id = SCTP_DEFAULT_VRFID;
if (sa->sa_family != AF_INET ||
((struct sockaddr_in *)sa)->sin_addr.s_addr == INADDR_ANY) {
return;
}
if (PRC_IS_REDIRECT(cmd)) {
ip = 0;
} else if ((unsigned)cmd >= PRC_NCMDS || inetctlerrmap[cmd] == 0) {
return;
}
if (ip) {
struct sctp_inpcb *inp = NULL;
struct sctp_tcb *stcb = NULL;
struct sctp_nets *net = NULL;
struct sockaddr_in to, from;
sh = (struct sctphdr *)((caddr_t)ip + (ip->ip_hl << 2));
bzero(&to, sizeof(to));
bzero(&from, sizeof(from));
from.sin_family = to.sin_family = AF_INET;
from.sin_len = to.sin_len = sizeof(to);
from.sin_port = sh->src_port;
from.sin_addr = ip->ip_src;
to.sin_port = sh->dest_port;
to.sin_addr = ip->ip_dst;
/*
* 'to' holds the dest of the packet that failed to be sent.
* 'from' holds our local endpoint address. Thus we reverse
* the to and the from in the lookup.
*/
stcb = sctp_findassociation_addr_sa((struct sockaddr *)&to,
(struct sockaddr *)&from,
&inp, &net, 1, vrf_id);
if (stcb != NULL && inp && (inp->sctp_socket != NULL)) {
if (cmd != PRC_MSGSIZE) {
sctp_notify(inp, ip, sh,
(struct sockaddr *)&to, stcb,
net);
} else {
/* handle possible ICMP size messages */
sctp_notify_mbuf(inp, stcb, net, ip, sh);
}
} else {
if ((stcb == NULL) && (inp != NULL)) {
/* reduce ref-count */
SCTP_INP_WLOCK(inp);
SCTP_INP_DECR_REF(inp);
SCTP_INP_WUNLOCK(inp);
}
if (stcb) {
SCTP_TCB_UNLOCK(stcb);
}
}
}
return;
}
#endif
static int
sctp_getcred(SYSCTL_HANDLER_ARGS)
{
struct xucred xuc;
struct sockaddr_in addrs[2];
struct sctp_inpcb *inp;
struct sctp_nets *net;
struct sctp_tcb *stcb;
int error;
uint32_t vrf_id;
/* FIX, for non-bsd is this right? */
vrf_id = SCTP_DEFAULT_VRFID;
error = priv_check(req->td, PRIV_NETINET_GETCRED);
if (error)
return (error);
error = SYSCTL_IN(req, addrs, sizeof(addrs));
if (error)
return (error);
stcb = sctp_findassociation_addr_sa(sintosa(&addrs[1]),
sintosa(&addrs[0]),
&inp, &net, 1, vrf_id);
if (stcb == NULL || inp == NULL || inp->sctp_socket == NULL) {
if ((inp != NULL) && (stcb == NULL)) {
/* reduce ref-count */
SCTP_INP_WLOCK(inp);
SCTP_INP_DECR_REF(inp);
goto cred_can_cont;
}
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTP_USRREQ, ENOENT);
error = ENOENT;
goto out;
}
SCTP_TCB_UNLOCK(stcb);
/*
* We use the write lock here, only since in the error leg we need
* it. If we used RLOCK, then we would have to
* wlock/decr/unlock/rlock. Which in theory could create a hole.
* Better to use higher wlock.
*/
SCTP_INP_WLOCK(inp);
cred_can_cont:
error = cr_canseesocket(req->td->td_ucred, inp->sctp_socket);
if (error) {
SCTP_INP_WUNLOCK(inp);
goto out;
}
cru2x(inp->sctp_socket->so_cred, &xuc);
SCTP_INP_WUNLOCK(inp);
error = SYSCTL_OUT(req, &xuc, sizeof(struct xucred));
out:
return (error);
}
SYSCTL_PROC(_net_inet_sctp, OID_AUTO, getcred, CTLTYPE_OPAQUE | CTLFLAG_RW,
0, 0, sctp_getcred, "S,ucred", "Get the ucred of a SCTP connection");
#ifdef INET
static void
sctp_abort(struct socket *so)
{
struct sctp_inpcb *inp;
uint32_t flags;
inp = (struct sctp_inpcb *)so->so_pcb;
if (inp == NULL) {
return;
}
sctp_must_try_again:
flags = inp->sctp_flags;
#ifdef SCTP_LOG_CLOSING
sctp_log_closing(inp, NULL, 17);
#endif
if (((flags & SCTP_PCB_FLAGS_SOCKET_GONE) == 0) &&
(atomic_cmpset_int(&inp->sctp_flags, flags, (flags | SCTP_PCB_FLAGS_SOCKET_GONE | SCTP_PCB_FLAGS_CLOSE_IP)))) {
#ifdef SCTP_LOG_CLOSING
sctp_log_closing(inp, NULL, 16);
#endif
sctp_inpcb_free(inp, SCTP_FREE_SHOULD_USE_ABORT,
SCTP_CALLED_AFTER_CMPSET_OFCLOSE);
SOCK_LOCK(so);
SCTP_SB_CLEAR(so->so_snd);
/*
* same for the rcv ones, they are only here for the
* accounting/select.
*/
SCTP_SB_CLEAR(so->so_rcv);
/* Now null out the reference, we are completely detached. */
so->so_pcb = NULL;
SOCK_UNLOCK(so);
} else {
flags = inp->sctp_flags;
if ((flags & SCTP_PCB_FLAGS_SOCKET_GONE) == 0) {
goto sctp_must_try_again;
}
}
return;
}
static int
sctp_attach(struct socket *so, int proto SCTP_UNUSED, struct thread *p SCTP_UNUSED)
{
struct sctp_inpcb *inp;
struct inpcb *ip_inp;
int error;
uint32_t vrf_id = SCTP_DEFAULT_VRFID;
#ifdef IPSEC
uint32_t flags;
#endif
inp = (struct sctp_inpcb *)so->so_pcb;
if (inp != 0) {
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTP_USRREQ, EINVAL);
return (EINVAL);
}
if (so->so_snd.sb_hiwat == 0 || so->so_rcv.sb_hiwat == 0) {
error = SCTP_SORESERVE(so, SCTP_BASE_SYSCTL(sctp_sendspace), SCTP_BASE_SYSCTL(sctp_recvspace));
if (error) {
return (error);
}
}
error = sctp_inpcb_alloc(so, vrf_id);
if (error) {
return (error);
}
inp = (struct sctp_inpcb *)so->so_pcb;
SCTP_INP_WLOCK(inp);
inp->sctp_flags &= ~SCTP_PCB_FLAGS_BOUND_V6; /* I'm not v6! */
ip_inp = &inp->ip_inp.inp;
ip_inp->inp_vflag |= INP_IPV4;
ip_inp->inp_ip_ttl = MODULE_GLOBAL(ip_defttl);
#ifdef IPSEC
error = ipsec_init_policy(so, &ip_inp->inp_sp);
#ifdef SCTP_LOG_CLOSING
sctp_log_closing(inp, NULL, 17);
#endif
if (error != 0) {
try_again:
flags = inp->sctp_flags;
if (((flags & SCTP_PCB_FLAGS_SOCKET_GONE) == 0) &&
(atomic_cmpset_int(&inp->sctp_flags, flags, (flags | SCTP_PCB_FLAGS_SOCKET_GONE | SCTP_PCB_FLAGS_CLOSE_IP)))) {
#ifdef SCTP_LOG_CLOSING
sctp_log_closing(inp, NULL, 15);
#endif
SCTP_INP_WUNLOCK(inp);
sctp_inpcb_free(inp, SCTP_FREE_SHOULD_USE_ABORT,
SCTP_CALLED_AFTER_CMPSET_OFCLOSE);
} else {
flags = inp->sctp_flags;
if ((flags & SCTP_PCB_FLAGS_SOCKET_GONE) == 0) {
goto try_again;
} else {
SCTP_INP_WUNLOCK(inp);
}
}
return (error);
}
#endif /* IPSEC */
SCTP_INP_WUNLOCK(inp);
return (0);
}
static int
sctp_bind(struct socket *so, struct sockaddr *addr, struct thread *p)
{
struct sctp_inpcb *inp;
inp = (struct sctp_inpcb *)so->so_pcb;
if (inp == NULL) {
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTP_USRREQ, EINVAL);
return (EINVAL);
}
if (addr != NULL) {
if ((addr->sa_family != AF_INET) ||
(addr->sa_len != sizeof(struct sockaddr_in))) {
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTP_USRREQ, EINVAL);
return (EINVAL);
}
}
return (sctp_inpcb_bind(so, addr, NULL, p));
}
#endif
void
sctp_close(struct socket *so)
{
struct sctp_inpcb *inp;
uint32_t flags;
inp = (struct sctp_inpcb *)so->so_pcb;
if (inp == NULL)
return;
/*
* Inform all the lower layer assoc that we are done.
*/
sctp_must_try_again:
flags = inp->sctp_flags;
#ifdef SCTP_LOG_CLOSING
sctp_log_closing(inp, NULL, 17);
#endif
if (((flags & SCTP_PCB_FLAGS_SOCKET_GONE) == 0) &&
(atomic_cmpset_int(&inp->sctp_flags, flags, (flags | SCTP_PCB_FLAGS_SOCKET_GONE | SCTP_PCB_FLAGS_CLOSE_IP)))) {
if (((so->so_options & SO_LINGER) && (so->so_linger == 0)) ||
(so->so_rcv.sb_cc > 0)) {
#ifdef SCTP_LOG_CLOSING
sctp_log_closing(inp, NULL, 13);
#endif
sctp_inpcb_free(inp, SCTP_FREE_SHOULD_USE_ABORT,
SCTP_CALLED_AFTER_CMPSET_OFCLOSE);
} else {
#ifdef SCTP_LOG_CLOSING
sctp_log_closing(inp, NULL, 14);
#endif
sctp_inpcb_free(inp, SCTP_FREE_SHOULD_USE_GRACEFUL_CLOSE,
SCTP_CALLED_AFTER_CMPSET_OFCLOSE);
}
/*
* The socket is now detached, no matter what the state of
* the SCTP association.
*/
SOCK_LOCK(so);
SCTP_SB_CLEAR(so->so_snd);
/*
* same for the rcv ones, they are only here for the
* accounting/select.
*/
SCTP_SB_CLEAR(so->so_rcv);
/* Now null out the reference, we are completely detached. */
so->so_pcb = NULL;
SOCK_UNLOCK(so);
} else {
flags = inp->sctp_flags;
if ((flags & SCTP_PCB_FLAGS_SOCKET_GONE) == 0) {
goto sctp_must_try_again;
}
}
return;
}
int
sctp_sendm(struct socket *so, int flags, struct mbuf *m, struct sockaddr *addr,
struct mbuf *control, struct thread *p);
int
sctp_sendm(struct socket *so, int flags, struct mbuf *m, struct sockaddr *addr,
struct mbuf *control, struct thread *p)
{
struct sctp_inpcb *inp;
int error;
inp = (struct sctp_inpcb *)so->so_pcb;
if (inp == NULL) {
if (control) {
sctp_m_freem(control);
control = NULL;
}
SCTP_LTRACE_ERR_RET_PKT(m, inp, NULL, NULL, SCTP_FROM_SCTP_USRREQ, EINVAL);
sctp_m_freem(m);
return (EINVAL);
}
/* Got to have an to address if we are NOT a connected socket */
if ((addr == NULL) &&
((inp->sctp_flags & SCTP_PCB_FLAGS_CONNECTED) ||
(inp->sctp_flags & SCTP_PCB_FLAGS_TCPTYPE))) {
goto connected_type;
} else if (addr == NULL) {
SCTP_LTRACE_ERR_RET_PKT(m, inp, NULL, NULL, SCTP_FROM_SCTP_USRREQ, EDESTADDRREQ);
error = EDESTADDRREQ;
sctp_m_freem(m);
if (control) {
sctp_m_freem(control);
control = NULL;
}
return (error);
}
#ifdef INET6
if (addr->sa_family != AF_INET) {
/* must be a v4 address! */
SCTP_LTRACE_ERR_RET_PKT(m, inp, NULL, NULL, SCTP_FROM_SCTP_USRREQ, EDESTADDRREQ);
sctp_m_freem(m);
if (control) {
sctp_m_freem(control);
control = NULL;
}
error = EDESTADDRREQ;
return (error);
}
#endif /* INET6 */
connected_type:
/* now what about control */
if (control) {
if (inp->control) {
SCTP_PRINTF("huh? control set?\n");
sctp_m_freem(inp->control);
inp->control = NULL;
}
inp->control = control;
}
/* Place the data */
if (inp->pkt) {
SCTP_BUF_NEXT(inp->pkt_last) = m;
inp->pkt_last = m;
} else {
inp->pkt_last = inp->pkt = m;
}
if (
/* FreeBSD uses a flag passed */
((flags & PRUS_MORETOCOME) == 0)
) {
/*
* note with the current version this code will only be used
* by OpenBSD-- NetBSD, FreeBSD, and MacOS have methods for
* re-defining sosend to use the sctp_sosend. One can
* optionally switch back to this code (by changing back the
* definitions) but this is not advisable. This code is used
* by FreeBSD when sending a file with sendfile() though.
*/
int ret;
ret = sctp_output(inp, inp->pkt, addr, inp->control, p, flags);
inp->pkt = NULL;
inp->control = NULL;
return (ret);
} else {
return (0);
}
}
int
sctp_disconnect(struct socket *so)
{
struct sctp_inpcb *inp;
inp = (struct sctp_inpcb *)so->so_pcb;
if (inp == NULL) {
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTP_USRREQ, ENOTCONN);
return (ENOTCONN);
}
SCTP_INP_RLOCK(inp);
if ((inp->sctp_flags & SCTP_PCB_FLAGS_TCPTYPE) ||
(inp->sctp_flags & SCTP_PCB_FLAGS_IN_TCPPOOL)) {
if (LIST_EMPTY(&inp->sctp_asoc_list)) {
/* No connection */
SCTP_INP_RUNLOCK(inp);
return (0);
} else {
struct sctp_association *asoc;
struct sctp_tcb *stcb;
stcb = LIST_FIRST(&inp->sctp_asoc_list);
if (stcb == NULL) {
SCTP_INP_RUNLOCK(inp);
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTP_USRREQ, EINVAL);
return (EINVAL);
}
SCTP_TCB_LOCK(stcb);
asoc = &stcb->asoc;
if (stcb->asoc.state & SCTP_STATE_ABOUT_TO_BE_FREED) {
/* We are about to be freed, out of here */
SCTP_TCB_UNLOCK(stcb);
SCTP_INP_RUNLOCK(inp);
return (0);
}
if (((so->so_options & SO_LINGER) &&
(so->so_linger == 0)) ||
(so->so_rcv.sb_cc > 0)) {
if (SCTP_GET_STATE(asoc) !=
SCTP_STATE_COOKIE_WAIT) {
/* Left with Data unread */
struct mbuf *err;
err = sctp_get_mbuf_for_msg(sizeof(struct sctp_paramhdr), 0, M_NOWAIT, 1, MT_DATA);
if (err) {
/*
* Fill in the user
* initiated abort
*/
struct sctp_paramhdr *ph;
ph = mtod(err, struct sctp_paramhdr *);
SCTP_BUF_LEN(err) = sizeof(struct sctp_paramhdr);
ph->param_type = htons(SCTP_CAUSE_USER_INITIATED_ABT);
ph->param_length = htons(SCTP_BUF_LEN(err));
}
sctp_send_abort_tcb(stcb, err, SCTP_SO_LOCKED);
SCTP_STAT_INCR_COUNTER32(sctps_aborted);
}
SCTP_INP_RUNLOCK(inp);
if ((SCTP_GET_STATE(&stcb->asoc) == SCTP_STATE_OPEN) ||
(SCTP_GET_STATE(&stcb->asoc) == SCTP_STATE_SHUTDOWN_RECEIVED)) {
SCTP_STAT_DECR_GAUGE32(sctps_currestab);
}
(void)sctp_free_assoc(inp, stcb, SCTP_NORMAL_PROC, SCTP_FROM_SCTP_USRREQ + SCTP_LOC_3);
/* No unlock tcb assoc is gone */
return (0);
}
if (TAILQ_EMPTY(&asoc->send_queue) &&
TAILQ_EMPTY(&asoc->sent_queue) &&
(asoc->stream_queue_cnt == 0)) {
/* there is nothing queued to send, so done */
if (asoc->locked_on_sending) {
goto abort_anyway;
}
if ((SCTP_GET_STATE(asoc) != SCTP_STATE_SHUTDOWN_SENT) &&
(SCTP_GET_STATE(asoc) != SCTP_STATE_SHUTDOWN_ACK_SENT)) {
/* only send SHUTDOWN 1st time thru */
struct sctp_nets *netp;
if (stcb->asoc.alternate) {
netp = stcb->asoc.alternate;
} else {
netp = stcb->asoc.primary_destination;
}
sctp_stop_timers_for_shutdown(stcb);
sctp_send_shutdown(stcb, netp);
sctp_chunk_output(stcb->sctp_ep, stcb, SCTP_OUTPUT_FROM_T3, SCTP_SO_LOCKED);
if ((SCTP_GET_STATE(asoc) == SCTP_STATE_OPEN) ||
(SCTP_GET_STATE(asoc) == SCTP_STATE_SHUTDOWN_RECEIVED)) {
SCTP_STAT_DECR_GAUGE32(sctps_currestab);
}
SCTP_SET_STATE(asoc, SCTP_STATE_SHUTDOWN_SENT);
SCTP_CLEAR_SUBSTATE(asoc, SCTP_STATE_SHUTDOWN_PENDING);
sctp_timer_start(SCTP_TIMER_TYPE_SHUTDOWN,
stcb->sctp_ep, stcb, netp);
sctp_timer_start(SCTP_TIMER_TYPE_SHUTDOWNGUARD,
stcb->sctp_ep, stcb, netp);
}
} else {
/*
* we still got (or just got) data to send,
* so set SHUTDOWN_PENDING
*/
/*
* XXX sockets draft says that SCTP_EOF
* should be sent with no data. currently,
* we will allow user data to be sent first
* and move to SHUTDOWN-PENDING
*/
struct sctp_nets *netp;
if (stcb->asoc.alternate) {
netp = stcb->asoc.alternate;
} else {
netp = stcb->asoc.primary_destination;
}
asoc->state |= SCTP_STATE_SHUTDOWN_PENDING;
sctp_timer_start(SCTP_TIMER_TYPE_SHUTDOWNGUARD, stcb->sctp_ep, stcb,
netp);
if (asoc->locked_on_sending) {
/* Locked to send out the data */
struct sctp_stream_queue_pending *sp;
sp = TAILQ_LAST(&asoc->locked_on_sending->outqueue, sctp_streamhead);
if (sp == NULL) {
SCTP_PRINTF("Error, sp is NULL, locked on sending is non-null strm:%d\n",
asoc->locked_on_sending->stream_no);
} else {
if ((sp->length == 0) && (sp->msg_is_complete == 0))
asoc->state |= SCTP_STATE_PARTIAL_MSG_LEFT;
}
}
if (TAILQ_EMPTY(&asoc->send_queue) &&
TAILQ_EMPTY(&asoc->sent_queue) &&
(asoc->state & SCTP_STATE_PARTIAL_MSG_LEFT)) {
struct mbuf *op_err;
abort_anyway:
op_err = sctp_get_mbuf_for_msg((sizeof(struct sctp_paramhdr) + sizeof(uint32_t)),
0, M_NOWAIT, 1, MT_DATA);
if (op_err) {
/*
* Fill in the user
* initiated abort
*/
struct sctp_paramhdr *ph;
uint32_t *ippp;
SCTP_BUF_LEN(op_err) =
(sizeof(struct sctp_paramhdr) + sizeof(uint32_t));
ph = mtod(op_err,
struct sctp_paramhdr *);
ph->param_type = htons(
SCTP_CAUSE_USER_INITIATED_ABT);
ph->param_length = htons(SCTP_BUF_LEN(op_err));
ippp = (uint32_t *) (ph + 1);
*ippp = htonl(SCTP_FROM_SCTP_USRREQ + SCTP_LOC_4);
}
stcb->sctp_ep->last_abort_code = SCTP_FROM_SCTP_USRREQ + SCTP_LOC_4;
sctp_send_abort_tcb(stcb, op_err, SCTP_SO_LOCKED);
SCTP_STAT_INCR_COUNTER32(sctps_aborted);
if ((SCTP_GET_STATE(&stcb->asoc) == SCTP_STATE_OPEN) ||
(SCTP_GET_STATE(&stcb->asoc) == SCTP_STATE_SHUTDOWN_RECEIVED)) {
SCTP_STAT_DECR_GAUGE32(sctps_currestab);
}
SCTP_INP_RUNLOCK(inp);
(void)sctp_free_assoc(inp, stcb, SCTP_NORMAL_PROC, SCTP_FROM_SCTP_USRREQ + SCTP_LOC_5);
return (0);
} else {
sctp_chunk_output(inp, stcb, SCTP_OUTPUT_FROM_CLOSING, SCTP_SO_LOCKED);
}
}
soisdisconnecting(so);
SCTP_TCB_UNLOCK(stcb);
SCTP_INP_RUNLOCK(inp);
return (0);
}
/* not reached */
} else {
/* UDP model does not support this */
SCTP_INP_RUNLOCK(inp);
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTP_USRREQ, EOPNOTSUPP);
return (EOPNOTSUPP);
}
}
int
sctp_flush(struct socket *so, int how)
{
/*
* We will just clear out the values and let subsequent close clear
* out the data, if any. Note if the user did a shutdown(SHUT_RD)
* they will not be able to read the data, the socket will block
* that from happening.
*/
struct sctp_inpcb *inp;
inp = (struct sctp_inpcb *)so->so_pcb;
if (inp == NULL) {
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTP_USRREQ, EINVAL);
return (EINVAL);
}
SCTP_INP_RLOCK(inp);
/* For the 1 to many model this does nothing */
if (inp->sctp_flags & SCTP_PCB_FLAGS_UDPTYPE) {
SCTP_INP_RUNLOCK(inp);
return (0);
}
SCTP_INP_RUNLOCK(inp);
if ((how == PRU_FLUSH_RD) || (how == PRU_FLUSH_RDWR)) {
/*
* First make sure the sb will be happy, we don't use these
* except maybe the count
*/
SCTP_INP_WLOCK(inp);
SCTP_INP_READ_LOCK(inp);
inp->sctp_flags |= SCTP_PCB_FLAGS_SOCKET_CANT_READ;
SCTP_INP_READ_UNLOCK(inp);
SCTP_INP_WUNLOCK(inp);
so->so_rcv.sb_cc = 0;
so->so_rcv.sb_mbcnt = 0;
so->so_rcv.sb_mb = NULL;
}
if ((how == PRU_FLUSH_WR) || (how == PRU_FLUSH_RDWR)) {
/*
* First make sure the sb will be happy, we don't use these
* except maybe the count
*/
so->so_snd.sb_cc = 0;
so->so_snd.sb_mbcnt = 0;
so->so_snd.sb_mb = NULL;
}
return (0);
}
int
sctp_shutdown(struct socket *so)
{
struct sctp_inpcb *inp;
inp = (struct sctp_inpcb *)so->so_pcb;
if (inp == NULL) {
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTP_USRREQ, EINVAL);
return (EINVAL);
}
SCTP_INP_RLOCK(inp);
/* For UDP model this is a invalid call */
if (!((inp->sctp_flags & SCTP_PCB_FLAGS_TCPTYPE) ||
(inp->sctp_flags & SCTP_PCB_FLAGS_IN_TCPPOOL))) {
/* Restore the flags that the soshutdown took away. */
SOCKBUF_LOCK(&so->so_rcv);
so->so_rcv.sb_state &= ~SBS_CANTRCVMORE;
SOCKBUF_UNLOCK(&so->so_rcv);
/* This proc will wakeup for read and do nothing (I hope) */
SCTP_INP_RUNLOCK(inp);
SCTP_LTRACE_ERR_RET(inp, NULL, NULL, SCTP_FROM_SCTP_USRREQ, EOPNOTSUPP);
return (EOPNOTSUPP);
}
/*
* Ok if we reach here its the TCP model and it is either a SHUT_WR
* or SHUT_RDWR. This means we put the shutdown flag against it.
*/
{
struct sctp_tcb *stcb;
struct sctp_association *asoc;
if ((so->so_state &
(SS_ISCONNECTED | SS_ISCONNECTING | SS_ISDISCONNECTING)) == 0) {
SCTP_INP_RUNLOCK(inp);
return (ENOTCONN);
}
socantsendmore(so);
stcb = LIST_FIRST(&inp->sctp_asoc_list);
if (stcb == NULL) {
/*
* Ok we hit the case that the shutdown call was
* made after an abort or something. Nothing to do
* now.
*/