/
uipc_socket.c
3573 lines (3246 loc) · 91 KB
/
uipc_socket.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) 1982, 1986, 1988, 1990, 1993
* The Regents of the University of California.
* Copyright (c) 2004 The FreeBSD Foundation
* Copyright (c) 2004-2008 Robert N. M. Watson
* 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.
* 4. Neither the name of the University 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 REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* @(#)uipc_socket.c 8.3 (Berkeley) 4/15/94
*/
/*
* Comments on the socket life cycle:
*
* soalloc() sets of socket layer state for a socket, called only by
* socreate() and sonewconn(). Socket layer private.
*
* sodealloc() tears down socket layer state for a socket, called only by
* sofree() and sonewconn(). Socket layer private.
*
* pru_attach() associates protocol layer state with an allocated socket;
* called only once, may fail, aborting socket allocation. This is called
* from socreate() and sonewconn(). Socket layer private.
*
* pru_detach() disassociates protocol layer state from an attached socket,
* and will be called exactly once for sockets in which pru_attach() has
* been successfully called. If pru_attach() returned an error,
* pru_detach() will not be called. Socket layer private.
*
* pru_abort() and pru_close() notify the protocol layer that the last
* consumer of a socket is starting to tear down the socket, and that the
* protocol should terminate the connection. Historically, pru_abort() also
* detached protocol state from the socket state, but this is no longer the
* case.
*
* socreate() creates a socket and attaches protocol state. This is a public
* interface that may be used by socket layer consumers to create new
* sockets.
*
* sonewconn() creates a socket and attaches protocol state. This is a
* public interface that may be used by protocols to create new sockets when
* a new connection is received and will be available for accept() on a
* listen socket.
*
* soclose() destroys a socket after possibly waiting for it to disconnect.
* This is a public interface that socket consumers should use to close and
* release a socket when done with it.
*
* soabort() destroys a socket without waiting for it to disconnect (used
* only for incoming connections that are already partially or fully
* connected). This is used internally by the socket layer when clearing
* listen socket queues (due to overflow or close on the listen socket), but
* is also a public interface protocols may use to abort connections in
* their incomplete listen queues should they no longer be required. Sockets
* placed in completed connection listen queues should not be aborted for
* reasons described in the comment above the soclose() implementation. This
* is not a general purpose close routine, and except in the specific
* circumstances described here, should not be used.
*
* sofree() will free a socket and its protocol state if all references on
* the socket have been released, and is the public interface to attempt to
* free a socket when a reference is removed. This is a socket layer private
* interface.
*
* NOTE: In addition to socreate() and soclose(), which provide a single
* socket reference to the consumer to be managed as required, there are two
* calls to explicitly manage socket references, soref(), and sorele().
* Currently, these are generally required only when transitioning a socket
* from a listen queue to a file descriptor, in order to prevent garbage
* collection of the socket at an untimely moment. For a number of reasons,
* these interfaces are not preferred, and should be avoided.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include "opt_inet.h"
#include "opt_inet6.h"
#include "opt_zero.h"
#include "opt_compat.h"
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/fcntl.h>
#include <sys/limits.h>
#include <sys/lock.h>
#include <sys/mac.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/mutex.h>
#include <sys/domain.h>
#include <sys/file.h> /* for struct knote */
#include <sys/kernel.h>
#include <sys/event.h>
#include <sys/eventhandler.h>
#include <sys/poll.h>
#include <sys/proc.h>
#include <sys/protosw.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/resourcevar.h>
#include <net/route.h>
#include <sys/signalvar.h>
#include <sys/stat.h>
#include <sys/sx.h>
#include <sys/sysctl.h>
#include <sys/uio.h>
#include <sys/jail.h>
#include <net/vnet.h>
#include <security/mac/mac_framework.h>
#include <vm/uma.h>
#ifdef COMPAT_FREEBSD32
#include <sys/mount.h>
#include <sys/sysent.h>
#include <compat/freebsd32/freebsd32.h>
#endif
static int soreceive_rcvoob(struct socket *so, struct uio *uio,
int flags);
static void filt_sordetach(struct knote *kn);
static int filt_soread(struct knote *kn, long hint);
static void filt_sowdetach(struct knote *kn);
static int filt_sowrite(struct knote *kn, long hint);
static int filt_solisten(struct knote *kn, long hint);
static struct filterops solisten_filtops =
{ 1, NULL, filt_sordetach, filt_solisten };
static struct filterops soread_filtops =
{ 1, NULL, filt_sordetach, filt_soread };
static struct filterops sowrite_filtops =
{ 1, NULL, filt_sowdetach, filt_sowrite };
uma_zone_t socket_zone;
so_gen_t so_gencnt; /* generation count for sockets */
int maxsockets;
MALLOC_DEFINE(M_SONAME, "soname", "socket name");
MALLOC_DEFINE(M_PCB, "pcb", "protocol control block");
static int somaxconn = SOMAXCONN;
static int sysctl_somaxconn(SYSCTL_HANDLER_ARGS);
/* XXX: we dont have SYSCTL_USHORT */
SYSCTL_PROC(_kern_ipc, KIPC_SOMAXCONN, somaxconn, CTLTYPE_UINT | CTLFLAG_RW,
0, sizeof(int), sysctl_somaxconn, "I", "Maximum pending socket connection "
"queue size");
static int numopensockets;
SYSCTL_INT(_kern_ipc, OID_AUTO, numopensockets, CTLFLAG_RD,
&numopensockets, 0, "Number of open sockets");
#ifdef ZERO_COPY_SOCKETS
/* These aren't static because they're used in other files. */
int so_zero_copy_send = 1;
int so_zero_copy_receive = 1;
SYSCTL_NODE(_kern_ipc, OID_AUTO, zero_copy, CTLFLAG_RD, 0,
"Zero copy controls");
SYSCTL_INT(_kern_ipc_zero_copy, OID_AUTO, receive, CTLFLAG_RW,
&so_zero_copy_receive, 0, "Enable zero copy receive");
SYSCTL_INT(_kern_ipc_zero_copy, OID_AUTO, send, CTLFLAG_RW,
&so_zero_copy_send, 0, "Enable zero copy send");
#endif /* ZERO_COPY_SOCKETS */
/*
* accept_mtx locks down per-socket fields relating to accept queues. See
* socketvar.h for an annotation of the protected fields of struct socket.
*/
struct mtx accept_mtx;
MTX_SYSINIT(accept_mtx, &accept_mtx, "accept", MTX_DEF);
/*
* so_global_mtx protects so_gencnt, numopensockets, and the per-socket
* so_gencnt field.
*/
static struct mtx so_global_mtx;
MTX_SYSINIT(so_global_mtx, &so_global_mtx, "so_glabel", MTX_DEF);
/*
* General IPC sysctl name space, used by sockets and a variety of other IPC
* types.
*/
SYSCTL_NODE(_kern, KERN_IPC, ipc, CTLFLAG_RW, 0, "IPC");
/*
* Sysctl to get and set the maximum global sockets limit. Notify protocols
* of the change so that they can update their dependent limits as required.
*/
static int
sysctl_maxsockets(SYSCTL_HANDLER_ARGS)
{
int error, newmaxsockets;
newmaxsockets = maxsockets;
error = sysctl_handle_int(oidp, &newmaxsockets, 0, req);
if (error == 0 && req->newptr) {
if (newmaxsockets > maxsockets) {
maxsockets = newmaxsockets;
if (maxsockets > ((maxfiles / 4) * 3)) {
maxfiles = (maxsockets * 5) / 4;
maxfilesperproc = (maxfiles * 9) / 10;
}
EVENTHANDLER_INVOKE(maxsockets_change);
} else
error = EINVAL;
}
return (error);
}
SYSCTL_PROC(_kern_ipc, OID_AUTO, maxsockets, CTLTYPE_INT|CTLFLAG_RW,
&maxsockets, 0, sysctl_maxsockets, "IU",
"Maximum number of sockets avaliable");
/*
* Initialise maxsockets. This SYSINIT must be run after
* tunable_mbinit().
*/
static void
init_maxsockets(void *ignored)
{
TUNABLE_INT_FETCH("kern.ipc.maxsockets", &maxsockets);
maxsockets = imax(maxsockets, imax(maxfiles, nmbclusters));
}
SYSINIT(param, SI_SUB_TUNABLES, SI_ORDER_ANY, init_maxsockets, NULL);
/*
* Socket operation routines. These routines are called by the routines in
* sys_socket.c or from a system process, and implement the semantics of
* socket operations by switching out to the protocol specific routines.
*/
/*
* Get a socket structure from our zone, and initialize it. Note that it
* would probably be better to allocate socket and PCB at the same time, but
* I'm not convinced that all the protocols can be easily modified to do
* this.
*
* soalloc() returns a socket with a ref count of 0.
*/
static struct socket *
soalloc(struct vnet *vnet)
{
struct socket *so;
so = uma_zalloc(socket_zone, M_NOWAIT | M_ZERO);
if (so == NULL)
return (NULL);
#ifdef MAC
if (mac_socket_init(so, M_NOWAIT) != 0) {
uma_zfree(socket_zone, so);
return (NULL);
}
#endif
SOCKBUF_LOCK_INIT(&so->so_snd, "so_snd");
SOCKBUF_LOCK_INIT(&so->so_rcv, "so_rcv");
sx_init(&so->so_snd.sb_sx, "so_snd_sx");
sx_init(&so->so_rcv.sb_sx, "so_rcv_sx");
TAILQ_INIT(&so->so_aiojobq);
mtx_lock(&so_global_mtx);
so->so_gencnt = ++so_gencnt;
++numopensockets;
#ifdef VIMAGE
vnet->vnet_sockcnt++;
so->so_vnet = vnet;
#endif
mtx_unlock(&so_global_mtx);
return (so);
}
/*
* Free the storage associated with a socket at the socket layer, tear down
* locks, labels, etc. All protocol state is assumed already to have been
* torn down (and possibly never set up) by the caller.
*/
static void
sodealloc(struct socket *so)
{
KASSERT(so->so_count == 0, ("sodealloc(): so_count %d", so->so_count));
KASSERT(so->so_pcb == NULL, ("sodealloc(): so_pcb != NULL"));
mtx_lock(&so_global_mtx);
so->so_gencnt = ++so_gencnt;
--numopensockets; /* Could be below, but faster here. */
#ifdef VIMAGE
so->so_vnet->vnet_sockcnt--;
#endif
mtx_unlock(&so_global_mtx);
if (so->so_rcv.sb_hiwat)
(void)chgsbsize(so->so_cred->cr_uidinfo,
&so->so_rcv.sb_hiwat, 0, RLIM_INFINITY);
if (so->so_snd.sb_hiwat)
(void)chgsbsize(so->so_cred->cr_uidinfo,
&so->so_snd.sb_hiwat, 0, RLIM_INFINITY);
#ifdef INET
/* remove acccept filter if one is present. */
if (so->so_accf != NULL)
do_setopt_accept_filter(so, NULL);
#endif
#ifdef MAC
mac_socket_destroy(so);
#endif
crfree(so->so_cred);
sx_destroy(&so->so_snd.sb_sx);
sx_destroy(&so->so_rcv.sb_sx);
SOCKBUF_LOCK_DESTROY(&so->so_snd);
SOCKBUF_LOCK_DESTROY(&so->so_rcv);
uma_zfree(socket_zone, so);
}
/*
* socreate returns a socket with a ref count of 1. The socket should be
* closed with soclose().
*/
int
socreate(int dom, struct socket **aso, int type, int proto,
struct ucred *cred, struct thread *td)
{
struct protosw *prp;
struct socket *so;
int error;
if (proto)
prp = pffindproto(dom, proto, type);
else
prp = pffindtype(dom, type);
if (prp == NULL || prp->pr_usrreqs->pru_attach == NULL ||
prp->pr_usrreqs->pru_attach == pru_attach_notsupp)
return (EPROTONOSUPPORT);
if (prison_check_af(cred, prp->pr_domain->dom_family) != 0)
return (EPROTONOSUPPORT);
if (prp->pr_type != type)
return (EPROTOTYPE);
so = soalloc(CRED_TO_VNET(cred));
if (so == NULL)
return (ENOBUFS);
TAILQ_INIT(&so->so_incomp);
TAILQ_INIT(&so->so_comp);
so->so_type = type;
so->so_cred = crhold(cred);
if ((prp->pr_domain->dom_family == PF_INET) ||
(prp->pr_domain->dom_family == PF_ROUTE))
so->so_fibnum = td->td_proc->p_fibnum;
else
so->so_fibnum = 0;
so->so_proto = prp;
#ifdef MAC
mac_socket_create(cred, so);
#endif
knlist_init_mtx(&so->so_rcv.sb_sel.si_note, SOCKBUF_MTX(&so->so_rcv));
knlist_init_mtx(&so->so_snd.sb_sel.si_note, SOCKBUF_MTX(&so->so_snd));
so->so_count = 1;
/*
* Auto-sizing of socket buffers is managed by the protocols and
* the appropriate flags must be set in the pru_attach function.
*/
CURVNET_SET(so->so_vnet);
error = (*prp->pr_usrreqs->pru_attach)(so, proto, td);
CURVNET_RESTORE();
if (error) {
KASSERT(so->so_count == 1, ("socreate: so_count %d",
so->so_count));
so->so_count = 0;
sodealloc(so);
return (error);
}
*aso = so;
return (0);
}
#ifdef REGRESSION
static int regression_sonewconn_earlytest = 1;
SYSCTL_INT(_regression, OID_AUTO, sonewconn_earlytest, CTLFLAG_RW,
®ression_sonewconn_earlytest, 0, "Perform early sonewconn limit test");
#endif
/*
* When an attempt at a new connection is noted on a socket which accepts
* connections, sonewconn is called. If the connection is possible (subject
* to space constraints, etc.) then we allocate a new structure, propoerly
* linked into the data structure of the original socket, and return this.
* Connstatus may be 0, or SO_ISCONFIRMING, or SO_ISCONNECTED.
*
* Note: the ref count on the socket is 0 on return.
*/
struct socket *
sonewconn(struct socket *head, int connstatus)
{
struct socket *so;
int over;
ACCEPT_LOCK();
over = (head->so_qlen > 3 * head->so_qlimit / 2);
ACCEPT_UNLOCK();
#ifdef REGRESSION
if (regression_sonewconn_earlytest && over)
#else
if (over)
#endif
return (NULL);
VNET_ASSERT(head->so_vnet);
so = soalloc(head->so_vnet);
if (so == NULL)
return (NULL);
if ((head->so_options & SO_ACCEPTFILTER) != 0)
connstatus = 0;
so->so_head = head;
so->so_type = head->so_type;
so->so_options = head->so_options &~ SO_ACCEPTCONN;
so->so_linger = head->so_linger;
so->so_state = head->so_state | SS_NOFDREF;
so->so_fibnum = head->so_fibnum;
so->so_proto = head->so_proto;
so->so_cred = crhold(head->so_cred);
#ifdef MAC
mac_socket_newconn(head, so);
#endif
knlist_init_mtx(&so->so_rcv.sb_sel.si_note, SOCKBUF_MTX(&so->so_rcv));
knlist_init_mtx(&so->so_snd.sb_sel.si_note, SOCKBUF_MTX(&so->so_snd));
if (soreserve(so, head->so_snd.sb_hiwat, head->so_rcv.sb_hiwat) ||
(*so->so_proto->pr_usrreqs->pru_attach)(so, 0, NULL)) {
sodealloc(so);
return (NULL);
}
so->so_rcv.sb_lowat = head->so_rcv.sb_lowat;
so->so_snd.sb_lowat = head->so_snd.sb_lowat;
so->so_rcv.sb_timeo = head->so_rcv.sb_timeo;
so->so_snd.sb_timeo = head->so_snd.sb_timeo;
so->so_rcv.sb_flags |= head->so_rcv.sb_flags & SB_AUTOSIZE;
so->so_snd.sb_flags |= head->so_snd.sb_flags & SB_AUTOSIZE;
so->so_state |= connstatus;
ACCEPT_LOCK();
if (connstatus) {
TAILQ_INSERT_TAIL(&head->so_comp, so, so_list);
so->so_qstate |= SQ_COMP;
head->so_qlen++;
} else {
/*
* Keep removing sockets from the head until there's room for
* us to insert on the tail. In pre-locking revisions, this
* was a simple if(), but as we could be racing with other
* threads and soabort() requires dropping locks, we must
* loop waiting for the condition to be true.
*/
while (head->so_incqlen > head->so_qlimit) {
struct socket *sp;
sp = TAILQ_FIRST(&head->so_incomp);
TAILQ_REMOVE(&head->so_incomp, sp, so_list);
head->so_incqlen--;
sp->so_qstate &= ~SQ_INCOMP;
sp->so_head = NULL;
ACCEPT_UNLOCK();
soabort(sp);
ACCEPT_LOCK();
}
TAILQ_INSERT_TAIL(&head->so_incomp, so, so_list);
so->so_qstate |= SQ_INCOMP;
head->so_incqlen++;
}
ACCEPT_UNLOCK();
if (connstatus) {
sorwakeup(head);
wakeup_one(&head->so_timeo);
}
return (so);
}
int
sobind(struct socket *so, struct sockaddr *nam, struct thread *td)
{
int error;
CURVNET_SET(so->so_vnet);
error = (*so->so_proto->pr_usrreqs->pru_bind)(so, nam, td);
CURVNET_RESTORE();
return error;
}
/*
* solisten() transitions a socket from a non-listening state to a listening
* state, but can also be used to update the listen queue depth on an
* existing listen socket. The protocol will call back into the sockets
* layer using solisten_proto_check() and solisten_proto() to check and set
* socket-layer listen state. Call backs are used so that the protocol can
* acquire both protocol and socket layer locks in whatever order is required
* by the protocol.
*
* Protocol implementors are advised to hold the socket lock across the
* socket-layer test and set to avoid races at the socket layer.
*/
int
solisten(struct socket *so, int backlog, struct thread *td)
{
return ((*so->so_proto->pr_usrreqs->pru_listen)(so, backlog, td));
}
int
solisten_proto_check(struct socket *so)
{
SOCK_LOCK_ASSERT(so);
if (so->so_state & (SS_ISCONNECTED | SS_ISCONNECTING |
SS_ISDISCONNECTING))
return (EINVAL);
return (0);
}
void
solisten_proto(struct socket *so, int backlog)
{
SOCK_LOCK_ASSERT(so);
if (backlog < 0 || backlog > somaxconn)
backlog = somaxconn;
so->so_qlimit = backlog;
so->so_options |= SO_ACCEPTCONN;
}
/*
* Attempt to free a socket. This should really be sotryfree().
*
* sofree() will succeed if:
*
* - There are no outstanding file descriptor references or related consumers
* (so_count == 0).
*
* - The socket has been closed by user space, if ever open (SS_NOFDREF).
*
* - The protocol does not have an outstanding strong reference on the socket
* (SS_PROTOREF).
*
* - The socket is not in a completed connection queue, so a process has been
* notified that it is present. If it is removed, the user process may
* block in accept() despite select() saying the socket was ready.
*
* Otherwise, it will quietly abort so that a future call to sofree(), when
* conditions are right, can succeed.
*/
void
sofree(struct socket *so)
{
struct protosw *pr = so->so_proto;
struct socket *head;
ACCEPT_LOCK_ASSERT();
SOCK_LOCK_ASSERT(so);
if ((so->so_state & SS_NOFDREF) == 0 || so->so_count != 0 ||
(so->so_state & SS_PROTOREF) || (so->so_qstate & SQ_COMP)) {
SOCK_UNLOCK(so);
ACCEPT_UNLOCK();
return;
}
head = so->so_head;
if (head != NULL) {
KASSERT((so->so_qstate & SQ_COMP) != 0 ||
(so->so_qstate & SQ_INCOMP) != 0,
("sofree: so_head != NULL, but neither SQ_COMP nor "
"SQ_INCOMP"));
KASSERT((so->so_qstate & SQ_COMP) == 0 ||
(so->so_qstate & SQ_INCOMP) == 0,
("sofree: so->so_qstate is SQ_COMP and also SQ_INCOMP"));
TAILQ_REMOVE(&head->so_incomp, so, so_list);
head->so_incqlen--;
so->so_qstate &= ~SQ_INCOMP;
so->so_head = NULL;
}
KASSERT((so->so_qstate & SQ_COMP) == 0 &&
(so->so_qstate & SQ_INCOMP) == 0,
("sofree: so_head == NULL, but still SQ_COMP(%d) or SQ_INCOMP(%d)",
so->so_qstate & SQ_COMP, so->so_qstate & SQ_INCOMP));
if (so->so_options & SO_ACCEPTCONN) {
KASSERT((TAILQ_EMPTY(&so->so_comp)), ("sofree: so_comp populated"));
KASSERT((TAILQ_EMPTY(&so->so_incomp)), ("sofree: so_comp populated"));
}
SOCK_UNLOCK(so);
ACCEPT_UNLOCK();
if (pr->pr_flags & PR_RIGHTS && pr->pr_domain->dom_dispose != NULL)
(*pr->pr_domain->dom_dispose)(so->so_rcv.sb_mb);
if (pr->pr_usrreqs->pru_detach != NULL)
(*pr->pr_usrreqs->pru_detach)(so);
/*
* From this point on, we assume that no other references to this
* socket exist anywhere else in the stack. Therefore, no locks need
* to be acquired or held.
*
* We used to do a lot of socket buffer and socket locking here, as
* well as invoke sorflush() and perform wakeups. The direct call to
* dom_dispose() and sbrelease_internal() are an inlining of what was
* necessary from sorflush().
*
* Notice that the socket buffer and kqueue state are torn down
* before calling pru_detach. This means that protocols shold not
* assume they can perform socket wakeups, etc, in their detach code.
*/
sbdestroy(&so->so_snd, so);
sbdestroy(&so->so_rcv, so);
knlist_destroy(&so->so_rcv.sb_sel.si_note);
knlist_destroy(&so->so_snd.sb_sel.si_note);
sodealloc(so);
}
/*
* Close a socket on last file table reference removal. Initiate disconnect
* if connected. Free socket when disconnect complete.
*
* This function will sorele() the socket. Note that soclose() may be called
* prior to the ref count reaching zero. The actual socket structure will
* not be freed until the ref count reaches zero.
*/
int
soclose(struct socket *so)
{
int error = 0;
KASSERT(!(so->so_state & SS_NOFDREF), ("soclose: SS_NOFDREF on enter"));
CURVNET_SET(so->so_vnet);
funsetown(&so->so_sigio);
if (so->so_state & SS_ISCONNECTED) {
if ((so->so_state & SS_ISDISCONNECTING) == 0) {
error = sodisconnect(so);
if (error) {
if (error == ENOTCONN)
error = 0;
goto drop;
}
}
if (so->so_options & SO_LINGER) {
if ((so->so_state & SS_ISDISCONNECTING) &&
(so->so_state & SS_NBIO))
goto drop;
while (so->so_state & SS_ISCONNECTED) {
error = tsleep(&so->so_timeo,
PSOCK | PCATCH, "soclos", so->so_linger * hz);
if (error)
break;
}
}
}
drop:
if (so->so_proto->pr_usrreqs->pru_close != NULL)
(*so->so_proto->pr_usrreqs->pru_close)(so);
if (so->so_options & SO_ACCEPTCONN) {
struct socket *sp;
ACCEPT_LOCK();
while ((sp = TAILQ_FIRST(&so->so_incomp)) != NULL) {
TAILQ_REMOVE(&so->so_incomp, sp, so_list);
so->so_incqlen--;
sp->so_qstate &= ~SQ_INCOMP;
sp->so_head = NULL;
ACCEPT_UNLOCK();
soabort(sp);
ACCEPT_LOCK();
}
while ((sp = TAILQ_FIRST(&so->so_comp)) != NULL) {
TAILQ_REMOVE(&so->so_comp, sp, so_list);
so->so_qlen--;
sp->so_qstate &= ~SQ_COMP;
sp->so_head = NULL;
ACCEPT_UNLOCK();
soabort(sp);
ACCEPT_LOCK();
}
ACCEPT_UNLOCK();
}
ACCEPT_LOCK();
SOCK_LOCK(so);
KASSERT((so->so_state & SS_NOFDREF) == 0, ("soclose: NOFDREF"));
so->so_state |= SS_NOFDREF;
sorele(so);
CURVNET_RESTORE();
return (error);
}
/*
* soabort() is used to abruptly tear down a connection, such as when a
* resource limit is reached (listen queue depth exceeded), or if a listen
* socket is closed while there are sockets waiting to be accepted.
*
* This interface is tricky, because it is called on an unreferenced socket,
* and must be called only by a thread that has actually removed the socket
* from the listen queue it was on, or races with other threads are risked.
*
* This interface will call into the protocol code, so must not be called
* with any socket locks held. Protocols do call it while holding their own
* recursible protocol mutexes, but this is something that should be subject
* to review in the future.
*/
void
soabort(struct socket *so)
{
/*
* In as much as is possible, assert that no references to this
* socket are held. This is not quite the same as asserting that the
* current thread is responsible for arranging for no references, but
* is as close as we can get for now.
*/
KASSERT(so->so_count == 0, ("soabort: so_count"));
KASSERT((so->so_state & SS_PROTOREF) == 0, ("soabort: SS_PROTOREF"));
KASSERT(so->so_state & SS_NOFDREF, ("soabort: !SS_NOFDREF"));
KASSERT((so->so_state & SQ_COMP) == 0, ("soabort: SQ_COMP"));
KASSERT((so->so_state & SQ_INCOMP) == 0, ("soabort: SQ_INCOMP"));
if (so->so_proto->pr_usrreqs->pru_abort != NULL)
(*so->so_proto->pr_usrreqs->pru_abort)(so);
ACCEPT_LOCK();
SOCK_LOCK(so);
sofree(so);
}
int
soaccept(struct socket *so, struct sockaddr **nam)
{
int error;
SOCK_LOCK(so);
KASSERT((so->so_state & SS_NOFDREF) != 0, ("soaccept: !NOFDREF"));
so->so_state &= ~SS_NOFDREF;
SOCK_UNLOCK(so);
error = (*so->so_proto->pr_usrreqs->pru_accept)(so, nam);
return (error);
}
int
soconnect(struct socket *so, struct sockaddr *nam, struct thread *td)
{
int error;
if (so->so_options & SO_ACCEPTCONN)
return (EOPNOTSUPP);
CURVNET_SET(so->so_vnet);
/*
* If protocol is connection-based, can only connect once.
* Otherwise, if connected, try to disconnect first. This allows
* user to disconnect by connecting to, e.g., a null address.
*/
if (so->so_state & (SS_ISCONNECTED|SS_ISCONNECTING) &&
((so->so_proto->pr_flags & PR_CONNREQUIRED) ||
(error = sodisconnect(so)))) {
error = EISCONN;
} else {
/*
* Prevent accumulated error from previous connection from
* biting us.
*/
so->so_error = 0;
error = (*so->so_proto->pr_usrreqs->pru_connect)(so, nam, td);
}
CURVNET_RESTORE();
return (error);
}
int
soconnect2(struct socket *so1, struct socket *so2)
{
return ((*so1->so_proto->pr_usrreqs->pru_connect2)(so1, so2));
}
int
sodisconnect(struct socket *so)
{
int error;
if ((so->so_state & SS_ISCONNECTED) == 0)
return (ENOTCONN);
if (so->so_state & SS_ISDISCONNECTING)
return (EALREADY);
error = (*so->so_proto->pr_usrreqs->pru_disconnect)(so);
return (error);
}
#ifdef ZERO_COPY_SOCKETS
struct so_zerocopy_stats{
int size_ok;
int align_ok;
int found_ifp;
};
struct so_zerocopy_stats so_zerocp_stats = {0,0,0};
#include <netinet/in.h>
#include <net/route.h>
#include <netinet/in_pcb.h>
#include <vm/vm.h>
#include <vm/vm_page.h>
#include <vm/vm_object.h>
/*
* sosend_copyin() is only used if zero copy sockets are enabled. Otherwise
* sosend_dgram() and sosend_generic() use m_uiotombuf().
*
* sosend_copyin() accepts a uio and prepares an mbuf chain holding part or
* all of the data referenced by the uio. If desired, it uses zero-copy.
* *space will be updated to reflect data copied in.
*
* NB: If atomic I/O is requested, the caller must already have checked that
* space can hold resid bytes.
*
* NB: In the event of an error, the caller may need to free the partial
* chain pointed to by *mpp. The contents of both *uio and *space may be
* modified even in the case of an error.
*/
static int
sosend_copyin(struct uio *uio, struct mbuf **retmp, int atomic, long *space,
int flags)
{
struct mbuf *m, **mp, *top;
long len, resid;
int error;
#ifdef ZERO_COPY_SOCKETS
int cow_send;
#endif
*retmp = top = NULL;
mp = ⊤
len = 0;
resid = uio->uio_resid;
error = 0;
do {
#ifdef ZERO_COPY_SOCKETS
cow_send = 0;
#endif /* ZERO_COPY_SOCKETS */
if (resid >= MINCLSIZE) {
#ifdef ZERO_COPY_SOCKETS
if (top == NULL) {
m = m_gethdr(M_WAITOK, MT_DATA);
m->m_pkthdr.len = 0;
m->m_pkthdr.rcvif = NULL;
} else
m = m_get(M_WAITOK, MT_DATA);
if (so_zero_copy_send &&
resid>=PAGE_SIZE &&
*space>=PAGE_SIZE &&
uio->uio_iov->iov_len>=PAGE_SIZE) {
so_zerocp_stats.size_ok++;
so_zerocp_stats.align_ok++;
cow_send = socow_setup(m, uio);
len = cow_send;
}
if (!cow_send) {
m_clget(m, M_WAITOK);
len = min(min(MCLBYTES, resid), *space);
}
#else /* ZERO_COPY_SOCKETS */
if (top == NULL) {
m = m_getcl(M_WAIT, MT_DATA, M_PKTHDR);
m->m_pkthdr.len = 0;
m->m_pkthdr.rcvif = NULL;
} else
m = m_getcl(M_WAIT, MT_DATA, 0);
len = min(min(MCLBYTES, resid), *space);
#endif /* ZERO_COPY_SOCKETS */
} else {
if (top == NULL) {
m = m_gethdr(M_WAIT, MT_DATA);
m->m_pkthdr.len = 0;
m->m_pkthdr.rcvif = NULL;
len = min(min(MHLEN, resid), *space);
/*
* For datagram protocols, leave room
* for protocol headers in first mbuf.
*/
if (atomic && m && len < MHLEN)
MH_ALIGN(m, len);
} else {
m = m_get(M_WAIT, MT_DATA);
len = min(min(MLEN, resid), *space);
}
}
if (m == NULL) {
error = ENOBUFS;
goto out;
}
*space -= len;
#ifdef ZERO_COPY_SOCKETS
if (cow_send)
error = 0;
else
#endif /* ZERO_COPY_SOCKETS */
error = uiomove(mtod(m, void *), (int)len, uio);
resid = uio->uio_resid;
m->m_len = len;
*mp = m;
top->m_pkthdr.len += len;
if (error)
goto out;
mp = &m->m_next;
if (resid <= 0) {
if (flags & MSG_EOR)
top->m_flags |= M_EOR;
break;
}
} while (*space > 0 && atomic);
out:
*retmp = top;
return (error);
}
#endif /*ZERO_COPY_SOCKETS*/
#define SBLOCKWAIT(f) (((f) & MSG_DONTWAIT) ? 0 : SBL_WAIT)
int
sosend_dgram(struct socket *so, struct sockaddr *addr, struct uio *uio,
struct mbuf *top, struct mbuf *control, int flags, struct thread *td)
{
long space, resid;
int clen = 0, error, dontroute;
#ifdef ZERO_COPY_SOCKETS
int atomic = sosendallatonce(so) || top;
#endif
KASSERT(so->so_type == SOCK_DGRAM, ("sodgram_send: !SOCK_DGRAM"));
KASSERT(so->so_proto->pr_flags & PR_ATOMIC,
("sodgram_send: !PR_ATOMIC"));
if (uio != NULL)
resid = uio->uio_resid;
else
resid = top->m_pkthdr.len;
/*
* In theory resid should be unsigned. However, space must be
* signed, as it might be less than 0 if we over-committed, and we
* must use a signed comparison of space and resid. On the other
* hand, a negative resid causes us to loop sending 0-length
* segments to the protocol.
*/
if (resid < 0) {
error = EINVAL;
goto out;
}
dontroute =
(flags & MSG_DONTROUTE) && (so->so_options & SO_DONTROUTE) == 0;
if (td != NULL)
td->td_ru.ru_msgsnd++;
if (control != NULL)
clen = control->m_len;
SOCKBUF_LOCK(&so->so_snd);
if (so->so_snd.sb_state & SBS_CANTSENDMORE) {
SOCKBUF_UNLOCK(&so->so_snd);
error = EPIPE;
goto out;
}
if (so->so_error) {
error = so->so_error;
so->so_error = 0;
SOCKBUF_UNLOCK(&so->so_snd);
goto out;
}
if ((so->so_state & SS_ISCONNECTED) == 0) {
/*
* `sendto' and `sendmsg' is allowed on a connection-based
* socket if it supports implied connect. Return ENOTCONN if
* not connected and no address is supplied.
*/
if ((so->so_proto->pr_flags & PR_CONNREQUIRED) &&
(so->so_proto->pr_flags & PR_IMPLOPCL) == 0) {