/
uipc_usrreq.c
2062 lines (1854 loc) · 50.5 KB
/
uipc_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
/* $NetBSD: uipc_usrreq.c,v 1.203 2022/05/28 22:08:46 andvar Exp $ */
/*-
* Copyright (c) 1998, 2000, 2004, 2008, 2009, 2020 The NetBSD Foundation, Inc.
* All rights reserved.
*
* This code is derived from software contributed to The NetBSD Foundation
* by Jason R. Thorpe of the Numerical Aerospace Simulation Facility,
* NASA Ames Research Center, and by Andrew Doran.
*
* 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.
*/
/*
* Copyright (c) 1982, 1986, 1989, 1991, 1993
* The Regents of the University of California. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the name of the 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_usrreq.c 8.9 (Berkeley) 5/14/95
*/
/*
* Copyright (c) 1997 Christopher G. Demetriou. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by the University of
* California, Berkeley and its contributors.
* 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_usrreq.c 8.9 (Berkeley) 5/14/95
*/
#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: uipc_usrreq.c,v 1.203 2022/05/28 22:08:46 andvar Exp $");
#ifdef _KERNEL_OPT
#include "opt_compat_netbsd.h"
#endif
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/proc.h>
#include <sys/filedesc.h>
#include <sys/domain.h>
#include <sys/protosw.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/unpcb.h>
#include <sys/un.h>
#include <sys/namei.h>
#include <sys/vnode.h>
#include <sys/file.h>
#include <sys/stat.h>
#include <sys/mbuf.h>
#include <sys/kauth.h>
#include <sys/kmem.h>
#include <sys/atomic.h>
#include <sys/uidinfo.h>
#include <sys/kernel.h>
#include <sys/kthread.h>
#include <sys/compat_stub.h>
#include <compat/sys/socket.h>
#include <compat/net/route_70.h>
/*
* Unix communications domain.
*
* TODO:
* RDM
* rethink name space problems
* need a proper out-of-band
*
* Notes on locking:
*
* The generic rules noted in uipc_socket2.c apply. In addition:
*
* o We have a global lock, uipc_lock.
*
* o All datagram sockets are locked by uipc_lock.
*
* o For stream socketpairs, the two endpoints are created sharing the same
* independent lock. Sockets presented to PRU_CONNECT2 must already have
* matching locks.
*
* o Stream sockets created via socket() start life with their own
* independent lock.
*
* o Stream connections to a named endpoint are slightly more complicated.
* Sockets that have called listen() have their lock pointer mutated to
* the global uipc_lock. When establishing a connection, the connecting
* socket also has its lock mutated to uipc_lock, which matches the head
* (listening socket). We create a new socket for accept() to return, and
* that also shares the head's lock. Until the connection is completely
* done on both ends, all three sockets are locked by uipc_lock. Once the
* connection is complete, the association with the head's lock is broken.
* The connecting socket and the socket returned from accept() have their
* lock pointers mutated away from uipc_lock, and back to the connecting
* socket's original, independent lock. The head continues to be locked
* by uipc_lock.
*
* o If uipc_lock is determined to be a significant source of contention,
* it could easily be hashed out. It is difficult to simply make it an
* independent lock because of visibility / garbage collection issues:
* if a socket has been associated with a lock at any point, that lock
* must remain valid until the socket is no longer visible in the system.
* The lock must not be freed or otherwise destroyed until any sockets
* that had referenced it have also been destroyed.
*/
const struct sockaddr_un sun_noname = {
.sun_len = offsetof(struct sockaddr_un, sun_path),
.sun_family = AF_LOCAL,
};
ino_t unp_ino; /* prototype for fake inode numbers */
static struct mbuf * unp_addsockcred(struct lwp *, struct mbuf *);
static void unp_discard_later(file_t *);
static void unp_discard_now(file_t *);
static void unp_disconnect1(struct unpcb *);
static bool unp_drop(struct unpcb *, int);
static int unp_internalize(struct mbuf **);
static void unp_mark(file_t *);
static void unp_scan(struct mbuf *, void (*)(file_t *), int);
static void unp_shutdown1(struct unpcb *);
static void unp_thread(void *);
static void unp_thread_kick(void);
static kmutex_t *uipc_lock;
static kcondvar_t unp_thread_cv;
static lwp_t *unp_thread_lwp;
static SLIST_HEAD(,file) unp_thread_discard;
static int unp_defer;
static struct sysctllog *usrreq_sysctllog;
static void unp_sysctl_create(void);
/* Compat interface */
struct mbuf * stub_compat_70_unp_addsockcred(lwp_t *, struct mbuf *);
struct mbuf * stub_compat_70_unp_addsockcred(struct lwp *lwp,
struct mbuf *control)
{
/* just copy our initial argument */
return control;
}
bool compat70_ocreds_valid = false;
/*
* Initialize Unix protocols.
*/
void
uipc_init(void)
{
int error;
unp_sysctl_create();
uipc_lock = mutex_obj_alloc(MUTEX_DEFAULT, IPL_NONE);
cv_init(&unp_thread_cv, "unpgc");
error = kthread_create(PRI_NONE, KTHREAD_MPSAFE, NULL, unp_thread,
NULL, &unp_thread_lwp, "unpgc");
if (error != 0)
panic("uipc_init %d", error);
}
static void
unp_connid(struct lwp *l, struct unpcb *unp, int flags)
{
unp->unp_connid.unp_pid = l->l_proc->p_pid;
unp->unp_connid.unp_euid = kauth_cred_geteuid(l->l_cred);
unp->unp_connid.unp_egid = kauth_cred_getegid(l->l_cred);
unp->unp_flags |= flags;
}
/*
* A connection succeeded: disassociate both endpoints from the head's
* lock, and make them share their own lock. There is a race here: for
* a very brief time one endpoint will be locked by a different lock
* than the other end. However, since the current thread holds the old
* lock (the listening socket's lock, the head) access can still only be
* made to one side of the connection.
*/
static void
unp_setpeerlocks(struct socket *so, struct socket *so2)
{
struct unpcb *unp;
kmutex_t *lock;
KASSERT(solocked2(so, so2));
/*
* Bail out if either end of the socket is not yet fully
* connected or accepted. We only break the lock association
* with the head when the pair of sockets stand completely
* on their own.
*/
KASSERT(so->so_head == NULL);
if (so2->so_head != NULL)
return;
/*
* Drop references to old lock. A third reference (from the
* queue head) must be held as we still hold its lock. Bonus:
* we don't need to worry about garbage collecting the lock.
*/
lock = so->so_lock;
KASSERT(lock == uipc_lock);
mutex_obj_free(lock);
mutex_obj_free(lock);
/*
* Grab stream lock from the initiator and share between the two
* endpoints. Issue memory barrier to ensure all modifications
* become globally visible before the lock change. so2 is
* assumed not to have a stream lock, because it was created
* purely for the server side to accept this connection and
* started out life using the domain-wide lock.
*/
unp = sotounpcb(so);
KASSERT(unp->unp_streamlock != NULL);
KASSERT(sotounpcb(so2)->unp_streamlock == NULL);
lock = unp->unp_streamlock;
unp->unp_streamlock = NULL;
mutex_obj_hold(lock);
/*
* Ensure lock is initialized before publishing it with
* solockreset. Pairs with atomic_load_consume in solock and
* various loops to reacquire lock after wakeup.
*/
membar_release();
/*
* possible race if lock is not held - see comment in
* uipc_usrreq(PRU_ACCEPT).
*/
KASSERT(mutex_owned(lock));
solockreset(so, lock);
solockreset(so2, lock);
}
/*
* Reset a socket's lock back to the domain-wide lock.
*/
static void
unp_resetlock(struct socket *so)
{
kmutex_t *olock, *nlock;
struct unpcb *unp;
KASSERT(solocked(so));
olock = so->so_lock;
nlock = uipc_lock;
if (olock == nlock)
return;
unp = sotounpcb(so);
KASSERT(unp->unp_streamlock == NULL);
unp->unp_streamlock = olock;
mutex_obj_hold(nlock);
mutex_enter(nlock);
solockreset(so, nlock);
mutex_exit(olock);
}
static void
unp_free(struct unpcb *unp)
{
if (unp->unp_addr)
free(unp->unp_addr, M_SONAME);
if (unp->unp_streamlock != NULL)
mutex_obj_free(unp->unp_streamlock);
kmem_free(unp, sizeof(*unp));
}
static int
unp_output(struct mbuf *m, struct mbuf *control, struct unpcb *unp)
{
struct socket *so2;
const struct sockaddr_un *sun;
/* XXX: server side closed the socket */
if (unp->unp_conn == NULL)
return ECONNREFUSED;
so2 = unp->unp_conn->unp_socket;
KASSERT(solocked(so2));
if (unp->unp_addr)
sun = unp->unp_addr;
else
sun = &sun_noname;
if (unp->unp_conn->unp_flags & UNP_WANTCRED)
control = unp_addsockcred(curlwp, control);
if (unp->unp_conn->unp_flags & UNP_OWANTCRED)
MODULE_HOOK_CALL(uipc_unp_70_hook, (curlwp, control),
stub_compat_70_unp_addsockcred(curlwp, control), control);
if (sbappendaddr(&so2->so_rcv, (const struct sockaddr *)sun, m,
control) == 0) {
unp_dispose(control);
m_freem(control);
m_freem(m);
/* Don't call soroverflow because we're returning this
* error directly to the sender. */
so2->so_rcv.sb_overflowed++;
return ENOBUFS;
} else {
sorwakeup(so2);
return 0;
}
}
static void
unp_setaddr(struct socket *so, struct sockaddr *nam, bool peeraddr)
{
const struct sockaddr_un *sun = NULL;
struct unpcb *unp;
KASSERT(solocked(so));
unp = sotounpcb(so);
if (peeraddr) {
if (unp->unp_conn && unp->unp_conn->unp_addr)
sun = unp->unp_conn->unp_addr;
} else {
if (unp->unp_addr)
sun = unp->unp_addr;
}
if (sun == NULL)
sun = &sun_noname;
memcpy(nam, sun, sun->sun_len);
}
static int
unp_rcvd(struct socket *so, int flags, struct lwp *l)
{
struct unpcb *unp = sotounpcb(so);
struct socket *so2;
u_int newhiwat;
KASSERT(solocked(so));
KASSERT(unp != NULL);
switch (so->so_type) {
case SOCK_DGRAM:
panic("uipc 1");
/*NOTREACHED*/
case SOCK_SEQPACKET: /* FALLTHROUGH */
case SOCK_STREAM:
#define rcv (&so->so_rcv)
#define snd (&so2->so_snd)
if (unp->unp_conn == 0)
break;
so2 = unp->unp_conn->unp_socket;
KASSERT(solocked2(so, so2));
/*
* Adjust backpressure on sender
* and wakeup any waiting to write.
*/
snd->sb_mbmax += unp->unp_mbcnt - rcv->sb_mbcnt;
unp->unp_mbcnt = rcv->sb_mbcnt;
newhiwat = snd->sb_hiwat + unp->unp_cc - rcv->sb_cc;
(void)chgsbsize(so2->so_uidinfo,
&snd->sb_hiwat, newhiwat, RLIM_INFINITY);
unp->unp_cc = rcv->sb_cc;
sowwakeup(so2);
#undef snd
#undef rcv
break;
default:
panic("uipc 2");
}
return 0;
}
static int
unp_recvoob(struct socket *so, struct mbuf *m, int flags)
{
KASSERT(solocked(so));
return EOPNOTSUPP;
}
static int
unp_send(struct socket *so, struct mbuf *m, struct sockaddr *nam,
struct mbuf *control, struct lwp *l)
{
struct unpcb *unp = sotounpcb(so);
int error = 0;
u_int newhiwat;
struct socket *so2;
KASSERT(solocked(so));
KASSERT(unp != NULL);
KASSERT(m != NULL);
/*
* Note: unp_internalize() rejects any control message
* other than SCM_RIGHTS, and only allows one. This
* has the side-effect of preventing a caller from
* forging SCM_CREDS.
*/
if (control) {
sounlock(so);
error = unp_internalize(&control);
solock(so);
if (error != 0) {
m_freem(control);
m_freem(m);
return error;
}
}
switch (so->so_type) {
case SOCK_DGRAM: {
KASSERT(so->so_lock == uipc_lock);
if (nam) {
if ((so->so_state & SS_ISCONNECTED) != 0)
error = EISCONN;
else {
/*
* Note: once connected, the
* socket's lock must not be
* dropped until we have sent
* the message and disconnected.
* This is necessary to prevent
* intervening control ops, like
* another connection.
*/
error = unp_connect(so, nam, l);
}
} else {
if ((so->so_state & SS_ISCONNECTED) == 0)
error = ENOTCONN;
}
if (error) {
unp_dispose(control);
m_freem(control);
m_freem(m);
return error;
}
error = unp_output(m, control, unp);
if (nam)
unp_disconnect1(unp);
break;
}
case SOCK_SEQPACKET: /* FALLTHROUGH */
case SOCK_STREAM:
#define rcv (&so2->so_rcv)
#define snd (&so->so_snd)
if (unp->unp_conn == NULL) {
error = ENOTCONN;
break;
}
so2 = unp->unp_conn->unp_socket;
KASSERT(solocked2(so, so2));
if (unp->unp_conn->unp_flags & UNP_WANTCRED) {
/*
* Credentials are passed only once on
* SOCK_STREAM and SOCK_SEQPACKET.
*/
unp->unp_conn->unp_flags &= ~UNP_WANTCRED;
control = unp_addsockcred(l, control);
}
if (unp->unp_conn->unp_flags & UNP_OWANTCRED) {
/*
* Credentials are passed only once on
* SOCK_STREAM and SOCK_SEQPACKET.
*/
unp->unp_conn->unp_flags &= ~UNP_OWANTCRED;
MODULE_HOOK_CALL(uipc_unp_70_hook, (curlwp, control),
stub_compat_70_unp_addsockcred(curlwp, control),
control);
}
/*
* Send to paired receive port, and then reduce
* send buffer hiwater marks to maintain backpressure.
* Wake up readers.
*/
if (control) {
if (sbappendcontrol(rcv, m, control) != 0)
control = NULL;
} else {
switch(so->so_type) {
case SOCK_SEQPACKET:
sbappendrecord(rcv, m);
break;
case SOCK_STREAM:
sbappend(rcv, m);
break;
default:
panic("uipc_usrreq");
break;
}
}
snd->sb_mbmax -=
rcv->sb_mbcnt - unp->unp_conn->unp_mbcnt;
unp->unp_conn->unp_mbcnt = rcv->sb_mbcnt;
newhiwat = snd->sb_hiwat -
(rcv->sb_cc - unp->unp_conn->unp_cc);
(void)chgsbsize(so->so_uidinfo,
&snd->sb_hiwat, newhiwat, RLIM_INFINITY);
unp->unp_conn->unp_cc = rcv->sb_cc;
sorwakeup(so2);
#undef snd
#undef rcv
if (control != NULL) {
unp_dispose(control);
m_freem(control);
}
break;
default:
panic("uipc 4");
}
return error;
}
static int
unp_sendoob(struct socket *so, struct mbuf *m, struct mbuf * control)
{
KASSERT(solocked(so));
m_freem(m);
m_freem(control);
return EOPNOTSUPP;
}
/*
* Unix domain socket option processing.
*/
int
uipc_ctloutput(int op, struct socket *so, struct sockopt *sopt)
{
struct unpcb *unp = sotounpcb(so);
int optval = 0, error = 0;
KASSERT(solocked(so));
if (sopt->sopt_level != SOL_LOCAL) {
error = ENOPROTOOPT;
} else switch (op) {
case PRCO_SETOPT:
switch (sopt->sopt_name) {
case LOCAL_OCREDS:
if (!compat70_ocreds_valid) {
error = ENOPROTOOPT;
break;
}
/* FALLTHROUGH */
case LOCAL_CREDS:
case LOCAL_CONNWAIT:
error = sockopt_getint(sopt, &optval);
if (error)
break;
switch (sopt->sopt_name) {
#define OPTSET(bit) \
if (optval) \
unp->unp_flags |= (bit); \
else \
unp->unp_flags &= ~(bit);
case LOCAL_CREDS:
OPTSET(UNP_WANTCRED);
break;
case LOCAL_CONNWAIT:
OPTSET(UNP_CONNWAIT);
break;
case LOCAL_OCREDS:
OPTSET(UNP_OWANTCRED);
break;
}
break;
#undef OPTSET
default:
error = ENOPROTOOPT;
break;
}
break;
case PRCO_GETOPT:
sounlock(so);
switch (sopt->sopt_name) {
case LOCAL_PEEREID:
if (unp->unp_flags & UNP_EIDSVALID) {
error = sockopt_set(sopt, &unp->unp_connid,
sizeof(unp->unp_connid));
} else {
error = EINVAL;
}
break;
case LOCAL_CREDS:
#define OPTBIT(bit) (unp->unp_flags & (bit) ? 1 : 0)
optval = OPTBIT(UNP_WANTCRED);
error = sockopt_setint(sopt, optval);
break;
case LOCAL_OCREDS:
if (compat70_ocreds_valid) {
optval = OPTBIT(UNP_OWANTCRED);
error = sockopt_setint(sopt, optval);
break;
}
#undef OPTBIT
/* FALLTHROUGH */
default:
error = ENOPROTOOPT;
break;
}
solock(so);
break;
}
return (error);
}
/*
* Both send and receive buffers are allocated PIPSIZ bytes of buffering
* for stream sockets, although the total for sender and receiver is
* actually only PIPSIZ.
* Datagram sockets really use the sendspace as the maximum datagram size,
* and don't really want to reserve the sendspace. Their recvspace should
* be large enough for at least one max-size datagram plus address.
*/
#ifndef PIPSIZ
#define PIPSIZ 8192
#endif
u_long unpst_sendspace = PIPSIZ;
u_long unpst_recvspace = PIPSIZ;
u_long unpdg_sendspace = 2*1024; /* really max datagram size */
u_long unpdg_recvspace = 16*1024;
u_int unp_rights; /* files in flight */
u_int unp_rights_ratio = 2; /* limit, fraction of maxfiles */
static int
unp_attach(struct socket *so, int proto)
{
struct unpcb *unp = sotounpcb(so);
u_long sndspc, rcvspc;
int error;
KASSERT(unp == NULL);
switch (so->so_type) {
case SOCK_SEQPACKET:
/* FALLTHROUGH */
case SOCK_STREAM:
if (so->so_lock == NULL) {
so->so_lock = mutex_obj_alloc(MUTEX_DEFAULT, IPL_NONE);
solock(so);
}
sndspc = unpst_sendspace;
rcvspc = unpst_recvspace;
break;
case SOCK_DGRAM:
if (so->so_lock == NULL) {
mutex_obj_hold(uipc_lock);
so->so_lock = uipc_lock;
solock(so);
}
sndspc = unpdg_sendspace;
rcvspc = unpdg_recvspace;
break;
default:
panic("unp_attach");
}
if (so->so_snd.sb_hiwat == 0 || so->so_rcv.sb_hiwat == 0) {
error = soreserve(so, sndspc, rcvspc);
if (error) {
return error;
}
}
unp = kmem_zalloc(sizeof(*unp), KM_SLEEP);
nanotime(&unp->unp_ctime);
unp->unp_socket = so;
so->so_pcb = unp;
KASSERT(solocked(so));
return 0;
}
static void
unp_detach(struct socket *so)
{
struct unpcb *unp;
vnode_t *vp;
unp = sotounpcb(so);
KASSERT(unp != NULL);
KASSERT(solocked(so));
retry:
if ((vp = unp->unp_vnode) != NULL) {
sounlock(so);
/* Acquire v_interlock to protect against unp_connect(). */
/* XXXAD racy */
mutex_enter(vp->v_interlock);
vp->v_socket = NULL;
mutex_exit(vp->v_interlock);
vrele(vp);
solock(so);
unp->unp_vnode = NULL;
}
if (unp->unp_conn)
unp_disconnect1(unp);
while (unp->unp_refs) {
KASSERT(solocked2(so, unp->unp_refs->unp_socket));
if (unp_drop(unp->unp_refs, ECONNRESET)) {
solock(so);
goto retry;
}
}
soisdisconnected(so);
so->so_pcb = NULL;
if (unp_rights) {
/*
* Normally the receive buffer is flushed later, in sofree,
* but if our receive buffer holds references to files that
* are now garbage, we will enqueue those file references to
* the garbage collector and kick it into action.
*/
sorflush(so);
unp_free(unp);
unp_thread_kick();
} else
unp_free(unp);
}
static int
unp_accept(struct socket *so, struct sockaddr *nam)
{
struct unpcb *unp = sotounpcb(so);
struct socket *so2;
KASSERT(solocked(so));
KASSERT(nam != NULL);
/* XXX code review required to determine if unp can ever be NULL */
if (unp == NULL)
return EINVAL;
KASSERT(so->so_lock == uipc_lock);
/*
* Mark the initiating STREAM socket as connected *ONLY*
* after it's been accepted. This prevents a client from
* overrunning a server and receiving ECONNREFUSED.
*/
if (unp->unp_conn == NULL) {
/*
* This will use the empty socket and will not
* allocate.
*/
unp_setaddr(so, nam, true);
return 0;
}
so2 = unp->unp_conn->unp_socket;
if (so2->so_state & SS_ISCONNECTING) {
KASSERT(solocked2(so, so->so_head));
KASSERT(solocked2(so2, so->so_head));
soisconnected(so2);
}
/*
* If the connection is fully established, break the
* association with uipc_lock and give the connected
* pair a separate lock to share.
* There is a race here: sotounpcb(so2)->unp_streamlock
* is not locked, so when changing so2->so_lock
* another thread can grab it while so->so_lock is still
* pointing to the (locked) uipc_lock.
* this should be harmless, except that this makes
* solocked2() and solocked() unreliable.
* Another problem is that unp_setaddr() expects the
* the socket locked. Grabbing sotounpcb(so2)->unp_streamlock
* fixes both issues.
*/
mutex_enter(sotounpcb(so2)->unp_streamlock);
unp_setpeerlocks(so2, so);
/*
* Only now return peer's address, as we may need to
* block in order to allocate memory.
*
* XXX Minor race: connection can be broken while
* lock is dropped in unp_setaddr(). We will return
* error == 0 and sun_noname as the peer address.
*/
unp_setaddr(so, nam, true);
/* so_lock now points to unp_streamlock */
mutex_exit(so2->so_lock);
return 0;
}
static int
unp_ioctl(struct socket *so, u_long cmd, void *nam, struct ifnet *ifp)
{
return EOPNOTSUPP;
}
static int
unp_stat(struct socket *so, struct stat *ub)
{
struct unpcb *unp;
struct socket *so2;
KASSERT(solocked(so));
unp = sotounpcb(so);
if (unp == NULL)
return EINVAL;
ub->st_blksize = so->so_snd.sb_hiwat;
switch (so->so_type) {
case SOCK_SEQPACKET: /* FALLTHROUGH */
case SOCK_STREAM:
if (unp->unp_conn == 0)
break;
so2 = unp->unp_conn->unp_socket;
KASSERT(solocked2(so, so2));
ub->st_blksize += so2->so_rcv.sb_cc;
break;
default:
break;
}
ub->st_dev = NODEV;
if (unp->unp_ino == 0)
unp->unp_ino = unp_ino++;
ub->st_atimespec = ub->st_mtimespec = ub->st_ctimespec = unp->unp_ctime;
ub->st_ino = unp->unp_ino;
ub->st_uid = so->so_uidinfo->ui_uid;
ub->st_gid = so->so_egid;
return (0);
}
static int
unp_peeraddr(struct socket *so, struct sockaddr *nam)
{
KASSERT(solocked(so));
KASSERT(sotounpcb(so) != NULL);
KASSERT(nam != NULL);
unp_setaddr(so, nam, true);
return 0;
}
static int
unp_sockaddr(struct socket *so, struct sockaddr *nam)
{
KASSERT(solocked(so));
KASSERT(sotounpcb(so) != NULL);
KASSERT(nam != NULL);
unp_setaddr(so, nam, false);
return 0;
}
/*
* we only need to perform this allocation until syscalls other than
* bind are adjusted to use sockaddr_big.
*/
static struct sockaddr_un *
makeun_sb(struct sockaddr *nam, size_t *addrlen)
{
struct sockaddr_un *sun;
*addrlen = nam->sa_len + 1;
sun = malloc(*addrlen, M_SONAME, M_WAITOK);
memcpy(sun, nam, nam->sa_len);
*(((char *)sun) + nam->sa_len) = '\0';
return sun;
}
static int
unp_bind(struct socket *so, struct sockaddr *nam, struct lwp *l)
{
struct sockaddr_un *sun;
struct unpcb *unp;
vnode_t *vp;
struct vattr vattr;
size_t addrlen;
int error;
struct pathbuf *pb;
struct nameidata nd;
proc_t *p;
unp = sotounpcb(so);
KASSERT(solocked(so));
KASSERT(unp != NULL);
KASSERT(nam != NULL);
if (unp->unp_vnode != NULL)
return (EINVAL);
if ((unp->unp_flags & UNP_BUSY) != 0) {
/*
* EALREADY may not be strictly accurate, but since this
* is a major application error it's hardly a big deal.
*/
return (EALREADY);
}
unp->unp_flags |= UNP_BUSY;
sounlock(so);
p = l->l_proc;
sun = makeun_sb(nam, &addrlen);
pb = pathbuf_create(sun->sun_path);
if (pb == NULL) {
error = ENOMEM;
goto bad;
}
NDINIT(&nd, CREATE, FOLLOW | LOCKPARENT | TRYEMULROOT, pb);
/* SHOULD BE ABLE TO ADOPT EXISTING AND wakeup() ALA FIFO's */
if ((error = namei(&nd)) != 0) {
pathbuf_destroy(pb);
goto bad;
}
vp = nd.ni_vp;
if (vp != NULL) {