-
Notifications
You must be signed in to change notification settings - Fork 273
/
if.c
4039 lines (3446 loc) · 91.7 KB
/
if.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: if.c,v 1.496 2021/09/30 03:51:05 yamaguchi Exp $ */
/*-
* Copyright (c) 1999, 2000, 2001, 2008 The NetBSD Foundation, Inc.
* All rights reserved.
*
* This code is derived from software contributed to The NetBSD Foundation
* by William Studenmund and Jason R. Thorpe.
*
* 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) 1995, 1996, 1997, and 1998 WIDE Project.
* 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 project 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 PROJECT 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 PROJECT 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) 1980, 1986, 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.
*
* @(#)if.c 8.5 (Berkeley) 1/9/95
*/
#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: if.c,v 1.496 2021/09/30 03:51:05 yamaguchi Exp $");
#if defined(_KERNEL_OPT)
#include "opt_inet.h"
#include "opt_ipsec.h"
#include "opt_atalk.h"
#include "opt_wlan.h"
#include "opt_net_mpsafe.h"
#include "opt_mrouting.h"
#endif
#include <sys/param.h>
#include <sys/mbuf.h>
#include <sys/systm.h>
#include <sys/callout.h>
#include <sys/proc.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/domain.h>
#include <sys/protosw.h>
#include <sys/kernel.h>
#include <sys/ioctl.h>
#include <sys/sysctl.h>
#include <sys/syslog.h>
#include <sys/kauth.h>
#include <sys/kmem.h>
#include <sys/xcall.h>
#include <sys/cpu.h>
#include <sys/intr.h>
#include <sys/module_hook.h>
#include <sys/compat_stub.h>
#include <sys/msan.h>
#include <sys/hook.h>
#include <net/if.h>
#include <net/if_dl.h>
#include <net/if_ether.h>
#include <net/if_media.h>
#include <net80211/ieee80211.h>
#include <net80211/ieee80211_ioctl.h>
#include <net/if_types.h>
#include <net/route.h>
#include <net/netisr.h>
#include <sys/module.h>
#ifdef NETATALK
#include <netatalk/at_extern.h>
#include <netatalk/at.h>
#endif
#include <net/pfil.h>
#include <netinet/in.h>
#include <netinet/in_var.h>
#include <netinet/ip_encap.h>
#include <net/bpf.h>
#ifdef INET6
#include <netinet6/in6_var.h>
#include <netinet6/nd6.h>
#endif
#include "ether.h"
#include "bridge.h"
#if NBRIDGE > 0
#include <net/if_bridgevar.h>
#endif
#include "carp.h"
#if NCARP > 0
#include <netinet/ip_carp.h>
#endif
#include <compat/sys/sockio.h>
MALLOC_DEFINE(M_IFADDR, "ifaddr", "interface address");
MALLOC_DEFINE(M_IFMADDR, "ether_multi", "link-level multicast address");
/*
* XXX reusing (ifp)->if_snd->ifq_lock rather than having another spin mutex
* for each ifnet. It doesn't matter because:
* - if IFEF_MPSAFE is enabled, if_snd isn't used and lock contentions on
* ifq_lock don't happen
* - if IFEF_MPSAFE is disabled, there is no lock contention on ifq_lock
* because if_snd, if_link_state_change and if_link_state_change_process
* are all called with KERNEL_LOCK
*/
#define IF_LINK_STATE_CHANGE_LOCK(ifp) \
mutex_enter((ifp)->if_snd.ifq_lock)
#define IF_LINK_STATE_CHANGE_UNLOCK(ifp) \
mutex_exit((ifp)->if_snd.ifq_lock)
/*
* Global list of interfaces.
*/
/* DEPRECATED. Remove it once kvm(3) users disappeared */
struct ifnet_head ifnet_list;
struct pslist_head ifnet_pslist;
static ifnet_t ** ifindex2ifnet = NULL;
static u_int if_index = 1;
static size_t if_indexlim = 0;
static uint64_t index_gen;
/* Mutex to protect the above objects. */
kmutex_t ifnet_mtx __cacheline_aligned;
static struct psref_class *ifnet_psref_class __read_mostly;
static pserialize_t ifnet_psz;
static struct workqueue *ifnet_link_state_wq __read_mostly;
static kmutex_t if_clone_mtx;
struct ifnet *lo0ifp;
int ifqmaxlen = IFQ_MAXLEN;
struct psref_class *ifa_psref_class __read_mostly;
static int if_delroute_matcher(struct rtentry *, void *);
static bool if_is_unit(const char *);
static struct if_clone *if_clone_lookup(const char *, int *);
static LIST_HEAD(, if_clone) if_cloners = LIST_HEAD_INITIALIZER(if_cloners);
static int if_cloners_count;
/* Packet filtering hook for interfaces. */
pfil_head_t * if_pfil __read_mostly;
static kauth_listener_t if_listener;
static int doifioctl(struct socket *, u_long, void *, struct lwp *);
static void if_detach_queues(struct ifnet *, struct ifqueue *);
static void sysctl_sndq_setup(struct sysctllog **, const char *,
struct ifaltq *);
static void if_slowtimo(void *);
static void if_attachdomain1(struct ifnet *);
static int ifconf(u_long, void *);
static int if_transmit(struct ifnet *, struct mbuf *);
static int if_clone_create(const char *);
static int if_clone_destroy(const char *);
static void if_link_state_change_work(struct work *, void *);
static void if_up_locked(struct ifnet *);
static void _if_down(struct ifnet *);
static void if_down_deactivated(struct ifnet *);
struct if_percpuq {
struct ifnet *ipq_ifp;
void *ipq_si;
struct percpu *ipq_ifqs; /* struct ifqueue */
};
static struct mbuf *if_percpuq_dequeue(struct if_percpuq *);
static void if_percpuq_drops(void *, void *, struct cpu_info *);
static int sysctl_percpuq_drops_handler(SYSCTLFN_PROTO);
static void sysctl_percpuq_setup(struct sysctllog **, const char *,
struct if_percpuq *);
struct if_deferred_start {
struct ifnet *ids_ifp;
void (*ids_if_start)(struct ifnet *);
void *ids_si;
};
static void if_deferred_start_softint(void *);
static void if_deferred_start_common(struct ifnet *);
static void if_deferred_start_destroy(struct ifnet *);
#if defined(INET) || defined(INET6)
static void sysctl_net_pktq_setup(struct sysctllog **, int);
#endif
/*
* Hook for if_vlan - needed by if_agr
*/
struct if_vlan_vlan_input_hook_t if_vlan_vlan_input_hook;
static void if_sysctl_setup(struct sysctllog **);
static int
if_listener_cb(kauth_cred_t cred, kauth_action_t action, void *cookie,
void *arg0, void *arg1, void *arg2, void *arg3)
{
int result;
enum kauth_network_req req;
result = KAUTH_RESULT_DEFER;
req = (enum kauth_network_req)(uintptr_t)arg1;
if (action != KAUTH_NETWORK_INTERFACE)
return result;
if ((req == KAUTH_REQ_NETWORK_INTERFACE_GET) ||
(req == KAUTH_REQ_NETWORK_INTERFACE_SET))
result = KAUTH_RESULT_ALLOW;
return result;
}
/*
* Network interface utility routines.
*
* Routines with ifa_ifwith* names take sockaddr *'s as
* parameters.
*/
void
ifinit(void)
{
#if (defined(INET) || defined(INET6))
encapinit();
#endif
if_listener = kauth_listen_scope(KAUTH_SCOPE_NETWORK,
if_listener_cb, NULL);
/* interfaces are available, inform socket code */
ifioctl = doifioctl;
}
/*
* XXX Initialization before configure().
* XXX hack to get pfil_add_hook working in autoconf.
*/
void
ifinit1(void)
{
int error __diagused;
#ifdef NET_MPSAFE
printf("NET_MPSAFE enabled\n");
#endif
mutex_init(&if_clone_mtx, MUTEX_DEFAULT, IPL_NONE);
TAILQ_INIT(&ifnet_list);
mutex_init(&ifnet_mtx, MUTEX_DEFAULT, IPL_NONE);
ifnet_psz = pserialize_create();
ifnet_psref_class = psref_class_create("ifnet", IPL_SOFTNET);
ifa_psref_class = psref_class_create("ifa", IPL_SOFTNET);
error = workqueue_create(&ifnet_link_state_wq, "iflnkst",
if_link_state_change_work, NULL, PRI_SOFTNET, IPL_NET,
WQ_MPSAFE);
KASSERT(error == 0);
PSLIST_INIT(&ifnet_pslist);
if_indexlim = 8;
if_pfil = pfil_head_create(PFIL_TYPE_IFNET, NULL);
KASSERT(if_pfil != NULL);
#if NETHER > 0 || defined(NETATALK) || defined(WLAN)
etherinit();
#endif
}
/* XXX must be after domaininit() */
void
ifinit_post(void)
{
if_sysctl_setup(NULL);
}
ifnet_t *
if_alloc(u_char type)
{
return kmem_zalloc(sizeof(ifnet_t), KM_SLEEP);
}
void
if_free(ifnet_t *ifp)
{
kmem_free(ifp, sizeof(ifnet_t));
}
void
if_initname(struct ifnet *ifp, const char *name, int unit)
{
(void)snprintf(ifp->if_xname, sizeof(ifp->if_xname),
"%s%d", name, unit);
}
/*
* Null routines used while an interface is going away. These routines
* just return an error.
*/
int
if_nulloutput(struct ifnet *ifp, struct mbuf *m,
const struct sockaddr *so, const struct rtentry *rt)
{
return ENXIO;
}
void
if_nullinput(struct ifnet *ifp, struct mbuf *m)
{
/* Nothing. */
}
void
if_nullstart(struct ifnet *ifp)
{
/* Nothing. */
}
int
if_nulltransmit(struct ifnet *ifp, struct mbuf *m)
{
m_freem(m);
return ENXIO;
}
int
if_nullioctl(struct ifnet *ifp, u_long cmd, void *data)
{
return ENXIO;
}
int
if_nullinit(struct ifnet *ifp)
{
return ENXIO;
}
void
if_nullstop(struct ifnet *ifp, int disable)
{
/* Nothing. */
}
void
if_nullslowtimo(struct ifnet *ifp)
{
/* Nothing. */
}
void
if_nulldrain(struct ifnet *ifp)
{
/* Nothing. */
}
void
if_set_sadl(struct ifnet *ifp, const void *lla, u_char addrlen, bool factory)
{
struct ifaddr *ifa;
struct sockaddr_dl *sdl;
ifp->if_addrlen = addrlen;
if_alloc_sadl(ifp);
ifa = ifp->if_dl;
sdl = satosdl(ifa->ifa_addr);
(void)sockaddr_dl_setaddr(sdl, sdl->sdl_len, lla, ifp->if_addrlen);
if (factory) {
KASSERT(ifp->if_hwdl == NULL);
ifp->if_hwdl = ifp->if_dl;
ifaref(ifp->if_hwdl);
}
/* TBD routing socket */
}
struct ifaddr *
if_dl_create(const struct ifnet *ifp, const struct sockaddr_dl **sdlp)
{
unsigned socksize, ifasize;
int addrlen, namelen;
struct sockaddr_dl *mask, *sdl;
struct ifaddr *ifa;
namelen = strlen(ifp->if_xname);
addrlen = ifp->if_addrlen;
socksize = roundup(sockaddr_dl_measure(namelen, addrlen), sizeof(long));
ifasize = sizeof(*ifa) + 2 * socksize;
ifa = malloc(ifasize, M_IFADDR, M_WAITOK | M_ZERO);
sdl = (struct sockaddr_dl *)(ifa + 1);
mask = (struct sockaddr_dl *)(socksize + (char *)sdl);
sockaddr_dl_init(sdl, socksize, ifp->if_index, ifp->if_type,
ifp->if_xname, namelen, NULL, addrlen);
mask->sdl_family = AF_LINK;
mask->sdl_len = sockaddr_dl_measure(namelen, 0);
memset(&mask->sdl_data[0], 0xff, namelen);
ifa->ifa_rtrequest = link_rtrequest;
ifa->ifa_addr = (struct sockaddr *)sdl;
ifa->ifa_netmask = (struct sockaddr *)mask;
ifa_psref_init(ifa);
*sdlp = sdl;
return ifa;
}
static void
if_sadl_setrefs(struct ifnet *ifp, struct ifaddr *ifa)
{
const struct sockaddr_dl *sdl;
ifp->if_dl = ifa;
ifaref(ifa);
sdl = satosdl(ifa->ifa_addr);
ifp->if_sadl = sdl;
}
/*
* Allocate the link level name for the specified interface. This
* is an attachment helper. It must be called after ifp->if_addrlen
* is initialized, which may not be the case when if_attach() is
* called.
*/
void
if_alloc_sadl(struct ifnet *ifp)
{
struct ifaddr *ifa;
const struct sockaddr_dl *sdl;
/*
* If the interface already has a link name, release it
* now. This is useful for interfaces that can change
* link types, and thus switch link names often.
*/
if (ifp->if_sadl != NULL)
if_free_sadl(ifp, 0);
ifa = if_dl_create(ifp, &sdl);
ifa_insert(ifp, ifa);
if_sadl_setrefs(ifp, ifa);
}
static void
if_deactivate_sadl(struct ifnet *ifp)
{
struct ifaddr *ifa;
KASSERT(ifp->if_dl != NULL);
ifa = ifp->if_dl;
ifp->if_sadl = NULL;
ifp->if_dl = NULL;
ifafree(ifa);
}
static void
if_replace_sadl(struct ifnet *ifp, struct ifaddr *ifa)
{
struct ifaddr *old;
KASSERT(ifp->if_dl != NULL);
old = ifp->if_dl;
ifaref(ifa);
/* XXX Update if_dl and if_sadl atomically */
ifp->if_dl = ifa;
ifp->if_sadl = satosdl(ifa->ifa_addr);
ifafree(old);
}
void
if_activate_sadl(struct ifnet *ifp, struct ifaddr *ifa0,
const struct sockaddr_dl *sdl)
{
int s, ss;
struct ifaddr *ifa;
int bound = curlwp_bind();
KASSERT(ifa_held(ifa0));
s = splsoftnet();
if_replace_sadl(ifp, ifa0);
ss = pserialize_read_enter();
IFADDR_READER_FOREACH(ifa, ifp) {
struct psref psref;
ifa_acquire(ifa, &psref);
pserialize_read_exit(ss);
rtinit(ifa, RTM_LLINFO_UPD, 0);
ss = pserialize_read_enter();
ifa_release(ifa, &psref);
}
pserialize_read_exit(ss);
splx(s);
curlwp_bindx(bound);
}
/*
* Free the link level name for the specified interface. This is
* a detach helper. This is called from if_detach().
*/
void
if_free_sadl(struct ifnet *ifp, int factory)
{
struct ifaddr *ifa;
int s;
if (factory && ifp->if_hwdl != NULL) {
ifa = ifp->if_hwdl;
ifp->if_hwdl = NULL;
ifafree(ifa);
}
ifa = ifp->if_dl;
if (ifa == NULL) {
KASSERT(ifp->if_sadl == NULL);
return;
}
KASSERT(ifp->if_sadl != NULL);
s = splsoftnet();
KASSERT(ifa->ifa_addr->sa_family == AF_LINK);
ifa_remove(ifp, ifa);
if_deactivate_sadl(ifp);
splx(s);
}
static void
if_getindex(ifnet_t *ifp)
{
bool hitlimit = false;
char xnamebuf[HOOKNAMSIZ];
ifp->if_index_gen = index_gen++;
snprintf(xnamebuf, sizeof(xnamebuf),
"%s-lshk", ifp->if_xname);
ifp->if_linkstate_hooks = simplehook_create(IPL_NET,
xnamebuf);
ifp->if_index = if_index;
if (ifindex2ifnet == NULL) {
if_index++;
goto skip;
}
while (if_byindex(ifp->if_index)) {
/*
* If we hit USHRT_MAX, we skip back to 0 since
* there are a number of places where the value
* of if_index or if_index itself is compared
* to or stored in an unsigned short. By
* jumping back, we won't botch those assignments
* or comparisons.
*/
if (++if_index == 0) {
if_index = 1;
} else if (if_index == USHRT_MAX) {
/*
* However, if we have to jump back to
* zero *twice* without finding an empty
* slot in ifindex2ifnet[], then there
* there are too many (>65535) interfaces.
*/
if (hitlimit) {
panic("too many interfaces");
}
hitlimit = true;
if_index = 1;
}
ifp->if_index = if_index;
}
skip:
/*
* ifindex2ifnet is indexed by if_index. Since if_index will
* grow dynamically, it should grow too.
*/
if (ifindex2ifnet == NULL || ifp->if_index >= if_indexlim) {
size_t m, n, oldlim;
void *q;
oldlim = if_indexlim;
while (ifp->if_index >= if_indexlim)
if_indexlim <<= 1;
/* grow ifindex2ifnet */
m = oldlim * sizeof(struct ifnet *);
n = if_indexlim * sizeof(struct ifnet *);
q = malloc(n, M_IFADDR, M_WAITOK | M_ZERO);
if (ifindex2ifnet != NULL) {
memcpy(q, ifindex2ifnet, m);
free(ifindex2ifnet, M_IFADDR);
}
ifindex2ifnet = (struct ifnet **)q;
}
ifindex2ifnet[ifp->if_index] = ifp;
}
/*
* Initialize an interface and assign an index for it.
*
* It must be called prior to a device specific attach routine
* (e.g., ether_ifattach and ieee80211_ifattach) or if_alloc_sadl,
* and be followed by if_register:
*
* if_initialize(ifp);
* ether_ifattach(ifp, enaddr);
* if_register(ifp);
*/
void
if_initialize(ifnet_t *ifp)
{
KASSERT(if_indexlim > 0);
TAILQ_INIT(&ifp->if_addrlist);
/*
* Link level name is allocated later by a separate call to
* if_alloc_sadl().
*/
if (ifp->if_snd.ifq_maxlen == 0)
ifp->if_snd.ifq_maxlen = ifqmaxlen;
ifp->if_broadcastaddr = 0; /* reliably crash if used uninitialized */
ifp->if_link_state = LINK_STATE_UNKNOWN;
ifp->if_link_queue = -1; /* all bits set, see link_state_change() */
ifp->if_link_scheduled = false;
ifp->if_capenable = 0;
ifp->if_csum_flags_tx = 0;
ifp->if_csum_flags_rx = 0;
#ifdef ALTQ
ifp->if_snd.altq_type = 0;
ifp->if_snd.altq_disc = NULL;
ifp->if_snd.altq_flags &= ALTQF_CANTCHANGE;
ifp->if_snd.altq_tbr = NULL;
ifp->if_snd.altq_ifp = ifp;
#endif
IFQ_LOCK_INIT(&ifp->if_snd);
ifp->if_pfil = pfil_head_create(PFIL_TYPE_IFNET, ifp);
pfil_run_ifhooks(if_pfil, PFIL_IFNET_ATTACH, ifp);
IF_AFDATA_LOCK_INIT(ifp);
PSLIST_ENTRY_INIT(ifp, if_pslist_entry);
PSLIST_INIT(&ifp->if_addr_pslist);
psref_target_init(&ifp->if_psref, ifnet_psref_class);
ifp->if_ioctl_lock = mutex_obj_alloc(MUTEX_DEFAULT, IPL_NONE);
LIST_INIT(&ifp->if_multiaddrs);
if_stats_init(ifp);
IFNET_GLOBAL_LOCK();
if_getindex(ifp);
IFNET_GLOBAL_UNLOCK();
}
/*
* Register an interface to the list of "active" interfaces.
*/
void
if_register(ifnet_t *ifp)
{
/*
* If the driver has not supplied its own if_ioctl or if_stop,
* then supply the default.
*/
if (ifp->if_ioctl == NULL)
ifp->if_ioctl = ifioctl_common;
if (ifp->if_stop == NULL)
ifp->if_stop = if_nullstop;
sysctl_sndq_setup(&ifp->if_sysctl_log, ifp->if_xname, &ifp->if_snd);
if (!STAILQ_EMPTY(&domains))
if_attachdomain1(ifp);
/* Announce the interface. */
rt_ifannouncemsg(ifp, IFAN_ARRIVAL);
if (ifp->if_slowtimo != NULL) {
ifp->if_slowtimo_ch =
kmem_zalloc(sizeof(*ifp->if_slowtimo_ch), KM_SLEEP);
callout_init(ifp->if_slowtimo_ch, 0);
callout_setfunc(ifp->if_slowtimo_ch, if_slowtimo, ifp);
if_slowtimo(ifp);
}
if (ifp->if_transmit == NULL || ifp->if_transmit == if_nulltransmit)
ifp->if_transmit = if_transmit;
IFNET_GLOBAL_LOCK();
TAILQ_INSERT_TAIL(&ifnet_list, ifp, if_list);
IFNET_WRITER_INSERT_TAIL(ifp);
IFNET_GLOBAL_UNLOCK();
}
/*
* The if_percpuq framework
*
* It allows network device drivers to execute the network stack
* in softint (so called softint-based if_input). It utilizes
* softint and percpu ifqueue. It doesn't distribute any packets
* between CPUs, unlike pktqueue(9).
*
* Currently we support two options for device drivers to apply the framework:
* - Use it implicitly with less changes
* - If you use if_attach in driver's _attach function and if_input in
* driver's Rx interrupt handler, a packet is queued and a softint handles
* the packet implicitly
* - Use it explicitly in each driver (recommended)
* - You can use if_percpuq_* directly in your driver
* - In this case, you need to allocate struct if_percpuq in driver's softc
* - See wm(4) as a reference implementation
*/
static void
if_percpuq_softint(void *arg)
{
struct if_percpuq *ipq = arg;
struct ifnet *ifp = ipq->ipq_ifp;
struct mbuf *m;
while ((m = if_percpuq_dequeue(ipq)) != NULL) {
if_statinc(ifp, if_ipackets);
bpf_mtap(ifp, m, BPF_D_IN);
ifp->_if_input(ifp, m);
}
}
static void
if_percpuq_init_ifq(void *p, void *arg __unused, struct cpu_info *ci __unused)
{
struct ifqueue *const ifq = p;
memset(ifq, 0, sizeof(*ifq));
ifq->ifq_maxlen = IFQ_MAXLEN;
}
struct if_percpuq *
if_percpuq_create(struct ifnet *ifp)
{
struct if_percpuq *ipq;
u_int flags = SOFTINT_NET;
flags |= if_is_mpsafe(ifp) ? SOFTINT_MPSAFE : 0;
ipq = kmem_zalloc(sizeof(*ipq), KM_SLEEP);
ipq->ipq_ifp = ifp;
ipq->ipq_si = softint_establish(flags, if_percpuq_softint, ipq);
ipq->ipq_ifqs = percpu_alloc(sizeof(struct ifqueue));
percpu_foreach(ipq->ipq_ifqs, &if_percpuq_init_ifq, NULL);
sysctl_percpuq_setup(&ifp->if_sysctl_log, ifp->if_xname, ipq);
return ipq;
}
static struct mbuf *
if_percpuq_dequeue(struct if_percpuq *ipq)
{
struct mbuf *m;
struct ifqueue *ifq;
int s;
s = splnet();
ifq = percpu_getref(ipq->ipq_ifqs);
IF_DEQUEUE(ifq, m);
percpu_putref(ipq->ipq_ifqs);
splx(s);
return m;
}
static void
if_percpuq_purge_ifq(void *p, void *arg __unused, struct cpu_info *ci __unused)
{
struct ifqueue *const ifq = p;
IF_PURGE(ifq);
}
void
if_percpuq_destroy(struct if_percpuq *ipq)
{
/* if_detach may already destroy it */
if (ipq == NULL)
return;
softint_disestablish(ipq->ipq_si);
percpu_foreach(ipq->ipq_ifqs, &if_percpuq_purge_ifq, NULL);
percpu_free(ipq->ipq_ifqs, sizeof(struct ifqueue));
kmem_free(ipq, sizeof(*ipq));
}
void
if_percpuq_enqueue(struct if_percpuq *ipq, struct mbuf *m)
{
struct ifqueue *ifq;
int s;
KASSERT(ipq != NULL);
s = splnet();
ifq = percpu_getref(ipq->ipq_ifqs);
if (IF_QFULL(ifq)) {
IF_DROP(ifq);
percpu_putref(ipq->ipq_ifqs);
m_freem(m);
goto out;
}
IF_ENQUEUE(ifq, m);
percpu_putref(ipq->ipq_ifqs);
softint_schedule(ipq->ipq_si);
out:
splx(s);
}
static void
if_percpuq_drops(void *p, void *arg, struct cpu_info *ci __unused)
{
struct ifqueue *const ifq = p;
int *sum = arg;
*sum += ifq->ifq_drops;
}
static int
sysctl_percpuq_drops_handler(SYSCTLFN_ARGS)
{
struct sysctlnode node;
struct if_percpuq *ipq;
int sum = 0;
int error;
node = *rnode;
ipq = node.sysctl_data;
percpu_foreach(ipq->ipq_ifqs, if_percpuq_drops, &sum);
node.sysctl_data = ∑
error = sysctl_lookup(SYSCTLFN_CALL(&node));
if (error != 0 || newp == NULL)
return error;
return 0;
}
static void
sysctl_percpuq_setup(struct sysctllog **clog, const char* ifname,
struct if_percpuq *ipq)
{
const struct sysctlnode *cnode, *rnode;
if (sysctl_createv(clog, 0, NULL, &rnode,
CTLFLAG_PERMANENT,
CTLTYPE_NODE, "interfaces",
SYSCTL_DESCR("Per-interface controls"),
NULL, 0, NULL, 0,
CTL_NET, CTL_CREATE, CTL_EOL) != 0)
goto bad;
if (sysctl_createv(clog, 0, &rnode, &rnode,
CTLFLAG_PERMANENT,
CTLTYPE_NODE, ifname,
SYSCTL_DESCR("Interface controls"),
NULL, 0, NULL, 0,
CTL_CREATE, CTL_EOL) != 0)
goto bad;
if (sysctl_createv(clog, 0, &rnode, &rnode,
CTLFLAG_PERMANENT,
CTLTYPE_NODE, "rcvq",
SYSCTL_DESCR("Interface input queue controls"),
NULL, 0, NULL, 0,
CTL_CREATE, CTL_EOL) != 0)
goto bad;
#ifdef NOTYET
/* XXX Should show each per-CPU queue length? */
if (sysctl_createv(clog, 0, &rnode, &rnode,
CTLFLAG_PERMANENT,
CTLTYPE_INT, "len",
SYSCTL_DESCR("Current input queue length"),
sysctl_percpuq_len, 0, NULL, 0,
CTL_CREATE, CTL_EOL) != 0)
goto bad;
if (sysctl_createv(clog, 0, &rnode, &cnode,
CTLFLAG_PERMANENT | CTLFLAG_READWRITE,
CTLTYPE_INT, "maxlen",
SYSCTL_DESCR("Maximum allowed input queue length"),
sysctl_percpuq_maxlen_handler, 0, (void *)ipq, 0,
CTL_CREATE, CTL_EOL) != 0)
goto bad;
#endif