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
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
|
/* Connection state tracking for netfilter. This is separated from,
but required by, the NAT layer; it can also be used by an iptables
extension. */
/* (C) 1999-2001 Paul `Rusty' Russell
* (C) 2002-2006 Netfilter Core Team <coreteam@netfilter.org>
* (C) 2003,2004 USAGI/WIDE Project <http://www.linux-ipv6.org>
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation.
*
* 23 Apr 2001: Harald Welte <laforge@gnumonks.org>
* - new API and handling of conntrack/nat helpers
* - now capable of multiple expectations for one master
* 16 Jul 2002: Harald Welte <laforge@gnumonks.org>
* - add usage/reference counts to ip_conntrack_expect
* - export ip_conntrack[_expect]_{find_get,put} functions
* 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp>
* - generalize L3 protocol denendent part.
* 23 Mar 2004: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp>
* - add support various size of conntrack structures.
* 26 Jan 2006: Harald Welte <laforge@netfilter.org>
* - restructure nf_conn (introduce nf_conn_help)
* - redesign 'features' how they were originally intended
* 26 Feb 2006: Pablo Neira Ayuso <pablo@eurodev.net>
* - add support for L3 protocol module load on demand.
*
* Derived from net/ipv4/netfilter/ip_conntrack_core.c
*/
#include <linux/types.h>
#include <linux/netfilter.h>
#include <linux/module.h>
#include <linux/skbuff.h>
#include <linux/proc_fs.h>
#include <linux/vmalloc.h>
#include <linux/stddef.h>
#include <linux/slab.h>
#include <linux/random.h>
#include <linux/jhash.h>
#include <linux/err.h>
#include <linux/percpu.h>
#include <linux/moduleparam.h>
#include <linux/notifier.h>
#include <linux/kernel.h>
#include <linux/netdevice.h>
#include <linux/socket.h>
/* This rwlock protects the main hash table, protocol/helper/expected
registrations, conntrack timers*/
#define ASSERT_READ_LOCK(x)
#define ASSERT_WRITE_LOCK(x)
#include <net/netfilter/nf_conntrack.h>
#include <net/netfilter/nf_conntrack_l3proto.h>
#include <net/netfilter/nf_conntrack_protocol.h>
#include <net/netfilter/nf_conntrack_expect.h>
#include <net/netfilter/nf_conntrack_helper.h>
#include <net/netfilter/nf_conntrack_core.h>
#define NF_CONNTRACK_VERSION "0.5.0"
#if 0
#define DEBUGP printk
#else
#define DEBUGP(format, args...)
#endif
DEFINE_RWLOCK(nf_conntrack_lock);
/* nf_conntrack_standalone needs this */
atomic_t nf_conntrack_count = ATOMIC_INIT(0);
void (*nf_conntrack_destroyed)(struct nf_conn *conntrack) = NULL;
unsigned int nf_conntrack_htable_size __read_mostly;
int nf_conntrack_max __read_mostly;
struct list_head *nf_conntrack_hash __read_mostly;
struct nf_conn nf_conntrack_untracked __read_mostly;
unsigned int nf_ct_log_invalid __read_mostly;
LIST_HEAD(unconfirmed);
static int nf_conntrack_vmalloc __read_mostly;
static unsigned int nf_conntrack_next_id;
DEFINE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat);
EXPORT_PER_CPU_SYMBOL(nf_conntrack_stat);
/*
* This scheme offers various size of "struct nf_conn" dependent on
* features(helper, nat, ...)
*/
#define NF_CT_FEATURES_NAMELEN 256
static struct {
/* name of slab cache. printed in /proc/slabinfo */
char *name;
/* size of slab cache */
size_t size;
/* slab cache pointer */
kmem_cache_t *cachep;
/* allocated slab cache + modules which uses this slab cache */
int use;
} nf_ct_cache[NF_CT_F_NUM];
/* protect members of nf_ct_cache except of "use" */
DEFINE_RWLOCK(nf_ct_cache_lock);
/* This avoids calling kmem_cache_create() with same name simultaneously */
static DEFINE_MUTEX(nf_ct_cache_mutex);
static int nf_conntrack_hash_rnd_initted;
static unsigned int nf_conntrack_hash_rnd;
static u_int32_t __hash_conntrack(const struct nf_conntrack_tuple *tuple,
unsigned int size, unsigned int rnd)
{
unsigned int a, b;
a = jhash((void *)tuple->src.u3.all, sizeof(tuple->src.u3.all),
((tuple->src.l3num) << 16) | tuple->dst.protonum);
b = jhash((void *)tuple->dst.u3.all, sizeof(tuple->dst.u3.all),
(tuple->src.u.all << 16) | tuple->dst.u.all);
return jhash_2words(a, b, rnd) % size;
}
static inline u_int32_t hash_conntrack(const struct nf_conntrack_tuple *tuple)
{
return __hash_conntrack(tuple, nf_conntrack_htable_size,
nf_conntrack_hash_rnd);
}
int nf_conntrack_register_cache(u_int32_t features, const char *name,
size_t size)
{
int ret = 0;
char *cache_name;
kmem_cache_t *cachep;
DEBUGP("nf_conntrack_register_cache: features=0x%x, name=%s, size=%d\n",
features, name, size);
if (features < NF_CT_F_BASIC || features >= NF_CT_F_NUM) {
DEBUGP("nf_conntrack_register_cache: invalid features.: 0x%x\n",
features);
return -EINVAL;
}
mutex_lock(&nf_ct_cache_mutex);
write_lock_bh(&nf_ct_cache_lock);
/* e.g: multiple helpers are loaded */
if (nf_ct_cache[features].use > 0) {
DEBUGP("nf_conntrack_register_cache: already resisterd.\n");
if ((!strncmp(nf_ct_cache[features].name, name,
NF_CT_FEATURES_NAMELEN))
&& nf_ct_cache[features].size == size) {
DEBUGP("nf_conntrack_register_cache: reusing.\n");
nf_ct_cache[features].use++;
ret = 0;
} else
ret = -EBUSY;
write_unlock_bh(&nf_ct_cache_lock);
mutex_unlock(&nf_ct_cache_mutex);
return ret;
}
write_unlock_bh(&nf_ct_cache_lock);
/*
* The memory space for name of slab cache must be alive until
* cache is destroyed.
*/
cache_name = kmalloc(sizeof(char)*NF_CT_FEATURES_NAMELEN, GFP_ATOMIC);
if (cache_name == NULL) {
DEBUGP("nf_conntrack_register_cache: can't alloc cache_name\n");
ret = -ENOMEM;
goto out_up_mutex;
}
if (strlcpy(cache_name, name, NF_CT_FEATURES_NAMELEN)
>= NF_CT_FEATURES_NAMELEN) {
printk("nf_conntrack_register_cache: name too long\n");
ret = -EINVAL;
goto out_free_name;
}
cachep = kmem_cache_create(cache_name, size, 0, 0,
NULL, NULL);
if (!cachep) {
printk("nf_conntrack_register_cache: Can't create slab cache "
"for the features = 0x%x\n", features);
ret = -ENOMEM;
goto out_free_name;
}
write_lock_bh(&nf_ct_cache_lock);
nf_ct_cache[features].use = 1;
nf_ct_cache[features].size = size;
nf_ct_cache[features].cachep = cachep;
nf_ct_cache[features].name = cache_name;
write_unlock_bh(&nf_ct_cache_lock);
goto out_up_mutex;
out_free_name:
kfree(cache_name);
out_up_mutex:
mutex_unlock(&nf_ct_cache_mutex);
return ret;
}
/* FIXME: In the current, only nf_conntrack_cleanup() can call this function. */
void nf_conntrack_unregister_cache(u_int32_t features)
{
kmem_cache_t *cachep;
char *name;
/*
* This assures that kmem_cache_create() isn't called before destroying
* slab cache.
*/
DEBUGP("nf_conntrack_unregister_cache: 0x%04x\n", features);
mutex_lock(&nf_ct_cache_mutex);
write_lock_bh(&nf_ct_cache_lock);
if (--nf_ct_cache[features].use > 0) {
write_unlock_bh(&nf_ct_cache_lock);
mutex_unlock(&nf_ct_cache_mutex);
return;
}
cachep = nf_ct_cache[features].cachep;
name = nf_ct_cache[features].name;
nf_ct_cache[features].cachep = NULL;
nf_ct_cache[features].name = NULL;
nf_ct_cache[features].size = 0;
write_unlock_bh(&nf_ct_cache_lock);
synchronize_net();
kmem_cache_destroy(cachep);
kfree(name);
mutex_unlock(&nf_ct_cache_mutex);
}
int
nf_ct_get_tuple(const struct sk_buff *skb,
unsigned int nhoff,
unsigned int dataoff,
u_int16_t l3num,
u_int8_t protonum,
struct nf_conntrack_tuple *tuple,
const struct nf_conntrack_l3proto *l3proto,
const struct nf_conntrack_protocol *protocol)
{
NF_CT_TUPLE_U_BLANK(tuple);
tuple->src.l3num = l3num;
if (l3proto->pkt_to_tuple(skb, nhoff, tuple) == 0)
return 0;
tuple->dst.protonum = protonum;
tuple->dst.dir = IP_CT_DIR_ORIGINAL;
return protocol->pkt_to_tuple(skb, dataoff, tuple);
}
int
nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse,
const struct nf_conntrack_tuple *orig,
const struct nf_conntrack_l3proto *l3proto,
const struct nf_conntrack_protocol *protocol)
{
NF_CT_TUPLE_U_BLANK(inverse);
inverse->src.l3num = orig->src.l3num;
if (l3proto->invert_tuple(inverse, orig) == 0)
return 0;
inverse->dst.dir = !orig->dst.dir;
inverse->dst.protonum = orig->dst.protonum;
return protocol->invert_tuple(inverse, orig);
}
static void
clean_from_lists(struct nf_conn *ct)
{
DEBUGP("clean_from_lists(%p)\n", ct);
ASSERT_WRITE_LOCK(&nf_conntrack_lock);
list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list);
list_del(&ct->tuplehash[IP_CT_DIR_REPLY].list);
/* Destroy all pending expectations */
nf_ct_remove_expectations(ct);
}
static void
destroy_conntrack(struct nf_conntrack *nfct)
{
struct nf_conn *ct = (struct nf_conn *)nfct;
struct nf_conntrack_l3proto *l3proto;
struct nf_conntrack_protocol *proto;
DEBUGP("destroy_conntrack(%p)\n", ct);
NF_CT_ASSERT(atomic_read(&nfct->use) == 0);
NF_CT_ASSERT(!timer_pending(&ct->timeout));
nf_conntrack_event(IPCT_DESTROY, ct);
set_bit(IPS_DYING_BIT, &ct->status);
/* To make sure we don't get any weird locking issues here:
* destroy_conntrack() MUST NOT be called with a write lock
* to nf_conntrack_lock!!! -HW */
l3proto = __nf_ct_l3proto_find(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num);
if (l3proto && l3proto->destroy)
l3proto->destroy(ct);
proto = __nf_ct_proto_find(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num, ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.protonum);
if (proto && proto->destroy)
proto->destroy(ct);
if (nf_conntrack_destroyed)
nf_conntrack_destroyed(ct);
write_lock_bh(&nf_conntrack_lock);
/* Expectations will have been removed in clean_from_lists,
* except TFTP can create an expectation on the first packet,
* before connection is in the list, so we need to clean here,
* too. */
nf_ct_remove_expectations(ct);
/* We overload first tuple to link into unconfirmed list. */
if (!nf_ct_is_confirmed(ct)) {
BUG_ON(list_empty(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list));
list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list);
}
NF_CT_STAT_INC(delete);
write_unlock_bh(&nf_conntrack_lock);
if (ct->master)
nf_ct_put(ct->master);
DEBUGP("destroy_conntrack: returning ct=%p to slab\n", ct);
nf_conntrack_free(ct);
}
static void death_by_timeout(unsigned long ul_conntrack)
{
struct nf_conn *ct = (void *)ul_conntrack;
write_lock_bh(&nf_conntrack_lock);
/* Inside lock so preempt is disabled on module removal path.
* Otherwise we can get spurious warnings. */
NF_CT_STAT_INC(delete_list);
clean_from_lists(ct);
write_unlock_bh(&nf_conntrack_lock);
nf_ct_put(ct);
}
struct nf_conntrack_tuple_hash *
__nf_conntrack_find(const struct nf_conntrack_tuple *tuple,
const struct nf_conn *ignored_conntrack)
{
struct nf_conntrack_tuple_hash *h;
unsigned int hash = hash_conntrack(tuple);
ASSERT_READ_LOCK(&nf_conntrack_lock);
list_for_each_entry(h, &nf_conntrack_hash[hash], list) {
if (nf_ct_tuplehash_to_ctrack(h) != ignored_conntrack &&
nf_ct_tuple_equal(tuple, &h->tuple)) {
NF_CT_STAT_INC(found);
return h;
}
NF_CT_STAT_INC(searched);
}
return NULL;
}
/* Find a connection corresponding to a tuple. */
struct nf_conntrack_tuple_hash *
nf_conntrack_find_get(const struct nf_conntrack_tuple *tuple,
const struct nf_conn *ignored_conntrack)
{
struct nf_conntrack_tuple_hash *h;
read_lock_bh(&nf_conntrack_lock);
h = __nf_conntrack_find(tuple, ignored_conntrack);
if (h)
atomic_inc(&nf_ct_tuplehash_to_ctrack(h)->ct_general.use);
read_unlock_bh(&nf_conntrack_lock);
return h;
}
static void __nf_conntrack_hash_insert(struct nf_conn *ct,
unsigned int hash,
unsigned int repl_hash)
{
ct->id = ++nf_conntrack_next_id;
list_add(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list,
&nf_conntrack_hash[hash]);
list_add(&ct->tuplehash[IP_CT_DIR_REPLY].list,
&nf_conntrack_hash[repl_hash]);
}
void nf_conntrack_hash_insert(struct nf_conn *ct)
{
unsigned int hash, repl_hash;
hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
repl_hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple);
write_lock_bh(&nf_conntrack_lock);
__nf_conntrack_hash_insert(ct, hash, repl_hash);
write_unlock_bh(&nf_conntrack_lock);
}
/* Confirm a connection given skb; places it in hash table */
int
__nf_conntrack_confirm(struct sk_buff **pskb)
{
unsigned int hash, repl_hash;
struct nf_conntrack_tuple_hash *h;
struct nf_conn *ct;
struct nf_conn_help *help;
enum ip_conntrack_info ctinfo;
ct = nf_ct_get(*pskb, &ctinfo);
/* ipt_REJECT uses nf_conntrack_attach to attach related
ICMP/TCP RST packets in other direction. Actual packet
which created connection will be IP_CT_NEW or for an
expected connection, IP_CT_RELATED. */
if (CTINFO2DIR(ctinfo) != IP_CT_DIR_ORIGINAL)
return NF_ACCEPT;
hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple);
repl_hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple);
/* We're not in hash table, and we refuse to set up related
connections for unconfirmed conns. But packet copies and
REJECT will give spurious warnings here. */
/* NF_CT_ASSERT(atomic_read(&ct->ct_general.use) == 1); */
/* No external references means noone else could have
confirmed us. */
NF_CT_ASSERT(!nf_ct_is_confirmed(ct));
DEBUGP("Confirming conntrack %p\n", ct);
write_lock_bh(&nf_conntrack_lock);
/* See if there's one in the list already, including reverse:
NAT could have grabbed it without realizing, since we're
not in the hash. If there is, we lost race. */
list_for_each_entry(h, &nf_conntrack_hash[hash], list)
if (nf_ct_tuple_equal(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple,
&h->tuple))
goto out;
list_for_each_entry(h, &nf_conntrack_hash[repl_hash], list)
if (nf_ct_tuple_equal(&ct->tuplehash[IP_CT_DIR_REPLY].tuple,
&h->tuple))
goto out;
/* Remove from unconfirmed list */
list_del(&ct->tuplehash[IP_CT_DIR_ORIGINAL].list);
__nf_conntrack_hash_insert(ct, hash, repl_hash);
/* Timer relative to confirmation time, not original
setting time, otherwise we'd get timer wrap in
weird delay cases. */
ct->timeout.expires += jiffies;
add_timer(&ct->timeout);
atomic_inc(&ct->ct_general.use);
set_bit(IPS_CONFIRMED_BIT, &ct->status);
NF_CT_STAT_INC(insert);
write_unlock_bh(&nf_conntrack_lock);
help = nfct_help(ct);
if (help && help->helper)
nf_conntrack_event_cache(IPCT_HELPER, *pskb);
#ifdef CONFIG_NF_NAT_NEEDED
if (test_bit(IPS_SRC_NAT_DONE_BIT, &ct->status) ||
test_bit(IPS_DST_NAT_DONE_BIT, &ct->status))
nf_conntrack_event_cache(IPCT_NATINFO, *pskb);
#endif
nf_conntrack_event_cache(master_ct(ct) ?
IPCT_RELATED : IPCT_NEW, *pskb);
return NF_ACCEPT;
out:
NF_CT_STAT_INC(insert_failed);
write_unlock_bh(&nf_conntrack_lock);
return NF_DROP;
}
/* Returns true if a connection correspondings to the tuple (required
for NAT). */
int
nf_conntrack_tuple_taken(const struct nf_conntrack_tuple *tuple,
const struct nf_conn *ignored_conntrack)
{
struct nf_conntrack_tuple_hash *h;
read_lock_bh(&nf_conntrack_lock);
h = __nf_conntrack_find(tuple, ignored_conntrack);
read_unlock_bh(&nf_conntrack_lock);
return h != NULL;
}
/* There's a small race here where we may free a just-assured
connection. Too bad: we're in trouble anyway. */
static int early_drop(struct list_head *chain)
{
/* Traverse backwards: gives us oldest, which is roughly LRU */
struct nf_conntrack_tuple_hash *h;
struct nf_conn *ct = NULL, *tmp;
int dropped = 0;
read_lock_bh(&nf_conntrack_lock);
list_for_each_entry_reverse(h, chain, list) {
tmp = nf_ct_tuplehash_to_ctrack(h);
if (!test_bit(IPS_ASSURED_BIT, &tmp->status)) {
ct = tmp;
atomic_inc(&ct->ct_general.use);
break;
}
}
read_unlock_bh(&nf_conntrack_lock);
if (!ct)
return dropped;
if (del_timer(&ct->timeout)) {
death_by_timeout((unsigned long)ct);
dropped = 1;
NF_CT_STAT_INC(early_drop);
}
nf_ct_put(ct);
return dropped;
}
static struct nf_conn *
__nf_conntrack_alloc(const struct nf_conntrack_tuple *orig,
const struct nf_conntrack_tuple *repl,
const struct nf_conntrack_l3proto *l3proto)
{
struct nf_conn *conntrack = NULL;
u_int32_t features = 0;
struct nf_conntrack_helper *helper;
if (unlikely(!nf_conntrack_hash_rnd_initted)) {
get_random_bytes(&nf_conntrack_hash_rnd, 4);
nf_conntrack_hash_rnd_initted = 1;
}
/* We don't want any race condition at early drop stage */
atomic_inc(&nf_conntrack_count);
if (nf_conntrack_max
&& atomic_read(&nf_conntrack_count) > nf_conntrack_max) {
unsigned int hash = hash_conntrack(orig);
/* Try dropping from this hash chain. */
if (!early_drop(&nf_conntrack_hash[hash])) {
atomic_dec(&nf_conntrack_count);
if (net_ratelimit())
printk(KERN_WARNING
"nf_conntrack: table full, dropping"
" packet.\n");
return ERR_PTR(-ENOMEM);
}
}
/* find features needed by this conntrack. */
features = l3proto->get_features(orig);
/* FIXME: protect helper list per RCU */
read_lock_bh(&nf_conntrack_lock);
helper = __nf_ct_helper_find(repl);
if (helper)
features |= NF_CT_F_HELP;
read_unlock_bh(&nf_conntrack_lock);
DEBUGP("nf_conntrack_alloc: features=0x%x\n", features);
read_lock_bh(&nf_ct_cache_lock);
if (unlikely(!nf_ct_cache[features].use)) {
DEBUGP("nf_conntrack_alloc: not supported features = 0x%x\n",
features);
goto out;
}
conntrack = kmem_cache_alloc(nf_ct_cache[features].cachep, GFP_ATOMIC);
if (conntrack == NULL) {
DEBUGP("nf_conntrack_alloc: Can't alloc conntrack from cache\n");
goto out;
}
memset(conntrack, 0, nf_ct_cache[features].size);
conntrack->features = features;
atomic_set(&conntrack->ct_general.use, 1);
conntrack->ct_general.destroy = destroy_conntrack;
conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple = *orig;
conntrack->tuplehash[IP_CT_DIR_REPLY].tuple = *repl;
/* Don't set timer yet: wait for confirmation */
init_timer(&conntrack->timeout);
conntrack->timeout.data = (unsigned long)conntrack;
conntrack->timeout.function = death_by_timeout;
read_unlock_bh(&nf_ct_cache_lock);
return conntrack;
out:
read_unlock_bh(&nf_ct_cache_lock);
atomic_dec(&nf_conntrack_count);
return conntrack;
}
struct nf_conn *nf_conntrack_alloc(const struct nf_conntrack_tuple *orig,
const struct nf_conntrack_tuple *repl)
{
struct nf_conntrack_l3proto *l3proto;
l3proto = __nf_ct_l3proto_find(orig->src.l3num);
return __nf_conntrack_alloc(orig, repl, l3proto);
}
void nf_conntrack_free(struct nf_conn *conntrack)
{
u_int32_t features = conntrack->features;
NF_CT_ASSERT(features >= NF_CT_F_BASIC && features < NF_CT_F_NUM);
DEBUGP("nf_conntrack_free: features = 0x%x, conntrack=%p\n", features,
conntrack);
kmem_cache_free(nf_ct_cache[features].cachep, conntrack);
atomic_dec(&nf_conntrack_count);
}
/* Allocate a new conntrack: we return -ENOMEM if classification
failed due to stress. Otherwise it really is unclassifiable. */
static struct nf_conntrack_tuple_hash *
init_conntrack(const struct nf_conntrack_tuple *tuple,
struct nf_conntrack_l3proto *l3proto,
struct nf_conntrack_protocol *protocol,
struct sk_buff *skb,
unsigned int dataoff)
{
struct nf_conn *conntrack;
struct nf_conntrack_tuple repl_tuple;
struct nf_conntrack_expect *exp;
if (!nf_ct_invert_tuple(&repl_tuple, tuple, l3proto, protocol)) {
DEBUGP("Can't invert tuple.\n");
return NULL;
}
conntrack = __nf_conntrack_alloc(tuple, &repl_tuple, l3proto);
if (conntrack == NULL || IS_ERR(conntrack)) {
DEBUGP("Can't allocate conntrack.\n");
return (struct nf_conntrack_tuple_hash *)conntrack;
}
if (!protocol->new(conntrack, skb, dataoff)) {
nf_conntrack_free(conntrack);
DEBUGP("init conntrack: can't track with proto module\n");
return NULL;
}
write_lock_bh(&nf_conntrack_lock);
exp = find_expectation(tuple);
if (exp) {
DEBUGP("conntrack: expectation arrives ct=%p exp=%p\n",
conntrack, exp);
/* Welcome, Mr. Bond. We've been expecting you... */
__set_bit(IPS_EXPECTED_BIT, &conntrack->status);
conntrack->master = exp->master;
#ifdef CONFIG_NF_CONNTRACK_MARK
conntrack->mark = exp->master->mark;
#endif
#ifdef CONFIG_NF_CONNTRACK_SECMARK
conntrack->secmark = exp->master->secmark;
#endif
nf_conntrack_get(&conntrack->master->ct_general);
NF_CT_STAT_INC(expect_new);
} else {
struct nf_conn_help *help = nfct_help(conntrack);
if (help)
help->helper = __nf_ct_helper_find(&repl_tuple);
NF_CT_STAT_INC(new);
}
/* Overload tuple linked list to put us in unconfirmed list. */
list_add(&conntrack->tuplehash[IP_CT_DIR_ORIGINAL].list, &unconfirmed);
write_unlock_bh(&nf_conntrack_lock);
if (exp) {
if (exp->expectfn)
exp->expectfn(conntrack, exp);
nf_conntrack_expect_put(exp);
}
return &conntrack->tuplehash[IP_CT_DIR_ORIGINAL];
}
/* On success, returns conntrack ptr, sets skb->nfct and ctinfo */
static inline struct nf_conn *
resolve_normal_ct(struct sk_buff *skb,
unsigned int dataoff,
u_int16_t l3num,
u_int8_t protonum,
struct nf_conntrack_l3proto *l3proto,
struct nf_conntrack_protocol *proto,
int *set_reply,
enum ip_conntrack_info *ctinfo)
{
struct nf_conntrack_tuple tuple;
struct nf_conntrack_tuple_hash *h;
struct nf_conn *ct;
if (!nf_ct_get_tuple(skb, (unsigned int)(skb->nh.raw - skb->data),
dataoff, l3num, protonum, &tuple, l3proto,
proto)) {
DEBUGP("resolve_normal_ct: Can't get tuple\n");
return NULL;
}
/* look for tuple match */
h = nf_conntrack_find_get(&tuple, NULL);
if (!h) {
h = init_conntrack(&tuple, l3proto, proto, skb, dataoff);
if (!h)
return NULL;
if (IS_ERR(h))
return (void *)h;
}
ct = nf_ct_tuplehash_to_ctrack(h);
/* It exists; we have (non-exclusive) reference. */
if (NF_CT_DIRECTION(h) == IP_CT_DIR_REPLY) {
*ctinfo = IP_CT_ESTABLISHED + IP_CT_IS_REPLY;
/* Please set reply bit if this packet OK */
*set_reply = 1;
} else {
/* Once we've had two way comms, always ESTABLISHED. */
if (test_bit(IPS_SEEN_REPLY_BIT, &ct->status)) {
DEBUGP("nf_conntrack_in: normal packet for %p\n", ct);
*ctinfo = IP_CT_ESTABLISHED;
} else if (test_bit(IPS_EXPECTED_BIT, &ct->status)) {
DEBUGP("nf_conntrack_in: related packet for %p\n", ct);
*ctinfo = IP_CT_RELATED;
} else {
DEBUGP("nf_conntrack_in: new packet for %p\n", ct);
*ctinfo = IP_CT_NEW;
}
*set_reply = 0;
}
skb->nfct = &ct->ct_general;
skb->nfctinfo = *ctinfo;
return ct;
}
unsigned int
nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff **pskb)
{
struct nf_conn *ct;
enum ip_conntrack_info ctinfo;
struct nf_conntrack_l3proto *l3proto;
struct nf_conntrack_protocol *proto;
unsigned int dataoff;
u_int8_t protonum;
int set_reply = 0;
int ret;
/* Previously seen (loopback or untracked)? Ignore. */
if ((*pskb)->nfct) {
NF_CT_STAT_INC(ignore);
return NF_ACCEPT;
}
l3proto = __nf_ct_l3proto_find((u_int16_t)pf);
if ((ret = l3proto->prepare(pskb, hooknum, &dataoff, &protonum)) <= 0) {
DEBUGP("not prepared to track yet or error occured\n");
return -ret;
}
proto = __nf_ct_proto_find((u_int16_t)pf, protonum);
/* It may be an special packet, error, unclean...
* inverse of the return code tells to the netfilter
* core what to do with the packet. */
if (proto->error != NULL &&
(ret = proto->error(*pskb, dataoff, &ctinfo, pf, hooknum)) <= 0) {
NF_CT_STAT_INC(error);
NF_CT_STAT_INC(invalid);
return -ret;
}
ct = resolve_normal_ct(*pskb, dataoff, pf, protonum, l3proto, proto,
&set_reply, &ctinfo);
if (!ct) {
/* Not valid part of a connection */
NF_CT_STAT_INC(invalid);
return NF_ACCEPT;
}
if (IS_ERR(ct)) {
/* Too stressed to deal. */
NF_CT_STAT_INC(drop);
return NF_DROP;
}
NF_CT_ASSERT((*pskb)->nfct);
ret = proto->packet(ct, *pskb, dataoff, ctinfo, pf, hooknum);
if (ret < 0) {
/* Invalid: inverse of the return code tells
* the netfilter core what to do */
DEBUGP("nf_conntrack_in: Can't track with proto module\n");
nf_conntrack_put((*pskb)->nfct);
(*pskb)->nfct = NULL;
NF_CT_STAT_INC(invalid);
return -ret;
}
if (set_reply && !test_and_set_bit(IPS_SEEN_REPLY_BIT, &ct->status))
nf_conntrack_event_cache(IPCT_STATUS, *pskb);
return ret;
}
int nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse,
const struct nf_conntrack_tuple *orig)
{
return nf_ct_invert_tuple(inverse, orig,
__nf_ct_l3proto_find(orig->src.l3num),
__nf_ct_proto_find(orig->src.l3num,
orig->dst.protonum));
}
/* Refresh conntrack for this many jiffies and do accounting if do_acct is 1 */
void __nf_ct_refresh_acct(struct nf_conn *ct,
enum ip_conntrack_info ctinfo,
const struct sk_buff *skb,
unsigned long extra_jiffies,
int do_acct)
{
int event = 0;
NF_CT_ASSERT(ct->timeout.data == (unsigned long)ct);
NF_CT_ASSERT(skb);
write_lock_bh(&nf_conntrack_lock);
/* Only update if this is not a fixed timeout */
if (test_bit(IPS_FIXED_TIMEOUT_BIT, &ct->status)) {
write_unlock_bh(&nf_conntrack_lock);
return;
}
/* If not in hash table, timer will not be active yet */
if (!nf_ct_is_confirmed(ct)) {
ct->timeout.expires = extra_jiffies;
event = IPCT_REFRESH;
} else {
/* Need del_timer for race avoidance (may already be dying). */
if (del_timer(&ct->timeout)) {
ct->timeout.expires = jiffies + extra_jiffies;
add_timer(&ct->timeout);
event = IPCT_REFRESH;
}
}
#ifdef CONFIG_NF_CT_ACCT
if (do_acct) {
ct->counters[CTINFO2DIR(ctinfo)].packets++;
ct->counters[CTINFO2DIR(ctinfo)].bytes +=
skb->len - (unsigned int)(skb->nh.raw - skb->data);
if ((ct->counters[CTINFO2DIR(ctinfo)].packets & 0x80000000)
|| (ct->counters[CTINFO2DIR(ctinfo)].bytes & 0x80000000))
event |= IPCT_COUNTER_FILLING;
}
#endif
write_unlock_bh(&nf_conntrack_lock);
/* must be unlocked when calling event cache */
if (event)
nf_conntrack_event_cache(event, skb);
}
#if defined(CONFIG_NF_CT_NETLINK) || \
defined(CONFIG_NF_CT_NETLINK_MODULE)
#include <linux/netfilter/nfnetlink.h>
#include <linux/netfilter/nfnetlink_conntrack.h>
#include <linux/mutex.h>
/* Generic function for tcp/udp/sctp/dccp and alike. This needs to be
* in ip_conntrack_core, since we don't want the protocols to autoload
* or depend on ctnetlink */
int nf_ct_port_tuple_to_nfattr(struct sk_buff *skb,
const struct nf_conntrack_tuple *tuple)
{
NFA_PUT(skb, CTA_PROTO_SRC_PORT, sizeof(u_int16_t),
&tuple->src.u.tcp.port);
NFA_PUT(skb, CTA_PROTO_DST_PORT, sizeof(u_int16_t),
&tuple->dst.u.tcp.port);
return 0;
nfattr_failure:
return -1;
}
static const size_t cta_min_proto[CTA_PROTO_MAX] = {
[CTA_PROTO_SRC_PORT-1] = sizeof(u_int16_t),
[CTA_PROTO_DST_PORT-1] = sizeof(u_int16_t)
};
int nf_ct_port_nfattr_to_tuple(struct nfattr *tb[],
struct nf_conntrack_tuple *t)
{
if (!tb[CTA_PROTO_SRC_PORT-1] || !tb[CTA_PROTO_DST_PORT-1])
return -EINVAL;
if (nfattr_bad_size(tb, CTA_PROTO_MAX, cta_min_proto))
return -EINVAL;
t->src.u.tcp.port =
*(u_int16_t *)NFA_DATA(tb[CTA_PROTO_SRC_PORT-1]);
t->dst.u.tcp.port =
*(u_int16_t *)NFA_DATA(tb[CTA_PROTO_DST_PORT-1]);
return 0;
}
#endif
/* Used by ipt_REJECT and ip6t_REJECT. */
void __nf_conntrack_attach(struct sk_buff *nskb, struct sk_buff *skb)
{
struct nf_conn *ct;
enum ip_conntrack_info ctinfo;
/* This ICMP is in reverse direction to the packet which caused it */
ct = nf_ct_get(skb, &ctinfo);
if (CTINFO2DIR(ctinfo) == IP_CT_DIR_ORIGINAL)
ctinfo = IP_CT_RELATED + IP_CT_IS_REPLY;
else
ctinfo = IP_CT_RELATED;
/* Attach to new skbuff, and increment count */
nskb->nfct = &ct->ct_general;
nskb->nfctinfo = ctinfo;
nf_conntrack_get(nskb->nfct);
}
static inline int
do_iter(const struct nf_conntrack_tuple_hash *i,
int (*iter)(struct nf_conn *i, void *data),
void *data)
{
return iter(nf_ct_tuplehash_to_ctrack(i), data);
}
/* Bring out ya dead! */
static struct nf_conn *
get_next_corpse(int (*iter)(struct nf_conn *i, void *data),
void *data, unsigned int *bucket)
{
struct nf_conntrack_tuple_hash *h;
struct nf_conn *ct;
write_lock_bh(&nf_conntrack_lock);
for (; *bucket < nf_conntrack_htable_size; (*bucket)++) {
list_for_each_entry(h, &nf_conntrack_hash[*bucket], list) {
ct = nf_ct_tuplehash_to_ctrack(h);
if (iter(ct, data))
goto found;
}
}
list_for_each_entry(h, &unconfirmed, list) {
ct = nf_ct_tuplehash_to_ctrack(h);
if (iter(ct, data))
goto found;
}
write_unlock_bh(&nf_conntrack_lock);
return NULL;
found:
atomic_inc(&ct->ct_general.use);
write_unlock_bh(&nf_conntrack_lock);
return ct;
}
void
nf_ct_iterate_cleanup(int (*iter)(struct nf_conn *i, void *data), void *data)
{
struct nf_conn *ct;
unsigned int bucket = 0;
while ((ct = get_next_corpse(iter, data, &bucket)) != NULL) {
/* Time to push up daises... */
if (del_timer(&ct->timeout))
death_by_timeout((unsigned long)ct);
/* ... else the timer will get him soon. */
nf_ct_put(ct);
}
}
static int kill_all(struct nf_conn *i, void *data)
{
return 1;
}
static void free_conntrack_hash(struct list_head *hash, int vmalloced, int size)
{
if (vmalloced)
vfree(hash);
else
free_pages((unsigned long)hash,
get_order(sizeof(struct list_head) * size));
}
void nf_conntrack_flush()
{
nf_ct_iterate_cleanup(kill_all, NULL);
}
/* Mishearing the voices in his head, our hero wonders how he's
supposed to kill the mall. */
void nf_conntrack_cleanup(void)
{
int i;
ip_ct_attach = NULL;
/* This makes sure all current packets have passed through
netfilter framework. Roll on, two-stage module
delete... */
synchronize_net();
nf_ct_event_cache_flush();
i_see_dead_people:
nf_conntrack_flush();
if (atomic_read(&nf_conntrack_count) != 0) {
schedule();
goto i_see_dead_people;
}
/* wait until all references to nf_conntrack_untracked are dropped */
while (atomic_read(&nf_conntrack_untracked.ct_general.use) > 1)
schedule();
for (i = 0; i < NF_CT_F_NUM; i++) {
if (nf_ct_cache[i].use == 0)
continue;
NF_CT_ASSERT(nf_ct_cache[i].use == 1);
nf_ct_cache[i].use = 1;
nf_conntrack_unregister_cache(i);
}
kmem_cache_destroy(nf_conntrack_expect_cachep);
free_conntrack_hash(nf_conntrack_hash, nf_conntrack_vmalloc,
nf_conntrack_htable_size);
/* free l3proto protocol tables */
for (i = 0; i < PF_MAX; i++)
if (nf_ct_protos[i]) {
kfree(nf_ct_protos[i]);
nf_ct_protos[i] = NULL;
}
}
static struct list_head *alloc_hashtable(int size, int *vmalloced)
{
struct list_head *hash;
unsigned int i;
*vmalloced = 0;
hash = (void*)__get_free_pages(GFP_KERNEL,
get_order(sizeof(struct list_head)
* size));
if (!hash) {
*vmalloced = 1;
printk(KERN_WARNING "nf_conntrack: falling back to vmalloc.\n");
hash = vmalloc(sizeof(struct list_head) * size);
}
if (hash)
for (i = 0; i < size; i++)
INIT_LIST_HEAD(&hash[i]);
return hash;
}
int set_hashsize(const char *val, struct kernel_param *kp)
{
int i, bucket, hashsize, vmalloced;
int old_vmalloced, old_size;
int rnd;
struct list_head *hash, *old_hash;
struct nf_conntrack_tuple_hash *h;
/* On boot, we can set this without any fancy locking. */
if (!nf_conntrack_htable_size)
return param_set_uint(val, kp);
hashsize = simple_strtol(val, NULL, 0);
if (!hashsize)
return -EINVAL;
hash = alloc_hashtable(hashsize, &vmalloced);
if (!hash)
return -ENOMEM;
/* We have to rehahs for the new table anyway, so we also can
* use a newrandom seed */
get_random_bytes(&rnd, 4);
write_lock_bh(&nf_conntrack_lock);
for (i = 0; i < nf_conntrack_htable_size; i++) {
while (!list_empty(&nf_conntrack_hash[i])) {
h = list_entry(nf_conntrack_hash[i].next,
struct nf_conntrack_tuple_hash, list);
list_del(&h->list);
bucket = __hash_conntrack(&h->tuple, hashsize, rnd);
list_add_tail(&h->list, &hash[bucket]);
}
}
old_size = nf_conntrack_htable_size;
old_vmalloced = nf_conntrack_vmalloc;
old_hash = nf_conntrack_hash;
nf_conntrack_htable_size = hashsize;
nf_conntrack_vmalloc = vmalloced;
nf_conntrack_hash = hash;
nf_conntrack_hash_rnd = rnd;
write_unlock_bh(&nf_conntrack_lock);
free_conntrack_hash(old_hash, old_vmalloced, old_size);
return 0;
}
module_param_call(hashsize, set_hashsize, param_get_uint,
&nf_conntrack_htable_size, 0600);
int __init nf_conntrack_init(void)
{
unsigned int i;
int ret;
/* Idea from tcp.c: use 1/16384 of memory. On i386: 32MB
* machine has 256 buckets. >= 1GB machines have 8192 buckets. */
if (!nf_conntrack_htable_size) {
nf_conntrack_htable_size
= (((num_physpages << PAGE_SHIFT) / 16384)
/ sizeof(struct list_head));
if (num_physpages > (1024 * 1024 * 1024 / PAGE_SIZE))
nf_conntrack_htable_size = 8192;
if (nf_conntrack_htable_size < 16)
nf_conntrack_htable_size = 16;
}
nf_conntrack_max = 8 * nf_conntrack_htable_size;
printk("nf_conntrack version %s (%u buckets, %d max)\n",
NF_CONNTRACK_VERSION, nf_conntrack_htable_size,
nf_conntrack_max);
nf_conntrack_hash = alloc_hashtable(nf_conntrack_htable_size,
&nf_conntrack_vmalloc);
if (!nf_conntrack_hash) {
printk(KERN_ERR "Unable to create nf_conntrack_hash\n");
goto err_out;
}
ret = nf_conntrack_register_cache(NF_CT_F_BASIC, "nf_conntrack:basic",
sizeof(struct nf_conn));
if (ret < 0) {
printk(KERN_ERR "Unable to create nf_conn slab cache\n");
goto err_free_hash;
}
nf_conntrack_expect_cachep = kmem_cache_create("nf_conntrack_expect",
sizeof(struct nf_conntrack_expect),
0, 0, NULL, NULL);
if (!nf_conntrack_expect_cachep) {
printk(KERN_ERR "Unable to create nf_expect slab cache\n");
goto err_free_conntrack_slab;
}
/* Don't NEED lock here, but good form anyway. */
write_lock_bh(&nf_conntrack_lock);
for (i = 0; i < PF_MAX; i++)
nf_ct_l3protos[i] = &nf_conntrack_generic_l3proto;
write_unlock_bh(&nf_conntrack_lock);
/* For use by REJECT target */
ip_ct_attach = __nf_conntrack_attach;
/* Set up fake conntrack:
- to never be deleted, not in any hashes */
atomic_set(&nf_conntrack_untracked.ct_general.use, 1);
/* - and look it like as a confirmed connection */
set_bit(IPS_CONFIRMED_BIT, &nf_conntrack_untracked.status);
return ret;
err_free_conntrack_slab:
nf_conntrack_unregister_cache(NF_CT_F_BASIC);
err_free_hash:
free_conntrack_hash(nf_conntrack_hash, nf_conntrack_vmalloc,
nf_conntrack_htable_size);
err_out:
return -ENOMEM;
}
|