-
Notifications
You must be signed in to change notification settings - Fork 34
/
natcap_peer.c
5722 lines (5097 loc) · 179 KB
/
natcap_peer.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
/*
* Author: Chen Minqiang <[email protected]>
* Date : Thu, 30 Aug 2018 11:25:35 +0800
*
* This file is part of the natcap.
*
* natcap is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 3 of the License, or
* (at your option) any later version.
*
* natcap is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with natcap; see the file COPYING. If not, see
* <http://www.gnu.org/licenses/>.
*/
#include <linux/ctype.h>
#include <linux/cdev.h>
#include <linux/device.h>
#include <linux/if_ether.h>
#include <linux/if_packet.h>
#include <linux/if_arp.h>
#include <linux/init.h>
#include <linux/ip.h>
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/module.h>
#include <linux/netdevice.h>
#include <linux/inetdevice.h>
#include <linux/netfilter.h>
#include <linux/skbuff.h>
#include <linux/string.h>
#include <linux/tcp.h>
#include <linux/udp.h>
#include <linux/timer.h>
#include <linux/version.h>
#include <linux/vmalloc.h>
#include <asm/unaligned.h>
#include <net/netfilter/nf_conntrack.h>
#include <net/netfilter/nf_conntrack_helper.h>
#include <net/netfilter/nf_conntrack_acct.h>
#include "net/netfilter/nf_conntrack_seqadj.h"
#include "natcap_common.h"
#include "natcap_peer.h"
#include "natcap_client.h"
#include "natcap_knock.h"
static unsigned int peer_open_portmap = 0;
static unsigned int peer_mode = 0;
static unsigned int peer_max_pmtu = 1420;
static unsigned int peer_sni_ban = 0;
static struct in6_addr peer_local_ip6_addr;
struct peer_cache_node {
struct nf_conn *user;
struct sk_buff *skb;
unsigned long jiffies;
};
DEFINE_SPINLOCK(peer_cache_lock);
#if defined(CONFIG_64BIT) || defined(CONFIG_X86) || defined(CONFIG_X86_64) || defined(CONFIG_ARM) || defined(CONFIG_ARM64)
#define MAX_PEER_CACHE 1024
#else
#define MAX_PEER_CACHE 64
#endif
static unsigned short peer_cache_next_to_clean = 0;
static unsigned short peer_cache_next_to_use = 0;
static struct peer_cache_node peer_cache[MAX_PEER_CACHE];
#define PEER_CACHE_TIMEOUT 4
__be32 peer_pub_ip[PEER_PUB_NUM];
unsigned int peer_pub_active[PEER_PUB_NUM];
unsigned int peer_pub_idx = 0;
static unsigned int rt_out_magic = 0;
static struct natcap_fastpath_route peer_fastpath_route[MAX_PEER_NUM];
struct natcap_fastpath_route *natcap_pfr = peer_fastpath_route;
int is_fastpath_route_ready(struct natcap_fastpath_route *pfr)
{
return (!pfr->is_dead &&
pfr->rt_out_magic == rt_out_magic &&
pfr->rt_out.outdev != NULL);
}
static inline void peer_cache_init(void)
{
spin_lock_bh(&peer_cache_lock);
memset(peer_cache, 0, sizeof(struct peer_cache_node) * MAX_PEER_CACHE);
spin_unlock_bh(&peer_cache_lock);
}
static inline int peer_cache_attach(struct nf_conn *ct, struct sk_buff *skb)
{
struct natcap_session *ns = natcap_session_get(ct);
//XXX we use p.cache_index - 1 as index
if (ns == NULL || ns->p.cache_index != 0) {
return -EINVAL;
}
spin_lock_bh(&peer_cache_lock);
if ((peer_cache_next_to_use + 1) % MAX_PEER_CACHE == peer_cache_next_to_clean) {
spin_unlock_bh(&peer_cache_lock);
return -ENOSPC;
}
nf_conntrack_get(&ct->ct_general);
peer_cache[peer_cache_next_to_use].jiffies = jiffies;
peer_cache[peer_cache_next_to_use].user = ct;
peer_cache[peer_cache_next_to_use].skb = skb;
ns->p.cache_index = peer_cache_next_to_use + 1;
peer_cache_next_to_use = (peer_cache_next_to_use + 1) % MAX_PEER_CACHE;
spin_unlock_bh(&peer_cache_lock);
return 0;
}
static inline struct sk_buff *peer_cache_detach(struct nf_conn *ct)
{
unsigned short i;
struct sk_buff *skb = NULL;
struct natcap_session *ns = natcap_session_get(ct);
if (ns == NULL)
return NULL;
i = ns->p.cache_index;
if (i == 0 || i > MAX_PEER_CACHE)
return NULL;
i = i - 1;
spin_lock_bh(&peer_cache_lock);
if (peer_cache[i].user == ct) {
ns->p.cache_index = 0;
nf_ct_put(peer_cache[i].user);
peer_cache[i].user = NULL;
if (peer_cache[i].skb != NULL) {
skb = peer_cache[i].skb;
peer_cache[i].skb = NULL;
}
}
spin_unlock_bh(&peer_cache_lock);
return skb;
}
static inline void peer_cache_cleaner(void)
{
unsigned short i;
struct natcap_session *ns;
if (peer_cache[peer_cache_next_to_clean].user != NULL &&
!time_after(jiffies, peer_cache[peer_cache_next_to_clean].jiffies + PEER_CACHE_TIMEOUT * HZ))
return;
spin_lock_bh(&peer_cache_lock);
i = peer_cache_next_to_clean;
while (i != peer_cache_next_to_use) {
if (peer_cache[i].user == NULL) {
i = (i + 1) % MAX_PEER_CACHE;
continue;
}
if (!time_after(jiffies, peer_cache[i].jiffies + PEER_CACHE_TIMEOUT * HZ))
break;
ns = natcap_session_get(peer_cache[i].user);
ns->p.cache_index = 0;
nf_ct_put(peer_cache[i].user);
peer_cache[i].user = NULL;
if (peer_cache[i].skb != NULL) {
consume_skb(peer_cache[i].skb);
peer_cache[i].skb = NULL;
}
i = (i + 1) % MAX_PEER_CACHE;
}
peer_cache_next_to_clean = i;
spin_unlock_bh(&peer_cache_lock);
}
static inline void peer_cache_cleanup(void)
{
unsigned short i;
struct natcap_session *ns;
spin_lock_bh(&peer_cache_lock);
for (i = 0; i < MAX_PEER_CACHE; i++) {
if (peer_cache[i].user == NULL)
continue;
ns = natcap_session_get(peer_cache[i].user);
//BUG_ON(ns == NULL);
ns->p.cache_index = 0;
nf_ct_put(peer_cache[i].user);
peer_cache[i].user = NULL;
if (peer_cache[i].skb != NULL) {
consume_skb(peer_cache[i].skb);
peer_cache[i].skb = NULL;
}
}
spin_unlock_bh(&peer_cache_lock);
}
struct peer_sni_cache_node {
unsigned long active_jiffies;
__be32 src_ip;
__be16 src_port;
unsigned short add_data_len;
struct sk_buff *skb;
};
#define MAX_PEER_SNI_CACHE_NODE 64
static struct peer_sni_cache_node peer_sni_cache[NR_CPUS][MAX_PEER_SNI_CACHE_NODE];
static inline void peer_sni_cache_init(void)
{
int i, j;
for (i = 0; i < NR_CPUS; i++) {
for (j = 0; j < MAX_PEER_SNI_CACHE_NODE; j++) {
peer_sni_cache[i][j].skb = NULL;
}
}
}
static inline void peer_sni_cache_cleanup(void)
{
int i, j;
for (i = 0; i < NR_CPUS; i++) {
for (j = 0; j < MAX_PEER_SNI_CACHE_NODE; j++) {
if (peer_sni_cache[i][j].skb != NULL) {
consume_skb(peer_sni_cache[i][j].skb);
peer_sni_cache[i][j].skb = NULL;
}
}
}
}
static inline int peer_sni_cache_attach(__be32 src_ip, __be16 src_port, struct sk_buff *skb, unsigned short add_data_len)
{
int i = smp_processor_id();
int j;
int next_to_use = MAX_PEER_SNI_CACHE_NODE;
for (j = 0; j < MAX_PEER_SNI_CACHE_NODE; j++) {
if (peer_sni_cache[i][j].src_ip == src_ip) {
if (peer_sni_cache[i][j].src_port == src_port) {
return -EEXIST;
}
} else if (next_to_use == MAX_PEER_SNI_CACHE_NODE && peer_sni_cache[i][j].skb == NULL) {
next_to_use = j;
}
}
if (next_to_use == MAX_PEER_SNI_CACHE_NODE) {
return -ENOMEM;
}
peer_sni_cache[i][next_to_use].src_ip = src_ip;
peer_sni_cache[i][next_to_use].src_port = src_port;
peer_sni_cache[i][next_to_use].add_data_len = add_data_len;
peer_sni_cache[i][next_to_use].skb = skb;
peer_sni_cache[i][next_to_use].active_jiffies = (unsigned long)jiffies;
return 0;
}
static inline struct sk_buff *peer_sni_cache_detach(__be32 src_ip, __be16 src_port, unsigned short *add_data_len)
{
int i = smp_processor_id();
int j = 0;
struct sk_buff *skb = NULL;
for (j = 0; j < MAX_PEER_SNI_CACHE_NODE; j++) {
if (peer_sni_cache[i][j].skb != NULL) {
if (time_after(jiffies, peer_sni_cache[i][j].active_jiffies + PEER_CACHE_TIMEOUT * HZ)) {
consume_skb(peer_sni_cache[i][j].skb);
peer_sni_cache[i][j].skb = NULL;
} else if (peer_sni_cache[i][j].src_ip == src_ip) {
if (peer_sni_cache[i][j].src_port == src_port) {
skb = peer_sni_cache[i][j].skb;
*add_data_len = peer_sni_cache[i][j].add_data_len;
peer_sni_cache[i][j].skb = NULL;
break;
}
}
}
}
for (; j < MAX_PEER_SNI_CACHE_NODE; j++) {
if (peer_sni_cache[i][j].skb != NULL) {
if (time_after(jiffies, peer_sni_cache[i][j].active_jiffies + PEER_CACHE_TIMEOUT * HZ)) {
consume_skb(peer_sni_cache[i][j].skb);
peer_sni_cache[i][j].skb = NULL;
}
}
}
return skb;
}
static int peer_stop = 1;
static int peer_subtype = 0;
static unsigned int peer_sni_auth = 0;
static __be32 peer_sni_ip = __constant_htonl(0);
__be16 peer_sni_port = __constant_htons(991);
static void *peer_xsyn_last_dev = NULL;
__be32 peer_xsyn_enumerate_addr(void)
{
__be32 ip;
struct net_device *dev;
struct net_device *last_dev;
struct in_device *indev;
struct in_ifaddr *ifa;
last_dev = peer_xsyn_last_dev;
rcu_read_lock();
for(dev = first_net_device(&init_net); dev != NULL && dev != last_dev; dev = next_net_device(dev));
if (dev == peer_xsyn_last_dev && peer_xsyn_last_dev != NULL) {
dev = next_net_device(dev);
}
if (dev == NULL) {
dev = first_net_device(&init_net);
}
last_dev = dev;
for (; dev != NULL; dev = next_net_device(dev)) {
indev = __in_dev_get_rcu(dev);
if (indev && indev->ifa_list) {
ifa = indev->ifa_list;
ip = ifa->ifa_local;
if (!ipv4_is_loopback(ip)) {
peer_xsyn_last_dev = dev;
rcu_read_unlock();
return ip;
}
}
}
for (dev = first_net_device(&init_net); dev != NULL && dev != last_dev; dev = next_net_device(dev)) {
indev = __in_dev_get_rcu(dev);
if (indev && indev->ifa_list) {
ifa = indev->ifa_list;
ip = ifa->ifa_local;
if (!ipv4_is_loopback(ip)) {
peer_xsyn_last_dev = dev;
rcu_read_unlock();
return ip;
}
}
}
peer_xsyn_last_dev = last_dev;
rcu_read_unlock();
return 0;
}
static __be32 peer_knock_ip = __constant_htonl(0);
static __be16 peer_knock_port = __constant_htons(22);
static unsigned char peer_knock_mac[ETH_ALEN] = { };
__be16 peer_knock_local_port = __constant_htons(997);
static inline __be32 gen_seq_number(void)
{
__be32 s;
do {
s = get_random_u32();
} while (s == 0);
return s;
}
static __be32 peer_local_ip = __constant_htonl(0);
static __be16 peer_local_port = __constant_htons(443);
#define ICMP_PAYLOAD_LIMIT 1024
#define MAX_PEER_SERVER 8
struct peer_server_node peer_server[MAX_PEER_SERVER];
static inline __be16 peer_fakeuser_sport(struct nf_conn *user)
{
if (!user)
return 0;
return user->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u.all;
}
static inline __be16 peer_fakeuser_dport(struct nf_conn *user)
{
if (!user)
return 0;
return user->tuplehash[IP_CT_DIR_REPLY].tuple.src.u.all;
}
#define MAX_PEER_PORT_MAP 65536
static struct nf_conn **peer_port_map = NULL;
DEFINE_SPINLOCK(peer_port_map_lock);
static struct timer_list peer_timer;
#define NATCAP_PEER_EXPECT_TIMEOUT 5
#define NATCAP_PEER_USER_TIMEOUT_DEFAULT 180
unsigned int peer_port_map_timeout = NATCAP_PEER_USER_TIMEOUT_DEFAULT;
#define NATCAP_PEER_CONN_TIMEOUT_DEFAULT 180
unsigned int peer_conn_timeout = NATCAP_PEER_CONN_TIMEOUT_DEFAULT;
#define PEER_PORT_MAP_FLUSH_STEP 256
#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 15, 0)
static void peer_timer_flush(unsigned long ignore)
#else
static void peer_timer_flush(struct timer_list *ignore)
#endif
{
static unsigned short flush_idx = 0;
unsigned int i, j = 0;
struct nf_conn *user;
for (i = 0; i < PEER_PORT_MAP_FLUSH_STEP; i++) {
if (peer_port_map[flush_idx] == NULL) {
flush_idx++;
continue;
}
spin_lock_bh(&peer_port_map_lock);
user = peer_port_map[flush_idx];
if (user != NULL) {
unsigned char client_mac[ETH_ALEN];
struct user_expect *ue = peer_user_expect(user);
if (after(jiffies, ue->last_active + peer_port_map_timeout * HZ)) {
set_byte4(client_mac, get_byte4((void *)&user->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip));
set_byte2(client_mac + 4, get_byte2((void *)&user->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u.all));
NATCAP_INFO(DEBUG_FMT_PREFIX "C[%02x:%02x:%02x:%02x:%02x:%02x,%pI4,%pI4] P=%u [AS %ds] timeout drop\n", DEBUG_ARG_PREFIX,
client_mac[0], client_mac[1], client_mac[2], client_mac[3], client_mac[4], client_mac[5],
&ue->local_ip, &ue->ip, ntohs(ue->map_port), ue->last_active != 0 ? (uintmindiff(ue->last_active, jiffies) + HZ / 2) / HZ : (-1)
);
peer_port_map[flush_idx] = NULL;
nf_ct_put(user);
j++;
}
}
spin_unlock_bh(&peer_port_map_lock);
flush_idx++;
}
for (i = 0; i < MAX_PEER_SERVER; i++) {
struct peer_server_node *ps = &peer_server[i];
spin_lock_bh(&ps->lock);
for (j = 0; j < MAX_PEER_CONN; j++) {
user = ps->port_map[j];
if (user != NULL) {
struct fakeuser_expect *fue = peer_fakeuser_expect(user);
if (after(jiffies, fue->last_active + peer_conn_timeout * HZ)) {
NATCAP_INFO(DEBUG_FMT_PREFIX "conn[%u:%u] @N[[%pI4:%u] [AS %ds] timeout drop\n", DEBUG_ARG_PREFIX,
ntohs(peer_fakeuser_sport(user)), ntohs(peer_fakeuser_dport(user)),
&ps->ip, ntohs(ps->map_port), fue->last_active != 0 ?(uintmindiff(fue->last_active, jiffies) + HZ / 2) / HZ : (-1)
);
peer_server[i].port_map[j] = NULL;
nf_ct_put(user);
}
}
}
spin_unlock_bh(&ps->lock);
}
peer_cache_cleaner();
if (peer_stop) {
return;
}
mod_timer(&peer_timer, jiffies + HZ / 2);
}
static inline void peer_port_map_kill(unsigned short idx)
{
struct nf_conn *user;
spin_lock_bh(&peer_port_map_lock);
user = peer_port_map[idx];
if (user != NULL) {
unsigned char client_mac[ETH_ALEN];
struct user_expect *ue = peer_user_expect(user);
set_byte4(client_mac, get_byte4((void *)&user->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip));
set_byte2(client_mac + 4, get_byte2((void *)&user->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u.all));
NATCAP_INFO(DEBUG_FMT_PREFIX "C[%02x:%02x:%02x:%02x:%02x:%02x,%pI4,%pI4] P=%u [AS %ds] killed\n", DEBUG_ARG_PREFIX,
client_mac[0], client_mac[1], client_mac[2], client_mac[3], client_mac[4], client_mac[5],
&ue->local_ip, &ue->ip, ntohs(ue->map_port), ue->last_active != 0 ? (uintmindiff(ue->last_active, jiffies) + HZ / 2) / HZ : (-1)
);
peer_port_map[idx] = NULL;
nf_ct_put(user);
}
spin_unlock_bh(&peer_port_map_lock);
}
static int peer_timer_init(void)
{
struct timer_list *timer = &peer_timer;
#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 15, 0)
init_timer(timer);
timer->data = 0;
timer->function = peer_timer_flush;
#else
timer_setup(timer, peer_timer_flush, 0);
#endif
return 0;
}
static void peer_timer_start(void)
{
struct timer_list *timer = &peer_timer;
mod_timer(timer, jiffies + 8 * HZ);
}
static void peer_timer_exit(void)
{
del_timer(&peer_timer);
}
static inline struct nf_conn *get_peer_user(unsigned int port)
{
struct nf_conn *user;
if (port >= MAX_PEER_PORT_MAP || peer_port_map[port] == NULL)
return NULL;
spin_lock_bh(&peer_port_map_lock);
user = peer_port_map[port];
if (user) {
nf_conntrack_get(&user->ct_general);
}
spin_unlock_bh(&peer_port_map_lock);
return user;
}
static inline void put_peer_user(struct nf_conn *user)
{
nf_ct_put(user);
}
static __be16 alloc_peer_port(struct nf_conn *user, const unsigned char *mac)
{
static unsigned int seed_rnd;
unsigned short port;
unsigned int hash;
unsigned int data = get_byte4(mac);
get_random_once(&seed_rnd, sizeof(seed_rnd));
hash = jhash2(&data, 1, get_byte2(mac + 4)^seed_rnd);
port = 1024 + hash % (MAX_PEER_PORT_MAP - 1024);
for (; port < MAX_PEER_PORT_MAP - 1; port++) {
if (peer_port_map[port] == NULL) {
spin_lock_bh(&peer_port_map_lock);
//re-check-in-lock
if (peer_port_map[port] == NULL) {
peer_port_map[port] = user;
nf_conntrack_get(&user->ct_general);
spin_unlock_bh(&peer_port_map_lock);
return htons(port);
}
spin_unlock_bh(&peer_port_map_lock);
}
}
for (port = 1024; port < 1024 + hash % (MAX_PEER_PORT_MAP - 1024); port++) {
if (peer_port_map[port] == NULL) {
spin_lock_bh(&peer_port_map_lock);
//re-check-in-lock
if (peer_port_map[port] == NULL) {
peer_port_map[port] = user;
nf_conntrack_get(&user->ct_general);
spin_unlock_bh(&peer_port_map_lock);
return htons(port);
}
spin_unlock_bh(&peer_port_map_lock);
}
}
return 0;
}
static struct peer_server_node *peer_server_node_in(__be32 ip, unsigned short conn, int new)
{
unsigned int i;
unsigned long maxdiff = 0;
unsigned long last_jiffies = jiffies;
struct peer_server_node *ps = NULL;
if (conn <= 0)
conn = 1;
if (ip == 0)
return NULL;
for (i = 0; i < MAX_PEER_SERVER; i++) {
if (peer_server[i].ip == ip) {
spin_lock_bh(&peer_server[i].lock);
//re-check-in-lock
if (peer_server[i].ip == ip) {
if (new == 1 && peer_server[i].conn != conn) {
peer_server[i].conn = conn;
}
spin_unlock_bh(&peer_server[i].lock);
return &peer_server[i];
}
spin_unlock_bh(&peer_server[i].lock);
break;
}
}
if (new == 0)
return NULL;
for (i = 0; i < MAX_PEER_SERVER; i++) {
if (peer_server[i].ip == 0) {
spin_lock_bh(&peer_server[i].lock);
if (peer_server[i].ip == 0) {
ps = &peer_server[i];
goto init_out;
}
spin_unlock_bh(&peer_server[i].lock);
}
if (maxdiff < uintmindiff(peer_server[i].last_active, last_jiffies)) {
maxdiff = uintmindiff(peer_server[i].last_active, last_jiffies);
ps = &peer_server[i];
}
}
if (ps) {
spin_lock_bh(&ps->lock);
init_out:
if (ps->ip != 0) {
NATCAP_WARN(DEBUG_FMT_PREFIX "drop the old server %pI4 map_port=%u replace new=%pI4\n",
DEBUG_ARG_PREFIX, &ps->ip, ntohs(ps->map_port), &ip);
}
for (i = 0; i < MAX_PEER_CONN; i++) {
if (ps->port_map[i] != NULL) {
nf_ct_put(ps->port_map[i]);
ps->port_map[i] = NULL;
}
}
ps->ip = ip;
ps->map_port = 0;
ps->conn = conn;
ps->last_active = 0;
ps->last_inuse = 0;
spin_unlock_bh(&ps->lock);
}
return ps;
}
static void natcap_user_timeout_touch(struct nf_conn *ct, unsigned long timeout)
{
#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 9, 0)
unsigned long newtimeout = jiffies + timeout * HZ;
if (newtimeout - ct->timeout.expires > HZ) {
mod_timer_pending(&ct->timeout, newtimeout);
}
#else
ct->timeout = jiffies + timeout * HZ;
#endif
}
static struct nf_conn *peer_fakeuser_expect_new(__be32 saddr, __be32 daddr, __be16 sport, __be16 dport, int pmi)
{
struct fakeuser_expect *fue;
struct nf_conn *user;
struct nf_ct_ext *new = NULL;
enum ip_conntrack_info ctinfo;
unsigned int newoff = 0;
int ret;
struct sk_buff *uskb;
struct iphdr *iph;
struct udphdr *udph;
uskb = uskb_of_this_cpu();
if (uskb == NULL) {
return NULL;
}
skb_reset_transport_header(uskb);
skb_reset_network_header(uskb);
skb_reset_mac_len(uskb);
uskb->protocol = __constant_htons(ETH_P_IP);
skb_set_tail_pointer(uskb, PEER_USKB_SIZE);
uskb->len = PEER_USKB_SIZE;
uskb->pkt_type = PACKET_HOST;
uskb->transport_header = uskb->network_header + sizeof(struct iphdr);
iph = ip_hdr(uskb);
iph->version = 4;
iph->ihl = 5;
iph->saddr = saddr;
iph->daddr = daddr;
iph->tos = 0;
iph->tot_len = htons(PEER_USKB_SIZE);
iph->ttl=255;
iph->protocol = IPPROTO_UDP;
iph->id = __constant_htons(0xdead);
iph->frag_off = 0;
iph->check = 0;
iph->check = ip_fast_csum(iph, iph->ihl);
udph = (struct udphdr *)((char *)iph + sizeof(struct iphdr));
udph->source = sport;
udph->dest = dport;
udph->len = __constant_htons(sizeof(struct udphdr));
udph->check = 0;
ret = nf_conntrack_in_compat(&init_net, PF_INET, NF_INET_PRE_ROUTING, uskb);
if (ret != NF_ACCEPT) {
return NULL;
}
user = nf_ct_get(uskb, &ctinfo);
if (!user) {
NATCAP_ERROR("fakeuser create for ct[%pI4:%u->%pI4:%u] failed\n", &saddr, ntohs(sport), &daddr, ntohs(dport));
return NULL;
}
if (!user->ext) {
NATCAP_ERROR("fakeuser create for ct[%pI4:%u->%pI4:%u] failed, user->ext is NULL\n", &saddr, ntohs(sport), &daddr, ntohs(dport));
skb_nfct_reset(uskb);
return NULL;
}
if (nf_ct_is_confirmed(user)) {
skb_nfct_reset(uskb);
NATCAP_WARN("fakeuser create for ct[%pI4:%u->%pI4:%u] failed, user nf_ct_is_confirmed\n", &saddr, ntohs(sport), &daddr, ntohs(dport));
return NULL;
}
if (!(IPS_NATCAP_PEER & user->status) && !test_and_set_bit(IPS_NATCAP_PEER_BIT, &user->status)) {
newoff = ALIGN(user->ext->len, __ALIGN_64BITS);
#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 5, 0)
new = __krealloc(user->ext, newoff + sizeof(struct fakeuser_expect), GFP_ATOMIC);
#else
new = krealloc(user->ext, newoff + sizeof(struct fakeuser_expect), GFP_ATOMIC);
#endif
if (!new) {
NATCAP_ERROR("fakeuser create for ct[%pI4:%u->%pI4:%u] failed, realloc user->ext failed\n", &saddr, ntohs(sport), &daddr, ntohs(dport));
skb_nfct_reset(uskb);
return NULL;
}
if (user->ext != new) {
#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 9, 0)
kfree_rcu(user->ext, rcu);
user->ext = new;
#elif LINUX_VERSION_CODE < KERNEL_VERSION(5, 5, 0)
kfree_rcu(user->ext, rcu);
rcu_assign_pointer(user->ext, new);
#else
user->ext = new;
#endif
}
new->len = newoff;
memset((void *)new + newoff, 0, sizeof(struct fakeuser_expect));
fue = peer_fakeuser_expect(user);
fue->pmi = pmi;
fue->local_seq = ntohl(gen_seq_number());
fue->remote_seq = 0;
fue->last_active = jiffies;
}
ret = nf_conntrack_confirm(uskb);
if (ret != NF_ACCEPT) {
skb_nfct_reset(uskb);
return NULL;
}
user = nf_ct_get(uskb, &ctinfo);
nf_conntrack_get(&user->ct_general);
skb_nfct_reset(uskb);
natcap_user_timeout_touch(user, peer_conn_timeout);
fue = peer_fakeuser_expect(user);
NATCAP_DEBUG("fakeuser create user[%pI4:%u->%pI4:%u] pmi=%d upmi=%d\n", &saddr, ntohs(sport), &daddr, ntohs(dport), pmi, fue->pmi);
return user;
}
static struct nf_conn *peer_user_expect_in(int ttl, __be32 saddr, __be32 daddr, __be16 sport, __be16 dport, __be32 client_ip, const unsigned char *client_mac, struct peer_tuple **ppt)
{
int ret;
unsigned int i;
struct peer_tuple *pt = NULL;
struct user_expect *ue;
struct nf_conn *user;
struct nf_ct_ext *new;
enum ip_conntrack_info ctinfo;
unsigned int newoff;
struct sk_buff *uskb;
struct iphdr *iph;
struct udphdr *udph;
unsigned long last_jiffies = jiffies;
uskb = uskb_of_this_cpu();
if (uskb == NULL) {
return NULL;
}
skb_reset_transport_header(uskb);
skb_reset_network_header(uskb);
skb_reset_mac_len(uskb);
uskb->protocol = __constant_htons(ETH_P_IP);
skb_set_tail_pointer(uskb, PEER_USKB_SIZE);
uskb->len = PEER_USKB_SIZE;
uskb->pkt_type = PACKET_HOST;
uskb->transport_header = uskb->network_header + sizeof(struct iphdr);
iph = ip_hdr(uskb);
iph->version = 4;
iph->ihl = 5;
iph->saddr = get_byte4(client_mac);
iph->daddr = PEER_FAKEUSER_DADDR;
iph->tos = 0;
iph->tot_len = htons(PEER_USKB_SIZE);
iph->ttl=255;
iph->protocol = IPPROTO_UDP;
iph->id = __constant_htons(0xdead);
iph->frag_off = 0;
iph->check = 0;
iph->check = ip_fast_csum(iph, iph->ihl);
udph = (struct udphdr *)((char *)iph + sizeof(struct iphdr));
udph->source = get_byte2(client_mac + 4);
udph->dest = __constant_htons(65535);
udph->len = __constant_htons(sizeof(struct udphdr));
udph->check = 0;
ret = nf_conntrack_in_compat(&init_net, PF_INET, NF_INET_PRE_ROUTING, uskb);
if (ret != NF_ACCEPT) {
return NULL;
}
user = nf_ct_get(uskb, &ctinfo);
if (!user) {
NATCAP_ERROR("user [%02x:%02x:%02x:%02x:%02x:%02x] ct[%pI4:%u->%pI4:%u] failed\n",
client_mac[0], client_mac[1], client_mac[2], client_mac[3], client_mac[4], client_mac[5],
&saddr, ntohs(sport), &daddr, ntohs(dport));
return NULL;
}
if (!user->ext) {
NATCAP_ERROR("user [%02x:%02x:%02x:%02x:%02x:%02x] ct[%pI4:%u->%pI4:%u] failed, user->ext is NULL\n",
client_mac[0], client_mac[1], client_mac[2], client_mac[3], client_mac[4], client_mac[5],
&saddr, ntohs(sport), &daddr, ntohs(dport));
skb_nfct_reset(uskb);
return NULL;
}
if (!nf_ct_is_confirmed(user) && !(IPS_NATCAP_PEER & user->status) && !test_and_set_bit(IPS_NATCAP_PEER_BIT, &user->status)) {
newoff = ALIGN(user->ext->len, __ALIGN_64BITS);
#if LINUX_VERSION_CODE < KERNEL_VERSION(5, 5, 0)
new = __krealloc(user->ext, newoff + sizeof(struct user_expect), GFP_ATOMIC);
#else
new = krealloc(user->ext, newoff + sizeof(struct user_expect), GFP_ATOMIC);
#endif
if (!new) {
NATCAP_ERROR("user [%02x:%02x:%02x:%02x:%02x:%02x] ct[%pI4:%u->%pI4:%u] failed, realloc user->ext failed\n",
client_mac[0], client_mac[1], client_mac[2], client_mac[3], client_mac[4], client_mac[5],
&saddr, ntohs(sport), &daddr, ntohs(dport));
skb_nfct_reset(uskb);
return NULL;
}
if (user->ext != new) {
#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 9, 0)
kfree_rcu(user->ext, rcu);
user->ext = new;
#elif LINUX_VERSION_CODE < KERNEL_VERSION(5, 5, 0)
kfree_rcu(user->ext, rcu);
rcu_assign_pointer(user->ext, new);
#else
user->ext = new;
#endif
}
new->len = newoff;
memset((void *)new + newoff, 0, sizeof(struct user_expect));
//Repeated initialization cannot happen, it is safe.
ue = peer_user_expect(user);
spin_lock_init(&ue->lock);
spin_lock_bh(&ue->lock);
ue->ip = saddr;
ue->local_ip = client_ip;
ue->map_port = alloc_peer_port(user, client_mac);
spin_unlock_bh(&ue->lock);
user->mark = ntohl(saddr);
}
ret = nf_conntrack_confirm(uskb);
if (ret != NF_ACCEPT) {
skb_nfct_reset(uskb);
return NULL;
}
user = nf_ct_get(uskb, &ctinfo);
ue = peer_user_expect(user);
if (user != peer_port_map[ntohs(ue->map_port)]) {
//XXX this can only happen when alloc_peer_port get 0 or old user got timeout.
// so we re-alloc it
spin_lock_bh(&ue->lock);
//re-check-in-lock
if (user != peer_port_map[ntohs(ue->map_port)]) {
//re-alloc-map_port
ue->map_port = alloc_peer_port(user, client_mac);
}
spin_unlock_bh(&ue->lock);
if (user != peer_port_map[ntohs(ue->map_port)]) {
NATCAP_WARN("user [%02x:%02x:%02x:%02x:%02x:%02x] ct[%pI4:%u->%pI4:%u] alloc map_port fail\n",
client_mac[0], client_mac[1], client_mac[2], client_mac[3], client_mac[4], client_mac[5],
&saddr, ntohs(sport), &daddr, ntohs(dport));
/* alloc_peer_port fail: portmap would not work, but sni should work */
}
}
if (ntohl(saddr) != user->mark) {
__be32 old_ip = htonl(user->mark);
NATCAP_WARN("user [%02x:%02x:%02x:%02x:%02x:%02x] ct[%pI4:%u->%pI4:%u] change ip from %pI4(ttl=%u) to %pI4(ttl=%u) P=%u AS=%d\n",
client_mac[0], client_mac[1], client_mac[2], client_mac[3], client_mac[4], client_mac[5],
&saddr, ntohs(sport), &daddr, ntohs(dport),
&old_ip, (unsigned int)((user->status & 0xff000000) >> 24), &saddr, ttl, ntohs(ue->map_port),
ue->last_active != 0 ? (uintmindiff(ue->last_active, jiffies) + HZ / 2) / HZ : (-1));
user->mark = ntohl(saddr);
user->status &= ~(0xff << 24);
user->status |= ((ttl & 0xff) << 24);
}
natcap_user_timeout_touch(user, peer_port_map_timeout);
if (ue->ip != saddr) {
ue->ip = saddr;
short_clear_bit(PEER_SUBTYPE_PUB_BIT, &ue->status);
}
if (ue->local_ip != client_ip) {
ue->local_ip = client_ip;
short_clear_bit(PEER_SUBTYPE_PUB_BIT, &ue->status);
}
for (i = 0; i < MAX_PEER_TUPLE; i++) {
if (ue->tuple[i].sip == saddr && ue->tuple[i].dip == daddr && ue->tuple[i].sport == sport && ue->tuple[i].dport) {
spin_lock_bh(&ue->lock);
//re-check-in-lock
if (ue->tuple[i].sip == saddr && ue->tuple[i].dip == daddr && ue->tuple[i].sport == sport && ue->tuple[i].dport) {
pt = &ue->tuple[i];
spin_unlock_bh(&ue->lock);
break;
}
spin_unlock_bh(&ue->lock);
}
}
if (pt == NULL) {
unsigned long maxdiff = 0;
for (i = 0; i < MAX_PEER_TUPLE; i++) {
if (maxdiff < uintmindiff(last_jiffies, ue->tuple[i].last_active)) {
maxdiff = uintmindiff(last_jiffies, ue->tuple[i].last_active);
pt = &ue->tuple[i];
}
if (ue->tuple[i].sip == 0) {
pt = &ue->tuple[i];
break;
}
}
if (pt) {
spin_lock_bh(&ue->lock);
NATCAP_INFO("user [%02x:%02x:%02x:%02x:%02x:%02x] @map_port=%u use new-ct[%pI4:%u->%pI4:%u] replace old-ct[%pI4:%u->%pI4:%u] time=%u,%u\n",
client_mac[0], client_mac[1], client_mac[2], client_mac[3], client_mac[4], client_mac[5],
ntohs(ue->map_port), &saddr, ntohs(sport), &daddr, ntohs(dport),
&pt->sip, ntohs(pt->sport), &pt->dip, ntohs(pt->dport), pt->last_active, (unsigned int)last_jiffies);
pt->sip = saddr;
pt->dip = daddr;
pt->sport = sport;
pt->dport = dport;
pt->local_seq = 0;
pt->remote_seq = 0;
pt->connected = 0;
pt->mode = 0;
pt->last_active = 0;
spin_unlock_bh(&ue->lock);
}
}
if (ppt && pt) {
*ppt = pt;
}
nf_conntrack_get(&user->ct_general);
skb_nfct_reset(uskb);
return user;
}
static __be32 peer_upstream_auth_ip = 0;
static void natcap_auth_request_upstream(const unsigned char *client_mac, __be32 client_ip)
{
struct nf_conn *user;
struct sk_buff *skb;
struct iphdr *iph;
struct tcphdr *tcph;
struct natcap_TCPOPT *tcpopt;
struct peer_server_node *ps = NULL;
struct fakeuser_expect *fue;
u8 protocol = IPPROTO_TCP;
int opt_header_len = ALIGN(sizeof(struct natcap_TCPOPT_header) + sizeof(struct natcap_TCPOPT_peer), sizeof(unsigned int));