tipc: introduce jumbo frame support for broadcast
[firefly-linux-kernel-4.4.55.git] / net / tipc / link.c
1 /*
2  * net/tipc/link.c: TIPC link code
3  *
4  * Copyright (c) 1996-2007, 2012-2015, Ericsson AB
5  * Copyright (c) 2004-2007, 2010-2013, Wind River Systems
6  * All rights reserved.
7  *
8  * Redistribution and use in source and binary forms, with or without
9  * modification, are permitted provided that the following conditions are met:
10  *
11  * 1. Redistributions of source code must retain the above copyright
12  *    notice, this list of conditions and the following disclaimer.
13  * 2. Redistributions in binary form must reproduce the above copyright
14  *    notice, this list of conditions and the following disclaimer in the
15  *    documentation and/or other materials provided with the distribution.
16  * 3. Neither the names of the copyright holders nor the names of its
17  *    contributors may be used to endorse or promote products derived from
18  *    this software without specific prior written permission.
19  *
20  * Alternatively, this software may be distributed under the terms of the
21  * GNU General Public License ("GPL") version 2 as published by the Free
22  * Software Foundation.
23  *
24  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
25  * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
26  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
27  * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
28  * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
29  * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
30  * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
31  * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
32  * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
33  * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
34  * POSSIBILITY OF SUCH DAMAGE.
35  */
36
37 #include "core.h"
38 #include "subscr.h"
39 #include "link.h"
40 #include "bcast.h"
41 #include "socket.h"
42 #include "name_distr.h"
43 #include "discover.h"
44 #include "netlink.h"
45
46 #include <linux/pkt_sched.h>
47
48 /*
49  * Error message prefixes
50  */
51 static const char *link_co_err = "Link tunneling error, ";
52 static const char *link_rst_msg = "Resetting link ";
53 static const char tipc_bclink_name[] = "broadcast-link";
54
55 static const struct nla_policy tipc_nl_link_policy[TIPC_NLA_LINK_MAX + 1] = {
56         [TIPC_NLA_LINK_UNSPEC]          = { .type = NLA_UNSPEC },
57         [TIPC_NLA_LINK_NAME] = {
58                 .type = NLA_STRING,
59                 .len = TIPC_MAX_LINK_NAME
60         },
61         [TIPC_NLA_LINK_MTU]             = { .type = NLA_U32 },
62         [TIPC_NLA_LINK_BROADCAST]       = { .type = NLA_FLAG },
63         [TIPC_NLA_LINK_UP]              = { .type = NLA_FLAG },
64         [TIPC_NLA_LINK_ACTIVE]          = { .type = NLA_FLAG },
65         [TIPC_NLA_LINK_PROP]            = { .type = NLA_NESTED },
66         [TIPC_NLA_LINK_STATS]           = { .type = NLA_NESTED },
67         [TIPC_NLA_LINK_RX]              = { .type = NLA_U32 },
68         [TIPC_NLA_LINK_TX]              = { .type = NLA_U32 }
69 };
70
71 /* Properties valid for media, bearar and link */
72 static const struct nla_policy tipc_nl_prop_policy[TIPC_NLA_PROP_MAX + 1] = {
73         [TIPC_NLA_PROP_UNSPEC]          = { .type = NLA_UNSPEC },
74         [TIPC_NLA_PROP_PRIO]            = { .type = NLA_U32 },
75         [TIPC_NLA_PROP_TOL]             = { .type = NLA_U32 },
76         [TIPC_NLA_PROP_WIN]             = { .type = NLA_U32 }
77 };
78
79 /* Send states for broadcast NACKs
80  */
81 enum {
82         BC_NACK_SND_CONDITIONAL,
83         BC_NACK_SND_UNCONDITIONAL,
84         BC_NACK_SND_SUPPRESS,
85 };
86
87 /*
88  * Interval between NACKs when packets arrive out of order
89  */
90 #define TIPC_NACK_INTV (TIPC_MIN_LINK_WIN * 2)
91 /*
92  * Out-of-range value for link session numbers
93  */
94 #define WILDCARD_SESSION 0x10000
95
96 /* Link FSM states:
97  */
98 enum {
99         LINK_ESTABLISHED     = 0xe,
100         LINK_ESTABLISHING    = 0xe  << 4,
101         LINK_RESET           = 0x1  << 8,
102         LINK_RESETTING       = 0x2  << 12,
103         LINK_PEER_RESET      = 0xd  << 16,
104         LINK_FAILINGOVER     = 0xf  << 20,
105         LINK_SYNCHING        = 0xc  << 24
106 };
107
108 /* Link FSM state checking routines
109  */
110 static int link_is_up(struct tipc_link *l)
111 {
112         return l->state & (LINK_ESTABLISHED | LINK_SYNCHING);
113 }
114
115 static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
116                                struct sk_buff_head *xmitq);
117 static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe,
118                                       u16 rcvgap, int tolerance, int priority,
119                                       struct sk_buff_head *xmitq);
120 static void link_reset_statistics(struct tipc_link *l_ptr);
121 static void link_print(struct tipc_link *l_ptr, const char *str);
122 static void tipc_link_build_nack_msg(struct tipc_link *l,
123                                      struct sk_buff_head *xmitq);
124 static void tipc_link_build_bc_init_msg(struct tipc_link *l,
125                                         struct sk_buff_head *xmitq);
126 static bool tipc_link_release_pkts(struct tipc_link *l, u16 to);
127
128 /*
129  *  Simple non-static link routines (i.e. referenced outside this file)
130  */
131 bool tipc_link_is_up(struct tipc_link *l)
132 {
133         return link_is_up(l);
134 }
135
136 bool tipc_link_peer_is_down(struct tipc_link *l)
137 {
138         return l->state == LINK_PEER_RESET;
139 }
140
141 bool tipc_link_is_reset(struct tipc_link *l)
142 {
143         return l->state & (LINK_RESET | LINK_FAILINGOVER | LINK_ESTABLISHING);
144 }
145
146 bool tipc_link_is_establishing(struct tipc_link *l)
147 {
148         return l->state == LINK_ESTABLISHING;
149 }
150
151 bool tipc_link_is_synching(struct tipc_link *l)
152 {
153         return l->state == LINK_SYNCHING;
154 }
155
156 bool tipc_link_is_failingover(struct tipc_link *l)
157 {
158         return l->state == LINK_FAILINGOVER;
159 }
160
161 bool tipc_link_is_blocked(struct tipc_link *l)
162 {
163         return l->state & (LINK_RESETTING | LINK_PEER_RESET | LINK_FAILINGOVER);
164 }
165
166 bool link_is_bc_sndlink(struct tipc_link *l)
167 {
168         return !l->bc_sndlink;
169 }
170
171 bool link_is_bc_rcvlink(struct tipc_link *l)
172 {
173         return ((l->bc_rcvlink == l) && !link_is_bc_sndlink(l));
174 }
175
176 int tipc_link_is_active(struct tipc_link *l)
177 {
178         struct tipc_node *n = l->owner;
179
180         return (node_active_link(n, 0) == l) || (node_active_link(n, 1) == l);
181 }
182
183 void tipc_link_add_bc_peer(struct tipc_link *snd_l,
184                            struct tipc_link *uc_l,
185                            struct sk_buff_head *xmitq)
186 {
187         struct tipc_link *rcv_l = uc_l->bc_rcvlink;
188
189         snd_l->ackers++;
190         rcv_l->acked = snd_l->snd_nxt - 1;
191         tipc_link_build_bc_init_msg(uc_l, xmitq);
192 }
193
194 void tipc_link_remove_bc_peer(struct tipc_link *snd_l,
195                               struct tipc_link *rcv_l,
196                               struct sk_buff_head *xmitq)
197 {
198         u16 ack = snd_l->snd_nxt - 1;
199
200         snd_l->ackers--;
201         tipc_link_bc_ack_rcv(rcv_l, ack, xmitq);
202         tipc_link_reset(rcv_l);
203         rcv_l->state = LINK_RESET;
204         if (!snd_l->ackers) {
205                 tipc_link_reset(snd_l);
206                 __skb_queue_purge(xmitq);
207         }
208 }
209
210 int tipc_link_bc_peers(struct tipc_link *l)
211 {
212         return l->ackers;
213 }
214
215 void tipc_link_set_mtu(struct tipc_link *l, int mtu)
216 {
217         l->mtu = mtu;
218 }
219
220 int tipc_link_mtu(struct tipc_link *l)
221 {
222         return l->mtu;
223 }
224
225 static u32 link_own_addr(struct tipc_link *l)
226 {
227         return msg_prevnode(l->pmsg);
228 }
229
230 /**
231  * tipc_link_create - create a new link
232  * @n: pointer to associated node
233  * @if_name: associated interface name
234  * @bearer_id: id (index) of associated bearer
235  * @tolerance: link tolerance to be used by link
236  * @net_plane: network plane (A,B,c..) this link belongs to
237  * @mtu: mtu to be advertised by link
238  * @priority: priority to be used by link
239  * @window: send window to be used by link
240  * @session: session to be used by link
241  * @ownnode: identity of own node
242  * @peer: node id of peer node
243  * @peer_caps: bitmap describing peer node capabilities
244  * @maddr: media address to be used
245  * @bc_sndlink: the namespace global link used for broadcast sending
246  * @bc_rcvlink: the peer specific link used for broadcast reception
247  * @inputq: queue to put messages ready for delivery
248  * @namedq: queue to put binding table update messages ready for delivery
249  * @link: return value, pointer to put the created link
250  *
251  * Returns true if link was created, otherwise false
252  */
253 bool tipc_link_create(struct tipc_node *n, char *if_name, int bearer_id,
254                       int tolerance, char net_plane, u32 mtu, int priority,
255                       int window, u32 session, u32 ownnode, u32 peer,
256                       u16 peer_caps,
257                       struct tipc_media_addr *maddr,
258                       struct tipc_link *bc_sndlink,
259                       struct tipc_link *bc_rcvlink,
260                       struct sk_buff_head *inputq,
261                       struct sk_buff_head *namedq,
262                       struct tipc_link **link)
263 {
264         struct tipc_link *l;
265         struct tipc_msg *hdr;
266
267         l = kzalloc(sizeof(*l), GFP_ATOMIC);
268         if (!l)
269                 return false;
270         *link = l;
271         l->pmsg = (struct tipc_msg *)&l->proto_msg;
272         hdr = l->pmsg;
273         tipc_msg_init(ownnode, hdr, LINK_PROTOCOL, RESET_MSG, INT_H_SIZE, peer);
274         msg_set_size(hdr, sizeof(l->proto_msg));
275         msg_set_session(hdr, session);
276         msg_set_bearer_id(hdr, l->bearer_id);
277
278         /* Note: peer i/f name is completed by reset/activate message */
279         sprintf(l->name, "%u.%u.%u:%s-%u.%u.%u:unknown",
280                 tipc_zone(ownnode), tipc_cluster(ownnode), tipc_node(ownnode),
281                 if_name, tipc_zone(peer), tipc_cluster(peer), tipc_node(peer));
282         strcpy((char *)msg_data(hdr), if_name);
283
284         l->addr = peer;
285         l->peer_caps = peer_caps;
286         l->media_addr = maddr;
287         l->owner = n;
288         l->peer_session = WILDCARD_SESSION;
289         l->bearer_id = bearer_id;
290         l->tolerance = tolerance;
291         l->net_plane = net_plane;
292         l->advertised_mtu = mtu;
293         l->mtu = mtu;
294         l->priority = priority;
295         tipc_link_set_queue_limits(l, window);
296         l->ackers = 1;
297         l->bc_sndlink = bc_sndlink;
298         l->bc_rcvlink = bc_rcvlink;
299         l->inputq = inputq;
300         l->namedq = namedq;
301         l->state = LINK_RESETTING;
302         __skb_queue_head_init(&l->transmq);
303         __skb_queue_head_init(&l->backlogq);
304         __skb_queue_head_init(&l->deferdq);
305         skb_queue_head_init(&l->wakeupq);
306         skb_queue_head_init(l->inputq);
307         return true;
308 }
309
310 /**
311  * tipc_link_bc_create - create new link to be used for broadcast
312  * @n: pointer to associated node
313  * @mtu: mtu to be used
314  * @window: send window to be used
315  * @inputq: queue to put messages ready for delivery
316  * @namedq: queue to put binding table update messages ready for delivery
317  * @link: return value, pointer to put the created link
318  *
319  * Returns true if link was created, otherwise false
320  */
321 bool tipc_link_bc_create(struct tipc_node *n, u32 ownnode, u32 peer,
322                          int mtu, int window, u16 peer_caps,
323                          struct sk_buff_head *inputq,
324                          struct sk_buff_head *namedq,
325                          struct tipc_link *bc_sndlink,
326                          struct tipc_link **link)
327 {
328         struct tipc_link *l;
329
330         if (!tipc_link_create(n, "", MAX_BEARERS, 0, 'Z', mtu, 0, window,
331                               0, ownnode, peer, peer_caps, NULL, bc_sndlink,
332                               NULL, inputq, namedq, link))
333                 return false;
334
335         l = *link;
336         strcpy(l->name, tipc_bclink_name);
337         tipc_link_reset(l);
338         l->state = LINK_RESET;
339         l->ackers = 0;
340         l->bc_rcvlink = l;
341
342         /* Broadcast send link is always up */
343         if (link_is_bc_sndlink(l))
344                 l->state = LINK_ESTABLISHED;
345
346         return true;
347 }
348
349 /**
350  * tipc_link_fsm_evt - link finite state machine
351  * @l: pointer to link
352  * @evt: state machine event to be processed
353  */
354 int tipc_link_fsm_evt(struct tipc_link *l, int evt)
355 {
356         int rc = 0;
357
358         switch (l->state) {
359         case LINK_RESETTING:
360                 switch (evt) {
361                 case LINK_PEER_RESET_EVT:
362                         l->state = LINK_PEER_RESET;
363                         break;
364                 case LINK_RESET_EVT:
365                         l->state = LINK_RESET;
366                         break;
367                 case LINK_FAILURE_EVT:
368                 case LINK_FAILOVER_BEGIN_EVT:
369                 case LINK_ESTABLISH_EVT:
370                 case LINK_FAILOVER_END_EVT:
371                 case LINK_SYNCH_BEGIN_EVT:
372                 case LINK_SYNCH_END_EVT:
373                 default:
374                         goto illegal_evt;
375                 }
376                 break;
377         case LINK_RESET:
378                 switch (evt) {
379                 case LINK_PEER_RESET_EVT:
380                         l->state = LINK_ESTABLISHING;
381                         break;
382                 case LINK_FAILOVER_BEGIN_EVT:
383                         l->state = LINK_FAILINGOVER;
384                 case LINK_FAILURE_EVT:
385                 case LINK_RESET_EVT:
386                 case LINK_ESTABLISH_EVT:
387                 case LINK_FAILOVER_END_EVT:
388                         break;
389                 case LINK_SYNCH_BEGIN_EVT:
390                 case LINK_SYNCH_END_EVT:
391                 default:
392                         goto illegal_evt;
393                 }
394                 break;
395         case LINK_PEER_RESET:
396                 switch (evt) {
397                 case LINK_RESET_EVT:
398                         l->state = LINK_ESTABLISHING;
399                         break;
400                 case LINK_PEER_RESET_EVT:
401                 case LINK_ESTABLISH_EVT:
402                 case LINK_FAILURE_EVT:
403                         break;
404                 case LINK_SYNCH_BEGIN_EVT:
405                 case LINK_SYNCH_END_EVT:
406                 case LINK_FAILOVER_BEGIN_EVT:
407                 case LINK_FAILOVER_END_EVT:
408                 default:
409                         goto illegal_evt;
410                 }
411                 break;
412         case LINK_FAILINGOVER:
413                 switch (evt) {
414                 case LINK_FAILOVER_END_EVT:
415                         l->state = LINK_RESET;
416                         break;
417                 case LINK_PEER_RESET_EVT:
418                 case LINK_RESET_EVT:
419                 case LINK_ESTABLISH_EVT:
420                 case LINK_FAILURE_EVT:
421                         break;
422                 case LINK_FAILOVER_BEGIN_EVT:
423                 case LINK_SYNCH_BEGIN_EVT:
424                 case LINK_SYNCH_END_EVT:
425                 default:
426                         goto illegal_evt;
427                 }
428                 break;
429         case LINK_ESTABLISHING:
430                 switch (evt) {
431                 case LINK_ESTABLISH_EVT:
432                         l->state = LINK_ESTABLISHED;
433                         break;
434                 case LINK_FAILOVER_BEGIN_EVT:
435                         l->state = LINK_FAILINGOVER;
436                         break;
437                 case LINK_RESET_EVT:
438                         l->state = LINK_RESET;
439                         break;
440                 case LINK_FAILURE_EVT:
441                 case LINK_PEER_RESET_EVT:
442                 case LINK_SYNCH_BEGIN_EVT:
443                 case LINK_FAILOVER_END_EVT:
444                         break;
445                 case LINK_SYNCH_END_EVT:
446                 default:
447                         goto illegal_evt;
448                 }
449                 break;
450         case LINK_ESTABLISHED:
451                 switch (evt) {
452                 case LINK_PEER_RESET_EVT:
453                         l->state = LINK_PEER_RESET;
454                         rc |= TIPC_LINK_DOWN_EVT;
455                         break;
456                 case LINK_FAILURE_EVT:
457                         l->state = LINK_RESETTING;
458                         rc |= TIPC_LINK_DOWN_EVT;
459                         break;
460                 case LINK_RESET_EVT:
461                         l->state = LINK_RESET;
462                         break;
463                 case LINK_ESTABLISH_EVT:
464                 case LINK_SYNCH_END_EVT:
465                         break;
466                 case LINK_SYNCH_BEGIN_EVT:
467                         l->state = LINK_SYNCHING;
468                         break;
469                 case LINK_FAILOVER_BEGIN_EVT:
470                 case LINK_FAILOVER_END_EVT:
471                 default:
472                         goto illegal_evt;
473                 }
474                 break;
475         case LINK_SYNCHING:
476                 switch (evt) {
477                 case LINK_PEER_RESET_EVT:
478                         l->state = LINK_PEER_RESET;
479                         rc |= TIPC_LINK_DOWN_EVT;
480                         break;
481                 case LINK_FAILURE_EVT:
482                         l->state = LINK_RESETTING;
483                         rc |= TIPC_LINK_DOWN_EVT;
484                         break;
485                 case LINK_RESET_EVT:
486                         l->state = LINK_RESET;
487                         break;
488                 case LINK_ESTABLISH_EVT:
489                 case LINK_SYNCH_BEGIN_EVT:
490                         break;
491                 case LINK_SYNCH_END_EVT:
492                         l->state = LINK_ESTABLISHED;
493                         break;
494                 case LINK_FAILOVER_BEGIN_EVT:
495                 case LINK_FAILOVER_END_EVT:
496                 default:
497                         goto illegal_evt;
498                 }
499                 break;
500         default:
501                 pr_err("Unknown FSM state %x in %s\n", l->state, l->name);
502         }
503         return rc;
504 illegal_evt:
505         pr_err("Illegal FSM event %x in state %x on link %s\n",
506                evt, l->state, l->name);
507         return rc;
508 }
509
510 /* link_profile_stats - update statistical profiling of traffic
511  */
512 static void link_profile_stats(struct tipc_link *l)
513 {
514         struct sk_buff *skb;
515         struct tipc_msg *msg;
516         int length;
517
518         /* Update counters used in statistical profiling of send traffic */
519         l->stats.accu_queue_sz += skb_queue_len(&l->transmq);
520         l->stats.queue_sz_counts++;
521
522         skb = skb_peek(&l->transmq);
523         if (!skb)
524                 return;
525         msg = buf_msg(skb);
526         length = msg_size(msg);
527
528         if (msg_user(msg) == MSG_FRAGMENTER) {
529                 if (msg_type(msg) != FIRST_FRAGMENT)
530                         return;
531                 length = msg_size(msg_get_wrapped(msg));
532         }
533         l->stats.msg_lengths_total += length;
534         l->stats.msg_length_counts++;
535         if (length <= 64)
536                 l->stats.msg_length_profile[0]++;
537         else if (length <= 256)
538                 l->stats.msg_length_profile[1]++;
539         else if (length <= 1024)
540                 l->stats.msg_length_profile[2]++;
541         else if (length <= 4096)
542                 l->stats.msg_length_profile[3]++;
543         else if (length <= 16384)
544                 l->stats.msg_length_profile[4]++;
545         else if (length <= 32768)
546                 l->stats.msg_length_profile[5]++;
547         else
548                 l->stats.msg_length_profile[6]++;
549 }
550
551 /* tipc_link_timeout - perform periodic task as instructed from node timeout
552  */
553 /* tipc_link_timeout - perform periodic task as instructed from node timeout
554  */
555 int tipc_link_timeout(struct tipc_link *l, struct sk_buff_head *xmitq)
556 {
557         int rc = 0;
558         int mtyp = STATE_MSG;
559         bool xmit = false;
560         bool prb = false;
561         u16 bc_snt = l->bc_sndlink->snd_nxt - 1;
562         u16 bc_acked = l->bc_rcvlink->acked;
563         bool bc_up = link_is_up(l->bc_rcvlink);
564
565         link_profile_stats(l);
566
567         switch (l->state) {
568         case LINK_ESTABLISHED:
569         case LINK_SYNCHING:
570                 if (!l->silent_intv_cnt) {
571                         if (bc_up && (bc_acked != bc_snt))
572                                 xmit = true;
573                 } else if (l->silent_intv_cnt <= l->abort_limit) {
574                         xmit = true;
575                         prb = true;
576                 } else {
577                         rc |= tipc_link_fsm_evt(l, LINK_FAILURE_EVT);
578                 }
579                 l->silent_intv_cnt++;
580                 break;
581         case LINK_RESET:
582                 xmit = true;
583                 mtyp = RESET_MSG;
584                 break;
585         case LINK_ESTABLISHING:
586                 xmit = true;
587                 mtyp = ACTIVATE_MSG;
588                 break;
589         case LINK_PEER_RESET:
590         case LINK_RESETTING:
591         case LINK_FAILINGOVER:
592                 break;
593         default:
594                 break;
595         }
596
597         if (xmit)
598                 tipc_link_build_proto_msg(l, mtyp, prb, 0, 0, 0, xmitq);
599
600         return rc;
601 }
602
603 /**
604  * link_schedule_user - schedule a message sender for wakeup after congestion
605  * @link: congested link
606  * @list: message that was attempted sent
607  * Create pseudo msg to send back to user when congestion abates
608  * Does not consume buffer list
609  */
610 static int link_schedule_user(struct tipc_link *link, struct sk_buff_head *list)
611 {
612         struct tipc_msg *msg = buf_msg(skb_peek(list));
613         int imp = msg_importance(msg);
614         u32 oport = msg_origport(msg);
615         u32 addr = link_own_addr(link);
616         struct sk_buff *skb;
617
618         /* This really cannot happen...  */
619         if (unlikely(imp > TIPC_CRITICAL_IMPORTANCE)) {
620                 pr_warn("%s<%s>, send queue full", link_rst_msg, link->name);
621                 return -ENOBUFS;
622         }
623         /* Non-blocking sender: */
624         if (TIPC_SKB_CB(skb_peek(list))->wakeup_pending)
625                 return -ELINKCONG;
626
627         /* Create and schedule wakeup pseudo message */
628         skb = tipc_msg_create(SOCK_WAKEUP, 0, INT_H_SIZE, 0,
629                               addr, addr, oport, 0, 0);
630         if (!skb)
631                 return -ENOBUFS;
632         TIPC_SKB_CB(skb)->chain_sz = skb_queue_len(list);
633         TIPC_SKB_CB(skb)->chain_imp = imp;
634         skb_queue_tail(&link->wakeupq, skb);
635         link->stats.link_congs++;
636         return -ELINKCONG;
637 }
638
639 /**
640  * link_prepare_wakeup - prepare users for wakeup after congestion
641  * @link: congested link
642  * Move a number of waiting users, as permitted by available space in
643  * the send queue, from link wait queue to node wait queue for wakeup
644  */
645 void link_prepare_wakeup(struct tipc_link *l)
646 {
647         int pnd[TIPC_SYSTEM_IMPORTANCE + 1] = {0,};
648         int imp, lim;
649         struct sk_buff *skb, *tmp;
650
651         skb_queue_walk_safe(&l->wakeupq, skb, tmp) {
652                 imp = TIPC_SKB_CB(skb)->chain_imp;
653                 lim = l->window + l->backlog[imp].limit;
654                 pnd[imp] += TIPC_SKB_CB(skb)->chain_sz;
655                 if ((pnd[imp] + l->backlog[imp].len) >= lim)
656                         break;
657                 skb_unlink(skb, &l->wakeupq);
658                 skb_queue_tail(l->inputq, skb);
659         }
660 }
661
662 /**
663  * tipc_link_reset_fragments - purge link's inbound message fragments queue
664  * @l_ptr: pointer to link
665  */
666 void tipc_link_reset_fragments(struct tipc_link *l_ptr)
667 {
668         kfree_skb(l_ptr->reasm_buf);
669         l_ptr->reasm_buf = NULL;
670 }
671
672 void tipc_link_purge_backlog(struct tipc_link *l)
673 {
674         __skb_queue_purge(&l->backlogq);
675         l->backlog[TIPC_LOW_IMPORTANCE].len = 0;
676         l->backlog[TIPC_MEDIUM_IMPORTANCE].len = 0;
677         l->backlog[TIPC_HIGH_IMPORTANCE].len = 0;
678         l->backlog[TIPC_CRITICAL_IMPORTANCE].len = 0;
679         l->backlog[TIPC_SYSTEM_IMPORTANCE].len = 0;
680 }
681
682 /**
683  * tipc_link_purge_queues - purge all pkt queues associated with link
684  * @l_ptr: pointer to link
685  */
686 void tipc_link_purge_queues(struct tipc_link *l_ptr)
687 {
688         __skb_queue_purge(&l_ptr->deferdq);
689         __skb_queue_purge(&l_ptr->transmq);
690         tipc_link_purge_backlog(l_ptr);
691         tipc_link_reset_fragments(l_ptr);
692 }
693
694 void tipc_link_reset(struct tipc_link *l)
695 {
696         /* Link is down, accept any session */
697         l->peer_session = WILDCARD_SESSION;
698
699         /* If peer is up, it only accepts an incremented session number */
700         msg_set_session(l->pmsg, msg_session(l->pmsg) + 1);
701
702         /* Prepare for renewed mtu size negotiation */
703         l->mtu = l->advertised_mtu;
704
705         /* Clean up all queues: */
706         __skb_queue_purge(&l->transmq);
707         __skb_queue_purge(&l->deferdq);
708         skb_queue_splice_init(&l->wakeupq, l->inputq);
709
710         tipc_link_purge_backlog(l);
711         kfree_skb(l->reasm_buf);
712         kfree_skb(l->failover_reasm_skb);
713         l->reasm_buf = NULL;
714         l->failover_reasm_skb = NULL;
715         l->rcv_unacked = 0;
716         l->snd_nxt = 1;
717         l->rcv_nxt = 1;
718         l->acked = 0;
719         l->silent_intv_cnt = 0;
720         l->stats.recv_info = 0;
721         l->stale_count = 0;
722         l->bc_peer_is_up = false;
723         link_reset_statistics(l);
724 }
725
726 /**
727  * __tipc_link_xmit(): same as tipc_link_xmit, but destlink is known & locked
728  * @link: link to use
729  * @list: chain of buffers containing message
730  *
731  * Consumes the buffer chain, except when returning an error code,
732  * Returns 0 if success, or errno: -ELINKCONG, -EMSGSIZE or -ENOBUFS
733  * Messages at TIPC_SYSTEM_IMPORTANCE are always accepted
734  */
735 int __tipc_link_xmit(struct net *net, struct tipc_link *link,
736                      struct sk_buff_head *list)
737 {
738         struct tipc_msg *msg = buf_msg(skb_peek(list));
739         unsigned int maxwin = link->window;
740         unsigned int i, imp = msg_importance(msg);
741         uint mtu = link->mtu;
742         u16 ack = mod(link->rcv_nxt - 1);
743         u16 seqno = link->snd_nxt;
744         u16 bc_ack = link->bc_rcvlink->rcv_nxt - 1;
745         struct tipc_media_addr *addr = link->media_addr;
746         struct sk_buff_head *transmq = &link->transmq;
747         struct sk_buff_head *backlogq = &link->backlogq;
748         struct sk_buff *skb, *bskb;
749
750         /* Match msg importance against this and all higher backlog limits: */
751         for (i = imp; i <= TIPC_SYSTEM_IMPORTANCE; i++) {
752                 if (unlikely(link->backlog[i].len >= link->backlog[i].limit))
753                         return link_schedule_user(link, list);
754         }
755         if (unlikely(msg_size(msg) > mtu))
756                 return -EMSGSIZE;
757
758         /* Prepare each packet for sending, and add to relevant queue: */
759         while (skb_queue_len(list)) {
760                 skb = skb_peek(list);
761                 msg = buf_msg(skb);
762                 msg_set_seqno(msg, seqno);
763                 msg_set_ack(msg, ack);
764                 msg_set_bcast_ack(msg, bc_ack);
765
766                 if (likely(skb_queue_len(transmq) < maxwin)) {
767                         __skb_dequeue(list);
768                         __skb_queue_tail(transmq, skb);
769                         tipc_bearer_send(net, link->bearer_id, skb, addr);
770                         link->rcv_unacked = 0;
771                         seqno++;
772                         continue;
773                 }
774                 if (tipc_msg_bundle(skb_peek_tail(backlogq), msg, mtu)) {
775                         kfree_skb(__skb_dequeue(list));
776                         link->stats.sent_bundled++;
777                         continue;
778                 }
779                 if (tipc_msg_make_bundle(&bskb, msg, mtu, link->addr)) {
780                         kfree_skb(__skb_dequeue(list));
781                         __skb_queue_tail(backlogq, bskb);
782                         link->backlog[msg_importance(buf_msg(bskb))].len++;
783                         link->stats.sent_bundled++;
784                         link->stats.sent_bundles++;
785                         continue;
786                 }
787                 link->backlog[imp].len += skb_queue_len(list);
788                 skb_queue_splice_tail_init(list, backlogq);
789         }
790         link->snd_nxt = seqno;
791         return 0;
792 }
793
794 /**
795  * tipc_link_xmit(): enqueue buffer list according to queue situation
796  * @link: link to use
797  * @list: chain of buffers containing message
798  * @xmitq: returned list of packets to be sent by caller
799  *
800  * Consumes the buffer chain, except when returning -ELINKCONG,
801  * since the caller then may want to make more send attempts.
802  * Returns 0 if success, or errno: -ELINKCONG, -EMSGSIZE or -ENOBUFS
803  * Messages at TIPC_SYSTEM_IMPORTANCE are always accepted
804  */
805 int tipc_link_xmit(struct tipc_link *l, struct sk_buff_head *list,
806                    struct sk_buff_head *xmitq)
807 {
808         struct tipc_msg *hdr = buf_msg(skb_peek(list));
809         unsigned int maxwin = l->window;
810         unsigned int i, imp = msg_importance(hdr);
811         unsigned int mtu = l->mtu;
812         u16 ack = l->rcv_nxt - 1;
813         u16 seqno = l->snd_nxt;
814         u16 bc_ack = l->bc_rcvlink->rcv_nxt - 1;
815         struct sk_buff_head *transmq = &l->transmq;
816         struct sk_buff_head *backlogq = &l->backlogq;
817         struct sk_buff *skb, *_skb, *bskb;
818
819         /* Match msg importance against this and all higher backlog limits: */
820         for (i = imp; i <= TIPC_SYSTEM_IMPORTANCE; i++) {
821                 if (unlikely(l->backlog[i].len >= l->backlog[i].limit))
822                         return link_schedule_user(l, list);
823         }
824         if (unlikely(msg_size(hdr) > mtu))
825                 return -EMSGSIZE;
826
827         /* Prepare each packet for sending, and add to relevant queue: */
828         while (skb_queue_len(list)) {
829                 skb = skb_peek(list);
830                 hdr = buf_msg(skb);
831                 msg_set_seqno(hdr, seqno);
832                 msg_set_ack(hdr, ack);
833                 msg_set_bcast_ack(hdr, bc_ack);
834
835                 if (likely(skb_queue_len(transmq) < maxwin)) {
836                         _skb = skb_clone(skb, GFP_ATOMIC);
837                         if (!_skb)
838                                 return -ENOBUFS;
839                         __skb_dequeue(list);
840                         __skb_queue_tail(transmq, skb);
841                         __skb_queue_tail(xmitq, _skb);
842                         TIPC_SKB_CB(skb)->ackers = l->ackers;
843                         l->rcv_unacked = 0;
844                         seqno++;
845                         continue;
846                 }
847                 if (tipc_msg_bundle(skb_peek_tail(backlogq), hdr, mtu)) {
848                         kfree_skb(__skb_dequeue(list));
849                         l->stats.sent_bundled++;
850                         continue;
851                 }
852                 if (tipc_msg_make_bundle(&bskb, hdr, mtu, l->addr)) {
853                         kfree_skb(__skb_dequeue(list));
854                         __skb_queue_tail(backlogq, bskb);
855                         l->backlog[msg_importance(buf_msg(bskb))].len++;
856                         l->stats.sent_bundled++;
857                         l->stats.sent_bundles++;
858                         continue;
859                 }
860                 l->backlog[imp].len += skb_queue_len(list);
861                 skb_queue_splice_tail_init(list, backlogq);
862         }
863         l->snd_nxt = seqno;
864         return 0;
865 }
866
867 /*
868  * tipc_link_push_packets - push unsent packets to bearer
869  *
870  * Push out the unsent messages of a link where congestion
871  * has abated. Node is locked.
872  *
873  * Called with node locked
874  */
875 void tipc_link_push_packets(struct tipc_link *link)
876 {
877         struct sk_buff *skb;
878         struct tipc_msg *msg;
879         u16 seqno = link->snd_nxt;
880         u16 ack = mod(link->rcv_nxt - 1);
881
882         while (skb_queue_len(&link->transmq) < link->window) {
883                 skb = __skb_dequeue(&link->backlogq);
884                 if (!skb)
885                         break;
886                 TIPC_SKB_CB(skb)->ackers = link->ackers;
887                 msg = buf_msg(skb);
888                 link->backlog[msg_importance(msg)].len--;
889                 msg_set_ack(msg, ack);
890                 msg_set_seqno(msg, seqno);
891                 seqno = mod(seqno + 1);
892                 msg_set_bcast_ack(msg, link->owner->bclink.last_in);
893                 link->rcv_unacked = 0;
894                 __skb_queue_tail(&link->transmq, skb);
895                 tipc_bearer_send(link->owner->net, link->bearer_id,
896                                  skb, link->media_addr);
897         }
898         link->snd_nxt = seqno;
899 }
900
901 void tipc_link_advance_backlog(struct tipc_link *l, struct sk_buff_head *xmitq)
902 {
903         struct sk_buff *skb, *_skb;
904         struct tipc_msg *hdr;
905         u16 seqno = l->snd_nxt;
906         u16 ack = l->rcv_nxt - 1;
907         u16 bc_ack = l->bc_rcvlink->rcv_nxt - 1;
908
909         while (skb_queue_len(&l->transmq) < l->window) {
910                 skb = skb_peek(&l->backlogq);
911                 if (!skb)
912                         break;
913                 _skb = skb_clone(skb, GFP_ATOMIC);
914                 if (!_skb)
915                         break;
916                 __skb_dequeue(&l->backlogq);
917                 hdr = buf_msg(skb);
918                 l->backlog[msg_importance(hdr)].len--;
919                 __skb_queue_tail(&l->transmq, skb);
920                 __skb_queue_tail(xmitq, _skb);
921                 TIPC_SKB_CB(skb)->ackers = l->ackers;
922                 msg_set_seqno(hdr, seqno);
923                 msg_set_ack(hdr, ack);
924                 msg_set_bcast_ack(hdr, bc_ack);
925                 l->rcv_unacked = 0;
926                 seqno++;
927         }
928         l->snd_nxt = seqno;
929 }
930
931 static void link_retransmit_failure(struct tipc_link *l, struct sk_buff *skb)
932 {
933         struct tipc_msg *hdr = buf_msg(skb);
934
935         pr_warn("Retransmission failure on link <%s>\n", l->name);
936         link_print(l, "Resetting link ");
937         pr_info("Failed msg: usr %u, typ %u, len %u, err %u\n",
938                 msg_user(hdr), msg_type(hdr), msg_size(hdr), msg_errcode(hdr));
939         pr_info("sqno %u, prev: %x, src: %x\n",
940                 msg_seqno(hdr), msg_prevnode(hdr), msg_orignode(hdr));
941 }
942
943 void tipc_link_retransmit(struct tipc_link *l_ptr, struct sk_buff *skb,
944                           u32 retransmits)
945 {
946         struct tipc_msg *msg;
947
948         if (!skb)
949                 return;
950
951         msg = buf_msg(skb);
952
953         /* Detect repeated retransmit failures */
954         if (l_ptr->last_retransm == msg_seqno(msg)) {
955                 if (++l_ptr->stale_count > 100) {
956                         link_retransmit_failure(l_ptr, skb);
957                         return;
958                 }
959         } else {
960                 l_ptr->last_retransm = msg_seqno(msg);
961                 l_ptr->stale_count = 1;
962         }
963
964         skb_queue_walk_from(&l_ptr->transmq, skb) {
965                 if (!retransmits)
966                         break;
967                 msg = buf_msg(skb);
968                 msg_set_ack(msg, mod(l_ptr->rcv_nxt - 1));
969                 msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in);
970                 tipc_bearer_send(l_ptr->owner->net, l_ptr->bearer_id, skb,
971                                  l_ptr->media_addr);
972                 retransmits--;
973                 l_ptr->stats.retransmitted++;
974         }
975 }
976
977 int tipc_link_retrans(struct tipc_link *l, u16 from, u16 to,
978                       struct sk_buff_head *xmitq)
979 {
980         struct sk_buff *_skb, *skb = skb_peek(&l->transmq);
981         struct tipc_msg *hdr;
982         u16 ack = l->rcv_nxt - 1;
983         u16 bc_ack = l->bc_rcvlink->rcv_nxt - 1;
984
985         if (!skb)
986                 return 0;
987
988         /* Detect repeated retransmit failures on same packet */
989         if (likely(l->last_retransm != buf_seqno(skb))) {
990                 l->last_retransm = buf_seqno(skb);
991                 l->stale_count = 1;
992         } else if (++l->stale_count > 100) {
993                 link_retransmit_failure(l, skb);
994                 return tipc_link_fsm_evt(l, LINK_FAILURE_EVT);
995         }
996
997         /* Move forward to where retransmission should start */
998         skb_queue_walk(&l->transmq, skb) {
999                 if (!less(buf_seqno(skb), from))
1000                         break;
1001         }
1002
1003         skb_queue_walk_from(&l->transmq, skb) {
1004                 if (more(buf_seqno(skb), to))
1005                         break;
1006                 hdr = buf_msg(skb);
1007                 _skb = __pskb_copy(skb, MIN_H_SIZE, GFP_ATOMIC);
1008                 if (!_skb)
1009                         return 0;
1010                 hdr = buf_msg(_skb);
1011                 msg_set_ack(hdr, ack);
1012                 msg_set_bcast_ack(hdr, bc_ack);
1013                 _skb->priority = TC_PRIO_CONTROL;
1014                 __skb_queue_tail(xmitq, _skb);
1015                 l->stats.retransmitted++;
1016         }
1017         return 0;
1018 }
1019
1020 /* tipc_data_input - deliver data and name distr msgs to upper layer
1021  *
1022  * Consumes buffer if message is of right type
1023  * Node lock must be held
1024  */
1025 static bool tipc_data_input(struct tipc_link *l, struct sk_buff *skb,
1026                             struct sk_buff_head *inputq)
1027 {
1028         switch (msg_user(buf_msg(skb))) {
1029         case TIPC_LOW_IMPORTANCE:
1030         case TIPC_MEDIUM_IMPORTANCE:
1031         case TIPC_HIGH_IMPORTANCE:
1032         case TIPC_CRITICAL_IMPORTANCE:
1033         case CONN_MANAGER:
1034                 skb_queue_tail(inputq, skb);
1035                 return true;
1036         case NAME_DISTRIBUTOR:
1037                 l->bc_rcvlink->state = LINK_ESTABLISHED;
1038                 skb_queue_tail(l->namedq, skb);
1039                 return true;
1040         case MSG_BUNDLER:
1041         case TUNNEL_PROTOCOL:
1042         case MSG_FRAGMENTER:
1043         case BCAST_PROTOCOL:
1044                 return false;
1045         default:
1046                 pr_warn("Dropping received illegal msg type\n");
1047                 kfree_skb(skb);
1048                 return false;
1049         };
1050 }
1051
1052 /* tipc_link_input - process packet that has passed link protocol check
1053  *
1054  * Consumes buffer
1055  */
1056 static int tipc_link_input(struct tipc_link *l, struct sk_buff *skb,
1057                            struct sk_buff_head *inputq)
1058 {
1059         struct tipc_msg *hdr = buf_msg(skb);
1060         struct sk_buff **reasm_skb = &l->reasm_buf;
1061         struct sk_buff *iskb;
1062         struct sk_buff_head tmpq;
1063         int usr = msg_user(hdr);
1064         int rc = 0;
1065         int pos = 0;
1066         int ipos = 0;
1067
1068         if (unlikely(usr == TUNNEL_PROTOCOL)) {
1069                 if (msg_type(hdr) == SYNCH_MSG) {
1070                         __skb_queue_purge(&l->deferdq);
1071                         goto drop;
1072                 }
1073                 if (!tipc_msg_extract(skb, &iskb, &ipos))
1074                         return rc;
1075                 kfree_skb(skb);
1076                 skb = iskb;
1077                 hdr = buf_msg(skb);
1078                 if (less(msg_seqno(hdr), l->drop_point))
1079                         goto drop;
1080                 if (tipc_data_input(l, skb, inputq))
1081                         return rc;
1082                 usr = msg_user(hdr);
1083                 reasm_skb = &l->failover_reasm_skb;
1084         }
1085
1086         if (usr == MSG_BUNDLER) {
1087                 skb_queue_head_init(&tmpq);
1088                 l->stats.recv_bundles++;
1089                 l->stats.recv_bundled += msg_msgcnt(hdr);
1090                 while (tipc_msg_extract(skb, &iskb, &pos))
1091                         tipc_data_input(l, iskb, &tmpq);
1092                 tipc_skb_queue_splice_tail(&tmpq, inputq);
1093                 return 0;
1094         } else if (usr == MSG_FRAGMENTER) {
1095                 l->stats.recv_fragments++;
1096                 if (tipc_buf_append(reasm_skb, &skb)) {
1097                         l->stats.recv_fragmented++;
1098                         tipc_data_input(l, skb, inputq);
1099                 } else if (!*reasm_skb && !link_is_bc_rcvlink(l)) {
1100                         pr_warn_ratelimited("Unable to build fragment list\n");
1101                         return tipc_link_fsm_evt(l, LINK_FAILURE_EVT);
1102                 }
1103                 return 0;
1104         } else if (usr == BCAST_PROTOCOL) {
1105                 tipc_bcast_lock(l->owner->net);
1106                 tipc_link_bc_init_rcv(l->bc_rcvlink, hdr);
1107                 tipc_bcast_unlock(l->owner->net);
1108         }
1109 drop:
1110         kfree_skb(skb);
1111         return 0;
1112 }
1113
1114 static bool tipc_link_release_pkts(struct tipc_link *l, u16 acked)
1115 {
1116         bool released = false;
1117         struct sk_buff *skb, *tmp;
1118
1119         skb_queue_walk_safe(&l->transmq, skb, tmp) {
1120                 if (more(buf_seqno(skb), acked))
1121                         break;
1122                 __skb_unlink(skb, &l->transmq);
1123                 kfree_skb(skb);
1124                 released = true;
1125         }
1126         return released;
1127 }
1128
1129 /* tipc_link_build_ack_msg: prepare link acknowledge message for transmission
1130  *
1131  * Note that sending of broadcast ack is coordinated among nodes, to reduce
1132  * risk of ack storms towards the sender
1133  */
1134 int tipc_link_build_ack_msg(struct tipc_link *l, struct sk_buff_head *xmitq)
1135 {
1136         if (!l)
1137                 return 0;
1138
1139         /* Broadcast ACK must be sent via a unicast link => defer to caller */
1140         if (link_is_bc_rcvlink(l)) {
1141                 if (((l->rcv_nxt ^ link_own_addr(l)) & 0xf) != 0xf)
1142                         return 0;
1143                 l->rcv_unacked = 0;
1144                 return TIPC_LINK_SND_BC_ACK;
1145         }
1146
1147         /* Unicast ACK */
1148         l->rcv_unacked = 0;
1149         l->stats.sent_acks++;
1150         tipc_link_build_proto_msg(l, STATE_MSG, 0, 0, 0, 0, xmitq);
1151         return 0;
1152 }
1153
1154 /* tipc_link_build_reset_msg: prepare link RESET or ACTIVATE message
1155  */
1156 void tipc_link_build_reset_msg(struct tipc_link *l, struct sk_buff_head *xmitq)
1157 {
1158         int mtyp = RESET_MSG;
1159
1160         if (l->state == LINK_ESTABLISHING)
1161                 mtyp = ACTIVATE_MSG;
1162
1163         tipc_link_build_proto_msg(l, mtyp, 0, 0, 0, 0, xmitq);
1164 }
1165
1166 /* tipc_link_build_nack_msg: prepare link nack message for transmission
1167  */
1168 static void tipc_link_build_nack_msg(struct tipc_link *l,
1169                                      struct sk_buff_head *xmitq)
1170 {
1171         u32 def_cnt = ++l->stats.deferred_recv;
1172
1173         if (link_is_bc_rcvlink(l))
1174                 return;
1175
1176         if ((skb_queue_len(&l->deferdq) == 1) || !(def_cnt % TIPC_NACK_INTV))
1177                 tipc_link_build_proto_msg(l, STATE_MSG, 0, 0, 0, 0, xmitq);
1178 }
1179
1180 /* tipc_link_rcv - process TIPC packets/messages arriving from off-node
1181  * @l: the link that should handle the message
1182  * @skb: TIPC packet
1183  * @xmitq: queue to place packets to be sent after this call
1184  */
1185 int tipc_link_rcv(struct tipc_link *l, struct sk_buff *skb,
1186                   struct sk_buff_head *xmitq)
1187 {
1188         struct sk_buff_head *defq = &l->deferdq;
1189         struct tipc_msg *hdr;
1190         u16 seqno, rcv_nxt, win_lim;
1191         int rc = 0;
1192
1193         do {
1194                 hdr = buf_msg(skb);
1195                 seqno = msg_seqno(hdr);
1196                 rcv_nxt = l->rcv_nxt;
1197                 win_lim = rcv_nxt + TIPC_MAX_LINK_WIN;
1198
1199                 /* Verify and update link state */
1200                 if (unlikely(msg_user(hdr) == LINK_PROTOCOL))
1201                         return tipc_link_proto_rcv(l, skb, xmitq);
1202
1203                 if (unlikely(!link_is_up(l))) {
1204                         if (l->state == LINK_ESTABLISHING)
1205                                 rc = TIPC_LINK_UP_EVT;
1206                         goto drop;
1207                 }
1208
1209                 /* Don't send probe at next timeout expiration */
1210                 l->silent_intv_cnt = 0;
1211
1212                 /* Drop if outside receive window */
1213                 if (unlikely(less(seqno, rcv_nxt) || more(seqno, win_lim))) {
1214                         l->stats.duplicates++;
1215                         goto drop;
1216                 }
1217
1218                 /* Forward queues and wake up waiting users */
1219                 if (likely(tipc_link_release_pkts(l, msg_ack(hdr)))) {
1220                         tipc_link_advance_backlog(l, xmitq);
1221                         if (unlikely(!skb_queue_empty(&l->wakeupq)))
1222                                 link_prepare_wakeup(l);
1223                 }
1224
1225                 /* Defer delivery if sequence gap */
1226                 if (unlikely(seqno != rcv_nxt)) {
1227                         __tipc_skb_queue_sorted(defq, seqno, skb);
1228                         tipc_link_build_nack_msg(l, xmitq);
1229                         break;
1230                 }
1231
1232                 /* Deliver packet */
1233                 l->rcv_nxt++;
1234                 l->stats.recv_info++;
1235                 if (!tipc_data_input(l, skb, l->inputq))
1236                         rc |= tipc_link_input(l, skb, l->inputq);
1237                 if (unlikely(++l->rcv_unacked >= TIPC_MIN_LINK_WIN))
1238                         rc |= tipc_link_build_ack_msg(l, xmitq);
1239                 if (unlikely(rc & ~TIPC_LINK_SND_BC_ACK))
1240                         break;
1241         } while ((skb = __skb_dequeue(defq)));
1242
1243         return rc;
1244 drop:
1245         kfree_skb(skb);
1246         return rc;
1247 }
1248
1249 /**
1250  * tipc_link_defer_pkt - Add out-of-sequence message to deferred reception queue
1251  *
1252  * Returns increase in queue length (i.e. 0 or 1)
1253  */
1254 u32 tipc_link_defer_pkt(struct sk_buff_head *list, struct sk_buff *skb)
1255 {
1256         struct sk_buff *skb1;
1257         u16 seq_no = buf_seqno(skb);
1258
1259         /* Empty queue ? */
1260         if (skb_queue_empty(list)) {
1261                 __skb_queue_tail(list, skb);
1262                 return 1;
1263         }
1264
1265         /* Last ? */
1266         if (less(buf_seqno(skb_peek_tail(list)), seq_no)) {
1267                 __skb_queue_tail(list, skb);
1268                 return 1;
1269         }
1270
1271         /* Locate insertion point in queue, then insert; discard if duplicate */
1272         skb_queue_walk(list, skb1) {
1273                 u16 curr_seqno = buf_seqno(skb1);
1274
1275                 if (seq_no == curr_seqno) {
1276                         kfree_skb(skb);
1277                         return 0;
1278                 }
1279
1280                 if (less(seq_no, curr_seqno))
1281                         break;
1282         }
1283
1284         __skb_queue_before(list, skb1, skb);
1285         return 1;
1286 }
1287
1288 /*
1289  * Send protocol message to the other endpoint.
1290  */
1291 void tipc_link_proto_xmit(struct tipc_link *l, u32 msg_typ, int probe_msg,
1292                           u32 gap, u32 tolerance, u32 priority)
1293 {
1294         struct sk_buff *skb = NULL;
1295         struct sk_buff_head xmitq;
1296
1297         __skb_queue_head_init(&xmitq);
1298         tipc_link_build_proto_msg(l, msg_typ, probe_msg, gap,
1299                                   tolerance, priority, &xmitq);
1300         skb = __skb_dequeue(&xmitq);
1301         if (!skb)
1302                 return;
1303         tipc_bearer_send(l->owner->net, l->bearer_id, skb, l->media_addr);
1304         l->rcv_unacked = 0;
1305         kfree_skb(skb);
1306 }
1307
1308 static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe,
1309                                       u16 rcvgap, int tolerance, int priority,
1310                                       struct sk_buff_head *xmitq)
1311 {
1312         struct sk_buff *skb = NULL;
1313         struct tipc_msg *hdr = l->pmsg;
1314         bool node_up = link_is_up(l->bc_rcvlink);
1315
1316         /* Don't send protocol message during reset or link failover */
1317         if (tipc_link_is_blocked(l))
1318                 return;
1319
1320         msg_set_type(hdr, mtyp);
1321         msg_set_net_plane(hdr, l->net_plane);
1322         msg_set_next_sent(hdr, l->snd_nxt);
1323         msg_set_ack(hdr, l->rcv_nxt - 1);
1324         msg_set_bcast_ack(hdr, l->bc_rcvlink->rcv_nxt - 1);
1325         msg_set_last_bcast(hdr, l->bc_sndlink->snd_nxt - 1);
1326         msg_set_link_tolerance(hdr, tolerance);
1327         msg_set_linkprio(hdr, priority);
1328         msg_set_redundant_link(hdr, node_up);
1329         msg_set_seq_gap(hdr, 0);
1330
1331         /* Compatibility: created msg must not be in sequence with pkt flow */
1332         msg_set_seqno(hdr, l->snd_nxt + U16_MAX / 2);
1333
1334         if (mtyp == STATE_MSG) {
1335                 if (!tipc_link_is_up(l))
1336                         return;
1337
1338                 /* Override rcvgap if there are packets in deferred queue */
1339                 if (!skb_queue_empty(&l->deferdq))
1340                         rcvgap = buf_seqno(skb_peek(&l->deferdq)) - l->rcv_nxt;
1341                 if (rcvgap) {
1342                         msg_set_seq_gap(hdr, rcvgap);
1343                         l->stats.sent_nacks++;
1344                 }
1345                 msg_set_probe(hdr, probe);
1346                 if (probe)
1347                         l->stats.sent_probes++;
1348                 l->stats.sent_states++;
1349                 l->rcv_unacked = 0;
1350         } else {
1351                 /* RESET_MSG or ACTIVATE_MSG */
1352                 msg_set_max_pkt(hdr, l->advertised_mtu);
1353                 msg_set_ack(hdr, l->rcv_nxt - 1);
1354                 msg_set_next_sent(hdr, 1);
1355         }
1356         skb = tipc_buf_acquire(msg_size(hdr));
1357         if (!skb)
1358                 return;
1359         skb_copy_to_linear_data(skb, hdr, msg_size(hdr));
1360         skb->priority = TC_PRIO_CONTROL;
1361         __skb_queue_tail(xmitq, skb);
1362 }
1363
1364 /* tipc_link_tnl_prepare(): prepare and return a list of tunnel packets
1365  * with contents of the link's transmit and backlog queues.
1366  */
1367 void tipc_link_tnl_prepare(struct tipc_link *l, struct tipc_link *tnl,
1368                            int mtyp, struct sk_buff_head *xmitq)
1369 {
1370         struct sk_buff *skb, *tnlskb;
1371         struct tipc_msg *hdr, tnlhdr;
1372         struct sk_buff_head *queue = &l->transmq;
1373         struct sk_buff_head tmpxq, tnlq;
1374         u16 pktlen, pktcnt, seqno = l->snd_nxt;
1375
1376         if (!tnl)
1377                 return;
1378
1379         skb_queue_head_init(&tnlq);
1380         skb_queue_head_init(&tmpxq);
1381
1382         /* At least one packet required for safe algorithm => add dummy */
1383         skb = tipc_msg_create(TIPC_LOW_IMPORTANCE, TIPC_DIRECT_MSG,
1384                               BASIC_H_SIZE, 0, l->addr, link_own_addr(l),
1385                               0, 0, TIPC_ERR_NO_PORT);
1386         if (!skb) {
1387                 pr_warn("%sunable to create tunnel packet\n", link_co_err);
1388                 return;
1389         }
1390         skb_queue_tail(&tnlq, skb);
1391         tipc_link_xmit(l, &tnlq, &tmpxq);
1392         __skb_queue_purge(&tmpxq);
1393
1394         /* Initialize reusable tunnel packet header */
1395         tipc_msg_init(link_own_addr(l), &tnlhdr, TUNNEL_PROTOCOL,
1396                       mtyp, INT_H_SIZE, l->addr);
1397         pktcnt = skb_queue_len(&l->transmq) + skb_queue_len(&l->backlogq);
1398         msg_set_msgcnt(&tnlhdr, pktcnt);
1399         msg_set_bearer_id(&tnlhdr, l->peer_bearer_id);
1400 tnl:
1401         /* Wrap each packet into a tunnel packet */
1402         skb_queue_walk(queue, skb) {
1403                 hdr = buf_msg(skb);
1404                 if (queue == &l->backlogq)
1405                         msg_set_seqno(hdr, seqno++);
1406                 pktlen = msg_size(hdr);
1407                 msg_set_size(&tnlhdr, pktlen + INT_H_SIZE);
1408                 tnlskb = tipc_buf_acquire(pktlen + INT_H_SIZE);
1409                 if (!tnlskb) {
1410                         pr_warn("%sunable to send packet\n", link_co_err);
1411                         return;
1412                 }
1413                 skb_copy_to_linear_data(tnlskb, &tnlhdr, INT_H_SIZE);
1414                 skb_copy_to_linear_data_offset(tnlskb, INT_H_SIZE, hdr, pktlen);
1415                 __skb_queue_tail(&tnlq, tnlskb);
1416         }
1417         if (queue != &l->backlogq) {
1418                 queue = &l->backlogq;
1419                 goto tnl;
1420         }
1421
1422         tipc_link_xmit(tnl, &tnlq, xmitq);
1423
1424         if (mtyp == FAILOVER_MSG) {
1425                 tnl->drop_point = l->rcv_nxt;
1426                 tnl->failover_reasm_skb = l->reasm_buf;
1427                 l->reasm_buf = NULL;
1428         }
1429 }
1430
1431 /* tipc_link_proto_rcv(): receive link level protocol message :
1432  * Note that network plane id propagates through the network, and may
1433  * change at any time. The node with lowest numerical id determines
1434  * network plane
1435  */
1436 static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
1437                                struct sk_buff_head *xmitq)
1438 {
1439         struct tipc_msg *hdr = buf_msg(skb);
1440         u16 rcvgap = 0;
1441         u16 ack = msg_ack(hdr);
1442         u16 gap = msg_seq_gap(hdr);
1443         u16 peers_snd_nxt =  msg_next_sent(hdr);
1444         u16 peers_tol = msg_link_tolerance(hdr);
1445         u16 peers_prio = msg_linkprio(hdr);
1446         u16 rcv_nxt = l->rcv_nxt;
1447         int mtyp = msg_type(hdr);
1448         char *if_name;
1449         int rc = 0;
1450
1451         if (tipc_link_is_blocked(l) || !xmitq)
1452                 goto exit;
1453
1454         if (link_own_addr(l) > msg_prevnode(hdr))
1455                 l->net_plane = msg_net_plane(hdr);
1456
1457         switch (mtyp) {
1458         case RESET_MSG:
1459
1460                 /* Ignore duplicate RESET with old session number */
1461                 if ((less_eq(msg_session(hdr), l->peer_session)) &&
1462                     (l->peer_session != WILDCARD_SESSION))
1463                         break;
1464                 /* fall thru' */
1465
1466         case ACTIVATE_MSG:
1467
1468                 /* Complete own link name with peer's interface name */
1469                 if_name =  strrchr(l->name, ':') + 1;
1470                 if (sizeof(l->name) - (if_name - l->name) <= TIPC_MAX_IF_NAME)
1471                         break;
1472                 if (msg_data_sz(hdr) < TIPC_MAX_IF_NAME)
1473                         break;
1474                 strncpy(if_name, msg_data(hdr), TIPC_MAX_IF_NAME);
1475
1476                 /* Update own tolerance if peer indicates a non-zero value */
1477                 if (in_range(peers_tol, TIPC_MIN_LINK_TOL, TIPC_MAX_LINK_TOL))
1478                         l->tolerance = peers_tol;
1479
1480                 /* Update own priority if peer's priority is higher */
1481                 if (in_range(peers_prio, l->priority + 1, TIPC_MAX_LINK_PRI))
1482                         l->priority = peers_prio;
1483
1484                 /* ACTIVATE_MSG serves as PEER_RESET if link is already down */
1485                 if ((mtyp == RESET_MSG) || !link_is_up(l))
1486                         rc = tipc_link_fsm_evt(l, LINK_PEER_RESET_EVT);
1487
1488                 /* ACTIVATE_MSG takes up link if it was already locally reset */
1489                 if ((mtyp == ACTIVATE_MSG) && (l->state == LINK_ESTABLISHING))
1490                         rc = TIPC_LINK_UP_EVT;
1491
1492                 l->peer_session = msg_session(hdr);
1493                 l->peer_bearer_id = msg_bearer_id(hdr);
1494                 if (l->mtu > msg_max_pkt(hdr))
1495                         l->mtu = msg_max_pkt(hdr);
1496                 break;
1497
1498         case STATE_MSG:
1499
1500                 /* Update own tolerance if peer indicates a non-zero value */
1501                 if (in_range(peers_tol, TIPC_MIN_LINK_TOL, TIPC_MAX_LINK_TOL))
1502                         l->tolerance = peers_tol;
1503
1504                 l->silent_intv_cnt = 0;
1505                 l->stats.recv_states++;
1506                 if (msg_probe(hdr))
1507                         l->stats.recv_probes++;
1508
1509                 if (!link_is_up(l)) {
1510                         if (l->state == LINK_ESTABLISHING)
1511                                 rc = TIPC_LINK_UP_EVT;
1512                         break;
1513                 }
1514
1515                 /* Send NACK if peer has sent pkts we haven't received yet */
1516                 if (more(peers_snd_nxt, rcv_nxt) && !tipc_link_is_synching(l))
1517                         rcvgap = peers_snd_nxt - l->rcv_nxt;
1518                 if (rcvgap || (msg_probe(hdr)))
1519                         tipc_link_build_proto_msg(l, STATE_MSG, 0, rcvgap,
1520                                                   0, 0, xmitq);
1521                 tipc_link_release_pkts(l, ack);
1522
1523                 /* If NACK, retransmit will now start at right position */
1524                 if (gap) {
1525                         rc = tipc_link_retrans(l, ack + 1, ack + gap, xmitq);
1526                         l->stats.recv_nacks++;
1527                 }
1528
1529                 tipc_link_advance_backlog(l, xmitq);
1530                 if (unlikely(!skb_queue_empty(&l->wakeupq)))
1531                         link_prepare_wakeup(l);
1532         }
1533 exit:
1534         kfree_skb(skb);
1535         return rc;
1536 }
1537
1538 /* tipc_link_build_bc_proto_msg() - create broadcast protocol message
1539  */
1540 static bool tipc_link_build_bc_proto_msg(struct tipc_link *l, bool bcast,
1541                                          u16 peers_snd_nxt,
1542                                          struct sk_buff_head *xmitq)
1543 {
1544         struct sk_buff *skb;
1545         struct tipc_msg *hdr;
1546         struct sk_buff *dfrd_skb = skb_peek(&l->deferdq);
1547         u16 ack = l->rcv_nxt - 1;
1548         u16 gap_to = peers_snd_nxt - 1;
1549
1550         skb = tipc_msg_create(BCAST_PROTOCOL, STATE_MSG, INT_H_SIZE,
1551                               0, l->addr, link_own_addr(l), 0, 0, 0);
1552         if (!skb)
1553                 return false;
1554         hdr = buf_msg(skb);
1555         msg_set_last_bcast(hdr, l->bc_sndlink->snd_nxt - 1);
1556         msg_set_bcast_ack(hdr, ack);
1557         msg_set_bcgap_after(hdr, ack);
1558         if (dfrd_skb)
1559                 gap_to = buf_seqno(dfrd_skb) - 1;
1560         msg_set_bcgap_to(hdr, gap_to);
1561         msg_set_non_seq(hdr, bcast);
1562         __skb_queue_tail(xmitq, skb);
1563         return true;
1564 }
1565
1566 /* tipc_link_build_bc_init_msg() - synchronize broadcast link endpoints.
1567  *
1568  * Give a newly added peer node the sequence number where it should
1569  * start receiving and acking broadcast packets.
1570  */
1571 void tipc_link_build_bc_init_msg(struct tipc_link *l,
1572                                  struct sk_buff_head *xmitq)
1573 {
1574         struct sk_buff_head list;
1575
1576         __skb_queue_head_init(&list);
1577         if (!tipc_link_build_bc_proto_msg(l->bc_rcvlink, false, 0, &list))
1578                 return;
1579         tipc_link_xmit(l, &list, xmitq);
1580 }
1581
1582 /* tipc_link_bc_init_rcv - receive initial broadcast synch data from peer
1583  */
1584 void tipc_link_bc_init_rcv(struct tipc_link *l, struct tipc_msg *hdr)
1585 {
1586         int mtyp = msg_type(hdr);
1587         u16 peers_snd_nxt = msg_bc_snd_nxt(hdr);
1588
1589         if (link_is_up(l))
1590                 return;
1591
1592         if (msg_user(hdr) == BCAST_PROTOCOL) {
1593                 l->rcv_nxt = peers_snd_nxt;
1594                 l->state = LINK_ESTABLISHED;
1595                 return;
1596         }
1597
1598         if (l->peer_caps & TIPC_BCAST_SYNCH)
1599                 return;
1600
1601         if (msg_peer_node_is_up(hdr))
1602                 return;
1603
1604         /* Compatibility: accept older, less safe initial synch data */
1605         if ((mtyp == RESET_MSG) || (mtyp == ACTIVATE_MSG))
1606                 l->rcv_nxt = peers_snd_nxt;
1607 }
1608
1609 /* tipc_link_bc_sync_rcv - update rcv link according to peer's send state
1610  */
1611 void tipc_link_bc_sync_rcv(struct tipc_link *l, struct tipc_msg *hdr,
1612                            struct sk_buff_head *xmitq)
1613 {
1614         u16 peers_snd_nxt = msg_bc_snd_nxt(hdr);
1615
1616         if (!link_is_up(l))
1617                 return;
1618
1619         if (!msg_peer_node_is_up(hdr))
1620                 return;
1621
1622         l->bc_peer_is_up = true;
1623
1624         /* Ignore if peers_snd_nxt goes beyond receive window */
1625         if (more(peers_snd_nxt, l->rcv_nxt + l->window))
1626                 return;
1627
1628         if (!more(peers_snd_nxt, l->rcv_nxt)) {
1629                 l->nack_state = BC_NACK_SND_CONDITIONAL;
1630                 return;
1631         }
1632
1633         /* Don't NACK if one was recently sent or peeked */
1634         if (l->nack_state == BC_NACK_SND_SUPPRESS) {
1635                 l->nack_state = BC_NACK_SND_UNCONDITIONAL;
1636                 return;
1637         }
1638
1639         /* Conditionally delay NACK sending until next synch rcv */
1640         if (l->nack_state == BC_NACK_SND_CONDITIONAL) {
1641                 l->nack_state = BC_NACK_SND_UNCONDITIONAL;
1642                 if ((peers_snd_nxt - l->rcv_nxt) < TIPC_MIN_LINK_WIN)
1643                         return;
1644         }
1645
1646         /* Send NACK now but suppress next one */
1647         tipc_link_build_bc_proto_msg(l, true, peers_snd_nxt, xmitq);
1648         l->nack_state = BC_NACK_SND_SUPPRESS;
1649 }
1650
1651 void tipc_link_bc_ack_rcv(struct tipc_link *l, u16 acked,
1652                           struct sk_buff_head *xmitq)
1653 {
1654         struct sk_buff *skb, *tmp;
1655         struct tipc_link *snd_l = l->bc_sndlink;
1656
1657         if (!link_is_up(l) || !l->bc_peer_is_up)
1658                 return;
1659
1660         if (!more(acked, l->acked))
1661                 return;
1662
1663         /* Skip over packets peer has already acked */
1664         skb_queue_walk(&snd_l->transmq, skb) {
1665                 if (more(buf_seqno(skb), l->acked))
1666                         break;
1667         }
1668
1669         /* Update/release the packets peer is acking now */
1670         skb_queue_walk_from_safe(&snd_l->transmq, skb, tmp) {
1671                 if (more(buf_seqno(skb), acked))
1672                         break;
1673                 if (!--TIPC_SKB_CB(skb)->ackers) {
1674                         __skb_unlink(skb, &snd_l->transmq);
1675                         kfree_skb(skb);
1676                 }
1677         }
1678         l->acked = acked;
1679         tipc_link_advance_backlog(snd_l, xmitq);
1680         if (unlikely(!skb_queue_empty(&snd_l->wakeupq)))
1681                 link_prepare_wakeup(snd_l);
1682 }
1683
1684 /* tipc_link_bc_nack_rcv(): receive broadcast nack message
1685  */
1686 int tipc_link_bc_nack_rcv(struct tipc_link *l, struct sk_buff *skb,
1687                           struct sk_buff_head *xmitq)
1688 {
1689         struct tipc_msg *hdr = buf_msg(skb);
1690         u32 dnode = msg_destnode(hdr);
1691         int mtyp = msg_type(hdr);
1692         u16 acked = msg_bcast_ack(hdr);
1693         u16 from = acked + 1;
1694         u16 to = msg_bcgap_to(hdr);
1695         u16 peers_snd_nxt = to + 1;
1696         int rc = 0;
1697
1698         kfree_skb(skb);
1699
1700         if (!tipc_link_is_up(l) || !l->bc_peer_is_up)
1701                 return 0;
1702
1703         if (mtyp != STATE_MSG)
1704                 return 0;
1705
1706         if (dnode == link_own_addr(l)) {
1707                 tipc_link_bc_ack_rcv(l, acked, xmitq);
1708                 rc = tipc_link_retrans(l->bc_sndlink, from, to, xmitq);
1709                 l->stats.recv_nacks++;
1710                 return rc;
1711         }
1712
1713         /* Msg for other node => suppress own NACK at next sync if applicable */
1714         if (more(peers_snd_nxt, l->rcv_nxt) && !less(l->rcv_nxt, from))
1715                 l->nack_state = BC_NACK_SND_SUPPRESS;
1716
1717         return 0;
1718 }
1719
1720 void tipc_link_set_queue_limits(struct tipc_link *l, u32 win)
1721 {
1722         int max_bulk = TIPC_MAX_PUBLICATIONS / (l->mtu / ITEM_SIZE);
1723
1724         l->window = win;
1725         l->backlog[TIPC_LOW_IMPORTANCE].limit      = win / 2;
1726         l->backlog[TIPC_MEDIUM_IMPORTANCE].limit   = win;
1727         l->backlog[TIPC_HIGH_IMPORTANCE].limit     = win / 2 * 3;
1728         l->backlog[TIPC_CRITICAL_IMPORTANCE].limit = win * 2;
1729         l->backlog[TIPC_SYSTEM_IMPORTANCE].limit   = max_bulk;
1730 }
1731
1732 /* tipc_link_find_owner - locate owner node of link by link's name
1733  * @net: the applicable net namespace
1734  * @name: pointer to link name string
1735  * @bearer_id: pointer to index in 'node->links' array where the link was found.
1736  *
1737  * Returns pointer to node owning the link, or 0 if no matching link is found.
1738  */
1739 static struct tipc_node *tipc_link_find_owner(struct net *net,
1740                                               const char *link_name,
1741                                               unsigned int *bearer_id)
1742 {
1743         struct tipc_net *tn = net_generic(net, tipc_net_id);
1744         struct tipc_link *l_ptr;
1745         struct tipc_node *n_ptr;
1746         struct tipc_node *found_node = NULL;
1747         int i;
1748
1749         *bearer_id = 0;
1750         rcu_read_lock();
1751         list_for_each_entry_rcu(n_ptr, &tn->node_list, list) {
1752                 tipc_node_lock(n_ptr);
1753                 for (i = 0; i < MAX_BEARERS; i++) {
1754                         l_ptr = n_ptr->links[i].link;
1755                         if (l_ptr && !strcmp(l_ptr->name, link_name)) {
1756                                 *bearer_id = i;
1757                                 found_node = n_ptr;
1758                                 break;
1759                         }
1760                 }
1761                 tipc_node_unlock(n_ptr);
1762                 if (found_node)
1763                         break;
1764         }
1765         rcu_read_unlock();
1766
1767         return found_node;
1768 }
1769
1770 /**
1771  * link_reset_statistics - reset link statistics
1772  * @l_ptr: pointer to link
1773  */
1774 static void link_reset_statistics(struct tipc_link *l_ptr)
1775 {
1776         memset(&l_ptr->stats, 0, sizeof(l_ptr->stats));
1777         l_ptr->stats.sent_info = l_ptr->snd_nxt;
1778         l_ptr->stats.recv_info = l_ptr->rcv_nxt;
1779 }
1780
1781 static void link_print(struct tipc_link *l, const char *str)
1782 {
1783         struct sk_buff *hskb = skb_peek(&l->transmq);
1784         u16 head = hskb ? msg_seqno(buf_msg(hskb)) : l->snd_nxt - 1;
1785         u16 tail = l->snd_nxt - 1;
1786
1787         pr_info("%s Link <%s> state %x\n", str, l->name, l->state);
1788         pr_info("XMTQ: %u [%u-%u], BKLGQ: %u, SNDNX: %u, RCVNX: %u\n",
1789                 skb_queue_len(&l->transmq), head, tail,
1790                 skb_queue_len(&l->backlogq), l->snd_nxt, l->rcv_nxt);
1791 }
1792
1793 /* Parse and validate nested (link) properties valid for media, bearer and link
1794  */
1795 int tipc_nl_parse_link_prop(struct nlattr *prop, struct nlattr *props[])
1796 {
1797         int err;
1798
1799         err = nla_parse_nested(props, TIPC_NLA_PROP_MAX, prop,
1800                                tipc_nl_prop_policy);
1801         if (err)
1802                 return err;
1803
1804         if (props[TIPC_NLA_PROP_PRIO]) {
1805                 u32 prio;
1806
1807                 prio = nla_get_u32(props[TIPC_NLA_PROP_PRIO]);
1808                 if (prio > TIPC_MAX_LINK_PRI)
1809                         return -EINVAL;
1810         }
1811
1812         if (props[TIPC_NLA_PROP_TOL]) {
1813                 u32 tol;
1814
1815                 tol = nla_get_u32(props[TIPC_NLA_PROP_TOL]);
1816                 if ((tol < TIPC_MIN_LINK_TOL) || (tol > TIPC_MAX_LINK_TOL))
1817                         return -EINVAL;
1818         }
1819
1820         if (props[TIPC_NLA_PROP_WIN]) {
1821                 u32 win;
1822
1823                 win = nla_get_u32(props[TIPC_NLA_PROP_WIN]);
1824                 if ((win < TIPC_MIN_LINK_WIN) || (win > TIPC_MAX_LINK_WIN))
1825                         return -EINVAL;
1826         }
1827
1828         return 0;
1829 }
1830
1831 int tipc_nl_link_set(struct sk_buff *skb, struct genl_info *info)
1832 {
1833         int err;
1834         int res = 0;
1835         int bearer_id;
1836         char *name;
1837         struct tipc_link *link;
1838         struct tipc_node *node;
1839         struct nlattr *attrs[TIPC_NLA_LINK_MAX + 1];
1840         struct net *net = sock_net(skb->sk);
1841
1842         if (!info->attrs[TIPC_NLA_LINK])
1843                 return -EINVAL;
1844
1845         err = nla_parse_nested(attrs, TIPC_NLA_LINK_MAX,
1846                                info->attrs[TIPC_NLA_LINK],
1847                                tipc_nl_link_policy);
1848         if (err)
1849                 return err;
1850
1851         if (!attrs[TIPC_NLA_LINK_NAME])
1852                 return -EINVAL;
1853
1854         name = nla_data(attrs[TIPC_NLA_LINK_NAME]);
1855
1856         if (strcmp(name, tipc_bclink_name) == 0)
1857                 return tipc_nl_bc_link_set(net, attrs);
1858
1859         node = tipc_link_find_owner(net, name, &bearer_id);
1860         if (!node)
1861                 return -EINVAL;
1862
1863         tipc_node_lock(node);
1864
1865         link = node->links[bearer_id].link;
1866         if (!link) {
1867                 res = -EINVAL;
1868                 goto out;
1869         }
1870
1871         if (attrs[TIPC_NLA_LINK_PROP]) {
1872                 struct nlattr *props[TIPC_NLA_PROP_MAX + 1];
1873
1874                 err = tipc_nl_parse_link_prop(attrs[TIPC_NLA_LINK_PROP],
1875                                               props);
1876                 if (err) {
1877                         res = err;
1878                         goto out;
1879                 }
1880
1881                 if (props[TIPC_NLA_PROP_TOL]) {
1882                         u32 tol;
1883
1884                         tol = nla_get_u32(props[TIPC_NLA_PROP_TOL]);
1885                         link->tolerance = tol;
1886                         tipc_link_proto_xmit(link, STATE_MSG, 0, 0, tol, 0);
1887                 }
1888                 if (props[TIPC_NLA_PROP_PRIO]) {
1889                         u32 prio;
1890
1891                         prio = nla_get_u32(props[TIPC_NLA_PROP_PRIO]);
1892                         link->priority = prio;
1893                         tipc_link_proto_xmit(link, STATE_MSG, 0, 0, 0, prio);
1894                 }
1895                 if (props[TIPC_NLA_PROP_WIN]) {
1896                         u32 win;
1897
1898                         win = nla_get_u32(props[TIPC_NLA_PROP_WIN]);
1899                         tipc_link_set_queue_limits(link, win);
1900                 }
1901         }
1902
1903 out:
1904         tipc_node_unlock(node);
1905
1906         return res;
1907 }
1908
1909 static int __tipc_nl_add_stats(struct sk_buff *skb, struct tipc_stats *s)
1910 {
1911         int i;
1912         struct nlattr *stats;
1913
1914         struct nla_map {
1915                 u32 key;
1916                 u32 val;
1917         };
1918
1919         struct nla_map map[] = {
1920                 {TIPC_NLA_STATS_RX_INFO, s->recv_info},
1921                 {TIPC_NLA_STATS_RX_FRAGMENTS, s->recv_fragments},
1922                 {TIPC_NLA_STATS_RX_FRAGMENTED, s->recv_fragmented},
1923                 {TIPC_NLA_STATS_RX_BUNDLES, s->recv_bundles},
1924                 {TIPC_NLA_STATS_RX_BUNDLED, s->recv_bundled},
1925                 {TIPC_NLA_STATS_TX_INFO, s->sent_info},
1926                 {TIPC_NLA_STATS_TX_FRAGMENTS, s->sent_fragments},
1927                 {TIPC_NLA_STATS_TX_FRAGMENTED, s->sent_fragmented},
1928                 {TIPC_NLA_STATS_TX_BUNDLES, s->sent_bundles},
1929                 {TIPC_NLA_STATS_TX_BUNDLED, s->sent_bundled},
1930                 {TIPC_NLA_STATS_MSG_PROF_TOT, (s->msg_length_counts) ?
1931                         s->msg_length_counts : 1},
1932                 {TIPC_NLA_STATS_MSG_LEN_CNT, s->msg_length_counts},
1933                 {TIPC_NLA_STATS_MSG_LEN_TOT, s->msg_lengths_total},
1934                 {TIPC_NLA_STATS_MSG_LEN_P0, s->msg_length_profile[0]},
1935                 {TIPC_NLA_STATS_MSG_LEN_P1, s->msg_length_profile[1]},
1936                 {TIPC_NLA_STATS_MSG_LEN_P2, s->msg_length_profile[2]},
1937                 {TIPC_NLA_STATS_MSG_LEN_P3, s->msg_length_profile[3]},
1938                 {TIPC_NLA_STATS_MSG_LEN_P4, s->msg_length_profile[4]},
1939                 {TIPC_NLA_STATS_MSG_LEN_P5, s->msg_length_profile[5]},
1940                 {TIPC_NLA_STATS_MSG_LEN_P6, s->msg_length_profile[6]},
1941                 {TIPC_NLA_STATS_RX_STATES, s->recv_states},
1942                 {TIPC_NLA_STATS_RX_PROBES, s->recv_probes},
1943                 {TIPC_NLA_STATS_RX_NACKS, s->recv_nacks},
1944                 {TIPC_NLA_STATS_RX_DEFERRED, s->deferred_recv},
1945                 {TIPC_NLA_STATS_TX_STATES, s->sent_states},
1946                 {TIPC_NLA_STATS_TX_PROBES, s->sent_probes},
1947                 {TIPC_NLA_STATS_TX_NACKS, s->sent_nacks},
1948                 {TIPC_NLA_STATS_TX_ACKS, s->sent_acks},
1949                 {TIPC_NLA_STATS_RETRANSMITTED, s->retransmitted},
1950                 {TIPC_NLA_STATS_DUPLICATES, s->duplicates},
1951                 {TIPC_NLA_STATS_LINK_CONGS, s->link_congs},
1952                 {TIPC_NLA_STATS_MAX_QUEUE, s->max_queue_sz},
1953                 {TIPC_NLA_STATS_AVG_QUEUE, s->queue_sz_counts ?
1954                         (s->accu_queue_sz / s->queue_sz_counts) : 0}
1955         };
1956
1957         stats = nla_nest_start(skb, TIPC_NLA_LINK_STATS);
1958         if (!stats)
1959                 return -EMSGSIZE;
1960
1961         for (i = 0; i <  ARRAY_SIZE(map); i++)
1962                 if (nla_put_u32(skb, map[i].key, map[i].val))
1963                         goto msg_full;
1964
1965         nla_nest_end(skb, stats);
1966
1967         return 0;
1968 msg_full:
1969         nla_nest_cancel(skb, stats);
1970
1971         return -EMSGSIZE;
1972 }
1973
1974 /* Caller should hold appropriate locks to protect the link */
1975 static int __tipc_nl_add_link(struct net *net, struct tipc_nl_msg *msg,
1976                               struct tipc_link *link, int nlflags)
1977 {
1978         int err;
1979         void *hdr;
1980         struct nlattr *attrs;
1981         struct nlattr *prop;
1982         struct tipc_net *tn = net_generic(net, tipc_net_id);
1983
1984         hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family,
1985                           nlflags, TIPC_NL_LINK_GET);
1986         if (!hdr)
1987                 return -EMSGSIZE;
1988
1989         attrs = nla_nest_start(msg->skb, TIPC_NLA_LINK);
1990         if (!attrs)
1991                 goto msg_full;
1992
1993         if (nla_put_string(msg->skb, TIPC_NLA_LINK_NAME, link->name))
1994                 goto attr_msg_full;
1995         if (nla_put_u32(msg->skb, TIPC_NLA_LINK_DEST,
1996                         tipc_cluster_mask(tn->own_addr)))
1997                 goto attr_msg_full;
1998         if (nla_put_u32(msg->skb, TIPC_NLA_LINK_MTU, link->mtu))
1999                 goto attr_msg_full;
2000         if (nla_put_u32(msg->skb, TIPC_NLA_LINK_RX, link->rcv_nxt))
2001                 goto attr_msg_full;
2002         if (nla_put_u32(msg->skb, TIPC_NLA_LINK_TX, link->snd_nxt))
2003                 goto attr_msg_full;
2004
2005         if (tipc_link_is_up(link))
2006                 if (nla_put_flag(msg->skb, TIPC_NLA_LINK_UP))
2007                         goto attr_msg_full;
2008         if (tipc_link_is_active(link))
2009                 if (nla_put_flag(msg->skb, TIPC_NLA_LINK_ACTIVE))
2010                         goto attr_msg_full;
2011
2012         prop = nla_nest_start(msg->skb, TIPC_NLA_LINK_PROP);
2013         if (!prop)
2014                 goto attr_msg_full;
2015         if (nla_put_u32(msg->skb, TIPC_NLA_PROP_PRIO, link->priority))
2016                 goto prop_msg_full;
2017         if (nla_put_u32(msg->skb, TIPC_NLA_PROP_TOL, link->tolerance))
2018                 goto prop_msg_full;
2019         if (nla_put_u32(msg->skb, TIPC_NLA_PROP_WIN,
2020                         link->window))
2021                 goto prop_msg_full;
2022         if (nla_put_u32(msg->skb, TIPC_NLA_PROP_PRIO, link->priority))
2023                 goto prop_msg_full;
2024         nla_nest_end(msg->skb, prop);
2025
2026         err = __tipc_nl_add_stats(msg->skb, &link->stats);
2027         if (err)
2028                 goto attr_msg_full;
2029
2030         nla_nest_end(msg->skb, attrs);
2031         genlmsg_end(msg->skb, hdr);
2032
2033         return 0;
2034
2035 prop_msg_full:
2036         nla_nest_cancel(msg->skb, prop);
2037 attr_msg_full:
2038         nla_nest_cancel(msg->skb, attrs);
2039 msg_full:
2040         genlmsg_cancel(msg->skb, hdr);
2041
2042         return -EMSGSIZE;
2043 }
2044
2045 /* Caller should hold node lock  */
2046 static int __tipc_nl_add_node_links(struct net *net, struct tipc_nl_msg *msg,
2047                                     struct tipc_node *node, u32 *prev_link)
2048 {
2049         u32 i;
2050         int err;
2051
2052         for (i = *prev_link; i < MAX_BEARERS; i++) {
2053                 *prev_link = i;
2054
2055                 if (!node->links[i].link)
2056                         continue;
2057
2058                 err = __tipc_nl_add_link(net, msg,
2059                                          node->links[i].link, NLM_F_MULTI);
2060                 if (err)
2061                         return err;
2062         }
2063         *prev_link = 0;
2064
2065         return 0;
2066 }
2067
2068 int tipc_nl_link_dump(struct sk_buff *skb, struct netlink_callback *cb)
2069 {
2070         struct net *net = sock_net(skb->sk);
2071         struct tipc_net *tn = net_generic(net, tipc_net_id);
2072         struct tipc_node *node;
2073         struct tipc_nl_msg msg;
2074         u32 prev_node = cb->args[0];
2075         u32 prev_link = cb->args[1];
2076         int done = cb->args[2];
2077         int err;
2078
2079         if (done)
2080                 return 0;
2081
2082         msg.skb = skb;
2083         msg.portid = NETLINK_CB(cb->skb).portid;
2084         msg.seq = cb->nlh->nlmsg_seq;
2085
2086         rcu_read_lock();
2087         if (prev_node) {
2088                 node = tipc_node_find(net, prev_node);
2089                 if (!node) {
2090                         /* We never set seq or call nl_dump_check_consistent()
2091                          * this means that setting prev_seq here will cause the
2092                          * consistence check to fail in the netlink callback
2093                          * handler. Resulting in the last NLMSG_DONE message
2094                          * having the NLM_F_DUMP_INTR flag set.
2095                          */
2096                         cb->prev_seq = 1;
2097                         goto out;
2098                 }
2099                 tipc_node_put(node);
2100
2101                 list_for_each_entry_continue_rcu(node, &tn->node_list,
2102                                                  list) {
2103                         tipc_node_lock(node);
2104                         err = __tipc_nl_add_node_links(net, &msg, node,
2105                                                        &prev_link);
2106                         tipc_node_unlock(node);
2107                         if (err)
2108                                 goto out;
2109
2110                         prev_node = node->addr;
2111                 }
2112         } else {
2113                 err = tipc_nl_add_bc_link(net, &msg);
2114                 if (err)
2115                         goto out;
2116
2117                 list_for_each_entry_rcu(node, &tn->node_list, list) {
2118                         tipc_node_lock(node);
2119                         err = __tipc_nl_add_node_links(net, &msg, node,
2120                                                        &prev_link);
2121                         tipc_node_unlock(node);
2122                         if (err)
2123                                 goto out;
2124
2125                         prev_node = node->addr;
2126                 }
2127         }
2128         done = 1;
2129 out:
2130         rcu_read_unlock();
2131
2132         cb->args[0] = prev_node;
2133         cb->args[1] = prev_link;
2134         cb->args[2] = done;
2135
2136         return skb->len;
2137 }
2138
2139 int tipc_nl_link_get(struct sk_buff *skb, struct genl_info *info)
2140 {
2141         struct net *net = genl_info_net(info);
2142         struct tipc_nl_msg msg;
2143         char *name;
2144         int err;
2145
2146         msg.portid = info->snd_portid;
2147         msg.seq = info->snd_seq;
2148
2149         if (!info->attrs[TIPC_NLA_LINK_NAME])
2150                 return -EINVAL;
2151         name = nla_data(info->attrs[TIPC_NLA_LINK_NAME]);
2152
2153         msg.skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
2154         if (!msg.skb)
2155                 return -ENOMEM;
2156
2157         if (strcmp(name, tipc_bclink_name) == 0) {
2158                 err = tipc_nl_add_bc_link(net, &msg);
2159                 if (err) {
2160                         nlmsg_free(msg.skb);
2161                         return err;
2162                 }
2163         } else {
2164                 int bearer_id;
2165                 struct tipc_node *node;
2166                 struct tipc_link *link;
2167
2168                 node = tipc_link_find_owner(net, name, &bearer_id);
2169                 if (!node)
2170                         return -EINVAL;
2171
2172                 tipc_node_lock(node);
2173                 link = node->links[bearer_id].link;
2174                 if (!link) {
2175                         tipc_node_unlock(node);
2176                         nlmsg_free(msg.skb);
2177                         return -EINVAL;
2178                 }
2179
2180                 err = __tipc_nl_add_link(net, &msg, link, 0);
2181                 tipc_node_unlock(node);
2182                 if (err) {
2183                         nlmsg_free(msg.skb);
2184                         return err;
2185                 }
2186         }
2187
2188         return genlmsg_reply(msg.skb, info);
2189 }
2190
2191 int tipc_nl_link_reset_stats(struct sk_buff *skb, struct genl_info *info)
2192 {
2193         int err;
2194         char *link_name;
2195         unsigned int bearer_id;
2196         struct tipc_link *link;
2197         struct tipc_node *node;
2198         struct nlattr *attrs[TIPC_NLA_LINK_MAX + 1];
2199         struct net *net = sock_net(skb->sk);
2200
2201         if (!info->attrs[TIPC_NLA_LINK])
2202                 return -EINVAL;
2203
2204         err = nla_parse_nested(attrs, TIPC_NLA_LINK_MAX,
2205                                info->attrs[TIPC_NLA_LINK],
2206                                tipc_nl_link_policy);
2207         if (err)
2208                 return err;
2209
2210         if (!attrs[TIPC_NLA_LINK_NAME])
2211                 return -EINVAL;
2212
2213         link_name = nla_data(attrs[TIPC_NLA_LINK_NAME]);
2214
2215         if (strcmp(link_name, tipc_bclink_name) == 0) {
2216                 err = tipc_bclink_reset_stats(net);
2217                 if (err)
2218                         return err;
2219                 return 0;
2220         }
2221
2222         node = tipc_link_find_owner(net, link_name, &bearer_id);
2223         if (!node)
2224                 return -EINVAL;
2225
2226         tipc_node_lock(node);
2227
2228         link = node->links[bearer_id].link;
2229         if (!link) {
2230                 tipc_node_unlock(node);
2231                 return -EINVAL;
2232         }
2233
2234         link_reset_statistics(link);
2235
2236         tipc_node_unlock(node);
2237
2238         return 0;
2239 }