aboutsummaryrefslogtreecommitdiffstats
path: root/fs/dlm/lowcomms-tcp.c
blob: 7289e59b4bd37d2b153f8856e34a3ef65949883d (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
/******************************************************************************
*******************************************************************************
**
**  Copyright (C) Sistina Software, Inc.  1997-2003  All rights reserved.
**  Copyright (C) 2004-2006 Red Hat, Inc.  All rights reserved.
**
**  This copyrighted material is made available to anyone wishing to use,
**  modify, copy, or redistribute it subject to the terms and conditions
**  of the GNU General Public License v.2.
**
*******************************************************************************
******************************************************************************/

/*
 * lowcomms.c
 *
 * This is the "low-level" comms layer.
 *
 * It is responsible for sending/receiving messages
 * from other nodes in the cluster.
 *
 * Cluster nodes are referred to by their nodeids. nodeids are
 * simply 32 bit numbers to the locking module - if they need to
 * be expanded for the cluster infrastructure then that is it's
 * responsibility. It is this layer's
 * responsibility to resolve these into IP address or
 * whatever it needs for inter-node communication.
 *
 * The comms level is two kernel threads that deal mainly with
 * the receiving of messages from other nodes and passing them
 * up to the mid-level comms layer (which understands the
 * message format) for execution by the locking core, and
 * a send thread which does all the setting up of connections
 * to remote nodes and the sending of data. Threads are not allowed
 * to send their own data because it may cause them to wait in times
 * of high load. Also, this way, the sending thread can collect together
 * messages bound for one node and send them in one block.
 *
 * I don't see any problem with the recv thread executing the locking
 * code on behalf of remote processes as the locking code is
 * short, efficient and never waits.
 *
 */


#include <asm/ioctls.h>
#include <net/sock.h>
#include <net/tcp.h>
#include <linux/pagemap.h>

#include "dlm_internal.h"
#include "lowcomms.h"
#include "midcomms.h"
#include "config.h"

struct cbuf {
	unsigned base;
	unsigned len;
	unsigned mask;
};

#ifndef FALSE
#define FALSE 0
#define TRUE 1
#endif
#define NODE_INCREMENT 32

#define CBUF_INIT(cb, size) do { (cb)->base = (cb)->len = 0; (cb)->mask = ((size)-1); } while(0)
#define CBUF_ADD(cb, n) do { (cb)->len += n; } while(0)
#define CBUF_EMPTY(cb) ((cb)->len == 0)
#define CBUF_MAY_ADD(cb, n) (((cb)->len + (n)) < ((cb)->mask + 1))
#define CBUF_EAT(cb, n) do { (cb)->len  -= (n); \
                             (cb)->base += (n); (cb)->base &= (cb)->mask; } while(0)
#define CBUF_DATA(cb) (((cb)->base + (cb)->len) & (cb)->mask)

/* Maximum number of incoming messages to process before
   doing a schedule()
*/
#define MAX_RX_MSG_COUNT 25

struct connection {
	struct socket *sock;	/* NULL if not connected */
	uint32_t nodeid;	/* So we know who we are in the list */
	struct rw_semaphore sock_sem;	/* Stop connect races */
	struct list_head read_list;	/* On this list when ready for reading */
	struct list_head write_list;	/* On this list when ready for writing */
	struct list_head state_list;	/* On this list when ready to connect */
	unsigned long flags;	/* bit 1,2 = We are on the read/write lists */
#define CF_READ_PENDING 1
#define CF_WRITE_PENDING 2
#define CF_CONNECT_PENDING 3
#define CF_IS_OTHERCON 4
	struct list_head writequeue;	/* List of outgoing writequeue_entries */
	struct list_head listenlist;    /* List of allocated listening sockets */
	spinlock_t writequeue_lock;
	int (*rx_action) (struct connection *);	/* What to do when active */
	struct page *rx_page;
	struct cbuf cb;
	int retries;
	atomic_t waiting_requests;
#define MAX_CONNECT_RETRIES 3
	struct connection *othercon;
};
#define sock2con(x) ((struct connection *)(x)->sk_user_data)

/* An entry waiting to be sent */
struct writequeue_entry {
	struct list_head list;
	struct page *page;
	int offset;
	int len;
	int end;
	int users;
	struct connection *con;
};

static struct sockaddr_storage dlm_local_addr;

/* Manage daemons */
static struct task_struct *recv_task;
static struct task_struct *send_task;

static wait_queue_t lowcomms_send_waitq_head;
static wait_queue_head_t lowcomms_send_waitq;
static wait_queue_t lowcomms_recv_waitq_head;
static wait_queue_head_t lowcomms_recv_waitq;

/* An array of pointers to connections, indexed by NODEID */
static struct connection **connections;
static struct semaphore connections_lock;
static kmem_cache_t *con_cache;
static int conn_array_size;
static atomic_t accepting;

/* List of sockets that have reads pending */
static struct list_head read_sockets;
static spinlock_t read_sockets_lock;

/* List of sockets which have writes pending */
static struct list_head write_sockets;
static spinlock_t write_sockets_lock;

/* List of sockets which have connects pending */
static struct list_head state_sockets;
static spinlock_t state_sockets_lock;

static struct connection *nodeid2con(int nodeid, gfp_t allocation)
{
	struct connection *con = NULL;

	down(&connections_lock);
	if (nodeid >= conn_array_size) {
		int new_size = nodeid + NODE_INCREMENT;
		struct connection **new_conns;

		new_conns = kmalloc(sizeof(struct connection *) *
				    new_size, allocation);
		if (!new_conns)
			goto finish;

		memset(new_conns, 0, sizeof(struct connection *) * new_size);
		memcpy(new_conns, connections,  sizeof(struct connection *) * conn_array_size);
		conn_array_size = new_size;
		kfree(connections);
		connections = new_conns;

	}

	con = connections[nodeid];
	if (con == NULL && allocation) {
		con = kmem_cache_alloc(con_cache, allocation);
		if (!con)
			goto finish;

		memset(con, 0, sizeof(*con));
		con->nodeid = nodeid;
		init_rwsem(&con->sock_sem);
		INIT_LIST_HEAD(&con->writequeue);
		spin_lock_init(&con->writequeue_lock);

		connections[nodeid] = con;
	}

 finish:
	up(&connections_lock);
	return con;
}

/* Data available on socket or listen socket received a connect */
static void lowcomms_data_ready(struct sock *sk, int count_unused)
{
	struct connection *con = sock2con(sk);

	atomic_inc(&con->waiting_requests);
	if (test_and_set_bit(CF_READ_PENDING, &con->flags))
		return;

	spin_lock_bh(&read_sockets_lock);
	list_add_tail(&con->read_list, &read_sockets);
	spin_unlock_bh(&read_sockets_lock);

	wake_up_interruptible(&lowcomms_recv_waitq);
}

static void lowcomms_write_space(struct sock *sk)
{
	struct connection *con = sock2con(sk);

	if (test_and_set_bit(CF_WRITE_PENDING, &con->flags))
		return;

	spin_lock_bh(&write_sockets_lock);
	list_add_tail(&con->write_list, &write_sockets);
	spin_unlock_bh(&write_sockets_lock);

	wake_up_interruptible(&lowcomms_send_waitq);
}

static inline void lowcomms_connect_sock(struct connection *con)
{
	if (test_and_set_bit(CF_CONNECT_PENDING, &con->flags))
		return;
	if (!atomic_read(&accepting))
		return;

	spin_lock_bh(&state_sockets_lock);
	list_add_tail(&con->state_list, &state_sockets);
	spin_unlock_bh(&state_sockets_lock);

	wake_up_interruptible(&lowcomms_send_waitq);
}

static void lowcomms_state_change(struct sock *sk)
{
/*	struct connection *con = sock2con(sk); */

	switch (sk->sk_state) {
	case TCP_ESTABLISHED:
		lowcomms_write_space(sk);
		break;

	case TCP_FIN_WAIT1:
	case TCP_FIN_WAIT2:
	case TCP_TIME_WAIT:
	case TCP_CLOSE:
	case TCP_CLOSE_WAIT:
	case TCP_LAST_ACK:
	case TCP_CLOSING:
		/* FIXME: I think this causes more trouble than it solves.
		   lowcomms wil reconnect anyway when there is something to
		   send. This just attempts reconnection if a node goes down!
		*/
		/* lowcomms_connect_sock(con); */
		break;

	default:
		printk("dlm: lowcomms_state_change: state=%d\n", sk->sk_state);
		break;
	}
}

/* Make a socket active */
static int add_sock(struct socket *sock, struct connection *con)
{
	con->sock = sock;

	/* Install a data_ready callback */
	con->sock->sk->sk_data_ready = lowcomms_data_ready;
	con->sock->sk->sk_write_space = lowcomms_write_space;
	con->sock->sk->sk_state_change = lowcomms_state_change;

	return 0;
}

/* Add the port number to an IP6 or 4 sockaddr and return the address
   length */
static void make_sockaddr(struct sockaddr_storage *saddr, uint16_t port,
			  int *addr_len)
{
        saddr->ss_family =  dlm_local_addr.ss_family;
        if (saddr->ss_family == AF_INET) {
		struct sockaddr_in *in4_addr = (struct sockaddr_in *)saddr;
		in4_addr->sin_port = cpu_to_be16(port);
		*addr_len = sizeof(struct sockaddr_in);
	}
	else {
		struct sockaddr_in6 *in6_addr = (struct sockaddr_in6 *)saddr;
		in6_addr->sin6_port = cpu_to_be16(port);
		*addr_len = sizeof(struct sockaddr_in6);
	}
}

/* Close a remote connection and tidy up */
static void close_connection(struct connection *con, int and_other)
{
	down_write(&con->sock_sem);

	if (con->sock) {
		sock_release(con->sock);
		con->sock = NULL;
	}
	if (con->othercon && and_other) {
		/* Argh! recursion in kernel code!
		   Actually, this isn't a list so it
		   will only re-enter once.
		*/
		close_connection(con->othercon, FALSE);
	}
	if (con->rx_page) {
		__free_page(con->rx_page);
		con->rx_page = NULL;
	}
	con->retries = 0;
	up_write(&con->sock_sem);
}

/* Data received from remote end */
static int receive_from_sock(struct connection *con)
{
	int ret = 0;
	struct msghdr msg;
	struct iovec iov[2];
	mm_segment_t fs;
	unsigned len;
	int r;
	int call_again_soon = 0;

	down_read(&con->sock_sem);

	if (con->sock == NULL)
		goto out;
	if (con->rx_page == NULL) {
		/*
		 * This doesn't need to be atomic, but I think it should
		 * improve performance if it is.
		 */
		con->rx_page = alloc_page(GFP_ATOMIC);
		if (con->rx_page == NULL)
			goto out_resched;
		CBUF_INIT(&con->cb, PAGE_CACHE_SIZE);
	}

	msg.msg_control = NULL;
	msg.msg_controllen = 0;
	msg.msg_iovlen = 1;
	msg.msg_iov = iov;
	msg.msg_name = NULL;
	msg.msg_namelen = 0;
	msg.msg_flags = 0;

	/*
	 * iov[0] is the bit of the circular buffer between the current end
	 * point (cb.base + cb.len) and the end of the buffer.
	 */
	iov[0].iov_len = con->cb.base - CBUF_DATA(&con->cb);
	iov[0].iov_base = page_address(con->rx_page) + CBUF_DATA(&con->cb);
	iov[1].iov_len = 0;

	/*
	 * iov[1] is the bit of the circular buffer between the start of the
	 * buffer and the start of the currently used section (cb.base)
	 */
	if (CBUF_DATA(&con->cb) >= con->cb.base) {
		iov[0].iov_len = PAGE_CACHE_SIZE - CBUF_DATA(&con->cb);
		iov[1].iov_len = con->cb.base;
		iov[1].iov_base = page_address(con->rx_page);
		msg.msg_iovlen = 2;
	}
	len = iov[0].iov_len + iov[1].iov_len;

	fs = get_fs();
	set_fs(get_ds());
	r = ret = sock_recvmsg(con->sock, &msg, len,
			       MSG_DONTWAIT | MSG_NOSIGNAL);
	set_fs(fs);

	if (ret <= 0)
		goto out_close;
	if (ret == len)
		call_again_soon = 1;
	CBUF_ADD(&con->cb, ret);
	ret = dlm_process_incoming_buffer(con->nodeid,
					  page_address(con->rx_page),
					  con->cb.base, con->cb.len,
					  PAGE_CACHE_SIZE);
	if (ret == -EBADMSG) {
		printk(KERN_INFO "dlm: lowcomms: addr=%p, base=%u, len=%u, "
		       "iov_len=%u, iov_base[0]=%p, read=%d\n",
		       page_address(con->rx_page), con->cb.base, con->cb.len,
		       len, iov[0].iov_base, r);
	}
	if (ret < 0)
		goto out_close;
	CBUF_EAT(&con->cb, ret);

	if (CBUF_EMPTY(&con->cb) && !call_again_soon) {
		__free_page(con->rx_page);
		con->rx_page = NULL;
	}

      out:
	if (call_again_soon)
		goto out_resched;
	up_read(&con->sock_sem);
	ret = 0;
	goto out_ret;

      out_resched:
	lowcomms_data_ready(con->sock->sk, 0);
	up_read(&con->sock_sem);
	ret = 0;
	schedule();
	goto out_ret;

      out_close:
	up_read(&con->sock_sem);
	if (ret != -EAGAIN && !test_bit(CF_IS_OTHERCON, &con->flags)) {
		close_connection(con, FALSE);
		/* Reconnect when there is something to send */
	}

      out_ret:
	return ret;
}

/* Listening socket is busy, accept a connection */
static int accept_from_sock(struct connection *con)
{
	int result;
	struct sockaddr_storage peeraddr;
	struct socket *newsock;
	int len;
	int nodeid;
	struct connection *newcon;

	memset(&peeraddr, 0, sizeof(peeraddr));
	result = sock_create_kern(dlm_local_addr.ss_family, SOCK_STREAM, IPPROTO_TCP, &newsock);
	if (result < 0)
		return -ENOMEM;

	down_read(&con->sock_sem);

	result = -ENOTCONN;
	if (con->sock == NULL)
		goto accept_err;

	newsock->type = con->sock->type;
	newsock->ops = con->sock->ops;

	result = con->sock->ops->accept(con->sock, newsock, O_NONBLOCK);
	if (result < 0)
		goto accept_err;

	/* Get the connected socket's peer */
	memset(&peeraddr, 0, sizeof(peeraddr));
	if (newsock->ops->getname(newsock, (struct sockaddr *)&peeraddr,
				  &len, 2)) {
		result = -ECONNABORTED;
		goto accept_err;
	}

	/* Get the new node's NODEID */
	make_sockaddr(&peeraddr, 0, &len);
	if (dlm_addr_to_nodeid(&peeraddr, &nodeid)) {
	    	printk("dlm: connect from non cluster node\n");
		sock_release(newsock);
		up_read(&con->sock_sem);
		return -1;
	}

	log_print("got connection from %d", nodeid);

	/*  Check to see if we already have a connection to this node. This
	 *  could happen if the two nodes initiate a connection at roughly
	 *  the same time and the connections cross on the wire.
	 * TEMPORARY FIX:
	 *  In this case we store the incoming one in "othercon"
	 */
	newcon = nodeid2con(nodeid, GFP_KERNEL);
	if (!newcon) {
		result = -ENOMEM;
		goto accept_err;
	}
	down_write(&newcon->sock_sem);
	if (newcon->sock) {
	        struct connection *othercon = newcon->othercon;

		if (!othercon) {
			othercon = kmem_cache_alloc(con_cache, GFP_KERNEL);
			if (!othercon) {
				printk("dlm: failed to allocate incoming socket\n");
				up_write(&newcon->sock_sem);
				result = -ENOMEM;
				goto accept_err;
			}
			memset(othercon, 0, sizeof(*othercon));
			othercon->nodeid = nodeid;
			othercon->rx_action = receive_from_sock;
			init_rwsem(&othercon->sock_sem);
			set_bit(CF_IS_OTHERCON, &othercon->flags);
			newcon->othercon = othercon;
		}
		othercon->sock = newsock;
		newsock->sk->sk_user_data = othercon;
		add_sock(newsock, othercon);
	}
	else {
		newsock->sk->sk_user_data = newcon;
		newcon->rx_action = receive_from_sock;
		add_sock(newsock, newcon);

	}

	up_write(&newcon->sock_sem);

	/*
	 * Add it to the active queue in case we got data
	 * beween processing the accept adding the socket
	 * to the read_sockets list
	 */
	lowcomms_data_ready(newsock->sk, 0);
	up_read(&con->sock_sem);

	return 0;

      accept_err:
	up_read(&con->sock_sem);
	sock_release(newsock);

	if (result != -EAGAIN)
		printk("dlm: error accepting connection from node: %d\n", result);
	return result;
}

/* Connect a new socket to its peer */
static int connect_to_sock(struct connection *con)
{
	int result = -EHOSTUNREACH;
	struct sockaddr_storage saddr;
	int addr_len;
	struct socket *sock;

	if (con->nodeid == 0) {
		log_print("attempt to connect sock 0 foiled");
		return 0;
	}

	down_write(&con->sock_sem);
	if (con->retries++ > MAX_CONNECT_RETRIES)
		goto out;

	/* Some odd races can cause double-connects, ignore them */
	if (con->sock) {
		result = 0;
		goto out;
	}

	/* Create a socket to communicate with */
	result = sock_create_kern(dlm_local_addr.ss_family, SOCK_STREAM, IPPROTO_TCP, &sock);
	if (result < 0)
		goto out_err;

	memset(&saddr, 0, sizeof(saddr));
	if (dlm_nodeid_to_addr(con->nodeid, &saddr))
	        goto out_err;

	sock->sk->sk_user_data = con;
	con->rx_action = receive_from_sock;

	make_sockaddr(&saddr, dlm_config.tcp_port, &addr_len);

	add_sock(sock, con);

	log_print("connecting to %d", con->nodeid);
	result =
		sock->ops->connect(sock, (struct sockaddr *)&saddr, addr_len,
			       O_NONBLOCK);
	if (result == -EINPROGRESS)
		result = 0;
	if (result != 0)
		goto out_err;

      out:
	up_write(&con->sock_sem);
	/*
	 * Returning an error here means we've given up trying to connect to
	 * a remote node, otherwise we return 0 and reschedule the connetion
	 * attempt
	 */
	return result;

      out_err:
	if (con->sock) {
		sock_release(con->sock);
		con->sock = NULL;
	}
	/*
	 * Some errors are fatal and this list might need adjusting. For other
	 * errors we try again until the max number of retries is reached.
	 */
	if (result != -EHOSTUNREACH && result != -ENETUNREACH &&
	    result != -ENETDOWN && result != EINVAL
	    && result != -EPROTONOSUPPORT) {
		lowcomms_connect_sock(con);
		result = 0;
	}
	goto out;
}

static struct socket *create_listen_sock(struct connection *con, struct sockaddr_storage *saddr)
{
        struct socket *sock = NULL;
	mm_segment_t fs;
	int result = 0;
	int one = 1;
	int addr_len;

	if (dlm_local_addr.ss_family == AF_INET)
		addr_len = sizeof(struct sockaddr_in);
	else
		addr_len = sizeof(struct sockaddr_in6);

	/* Create a socket to communicate with */
	result = sock_create_kern(dlm_local_addr.ss_family, SOCK_STREAM, IPPROTO_TCP, &sock);
	if (result < 0) {
		printk("dlm: Can't create listening comms socket\n");
		goto create_out;
	}

	fs = get_fs();
	set_fs(get_ds());
	result = sock_setsockopt(sock, SOL_SOCKET, SO_REUSEADDR, (char *)&one, sizeof(one));
	set_fs(fs);
	if (result < 0) {
		printk("dlm: Failed to set SO_REUSEADDR on socket: result=%d\n",result);
	}
	sock->sk->sk_user_data = con;
	con->rx_action = accept_from_sock;
	con->sock = sock;

	/* Bind to our port */
	make_sockaddr(saddr, dlm_config.tcp_port, &addr_len);
	result = sock->ops->bind(sock, (struct sockaddr *) saddr, addr_len);
	if (result < 0) {
		printk("dlm: Can't bind to port %d\n", dlm_config.tcp_port);
		sock_release(sock);
		sock = NULL;
		con->sock = NULL;
		goto create_out;
	}

	fs = get_fs();
	set_fs(get_ds());

	result = sock_setsockopt(sock, SOL_SOCKET, SO_KEEPALIVE, (char *)&one, sizeof(one));
	set_fs(fs);
	if (result < 0) {
		printk("dlm: Set keepalive failed: %d\n", result);
	}

	result = sock->ops->listen(sock, 5);
	if (result < 0) {
		printk("dlm: Can't listen on port %d\n", dlm_config.tcp_port);
		sock_release(sock);
		sock = NULL;
		goto create_out;
	}

      create_out:
	return sock;
}


/* Listen on all interfaces */
static int listen_for_all(void)
{
	struct socket *sock = NULL;
	struct connection *con = nodeid2con(0, GFP_KERNEL);
	int result = -EINVAL;

	/* We don't support multi-homed hosts */
	memset(con, 0, sizeof(*con));
	init_rwsem(&con->sock_sem);
	spin_lock_init(&con->writequeue_lock);
	INIT_LIST_HEAD(&con->writequeue);
	set_bit(CF_IS_OTHERCON, &con->flags);

	sock = create_listen_sock(con, &dlm_local_addr);
	if (sock) {
		add_sock(sock, con);
		result = 0;
	}
	else {
		result = -EADDRINUSE;
	}

	return result;
}



static struct writequeue_entry *new_writequeue_entry(struct connection *con,
						     gfp_t allocation)
{
	struct writequeue_entry *entry;

	entry = kmalloc(sizeof(struct writequeue_entry), allocation);
	if (!entry)
		return NULL;

	entry->page = alloc_page(allocation);
	if (!entry->page) {
		kfree(entry);
		return NULL;
	}

	entry->offset = 0;
	entry->len = 0;
	entry->end = 0;
	entry->users = 0;
	entry->con = con;

	return entry;
}

void *dlm_lowcomms_get_buffer(int nodeid, int len,
			      gfp_t allocation, char **ppc)
{
	struct connection *con;
	struct writequeue_entry *e;
	int offset = 0;
	int users = 0;

	if (!atomic_read(&accepting))
		return NULL;

	con = nodeid2con(nodeid, allocation);
	if (!con)
		return NULL;

	spin_lock(&con->writequeue_lock);
	e = list_entry(con->writequeue.prev, struct writequeue_entry, list);
	if (((struct list_head *) e == &con->writequeue) ||
	    (PAGE_CACHE_SIZE - e->end < len)) {
		e = NULL;
	} else {
		offset = e->end;
		e->end += len;
		users = e->users++;
	}
	spin_unlock(&con->writequeue_lock);

	if (e) {
	      got_one:
		if (users == 0)
			kmap(e->page);
		*ppc = page_address(e->page) + offset;
		return e;
	}

	e = new_writequeue_entry(con, allocation);
	if (e) {
		spin_lock(&con->writequeue_lock);
		offset = e->end;
		e->end += len;
		users = e->users++;
		list_add_tail(&e->list, &con->writequeue);
		spin_unlock(&con->writequeue_lock);
		goto got_one;
	}
	return NULL;
}

void dlm_lowcomms_commit_buffer(void *mh)
{
	struct writequeue_entry *e = (struct writequeue_entry *)mh;
	struct connection *con = e->con;
	int users;

	if (!atomic_read(&accepting))
		return;

	spin_lock(&con->writequeue_lock);
	users = --e->users;
	if (users)
		goto out;
	e->len = e->end - e->offset;
	kunmap(e->page);
	spin_unlock(&con->writequeue_lock);

	if (test_and_set_bit(CF_WRITE_PENDING, &con->flags) == 0) {
		spin_lock_bh(&write_sockets_lock);
		list_add_tail(&con->write_list, &write_sockets);
		spin_unlock_bh(&write_sockets_lock);

		wake_up_interruptible(&lowcomms_send_waitq);
	}
	return;

      out:
	spin_unlock(&con->writequeue_lock);
	return;
}

static void free_entry(struct writequeue_entry *e)
{
	__free_page(e->page);
	kfree(e);
}

/* Send a message */
static int send_to_sock(struct connection *con)
{
	int ret = 0;
	ssize_t(*sendpage) (struct socket *, struct page *, int, size_t, int);
	const int msg_flags = MSG_DONTWAIT | MSG_NOSIGNAL;
	struct writequeue_entry *e;
	int len, offset;

	down_read(&con->sock_sem);
	if (con->sock == NULL)
		goto out_connect;

	sendpage = con->sock->ops->sendpage;

	spin_lock(&con->writequeue_lock);
	for (;;) {
		e = list_entry(con->writequeue.next, struct writequeue_entry,
			       list);
		if ((struct list_head *) e == &con->writequeue)
			break;

		len = e->len;
		offset = e->offset;
		BUG_ON(len == 0 && e->users == 0);
		spin_unlock(&con->writequeue_lock);

		ret = 0;
		if (len) {
			ret = sendpage(con->sock, e->page, offset, len,
				       msg_flags);
			if (ret == -EAGAIN || ret == 0)
				goto out;
			if (ret <= 0)
				goto send_error;
		}
		else {
			/* Don't starve people filling buffers */
			schedule();
		}

		spin_lock(&con->writequeue_lock);
		e->offset += ret;
		e->len -= ret;

		if (e->len == 0 && e->users == 0) {
			list_del(&e->list);
			free_entry(e);
			continue;
		}
	}
	spin_unlock(&con->writequeue_lock);
      out:
	up_read(&con->sock_sem);
	return ret;

      send_error:
	up_read(&con->sock_sem);
	close_connection(con, FALSE);
	lowcomms_connect_sock(con);
	return ret;

      out_connect:
	up_read(&con->sock_sem);
	lowcomms_connect_sock(con);
	return 0;
}

static void clean_one_writequeue(struct connection *con)
{
	struct list_head *list;
	struct list_head *temp;

	spin_lock(&con->writequeue_lock);
	list_for_each_safe(list, temp, &con->writequeue) {
		struct writequeue_entry *e =
			list_entry(list, struct writequeue_entry, list);
		list_del(&e->list);
		free_entry(e);
	}
	spin_unlock(&con->writequeue_lock);
}

/* Called from recovery when it knows that a node has
   left the cluster */
int dlm_lowcomms_close(int nodeid)
{
	struct connection *con;

	if (!connections)
		goto out;

	log_print("closing connection to node %d", nodeid);
	con = nodeid2con(nodeid, 0);
	if (con) {
		clean_one_writequeue(con);
		close_connection(con, TRUE);
		atomic_set(&con->waiting_requests, 0);
	}
	return 0;

      out:
	return -1;
}

/* API send message call, may queue the request */
/* N.B. This is the old interface - use the new one for new calls */
int lowcomms_send_message(int nodeid, char *buf, int len, gfp_t allocation)
{
	struct writequeue_entry *e;
	char *b;

	e = dlm_lowcomms_get_buffer(nodeid, len, allocation, &b);
	if (e) {
		memcpy(b, buf, len);
		dlm_lowcomms_commit_buffer(e);
		return 0;
	}
	return -ENOBUFS;
}

/* Look for activity on active sockets */
static void process_sockets(void)
{
	struct list_head *list;
	struct list_head *temp;
	int count = 0;

	spin_lock_bh(&read_sockets_lock);
	list_for_each_safe(list, temp, &read_sockets) {

		struct connection *con =
		    list_entry(list, struct connection, read_list);
		list_del(&con->read_list);
		clear_bit(CF_READ_PENDING, &con->flags);

		spin_unlock_bh(&read_sockets_lock);

		/* This can reach zero if we are processing requests
		 * as they come in.
		 */
		if (atomic_read(&con->waiting_requests) == 0) {
			spin_lock_bh(&read_sockets_lock);
			continue;
		}

		do {
			con->rx_action(con);

			/* Don't starve out everyone else */
			if (++count >= MAX_RX_MSG_COUNT) {
				schedule();
				count = 0;
			}

		} while (!atomic_dec_and_test(&con->waiting_requests) &&
			 !kthread_should_stop());

		spin_lock_bh(&read_sockets_lock);
	}
	spin_unlock_bh(&read_sockets_lock);
}

/* Try to send any messages that are pending
 */
static void process_output_queue(void)
{
	struct list_head *list;
	struct list_head *temp;
	int ret;

	spin_lock_bh(&write_sockets_lock);
	list_for_each_safe(list, temp, &write_sockets) {
		struct connection *con =
		    list_entry(list, struct connection, write_list);
		clear_bit(CF_WRITE_PENDING, &con->flags);
		list_del(&con->write_list);

		spin_unlock_bh(&write_sockets_lock);

		ret = send_to_sock(con);
		if (ret < 0) {
		}
		spin_lock_bh(&write_sockets_lock);
	}
	spin_unlock_bh(&write_sockets_lock);
}

static void process_state_queue(void)
{
	struct list_head *list;
	struct list_head *temp;
	int ret;

	spin_lock_bh(&state_sockets_lock);
	list_for_each_safe(list, temp, &state_sockets) {
		struct connection *con =
		    list_entry(list, struct connection, state_list);
		list_del(&con->state_list);
		clear_bit(CF_CONNECT_PENDING, &con->flags);
		spin_unlock_bh(&state_sockets_lock);

		ret = connect_to_sock(con);
		if (ret < 0) {
		}
		spin_lock_bh(&state_sockets_lock);
	}
	spin_unlock_bh(&state_sockets_lock);
}


/* Discard all entries on the write queues */
static void clean_writequeues(void)
{
	int nodeid;

	for (nodeid = 1; nodeid < conn_array_size; nodeid++) {
		struct connection *con = nodeid2con(nodeid, 0);

		if (con)
			clean_one_writequeue(con);
	}
}

static int read_list_empty(void)
{
	int status;

	spin_lock_bh(&read_sockets_lock);
	status = list_empty(&read_sockets);
	spin_unlock_bh(&read_sockets_lock);

	return status;
}

/* DLM Transport comms receive daemon */
static int dlm_recvd(void *data)
{
	init_waitqueue_head(&lowcomms_recv_waitq);
	init_waitqueue_entry(&lowcomms_recv_waitq_head, current);
	add_wait_queue(&lowcomms_recv_waitq, &lowcomms_recv_waitq_head);

	while (!kthread_should_stop()) {
		set_current_state(TASK_INTERRUPTIBLE);
		if (read_list_empty())
			schedule();
		set_current_state(TASK_RUNNING);

		process_sockets();
	}

	return 0;
}

static int write_and_state_lists_empty(void)
{
	int status;

	spin_lock_bh(&write_sockets_lock);
	status = list_empty(&write_sockets);
	spin_unlock_bh(&write_sockets_lock);

	spin_lock_bh(&state_sockets_lock);
	if (list_empty(&state_sockets) == 0)
		status = 0;
	spin_unlock_bh(&state_sockets_lock);

	return status;
}

/* DLM Transport send daemon */
static int dlm_sendd(void *data)
{
	init_waitqueue_head(&lowcomms_send_waitq);
	init_waitqueue_entry(&lowcomms_send_waitq_head, current);
	add_wait_queue(&lowcomms_send_waitq, &lowcomms_send_waitq_head);

	while (!kthread_should_stop()) {
		set_current_state(TASK_INTERRUPTIBLE);
		if (write_and_state_lists_empty())
			schedule();
		set_current_state(TASK_RUNNING);

		process_state_queue();
		process_output_queue();
	}

	return 0;
}

static void daemons_stop(void)
{
	kthread_stop(recv_task);
	kthread_stop(send_task);
}

static int daemons_start(void)
{
	struct task_struct *p;
	int error;

	p = kthread_run(dlm_recvd, NULL, "dlm_recvd");
	error = IS_ERR(p);
       	if (error) {
		log_print("can't start dlm_recvd %d", error);
		return error;
	}
	recv_task = p;

	p = kthread_run(dlm_sendd, NULL, "dlm_sendd");
	error = IS_ERR(p);
       	if (error) {
		log_print("can't start dlm_sendd %d", error);
		kthread_stop(recv_task);
		return error;
	}
	send_task = p;

	return 0;
}

/*
 * Return the largest buffer size we can cope with.
 */
int lowcomms_max_buffer_size(void)
{
	return PAGE_CACHE_SIZE;
}

void dlm_lowcomms_stop(void)
{
	int i;

	atomic_set(&accepting, 0);

	/* Set all the activity flags to prevent any
	   socket activity.
	*/
	for (i = 0; i < conn_array_size; i++) {
		if (connections[i])
			connections[i]->flags |= 0x7;
	}
	daemons_stop();
	clean_writequeues();

	for (i = 0; i < conn_array_size; i++) {
		if (connections[i]) {
			close_connection(connections[i], TRUE);
			if (connections[i]->othercon)
				kmem_cache_free(con_cache, connections[i]->othercon);
			kmem_cache_free(con_cache, connections[i]);
		}
	}

	kfree(connections);
	connections = NULL;

	kmem_cache_destroy(con_cache);
}

/* This is quite likely to sleep... */
int dlm_lowcomms_start(void)
{
	int error = 0;

	error = -ENOTCONN;

	/*
	 * Temporarily initialise the waitq head so that lowcomms_send_message
	 * doesn't crash if it gets called before the thread is fully
	 * initialised
	 */
	init_waitqueue_head(&lowcomms_send_waitq);

	error = -ENOMEM;
	connections = kmalloc(sizeof(struct connection *) *
			      NODE_INCREMENT, GFP_KERNEL);
	if (!connections)
		goto out;

	memset(connections, 0,
	       sizeof(struct connection *) * NODE_INCREMENT);

	conn_array_size = NODE_INCREMENT;

	if (dlm_our_addr(&dlm_local_addr, 0)) {
		log_print("no local IP address has been set");
		goto fail_free_conn;
	}
	if (!dlm_our_addr(&dlm_local_addr, 1)) {
		log_print("This dlm comms module does not support multi-homed clustering");
		goto fail_free_conn;
	}

	con_cache = kmem_cache_create("dlm_conn", sizeof(struct connection),
				      __alignof__(struct connection), 0, NULL, NULL);
	if (!con_cache)
		goto fail_free_conn;


	/* Start listening */
	error = listen_for_all();
	if (error)
		goto fail_unlisten;

	error = daemons_start();
	if (error)
		goto fail_unlisten;

	atomic_set(&accepting, 1);

	return 0;

      fail_unlisten:
	close_connection(connections[0], 0);
	kmem_cache_free(con_cache, connections[0]);
	kmem_cache_destroy(con_cache);

      fail_free_conn:
	kfree(connections);

      out:
	return error;
}

int dlm_lowcomms_init(void)
{
	INIT_LIST_HEAD(&read_sockets);
	INIT_LIST_HEAD(&write_sockets);
	INIT_LIST_HEAD(&state_sockets);

	spin_lock_init(&read_sockets_lock);
	spin_lock_init(&write_sockets_lock);
	spin_lock_init(&state_sockets_lock);
	init_MUTEX(&connections_lock);

	return 0;
}

void dlm_lowcomms_exit(void)
{
}

/*
 * Overrides for Emacs so that we follow Linus's tabbing style.
 * Emacs will notice this stuff at the end of the file and automatically
 * adjust the settings for this buffer only.  This must remain at the end
 * of the file.
 * ---------------------------------------------------------------------------
 * Local variables:
 * c-file-style: "linux"
 * End:
 */