fs/dlm/lowcomms.c

   1 /******************************************************************************
   2 *******************************************************************************
   3 **
   4 **  Copyright (C) Sistina Software, Inc.  1997-2003  All rights reserved.
   5 **  Copyright (C) 2004-2009 Red Hat, Inc.  All rights reserved.
   6 **
   7 **  This copyrighted material is made available to anyone wishing to use,
   8 **  modify, copy, or redistribute it subject to the terms and conditions
   9 **  of the GNU General Public License v.2.
  10 **
  11 *******************************************************************************
  12 ******************************************************************************/
  13
  14 /*
  15  * lowcomms.c
  16  *
  17  * This is the "low-level" comms layer.
  18  *
  19  * It is responsible for sending/receiving messages
  20  * from other nodes in the cluster.
  21  *
  22  * Cluster nodes are referred to by their nodeids. nodeids are
  23  * simply 32 bit numbers to the locking module - if they need to
  24  * be expanded for the cluster infrastructure then that is its
  25  * responsibility. It is this layer's
  26  * responsibility to resolve these into IP address or
  27  * whatever it needs for inter-node communication.
  28  *
  29  * The comms level is two kernel threads that deal mainly with
  30  * the receiving of messages from other nodes and passing them
  31  * up to the mid-level comms layer (which understands the
  32  * message format) for execution by the locking core, and
  33  * a send thread which does all the setting up of connections
  34  * to remote nodes and the sending of data. Threads are not allowed
  35  * to send their own data because it may cause them to wait in times
  36  * of high load. Also, this way, the sending thread can collect together
  37  * messages bound for one node and send them in one block.
  38  *
  39  * lowcomms will choose to use either TCP or SCTP as its transport layer
  40  * depending on the configuration variable 'protocol'. This should be set
  41  * to 0 (default) for TCP or 1 for SCTP. It should be configured using a
  42  * cluster-wide mechanism as it must be the same on all nodes of the cluster
  43  * for the DLM to function.
  44  *
  45  */
  46
  47 #include <asm/ioctls.h>
  48 #include <net/sock.h>
  49 #include <net/tcp.h>
  50 #include <linux/pagemap.h>
  51 #include <linux/file.h>
  52 #include <linux/mutex.h>
  53 #include <linux/sctp.h>
  54 #include <linux/slab.h>
  55 #include <net/sctp/sctp.h>
  56 #include <net/ipv6.h>
  57
  58 #include "dlm_internal.h"
  59 #include "lowcomms.h"
  60 #include "midcomms.h"
  61 #include "config.h"
  62
  63 #define NEEDED_RMEM (4*1024*1024)
  64 #define CONN_HASH_SIZE 32
  65
  66 /* Number of messages to send before rescheduling */
  67 #define MAX_SEND_MSG_COUNT 25
  68
  69 struct cbuf {
  70         unsigned int base;
  71         unsigned int len;
  72         unsigned int mask;
  73 };
  74
  75 static void cbuf_add(struct cbuf *cb, int n)
  76 {
  77         cb->len += n;
  78 }
  79
  80 static int cbuf_data(struct cbuf *cb)
  81 {
  82         return ((cb->base + cb->len) & cb->mask);
  83 }
  84
  85 static void cbuf_init(struct cbuf *cb, int size)
  86 {
  87         cb->base = cb->len = 0;
  88         cb->mask = size-1;
  89 }
  90
  91 static void cbuf_eat(struct cbuf *cb, int n)
  92 {
  93         cb->len  -= n;
  94         cb->base += n;
  95         cb->base &= cb->mask;
  96 }
  97
  98 static bool cbuf_empty(struct cbuf *cb)
  99 {
 100         return cb->len == 0;
 101 }
 102
 103 struct connection {
 104         struct socket *sock;    /* NULL if not connected */
 105         uint32_t nodeid;        /* So we know who we are in the list */
 106         struct mutex sock_mutex;
 107         unsigned long flags;
 108 #define CF_READ_PENDING 1
 109 #define CF_WRITE_PENDING 2
 110 #define CF_CONNECT_PENDING 3
 111 #define CF_INIT_PENDING 4
 112 #define CF_IS_OTHERCON 5
 113 #define CF_CLOSE 6
 114 #define CF_APP_LIMITED 7
 115         struct list_head writequeue;  /* List of outgoing writequeue_entries */
 116         spinlock_t writequeue_lock;
 117         int (*rx_action) (struct connection *); /* What to do when active */
 118         void (*connect_action) (struct connection *);   /* What to do to connect */
 119         struct page *rx_page;
 120         struct cbuf cb;
 121         int retries;
 122 #define MAX_CONNECT_RETRIES 3
 123         int sctp_assoc;
 124         struct hlist_node list;
 125         struct connection *othercon;
 126         struct work_struct rwork; /* Receive workqueue */
 127         struct work_struct swork; /* Send workqueue */
 128         bool try_new_addr;
 129 };
 130 #define sock2con(x) ((struct connection *)(x)->sk_user_data)
 131
 132 /* An entry waiting to be sent */
 133 struct writequeue_entry {
 134         struct list_head list;
 135         struct page *page;
 136         int offset;
 137         int len;
 138         int end;
 139         int users;
 140         struct connection *con;
 141 };
 142
 143 struct dlm_node_addr {
 144         struct list_head list;
 145         int nodeid;
 146         int addr_count;
 147         int curr_addr_index;
 148         struct sockaddr_storage *addr[DLM_MAX_ADDR_COUNT];
 149 };
 150
 151 static LIST_HEAD(dlm_node_addrs);
 152 static DEFINE_SPINLOCK(dlm_node_addrs_spin);
 153
 154 static struct sockaddr_storage *dlm_local_addr[DLM_MAX_ADDR_COUNT];
 155 static int dlm_local_count;
 156 static int dlm_allow_conn;
 157
 158 /* Work queues */
 159 static struct workqueue_struct *recv_workqueue;
 160 static struct workqueue_struct *send_workqueue;
 161
 162 static struct hlist_head connection_hash[CONN_HASH_SIZE];
 163 static DEFINE_MUTEX(connections_lock);
 164 static struct kmem_cache *con_cache;
 165
 166 static void process_recv_sockets(struct work_struct *work);
 167 static void process_send_sockets(struct work_struct *work);
 168
 169
 170 /* This is deliberately very simple because most clusters have simple
 171    sequential nodeids, so we should be able to go straight to a connection
 172    struct in the array */
 173 static inline int nodeid_hash(int nodeid)
 174 {
 175         return nodeid & (CONN_HASH_SIZE-1);
 176 }
 177
 178 static struct connection *__find_con(int nodeid)
 179 {
 180         int r;
 181         struct connection *con;
 182
 183         r = nodeid_hash(nodeid);
 184
 185         hlist_for_each_entry(con, &connection_hash[r], list) {
 186                 if (con->nodeid == nodeid)
 187                         return con;
 188         }
 189         return NULL;
 190 }
 191
 192 /*
 193  * If 'allocation' is zero then we don't attempt to create a new
 194  * connection structure for this node.
 195  */
 196 static struct connection *__nodeid2con(int nodeid, gfp_t alloc)
 197 {
 198         struct connection *con = NULL;
 199         int r;
 200
 201         con = __find_con(nodeid);
 202         if (con || !alloc)
 203                 return con;
 204
 205         con = kmem_cache_zalloc(con_cache, alloc);
 206         if (!con)
 207                 return NULL;
 208
 209         r = nodeid_hash(nodeid);
 210         hlist_add_head(&con->list, &connection_hash[r]);
 211
 212         con->nodeid = nodeid;
 213         mutex_init(&con->sock_mutex);
 214         INIT_LIST_HEAD(&con->writequeue);
 215         spin_lock_init(&con->writequeue_lock);
 216         INIT_WORK(&con->swork, process_send_sockets);
 217         INIT_WORK(&con->rwork, process_recv_sockets);
 218
 219         /* Setup action pointers for child sockets */
 220         if (con->nodeid) {
 221                 struct connection *zerocon = __find_con(0);
 222
 223                 con->connect_action = zerocon->connect_action;
 224                 if (!con->rx_action)
 225                         con->rx_action = zerocon->rx_action;
 226         }
 227
 228         return con;
 229 }
 230
 231 /* Loop round all connections */
 232 static void foreach_conn(void (*conn_func)(struct connection *c))
 233 {
 234         int i;
 235         struct hlist_node *n;
 236         struct connection *con;
 237
 238         for (i = 0; i < CONN_HASH_SIZE; i++) {
 239                 hlist_for_each_entry_safe(con, n, &connection_hash[i], list)
 240                         conn_func(con);
 241         }
 242 }
 243
 244 static struct connection *nodeid2con(int nodeid, gfp_t allocation)
 245 {
 246         struct connection *con;
 247
 248         mutex_lock(&connections_lock);
 249         con = __nodeid2con(nodeid, allocation);
 250         mutex_unlock(&connections_lock);
 251
 252         return con;
 253 }
 254
 255 /* This is a bit drastic, but only called when things go wrong */
 256 static struct connection *assoc2con(int assoc_id)
 257 {
 258         int i;
 259         struct connection *con;
 260
 261         mutex_lock(&connections_lock);
 262
 263         for (i = 0 ; i < CONN_HASH_SIZE; i++) {
 264                 hlist_for_each_entry(con, &connection_hash[i], list) {
 265                         if (con->sctp_assoc == assoc_id) {
 266                                 mutex_unlock(&connections_lock);
 267                                 return con;
 268                         }
 269                 }
 270         }
 271         mutex_unlock(&connections_lock);
 272         return NULL;
 273 }
 274
 275 static struct dlm_node_addr *find_node_addr(int nodeid)
 276 {
 277         struct dlm_node_addr *na;
 278
 279         list_for_each_entry(na, &dlm_node_addrs, list) {
 280                 if (na->nodeid == nodeid)
 281                         return na;
 282         }
 283         return NULL;
 284 }
 285
 286 static int addr_compare(struct sockaddr_storage *x, struct sockaddr_storage *y)
 287 {
 288         switch (x->ss_family) {
 289         case AF_INET: {
 290                 struct sockaddr_in *sinx = (struct sockaddr_in *)x;
 291                 struct sockaddr_in *siny = (struct sockaddr_in *)y;
 292                 if (sinx->sin_addr.s_addr != siny->sin_addr.s_addr)
 293                         return 0;
 294                 if (sinx->sin_port != siny->sin_port)
 295                         return 0;
 296                 break;
 297         }
 298         case AF_INET6: {
 299                 struct sockaddr_in6 *sinx = (struct sockaddr_in6 *)x;
 300                 struct sockaddr_in6 *siny = (struct sockaddr_in6 *)y;
 301                 if (!ipv6_addr_equal(&sinx->sin6_addr, &siny->sin6_addr))
 302                         return 0;
 303                 if (sinx->sin6_port != siny->sin6_port)
 304                         return 0;
 305                 break;
 306         }
 307         default:
 308                 return 0;
 309         }
 310         return 1;
 311 }
 312
 313 static int nodeid_to_addr(int nodeid, struct sockaddr_storage *sas_out,
 314                           struct sockaddr *sa_out, bool try_new_addr)
 315 {
 316         struct sockaddr_storage sas;
 317         struct dlm_node_addr *na;
 318
 319         if (!dlm_local_count)
 320                 return -1;
 321
 322         spin_lock(&dlm_node_addrs_spin);
 323         na = find_node_addr(nodeid);
 324         if (na && na->addr_count) {
 325                 if (try_new_addr) {
 326                         na->curr_addr_index++;
 327                         if (na->curr_addr_index == na->addr_count)
 328                                 na->curr_addr_index = 0;
 329                 }
 330
 331                 memcpy(&sas, na->addr[na->curr_addr_index ],
 332                         sizeof(struct sockaddr_storage));
 333         }
 334         spin_unlock(&dlm_node_addrs_spin);
 335
 336         if (!na)
 337                 return -EEXIST;
 338
 339         if (!na->addr_count)
 340                 return -ENOENT;
 341
 342         if (sas_out)
 343                 memcpy(sas_out, &sas, sizeof(struct sockaddr_storage));
 344
 345         if (!sa_out)
 346                 return 0;
 347
 348         if (dlm_local_addr[0]->ss_family == AF_INET) {
 349                 struct sockaddr_in *in4  = (struct sockaddr_in *) &sas;
 350                 struct sockaddr_in *ret4 = (struct sockaddr_in *) sa_out;
 351                 ret4->sin_addr.s_addr = in4->sin_addr.s_addr;
 352         } else {
 353                 struct sockaddr_in6 *in6  = (struct sockaddr_in6 *) &sas;
 354                 struct sockaddr_in6 *ret6 = (struct sockaddr_in6 *) sa_out;
 355                 ret6->sin6_addr = in6->sin6_addr;
 356         }
 357
 358         return 0;
 359 }
 360
 361 static int addr_to_nodeid(struct sockaddr_storage *addr, int *nodeid)
 362 {
 363         struct dlm_node_addr *na;
 364         int rv = -EEXIST;
 365         int addr_i;
 366
 367         spin_lock(&dlm_node_addrs_spin);
 368         list_for_each_entry(na, &dlm_node_addrs, list) {
 369                 if (!na->addr_count)
 370                         continue;
 371
 372                 for (addr_i = 0; addr_i < na->addr_count; addr_i++) {
 373                         if (addr_compare(na->addr[addr_i], addr)) {
 374                                 *nodeid = na->nodeid;
 375                                 rv = 0;
 376                                 goto unlock;
 377                         }
 378                 }
 379         }
 380 unlock:
 381         spin_unlock(&dlm_node_addrs_spin);
 382         return rv;
 383 }
 384
 385 int dlm_lowcomms_addr(int nodeid, struct sockaddr_storage *addr, int len)
 386 {
 387         struct sockaddr_storage *new_addr;
 388         struct dlm_node_addr *new_node, *na;
 389
 390         new_node = kzalloc(sizeof(struct dlm_node_addr), GFP_NOFS);
 391         if (!new_node)
 392                 return -ENOMEM;
 393
 394         new_addr = kzalloc(sizeof(struct sockaddr_storage), GFP_NOFS);
 395         if (!new_addr) {
 396                 kfree(new_node);
 397                 return -ENOMEM;
 398         }
 399
 400         memcpy(new_addr, addr, len);
 401
 402         spin_lock(&dlm_node_addrs_spin);
 403         na = find_node_addr(nodeid);
 404         if (!na) {
 405                 new_node->nodeid = nodeid;
 406                 new_node->addr[0] = new_addr;
 407                 new_node->addr_count = 1;
 408                 list_add(&new_node->list, &dlm_node_addrs);
 409                 spin_unlock(&dlm_node_addrs_spin);
 410                 return 0;
 411         }
 412
 413         if (na->addr_count >= DLM_MAX_ADDR_COUNT) {
 414                 spin_unlock(&dlm_node_addrs_spin);
 415                 kfree(new_addr);
 416                 kfree(new_node);
 417                 return -ENOSPC;
 418         }
 419
 420         na->addr[na->addr_count++] = new_addr;
 421         spin_unlock(&dlm_node_addrs_spin);
 422         kfree(new_node);
 423         return 0;
 424 }
 425
 426 /* Data available on socket or listen socket received a connect */
 427 static void lowcomms_data_ready(struct sock *sk)
 428 {
 429         struct connection *con = sock2con(sk);
 430         if (con && !test_and_set_bit(CF_READ_PENDING, &con->flags))
 431                 queue_work(recv_workqueue, &con->rwork);
 432 }
 433
 434 static void lowcomms_write_space(struct sock *sk)
 435 {
 436         struct connection *con = sock2con(sk);
 437
 438         if (!con)
 439                 return;
 440
 441         clear_bit(SOCK_NOSPACE, &con->sock->flags);
 442
 443         if (test_and_clear_bit(CF_APP_LIMITED, &con->flags)) {
 444                 con->sock->sk->sk_write_pending--;
 445                 clear_bit(SOCK_ASYNC_NOSPACE, &con->sock->flags);
 446         }
 447
 448         if (!test_and_set_bit(CF_WRITE_PENDING, &con->flags))
 449                 queue_work(send_workqueue, &con->swork);
 450 }
 451
 452 static inline void lowcomms_connect_sock(struct connection *con)
 453 {
 454         if (test_bit(CF_CLOSE, &con->flags))
 455                 return;
 456         if (!test_and_set_bit(CF_CONNECT_PENDING, &con->flags))
 457                 queue_work(send_workqueue, &con->swork);
 458 }
 459
 460 static void lowcomms_state_change(struct sock *sk)
 461 {
 462         if (sk->sk_state == TCP_ESTABLISHED)
 463                 lowcomms_write_space(sk);
 464 }
 465
 466 int dlm_lowcomms_connect_node(int nodeid)
 467 {
 468         struct connection *con;
 469
 470         /* with sctp there's no connecting without sending */
 471         if (dlm_config.ci_protocol != 0)
 472                 return 0;
 473
 474         if (nodeid == dlm_our_nodeid())
 475                 return 0;
 476
 477         con = nodeid2con(nodeid, GFP_NOFS);
 478         if (!con)
 479                 return -ENOMEM;
 480         lowcomms_connect_sock(con);
 481         return 0;
 482 }
 483
 484 /* Make a socket active */
 485 static void add_sock(struct socket *sock, struct connection *con)
 486 {
 487         con->sock = sock;
 488
 489         /* Install a data_ready callback */
 490         con->sock->sk->sk_data_ready = lowcomms_data_ready;
 491         con->sock->sk->sk_write_space = lowcomms_write_space;
 492         con->sock->sk->sk_state_change = lowcomms_state_change;
 493         con->sock->sk->sk_user_data = con;
 494         con->sock->sk->sk_allocation = GFP_NOFS;
 495 }
 496
 497 /* Add the port number to an IPv6 or 4 sockaddr and return the address
 498    length */
 499 static void make_sockaddr(struct sockaddr_storage *saddr, uint16_t port,
 500                           int *addr_len)
 501 {
 502         saddr->ss_family =  dlm_local_addr[0]->ss_family;
 503         if (saddr->ss_family == AF_INET) {
 504                 struct sockaddr_in *in4_addr = (struct sockaddr_in *)saddr;
 505                 in4_addr->sin_port = cpu_to_be16(port);
 506                 *addr_len = sizeof(struct sockaddr_in);
 507                 memset(&in4_addr->sin_zero, 0, sizeof(in4_addr->sin_zero));
 508         } else {
 509                 struct sockaddr_in6 *in6_addr = (struct sockaddr_in6 *)saddr;
 510                 in6_addr->sin6_port = cpu_to_be16(port);
 511                 *addr_len = sizeof(struct sockaddr_in6);
 512         }
 513         memset((char *)saddr + *addr_len, 0, sizeof(struct sockaddr_storage) - *addr_len);
 514 }
 515
 516 /* Close a remote connection and tidy up */
 517 static void close_connection(struct connection *con, bool and_other)
 518 {
 519         mutex_lock(&con->sock_mutex);
 520
 521         if (con->sock) {
 522                 sock_release(con->sock);
 523                 con->sock = NULL;
 524         }
 525         if (con->othercon && and_other) {
 526                 /* Will only re-enter once. */
 527                 close_connection(con->othercon, false);
 528         }
 529         if (con->rx_page) {
 530                 __free_page(con->rx_page);
 531                 con->rx_page = NULL;
 532         }
 533
 534         con->retries = 0;
 535         mutex_unlock(&con->sock_mutex);
 536 }
 537
 538 /* We only send shutdown messages to nodes that are not part of the cluster */
 539 static void sctp_send_shutdown(sctp_assoc_t associd)
 540 {
 541         static char outcmsg[CMSG_SPACE(sizeof(struct sctp_sndrcvinfo))];
 542         struct msghdr outmessage;
 543         struct cmsghdr *cmsg;
 544         struct sctp_sndrcvinfo *sinfo;
 545         int ret;
 546         struct connection *con;
 547
 548         con = nodeid2con(0,0);
 549         BUG_ON(con == NULL);
 550
 551         outmessage.msg_name = NULL;
 552         outmessage.msg_namelen = 0;
 553         outmessage.msg_control = outcmsg;
 554         outmessage.msg_controllen = sizeof(outcmsg);
 555         outmessage.msg_flags = MSG_EOR;
 556
 557         cmsg = CMSG_FIRSTHDR(&outmessage);
 558         cmsg->cmsg_level = IPPROTO_SCTP;
 559         cmsg->cmsg_type = SCTP_SNDRCV;
 560         cmsg->cmsg_len = CMSG_LEN(sizeof(struct sctp_sndrcvinfo));
 561         outmessage.msg_controllen = cmsg->cmsg_len;
 562         sinfo = CMSG_DATA(cmsg);
 563         memset(sinfo, 0x00, sizeof(struct sctp_sndrcvinfo));
 564
 565         sinfo->sinfo_flags |= MSG_EOF;
 566         sinfo->sinfo_assoc_id = associd;
 567
 568         ret = kernel_sendmsg(con->sock, &outmessage, NULL, 0, 0);
 569
 570         if (ret != 0)
 571                 log_print("send EOF to node failed: %d", ret);
 572 }
 573
 574 static void sctp_init_failed_foreach(struct connection *con)
 575 {
 576
 577         /*
 578          * Don't try to recover base con and handle race where the
 579          * other node's assoc init creates a assoc and we get that
 580          * notification, then we get a notification that our attempt
 581          * failed due. This happens when we are still trying the primary
 582          * address, but the other node has already tried secondary addrs
 583          * and found one that worked.
 584          */
 585         if (!con->nodeid || con->sctp_assoc)
 586                 return;
 587
 588         log_print("Retrying SCTP association init for node %d\n", con->nodeid);
 589
 590         con->try_new_addr = true;
 591         con->sctp_assoc = 0;
 592         if (test_and_clear_bit(CF_INIT_PENDING, &con->flags)) {
 593                 if (!test_and_set_bit(CF_WRITE_PENDING, &con->flags))
 594                         queue_work(send_workqueue, &con->swork);
 595         }
 596 }
 597
 598 /* INIT failed but we don't know which node...
 599    restart INIT on all pending nodes */
 600 static void sctp_init_failed(void)
 601 {
 602         mutex_lock(&connections_lock);
 603
 604         foreach_conn(sctp_init_failed_foreach);
 605
 606         mutex_unlock(&connections_lock);
 607 }
 608
 609 static void retry_failed_sctp_send(struct connection *recv_con,
 610                                    struct sctp_send_failed *sn_send_failed,
 611                                    char *buf)
 612 {
 613         int len = sn_send_failed->ssf_length - sizeof(struct sctp_send_failed);
 614         struct dlm_mhandle *mh;
 615         struct connection *con;
 616         char *retry_buf;
 617         int nodeid = sn_send_failed->ssf_info.sinfo_ppid;
 618
 619         log_print("Retry sending %d bytes to node id %d", len, nodeid);
 620
 621         con = nodeid2con(nodeid, 0);
 622         if (!con) {
 623                 log_print("Could not look up con for nodeid %d\n",
 624                           nodeid);
 625                 return;
 626         }
 627
 628         mh = dlm_lowcomms_get_buffer(nodeid, len, GFP_NOFS, &retry_buf);
 629         if (!mh) {
 630                 log_print("Could not allocate buf for retry.");
 631                 return;
 632         }
 633         memcpy(retry_buf, buf + sizeof(struct sctp_send_failed), len);
 634         dlm_lowcomms_commit_buffer(mh);
 635
 636         /*
 637          * If we got a assoc changed event before the send failed event then
 638          * we only need to retry the send.
 639          */
 640         if (con->sctp_assoc) {
 641                 if (!test_and_set_bit(CF_WRITE_PENDING, &con->flags))
 642                         queue_work(send_workqueue, &con->swork);
 643         } else
 644                 sctp_init_failed_foreach(con);
 645 }
 646
 647 /* Something happened to an association */
 648 static void process_sctp_notification(struct connection *con,
 649                                       struct msghdr *msg, char *buf)
 650 {
 651         union sctp_notification *sn = (union sctp_notification *)buf;
 652         struct linger linger;
 653
 654         switch (sn->sn_header.sn_type) {
 655         case SCTP_SEND_FAILED:
 656                 retry_failed_sctp_send(con, &sn->sn_send_failed, buf);
 657                 break;
 658         case SCTP_ASSOC_CHANGE:
 659                 switch (sn->sn_assoc_change.sac_state) {
 660                 case SCTP_COMM_UP:
 661                 case SCTP_RESTART:
 662                 {
 663                         /* Check that the new node is in the lockspace */
 664                         struct sctp_prim prim;
 665                         int nodeid;
 666                         int prim_len, ret;
 667                         int addr_len;
 668                         struct connection *new_con;
 669
 670                         /*
 671                          * We get this before any data for an association.
 672                          * We verify that the node is in the cluster and
 673                          * then peel off a socket for it.
 674                          */
 675                         if ((int)sn->sn_assoc_change.sac_assoc_id <= 0) {
 676                                 log_print("COMM_UP for invalid assoc ID %d",
 677                                          (int)sn->sn_assoc_change.sac_assoc_id);
 678                                 sctp_init_failed();
 679                                 return;
 680                         }
 681                         memset(&prim, 0, sizeof(struct sctp_prim));
 682                         prim_len = sizeof(struct sctp_prim);
 683                         prim.ssp_assoc_id = sn->sn_assoc_change.sac_assoc_id;
 684
 685                         ret = kernel_getsockopt(con->sock,
 686                                                 IPPROTO_SCTP,
 687                                                 SCTP_PRIMARY_ADDR,
 688                                                 (char*)&prim,
 689                                                 &prim_len);
 690                         if (ret < 0) {
 691                                 log_print("getsockopt/sctp_primary_addr on "
 692                                           "new assoc %d failed : %d",
 693                                           (int)sn->sn_assoc_change.sac_assoc_id,
 694                                           ret);
 695
 696                                 /* Retry INIT later */
 697                                 new_con = assoc2con(sn->sn_assoc_change.sac_assoc_id);
 698                                 if (new_con)
 699                                         clear_bit(CF_CONNECT_PENDING, &con->flags);
 700                                 return;
 701                         }
 702                         make_sockaddr(&prim.ssp_addr, 0, &addr_len);
 703                         if (addr_to_nodeid(&prim.ssp_addr, &nodeid)) {
 704                                 unsigned char *b=(unsigned char *)&prim.ssp_addr;
 705                                 log_print("reject connect from unknown addr");
 706                                 print_hex_dump_bytes("ss: ", DUMP_PREFIX_NONE,
 707                                                      b, sizeof(struct sockaddr_storage));
 708                                 sctp_send_shutdown(prim.ssp_assoc_id);
 709                                 return;
 710                         }
 711
 712                         new_con = nodeid2con(nodeid, GFP_NOFS);
 713                         if (!new_con)
 714                                 return;
 715
 716                         /* Peel off a new sock */
 717                         lock_sock(con->sock->sk);
 718                         ret = sctp_do_peeloff(con->sock->sk,
 719                                 sn->sn_assoc_change.sac_assoc_id,
 720                                 &new_con->sock);
 721                         release_sock(con->sock->sk);
 722                         if (ret < 0) {
 723                                 log_print("Can't peel off a socket for "
 724                                           "connection %d to node %d: err=%d",
 725                                           (int)sn->sn_assoc_change.sac_assoc_id,
 726                                           nodeid, ret);
 727                                 return;
 728                         }
 729                         add_sock(new_con->sock, new_con);
 730
 731                         linger.l_onoff = 1;
 732                         linger.l_linger = 0;
 733                         ret = kernel_setsockopt(new_con->sock, SOL_SOCKET, SO_LINGER,
 734                                                 (char *)&linger, sizeof(linger));
 735                         if (ret < 0)
 736                                 log_print("set socket option SO_LINGER failed");
 737
 738                         log_print("connecting to %d sctp association %d",
 739                                  nodeid, (int)sn->sn_assoc_change.sac_assoc_id);
 740
 741                         new_con->sctp_assoc = sn->sn_assoc_change.sac_assoc_id;
 742                         new_con->try_new_addr = false;
 743                         /* Send any pending writes */
 744                         clear_bit(CF_CONNECT_PENDING, &new_con->flags);
 745                         clear_bit(CF_INIT_PENDING, &new_con->flags);
 746                         if (!test_and_set_bit(CF_WRITE_PENDING, &new_con->flags)) {
 747                                 queue_work(send_workqueue, &new_con->swork);
 748                         }
 749                         if (!test_and_set_bit(CF_READ_PENDING, &new_con->flags))
 750                                 queue_work(recv_workqueue, &new_con->rwork);
 751                 }
 752                 break;
 753
 754                 case SCTP_COMM_LOST:
 755                 case SCTP_SHUTDOWN_COMP:
 756                 {
 757                         con = assoc2con(sn->sn_assoc_change.sac_assoc_id);
 758                         if (con) {
 759                                 con->sctp_assoc = 0;
 760                         }
 761                 }
 762                 break;
 763
 764                 case SCTP_CANT_STR_ASSOC:
 765                 {
 766                         /* Will retry init when we get the send failed notification */
 767                         log_print("Can't start SCTP association - retrying");
 768                 }
 769                 break;
 770
 771                 default:
 772                         log_print("unexpected SCTP assoc change id=%d state=%d",
 773                                   (int)sn->sn_assoc_change.sac_assoc_id,
 774                                   sn->sn_assoc_change.sac_state);
 775                 }
 776         default:
 777                 ; /* fall through */
 778         }
 779 }
 780
 781 /* Data received from remote end */
 782 static int receive_from_sock(struct connection *con)
 783 {
 784         int ret = 0;
 785         struct msghdr msg = {};
 786         struct kvec iov[2];
 787         unsigned len;
 788         int r;
 789         int call_again_soon = 0;
 790         int nvec;
 791         char incmsg[CMSG_SPACE(sizeof(struct sctp_sndrcvinfo))];
 792
 793         mutex_lock(&con->sock_mutex);
 794
 795         if (con->sock == NULL) {
 796                 ret = -EAGAIN;
 797                 goto out_close;
 798         }
 799
 800         if (con->rx_page == NULL) {
 801                 /*
 802                  * This doesn't need to be atomic, but I think it should
 803                  * improve performance if it is.
 804                  */
 805                 con->rx_page = alloc_page(GFP_ATOMIC);
 806                 if (con->rx_page == NULL)
 807                         goto out_resched;
 808                 cbuf_init(&con->cb, PAGE_CACHE_SIZE);
 809         }
 810
 811         /* Only SCTP needs these really */
 812         memset(&incmsg, 0, sizeof(incmsg));
 813         msg.msg_control = incmsg;
 814         msg.msg_controllen = sizeof(incmsg);
 815
 816         /*
 817          * iov[0] is the bit of the circular buffer between the current end
 818          * point (cb.base + cb.len) and the end of the buffer.
 819          */
 820         iov[0].iov_len = con->cb.base - cbuf_data(&con->cb);
 821         iov[0].iov_base = page_address(con->rx_page) + cbuf_data(&con->cb);
 822         iov[1].iov_len = 0;
 823         nvec = 1;
 824
 825         /*
 826          * iov[1] is the bit of the circular buffer between the start of the
 827          * buffer and the start of the currently used section (cb.base)
 828          */
 829         if (cbuf_data(&con->cb) >= con->cb.base) {
 830                 iov[0].iov_len = PAGE_CACHE_SIZE - cbuf_data(&con->cb);
 831                 iov[1].iov_len = con->cb.base;
 832                 iov[1].iov_base = page_address(con->rx_page);
 833                 nvec = 2;
 834         }
 835         len = iov[0].iov_len + iov[1].iov_len;
 836
 837         r = ret = kernel_recvmsg(con->sock, &msg, iov, nvec, len,
 838                                MSG_DONTWAIT | MSG_NOSIGNAL);
 839         if (ret <= 0)
 840                 goto out_close;
 841
 842         /* Process SCTP notifications */
 843         if (msg.msg_flags & MSG_NOTIFICATION) {
 844                 msg.msg_control = incmsg;
 845                 msg.msg_controllen = sizeof(incmsg);
 846
 847                 process_sctp_notification(con, &msg,
 848                                 page_address(con->rx_page) + con->cb.base);
 849                 mutex_unlock(&con->sock_mutex);
 850                 return 0;
 851         }
 852         BUG_ON(con->nodeid == 0);
 853
 854         if (ret == len)
 855                 call_again_soon = 1;
 856         cbuf_add(&con->cb, ret);
 857         ret = dlm_process_incoming_buffer(con->nodeid,
 858                                           page_address(con->rx_page),
 859                                           con->cb.base, con->cb.len,
 860                                           PAGE_CACHE_SIZE);
 861         if (ret == -EBADMSG) {
 862                 log_print("lowcomms: addr=%p, base=%u, len=%u, "
 863                           "iov_len=%u, iov_base[0]=%p, read=%d",
 864                           page_address(con->rx_page), con->cb.base, con->cb.len,
 865                           len, iov[0].iov_base, r);
 866         }
 867         if (ret < 0)
 868                 goto out_close;
 869         cbuf_eat(&con->cb, ret);
 870
 871         if (cbuf_empty(&con->cb) && !call_again_soon) {
 872                 __free_page(con->rx_page);
 873                 con->rx_page = NULL;
 874         }
 875
 876         if (call_again_soon)
 877                 goto out_resched;
 878         mutex_unlock(&con->sock_mutex);
 879         return 0;
 880
 881 out_resched:
 882         if (!test_and_set_bit(CF_READ_PENDING, &con->flags))
 883                 queue_work(recv_workqueue, &con->rwork);
 884         mutex_unlock(&con->sock_mutex);
 885         return -EAGAIN;
 886
 887 out_close:
 888         mutex_unlock(&con->sock_mutex);
 889         if (ret != -EAGAIN) {
 890                 close_connection(con, false);
 891                 /* Reconnect when there is something to send */
 892         }
 893         /* Don't return success if we really got EOF */
 894         if (ret == 0)
 895                 ret = -EAGAIN;
 896
 897         return ret;
 898 }
 899
 900 /* Listening socket is busy, accept a connection */
 901 static int tcp_accept_from_sock(struct connection *con)
 902 {
 903         int result;
 904         struct sockaddr_storage peeraddr;
 905         struct socket *newsock;
 906         int len;
 907         int nodeid;
 908         struct connection *newcon;
 909         struct connection *addcon;
 910
 911         mutex_lock(&connections_lock);
 912         if (!dlm_allow_conn) {
 913                 mutex_unlock(&connections_lock);
 914                 return -1;
 915         }
 916         mutex_unlock(&connections_lock);
 917
 918         memset(&peeraddr, 0, sizeof(peeraddr));
 919         result = sock_create_kern(dlm_local_addr[0]->ss_family, SOCK_STREAM,
 920                                   IPPROTO_TCP, &newsock);
 921         if (result < 0)
 922                 return -ENOMEM;
 923
 924         mutex_lock_nested(&con->sock_mutex, 0);
 925
 926         result = -ENOTCONN;
 927         if (con->sock == NULL)
 928                 goto accept_err;
 929
 930         newsock->type = con->sock->type;
 931         newsock->ops = con->sock->ops;
 932
 933         result = con->sock->ops->accept(con->sock, newsock, O_NONBLOCK);
 934         if (result < 0)
 935                 goto accept_err;
 936
 937         /* Get the connected socket's peer */
 938         memset(&peeraddr, 0, sizeof(peeraddr));
 939         if (newsock->ops->getname(newsock, (struct sockaddr *)&peeraddr,
 940                                   &len, 2)) {
 941                 result = -ECONNABORTED;
 942                 goto accept_err;
 943         }
 944
 945         /* Get the new node's NODEID */
 946         make_sockaddr(&peeraddr, 0, &len);
 947         if (addr_to_nodeid(&peeraddr, &nodeid)) {
 948                 unsigned char *b=(unsigned char *)&peeraddr;
 949                 log_print("connect from non cluster node");
 950                 print_hex_dump_bytes("ss: ", DUMP_PREFIX_NONE,
 951                                      b, sizeof(struct sockaddr_storage));
 952                 sock_release(newsock);
 953                 mutex_unlock(&con->sock_mutex);
 954                 return -1;
 955         }
 956
 957         log_print("got connection from %d", nodeid);
 958
 959         /*  Check to see if we already have a connection to this node. This
 960          *  could happen if the two nodes initiate a connection at roughly
 961          *  the same time and the connections cross on the wire.
 962          *  In this case we store the incoming one in "othercon"
 963          */
 964         newcon = nodeid2con(nodeid, GFP_NOFS);
 965         if (!newcon) {
 966                 result = -ENOMEM;
 967                 goto accept_err;
 968         }
 969         mutex_lock_nested(&newcon->sock_mutex, 1);
 970         if (newcon->sock) {
 971                 struct connection *othercon = newcon->othercon;
 972
 973                 if (!othercon) {
 974                         othercon = kmem_cache_zalloc(con_cache, GFP_NOFS);
 975                         if (!othercon) {
 976                                 log_print("failed to allocate incoming socket");
 977                                 mutex_unlock(&newcon->sock_mutex);
 978                                 result = -ENOMEM;
 979                                 goto accept_err;
 980                         }
 981                         othercon->nodeid = nodeid;
 982                         othercon->rx_action = receive_from_sock;
 983                         mutex_init(&othercon->sock_mutex);
 984                         INIT_WORK(&othercon->swork, process_send_sockets);
 985                         INIT_WORK(&othercon->rwork, process_recv_sockets);
 986                         set_bit(CF_IS_OTHERCON, &othercon->flags);
 987                 }
 988                 if (!othercon->sock) {
 989                         newcon->othercon = othercon;
 990                         othercon->sock = newsock;
 991                         newsock->sk->sk_user_data = othercon;
 992                         add_sock(newsock, othercon);
 993                         addcon = othercon;
 994                 }
 995                 else {
 996                         printk("Extra connection from node %d attempted\n", nodeid);
 997                         result = -EAGAIN;
 998                         mutex_unlock(&newcon->sock_mutex);
 999                         goto accept_err;
1000                 }
1001         }
1002         else {
1003                 newsock->sk->sk_user_data = newcon;
1004                 newcon->rx_action = receive_from_sock;
1005                 add_sock(newsock, newcon);
1006                 addcon = newcon;
1007         }
1008
1009         mutex_unlock(&newcon->sock_mutex);
1010
1011         /*
1012          * Add it to the active queue in case we got data
1013          * between processing the accept adding the socket
1014          * to the read_sockets list
1015          */
1016         if (!test_and_set_bit(CF_READ_PENDING, &addcon->flags))
1017                 queue_work(recv_workqueue, &addcon->rwork);
1018         mutex_unlock(&con->sock_mutex);
1019
1020         return 0;
1021
1022 accept_err:
1023         mutex_unlock(&con->sock_mutex);
1024         sock_release(newsock);
1025
1026         if (result != -EAGAIN)
1027                 log_print("error accepting connection from node: %d", result);
1028         return result;
1029 }
1030
1031 static void free_entry(struct writequeue_entry *e)
1032 {
1033         __free_page(e->page);
1034         kfree(e);
1035 }
1036
1037 /*
1038  * writequeue_entry_complete - try to delete and free write queue entry
1039  * @e: write queue entry to try to delete
1040  * @completed: bytes completed
1041  *
1042  * writequeue_lock must be held.
1043  */
1044 static void writequeue_entry_complete(struct writequeue_entry *e, int completed)
1045 {
1046         e->offset += completed;
1047         e->len -= completed;
1048
1049         if (e->len == 0 && e->users == 0) {
1050                 list_del(&e->list);
1051                 free_entry(e);
1052         }
1053 }
1054
1055 /* Initiate an SCTP association.
1056    This is a special case of send_to_sock() in that we don't yet have a
1057    peeled-off socket for this association, so we use the listening socket
1058    and add the primary IP address of the remote node.
1059  */
1060 static void sctp_init_assoc(struct connection *con)
1061 {
1062         struct sockaddr_storage rem_addr;
1063         char outcmsg[CMSG_SPACE(sizeof(struct sctp_sndrcvinfo))];
1064         struct msghdr outmessage;
1065         struct cmsghdr *cmsg;
1066         struct sctp_sndrcvinfo *sinfo;
1067         struct connection *base_con;
1068         struct writequeue_entry *e;
1069         int len, offset;
1070         int ret;
1071         int addrlen;
1072         struct kvec iov[1];
1073
1074         mutex_lock(&con->sock_mutex);
1075         if (test_and_set_bit(CF_INIT_PENDING, &con->flags))
1076                 goto unlock;
1077
1078         if (nodeid_to_addr(con->nodeid, NULL, (struct sockaddr *)&rem_addr,
1079                            con->try_new_addr)) {
1080                 log_print("no address for nodeid %d", con->nodeid);
1081                 goto unlock;
1082         }
1083         base_con = nodeid2con(0, 0);
1084         BUG_ON(base_con == NULL);
1085
1086         make_sockaddr(&rem_addr, dlm_config.ci_tcp_port, &addrlen);
1087
1088         outmessage.msg_name = &rem_addr;
1089         outmessage.msg_namelen = addrlen;
1090         outmessage.msg_control = outcmsg;
1091         outmessage.msg_controllen = sizeof(outcmsg);
1092         outmessage.msg_flags = MSG_EOR;
1093
1094         spin_lock(&con->writequeue_lock);
1095
1096         if (list_empty(&con->writequeue)) {
1097                 spin_unlock(&con->writequeue_lock);
1098                 log_print("writequeue empty for nodeid %d", con->nodeid);
1099                 goto unlock;
1100         }
1101
1102         e = list_first_entry(&con->writequeue, struct writequeue_entry, list);
1103         len = e->len;
1104         offset = e->offset;
1105
1106         /* Send the first block off the write queue */
1107         iov[0].iov_base = page_address(e->page)+offset;
1108         iov[0].iov_len = len;
1109         spin_unlock(&con->writequeue_lock);
1110
1111         if (rem_addr.ss_family == AF_INET) {
1112                 struct sockaddr_in *sin = (struct sockaddr_in *)&rem_addr;
1113                 log_print("Trying to connect to %pI4", &sin->sin_addr.s_addr);
1114         } else {
1115                 struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)&rem_addr;
1116                 log_print("Trying to connect to %pI6", &sin6->sin6_addr);
1117         }
1118
1119         cmsg = CMSG_FIRSTHDR(&outmessage);
1120         cmsg->cmsg_level = IPPROTO_SCTP;
1121         cmsg->cmsg_type = SCTP_SNDRCV;
1122         cmsg->cmsg_len = CMSG_LEN(sizeof(struct sctp_sndrcvinfo));
1123         sinfo = CMSG_DATA(cmsg);
1124         memset(sinfo, 0x00, sizeof(struct sctp_sndrcvinfo));
1125         sinfo->sinfo_ppid = cpu_to_le32(con->nodeid);
1126         outmessage.msg_controllen = cmsg->cmsg_len;
1127         sinfo->sinfo_flags |= SCTP_ADDR_OVER;
1128
1129         ret = kernel_sendmsg(base_con->sock, &outmessage, iov, 1, len);
1130         if (ret < 0) {
1131                 log_print("Send first packet to node %d failed: %d",
1132                           con->nodeid, ret);
1133
1134                 /* Try again later */
1135                 clear_bit(CF_CONNECT_PENDING, &con->flags);
1136                 clear_bit(CF_INIT_PENDING, &con->flags);
1137         }
1138         else {
1139                 spin_lock(&con->writequeue_lock);
1140                 writequeue_entry_complete(e, ret);
1141                 spin_unlock(&con->writequeue_lock);
1142         }
1143
1144 unlock:
1145         mutex_unlock(&con->sock_mutex);
1146 }
1147
1148 /* Connect a new socket to its peer */
1149 static void tcp_connect_to_sock(struct connection *con)
1150 {
1151         struct sockaddr_storage saddr, src_addr;
1152         int addr_len;
1153         struct socket *sock = NULL;
1154         int one = 1;
1155         int result;
1156
1157         if (con->nodeid == 0) {
1158                 log_print("attempt to connect sock 0 foiled");
1159                 return;
1160         }
1161
1162         mutex_lock(&con->sock_mutex);
1163         if (con->retries++ > MAX_CONNECT_RETRIES)
1164                 goto out;
1165
1166         /* Some odd races can cause double-connects, ignore them */
1167         if (con->sock)
1168                 goto out;
1169
1170         /* Create a socket to communicate with */
1171         result = sock_create_kern(dlm_local_addr[0]->ss_family, SOCK_STREAM,
1172                                   IPPROTO_TCP, &sock);
1173         if (result < 0)
1174                 goto out_err;
1175
1176         memset(&saddr, 0, sizeof(saddr));
1177         result = nodeid_to_addr(con->nodeid, &saddr, NULL, false);
1178         if (result < 0) {
1179                 log_print("no address for nodeid %d", con->nodeid);
1180                 goto out_err;
1181         }
1182
1183         sock->sk->sk_user_data = con;
1184         con->rx_action = receive_from_sock;
1185         con->connect_action = tcp_connect_to_sock;
1186         add_sock(sock, con);
1187
1188         /* Bind to our cluster-known address connecting to avoid
1189            routing problems */
1190         memcpy(&src_addr, dlm_local_addr[0], sizeof(src_addr));
1191         make_sockaddr(&src_addr, 0, &addr_len);
1192         result = sock->ops->bind(sock, (struct sockaddr *) &src_addr,
1193                                  addr_len);
1194         if (result < 0) {
1195                 log_print("could not bind for connect: %d", result);
1196                 /* This *may* not indicate a critical error */
1197         }
1198
1199         make_sockaddr(&saddr, dlm_config.ci_tcp_port, &addr_len);
1200
1201         log_print("connecting to %d", con->nodeid);
1202
1203         /* Turn off Nagle's algorithm */
1204         kernel_setsockopt(sock, SOL_TCP, TCP_NODELAY, (char *)&one,
1205                           sizeof(one));
1206
1207         result = sock->ops->connect(sock, (struct sockaddr *)&saddr, addr_len,
1208                                    O_NONBLOCK);
1209         if (result == -EINPROGRESS)
1210                 result = 0;
1211         if (result == 0)
1212                 goto out;
1213
1214 out_err:
1215         if (con->sock) {
1216                 sock_release(con->sock);
1217                 con->sock = NULL;
1218         } else if (sock) {
1219                 sock_release(sock);
1220         }
1221         /*
1222          * Some errors are fatal and this list might need adjusting. For other
1223          * errors we try again until the max number of retries is reached.
1224          */
1225         if (result != -EHOSTUNREACH &&
1226             result != -ENETUNREACH &&
1227             result != -ENETDOWN &&
1228             result != -EINVAL &&
1229             result != -EPROTONOSUPPORT) {
1230                 log_print("connect %d try %d error %d", con->nodeid,
1231                           con->retries, result);
1232                 mutex_unlock(&con->sock_mutex);
1233                 msleep(1000);
1234                 lowcomms_connect_sock(con);
1235                 return;
1236         }
1237 out:
1238         mutex_unlock(&con->sock_mutex);
1239         return;
1240 }
1241
1242 static struct socket *tcp_create_listen_sock(struct connection *con,
1243                                              struct sockaddr_storage *saddr)
1244 {
1245         struct socket *sock = NULL;
1246         int result = 0;
1247         int one = 1;
1248         int addr_len;
1249
1250         if (dlm_local_addr[0]->ss_family == AF_INET)
1251                 addr_len = sizeof(struct sockaddr_in);
1252         else
1253                 addr_len = sizeof(struct sockaddr_in6);
1254
1255         /* Create a socket to communicate with */
1256         result = sock_create_kern(dlm_local_addr[0]->ss_family, SOCK_STREAM,
1257                                   IPPROTO_TCP, &sock);
1258         if (result < 0) {
1259                 log_print("Can't create listening comms socket");
1260                 goto create_out;
1261         }
1262
1263         /* Turn off Nagle's algorithm */
1264         kernel_setsockopt(sock, SOL_TCP, TCP_NODELAY, (char *)&one,
1265                           sizeof(one));
1266
1267         result = kernel_setsockopt(sock, SOL_SOCKET, SO_REUSEADDR,
1268                                    (char *)&one, sizeof(one));
1269
1270         if (result < 0) {
1271                 log_print("Failed to set SO_REUSEADDR on socket: %d", result);
1272         }
1273         con->rx_action = tcp_accept_from_sock;
1274         con->connect_action = tcp_connect_to_sock;
1275
1276         /* Bind to our port */
1277         make_sockaddr(saddr, dlm_config.ci_tcp_port, &addr_len);
1278         result = sock->ops->bind(sock, (struct sockaddr *) saddr, addr_len);
1279         if (result < 0) {
1280                 log_print("Can't bind to port %d", dlm_config.ci_tcp_port);
1281                 sock_release(sock);
1282                 sock = NULL;
1283                 con->sock = NULL;
1284                 goto create_out;
1285         }
1286         result = kernel_setsockopt(sock, SOL_SOCKET, SO_KEEPALIVE,
1287                                  (char *)&one, sizeof(one));
1288         if (result < 0) {
1289                 log_print("Set keepalive failed: %d", result);
1290         }
1291
1292         result = sock->ops->listen(sock, 5);
1293         if (result < 0) {
1294                 log_print("Can't listen on port %d", dlm_config.ci_tcp_port);
1295                 sock_release(sock);
1296                 sock = NULL;
1297                 goto create_out;
1298         }
1299
1300 create_out:
1301         return sock;
1302 }
1303
1304 /* Get local addresses */
1305 static void init_local(void)
1306 {
1307         struct sockaddr_storage sas, *addr;
1308         int i;
1309
1310         dlm_local_count = 0;
1311         for (i = 0; i < DLM_MAX_ADDR_COUNT; i++) {
1312                 if (dlm_our_addr(&sas, i))
1313                         break;
1314
1315                 addr = kmalloc(sizeof(*addr), GFP_NOFS);
1316                 if (!addr)
1317                         break;
1318                 memcpy(addr, &sas, sizeof(*addr));
1319                 dlm_local_addr[dlm_local_count++] = addr;
1320         }
1321 }
1322
1323 /* Bind to an IP address. SCTP allows multiple address so it can do
1324    multi-homing */
1325 static int add_sctp_bind_addr(struct connection *sctp_con,
1326                               struct sockaddr_storage *addr,
1327                               int addr_len, int num)
1328 {
1329         int result = 0;
1330
1331         if (num == 1)
1332                 result = kernel_bind(sctp_con->sock,
1333                                      (struct sockaddr *) addr,
1334                                      addr_len);
1335         else
1336                 result = kernel_setsockopt(sctp_con->sock, SOL_SCTP,
1337                                            SCTP_SOCKOPT_BINDX_ADD,
1338                                            (char *)addr, addr_len);
1339
1340         if (result < 0)
1341                 log_print("Can't bind to port %d addr number %d",
1342                           dlm_config.ci_tcp_port, num);
1343
1344         return result;
1345 }
1346
1347 /* Initialise SCTP socket and bind to all interfaces */
1348 static int sctp_listen_for_all(void)
1349 {
1350         struct socket *sock = NULL;
1351         struct sockaddr_storage localaddr;
1352         struct sctp_event_subscribe subscribe;
1353         int result = -EINVAL, num = 1, i, addr_len;
1354         struct connection *con = nodeid2con(0, GFP_NOFS);
1355         int bufsize = NEEDED_RMEM;
1356         int one = 1;
1357
1358         if (!con)
1359                 return -ENOMEM;
1360
1361         log_print("Using SCTP for communications");
1362
1363         result = sock_create_kern(dlm_local_addr[0]->ss_family, SOCK_SEQPACKET,
1364                                   IPPROTO_SCTP, &sock);
1365         if (result < 0) {
1366                 log_print("Can't create comms socket, check SCTP is loaded");
1367                 goto out;
1368         }
1369
1370         /* Listen for events */
1371         memset(&subscribe, 0, sizeof(subscribe));
1372         subscribe.sctp_data_io_event = 1;
1373         subscribe.sctp_association_event = 1;
1374         subscribe.sctp_send_failure_event = 1;
1375         subscribe.sctp_shutdown_event = 1;
1376         subscribe.sctp_partial_delivery_event = 1;
1377
1378         result = kernel_setsockopt(sock, SOL_SOCKET, SO_RCVBUFFORCE,
1379                                  (char *)&bufsize, sizeof(bufsize));
1380         if (result)
1381                 log_print("Error increasing buffer space on socket %d", result);
1382
1383         result = kernel_setsockopt(sock, SOL_SCTP, SCTP_EVENTS,
1384                                    (char *)&subscribe, sizeof(subscribe));
1385         if (result < 0) {
1386                 log_print("Failed to set SCTP_EVENTS on socket: result=%d",
1387                           result);
1388                 goto create_delsock;
1389         }
1390
1391         result = kernel_setsockopt(sock, SOL_SCTP, SCTP_NODELAY, (char *)&one,
1392                                    sizeof(one));
1393         if (result < 0)
1394                 log_print("Could not set SCTP NODELAY error %d\n", result);
1395
1396         /* Init con struct */
1397         sock->sk->sk_user_data = con;
1398         con->sock = sock;
1399         con->sock->sk->sk_data_ready = lowcomms_data_ready;
1400         con->rx_action = receive_from_sock;
1401         con->connect_action = sctp_init_assoc;
1402
1403         /* Bind to all interfaces. */
1404         for (i = 0; i < dlm_local_count; i++) {
1405                 memcpy(&localaddr, dlm_local_addr[i], sizeof(localaddr));
1406                 make_sockaddr(&localaddr, dlm_config.ci_tcp_port, &addr_len);
1407
1408                 result = add_sctp_bind_addr(con, &localaddr, addr_len, num);
1409                 if (result)
1410                         goto create_delsock;
1411                 ++num;
1412         }
1413
1414         result = sock->ops->listen(sock, 5);
1415         if (result < 0) {
1416                 log_print("Can't set socket listening");
1417                 goto create_delsock;
1418         }
1419
1420         return 0;
1421
1422 create_delsock:
1423         sock_release(sock);
1424         con->sock = NULL;
1425 out:
1426         return result;
1427 }
1428
1429 static int tcp_listen_for_all(void)
1430 {
1431         struct socket *sock = NULL;
1432         struct connection *con = nodeid2con(0, GFP_NOFS);
1433         int result = -EINVAL;
1434
1435         if (!con)
1436                 return -ENOMEM;
1437
1438         /* We don't support multi-homed hosts */
1439         if (dlm_local_addr[1] != NULL) {
1440                 log_print("TCP protocol can't handle multi-homed hosts, "
1441                           "try SCTP");
1442                 return -EINVAL;
1443         }
1444
1445         log_print("Using TCP for communications");
1446
1447         sock = tcp_create_listen_sock(con, dlm_local_addr[0]);
1448         if (sock) {
1449                 add_sock(sock, con);
1450                 result = 0;
1451         }
1452         else {
1453                 result = -EADDRINUSE;
1454         }
1455
1456         return result;
1457 }
1458
1459
1460
1461 static struct writequeue_entry *new_writequeue_entry(struct connection *con,
1462                                                      gfp_t allocation)
1463 {
1464         struct writequeue_entry *entry;
1465
1466         entry = kmalloc(sizeof(struct writequeue_entry), allocation);
1467         if (!entry)
1468                 return NULL;
1469
1470         entry->page = alloc_page(allocation);
1471         if (!entry->page) {
1472                 kfree(entry);
1473                 return NULL;
1474         }
1475
1476         entry->offset = 0;
1477         entry->len = 0;
1478         entry->end = 0;
1479         entry->users = 0;
1480         entry->con = con;
1481
1482         return entry;
1483 }
1484
1485 void *dlm_lowcomms_get_buffer(int nodeid, int len, gfp_t allocation, char **ppc)
1486 {
1487         struct connection *con;
1488         struct writequeue_entry *e;
1489         int offset = 0;
1490
1491         con = nodeid2con(nodeid, allocation);
1492         if (!con)
1493                 return NULL;
1494
1495         spin_lock(&con->writequeue_lock);
1496         e = list_entry(con->writequeue.prev, struct writequeue_entry, list);
1497         if ((&e->list == &con->writequeue) ||
1498             (PAGE_CACHE_SIZE - e->end < len)) {
1499                 e = NULL;
1500         } else {
1501                 offset = e->end;
1502                 e->end += len;
1503                 e->users++;
1504         }
1505         spin_unlock(&con->writequeue_lock);
1506
1507         if (e) {
1508         got_one:
1509                 *ppc = page_address(e->page) + offset;
1510                 return e;
1511         }
1512
1513         e = new_writequeue_entry(con, allocation);
1514         if (e) {
1515                 spin_lock(&con->writequeue_lock);
1516                 offset = e->end;
1517                 e->end += len;
1518                 e->users++;
1519                 list_add_tail(&e->list, &con->writequeue);
1520                 spin_unlock(&con->writequeue_lock);
1521                 goto got_one;
1522         }
1523         return NULL;
1524 }
1525
1526 void dlm_lowcomms_commit_buffer(void *mh)
1527 {
1528         struct writequeue_entry *e = (struct writequeue_entry *)mh;
1529         struct connection *con = e->con;
1530         int users;
1531
1532         spin_lock(&con->writequeue_lock);
1533         users = --e->users;
1534         if (users)
1535                 goto out;
1536         e->len = e->end - e->offset;
1537         spin_unlock(&con->writequeue_lock);
1538
1539         if (!test_and_set_bit(CF_WRITE_PENDING, &con->flags)) {
1540                 queue_work(send_workqueue, &con->swork);
1541         }
1542         return;
1543
1544 out:
1545         spin_unlock(&con->writequeue_lock);
1546         return;
1547 }
1548
1549 /* Send a message */
1550 static void send_to_sock(struct connection *con)
1551 {
1552         int ret = 0;
1553         const int msg_flags = MSG_DONTWAIT | MSG_NOSIGNAL;
1554         struct writequeue_entry *e;
1555         int len, offset;
1556         int count = 0;
1557
1558         mutex_lock(&con->sock_mutex);
1559         if (con->sock == NULL)
1560                 goto out_connect;
1561
1562         spin_lock(&con->writequeue_lock);
1563         for (;;) {
1564                 e = list_entry(con->writequeue.next, struct writequeue_entry,
1565                                list);
1566                 if ((struct list_head *) e == &con->writequeue)
1567                         break;
1568
1569                 len = e->len;
1570                 offset = e->offset;
1571                 BUG_ON(len == 0 && e->users == 0);
1572                 spin_unlock(&con->writequeue_lock);
1573
1574                 ret = 0;
1575                 if (len) {
1576                         ret = kernel_sendpage(con->sock, e->page, offset, len,
1577                                               msg_flags);
1578                         if (ret == -EAGAIN || ret == 0) {
1579                                 if (ret == -EAGAIN &&
1580                                     test_bit(SOCK_ASYNC_NOSPACE, &con->sock->flags) &&
1581                                     !test_and_set_bit(CF_APP_LIMITED, &con->flags)) {
1582                                         /* Notify TCP that we're limited by the
1583                                          * application window size.
1584                                          */
1585                                         set_bit(SOCK_NOSPACE, &con->sock->flags);
1586                                         con->sock->sk->sk_write_pending++;
1587                                 }
1588                                 cond_resched();
1589                                 goto out;
1590                         } else if (ret < 0)
1591                                 goto send_error;
1592                 }
1593
1594                 /* Don't starve people filling buffers */
1595                 if (++count >= MAX_SEND_MSG_COUNT) {
1596                         cond_resched();
1597                         count = 0;
1598                 }
1599
1600                 spin_lock(&con->writequeue_lock);
1601                 writequeue_entry_complete(e, ret);
1602         }
1603         spin_unlock(&con->writequeue_lock);
1604 out:
1605         mutex_unlock(&con->sock_mutex);
1606         return;
1607
1608 send_error:
1609         mutex_unlock(&con->sock_mutex);
1610         close_connection(con, false);
1611         lowcomms_connect_sock(con);
1612         return;
1613
1614 out_connect:
1615         mutex_unlock(&con->sock_mutex);
1616         if (!test_bit(CF_INIT_PENDING, &con->flags))
1617                 lowcomms_connect_sock(con);
1618 }
1619
1620 static void clean_one_writequeue(struct connection *con)
1621 {
1622         struct writequeue_entry *e, *safe;
1623
1624         spin_lock(&con->writequeue_lock);
1625         list_for_each_entry_safe(e, safe, &con->writequeue, list) {
1626                 list_del(&e->list);
1627                 free_entry(e);
1628         }
1629         spin_unlock(&con->writequeue_lock);
1630 }
1631
1632 /* Called from recovery when it knows that a node has
1633    left the cluster */
1634 int dlm_lowcomms_close(int nodeid)
1635 {
1636         struct connection *con;
1637         struct dlm_node_addr *na;
1638
1639         log_print("closing connection to node %d", nodeid);
1640         con = nodeid2con(nodeid, 0);
1641         if (con) {
1642                 clear_bit(CF_CONNECT_PENDING, &con->flags);
1643                 clear_bit(CF_WRITE_PENDING, &con->flags);
1644                 set_bit(CF_CLOSE, &con->flags);
1645                 if (cancel_work_sync(&con->swork))
1646                         log_print("canceled swork for node %d", nodeid);
1647                 if (cancel_work_sync(&con->rwork))
1648                         log_print("canceled rwork for node %d", nodeid);
1649                 clean_one_writequeue(con);
1650                 close_connection(con, true);
1651         }
1652
1653         spin_lock(&dlm_node_addrs_spin);
1654         na = find_node_addr(nodeid);
1655         if (na) {
1656                 list_del(&na->list);
1657                 while (na->addr_count--)
1658                         kfree(na->addr[na->addr_count]);
1659                 kfree(na);
1660         }
1661         spin_unlock(&dlm_node_addrs_spin);
1662
1663         return 0;
1664 }
1665
1666 /* Receive workqueue function */
1667 static void process_recv_sockets(struct work_struct *work)
1668 {
1669         struct connection *con = container_of(work, struct connection, rwork);
1670         int err;
1671
1672         clear_bit(CF_READ_PENDING, &con->flags);
1673         do {
1674                 err = con->rx_action(con);
1675         } while (!err);
1676 }
1677
1678 /* Send workqueue function */
1679 static void process_send_sockets(struct work_struct *work)
1680 {
1681         struct connection *con = container_of(work, struct connection, swork);
1682
1683         if (test_and_clear_bit(CF_CONNECT_PENDING, &con->flags)) {
1684                 con->connect_action(con);
1685                 set_bit(CF_WRITE_PENDING, &con->flags);
1686         }
1687         if (test_and_clear_bit(CF_WRITE_PENDING, &con->flags))
1688                 send_to_sock(con);
1689 }
1690
1691
1692 /* Discard all entries on the write queues */
1693 static void clean_writequeues(void)
1694 {
1695         foreach_conn(clean_one_writequeue);
1696 }
1697
1698 static void work_stop(void)
1699 {
1700         destroy_workqueue(recv_workqueue);
1701         destroy_workqueue(send_workqueue);
1702 }
1703
1704 static int work_start(void)
1705 {
1706         recv_workqueue = alloc_workqueue("dlm_recv",
1707                                          WQ_UNBOUND | WQ_MEM_RECLAIM, 1);
1708         if (!recv_workqueue) {
1709                 log_print("can't start dlm_recv");
1710                 return -ENOMEM;
1711         }
1712
1713         send_workqueue = alloc_workqueue("dlm_send",
1714                                          WQ_UNBOUND | WQ_MEM_RECLAIM, 1);
1715         if (!send_workqueue) {
1716                 log_print("can't start dlm_send");
1717                 destroy_workqueue(recv_workqueue);
1718                 return -ENOMEM;
1719         }
1720
1721         return 0;
1722 }
1723
1724 static void stop_conn(struct connection *con)
1725 {
1726         con->flags |= 0x0F;
1727         if (con->sock && con->sock->sk)
1728                 con->sock->sk->sk_user_data = NULL;
1729 }
1730
1731 static void free_conn(struct connection *con)
1732 {
1733         close_connection(con, true);
1734         if (con->othercon)
1735                 kmem_cache_free(con_cache, con->othercon);
1736         hlist_del(&con->list);
1737         kmem_cache_free(con_cache, con);
1738 }
1739
1740 void dlm_lowcomms_stop(void)
1741 {
1742         /* Set all the flags to prevent any
1743            socket activity.
1744         */
1745         mutex_lock(&connections_lock);
1746         dlm_allow_conn = 0;
1747         foreach_conn(stop_conn);
1748         mutex_unlock(&connections_lock);
1749
1750         work_stop();
1751
1752         mutex_lock(&connections_lock);
1753         clean_writequeues();
1754
1755         foreach_conn(free_conn);
1756
1757         mutex_unlock(&connections_lock);
1758         kmem_cache_destroy(con_cache);
1759 }
1760
1761 int dlm_lowcomms_start(void)
1762 {
1763         int error = -EINVAL;
1764         struct connection *con;
1765         int i;
1766
1767         for (i = 0; i < CONN_HASH_SIZE; i++)
1768                 INIT_HLIST_HEAD(&connection_hash[i]);
1769
1770         init_local();
1771         if (!dlm_local_count) {
1772                 error = -ENOTCONN;
1773                 log_print("no local IP address has been set");
1774                 goto fail;
1775         }
1776
1777         error = -ENOMEM;
1778         con_cache = kmem_cache_create("dlm_conn", sizeof(struct connection),
1779                                       __alignof__(struct connection), 0,
1780                                       NULL);
1781         if (!con_cache)
1782                 goto fail;
1783
1784         error = work_start();
1785         if (error)
1786                 goto fail_destroy;
1787
1788         dlm_allow_conn = 1;
1789
1790         /* Start listening */
1791         if (dlm_config.ci_protocol == 0)
1792                 error = tcp_listen_for_all();
1793         else
1794                 error = sctp_listen_for_all();
1795         if (error)
1796                 goto fail_unlisten;
1797
1798         return 0;
1799
1800 fail_unlisten:
1801         dlm_allow_conn = 0;
1802         con = nodeid2con(0,0);
1803         if (con) {
1804                 close_connection(con, false);
1805                 kmem_cache_free(con_cache, con);
1806         }
1807 fail_destroy:
1808         kmem_cache_destroy(con_cache);
1809 fail:
1810         return error;
1811 }
1812
1813 void dlm_lowcomms_exit(void)
1814 {
1815         struct dlm_node_addr *na, *safe;
1816
1817         spin_lock(&dlm_node_addrs_spin);
1818         list_for_each_entry_safe(na, safe, &dlm_node_addrs, list) {
1819                 list_del(&na->list);
1820                 while (na->addr_count--)
1821                         kfree(na->addr[na->addr_count]);
1822                 kfree(na);
1823         }
1824         spin_unlock(&dlm_node_addrs_spin);
1825 }