1 // SPDX-License-Identifier: GPL-2.0-only 2 /****************************************************************************** 3 ******************************************************************************* 4 ** 5 ** Copyright (C) 2005-2011 Red Hat, Inc. All rights reserved. 6 ** 7 ** 8 ******************************************************************************* 9 ******************************************************************************/ 10 11 #include "dlm_internal.h" 12 #include "lockspace.h" 13 #include "member.h" 14 #include "recoverd.h" 15 #include "recover.h" 16 #include "rcom.h" 17 #include "config.h" 18 #include "midcomms.h" 19 #include "lowcomms.h" 20 21 int dlm_slots_version(const struct dlm_header *h) 22 { 23 if ((le32_to_cpu(h->h_version) & 0x0000FFFF) < DLM_HEADER_SLOTS) 24 return 0; 25 return 1; 26 } 27 28 void dlm_slot_save(struct dlm_ls *ls, struct dlm_rcom *rc, 29 struct dlm_member *memb) 30 { 31 struct rcom_config *rf = (struct rcom_config *)rc->rc_buf; 32 33 if (!dlm_slots_version(&rc->rc_header)) 34 return; 35 36 memb->slot = le16_to_cpu(rf->rf_our_slot); 37 memb->generation = le32_to_cpu(rf->rf_generation); 38 } 39 40 void dlm_slots_copy_out(struct dlm_ls *ls, struct dlm_rcom *rc) 41 { 42 struct dlm_slot *slot; 43 struct rcom_slot *ro; 44 int i; 45 46 ro = (struct rcom_slot *)(rc->rc_buf + sizeof(struct rcom_config)); 47 48 /* ls_slots array is sparse, but not rcom_slots */ 49 50 for (i = 0; i < ls->ls_slots_size; i++) { 51 slot = &ls->ls_slots[i]; 52 if (!slot->nodeid) 53 continue; 54 ro->ro_nodeid = cpu_to_le32(slot->nodeid); 55 ro->ro_slot = cpu_to_le16(slot->slot); 56 ro++; 57 } 58 } 59 60 #define SLOT_DEBUG_LINE 128 61 62 static void log_slots(struct dlm_ls *ls, uint32_t gen, int num_slots, 63 struct rcom_slot *ro0, struct dlm_slot *array, 64 int array_size) 65 { 66 char line[SLOT_DEBUG_LINE]; 67 int len = SLOT_DEBUG_LINE - 1; 68 int pos = 0; 69 int ret, i; 70 71 memset(line, 0, sizeof(line)); 72 73 if (array) { 74 for (i = 0; i < array_size; i++) { 75 if (!array[i].nodeid) 76 continue; 77 78 ret = snprintf(line + pos, len - pos, " %d:%d", 79 array[i].slot, array[i].nodeid); 80 if (ret >= len - pos) 81 break; 82 pos += ret; 83 } 84 } else if (ro0) { 85 for (i = 0; i < num_slots; i++) { 86 ret = snprintf(line + pos, len - pos, " %d:%d", 87 ro0[i].ro_slot, ro0[i].ro_nodeid); 88 if (ret >= len - pos) 89 break; 90 pos += ret; 91 } 92 } 93 94 log_rinfo(ls, "generation %u slots %d%s", gen, num_slots, line); 95 } 96 97 int dlm_slots_copy_in(struct dlm_ls *ls) 98 { 99 struct dlm_member *memb; 100 struct dlm_rcom *rc = ls->ls_recover_buf; 101 struct rcom_config *rf = (struct rcom_config *)rc->rc_buf; 102 struct rcom_slot *ro0, *ro; 103 int our_nodeid = dlm_our_nodeid(); 104 int i, num_slots; 105 uint32_t gen; 106 107 if (!dlm_slots_version(&rc->rc_header)) 108 return -1; 109 110 gen = le32_to_cpu(rf->rf_generation); 111 if (gen <= ls->ls_generation) { 112 log_error(ls, "dlm_slots_copy_in gen %u old %u", 113 gen, ls->ls_generation); 114 } 115 ls->ls_generation = gen; 116 117 num_slots = le16_to_cpu(rf->rf_num_slots); 118 if (!num_slots) 119 return -1; 120 121 ro0 = (struct rcom_slot *)(rc->rc_buf + sizeof(struct rcom_config)); 122 123 log_slots(ls, gen, num_slots, ro0, NULL, 0); 124 125 list_for_each_entry(memb, &ls->ls_nodes, list) { 126 for (i = 0, ro = ro0; i < num_slots; i++, ro++) { 127 if (le32_to_cpu(ro->ro_nodeid) != memb->nodeid) 128 continue; 129 memb->slot = le16_to_cpu(ro->ro_slot); 130 memb->slot_prev = memb->slot; 131 break; 132 } 133 134 if (memb->nodeid == our_nodeid) { 135 if (ls->ls_slot && ls->ls_slot != memb->slot) { 136 log_error(ls, "dlm_slots_copy_in our slot " 137 "changed %d %d", ls->ls_slot, 138 memb->slot); 139 return -1; 140 } 141 142 if (!ls->ls_slot) 143 ls->ls_slot = memb->slot; 144 } 145 146 if (!memb->slot) { 147 log_error(ls, "dlm_slots_copy_in nodeid %d no slot", 148 memb->nodeid); 149 return -1; 150 } 151 } 152 153 return 0; 154 } 155 156 /* for any nodes that do not support slots, we will not have set memb->slot 157 in wait_status_all(), so memb->slot will remain -1, and we will not 158 assign slots or set ls_num_slots here */ 159 160 int dlm_slots_assign(struct dlm_ls *ls, int *num_slots, int *slots_size, 161 struct dlm_slot **slots_out, uint32_t *gen_out) 162 { 163 struct dlm_member *memb; 164 struct dlm_slot *array; 165 int our_nodeid = dlm_our_nodeid(); 166 int array_size, max_slots, i; 167 int need = 0; 168 int max = 0; 169 int num = 0; 170 uint32_t gen = 0; 171 172 /* our own memb struct will have slot -1 gen 0 */ 173 174 list_for_each_entry(memb, &ls->ls_nodes, list) { 175 if (memb->nodeid == our_nodeid) { 176 memb->slot = ls->ls_slot; 177 memb->generation = ls->ls_generation; 178 break; 179 } 180 } 181 182 list_for_each_entry(memb, &ls->ls_nodes, list) { 183 if (memb->generation > gen) 184 gen = memb->generation; 185 186 /* node doesn't support slots */ 187 188 if (memb->slot == -1) 189 return -1; 190 191 /* node needs a slot assigned */ 192 193 if (!memb->slot) 194 need++; 195 196 /* node has a slot assigned */ 197 198 num++; 199 200 if (!max || max < memb->slot) 201 max = memb->slot; 202 203 /* sanity check, once slot is assigned it shouldn't change */ 204 205 if (memb->slot_prev && memb->slot && memb->slot_prev != memb->slot) { 206 log_error(ls, "nodeid %d slot changed %d %d", 207 memb->nodeid, memb->slot_prev, memb->slot); 208 return -1; 209 } 210 memb->slot_prev = memb->slot; 211 } 212 213 array_size = max + need; 214 array = kcalloc(array_size, sizeof(*array), GFP_NOFS); 215 if (!array) 216 return -ENOMEM; 217 218 num = 0; 219 220 /* fill in slots (offsets) that are used */ 221 222 list_for_each_entry(memb, &ls->ls_nodes, list) { 223 if (!memb->slot) 224 continue; 225 226 if (memb->slot > array_size) { 227 log_error(ls, "invalid slot number %d", memb->slot); 228 kfree(array); 229 return -1; 230 } 231 232 array[memb->slot - 1].nodeid = memb->nodeid; 233 array[memb->slot - 1].slot = memb->slot; 234 num++; 235 } 236 237 /* assign new slots from unused offsets */ 238 239 list_for_each_entry(memb, &ls->ls_nodes, list) { 240 if (memb->slot) 241 continue; 242 243 for (i = 0; i < array_size; i++) { 244 if (array[i].nodeid) 245 continue; 246 247 memb->slot = i + 1; 248 memb->slot_prev = memb->slot; 249 array[i].nodeid = memb->nodeid; 250 array[i].slot = memb->slot; 251 num++; 252 253 if (!ls->ls_slot && memb->nodeid == our_nodeid) 254 ls->ls_slot = memb->slot; 255 break; 256 } 257 258 if (!memb->slot) { 259 log_error(ls, "no free slot found"); 260 kfree(array); 261 return -1; 262 } 263 } 264 265 gen++; 266 267 log_slots(ls, gen, num, NULL, array, array_size); 268 269 max_slots = (DLM_MAX_APP_BUFSIZE - sizeof(struct dlm_rcom) - 270 sizeof(struct rcom_config)) / sizeof(struct rcom_slot); 271 272 if (num > max_slots) { 273 log_error(ls, "num_slots %d exceeds max_slots %d", 274 num, max_slots); 275 kfree(array); 276 return -1; 277 } 278 279 *gen_out = gen; 280 *slots_out = array; 281 *slots_size = array_size; 282 *num_slots = num; 283 return 0; 284 } 285 286 static void add_ordered_member(struct dlm_ls *ls, struct dlm_member *new) 287 { 288 struct dlm_member *memb = NULL; 289 struct list_head *tmp; 290 struct list_head *newlist = &new->list; 291 struct list_head *head = &ls->ls_nodes; 292 293 list_for_each(tmp, head) { 294 memb = list_entry(tmp, struct dlm_member, list); 295 if (new->nodeid < memb->nodeid) 296 break; 297 } 298 299 if (!memb) 300 list_add_tail(newlist, head); 301 else { 302 /* FIXME: can use list macro here */ 303 newlist->prev = tmp->prev; 304 newlist->next = tmp; 305 tmp->prev->next = newlist; 306 tmp->prev = newlist; 307 } 308 } 309 310 static int add_remote_member(int nodeid) 311 { 312 int error; 313 314 if (nodeid == dlm_our_nodeid()) 315 return 0; 316 317 error = dlm_lowcomms_connect_node(nodeid); 318 if (error < 0) 319 return error; 320 321 dlm_midcomms_add_member(nodeid); 322 return 0; 323 } 324 325 static int dlm_add_member(struct dlm_ls *ls, struct dlm_config_node *node) 326 { 327 struct dlm_member *memb; 328 int error; 329 330 memb = kzalloc(sizeof(*memb), GFP_NOFS); 331 if (!memb) 332 return -ENOMEM; 333 334 memb->nodeid = node->nodeid; 335 memb->weight = node->weight; 336 memb->comm_seq = node->comm_seq; 337 338 error = add_remote_member(node->nodeid); 339 if (error < 0) { 340 kfree(memb); 341 return error; 342 } 343 344 add_ordered_member(ls, memb); 345 ls->ls_num_nodes++; 346 return 0; 347 } 348 349 static struct dlm_member *find_memb(struct list_head *head, int nodeid) 350 { 351 struct dlm_member *memb; 352 353 list_for_each_entry(memb, head, list) { 354 if (memb->nodeid == nodeid) 355 return memb; 356 } 357 return NULL; 358 } 359 360 int dlm_is_member(struct dlm_ls *ls, int nodeid) 361 { 362 if (find_memb(&ls->ls_nodes, nodeid)) 363 return 1; 364 return 0; 365 } 366 367 int dlm_is_removed(struct dlm_ls *ls, int nodeid) 368 { 369 WARN_ON_ONCE(!nodeid || nodeid == -1); 370 371 if (find_memb(&ls->ls_nodes_gone, nodeid)) 372 return 1; 373 return 0; 374 } 375 376 static void clear_memb_list(struct list_head *head, 377 void (*after_del)(int nodeid)) 378 { 379 struct dlm_member *memb; 380 381 while (!list_empty(head)) { 382 memb = list_entry(head->next, struct dlm_member, list); 383 list_del(&memb->list); 384 if (after_del) 385 after_del(memb->nodeid); 386 kfree(memb); 387 } 388 } 389 390 static void remove_remote_member(int nodeid) 391 { 392 if (nodeid == dlm_our_nodeid()) 393 return; 394 395 dlm_midcomms_remove_member(nodeid); 396 } 397 398 void dlm_clear_members(struct dlm_ls *ls) 399 { 400 clear_memb_list(&ls->ls_nodes, remove_remote_member); 401 ls->ls_num_nodes = 0; 402 } 403 404 void dlm_clear_members_gone(struct dlm_ls *ls) 405 { 406 clear_memb_list(&ls->ls_nodes_gone, NULL); 407 } 408 409 static void make_member_array(struct dlm_ls *ls) 410 { 411 struct dlm_member *memb; 412 int i, w, x = 0, total = 0, all_zero = 0, *array; 413 414 kfree(ls->ls_node_array); 415 ls->ls_node_array = NULL; 416 417 list_for_each_entry(memb, &ls->ls_nodes, list) { 418 if (memb->weight) 419 total += memb->weight; 420 } 421 422 /* all nodes revert to weight of 1 if all have weight 0 */ 423 424 if (!total) { 425 total = ls->ls_num_nodes; 426 all_zero = 1; 427 } 428 429 ls->ls_total_weight = total; 430 array = kmalloc_array(total, sizeof(*array), GFP_NOFS); 431 if (!array) 432 return; 433 434 list_for_each_entry(memb, &ls->ls_nodes, list) { 435 if (!all_zero && !memb->weight) 436 continue; 437 438 if (all_zero) 439 w = 1; 440 else 441 w = memb->weight; 442 443 DLM_ASSERT(x < total, printk("total %d x %d\n", total, x);); 444 445 for (i = 0; i < w; i++) 446 array[x++] = memb->nodeid; 447 } 448 449 ls->ls_node_array = array; 450 } 451 452 /* send a status request to all members just to establish comms connections */ 453 454 static int ping_members(struct dlm_ls *ls, uint64_t seq) 455 { 456 struct dlm_member *memb; 457 int error = 0; 458 459 list_for_each_entry(memb, &ls->ls_nodes, list) { 460 if (dlm_recovery_stopped(ls)) { 461 error = -EINTR; 462 break; 463 } 464 error = dlm_rcom_status(ls, memb->nodeid, 0, seq); 465 if (error) 466 break; 467 } 468 if (error) 469 log_rinfo(ls, "ping_members aborted %d last nodeid %d", 470 error, ls->ls_recover_nodeid); 471 return error; 472 } 473 474 static void dlm_lsop_recover_prep(struct dlm_ls *ls) 475 { 476 if (!ls->ls_ops || !ls->ls_ops->recover_prep) 477 return; 478 ls->ls_ops->recover_prep(ls->ls_ops_arg); 479 } 480 481 static void dlm_lsop_recover_slot(struct dlm_ls *ls, struct dlm_member *memb) 482 { 483 struct dlm_slot slot; 484 uint32_t seq; 485 int error; 486 487 if (!ls->ls_ops || !ls->ls_ops->recover_slot) 488 return; 489 490 /* if there is no comms connection with this node 491 or the present comms connection is newer 492 than the one when this member was added, then 493 we consider the node to have failed (versus 494 being removed due to dlm_release_lockspace) */ 495 496 error = dlm_comm_seq(memb->nodeid, &seq); 497 498 if (!error && seq == memb->comm_seq) 499 return; 500 501 slot.nodeid = memb->nodeid; 502 slot.slot = memb->slot; 503 504 ls->ls_ops->recover_slot(ls->ls_ops_arg, &slot); 505 } 506 507 void dlm_lsop_recover_done(struct dlm_ls *ls) 508 { 509 struct dlm_member *memb; 510 struct dlm_slot *slots; 511 int i, num; 512 513 if (!ls->ls_ops || !ls->ls_ops->recover_done) 514 return; 515 516 num = ls->ls_num_nodes; 517 slots = kcalloc(num, sizeof(*slots), GFP_KERNEL); 518 if (!slots) 519 return; 520 521 i = 0; 522 list_for_each_entry(memb, &ls->ls_nodes, list) { 523 if (i == num) { 524 log_error(ls, "dlm_lsop_recover_done bad num %d", num); 525 goto out; 526 } 527 slots[i].nodeid = memb->nodeid; 528 slots[i].slot = memb->slot; 529 i++; 530 } 531 532 ls->ls_ops->recover_done(ls->ls_ops_arg, slots, num, 533 ls->ls_slot, ls->ls_generation); 534 out: 535 kfree(slots); 536 } 537 538 static struct dlm_config_node *find_config_node(struct dlm_recover *rv, 539 int nodeid) 540 { 541 int i; 542 543 for (i = 0; i < rv->nodes_count; i++) { 544 if (rv->nodes[i].nodeid == nodeid) 545 return &rv->nodes[i]; 546 } 547 return NULL; 548 } 549 550 int dlm_recover_members(struct dlm_ls *ls, struct dlm_recover *rv, int *neg_out) 551 { 552 struct dlm_member *memb, *safe; 553 struct dlm_config_node *node; 554 int i, error, neg = 0, low = -1; 555 556 /* previously removed members that we've not finished removing need to 557 * count as a negative change so the "neg" recovery steps will happen 558 * 559 * This functionality must report all member changes to lsops or 560 * midcomms layer and must never return before. 561 */ 562 563 list_for_each_entry(memb, &ls->ls_nodes_gone, list) { 564 log_rinfo(ls, "prev removed member %d", memb->nodeid); 565 neg++; 566 } 567 568 /* move departed members from ls_nodes to ls_nodes_gone */ 569 570 list_for_each_entry_safe(memb, safe, &ls->ls_nodes, list) { 571 node = find_config_node(rv, memb->nodeid); 572 if (node && !node->new) 573 continue; 574 575 if (!node) { 576 log_rinfo(ls, "remove member %d", memb->nodeid); 577 } else { 578 /* removed and re-added */ 579 log_rinfo(ls, "remove member %d comm_seq %u %u", 580 memb->nodeid, memb->comm_seq, node->comm_seq); 581 } 582 583 neg++; 584 list_move(&memb->list, &ls->ls_nodes_gone); 585 remove_remote_member(memb->nodeid); 586 ls->ls_num_nodes--; 587 dlm_lsop_recover_slot(ls, memb); 588 } 589 590 /* add new members to ls_nodes */ 591 592 for (i = 0; i < rv->nodes_count; i++) { 593 node = &rv->nodes[i]; 594 if (dlm_is_member(ls, node->nodeid)) 595 continue; 596 error = dlm_add_member(ls, node); 597 if (error) 598 return error; 599 600 log_rinfo(ls, "add member %d", node->nodeid); 601 } 602 603 list_for_each_entry(memb, &ls->ls_nodes, list) { 604 if (low == -1 || memb->nodeid < low) 605 low = memb->nodeid; 606 } 607 ls->ls_low_nodeid = low; 608 609 make_member_array(ls); 610 *neg_out = neg; 611 612 error = ping_members(ls, rv->seq); 613 log_rinfo(ls, "dlm_recover_members %d nodes", ls->ls_num_nodes); 614 return error; 615 } 616 617 /* Userspace guarantees that dlm_ls_stop() has completed on all nodes before 618 dlm_ls_start() is called on any of them to start the new recovery. */ 619 620 int dlm_ls_stop(struct dlm_ls *ls) 621 { 622 int new; 623 624 /* 625 * Prevent dlm_recv from being in the middle of something when we do 626 * the stop. This includes ensuring dlm_recv isn't processing a 627 * recovery message (rcom), while dlm_recoverd is aborting and 628 * resetting things from an in-progress recovery. i.e. we want 629 * dlm_recoverd to abort its recovery without worrying about dlm_recv 630 * processing an rcom at the same time. Stopping dlm_recv also makes 631 * it easy for dlm_receive_message() to check locking stopped and add a 632 * message to the requestqueue without races. 633 */ 634 635 write_lock_bh(&ls->ls_recv_active); 636 637 /* 638 * Abort any recovery that's in progress (see RECOVER_STOP, 639 * dlm_recovery_stopped()) and tell any other threads running in the 640 * dlm to quit any processing (see RUNNING, dlm_locking_stopped()). 641 */ 642 643 spin_lock_bh(&ls->ls_recover_lock); 644 set_bit(LSFL_RECOVER_STOP, &ls->ls_flags); 645 new = test_and_clear_bit(LSFL_RUNNING, &ls->ls_flags); 646 if (new) 647 timer_delete_sync(&ls->ls_scan_timer); 648 ls->ls_recover_seq++; 649 650 /* activate requestqueue and stop processing */ 651 write_lock_bh(&ls->ls_requestqueue_lock); 652 set_bit(LSFL_RECV_MSG_BLOCKED, &ls->ls_flags); 653 write_unlock_bh(&ls->ls_requestqueue_lock); 654 spin_unlock_bh(&ls->ls_recover_lock); 655 656 /* 657 * Let dlm_recv run again, now any normal messages will be saved on the 658 * requestqueue for later. 659 */ 660 661 write_unlock_bh(&ls->ls_recv_active); 662 663 /* 664 * This in_recovery lock does two things: 665 * 1) Keeps this function from returning until all threads are out 666 * of locking routines and locking is truly stopped. 667 * 2) Keeps any new requests from being processed until it's unlocked 668 * when recovery is complete. 669 */ 670 671 if (new) { 672 set_bit(LSFL_RECOVER_DOWN, &ls->ls_flags); 673 wake_up_process(ls->ls_recoverd_task); 674 wait_event(ls->ls_recover_lock_wait, 675 test_bit(LSFL_RECOVER_LOCK, &ls->ls_flags)); 676 } 677 678 /* 679 * The recoverd suspend/resume makes sure that dlm_recoverd (if 680 * running) has noticed RECOVER_STOP above and quit processing the 681 * previous recovery. 682 */ 683 684 dlm_recoverd_suspend(ls); 685 686 spin_lock_bh(&ls->ls_recover_lock); 687 kfree(ls->ls_slots); 688 ls->ls_slots = NULL; 689 ls->ls_num_slots = 0; 690 ls->ls_slots_size = 0; 691 ls->ls_recover_status = 0; 692 spin_unlock_bh(&ls->ls_recover_lock); 693 694 dlm_recoverd_resume(ls); 695 696 if (!ls->ls_recover_begin) 697 ls->ls_recover_begin = jiffies; 698 699 /* call recover_prep ops only once and not multiple times 700 * for each possible dlm_ls_stop() when recovery is already 701 * stopped. 702 * 703 * If we successful was able to clear LSFL_RUNNING bit and 704 * it was set we know it is the first dlm_ls_stop() call. 705 */ 706 if (new) 707 dlm_lsop_recover_prep(ls); 708 709 return 0; 710 } 711 712 int dlm_ls_start(struct dlm_ls *ls) 713 { 714 struct dlm_recover *rv, *rv_old; 715 struct dlm_config_node *nodes = NULL; 716 int error, count; 717 718 rv = kzalloc(sizeof(*rv), GFP_NOFS); 719 if (!rv) 720 return -ENOMEM; 721 722 error = dlm_config_nodes(ls->ls_name, &nodes, &count); 723 if (error < 0) 724 goto fail_rv; 725 726 spin_lock_bh(&ls->ls_recover_lock); 727 728 /* the lockspace needs to be stopped before it can be started */ 729 730 if (!dlm_locking_stopped(ls)) { 731 spin_unlock_bh(&ls->ls_recover_lock); 732 log_error(ls, "start ignored: lockspace running"); 733 error = -EINVAL; 734 goto fail; 735 } 736 737 rv->nodes = nodes; 738 rv->nodes_count = count; 739 rv->seq = ++ls->ls_recover_seq; 740 rv_old = ls->ls_recover_args; 741 ls->ls_recover_args = rv; 742 spin_unlock_bh(&ls->ls_recover_lock); 743 744 if (rv_old) { 745 log_error(ls, "unused recovery %llx %d", 746 (unsigned long long)rv_old->seq, rv_old->nodes_count); 747 kfree(rv_old->nodes); 748 kfree(rv_old); 749 } 750 751 set_bit(LSFL_RECOVER_WORK, &ls->ls_flags); 752 wake_up_process(ls->ls_recoverd_task); 753 return 0; 754 755 fail: 756 kfree(nodes); 757 fail_rv: 758 kfree(rv); 759 return error; 760 } 761 762