fs/btrfs/ref-verify.c

   1 // SPDX-License-Identifier: GPL-2.0
   2 /*
   3  * Copyright (C) 2014 Facebook.  All rights reserved.
   4  */
   5
   6 #include <linux/sched.h>
   7 #include <linux/stacktrace.h>
   8 #include "messages.h"
   9 #include "ctree.h"
  10 #include "disk-io.h"
  11 #include "locking.h"
  12 #include "delayed-ref.h"
  13 #include "ref-verify.h"
  14 #include "fs.h"
  15 #include "accessors.h"
  16
  17 /*
  18  * Used to keep track the roots and number of refs each root has for a given
  19  * bytenr.  This just tracks the number of direct references, no shared
  20  * references.
  21  */
  22 struct root_entry {
  23         u64 root_objectid;
  24         u64 num_refs;
  25         struct rb_node node;
  26 };
  27
  28 /*
  29  * These are meant to represent what should exist in the extent tree, these can
  30  * be used to verify the extent tree is consistent as these should all match
  31  * what the extent tree says.
  32  */
  33 struct ref_entry {
  34         u64 root_objectid;
  35         u64 parent;
  36         u64 owner;
  37         u64 offset;
  38         u64 num_refs;
  39         struct rb_node node;
  40 };
  41
  42 #define MAX_TRACE       16
  43
  44 /*
  45  * Whenever we add/remove a reference we record the action.  The action maps
  46  * back to the delayed ref action.  We hold the ref we are changing in the
  47  * action so we can account for the history properly, and we record the root we
  48  * were called with since it could be different from ref_root.  We also store
  49  * stack traces because that's how I roll.
  50  */
  51 struct ref_action {
  52         int action;
  53         u64 root;
  54         struct ref_entry ref;
  55         struct list_head list;
  56         unsigned long trace[MAX_TRACE];
  57         unsigned int trace_len;
  58 };
  59
  60 /*
  61  * One of these for every block we reference, it holds the roots and references
  62  * to it as well as all of the ref actions that have occurred to it.  We never
  63  * free it until we unmount the file system in order to make sure re-allocations
  64  * are happening properly.
  65  */
  66 struct block_entry {
  67         u64 bytenr;
  68         u64 len;
  69         u64 num_refs;
  70         int metadata;
  71         int from_disk;
  72         struct rb_root roots;
  73         struct rb_root refs;
  74         struct rb_node node;
  75         struct list_head actions;
  76 };
  77
  78 static struct block_entry *insert_block_entry(struct rb_root *root,
  79                                               struct block_entry *be)
  80 {
  81         struct rb_node **p = &root->rb_node;
  82         struct rb_node *parent_node = NULL;
  83         struct block_entry *entry;
  84
  85         while (*p) {
  86                 parent_node = *p;
  87                 entry = rb_entry(parent_node, struct block_entry, node);
  88                 if (entry->bytenr > be->bytenr)
  89                         p = &(*p)->rb_left;
  90                 else if (entry->bytenr < be->bytenr)
  91                         p = &(*p)->rb_right;
  92                 else
  93                         return entry;
  94         }
  95
  96         rb_link_node(&be->node, parent_node, p);
  97         rb_insert_color(&be->node, root);
  98         return NULL;
  99 }
 100
 101 static struct block_entry *lookup_block_entry(struct rb_root *root, u64 bytenr)
 102 {
 103         struct rb_node *n;
 104         struct block_entry *entry = NULL;
 105
 106         n = root->rb_node;
 107         while (n) {
 108                 entry = rb_entry(n, struct block_entry, node);
 109                 if (entry->bytenr < bytenr)
 110                         n = n->rb_right;
 111                 else if (entry->bytenr > bytenr)
 112                         n = n->rb_left;
 113                 else
 114                         return entry;
 115         }
 116         return NULL;
 117 }
 118
 119 static struct root_entry *insert_root_entry(struct rb_root *root,
 120                                             struct root_entry *re)
 121 {
 122         struct rb_node **p = &root->rb_node;
 123         struct rb_node *parent_node = NULL;
 124         struct root_entry *entry;
 125
 126         while (*p) {
 127                 parent_node = *p;
 128                 entry = rb_entry(parent_node, struct root_entry, node);
 129                 if (entry->root_objectid > re->root_objectid)
 130                         p = &(*p)->rb_left;
 131                 else if (entry->root_objectid < re->root_objectid)
 132                         p = &(*p)->rb_right;
 133                 else
 134                         return entry;
 135         }
 136
 137         rb_link_node(&re->node, parent_node, p);
 138         rb_insert_color(&re->node, root);
 139         return NULL;
 140
 141 }
 142
 143 static int comp_refs(struct ref_entry *ref1, struct ref_entry *ref2)
 144 {
 145         if (ref1->root_objectid < ref2->root_objectid)
 146                 return -1;
 147         if (ref1->root_objectid > ref2->root_objectid)
 148                 return 1;
 149         if (ref1->parent < ref2->parent)
 150                 return -1;
 151         if (ref1->parent > ref2->parent)
 152                 return 1;
 153         if (ref1->owner < ref2->owner)
 154                 return -1;
 155         if (ref1->owner > ref2->owner)
 156                 return 1;
 157         if (ref1->offset < ref2->offset)
 158                 return -1;
 159         if (ref1->offset > ref2->offset)
 160                 return 1;
 161         return 0;
 162 }
 163
 164 static struct ref_entry *insert_ref_entry(struct rb_root *root,
 165                                           struct ref_entry *ref)
 166 {
 167         struct rb_node **p = &root->rb_node;
 168         struct rb_node *parent_node = NULL;
 169         struct ref_entry *entry;
 170         int cmp;
 171
 172         while (*p) {
 173                 parent_node = *p;
 174                 entry = rb_entry(parent_node, struct ref_entry, node);
 175                 cmp = comp_refs(entry, ref);
 176                 if (cmp > 0)
 177                         p = &(*p)->rb_left;
 178                 else if (cmp < 0)
 179                         p = &(*p)->rb_right;
 180                 else
 181                         return entry;
 182         }
 183
 184         rb_link_node(&ref->node, parent_node, p);
 185         rb_insert_color(&ref->node, root);
 186         return NULL;
 187
 188 }
 189
 190 static struct root_entry *lookup_root_entry(struct rb_root *root, u64 objectid)
 191 {
 192         struct rb_node *n;
 193         struct root_entry *entry = NULL;
 194
 195         n = root->rb_node;
 196         while (n) {
 197                 entry = rb_entry(n, struct root_entry, node);
 198                 if (entry->root_objectid < objectid)
 199                         n = n->rb_right;
 200                 else if (entry->root_objectid > objectid)
 201                         n = n->rb_left;
 202                 else
 203                         return entry;
 204         }
 205         return NULL;
 206 }
 207
 208 #ifdef CONFIG_STACKTRACE
 209 static void __save_stack_trace(struct ref_action *ra)
 210 {
 211         ra->trace_len = stack_trace_save(ra->trace, MAX_TRACE, 2);
 212 }
 213
 214 static void __print_stack_trace(struct btrfs_fs_info *fs_info,
 215                                 struct ref_action *ra)
 216 {
 217         if (ra->trace_len == 0) {
 218                 btrfs_err(fs_info, "  ref-verify: no stacktrace");
 219                 return;
 220         }
 221         stack_trace_print(ra->trace, ra->trace_len, 2);
 222 }
 223 #else
 224 static inline void __save_stack_trace(struct ref_action *ra)
 225 {
 226 }
 227
 228 static inline void __print_stack_trace(struct btrfs_fs_info *fs_info,
 229                                        struct ref_action *ra)
 230 {
 231         btrfs_err(fs_info, "  ref-verify: no stacktrace support");
 232 }
 233 #endif
 234
 235 static void free_block_entry(struct block_entry *be)
 236 {
 237         struct root_entry *re;
 238         struct ref_entry *ref;
 239         struct ref_action *ra;
 240         struct rb_node *n;
 241
 242         while ((n = rb_first(&be->roots))) {
 243                 re = rb_entry(n, struct root_entry, node);
 244                 rb_erase(&re->node, &be->roots);
 245                 kfree(re);
 246         }
 247
 248         while((n = rb_first(&be->refs))) {
 249                 ref = rb_entry(n, struct ref_entry, node);
 250                 rb_erase(&ref->node, &be->refs);
 251                 kfree(ref);
 252         }
 253
 254         while (!list_empty(&be->actions)) {
 255                 ra = list_first_entry(&be->actions, struct ref_action,
 256                                       list);
 257                 list_del(&ra->list);
 258                 kfree(ra);
 259         }
 260         kfree(be);
 261 }
 262
 263 static struct block_entry *add_block_entry(struct btrfs_fs_info *fs_info,
 264                                            u64 bytenr, u64 len,
 265                                            u64 root_objectid)
 266 {
 267         struct block_entry *be = NULL, *exist;
 268         struct root_entry *re = NULL;
 269
 270         re = kzalloc(sizeof(struct root_entry), GFP_NOFS);
 271         be = kzalloc(sizeof(struct block_entry), GFP_NOFS);
 272         if (!be || !re) {
 273                 kfree(re);
 274                 kfree(be);
 275                 return ERR_PTR(-ENOMEM);
 276         }
 277         be->bytenr = bytenr;
 278         be->len = len;
 279
 280         re->root_objectid = root_objectid;
 281         re->num_refs = 0;
 282
 283         spin_lock(&fs_info->ref_verify_lock);
 284         exist = insert_block_entry(&fs_info->block_tree, be);
 285         if (exist) {
 286                 if (root_objectid) {
 287                         struct root_entry *exist_re;
 288
 289                         exist_re = insert_root_entry(&exist->roots, re);
 290                         if (exist_re)
 291                                 kfree(re);
 292                 } else {
 293                         kfree(re);
 294                 }
 295                 kfree(be);
 296                 return exist;
 297         }
 298
 299         be->num_refs = 0;
 300         be->metadata = 0;
 301         be->from_disk = 0;
 302         be->roots = RB_ROOT;
 303         be->refs = RB_ROOT;
 304         INIT_LIST_HEAD(&be->actions);
 305         if (root_objectid)
 306                 insert_root_entry(&be->roots, re);
 307         else
 308                 kfree(re);
 309         return be;
 310 }
 311
 312 static int add_tree_block(struct btrfs_fs_info *fs_info, u64 ref_root,
 313                           u64 parent, u64 bytenr, int level)
 314 {
 315         struct block_entry *be;
 316         struct root_entry *re;
 317         struct ref_entry *ref = NULL, *exist;
 318
 319         ref = kmalloc(sizeof(struct ref_entry), GFP_NOFS);
 320         if (!ref)
 321                 return -ENOMEM;
 322
 323         if (parent)
 324                 ref->root_objectid = 0;
 325         else
 326                 ref->root_objectid = ref_root;
 327         ref->parent = parent;
 328         ref->owner = level;
 329         ref->offset = 0;
 330         ref->num_refs = 1;
 331
 332         be = add_block_entry(fs_info, bytenr, fs_info->nodesize, ref_root);
 333         if (IS_ERR(be)) {
 334                 kfree(ref);
 335                 return PTR_ERR(be);
 336         }
 337         be->num_refs++;
 338         be->from_disk = 1;
 339         be->metadata = 1;
 340
 341         if (!parent) {
 342                 ASSERT(ref_root);
 343                 re = lookup_root_entry(&be->roots, ref_root);
 344                 ASSERT(re);
 345                 re->num_refs++;
 346         }
 347         exist = insert_ref_entry(&be->refs, ref);
 348         if (exist) {
 349                 exist->num_refs++;
 350                 kfree(ref);
 351         }
 352         spin_unlock(&fs_info->ref_verify_lock);
 353
 354         return 0;
 355 }
 356
 357 static int add_shared_data_ref(struct btrfs_fs_info *fs_info,
 358                                u64 parent, u32 num_refs, u64 bytenr,
 359                                u64 num_bytes)
 360 {
 361         struct block_entry *be;
 362         struct ref_entry *ref;
 363
 364         ref = kzalloc(sizeof(struct ref_entry), GFP_NOFS);
 365         if (!ref)
 366                 return -ENOMEM;
 367         be = add_block_entry(fs_info, bytenr, num_bytes, 0);
 368         if (IS_ERR(be)) {
 369                 kfree(ref);
 370                 return PTR_ERR(be);
 371         }
 372         be->num_refs += num_refs;
 373
 374         ref->parent = parent;
 375         ref->num_refs = num_refs;
 376         if (insert_ref_entry(&be->refs, ref)) {
 377                 spin_unlock(&fs_info->ref_verify_lock);
 378                 btrfs_err(fs_info, "existing shared ref when reading from disk?");
 379                 kfree(ref);
 380                 return -EINVAL;
 381         }
 382         spin_unlock(&fs_info->ref_verify_lock);
 383         return 0;
 384 }
 385
 386 static int add_extent_data_ref(struct btrfs_fs_info *fs_info,
 387                                struct extent_buffer *leaf,
 388                                struct btrfs_extent_data_ref *dref,
 389                                u64 bytenr, u64 num_bytes)
 390 {
 391         struct block_entry *be;
 392         struct ref_entry *ref;
 393         struct root_entry *re;
 394         u64 ref_root = btrfs_extent_data_ref_root(leaf, dref);
 395         u64 owner = btrfs_extent_data_ref_objectid(leaf, dref);
 396         u64 offset = btrfs_extent_data_ref_offset(leaf, dref);
 397         u32 num_refs = btrfs_extent_data_ref_count(leaf, dref);
 398
 399         ref = kzalloc(sizeof(struct ref_entry), GFP_NOFS);
 400         if (!ref)
 401                 return -ENOMEM;
 402         be = add_block_entry(fs_info, bytenr, num_bytes, ref_root);
 403         if (IS_ERR(be)) {
 404                 kfree(ref);
 405                 return PTR_ERR(be);
 406         }
 407         be->num_refs += num_refs;
 408
 409         ref->parent = 0;
 410         ref->owner = owner;
 411         ref->root_objectid = ref_root;
 412         ref->offset = offset;
 413         ref->num_refs = num_refs;
 414         if (insert_ref_entry(&be->refs, ref)) {
 415                 spin_unlock(&fs_info->ref_verify_lock);
 416                 btrfs_err(fs_info, "existing ref when reading from disk?");
 417                 kfree(ref);
 418                 return -EINVAL;
 419         }
 420
 421         re = lookup_root_entry(&be->roots, ref_root);
 422         if (!re) {
 423                 spin_unlock(&fs_info->ref_verify_lock);
 424                 btrfs_err(fs_info, "missing root in new block entry?");
 425                 return -EINVAL;
 426         }
 427         re->num_refs += num_refs;
 428         spin_unlock(&fs_info->ref_verify_lock);
 429         return 0;
 430 }
 431
 432 static int process_extent_item(struct btrfs_fs_info *fs_info,
 433                                struct btrfs_path *path, struct btrfs_key *key,
 434                                int slot, int *tree_block_level)
 435 {
 436         struct btrfs_extent_item *ei;
 437         struct btrfs_extent_inline_ref *iref;
 438         struct btrfs_extent_data_ref *dref;
 439         struct btrfs_shared_data_ref *sref;
 440         struct extent_buffer *leaf = path->nodes[0];
 441         u32 item_size = btrfs_item_size(leaf, slot);
 442         unsigned long end, ptr;
 443         u64 offset, flags, count;
 444         int type, ret;
 445
 446         ei = btrfs_item_ptr(leaf, slot, struct btrfs_extent_item);
 447         flags = btrfs_extent_flags(leaf, ei);
 448
 449         if ((key->type == BTRFS_EXTENT_ITEM_KEY) &&
 450             flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
 451                 struct btrfs_tree_block_info *info;
 452
 453                 info = (struct btrfs_tree_block_info *)(ei + 1);
 454                 *tree_block_level = btrfs_tree_block_level(leaf, info);
 455                 iref = (struct btrfs_extent_inline_ref *)(info + 1);
 456         } else {
 457                 if (key->type == BTRFS_METADATA_ITEM_KEY)
 458                         *tree_block_level = key->offset;
 459                 iref = (struct btrfs_extent_inline_ref *)(ei + 1);
 460         }
 461
 462         ptr = (unsigned long)iref;
 463         end = (unsigned long)ei + item_size;
 464         while (ptr < end) {
 465                 iref = (struct btrfs_extent_inline_ref *)ptr;
 466                 type = btrfs_extent_inline_ref_type(leaf, iref);
 467                 offset = btrfs_extent_inline_ref_offset(leaf, iref);
 468                 switch (type) {
 469                 case BTRFS_TREE_BLOCK_REF_KEY:
 470                         ret = add_tree_block(fs_info, offset, 0, key->objectid,
 471                                              *tree_block_level);
 472                         break;
 473                 case BTRFS_SHARED_BLOCK_REF_KEY:
 474                         ret = add_tree_block(fs_info, 0, offset, key->objectid,
 475                                              *tree_block_level);
 476                         break;
 477                 case BTRFS_EXTENT_DATA_REF_KEY:
 478                         dref = (struct btrfs_extent_data_ref *)(&iref->offset);
 479                         ret = add_extent_data_ref(fs_info, leaf, dref,
 480                                                   key->objectid, key->offset);
 481                         break;
 482                 case BTRFS_SHARED_DATA_REF_KEY:
 483                         sref = (struct btrfs_shared_data_ref *)(iref + 1);
 484                         count = btrfs_shared_data_ref_count(leaf, sref);
 485                         ret = add_shared_data_ref(fs_info, offset, count,
 486                                                   key->objectid, key->offset);
 487                         break;
 488                 default:
 489                         btrfs_err(fs_info, "invalid key type in iref");
 490                         ret = -EINVAL;
 491                         break;
 492                 }
 493                 if (ret)
 494                         break;
 495                 ptr += btrfs_extent_inline_ref_size(type);
 496         }
 497         return ret;
 498 }
 499
 500 static int process_leaf(struct btrfs_root *root,
 501                         struct btrfs_path *path, u64 *bytenr, u64 *num_bytes,
 502                         int *tree_block_level)
 503 {
 504         struct btrfs_fs_info *fs_info = root->fs_info;
 505         struct extent_buffer *leaf = path->nodes[0];
 506         struct btrfs_extent_data_ref *dref;
 507         struct btrfs_shared_data_ref *sref;
 508         u32 count;
 509         int i = 0, ret = 0;
 510         struct btrfs_key key;
 511         int nritems = btrfs_header_nritems(leaf);
 512
 513         for (i = 0; i < nritems; i++) {
 514                 btrfs_item_key_to_cpu(leaf, &key, i);
 515                 switch (key.type) {
 516                 case BTRFS_EXTENT_ITEM_KEY:
 517                         *num_bytes = key.offset;
 518                         fallthrough;
 519                 case BTRFS_METADATA_ITEM_KEY:
 520                         *bytenr = key.objectid;
 521                         ret = process_extent_item(fs_info, path, &key, i,
 522                                                   tree_block_level);
 523                         break;
 524                 case BTRFS_TREE_BLOCK_REF_KEY:
 525                         ret = add_tree_block(fs_info, key.offset, 0,
 526                                              key.objectid, *tree_block_level);
 527                         break;
 528                 case BTRFS_SHARED_BLOCK_REF_KEY:
 529                         ret = add_tree_block(fs_info, 0, key.offset,
 530                                              key.objectid, *tree_block_level);
 531                         break;
 532                 case BTRFS_EXTENT_DATA_REF_KEY:
 533                         dref = btrfs_item_ptr(leaf, i,
 534                                               struct btrfs_extent_data_ref);
 535                         ret = add_extent_data_ref(fs_info, leaf, dref, *bytenr,
 536                                                   *num_bytes);
 537                         break;
 538                 case BTRFS_SHARED_DATA_REF_KEY:
 539                         sref = btrfs_item_ptr(leaf, i,
 540                                               struct btrfs_shared_data_ref);
 541                         count = btrfs_shared_data_ref_count(leaf, sref);
 542                         ret = add_shared_data_ref(fs_info, key.offset, count,
 543                                                   *bytenr, *num_bytes);
 544                         break;
 545                 default:
 546                         break;
 547                 }
 548                 if (ret)
 549                         break;
 550         }
 551         return ret;
 552 }
 553
 554 /* Walk down to the leaf from the given level */
 555 static int walk_down_tree(struct btrfs_root *root, struct btrfs_path *path,
 556                           int level, u64 *bytenr, u64 *num_bytes,
 557                           int *tree_block_level)
 558 {
 559         struct extent_buffer *eb;
 560         int ret = 0;
 561
 562         while (level >= 0) {
 563                 if (level) {
 564                         eb = btrfs_read_node_slot(path->nodes[level],
 565                                                   path->slots[level]);
 566                         if (IS_ERR(eb))
 567                                 return PTR_ERR(eb);
 568                         btrfs_tree_read_lock(eb);
 569                         path->nodes[level-1] = eb;
 570                         path->slots[level-1] = 0;
 571                         path->locks[level-1] = BTRFS_READ_LOCK;
 572                 } else {
 573                         ret = process_leaf(root, path, bytenr, num_bytes,
 574                                            tree_block_level);
 575                         if (ret)
 576                                 break;
 577                 }
 578                 level--;
 579         }
 580         return ret;
 581 }
 582
 583 /* Walk up to the next node that needs to be processed */
 584 static int walk_up_tree(struct btrfs_path *path, int *level)
 585 {
 586         int l;
 587
 588         for (l = 0; l < BTRFS_MAX_LEVEL; l++) {
 589                 if (!path->nodes[l])
 590                         continue;
 591                 if (l) {
 592                         path->slots[l]++;
 593                         if (path->slots[l] <
 594                             btrfs_header_nritems(path->nodes[l])) {
 595                                 *level = l;
 596                                 return 0;
 597                         }
 598                 }
 599                 btrfs_tree_unlock_rw(path->nodes[l], path->locks[l]);
 600                 free_extent_buffer(path->nodes[l]);
 601                 path->nodes[l] = NULL;
 602                 path->slots[l] = 0;
 603                 path->locks[l] = 0;
 604         }
 605
 606         return 1;
 607 }
 608
 609 static void dump_ref_action(struct btrfs_fs_info *fs_info,
 610                             struct ref_action *ra)
 611 {
 612         btrfs_err(fs_info,
 613 "  Ref action %d, root %llu, ref_root %llu, parent %llu, owner %llu, offset %llu, num_refs %llu",
 614                   ra->action, ra->root, ra->ref.root_objectid, ra->ref.parent,
 615                   ra->ref.owner, ra->ref.offset, ra->ref.num_refs);
 616         __print_stack_trace(fs_info, ra);
 617 }
 618
 619 /*
 620  * Dumps all the information from the block entry to printk, it's going to be
 621  * awesome.
 622  */
 623 static void dump_block_entry(struct btrfs_fs_info *fs_info,
 624                              struct block_entry *be)
 625 {
 626         struct ref_entry *ref;
 627         struct root_entry *re;
 628         struct ref_action *ra;
 629         struct rb_node *n;
 630
 631         btrfs_err(fs_info,
 632 "dumping block entry [%llu %llu], num_refs %llu, metadata %d, from disk %d",
 633                   be->bytenr, be->len, be->num_refs, be->metadata,
 634                   be->from_disk);
 635
 636         for (n = rb_first(&be->refs); n; n = rb_next(n)) {
 637                 ref = rb_entry(n, struct ref_entry, node);
 638                 btrfs_err(fs_info,
 639 "  ref root %llu, parent %llu, owner %llu, offset %llu, num_refs %llu",
 640                           ref->root_objectid, ref->parent, ref->owner,
 641                           ref->offset, ref->num_refs);
 642         }
 643
 644         for (n = rb_first(&be->roots); n; n = rb_next(n)) {
 645                 re = rb_entry(n, struct root_entry, node);
 646                 btrfs_err(fs_info, "  root entry %llu, num_refs %llu",
 647                           re->root_objectid, re->num_refs);
 648         }
 649
 650         list_for_each_entry(ra, &be->actions, list)
 651                 dump_ref_action(fs_info, ra);
 652 }
 653
 654 /*
 655  * btrfs_ref_tree_mod: called when we modify a ref for a bytenr
 656  *
 657  * This will add an action item to the given bytenr and do sanity checks to make
 658  * sure we haven't messed something up.  If we are making a new allocation and
 659  * this block entry has history we will delete all previous actions as long as
 660  * our sanity checks pass as they are no longer needed.
 661  */
 662 int btrfs_ref_tree_mod(struct btrfs_fs_info *fs_info,
 663                        struct btrfs_ref *generic_ref)
 664 {
 665         struct ref_entry *ref = NULL, *exist;
 666         struct ref_action *ra = NULL;
 667         struct block_entry *be = NULL;
 668         struct root_entry *re = NULL;
 669         int action = generic_ref->action;
 670         int ret = 0;
 671         bool metadata;
 672         u64 bytenr = generic_ref->bytenr;
 673         u64 num_bytes = generic_ref->len;
 674         u64 parent = generic_ref->parent;
 675         u64 ref_root = 0;
 676         u64 owner = 0;
 677         u64 offset = 0;
 678
 679         if (!btrfs_test_opt(fs_info, REF_VERIFY))
 680                 return 0;
 681
 682         if (generic_ref->type == BTRFS_REF_METADATA) {
 683                 if (!parent)
 684                         ref_root = generic_ref->tree_ref.owning_root;
 685                 owner = generic_ref->tree_ref.level;
 686         } else if (!parent) {
 687                 ref_root = generic_ref->data_ref.owning_root;
 688                 owner = generic_ref->data_ref.ino;
 689                 offset = generic_ref->data_ref.offset;
 690         }
 691         metadata = owner < BTRFS_FIRST_FREE_OBJECTID;
 692
 693         ref = kzalloc(sizeof(struct ref_entry), GFP_NOFS);
 694         ra = kmalloc(sizeof(struct ref_action), GFP_NOFS);
 695         if (!ra || !ref) {
 696                 kfree(ref);
 697                 kfree(ra);
 698                 ret = -ENOMEM;
 699                 goto out;
 700         }
 701
 702         ref->parent = parent;
 703         ref->owner = owner;
 704         ref->root_objectid = ref_root;
 705         ref->offset = offset;
 706         ref->num_refs = (action == BTRFS_DROP_DELAYED_REF) ? -1 : 1;
 707
 708         memcpy(&ra->ref, ref, sizeof(struct ref_entry));
 709         /*
 710          * Save the extra info from the delayed ref in the ref action to make it
 711          * easier to figure out what is happening.  The real ref's we add to the
 712          * ref tree need to reflect what we save on disk so it matches any
 713          * on-disk refs we pre-loaded.
 714          */
 715         ra->ref.owner = owner;
 716         ra->ref.offset = offset;
 717         ra->ref.root_objectid = ref_root;
 718         __save_stack_trace(ra);
 719
 720         INIT_LIST_HEAD(&ra->list);
 721         ra->action = action;
 722         ra->root = generic_ref->real_root;
 723
 724         /*
 725          * This is an allocation, preallocate the block_entry in case we haven't
 726          * used it before.
 727          */
 728         ret = -EINVAL;
 729         if (action == BTRFS_ADD_DELAYED_EXTENT) {
 730                 /*
 731                  * For subvol_create we'll just pass in whatever the parent root
 732                  * is and the new root objectid, so let's not treat the passed
 733                  * in root as if it really has a ref for this bytenr.
 734                  */
 735                 be = add_block_entry(fs_info, bytenr, num_bytes, ref_root);
 736                 if (IS_ERR(be)) {
 737                         kfree(ref);
 738                         kfree(ra);
 739                         ret = PTR_ERR(be);
 740                         goto out;
 741                 }
 742                 be->num_refs++;
 743                 if (metadata)
 744                         be->metadata = 1;
 745
 746                 if (be->num_refs != 1) {
 747                         btrfs_err(fs_info,
 748                         "re-allocated a block that still has references to it!");
 749                         dump_block_entry(fs_info, be);
 750                         dump_ref_action(fs_info, ra);
 751                         kfree(ref);
 752                         kfree(ra);
 753                         goto out_unlock;
 754                 }
 755
 756                 while (!list_empty(&be->actions)) {
 757                         struct ref_action *tmp;
 758
 759                         tmp = list_first_entry(&be->actions, struct ref_action,
 760                                                list);
 761                         list_del(&tmp->list);
 762                         kfree(tmp);
 763                 }
 764         } else {
 765                 struct root_entry *tmp;
 766
 767                 if (!parent) {
 768                         re = kmalloc(sizeof(struct root_entry), GFP_NOFS);
 769                         if (!re) {
 770                                 kfree(ref);
 771                                 kfree(ra);
 772                                 ret = -ENOMEM;
 773                                 goto out;
 774                         }
 775                         /*
 776                          * This is the root that is modifying us, so it's the
 777                          * one we want to lookup below when we modify the
 778                          * re->num_refs.
 779                          */
 780                         ref_root = generic_ref->real_root;
 781                         re->root_objectid = generic_ref->real_root;
 782                         re->num_refs = 0;
 783                 }
 784
 785                 spin_lock(&fs_info->ref_verify_lock);
 786                 be = lookup_block_entry(&fs_info->block_tree, bytenr);
 787                 if (!be) {
 788                         btrfs_err(fs_info,
 789 "trying to do action %d to bytenr %llu num_bytes %llu but there is no existing entry!",
 790                                   action, bytenr, num_bytes);
 791                         dump_ref_action(fs_info, ra);
 792                         kfree(ref);
 793                         kfree(ra);
 794                         goto out_unlock;
 795                 } else if (be->num_refs == 0) {
 796                         btrfs_err(fs_info,
 797                 "trying to do action %d for a bytenr that has 0 total references",
 798                                 action);
 799                         dump_block_entry(fs_info, be);
 800                         dump_ref_action(fs_info, ra);
 801                         kfree(ref);
 802                         kfree(ra);
 803                         goto out_unlock;
 804                 }
 805
 806                 if (!parent) {
 807                         tmp = insert_root_entry(&be->roots, re);
 808                         if (tmp) {
 809                                 kfree(re);
 810                                 re = tmp;
 811                         }
 812                 }
 813         }
 814
 815         exist = insert_ref_entry(&be->refs, ref);
 816         if (exist) {
 817                 if (action == BTRFS_DROP_DELAYED_REF) {
 818                         if (exist->num_refs == 0) {
 819                                 btrfs_err(fs_info,
 820 "dropping a ref for a existing root that doesn't have a ref on the block");
 821                                 dump_block_entry(fs_info, be);
 822                                 dump_ref_action(fs_info, ra);
 823                                 kfree(ref);
 824                                 kfree(ra);
 825                                 goto out_unlock;
 826                         }
 827                         exist->num_refs--;
 828                         if (exist->num_refs == 0) {
 829                                 rb_erase(&exist->node, &be->refs);
 830                                 kfree(exist);
 831                         }
 832                 } else if (!be->metadata) {
 833                         exist->num_refs++;
 834                 } else {
 835                         btrfs_err(fs_info,
 836 "attempting to add another ref for an existing ref on a tree block");
 837                         dump_block_entry(fs_info, be);
 838                         dump_ref_action(fs_info, ra);
 839                         kfree(ref);
 840                         kfree(ra);
 841                         goto out_unlock;
 842                 }
 843                 kfree(ref);
 844         } else {
 845                 if (action == BTRFS_DROP_DELAYED_REF) {
 846                         btrfs_err(fs_info,
 847 "dropping a ref for a root that doesn't have a ref on the block");
 848                         dump_block_entry(fs_info, be);
 849                         dump_ref_action(fs_info, ra);
 850                         kfree(ref);
 851                         kfree(ra);
 852                         goto out_unlock;
 853                 }
 854         }
 855
 856         if (!parent && !re) {
 857                 re = lookup_root_entry(&be->roots, ref_root);
 858                 if (!re) {
 859                         /*
 860                          * This shouldn't happen because we will add our re
 861                          * above when we lookup the be with !parent, but just in
 862                          * case catch this case so we don't panic because I
 863                          * didn't think of some other corner case.
 864                          */
 865                         btrfs_err(fs_info, "failed to find root %llu for %llu",
 866                                   generic_ref->real_root, be->bytenr);
 867                         dump_block_entry(fs_info, be);
 868                         dump_ref_action(fs_info, ra);
 869                         kfree(ra);
 870                         goto out_unlock;
 871                 }
 872         }
 873         if (action == BTRFS_DROP_DELAYED_REF) {
 874                 if (re)
 875                         re->num_refs--;
 876                 be->num_refs--;
 877         } else if (action == BTRFS_ADD_DELAYED_REF) {
 878                 be->num_refs++;
 879                 if (re)
 880                         re->num_refs++;
 881         }
 882         list_add_tail(&ra->list, &be->actions);
 883         ret = 0;
 884 out_unlock:
 885         spin_unlock(&fs_info->ref_verify_lock);
 886 out:
 887         if (ret)
 888                 btrfs_clear_opt(fs_info->mount_opt, REF_VERIFY);
 889         return ret;
 890 }
 891
 892 /* Free up the ref cache */
 893 void btrfs_free_ref_cache(struct btrfs_fs_info *fs_info)
 894 {
 895         struct block_entry *be;
 896         struct rb_node *n;
 897
 898         if (!btrfs_test_opt(fs_info, REF_VERIFY))
 899                 return;
 900
 901         spin_lock(&fs_info->ref_verify_lock);
 902         while ((n = rb_first(&fs_info->block_tree))) {
 903                 be = rb_entry(n, struct block_entry, node);
 904                 rb_erase(&be->node, &fs_info->block_tree);
 905                 free_block_entry(be);
 906                 cond_resched_lock(&fs_info->ref_verify_lock);
 907         }
 908         spin_unlock(&fs_info->ref_verify_lock);
 909 }
 910
 911 void btrfs_free_ref_tree_range(struct btrfs_fs_info *fs_info, u64 start,
 912                                u64 len)
 913 {
 914         struct block_entry *be = NULL, *entry;
 915         struct rb_node *n;
 916
 917         if (!btrfs_test_opt(fs_info, REF_VERIFY))
 918                 return;
 919
 920         spin_lock(&fs_info->ref_verify_lock);
 921         n = fs_info->block_tree.rb_node;
 922         while (n) {
 923                 entry = rb_entry(n, struct block_entry, node);
 924                 if (entry->bytenr < start) {
 925                         n = n->rb_right;
 926                 } else if (entry->bytenr > start) {
 927                         n = n->rb_left;
 928                 } else {
 929                         be = entry;
 930                         break;
 931                 }
 932                 /* We want to get as close to start as possible */
 933                 if (be == NULL ||
 934                     (entry->bytenr < start && be->bytenr > start) ||
 935                     (entry->bytenr < start && entry->bytenr > be->bytenr))
 936                         be = entry;
 937         }
 938
 939         /*
 940          * Could have an empty block group, maybe have something to check for
 941          * this case to verify we were actually empty?
 942          */
 943         if (!be) {
 944                 spin_unlock(&fs_info->ref_verify_lock);
 945                 return;
 946         }
 947
 948         n = &be->node;
 949         while (n) {
 950                 be = rb_entry(n, struct block_entry, node);
 951                 n = rb_next(n);
 952                 if (be->bytenr < start && be->bytenr + be->len > start) {
 953                         btrfs_err(fs_info,
 954                                 "block entry overlaps a block group [%llu,%llu]!",
 955                                 start, len);
 956                         dump_block_entry(fs_info, be);
 957                         continue;
 958                 }
 959                 if (be->bytenr < start)
 960                         continue;
 961                 if (be->bytenr >= start + len)
 962                         break;
 963                 if (be->bytenr + be->len > start + len) {
 964                         btrfs_err(fs_info,
 965                                 "block entry overlaps a block group [%llu,%llu]!",
 966                                 start, len);
 967                         dump_block_entry(fs_info, be);
 968                 }
 969                 rb_erase(&be->node, &fs_info->block_tree);
 970                 free_block_entry(be);
 971         }
 972         spin_unlock(&fs_info->ref_verify_lock);
 973 }
 974
 975 /* Walk down all roots and build the ref tree, meant to be called at mount */
 976 int btrfs_build_ref_tree(struct btrfs_fs_info *fs_info)
 977 {
 978         struct btrfs_root *extent_root;
 979         struct btrfs_path *path;
 980         struct extent_buffer *eb;
 981         int tree_block_level = 0;
 982         u64 bytenr = 0, num_bytes = 0;
 983         int ret, level;
 984
 985         if (!btrfs_test_opt(fs_info, REF_VERIFY))
 986                 return 0;
 987
 988         path = btrfs_alloc_path();
 989         if (!path)
 990                 return -ENOMEM;
 991
 992         extent_root = btrfs_extent_root(fs_info, 0);
 993         eb = btrfs_read_lock_root_node(extent_root);
 994         level = btrfs_header_level(eb);
 995         path->nodes[level] = eb;
 996         path->slots[level] = 0;
 997         path->locks[level] = BTRFS_READ_LOCK;
 998
 999         while (1) {
1000                 /*
1001                  * We have to keep track of the bytenr/num_bytes we last hit
1002                  * because we could have run out of space for an inline ref, and
1003                  * would have had to added a ref key item which may appear on a
1004                  * different leaf from the original extent item.
1005                  */
1006                 ret = walk_down_tree(extent_root, path, level,
1007                                      &bytenr, &num_bytes, &tree_block_level);
1008                 if (ret)
1009                         break;
1010                 ret = walk_up_tree(path, &level);
1011                 if (ret < 0)
1012                         break;
1013                 if (ret > 0) {
1014                         ret = 0;
1015                         break;
1016                 }
1017         }
1018         if (ret) {
1019                 btrfs_clear_opt(fs_info->mount_opt, REF_VERIFY);
1020                 btrfs_free_ref_cache(fs_info);
1021         }
1022         btrfs_free_path(path);
1023         return ret;
1024 }