fs/btrfs/ctree.c

   1 /*
   2  * Copyright (C) 2007,2008 Oracle.  All rights reserved.
   3  *
   4  * This program is free software; you can redistribute it and/or
   5  * modify it under the terms of the GNU General Public
   6  * License v2 as published by the Free Software Foundation.
   7  *
   8  * This program is distributed in the hope that it will be useful,
   9  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  10  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  11  * General Public License for more details.
  12  *
  13  * You should have received a copy of the GNU General Public
  14  * License along with this program; if not, write to the
  15  * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  16  * Boston, MA 021110-1307, USA.
  17  */
  18
  19 #include <linux/sched.h>
  20 #include <linux/slab.h>
  21 #include <linux/rbtree.h>
  22 #include <linux/vmalloc.h>
  23 #include "ctree.h"
  24 #include "disk-io.h"
  25 #include "transaction.h"
  26 #include "print-tree.h"
  27 #include "locking.h"
  28
  29 static int split_node(struct btrfs_trans_handle *trans, struct btrfs_root
  30                       *root, struct btrfs_path *path, int level);
  31 static int split_leaf(struct btrfs_trans_handle *trans, struct btrfs_root
  32                       *root, struct btrfs_key *ins_key,
  33                       struct btrfs_path *path, int data_size, int extend);
  34 static int push_node_left(struct btrfs_trans_handle *trans,
  35                           struct btrfs_root *root, struct extent_buffer *dst,
  36                           struct extent_buffer *src, int empty);
  37 static int balance_node_right(struct btrfs_trans_handle *trans,
  38                               struct btrfs_root *root,
  39                               struct extent_buffer *dst_buf,
  40                               struct extent_buffer *src_buf);
  41 static void del_ptr(struct btrfs_root *root, struct btrfs_path *path,
  42                     int level, int slot);
  43 static int tree_mod_log_free_eb(struct btrfs_fs_info *fs_info,
  44                                  struct extent_buffer *eb);
  45
  46 struct btrfs_path *btrfs_alloc_path(void)
  47 {
  48         struct btrfs_path *path;
  49         path = kmem_cache_zalloc(btrfs_path_cachep, GFP_NOFS);
  50         return path;
  51 }
  52
  53 /*
  54  * set all locked nodes in the path to blocking locks.  This should
  55  * be done before scheduling
  56  */
  57 noinline void btrfs_set_path_blocking(struct btrfs_path *p)
  58 {
  59         int i;
  60         for (i = 0; i < BTRFS_MAX_LEVEL; i++) {
  61                 if (!p->nodes[i] || !p->locks[i])
  62                         continue;
  63                 btrfs_set_lock_blocking_rw(p->nodes[i], p->locks[i]);
  64                 if (p->locks[i] == BTRFS_READ_LOCK)
  65                         p->locks[i] = BTRFS_READ_LOCK_BLOCKING;
  66                 else if (p->locks[i] == BTRFS_WRITE_LOCK)
  67                         p->locks[i] = BTRFS_WRITE_LOCK_BLOCKING;
  68         }
  69 }
  70
  71 /*
  72  * reset all the locked nodes in the patch to spinning locks.
  73  *
  74  * held is used to keep lockdep happy, when lockdep is enabled
  75  * we set held to a blocking lock before we go around and
  76  * retake all the spinlocks in the path.  You can safely use NULL
  77  * for held
  78  */
  79 noinline void btrfs_clear_path_blocking(struct btrfs_path *p,
  80                                         struct extent_buffer *held, int held_rw)
  81 {
  82         int i;
  83
  84         if (held) {
  85                 btrfs_set_lock_blocking_rw(held, held_rw);
  86                 if (held_rw == BTRFS_WRITE_LOCK)
  87                         held_rw = BTRFS_WRITE_LOCK_BLOCKING;
  88                 else if (held_rw == BTRFS_READ_LOCK)
  89                         held_rw = BTRFS_READ_LOCK_BLOCKING;
  90         }
  91         btrfs_set_path_blocking(p);
  92
  93         for (i = BTRFS_MAX_LEVEL - 1; i >= 0; i--) {
  94                 if (p->nodes[i] && p->locks[i]) {
  95                         btrfs_clear_lock_blocking_rw(p->nodes[i], p->locks[i]);
  96                         if (p->locks[i] == BTRFS_WRITE_LOCK_BLOCKING)
  97                                 p->locks[i] = BTRFS_WRITE_LOCK;
  98                         else if (p->locks[i] == BTRFS_READ_LOCK_BLOCKING)
  99                                 p->locks[i] = BTRFS_READ_LOCK;
 100                 }
 101         }
 102
 103         if (held)
 104                 btrfs_clear_lock_blocking_rw(held, held_rw);
 105 }
 106
 107 /* this also releases the path */
 108 void btrfs_free_path(struct btrfs_path *p)
 109 {
 110         if (!p)
 111                 return;
 112         btrfs_release_path(p);
 113         kmem_cache_free(btrfs_path_cachep, p);
 114 }
 115
 116 /*
 117  * path release drops references on the extent buffers in the path
 118  * and it drops any locks held by this path
 119  *
 120  * It is safe to call this on paths that no locks or extent buffers held.
 121  */
 122 noinline void btrfs_release_path(struct btrfs_path *p)
 123 {
 124         int i;
 125
 126         for (i = 0; i < BTRFS_MAX_LEVEL; i++) {
 127                 p->slots[i] = 0;
 128                 if (!p->nodes[i])
 129                         continue;
 130                 if (p->locks[i]) {
 131                         btrfs_tree_unlock_rw(p->nodes[i], p->locks[i]);
 132                         p->locks[i] = 0;
 133                 }
 134                 free_extent_buffer(p->nodes[i]);
 135                 p->nodes[i] = NULL;
 136         }
 137 }
 138
 139 /*
 140  * safely gets a reference on the root node of a tree.  A lock
 141  * is not taken, so a concurrent writer may put a different node
 142  * at the root of the tree.  See btrfs_lock_root_node for the
 143  * looping required.
 144  *
 145  * The extent buffer returned by this has a reference taken, so
 146  * it won't disappear.  It may stop being the root of the tree
 147  * at any time because there are no locks held.
 148  */
 149 struct extent_buffer *btrfs_root_node(struct btrfs_root *root)
 150 {
 151         struct extent_buffer *eb;
 152
 153         while (1) {
 154                 rcu_read_lock();
 155                 eb = rcu_dereference(root->node);
 156
 157                 /*
 158                  * RCU really hurts here, we could free up the root node because
 159                  * it was cow'ed but we may not get the new root node yet so do
 160                  * the inc_not_zero dance and if it doesn't work then
 161                  * synchronize_rcu and try again.
 162                  */
 163                 if (atomic_inc_not_zero(&eb->refs)) {
 164                         rcu_read_unlock();
 165                         break;
 166                 }
 167                 rcu_read_unlock();
 168                 synchronize_rcu();
 169         }
 170         return eb;
 171 }
 172
 173 /* loop around taking references on and locking the root node of the
 174  * tree until you end up with a lock on the root.  A locked buffer
 175  * is returned, with a reference held.
 176  */
 177 struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root)
 178 {
 179         struct extent_buffer *eb;
 180
 181         while (1) {
 182                 eb = btrfs_root_node(root);
 183                 btrfs_tree_lock(eb);
 184                 if (eb == root->node)
 185                         break;
 186                 btrfs_tree_unlock(eb);
 187                 free_extent_buffer(eb);
 188         }
 189         return eb;
 190 }
 191
 192 /* loop around taking references on and locking the root node of the
 193  * tree until you end up with a lock on the root.  A locked buffer
 194  * is returned, with a reference held.
 195  */
 196 static struct extent_buffer *btrfs_read_lock_root_node(struct btrfs_root *root)
 197 {
 198         struct extent_buffer *eb;
 199
 200         while (1) {
 201                 eb = btrfs_root_node(root);
 202                 btrfs_tree_read_lock(eb);
 203                 if (eb == root->node)
 204                         break;
 205                 btrfs_tree_read_unlock(eb);
 206                 free_extent_buffer(eb);
 207         }
 208         return eb;
 209 }
 210
 211 /* cowonly root (everything not a reference counted cow subvolume), just get
 212  * put onto a simple dirty list.  transaction.c walks this to make sure they
 213  * get properly updated on disk.
 214  */
 215 static void add_root_to_dirty_list(struct btrfs_root *root)
 216 {
 217         if (test_bit(BTRFS_ROOT_DIRTY, &root->state) ||
 218             !test_bit(BTRFS_ROOT_TRACK_DIRTY, &root->state))
 219                 return;
 220
 221         spin_lock(&root->fs_info->trans_lock);
 222         if (!test_and_set_bit(BTRFS_ROOT_DIRTY, &root->state)) {
 223                 /* Want the extent tree to be the last on the list */
 224                 if (root->objectid == BTRFS_EXTENT_TREE_OBJECTID)
 225                         list_move_tail(&root->dirty_list,
 226                                        &root->fs_info->dirty_cowonly_roots);
 227                 else
 228                         list_move(&root->dirty_list,
 229                                   &root->fs_info->dirty_cowonly_roots);
 230         }
 231         spin_unlock(&root->fs_info->trans_lock);
 232 }
 233
 234 /*
 235  * used by snapshot creation to make a copy of a root for a tree with
 236  * a given objectid.  The buffer with the new root node is returned in
 237  * cow_ret, and this func returns zero on success or a negative error code.
 238  */
 239 int btrfs_copy_root(struct btrfs_trans_handle *trans,
 240                       struct btrfs_root *root,
 241                       struct extent_buffer *buf,
 242                       struct extent_buffer **cow_ret, u64 new_root_objectid)
 243 {
 244         struct extent_buffer *cow;
 245         int ret = 0;
 246         int level;
 247         struct btrfs_disk_key disk_key;
 248
 249         WARN_ON(test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
 250                 trans->transid != root->fs_info->running_transaction->transid);
 251         WARN_ON(test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
 252                 trans->transid != root->last_trans);
 253
 254         level = btrfs_header_level(buf);
 255         if (level == 0)
 256                 btrfs_item_key(buf, &disk_key, 0);
 257         else
 258                 btrfs_node_key(buf, &disk_key, 0);
 259
 260         cow = btrfs_alloc_tree_block(trans, root, 0, new_root_objectid,
 261                         &disk_key, level, buf->start, 0);
 262         if (IS_ERR(cow))
 263                 return PTR_ERR(cow);
 264
 265         copy_extent_buffer(cow, buf, 0, 0, cow->len);
 266         btrfs_set_header_bytenr(cow, cow->start);
 267         btrfs_set_header_generation(cow, trans->transid);
 268         btrfs_set_header_backref_rev(cow, BTRFS_MIXED_BACKREF_REV);
 269         btrfs_clear_header_flag(cow, BTRFS_HEADER_FLAG_WRITTEN |
 270                                      BTRFS_HEADER_FLAG_RELOC);
 271         if (new_root_objectid == BTRFS_TREE_RELOC_OBJECTID)
 272                 btrfs_set_header_flag(cow, BTRFS_HEADER_FLAG_RELOC);
 273         else
 274                 btrfs_set_header_owner(cow, new_root_objectid);
 275
 276         write_extent_buffer(cow, root->fs_info->fsid, btrfs_header_fsid(),
 277                             BTRFS_FSID_SIZE);
 278
 279         WARN_ON(btrfs_header_generation(buf) > trans->transid);
 280         if (new_root_objectid == BTRFS_TREE_RELOC_OBJECTID)
 281                 ret = btrfs_inc_ref(trans, root, cow, 1);
 282         else
 283                 ret = btrfs_inc_ref(trans, root, cow, 0);
 284
 285         if (ret)
 286                 return ret;
 287
 288         btrfs_mark_buffer_dirty(cow);
 289         *cow_ret = cow;
 290         return 0;
 291 }
 292
 293 enum mod_log_op {
 294         MOD_LOG_KEY_REPLACE,
 295         MOD_LOG_KEY_ADD,
 296         MOD_LOG_KEY_REMOVE,
 297         MOD_LOG_KEY_REMOVE_WHILE_FREEING,
 298         MOD_LOG_KEY_REMOVE_WHILE_MOVING,
 299         MOD_LOG_MOVE_KEYS,
 300         MOD_LOG_ROOT_REPLACE,
 301 };
 302
 303 struct tree_mod_move {
 304         int dst_slot;
 305         int nr_items;
 306 };
 307
 308 struct tree_mod_root {
 309         u64 logical;
 310         u8 level;
 311 };
 312
 313 struct tree_mod_elem {
 314         struct rb_node node;
 315         u64 logical;
 316         u64 seq;
 317         enum mod_log_op op;
 318
 319         /* this is used for MOD_LOG_KEY_* and MOD_LOG_MOVE_KEYS operations */
 320         int slot;
 321
 322         /* this is used for MOD_LOG_KEY* and MOD_LOG_ROOT_REPLACE */
 323         u64 generation;
 324
 325         /* those are used for op == MOD_LOG_KEY_{REPLACE,REMOVE} */
 326         struct btrfs_disk_key key;
 327         u64 blockptr;
 328
 329         /* this is used for op == MOD_LOG_MOVE_KEYS */
 330         struct tree_mod_move move;
 331
 332         /* this is used for op == MOD_LOG_ROOT_REPLACE */
 333         struct tree_mod_root old_root;
 334 };
 335
 336 static inline void tree_mod_log_read_lock(struct btrfs_fs_info *fs_info)
 337 {
 338         read_lock(&fs_info->tree_mod_log_lock);
 339 }
 340
 341 static inline void tree_mod_log_read_unlock(struct btrfs_fs_info *fs_info)
 342 {
 343         read_unlock(&fs_info->tree_mod_log_lock);
 344 }
 345
 346 static inline void tree_mod_log_write_lock(struct btrfs_fs_info *fs_info)
 347 {
 348         write_lock(&fs_info->tree_mod_log_lock);
 349 }
 350
 351 static inline void tree_mod_log_write_unlock(struct btrfs_fs_info *fs_info)
 352 {
 353         write_unlock(&fs_info->tree_mod_log_lock);
 354 }
 355
 356 /*
 357  * Pull a new tree mod seq number for our operation.
 358  */
 359 static inline u64 btrfs_inc_tree_mod_seq(struct btrfs_fs_info *fs_info)
 360 {
 361         return atomic64_inc_return(&fs_info->tree_mod_seq);
 362 }
 363
 364 /*
 365  * This adds a new blocker to the tree mod log's blocker list if the @elem
 366  * passed does not already have a sequence number set. So when a caller expects
 367  * to record tree modifications, it should ensure to set elem->seq to zero
 368  * before calling btrfs_get_tree_mod_seq.
 369  * Returns a fresh, unused tree log modification sequence number, even if no new
 370  * blocker was added.
 371  */
 372 u64 btrfs_get_tree_mod_seq(struct btrfs_fs_info *fs_info,
 373                            struct seq_list *elem)
 374 {
 375         tree_mod_log_write_lock(fs_info);
 376         spin_lock(&fs_info->tree_mod_seq_lock);
 377         if (!elem->seq) {
 378                 elem->seq = btrfs_inc_tree_mod_seq(fs_info);
 379                 list_add_tail(&elem->list, &fs_info->tree_mod_seq_list);
 380         }
 381         spin_unlock(&fs_info->tree_mod_seq_lock);
 382         tree_mod_log_write_unlock(fs_info);
 383
 384         return elem->seq;
 385 }
 386
 387 void btrfs_put_tree_mod_seq(struct btrfs_fs_info *fs_info,
 388                             struct seq_list *elem)
 389 {
 390         struct rb_root *tm_root;
 391         struct rb_node *node;
 392         struct rb_node *next;
 393         struct seq_list *cur_elem;
 394         struct tree_mod_elem *tm;
 395         u64 min_seq = (u64)-1;
 396         u64 seq_putting = elem->seq;
 397
 398         if (!seq_putting)
 399                 return;
 400
 401         spin_lock(&fs_info->tree_mod_seq_lock);
 402         list_del(&elem->list);
 403         elem->seq = 0;
 404
 405         list_for_each_entry(cur_elem, &fs_info->tree_mod_seq_list, list) {
 406                 if (cur_elem->seq < min_seq) {
 407                         if (seq_putting > cur_elem->seq) {
 408                                 /*
 409                                  * blocker with lower sequence number exists, we
 410                                  * cannot remove anything from the log
 411                                  */
 412                                 spin_unlock(&fs_info->tree_mod_seq_lock);
 413                                 return;
 414                         }
 415                         min_seq = cur_elem->seq;
 416                 }
 417         }
 418         spin_unlock(&fs_info->tree_mod_seq_lock);
 419
 420         /*
 421          * anything that's lower than the lowest existing (read: blocked)
 422          * sequence number can be removed from the tree.
 423          */
 424         tree_mod_log_write_lock(fs_info);
 425         tm_root = &fs_info->tree_mod_log;
 426         for (node = rb_first(tm_root); node; node = next) {
 427                 next = rb_next(node);
 428                 tm = container_of(node, struct tree_mod_elem, node);
 429                 if (tm->seq > min_seq)
 430                         continue;
 431                 rb_erase(node, tm_root);
 432                 kfree(tm);
 433         }
 434         tree_mod_log_write_unlock(fs_info);
 435 }
 436
 437 /*
 438  * key order of the log:
 439  *       node/leaf start address -> sequence
 440  *
 441  * The 'start address' is the logical address of the *new* root node
 442  * for root replace operations, or the logical address of the affected
 443  * block for all other operations.
 444  *
 445  * Note: must be called with write lock (tree_mod_log_write_lock).
 446  */
 447 static noinline int
 448 __tree_mod_log_insert(struct btrfs_fs_info *fs_info, struct tree_mod_elem *tm)
 449 {
 450         struct rb_root *tm_root;
 451         struct rb_node **new;
 452         struct rb_node *parent = NULL;
 453         struct tree_mod_elem *cur;
 454
 455         BUG_ON(!tm);
 456
 457         tm->seq = btrfs_inc_tree_mod_seq(fs_info);
 458
 459         tm_root = &fs_info->tree_mod_log;
 460         new = &tm_root->rb_node;
 461         while (*new) {
 462                 cur = container_of(*new, struct tree_mod_elem, node);
 463                 parent = *new;
 464                 if (cur->logical < tm->logical)
 465                         new = &((*new)->rb_left);
 466                 else if (cur->logical > tm->logical)
 467                         new = &((*new)->rb_right);
 468                 else if (cur->seq < tm->seq)
 469                         new = &((*new)->rb_left);
 470                 else if (cur->seq > tm->seq)
 471                         new = &((*new)->rb_right);
 472                 else
 473                         return -EEXIST;
 474         }
 475
 476         rb_link_node(&tm->node, parent, new);
 477         rb_insert_color(&tm->node, tm_root);
 478         return 0;
 479 }
 480
 481 /*
 482  * Determines if logging can be omitted. Returns 1 if it can. Otherwise, it
 483  * returns zero with the tree_mod_log_lock acquired. The caller must hold
 484  * this until all tree mod log insertions are recorded in the rb tree and then
 485  * call tree_mod_log_write_unlock() to release.
 486  */
 487 static inline int tree_mod_dont_log(struct btrfs_fs_info *fs_info,
 488                                     struct extent_buffer *eb) {
 489         smp_mb();
 490         if (list_empty(&(fs_info)->tree_mod_seq_list))
 491                 return 1;
 492         if (eb && btrfs_header_level(eb) == 0)
 493                 return 1;
 494
 495         tree_mod_log_write_lock(fs_info);
 496         if (list_empty(&(fs_info)->tree_mod_seq_list)) {
 497                 tree_mod_log_write_unlock(fs_info);
 498                 return 1;
 499         }
 500
 501         return 0;
 502 }
 503
 504 /* Similar to tree_mod_dont_log, but doesn't acquire any locks. */
 505 static inline int tree_mod_need_log(const struct btrfs_fs_info *fs_info,
 506                                     struct extent_buffer *eb)
 507 {
 508         smp_mb();
 509         if (list_empty(&(fs_info)->tree_mod_seq_list))
 510                 return 0;
 511         if (eb && btrfs_header_level(eb) == 0)
 512                 return 0;
 513
 514         return 1;
 515 }
 516
 517 static struct tree_mod_elem *
 518 alloc_tree_mod_elem(struct extent_buffer *eb, int slot,
 519                     enum mod_log_op op, gfp_t flags)
 520 {
 521         struct tree_mod_elem *tm;
 522
 523         tm = kzalloc(sizeof(*tm), flags);
 524         if (!tm)
 525                 return NULL;
 526
 527         tm->logical = eb->start;
 528         if (op != MOD_LOG_KEY_ADD) {
 529                 btrfs_node_key(eb, &tm->key, slot);
 530                 tm->blockptr = btrfs_node_blockptr(eb, slot);
 531         }
 532         tm->op = op;
 533         tm->slot = slot;
 534         tm->generation = btrfs_node_ptr_generation(eb, slot);
 535         RB_CLEAR_NODE(&tm->node);
 536
 537         return tm;
 538 }
 539
 540 static noinline int
 541 tree_mod_log_insert_key(struct btrfs_fs_info *fs_info,
 542                         struct extent_buffer *eb, int slot,
 543                         enum mod_log_op op, gfp_t flags)
 544 {
 545         struct tree_mod_elem *tm;
 546         int ret;
 547
 548         if (!tree_mod_need_log(fs_info, eb))
 549                 return 0;
 550
 551         tm = alloc_tree_mod_elem(eb, slot, op, flags);
 552         if (!tm)
 553                 return -ENOMEM;
 554
 555         if (tree_mod_dont_log(fs_info, eb)) {
 556                 kfree(tm);
 557                 return 0;
 558         }
 559
 560         ret = __tree_mod_log_insert(fs_info, tm);
 561         tree_mod_log_write_unlock(fs_info);
 562         if (ret)
 563                 kfree(tm);
 564
 565         return ret;
 566 }
 567
 568 static noinline int
 569 tree_mod_log_insert_move(struct btrfs_fs_info *fs_info,
 570                          struct extent_buffer *eb, int dst_slot, int src_slot,
 571                          int nr_items, gfp_t flags)
 572 {
 573         struct tree_mod_elem *tm = NULL;
 574         struct tree_mod_elem **tm_list = NULL;
 575         int ret = 0;
 576         int i;
 577         int locked = 0;
 578
 579         if (!tree_mod_need_log(fs_info, eb))
 580                 return 0;
 581
 582         tm_list = kcalloc(nr_items, sizeof(struct tree_mod_elem *), flags);
 583         if (!tm_list)
 584                 return -ENOMEM;
 585
 586         tm = kzalloc(sizeof(*tm), flags);
 587         if (!tm) {
 588                 ret = -ENOMEM;
 589                 goto free_tms;
 590         }
 591
 592         tm->logical = eb->start;
 593         tm->slot = src_slot;
 594         tm->move.dst_slot = dst_slot;
 595         tm->move.nr_items = nr_items;
 596         tm->op = MOD_LOG_MOVE_KEYS;
 597
 598         for (i = 0; i + dst_slot < src_slot && i < nr_items; i++) {
 599                 tm_list[i] = alloc_tree_mod_elem(eb, i + dst_slot,
 600                     MOD_LOG_KEY_REMOVE_WHILE_MOVING, flags);
 601                 if (!tm_list[i]) {
 602                         ret = -ENOMEM;
 603                         goto free_tms;
 604                 }
 605         }
 606
 607         if (tree_mod_dont_log(fs_info, eb))
 608                 goto free_tms;
 609         locked = 1;
 610
 611         /*
 612          * When we override something during the move, we log these removals.
 613          * This can only happen when we move towards the beginning of the
 614          * buffer, i.e. dst_slot < src_slot.
 615          */
 616         for (i = 0; i + dst_slot < src_slot && i < nr_items; i++) {
 617                 ret = __tree_mod_log_insert(fs_info, tm_list[i]);
 618                 if (ret)
 619                         goto free_tms;
 620         }
 621
 622         ret = __tree_mod_log_insert(fs_info, tm);
 623         if (ret)
 624                 goto free_tms;
 625         tree_mod_log_write_unlock(fs_info);
 626         kfree(tm_list);
 627
 628         return 0;
 629 free_tms:
 630         for (i = 0; i < nr_items; i++) {
 631                 if (tm_list[i] && !RB_EMPTY_NODE(&tm_list[i]->node))
 632                         rb_erase(&tm_list[i]->node, &fs_info->tree_mod_log);
 633                 kfree(tm_list[i]);
 634         }
 635         if (locked)
 636                 tree_mod_log_write_unlock(fs_info);
 637         kfree(tm_list);
 638         kfree(tm);
 639
 640         return ret;
 641 }
 642
 643 static inline int
 644 __tree_mod_log_free_eb(struct btrfs_fs_info *fs_info,
 645                        struct tree_mod_elem **tm_list,
 646                        int nritems)
 647 {
 648         int i, j;
 649         int ret;
 650
 651         for (i = nritems - 1; i >= 0; i--) {
 652                 ret = __tree_mod_log_insert(fs_info, tm_list[i]);
 653                 if (ret) {
 654                         for (j = nritems - 1; j > i; j--)
 655                                 rb_erase(&tm_list[j]->node,
 656                                          &fs_info->tree_mod_log);
 657                         return ret;
 658                 }
 659         }
 660
 661         return 0;
 662 }
 663
 664 static noinline int
 665 tree_mod_log_insert_root(struct btrfs_fs_info *fs_info,
 666                          struct extent_buffer *old_root,
 667                          struct extent_buffer *new_root, gfp_t flags,
 668                          int log_removal)
 669 {
 670         struct tree_mod_elem *tm = NULL;
 671         struct tree_mod_elem **tm_list = NULL;
 672         int nritems = 0;
 673         int ret = 0;
 674         int i;
 675
 676         if (!tree_mod_need_log(fs_info, NULL))
 677                 return 0;
 678
 679         if (log_removal && btrfs_header_level(old_root) > 0) {
 680                 nritems = btrfs_header_nritems(old_root);
 681                 tm_list = kcalloc(nritems, sizeof(struct tree_mod_elem *),
 682                                   flags);
 683                 if (!tm_list) {
 684                         ret = -ENOMEM;
 685                         goto free_tms;
 686                 }
 687                 for (i = 0; i < nritems; i++) {
 688                         tm_list[i] = alloc_tree_mod_elem(old_root, i,
 689                             MOD_LOG_KEY_REMOVE_WHILE_FREEING, flags);
 690                         if (!tm_list[i]) {
 691                                 ret = -ENOMEM;
 692                                 goto free_tms;
 693                         }
 694                 }
 695         }
 696
 697         tm = kzalloc(sizeof(*tm), flags);
 698         if (!tm) {
 699                 ret = -ENOMEM;
 700                 goto free_tms;
 701         }
 702
 703         tm->logical = new_root->start;
 704         tm->old_root.logical = old_root->start;
 705         tm->old_root.level = btrfs_header_level(old_root);
 706         tm->generation = btrfs_header_generation(old_root);
 707         tm->op = MOD_LOG_ROOT_REPLACE;
 708
 709         if (tree_mod_dont_log(fs_info, NULL))
 710                 goto free_tms;
 711
 712         if (tm_list)
 713                 ret = __tree_mod_log_free_eb(fs_info, tm_list, nritems);
 714         if (!ret)
 715                 ret = __tree_mod_log_insert(fs_info, tm);
 716
 717         tree_mod_log_write_unlock(fs_info);
 718         if (ret)
 719                 goto free_tms;
 720         kfree(tm_list);
 721
 722         return ret;
 723
 724 free_tms:
 725         if (tm_list) {
 726                 for (i = 0; i < nritems; i++)
 727                         kfree(tm_list[i]);
 728                 kfree(tm_list);
 729         }
 730         kfree(tm);
 731
 732         return ret;
 733 }
 734
 735 static struct tree_mod_elem *
 736 __tree_mod_log_search(struct btrfs_fs_info *fs_info, u64 start, u64 min_seq,
 737                       int smallest)
 738 {
 739         struct rb_root *tm_root;
 740         struct rb_node *node;
 741         struct tree_mod_elem *cur = NULL;
 742         struct tree_mod_elem *found = NULL;
 743
 744         tree_mod_log_read_lock(fs_info);
 745         tm_root = &fs_info->tree_mod_log;
 746         node = tm_root->rb_node;
 747         while (node) {
 748                 cur = container_of(node, struct tree_mod_elem, node);
 749                 if (cur->logical < start) {
 750                         node = node->rb_left;
 751                 } else if (cur->logical > start) {
 752                         node = node->rb_right;
 753                 } else if (cur->seq < min_seq) {
 754                         node = node->rb_left;
 755                 } else if (!smallest) {
 756                         /* we want the node with the highest seq */
 757                         if (found)
 758                                 BUG_ON(found->seq > cur->seq);
 759                         found = cur;
 760                         node = node->rb_left;
 761                 } else if (cur->seq > min_seq) {
 762                         /* we want the node with the smallest seq */
 763                         if (found)
 764                                 BUG_ON(found->seq < cur->seq);
 765                         found = cur;
 766                         node = node->rb_right;
 767                 } else {
 768                         found = cur;
 769                         break;
 770                 }
 771         }
 772         tree_mod_log_read_unlock(fs_info);
 773
 774         return found;
 775 }
 776
 777 /*
 778  * this returns the element from the log with the smallest time sequence
 779  * value that's in the log (the oldest log item). any element with a time
 780  * sequence lower than min_seq will be ignored.
 781  */
 782 static struct tree_mod_elem *
 783 tree_mod_log_search_oldest(struct btrfs_fs_info *fs_info, u64 start,
 784                            u64 min_seq)
 785 {
 786         return __tree_mod_log_search(fs_info, start, min_seq, 1);
 787 }
 788
 789 /*
 790  * this returns the element from the log with the largest time sequence
 791  * value that's in the log (the most recent log item). any element with
 792  * a time sequence lower than min_seq will be ignored.
 793  */
 794 static struct tree_mod_elem *
 795 tree_mod_log_search(struct btrfs_fs_info *fs_info, u64 start, u64 min_seq)
 796 {
 797         return __tree_mod_log_search(fs_info, start, min_seq, 0);
 798 }
 799
 800 static noinline int
 801 tree_mod_log_eb_copy(struct btrfs_fs_info *fs_info, struct extent_buffer *dst,
 802                      struct extent_buffer *src, unsigned long dst_offset,
 803                      unsigned long src_offset, int nr_items)
 804 {
 805         int ret = 0;
 806         struct tree_mod_elem **tm_list = NULL;
 807         struct tree_mod_elem **tm_list_add, **tm_list_rem;
 808         int i;
 809         int locked = 0;
 810
 811         if (!tree_mod_need_log(fs_info, NULL))
 812                 return 0;
 813
 814         if (btrfs_header_level(dst) == 0 && btrfs_header_level(src) == 0)
 815                 return 0;
 816
 817         tm_list = kcalloc(nr_items * 2, sizeof(struct tree_mod_elem *),
 818                           GFP_NOFS);
 819         if (!tm_list)
 820                 return -ENOMEM;
 821
 822         tm_list_add = tm_list;
 823         tm_list_rem = tm_list + nr_items;
 824         for (i = 0; i < nr_items; i++) {
 825                 tm_list_rem[i] = alloc_tree_mod_elem(src, i + src_offset,
 826                     MOD_LOG_KEY_REMOVE, GFP_NOFS);
 827                 if (!tm_list_rem[i]) {
 828                         ret = -ENOMEM;
 829                         goto free_tms;
 830                 }
 831
 832                 tm_list_add[i] = alloc_tree_mod_elem(dst, i + dst_offset,
 833                     MOD_LOG_KEY_ADD, GFP_NOFS);
 834                 if (!tm_list_add[i]) {
 835                         ret = -ENOMEM;
 836                         goto free_tms;
 837                 }
 838         }
 839
 840         if (tree_mod_dont_log(fs_info, NULL))
 841                 goto free_tms;
 842         locked = 1;
 843
 844         for (i = 0; i < nr_items; i++) {
 845                 ret = __tree_mod_log_insert(fs_info, tm_list_rem[i]);
 846                 if (ret)
 847                         goto free_tms;
 848                 ret = __tree_mod_log_insert(fs_info, tm_list_add[i]);
 849                 if (ret)
 850                         goto free_tms;
 851         }
 852
 853         tree_mod_log_write_unlock(fs_info);
 854         kfree(tm_list);
 855
 856         return 0;
 857
 858 free_tms:
 859         for (i = 0; i < nr_items * 2; i++) {
 860                 if (tm_list[i] && !RB_EMPTY_NODE(&tm_list[i]->node))
 861                         rb_erase(&tm_list[i]->node, &fs_info->tree_mod_log);
 862                 kfree(tm_list[i]);
 863         }
 864         if (locked)
 865                 tree_mod_log_write_unlock(fs_info);
 866         kfree(tm_list);
 867
 868         return ret;
 869 }
 870
 871 static inline void
 872 tree_mod_log_eb_move(struct btrfs_fs_info *fs_info, struct extent_buffer *dst,
 873                      int dst_offset, int src_offset, int nr_items)
 874 {
 875         int ret;
 876         ret = tree_mod_log_insert_move(fs_info, dst, dst_offset, src_offset,
 877                                        nr_items, GFP_NOFS);
 878         BUG_ON(ret < 0);
 879 }
 880
 881 static noinline void
 882 tree_mod_log_set_node_key(struct btrfs_fs_info *fs_info,
 883                           struct extent_buffer *eb, int slot, int atomic)
 884 {
 885         int ret;
 886
 887         ret = tree_mod_log_insert_key(fs_info, eb, slot,
 888                                         MOD_LOG_KEY_REPLACE,
 889                                         atomic ? GFP_ATOMIC : GFP_NOFS);
 890         BUG_ON(ret < 0);
 891 }
 892
 893 static noinline int
 894 tree_mod_log_free_eb(struct btrfs_fs_info *fs_info, struct extent_buffer *eb)
 895 {
 896         struct tree_mod_elem **tm_list = NULL;
 897         int nritems = 0;
 898         int i;
 899         int ret = 0;
 900
 901         if (btrfs_header_level(eb) == 0)
 902                 return 0;
 903
 904         if (!tree_mod_need_log(fs_info, NULL))
 905                 return 0;
 906
 907         nritems = btrfs_header_nritems(eb);
 908         tm_list = kcalloc(nritems, sizeof(struct tree_mod_elem *), GFP_NOFS);
 909         if (!tm_list)
 910                 return -ENOMEM;
 911
 912         for (i = 0; i < nritems; i++) {
 913                 tm_list[i] = alloc_tree_mod_elem(eb, i,
 914                     MOD_LOG_KEY_REMOVE_WHILE_FREEING, GFP_NOFS);
 915                 if (!tm_list[i]) {
 916                         ret = -ENOMEM;
 917                         goto free_tms;
 918                 }
 919         }
 920
 921         if (tree_mod_dont_log(fs_info, eb))
 922                 goto free_tms;
 923
 924         ret = __tree_mod_log_free_eb(fs_info, tm_list, nritems);
 925         tree_mod_log_write_unlock(fs_info);
 926         if (ret)
 927                 goto free_tms;
 928         kfree(tm_list);
 929
 930         return 0;
 931
 932 free_tms:
 933         for (i = 0; i < nritems; i++)
 934                 kfree(tm_list[i]);
 935         kfree(tm_list);
 936
 937         return ret;
 938 }
 939
 940 static noinline void
 941 tree_mod_log_set_root_pointer(struct btrfs_root *root,
 942                               struct extent_buffer *new_root_node,
 943                               int log_removal)
 944 {
 945         int ret;
 946         ret = tree_mod_log_insert_root(root->fs_info, root->node,
 947                                        new_root_node, GFP_NOFS, log_removal);
 948         BUG_ON(ret < 0);
 949 }
 950
 951 /*
 952  * check if the tree block can be shared by multiple trees
 953  */
 954 int btrfs_block_can_be_shared(struct btrfs_root *root,
 955                               struct extent_buffer *buf)
 956 {
 957         /*
 958          * Tree blocks not in refernece counted trees and tree roots
 959          * are never shared. If a block was allocated after the last
 960          * snapshot and the block was not allocated by tree relocation,
 961          * we know the block is not shared.
 962          */
 963         if (test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
 964             buf != root->node && buf != root->commit_root &&
 965             (btrfs_header_generation(buf) <=
 966              btrfs_root_last_snapshot(&root->root_item) ||
 967              btrfs_header_flag(buf, BTRFS_HEADER_FLAG_RELOC)))
 968                 return 1;
 969 #ifdef BTRFS_COMPAT_EXTENT_TREE_V0
 970         if (test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
 971             btrfs_header_backref_rev(buf) < BTRFS_MIXED_BACKREF_REV)
 972                 return 1;
 973 #endif
 974         return 0;
 975 }
 976
 977 static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans,
 978                                        struct btrfs_root *root,
 979                                        struct extent_buffer *buf,
 980                                        struct extent_buffer *cow,
 981                                        int *last_ref)
 982 {
 983         u64 refs;
 984         u64 owner;
 985         u64 flags;
 986         u64 new_flags = 0;
 987         int ret;
 988
 989         /*
 990          * Backrefs update rules:
 991          *
 992          * Always use full backrefs for extent pointers in tree block
 993          * allocated by tree relocation.
 994          *
 995          * If a shared tree block is no longer referenced by its owner
 996          * tree (btrfs_header_owner(buf) == root->root_key.objectid),
 997          * use full backrefs for extent pointers in tree block.
 998          *
 999          * If a tree block is been relocating
1000          * (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID),
1001          * use full backrefs for extent pointers in tree block.
1002          * The reason for this is some operations (such as drop tree)
1003          * are only allowed for blocks use full backrefs.
1004          */
1005
1006         if (btrfs_block_can_be_shared(root, buf)) {
1007                 ret = btrfs_lookup_extent_info(trans, root, buf->start,
1008                                                btrfs_header_level(buf), 1,
1009                                                &refs, &flags);
1010                 if (ret)
1011                         return ret;
1012                 if (refs == 0) {
1013                         ret = -EROFS;
1014                         btrfs_std_error(root->fs_info, ret, NULL);
1015                         return ret;
1016                 }
1017         } else {
1018                 refs = 1;
1019                 if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID ||
1020                     btrfs_header_backref_rev(buf) < BTRFS_MIXED_BACKREF_REV)
1021                         flags = BTRFS_BLOCK_FLAG_FULL_BACKREF;
1022                 else
1023                         flags = 0;
1024         }
1025
1026         owner = btrfs_header_owner(buf);
1027         BUG_ON(owner == BTRFS_TREE_RELOC_OBJECTID &&
1028                !(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF));
1029
1030         if (refs > 1) {
1031                 if ((owner == root->root_key.objectid ||
1032                      root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) &&
1033                     !(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF)) {
1034                         ret = btrfs_inc_ref(trans, root, buf, 1);
1035                         BUG_ON(ret); /* -ENOMEM */
1036
1037                         if (root->root_key.objectid ==
1038                             BTRFS_TREE_RELOC_OBJECTID) {
1039                                 ret = btrfs_dec_ref(trans, root, buf, 0);
1040                                 BUG_ON(ret); /* -ENOMEM */
1041                                 ret = btrfs_inc_ref(trans, root, cow, 1);
1042                                 BUG_ON(ret); /* -ENOMEM */
1043                         }
1044                         new_flags |= BTRFS_BLOCK_FLAG_FULL_BACKREF;
1045                 } else {
1046
1047                         if (root->root_key.objectid ==
1048                             BTRFS_TREE_RELOC_OBJECTID)
1049                                 ret = btrfs_inc_ref(trans, root, cow, 1);
1050                         else
1051                                 ret = btrfs_inc_ref(trans, root, cow, 0);
1052                         BUG_ON(ret); /* -ENOMEM */
1053                 }
1054                 if (new_flags != 0) {
1055                         int level = btrfs_header_level(buf);
1056
1057                         ret = btrfs_set_disk_extent_flags(trans, root,
1058                                                           buf->start,
1059                                                           buf->len,
1060                                                           new_flags, level, 0);
1061                         if (ret)
1062                                 return ret;
1063                 }
1064         } else {
1065                 if (flags & BTRFS_BLOCK_FLAG_FULL_BACKREF) {
1066                         if (root->root_key.objectid ==
1067                             BTRFS_TREE_RELOC_OBJECTID)
1068                                 ret = btrfs_inc_ref(trans, root, cow, 1);
1069                         else
1070                                 ret = btrfs_inc_ref(trans, root, cow, 0);
1071                         BUG_ON(ret); /* -ENOMEM */
1072                         ret = btrfs_dec_ref(trans, root, buf, 1);
1073                         BUG_ON(ret); /* -ENOMEM */
1074                 }
1075                 clean_tree_block(trans, root->fs_info, buf);
1076                 *last_ref = 1;
1077         }
1078         return 0;
1079 }
1080
1081 /*
1082  * does the dirty work in cow of a single block.  The parent block (if
1083  * supplied) is updated to point to the new cow copy.  The new buffer is marked
1084  * dirty and returned locked.  If you modify the block it needs to be marked
1085  * dirty again.
1086  *
1087  * search_start -- an allocation hint for the new block
1088  *
1089  * empty_size -- a hint that you plan on doing more cow.  This is the size in
1090  * bytes the allocator should try to find free next to the block it returns.
1091  * This is just a hint and may be ignored by the allocator.
1092  */
1093 static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
1094                              struct btrfs_root *root,
1095                              struct extent_buffer *buf,
1096                              struct extent_buffer *parent, int parent_slot,
1097                              struct extent_buffer **cow_ret,
1098                              u64 search_start, u64 empty_size)
1099 {
1100         struct btrfs_disk_key disk_key;
1101         struct extent_buffer *cow;
1102         int level, ret;
1103         int last_ref = 0;
1104         int unlock_orig = 0;
1105         u64 parent_start;
1106
1107         if (*cow_ret == buf)
1108                 unlock_orig = 1;
1109
1110         btrfs_assert_tree_locked(buf);
1111
1112         WARN_ON(test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
1113                 trans->transid != root->fs_info->running_transaction->transid);
1114         WARN_ON(test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
1115                 trans->transid != root->last_trans);
1116
1117         level = btrfs_header_level(buf);
1118
1119         if (level == 0)
1120                 btrfs_item_key(buf, &disk_key, 0);
1121         else
1122                 btrfs_node_key(buf, &disk_key, 0);
1123
1124         if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) {
1125                 if (parent)
1126                         parent_start = parent->start;
1127                 else
1128                         parent_start = 0;
1129         } else
1130                 parent_start = 0;
1131
1132         cow = btrfs_alloc_tree_block(trans, root, parent_start,
1133                         root->root_key.objectid, &disk_key, level,
1134                         search_start, empty_size);
1135         if (IS_ERR(cow))
1136                 return PTR_ERR(cow);
1137
1138         /* cow is set to blocking by btrfs_init_new_buffer */
1139
1140         copy_extent_buffer(cow, buf, 0, 0, cow->len);
1141         btrfs_set_header_bytenr(cow, cow->start);
1142         btrfs_set_header_generation(cow, trans->transid);
1143         btrfs_set_header_backref_rev(cow, BTRFS_MIXED_BACKREF_REV);
1144         btrfs_clear_header_flag(cow, BTRFS_HEADER_FLAG_WRITTEN |
1145                                      BTRFS_HEADER_FLAG_RELOC);
1146         if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID)
1147                 btrfs_set_header_flag(cow, BTRFS_HEADER_FLAG_RELOC);
1148         else
1149                 btrfs_set_header_owner(cow, root->root_key.objectid);
1150
1151         write_extent_buffer(cow, root->fs_info->fsid, btrfs_header_fsid(),
1152                             BTRFS_FSID_SIZE);
1153
1154         ret = update_ref_for_cow(trans, root, buf, cow, &last_ref);
1155         if (ret) {
1156                 btrfs_abort_transaction(trans, root, ret);
1157                 return ret;
1158         }
1159
1160         if (test_bit(BTRFS_ROOT_REF_COWS, &root->state)) {
1161                 ret = btrfs_reloc_cow_block(trans, root, buf, cow);
1162                 if (ret) {
1163                         btrfs_abort_transaction(trans, root, ret);
1164                         return ret;
1165                 }
1166         }
1167
1168         if (buf == root->node) {
1169                 WARN_ON(parent && parent != buf);
1170                 if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID ||
1171                     btrfs_header_backref_rev(buf) < BTRFS_MIXED_BACKREF_REV)
1172                         parent_start = buf->start;
1173                 else
1174                         parent_start = 0;
1175
1176                 extent_buffer_get(cow);
1177                 tree_mod_log_set_root_pointer(root, cow, 1);
1178                 rcu_assign_pointer(root->node, cow);
1179
1180                 btrfs_free_tree_block(trans, root, buf, parent_start,
1181                                       last_ref);
1182                 free_extent_buffer(buf);
1183                 add_root_to_dirty_list(root);
1184         } else {
1185                 if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID)
1186                         parent_start = parent->start;
1187                 else
1188                         parent_start = 0;
1189
1190                 WARN_ON(trans->transid != btrfs_header_generation(parent));
1191                 tree_mod_log_insert_key(root->fs_info, parent, parent_slot,
1192                                         MOD_LOG_KEY_REPLACE, GFP_NOFS);
1193                 btrfs_set_node_blockptr(parent, parent_slot,
1194                                         cow->start);
1195                 btrfs_set_node_ptr_generation(parent, parent_slot,
1196                                               trans->transid);
1197                 btrfs_mark_buffer_dirty(parent);
1198                 if (last_ref) {
1199                         ret = tree_mod_log_free_eb(root->fs_info, buf);
1200                         if (ret) {
1201                                 btrfs_abort_transaction(trans, root, ret);
1202                                 return ret;
1203                         }
1204                 }
1205                 btrfs_free_tree_block(trans, root, buf, parent_start,
1206                                       last_ref);
1207         }
1208         if (unlock_orig)
1209                 btrfs_tree_unlock(buf);
1210         free_extent_buffer_stale(buf);
1211         btrfs_mark_buffer_dirty(cow);
1212         *cow_ret = cow;
1213         return 0;
1214 }
1215
1216 /*
1217  * returns the logical address of the oldest predecessor of the given root.
1218  * entries older than time_seq are ignored.
1219  */
1220 static struct tree_mod_elem *
1221 __tree_mod_log_oldest_root(struct btrfs_fs_info *fs_info,
1222                            struct extent_buffer *eb_root, u64 time_seq)
1223 {
1224         struct tree_mod_elem *tm;
1225         struct tree_mod_elem *found = NULL;
1226         u64 root_logical = eb_root->start;
1227         int looped = 0;
1228
1229         if (!time_seq)
1230                 return NULL;
1231
1232         /*
1233          * the very last operation that's logged for a root is the
1234          * replacement operation (if it is replaced at all). this has
1235          * the logical address of the *new* root, making it the very
1236          * first operation that's logged for this root.
1237          */
1238         while (1) {
1239                 tm = tree_mod_log_search_oldest(fs_info, root_logical,
1240                                                 time_seq);
1241                 if (!looped && !tm)
1242                         return NULL;
1243                 /*
1244                  * if there are no tree operation for the oldest root, we simply
1245                  * return it. this should only happen if that (old) root is at
1246                  * level 0.
1247                  */
1248                 if (!tm)
1249                         break;
1250
1251                 /*
1252                  * if there's an operation that's not a root replacement, we
1253                  * found the oldest version of our root. normally, we'll find a
1254                  * MOD_LOG_KEY_REMOVE_WHILE_FREEING operation here.
1255                  */
1256                 if (tm->op != MOD_LOG_ROOT_REPLACE)
1257                         break;
1258
1259                 found = tm;
1260                 root_logical = tm->old_root.logical;
1261                 looped = 1;
1262         }
1263
1264         /* if there's no old root to return, return what we found instead */
1265         if (!found)
1266                 found = tm;
1267
1268         return found;
1269 }
1270
1271 /*
1272  * tm is a pointer to the first operation to rewind within eb. then, all
1273  * previous operations will be rewinded (until we reach something older than
1274  * time_seq).
1275  */
1276 static void
1277 __tree_mod_log_rewind(struct btrfs_fs_info *fs_info, struct extent_buffer *eb,
1278                       u64 time_seq, struct tree_mod_elem *first_tm)
1279 {
1280         u32 n;
1281         struct rb_node *next;
1282         struct tree_mod_elem *tm = first_tm;
1283         unsigned long o_dst;
1284         unsigned long o_src;
1285         unsigned long p_size = sizeof(struct btrfs_key_ptr);
1286
1287         n = btrfs_header_nritems(eb);
1288         tree_mod_log_read_lock(fs_info);
1289         while (tm && tm->seq >= time_seq) {
1290                 /*
1291                  * all the operations are recorded with the operator used for
1292                  * the modification. as we're going backwards, we do the
1293                  * opposite of each operation here.
1294                  */
1295                 switch (tm->op) {
1296                 case MOD_LOG_KEY_REMOVE_WHILE_FREEING:
1297                         BUG_ON(tm->slot < n);
1298                         /* Fallthrough */
1299                 case MOD_LOG_KEY_REMOVE_WHILE_MOVING:
1300                 case MOD_LOG_KEY_REMOVE:
1301                         btrfs_set_node_key(eb, &tm->key, tm->slot);
1302                         btrfs_set_node_blockptr(eb, tm->slot, tm->blockptr);
1303                         btrfs_set_node_ptr_generation(eb, tm->slot,
1304                                                       tm->generation);
1305                         n++;
1306                         break;
1307                 case MOD_LOG_KEY_REPLACE:
1308                         BUG_ON(tm->slot >= n);
1309                         btrfs_set_node_key(eb, &tm->key, tm->slot);
1310                         btrfs_set_node_blockptr(eb, tm->slot, tm->blockptr);
1311                         btrfs_set_node_ptr_generation(eb, tm->slot,
1312                                                       tm->generation);
1313                         break;
1314                 case MOD_LOG_KEY_ADD:
1315                         /* if a move operation is needed it's in the log */
1316                         n--;
1317                         break;
1318                 case MOD_LOG_MOVE_KEYS:
1319                         o_dst = btrfs_node_key_ptr_offset(tm->slot);
1320                         o_src = btrfs_node_key_ptr_offset(tm->move.dst_slot);
1321                         memmove_extent_buffer(eb, o_dst, o_src,
1322                                               tm->move.nr_items * p_size);
1323                         break;
1324                 case MOD_LOG_ROOT_REPLACE:
1325                         /*
1326                          * this operation is special. for roots, this must be
1327                          * handled explicitly before rewinding.
1328                          * for non-roots, this operation may exist if the node
1329                          * was a root: root A -> child B; then A gets empty and
1330                          * B is promoted to the new root. in the mod log, we'll
1331                          * have a root-replace operation for B, a tree block
1332                          * that is no root. we simply ignore that operation.
1333                          */
1334                         break;
1335                 }
1336                 next = rb_next(&tm->node);
1337                 if (!next)
1338                         break;
1339                 tm = container_of(next, struct tree_mod_elem, node);
1340                 if (tm->logical != first_tm->logical)
1341                         break;
1342         }
1343         tree_mod_log_read_unlock(fs_info);
1344         btrfs_set_header_nritems(eb, n);
1345 }
1346
1347 /*
1348  * Called with eb read locked. If the buffer cannot be rewinded, the same buffer
1349  * is returned. If rewind operations happen, a fresh buffer is returned. The
1350  * returned buffer is always read-locked. If the returned buffer is not the
1351  * input buffer, the lock on the input buffer is released and the input buffer
1352  * is freed (its refcount is decremented).
1353  */
1354 static struct extent_buffer *
1355 tree_mod_log_rewind(struct btrfs_fs_info *fs_info, struct btrfs_path *path,
1356                     struct extent_buffer *eb, u64 time_seq)
1357 {
1358         struct extent_buffer *eb_rewin;
1359         struct tree_mod_elem *tm;
1360
1361         if (!time_seq)
1362                 return eb;
1363
1364         if (btrfs_header_level(eb) == 0)
1365                 return eb;
1366
1367         tm = tree_mod_log_search(fs_info, eb->start, time_seq);
1368         if (!tm)
1369                 return eb;
1370
1371         btrfs_set_path_blocking(path);
1372         btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
1373
1374         if (tm->op == MOD_LOG_KEY_REMOVE_WHILE_FREEING) {
1375                 BUG_ON(tm->slot != 0);
1376                 eb_rewin = alloc_dummy_extent_buffer(fs_info, eb->start);
1377                 if (!eb_rewin) {
1378                         btrfs_tree_read_unlock_blocking(eb);
1379                         free_extent_buffer(eb);
1380                         return NULL;
1381                 }
1382                 btrfs_set_header_bytenr(eb_rewin, eb->start);
1383                 btrfs_set_header_backref_rev(eb_rewin,
1384                                              btrfs_header_backref_rev(eb));
1385                 btrfs_set_header_owner(eb_rewin, btrfs_header_owner(eb));
1386                 btrfs_set_header_level(eb_rewin, btrfs_header_level(eb));
1387         } else {
1388                 eb_rewin = btrfs_clone_extent_buffer(eb);
1389                 if (!eb_rewin) {
1390                         btrfs_tree_read_unlock_blocking(eb);
1391                         free_extent_buffer(eb);
1392                         return NULL;
1393                 }
1394         }
1395
1396         btrfs_clear_path_blocking(path, NULL, BTRFS_READ_LOCK);
1397         btrfs_tree_read_unlock_blocking(eb);
1398         free_extent_buffer(eb);
1399
1400         extent_buffer_get(eb_rewin);
1401         btrfs_tree_read_lock(eb_rewin);
1402         __tree_mod_log_rewind(fs_info, eb_rewin, time_seq, tm);
1403         WARN_ON(btrfs_header_nritems(eb_rewin) >
1404                 BTRFS_NODEPTRS_PER_BLOCK(fs_info->tree_root));
1405
1406         return eb_rewin;
1407 }
1408
1409 /*
1410  * get_old_root() rewinds the state of @root's root node to the given @time_seq
1411  * value. If there are no changes, the current root->root_node is returned. If
1412  * anything changed in between, there's a fresh buffer allocated on which the
1413  * rewind operations are done. In any case, the returned buffer is read locked.
1414  * Returns NULL on error (with no locks held).
1415  */
1416 static inline struct extent_buffer *
1417 get_old_root(struct btrfs_root *root, u64 time_seq)
1418 {
1419         struct tree_mod_elem *tm;
1420         struct extent_buffer *eb = NULL;
1421         struct extent_buffer *eb_root;
1422         struct extent_buffer *old;
1423         struct tree_mod_root *old_root = NULL;
1424         u64 old_generation = 0;
1425         u64 logical;
1426
1427         eb_root = btrfs_read_lock_root_node(root);
1428         tm = __tree_mod_log_oldest_root(root->fs_info, eb_root, time_seq);
1429         if (!tm)
1430                 return eb_root;
1431
1432         if (tm->op == MOD_LOG_ROOT_REPLACE) {
1433                 old_root = &tm->old_root;
1434                 old_generation = tm->generation;
1435                 logical = old_root->logical;
1436         } else {
1437                 logical = eb_root->start;
1438         }
1439
1440         tm = tree_mod_log_search(root->fs_info, logical, time_seq);
1441         if (old_root && tm && tm->op != MOD_LOG_KEY_REMOVE_WHILE_FREEING) {
1442                 btrfs_tree_read_unlock(eb_root);
1443                 free_extent_buffer(eb_root);
1444                 old = read_tree_block(root, logical, 0);
1445                 if (WARN_ON(IS_ERR(old) || !extent_buffer_uptodate(old))) {
1446                         if (!IS_ERR(old))
1447                                 free_extent_buffer(old);
1448                         btrfs_warn(root->fs_info,
1449                                 "failed to read tree block %llu from get_old_root", logical);
1450                 } else {
1451                         eb = btrfs_clone_extent_buffer(old);
1452                         free_extent_buffer(old);
1453                 }
1454         } else if (old_root) {
1455                 btrfs_tree_read_unlock(eb_root);
1456                 free_extent_buffer(eb_root);
1457                 eb = alloc_dummy_extent_buffer(root->fs_info, logical);
1458         } else {
1459                 btrfs_set_lock_blocking_rw(eb_root, BTRFS_READ_LOCK);
1460                 eb = btrfs_clone_extent_buffer(eb_root);
1461                 btrfs_tree_read_unlock_blocking(eb_root);
1462                 free_extent_buffer(eb_root);
1463         }
1464
1465         if (!eb)
1466                 return NULL;
1467         extent_buffer_get(eb);
1468         btrfs_tree_read_lock(eb);
1469         if (old_root) {
1470                 btrfs_set_header_bytenr(eb, eb->start);
1471                 btrfs_set_header_backref_rev(eb, BTRFS_MIXED_BACKREF_REV);
1472                 btrfs_set_header_owner(eb, btrfs_header_owner(eb_root));
1473                 btrfs_set_header_level(eb, old_root->level);
1474                 btrfs_set_header_generation(eb, old_generation);
1475         }
1476         if (tm)
1477                 __tree_mod_log_rewind(root->fs_info, eb, time_seq, tm);
1478         else
1479                 WARN_ON(btrfs_header_level(eb) != 0);
1480         WARN_ON(btrfs_header_nritems(eb) > BTRFS_NODEPTRS_PER_BLOCK(root));
1481
1482         return eb;
1483 }
1484
1485 int btrfs_old_root_level(struct btrfs_root *root, u64 time_seq)
1486 {
1487         struct tree_mod_elem *tm;
1488         int level;
1489         struct extent_buffer *eb_root = btrfs_root_node(root);
1490
1491         tm = __tree_mod_log_oldest_root(root->fs_info, eb_root, time_seq);
1492         if (tm && tm->op == MOD_LOG_ROOT_REPLACE) {
1493                 level = tm->old_root.level;
1494         } else {
1495                 level = btrfs_header_level(eb_root);
1496         }
1497         free_extent_buffer(eb_root);
1498
1499         return level;
1500 }
1501
1502 static inline int should_cow_block(struct btrfs_trans_handle *trans,
1503                                    struct btrfs_root *root,
1504                                    struct extent_buffer *buf)
1505 {
1506         if (btrfs_test_is_dummy_root(root))
1507                 return 0;
1508
1509         /* ensure we can see the force_cow */
1510         smp_rmb();
1511
1512         /*
1513          * We do not need to cow a block if
1514          * 1) this block is not created or changed in this transaction;
1515          * 2) this block does not belong to TREE_RELOC tree;
1516          * 3) the root is not forced COW.
1517          *
1518          * What is forced COW:
1519          *    when we create snapshot during commiting the transaction,
1520          *    after we've finished coping src root, we must COW the shared
1521          *    block to ensure the metadata consistency.
1522          */
1523         if (btrfs_header_generation(buf) == trans->transid &&
1524             !btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN) &&
1525             !(root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID &&
1526               btrfs_header_flag(buf, BTRFS_HEADER_FLAG_RELOC)) &&
1527             !test_bit(BTRFS_ROOT_FORCE_COW, &root->state))
1528                 return 0;
1529         return 1;
1530 }
1531
1532 /*
1533  * cows a single block, see __btrfs_cow_block for the real work.
1534  * This version of it has extra checks so that a block isn't cow'd more than
1535  * once per transaction, as long as it hasn't been written yet
1536  */
1537 noinline int btrfs_cow_block(struct btrfs_trans_handle *trans,
1538                     struct btrfs_root *root, struct extent_buffer *buf,
1539                     struct extent_buffer *parent, int parent_slot,
1540                     struct extent_buffer **cow_ret)
1541 {
1542         u64 search_start;
1543         int ret;
1544
1545         if (trans->transaction != root->fs_info->running_transaction)
1546                 WARN(1, KERN_CRIT "trans %llu running %llu\n",
1547                        trans->transid,
1548                        root->fs_info->running_transaction->transid);
1549
1550         if (trans->transid != root->fs_info->generation)
1551                 WARN(1, KERN_CRIT "trans %llu running %llu\n",
1552                        trans->transid, root->fs_info->generation);
1553
1554         if (!should_cow_block(trans, root, buf)) {
1555                 *cow_ret = buf;
1556                 return 0;
1557         }
1558
1559         search_start = buf->start & ~((u64)SZ_1G - 1);
1560
1561         if (parent)
1562                 btrfs_set_lock_blocking(parent);
1563         btrfs_set_lock_blocking(buf);
1564
1565         ret = __btrfs_cow_block(trans, root, buf, parent,
1566                                  parent_slot, cow_ret, search_start, 0);
1567
1568         trace_btrfs_cow_block(root, buf, *cow_ret);
1569
1570         return ret;
1571 }
1572
1573 /*
1574  * helper function for defrag to decide if two blocks pointed to by a
1575  * node are actually close by
1576  */
1577 static int close_blocks(u64 blocknr, u64 other, u32 blocksize)
1578 {
1579         if (blocknr < other && other - (blocknr + blocksize) < 32768)
1580                 return 1;
1581         if (blocknr > other && blocknr - (other + blocksize) < 32768)
1582                 return 1;
1583         return 0;
1584 }
1585
1586 /*
1587  * compare two keys in a memcmp fashion
1588  */
1589 static int comp_keys(struct btrfs_disk_key *disk, struct btrfs_key *k2)
1590 {
1591         struct btrfs_key k1;
1592
1593         btrfs_disk_key_to_cpu(&k1, disk);
1594
1595         return btrfs_comp_cpu_keys(&k1, k2);
1596 }
1597
1598 /*
1599  * same as comp_keys only with two btrfs_key's
1600  */
1601 int btrfs_comp_cpu_keys(struct btrfs_key *k1, struct btrfs_key *k2)
1602 {
1603         if (k1->objectid > k2->objectid)
1604                 return 1;
1605         if (k1->objectid < k2->objectid)
1606                 return -1;
1607         if (k1->type > k2->type)
1608                 return 1;
1609         if (k1->type < k2->type)
1610                 return -1;
1611         if (k1->offset > k2->offset)
1612                 return 1;
1613         if (k1->offset < k2->offset)
1614                 return -1;
1615         return 0;
1616 }
1617
1618 /*
1619  * this is used by the defrag code to go through all the
1620  * leaves pointed to by a node and reallocate them so that
1621  * disk order is close to key order
1622  */
1623 int btrfs_realloc_node(struct btrfs_trans_handle *trans,
1624                        struct btrfs_root *root, struct extent_buffer *parent,
1625                        int start_slot, u64 *last_ret,
1626                        struct btrfs_key *progress)
1627 {
1628         struct extent_buffer *cur;
1629         u64 blocknr;
1630         u64 gen;
1631         u64 search_start = *last_ret;
1632         u64 last_block = 0;
1633         u64 other;
1634         u32 parent_nritems;
1635         int end_slot;
1636         int i;
1637         int err = 0;
1638         int parent_level;
1639         int uptodate;
1640         u32 blocksize;
1641         int progress_passed = 0;
1642         struct btrfs_disk_key disk_key;
1643
1644         parent_level = btrfs_header_level(parent);
1645
1646         WARN_ON(trans->transaction != root->fs_info->running_transaction);
1647         WARN_ON(trans->transid != root->fs_info->generation);
1648
1649         parent_nritems = btrfs_header_nritems(parent);
1650         blocksize = root->nodesize;
1651         end_slot = parent_nritems - 1;
1652
1653         if (parent_nritems <= 1)
1654                 return 0;
1655
1656         btrfs_set_lock_blocking(parent);
1657
1658         for (i = start_slot; i <= end_slot; i++) {
1659                 int close = 1;
1660
1661                 btrfs_node_key(parent, &disk_key, i);
1662                 if (!progress_passed && comp_keys(&disk_key, progress) < 0)
1663                         continue;
1664
1665                 progress_passed = 1;
1666                 blocknr = btrfs_node_blockptr(parent, i);
1667                 gen = btrfs_node_ptr_generation(parent, i);
1668                 if (last_block == 0)
1669                         last_block = blocknr;
1670
1671                 if (i > 0) {
1672                         other = btrfs_node_blockptr(parent, i - 1);
1673                         close = close_blocks(blocknr, other, blocksize);
1674                 }
1675                 if (!close && i < end_slot) {
1676                         other = btrfs_node_blockptr(parent, i + 1);
1677                         close = close_blocks(blocknr, other, blocksize);
1678                 }
1679                 if (close) {
1680                         last_block = blocknr;
1681                         continue;
1682                 }
1683
1684                 cur = btrfs_find_tree_block(root->fs_info, blocknr);
1685                 if (cur)
1686                         uptodate = btrfs_buffer_uptodate(cur, gen, 0);
1687                 else
1688                         uptodate = 0;
1689                 if (!cur || !uptodate) {
1690                         if (!cur) {
1691                                 cur = read_tree_block(root, blocknr, gen);
1692                                 if (IS_ERR(cur)) {
1693                                         return PTR_ERR(cur);
1694                                 } else if (!extent_buffer_uptodate(cur)) {
1695                                         free_extent_buffer(cur);
1696                                         return -EIO;
1697                                 }
1698                         } else if (!uptodate) {
1699                                 err = btrfs_read_buffer(cur, gen);
1700                                 if (err) {
1701                                         free_extent_buffer(cur);
1702                                         return err;
1703                                 }
1704                         }
1705                 }
1706                 if (search_start == 0)
1707                         search_start = last_block;
1708
1709                 btrfs_tree_lock(cur);
1710                 btrfs_set_lock_blocking(cur);
1711                 err = __btrfs_cow_block(trans, root, cur, parent, i,
1712                                         &cur, search_start,
1713                                         min(16 * blocksize,
1714                                             (end_slot - i) * blocksize));
1715                 if (err) {
1716                         btrfs_tree_unlock(cur);
1717                         free_extent_buffer(cur);
1718                         break;
1719                 }
1720                 search_start = cur->start;
1721                 last_block = cur->start;
1722                 *last_ret = search_start;
1723                 btrfs_tree_unlock(cur);
1724                 free_extent_buffer(cur);
1725         }
1726         return err;
1727 }
1728
1729 /*
1730  * The leaf data grows from end-to-front in the node.
1731  * this returns the address of the start of the last item,
1732  * which is the stop of the leaf data stack
1733  */
1734 static inline unsigned int leaf_data_end(struct btrfs_root *root,
1735                                          struct extent_buffer *leaf)
1736 {
1737         u32 nr = btrfs_header_nritems(leaf);
1738         if (nr == 0)
1739                 return BTRFS_LEAF_DATA_SIZE(root);
1740         return btrfs_item_offset_nr(leaf, nr - 1);
1741 }
1742
1743
1744 /*
1745  * search for key in the extent_buffer.  The items start at offset p,
1746  * and they are item_size apart.  There are 'max' items in p.
1747  *
1748  * the slot in the array is returned via slot, and it points to
1749  * the place where you would insert key if it is not found in
1750  * the array.
1751  *
1752  * slot may point to max if the key is bigger than all of the keys
1753  */
1754 static noinline int generic_bin_search(struct extent_buffer *eb,
1755                                        unsigned long p,
1756                                        int item_size, struct btrfs_key *key,
1757                                        int max, int *slot)
1758 {
1759         int low = 0;
1760         int high = max;
1761         int mid;
1762         int ret;
1763         struct btrfs_disk_key *tmp = NULL;
1764         struct btrfs_disk_key unaligned;
1765         unsigned long offset;
1766         char *kaddr = NULL;
1767         unsigned long map_start = 0;
1768         unsigned long map_len = 0;
1769         int err;
1770
1771         while (low < high) {
1772                 mid = (low + high) / 2;
1773                 offset = p + mid * item_size;
1774
1775                 if (!kaddr || offset < map_start ||
1776                     (offset + sizeof(struct btrfs_disk_key)) >
1777                     map_start + map_len) {
1778
1779                         err = map_private_extent_buffer(eb, offset,
1780                                                 sizeof(struct btrfs_disk_key),
1781                                                 &kaddr, &map_start, &map_len);
1782
1783                         if (!err) {
1784                                 tmp = (struct btrfs_disk_key *)(kaddr + offset -
1785                                                         map_start);
1786                         } else {
1787                                 read_extent_buffer(eb, &unaligned,
1788                                                    offset, sizeof(unaligned));
1789                                 tmp = &unaligned;
1790                         }
1791
1792                 } else {
1793                         tmp = (struct btrfs_disk_key *)(kaddr + offset -
1794                                                         map_start);
1795                 }
1796                 ret = comp_keys(tmp, key);
1797
1798                 if (ret < 0)
1799                         low = mid + 1;
1800                 else if (ret > 0)
1801                         high = mid;
1802                 else {
1803                         *slot = mid;
1804                         return 0;
1805                 }
1806         }
1807         *slot = low;
1808         return 1;
1809 }
1810
1811 /*
1812  * simple bin_search frontend that does the right thing for
1813  * leaves vs nodes
1814  */
1815 static int bin_search(struct extent_buffer *eb, struct btrfs_key *key,
1816                       int level, int *slot)
1817 {
1818         if (level == 0)
1819                 return generic_bin_search(eb,
1820                                           offsetof(struct btrfs_leaf, items),
1821                                           sizeof(struct btrfs_item),
1822                                           key, btrfs_header_nritems(eb),
1823                                           slot);
1824         else
1825                 return generic_bin_search(eb,
1826                                           offsetof(struct btrfs_node, ptrs),
1827                                           sizeof(struct btrfs_key_ptr),
1828                                           key, btrfs_header_nritems(eb),
1829                                           slot);
1830 }
1831
1832 int btrfs_bin_search(struct extent_buffer *eb, struct btrfs_key *key,
1833                      int level, int *slot)
1834 {
1835         return bin_search(eb, key, level, slot);
1836 }
1837
1838 static void root_add_used(struct btrfs_root *root, u32 size)
1839 {
1840         spin_lock(&root->accounting_lock);
1841         btrfs_set_root_used(&root->root_item,
1842                             btrfs_root_used(&root->root_item) + size);
1843         spin_unlock(&root->accounting_lock);
1844 }
1845
1846 static void root_sub_used(struct btrfs_root *root, u32 size)
1847 {
1848         spin_lock(&root->accounting_lock);
1849         btrfs_set_root_used(&root->root_item,
1850                             btrfs_root_used(&root->root_item) - size);
1851         spin_unlock(&root->accounting_lock);
1852 }
1853
1854 /* given a node and slot number, this reads the blocks it points to.  The
1855  * extent buffer is returned with a reference taken (but unlocked).
1856  * NULL is returned on error.
1857  */
1858 static noinline struct extent_buffer *read_node_slot(struct btrfs_root *root,
1859                                    struct extent_buffer *parent, int slot)
1860 {
1861         int level = btrfs_header_level(parent);
1862         struct extent_buffer *eb;
1863
1864         if (slot < 0)
1865                 return NULL;
1866         if (slot >= btrfs_header_nritems(parent))
1867                 return NULL;
1868
1869         BUG_ON(level == 0);
1870
1871         eb = read_tree_block(root, btrfs_node_blockptr(parent, slot),
1872                              btrfs_node_ptr_generation(parent, slot));
1873         if (IS_ERR(eb) || !extent_buffer_uptodate(eb)) {
1874                 if (!IS_ERR(eb))
1875                         free_extent_buffer(eb);
1876                 eb = NULL;
1877         }
1878
1879         return eb;
1880 }
1881
1882 /*
1883  * node level balancing, used to make sure nodes are in proper order for
1884  * item deletion.  We balance from the top down, so we have to make sure
1885  * that a deletion won't leave an node completely empty later on.
1886  */
1887 static noinline int balance_level(struct btrfs_trans_handle *trans,
1888                          struct btrfs_root *root,
1889                          struct btrfs_path *path, int level)
1890 {
1891         struct extent_buffer *right = NULL;
1892         struct extent_buffer *mid;
1893         struct extent_buffer *left = NULL;
1894         struct extent_buffer *parent = NULL;
1895         int ret = 0;
1896         int wret;
1897         int pslot;
1898         int orig_slot = path->slots[level];
1899         u64 orig_ptr;
1900
1901         if (level == 0)
1902                 return 0;
1903
1904         mid = path->nodes[level];
1905
1906         WARN_ON(path->locks[level] != BTRFS_WRITE_LOCK &&
1907                 path->locks[level] != BTRFS_WRITE_LOCK_BLOCKING);
1908         WARN_ON(btrfs_header_generation(mid) != trans->transid);
1909
1910         orig_ptr = btrfs_node_blockptr(mid, orig_slot);
1911
1912         if (level < BTRFS_MAX_LEVEL - 1) {
1913                 parent = path->nodes[level + 1];
1914                 pslot = path->slots[level + 1];
1915         }
1916
1917         /*
1918          * deal with the case where there is only one pointer in the root
1919          * by promoting the node below to a root
1920          */
1921         if (!parent) {
1922                 struct extent_buffer *child;
1923
1924                 if (btrfs_header_nritems(mid) != 1)
1925                         return 0;
1926
1927                 /* promote the child to a root */
1928                 child = read_node_slot(root, mid, 0);
1929                 if (!child) {
1930                         ret = -EROFS;
1931                         btrfs_std_error(root->fs_info, ret, NULL);
1932                         goto enospc;
1933                 }
1934
1935                 btrfs_tree_lock(child);
1936                 btrfs_set_lock_blocking(child);
1937                 ret = btrfs_cow_block(trans, root, child, mid, 0, &child);
1938                 if (ret) {
1939                         btrfs_tree_unlock(child);
1940                         free_extent_buffer(child);
1941                         goto enospc;
1942                 }
1943
1944                 tree_mod_log_set_root_pointer(root, child, 1);
1945                 rcu_assign_pointer(root->node, child);
1946
1947                 add_root_to_dirty_list(root);
1948                 btrfs_tree_unlock(child);
1949
1950                 path->locks[level] = 0;
1951                 path->nodes[level] = NULL;
1952                 clean_tree_block(trans, root->fs_info, mid);
1953                 btrfs_tree_unlock(mid);
1954                 /* once for the path */
1955                 free_extent_buffer(mid);
1956
1957                 root_sub_used(root, mid->len);
1958                 btrfs_free_tree_block(trans, root, mid, 0, 1);
1959                 /* once for the root ptr */
1960                 free_extent_buffer_stale(mid);
1961                 return 0;
1962         }
1963         if (btrfs_header_nritems(mid) >
1964             BTRFS_NODEPTRS_PER_BLOCK(root) / 4)
1965                 return 0;
1966
1967         left = read_node_slot(root, parent, pslot - 1);
1968         if (left) {
1969                 btrfs_tree_lock(left);
1970                 btrfs_set_lock_blocking(left);
1971                 wret = btrfs_cow_block(trans, root, left,
1972                                        parent, pslot - 1, &left);
1973                 if (wret) {
1974                         ret = wret;
1975                         goto enospc;
1976                 }
1977         }
1978         right = read_node_slot(root, parent, pslot + 1);
1979         if (right) {
1980                 btrfs_tree_lock(right);
1981                 btrfs_set_lock_blocking(right);
1982                 wret = btrfs_cow_block(trans, root, right,
1983                                        parent, pslot + 1, &right);
1984                 if (wret) {
1985                         ret = wret;
1986                         goto enospc;
1987                 }
1988         }
1989
1990         /* first, try to make some room in the middle buffer */
1991         if (left) {
1992                 orig_slot += btrfs_header_nritems(left);
1993                 wret = push_node_left(trans, root, left, mid, 1);
1994                 if (wret < 0)
1995                         ret = wret;
1996         }
1997
1998         /*
1999          * then try to empty the right most buffer into the middle
2000          */
2001         if (right) {
2002                 wret = push_node_left(trans, root, mid, right, 1);
2003                 if (wret < 0 && wret != -ENOSPC)
2004                         ret = wret;
2005                 if (btrfs_header_nritems(right) == 0) {
2006                         clean_tree_block(trans, root->fs_info, right);
2007                         btrfs_tree_unlock(right);
2008                         del_ptr(root, path, level + 1, pslot + 1);
2009                         root_sub_used(root, right->len);
2010                         btrfs_free_tree_block(trans, root, right, 0, 1);
2011                         free_extent_buffer_stale(right);
2012                         right = NULL;
2013                 } else {
2014                         struct btrfs_disk_key right_key;
2015                         btrfs_node_key(right, &right_key, 0);
2016                         tree_mod_log_set_node_key(root->fs_info, parent,
2017                                                   pslot + 1, 0);
2018                         btrfs_set_node_key(parent, &right_key, pslot + 1);
2019                         btrfs_mark_buffer_dirty(parent);
2020                 }
2021         }
2022         if (btrfs_header_nritems(mid) == 1) {
2023                 /*
2024                  * we're not allowed to leave a node with one item in the
2025                  * tree during a delete.  A deletion from lower in the tree
2026                  * could try to delete the only pointer in this node.
2027                  * So, pull some keys from the left.
2028                  * There has to be a left pointer at this point because
2029                  * otherwise we would have pulled some pointers from the
2030                  * right
2031                  */
2032                 if (!left) {
2033                         ret = -EROFS;
2034                         btrfs_std_error(root->fs_info, ret, NULL);
2035                         goto enospc;
2036                 }
2037                 wret = balance_node_right(trans, root, mid, left);
2038                 if (wret < 0) {
2039                         ret = wret;
2040                         goto enospc;
2041                 }
2042                 if (wret == 1) {
2043                         wret = push_node_left(trans, root, left, mid, 1);
2044                         if (wret < 0)
2045                                 ret = wret;
2046                 }
2047                 BUG_ON(wret == 1);
2048         }
2049         if (btrfs_header_nritems(mid) == 0) {
2050                 clean_tree_block(trans, root->fs_info, mid);
2051                 btrfs_tree_unlock(mid);
2052                 del_ptr(root, path, level + 1, pslot);
2053                 root_sub_used(root, mid->len);
2054                 btrfs_free_tree_block(trans, root, mid, 0, 1);
2055                 free_extent_buffer_stale(mid);
2056                 mid = NULL;
2057         } else {
2058                 /* update the parent key to reflect our changes */
2059                 struct btrfs_disk_key mid_key;
2060                 btrfs_node_key(mid, &mid_key, 0);
2061                 tree_mod_log_set_node_key(root->fs_info, parent,
2062                                           pslot, 0);
2063                 btrfs_set_node_key(parent, &mid_key, pslot);
2064                 btrfs_mark_buffer_dirty(parent);
2065         }
2066
2067         /* update the path */
2068         if (left) {
2069                 if (btrfs_header_nritems(left) > orig_slot) {
2070                         extent_buffer_get(left);
2071                         /* left was locked after cow */
2072                         path->nodes[level] = left;
2073                         path->slots[level + 1] -= 1;
2074                         path->slots[level] = orig_slot;
2075                         if (mid) {
2076                                 btrfs_tree_unlock(mid);
2077                                 free_extent_buffer(mid);
2078                         }
2079                 } else {
2080                         orig_slot -= btrfs_header_nritems(left);
2081                         path->slots[level] = orig_slot;
2082                 }
2083         }
2084         /* double check we haven't messed things up */
2085         if (orig_ptr !=
2086             btrfs_node_blockptr(path->nodes[level], path->slots[level]))
2087                 BUG();
2088 enospc:
2089         if (right) {
2090                 btrfs_tree_unlock(right);
2091                 free_extent_buffer(right);
2092         }
2093         if (left) {
2094                 if (path->nodes[level] != left)
2095                         btrfs_tree_unlock(left);
2096                 free_extent_buffer(left);
2097         }
2098         return ret;
2099 }
2100
2101 /* Node balancing for insertion.  Here we only split or push nodes around
2102  * when they are completely full.  This is also done top down, so we
2103  * have to be pessimistic.
2104  */
2105 static noinline int push_nodes_for_insert(struct btrfs_trans_handle *trans,
2106                                           struct btrfs_root *root,
2107                                           struct btrfs_path *path, int level)
2108 {
2109         struct extent_buffer *right = NULL;
2110         struct extent_buffer *mid;
2111         struct extent_buffer *left = NULL;
2112         struct extent_buffer *parent = NULL;
2113         int ret = 0;
2114         int wret;
2115         int pslot;
2116         int orig_slot = path->slots[level];
2117
2118         if (level == 0)
2119                 return 1;
2120
2121         mid = path->nodes[level];
2122         WARN_ON(btrfs_header_generation(mid) != trans->transid);
2123
2124         if (level < BTRFS_MAX_LEVEL - 1) {
2125                 parent = path->nodes[level + 1];
2126                 pslot = path->slots[level + 1];
2127         }
2128
2129         if (!parent)
2130                 return 1;
2131
2132         left = read_node_slot(root, parent, pslot - 1);
2133
2134         /* first, try to make some room in the middle buffer */
2135         if (left) {
2136                 u32 left_nr;
2137
2138                 btrfs_tree_lock(left);
2139                 btrfs_set_lock_blocking(left);
2140
2141                 left_nr = btrfs_header_nritems(left);
2142                 if (left_nr >= BTRFS_NODEPTRS_PER_BLOCK(root) - 1) {
2143                         wret = 1;
2144                 } else {
2145                         ret = btrfs_cow_block(trans, root, left, parent,
2146                                               pslot - 1, &left);
2147                         if (ret)
2148                                 wret = 1;
2149                         else {
2150                                 wret = push_node_left(trans, root,
2151                                                       left, mid, 0);
2152                         }
2153                 }
2154                 if (wret < 0)
2155                         ret = wret;
2156                 if (wret == 0) {
2157                         struct btrfs_disk_key disk_key;
2158                         orig_slot += left_nr;
2159                         btrfs_node_key(mid, &disk_key, 0);
2160                         tree_mod_log_set_node_key(root->fs_info, parent,
2161                                                   pslot, 0);
2162                         btrfs_set_node_key(parent, &disk_key, pslot);
2163                         btrfs_mark_buffer_dirty(parent);
2164                         if (btrfs_header_nritems(left) > orig_slot) {
2165                                 path->nodes[level] = left;
2166                                 path->slots[level + 1] -= 1;
2167                                 path->slots[level] = orig_slot;
2168                                 btrfs_tree_unlock(mid);
2169                                 free_extent_buffer(mid);
2170                         } else {
2171                                 orig_slot -=
2172                                         btrfs_header_nritems(left);
2173                                 path->slots[level] = orig_slot;
2174                                 btrfs_tree_unlock(left);
2175                                 free_extent_buffer(left);
2176                         }
2177                         return 0;
2178                 }
2179                 btrfs_tree_unlock(left);
2180                 free_extent_buffer(left);
2181         }
2182         right = read_node_slot(root, parent, pslot + 1);
2183
2184         /*
2185          * then try to empty the right most buffer into the middle
2186          */
2187         if (right) {
2188                 u32 right_nr;
2189
2190                 btrfs_tree_lock(right);
2191                 btrfs_set_lock_blocking(right);
2192
2193                 right_nr = btrfs_header_nritems(right);
2194                 if (right_nr >= BTRFS_NODEPTRS_PER_BLOCK(root) - 1) {
2195                         wret = 1;
2196                 } else {
2197                         ret = btrfs_cow_block(trans, root, right,
2198                                               parent, pslot + 1,
2199                                               &right);
2200                         if (ret)
2201                                 wret = 1;
2202                         else {
2203                                 wret = balance_node_right(trans, root,
2204                                                           right, mid);
2205                         }
2206                 }
2207                 if (wret < 0)
2208                         ret = wret;
2209                 if (wret == 0) {
2210                         struct btrfs_disk_key disk_key;
2211
2212                         btrfs_node_key(right, &disk_key, 0);
2213                         tree_mod_log_set_node_key(root->fs_info, parent,
2214                                                   pslot + 1, 0);
2215                         btrfs_set_node_key(parent, &disk_key, pslot + 1);
2216                         btrfs_mark_buffer_dirty(parent);
2217
2218                         if (btrfs_header_nritems(mid) <= orig_slot) {
2219                                 path->nodes[level] = right;
2220                                 path->slots[level + 1] += 1;
2221                                 path->slots[level] = orig_slot -
2222                                         btrfs_header_nritems(mid);
2223                                 btrfs_tree_unlock(mid);
2224                                 free_extent_buffer(mid);
2225                         } else {
2226                                 btrfs_tree_unlock(right);
2227                                 free_extent_buffer(right);
2228                         }
2229                         return 0;
2230                 }
2231                 btrfs_tree_unlock(right);
2232                 free_extent_buffer(right);
2233         }
2234         return 1;
2235 }
2236
2237 /*
2238  * readahead one full node of leaves, finding things that are close
2239  * to the block in 'slot', and triggering ra on them.
2240  */
2241 static void reada_for_search(struct btrfs_root *root,
2242                              struct btrfs_path *path,
2243                              int level, int slot, u64 objectid)
2244 {
2245         struct extent_buffer *node;
2246         struct btrfs_disk_key disk_key;
2247         u32 nritems;
2248         u64 search;
2249         u64 target;
2250         u64 nread = 0;
2251         u64 gen;
2252         struct extent_buffer *eb;
2253         u32 nr;
2254         u32 blocksize;
2255         u32 nscan = 0;
2256
2257         if (level != 1)
2258                 return;
2259
2260         if (!path->nodes[level])
2261                 return;
2262
2263         node = path->nodes[level];
2264
2265         search = btrfs_node_blockptr(node, slot);
2266         blocksize = root->nodesize;
2267         eb = btrfs_find_tree_block(root->fs_info, search);
2268         if (eb) {
2269                 free_extent_buffer(eb);
2270                 return;
2271         }
2272
2273         target = search;
2274
2275         nritems = btrfs_header_nritems(node);
2276         nr = slot;
2277
2278         while (1) {
2279                 if (path->reada == READA_BACK) {
2280                         if (nr == 0)
2281                                 break;
2282                         nr--;
2283                 } else if (path->reada == READA_FORWARD) {
2284                         nr++;
2285                         if (nr >= nritems)
2286                                 break;
2287                 }
2288                 if (path->reada == READA_BACK && objectid) {
2289                         btrfs_node_key(node, &disk_key, nr);
2290                         if (btrfs_disk_key_objectid(&disk_key) != objectid)
2291                                 break;
2292                 }
2293                 search = btrfs_node_blockptr(node, nr);
2294                 if ((search <= target && target - search <= 65536) ||
2295                     (search > target && search - target <= 65536)) {
2296                         gen = btrfs_node_ptr_generation(node, nr);
2297                         readahead_tree_block(root, search);
2298                         nread += blocksize;
2299                 }
2300                 nscan++;
2301                 if ((nread > 65536 || nscan > 32))
2302                         break;
2303         }
2304 }
2305
2306 static noinline void reada_for_balance(struct btrfs_root *root,
2307                                        struct btrfs_path *path, int level)
2308 {
2309         int slot;
2310         int nritems;
2311         struct extent_buffer *parent;
2312         struct extent_buffer *eb;
2313         u64 gen;
2314         u64 block1 = 0;
2315         u64 block2 = 0;
2316
2317         parent = path->nodes[level + 1];
2318         if (!parent)
2319                 return;
2320
2321         nritems = btrfs_header_nritems(parent);
2322         slot = path->slots[level + 1];
2323
2324         if (slot > 0) {
2325                 block1 = btrfs_node_blockptr(parent, slot - 1);
2326                 gen = btrfs_node_ptr_generation(parent, slot - 1);
2327                 eb = btrfs_find_tree_block(root->fs_info, block1);
2328                 /*
2329                  * if we get -eagain from btrfs_buffer_uptodate, we
2330                  * don't want to return eagain here.  That will loop
2331                  * forever
2332                  */
2333                 if (eb && btrfs_buffer_uptodate(eb, gen, 1) != 0)
2334                         block1 = 0;
2335                 free_extent_buffer(eb);
2336         }
2337         if (slot + 1 < nritems) {
2338                 block2 = btrfs_node_blockptr(parent, slot + 1);
2339                 gen = btrfs_node_ptr_generation(parent, slot + 1);
2340                 eb = btrfs_find_tree_block(root->fs_info, block2);
2341                 if (eb && btrfs_buffer_uptodate(eb, gen, 1) != 0)
2342                         block2 = 0;
2343                 free_extent_buffer(eb);
2344         }
2345
2346         if (block1)
2347                 readahead_tree_block(root, block1);
2348         if (block2)
2349                 readahead_tree_block(root, block2);
2350 }
2351
2352
2353 /*
2354  * when we walk down the tree, it is usually safe to unlock the higher layers
2355  * in the tree.  The exceptions are when our path goes through slot 0, because
2356  * operations on the tree might require changing key pointers higher up in the
2357  * tree.
2358  *
2359  * callers might also have set path->keep_locks, which tells this code to keep
2360  * the lock if the path points to the last slot in the block.  This is part of
2361  * walking through the tree, and selecting the next slot in the higher block.
2362  *
2363  * lowest_unlock sets the lowest level in the tree we're allowed to unlock.  so
2364  * if lowest_unlock is 1, level 0 won't be unlocked
2365  */
2366 static noinline void unlock_up(struct btrfs_path *path, int level,
2367                                int lowest_unlock, int min_write_lock_level,
2368                                int *write_lock_level)
2369 {
2370         int i;
2371         int skip_level = level;
2372         int no_skips = 0;
2373         struct extent_buffer *t;
2374
2375         for (i = level; i < BTRFS_MAX_LEVEL; i++) {
2376                 if (!path->nodes[i])
2377                         break;
2378                 if (!path->locks[i])
2379                         break;
2380                 if (!no_skips && path->slots[i] == 0) {
2381                         skip_level = i + 1;
2382                         continue;
2383                 }
2384                 if (!no_skips && path->keep_locks) {
2385                         u32 nritems;
2386                         t = path->nodes[i];
2387                         nritems = btrfs_header_nritems(t);
2388                         if (nritems < 1 || path->slots[i] >= nritems - 1) {
2389                                 skip_level = i + 1;
2390                                 continue;
2391                         }
2392                 }
2393                 if (skip_level < i && i >= lowest_unlock)
2394                         no_skips = 1;
2395
2396                 t = path->nodes[i];
2397                 if (i >= lowest_unlock && i > skip_level && path->locks[i]) {
2398                         btrfs_tree_unlock_rw(t, path->locks[i]);
2399                         path->locks[i] = 0;
2400                         if (write_lock_level &&
2401                             i > min_write_lock_level &&
2402                             i <= *write_lock_level) {
2403                                 *write_lock_level = i - 1;
2404                         }
2405                 }
2406         }
2407 }
2408
2409 /*
2410  * This releases any locks held in the path starting at level and
2411  * going all the way up to the root.
2412  *
2413  * btrfs_search_slot will keep the lock held on higher nodes in a few
2414  * corner cases, such as COW of the block at slot zero in the node.  This
2415  * ignores those rules, and it should only be called when there are no
2416  * more updates to be done higher up in the tree.
2417  */
2418 noinline void btrfs_unlock_up_safe(struct btrfs_path *path, int level)
2419 {
2420         int i;
2421
2422         if (path->keep_locks)
2423                 return;
2424
2425         for (i = level; i < BTRFS_MAX_LEVEL; i++) {
2426                 if (!path->nodes[i])
2427                         continue;
2428                 if (!path->locks[i])
2429                         continue;
2430                 btrfs_tree_unlock_rw(path->nodes[i], path->locks[i]);
2431                 path->locks[i] = 0;
2432         }
2433 }
2434
2435 /*
2436  * helper function for btrfs_search_slot.  The goal is to find a block
2437  * in cache without setting the path to blocking.  If we find the block
2438  * we return zero and the path is unchanged.
2439  *
2440  * If we can't find the block, we set the path blocking and do some
2441  * reada.  -EAGAIN is returned and the search must be repeated.
2442  */
2443 static int
2444 read_block_for_search(struct btrfs_trans_handle *trans,
2445                        struct btrfs_root *root, struct btrfs_path *p,
2446                        struct extent_buffer **eb_ret, int level, int slot,
2447                        struct btrfs_key *key, u64 time_seq)
2448 {
2449         u64 blocknr;
2450         u64 gen;
2451         struct extent_buffer *b = *eb_ret;
2452         struct extent_buffer *tmp;
2453         int ret;
2454
2455         blocknr = btrfs_node_blockptr(b, slot);
2456         gen = btrfs_node_ptr_generation(b, slot);
2457
2458         tmp = btrfs_find_tree_block(root->fs_info, blocknr);
2459         if (tmp) {
2460                 /* first we do an atomic uptodate check */
2461                 if (btrfs_buffer_uptodate(tmp, gen, 1) > 0) {
2462                         *eb_ret = tmp;
2463                         return 0;
2464                 }
2465
2466                 /* the pages were up to date, but we failed
2467                  * the generation number check.  Do a full
2468                  * read for the generation number that is correct.
2469                  * We must do this without dropping locks so
2470                  * we can trust our generation number
2471                  */
2472                 btrfs_set_path_blocking(p);
2473
2474                 /* now we're allowed to do a blocking uptodate check */
2475                 ret = btrfs_read_buffer(tmp, gen);
2476                 if (!ret) {
2477                         *eb_ret = tmp;
2478                         return 0;
2479                 }
2480                 free_extent_buffer(tmp);
2481                 btrfs_release_path(p);
2482                 return -EIO;
2483         }
2484
2485         /*
2486          * reduce lock contention at high levels
2487          * of the btree by dropping locks before
2488          * we read.  Don't release the lock on the current
2489          * level because we need to walk this node to figure
2490          * out which blocks to read.
2491          */
2492         btrfs_unlock_up_safe(p, level + 1);
2493         btrfs_set_path_blocking(p);
2494
2495         free_extent_buffer(tmp);
2496         if (p->reada != READA_NONE)
2497                 reada_for_search(root, p, level, slot, key->objectid);
2498
2499         btrfs_release_path(p);
2500
2501         ret = -EAGAIN;
2502         tmp = read_tree_block(root, blocknr, 0);
2503         if (!IS_ERR(tmp)) {
2504                 /*
2505                  * If the read above didn't mark this buffer up to date,
2506                  * it will never end up being up to date.  Set ret to EIO now
2507                  * and give up so that our caller doesn't loop forever
2508                  * on our EAGAINs.
2509                  */
2510                 if (!btrfs_buffer_uptodate(tmp, 0, 0))
2511                         ret = -EIO;
2512                 free_extent_buffer(tmp);
2513         }
2514         return ret;
2515 }
2516
2517 /*
2518  * helper function for btrfs_search_slot.  This does all of the checks
2519  * for node-level blocks and does any balancing required based on
2520  * the ins_len.
2521  *
2522  * If no extra work was required, zero is returned.  If we had to
2523  * drop the path, -EAGAIN is returned and btrfs_search_slot must
2524  * start over
2525  */
2526 static int
2527 setup_nodes_for_search(struct btrfs_trans_handle *trans,
2528                        struct btrfs_root *root, struct btrfs_path *p,
2529                        struct extent_buffer *b, int level, int ins_len,
2530                        int *write_lock_level)
2531 {
2532         int ret;
2533         if ((p->search_for_split || ins_len > 0) && btrfs_header_nritems(b) >=
2534             BTRFS_NODEPTRS_PER_BLOCK(root) - 3) {
2535                 int sret;
2536
2537                 if (*write_lock_level < level + 1) {
2538                         *write_lock_level = level + 1;
2539                         btrfs_release_path(p);
2540                         goto again;
2541                 }
2542
2543                 btrfs_set_path_blocking(p);
2544                 reada_for_balance(root, p, level);
2545                 sret = split_node(trans, root, p, level);
2546                 btrfs_clear_path_blocking(p, NULL, 0);
2547
2548                 BUG_ON(sret > 0);
2549                 if (sret) {
2550                         ret = sret;
2551                         goto done;
2552                 }
2553                 b = p->nodes[level];
2554         } else if (ins_len < 0 && btrfs_header_nritems(b) <
2555                    BTRFS_NODEPTRS_PER_BLOCK(root) / 2) {
2556                 int sret;
2557
2558                 if (*write_lock_level < level + 1) {
2559                         *write_lock_level = level + 1;
2560                         btrfs_release_path(p);
2561                         goto again;
2562                 }
2563
2564                 btrfs_set_path_blocking(p);
2565                 reada_for_balance(root, p, level);
2566                 sret = balance_level(trans, root, p, level);
2567                 btrfs_clear_path_blocking(p, NULL, 0);
2568
2569                 if (sret) {
2570                         ret = sret;
2571                         goto done;
2572                 }
2573                 b = p->nodes[level];
2574                 if (!b) {
2575                         btrfs_release_path(p);
2576                         goto again;
2577                 }
2578                 BUG_ON(btrfs_header_nritems(b) == 1);
2579         }
2580         return 0;
2581
2582 again:
2583         ret = -EAGAIN;
2584 done:
2585         return ret;
2586 }
2587
2588 static void key_search_validate(struct extent_buffer *b,
2589                                 struct btrfs_key *key,
2590                                 int level)
2591 {
2592 #ifdef CONFIG_BTRFS_ASSERT
2593         struct btrfs_disk_key disk_key;
2594
2595         btrfs_cpu_key_to_disk(&disk_key, key);
2596
2597         if (level == 0)
2598                 ASSERT(!memcmp_extent_buffer(b, &disk_key,
2599                     offsetof(struct btrfs_leaf, items[0].key),
2600                     sizeof(disk_key)));
2601         else
2602                 ASSERT(!memcmp_extent_buffer(b, &disk_key,
2603                     offsetof(struct btrfs_node, ptrs[0].key),
2604                     sizeof(disk_key)));
2605 #endif
2606 }
2607
2608 static int key_search(struct extent_buffer *b, struct btrfs_key *key,
2609                       int level, int *prev_cmp, int *slot)
2610 {
2611         if (*prev_cmp != 0) {
2612                 *prev_cmp = bin_search(b, key, level, slot);
2613                 return *prev_cmp;
2614         }
2615
2616         key_search_validate(b, key, level);
2617         *slot = 0;
2618
2619         return 0;
2620 }
2621
2622 int btrfs_find_item(struct btrfs_root *fs_root, struct btrfs_path *path,
2623                 u64 iobjectid, u64 ioff, u8 key_type,
2624                 struct btrfs_key *found_key)
2625 {
2626         int ret;
2627         struct btrfs_key key;
2628         struct extent_buffer *eb;
2629
2630         ASSERT(path);
2631         ASSERT(found_key);
2632
2633         key.type = key_type;
2634         key.objectid = iobjectid;
2635         key.offset = ioff;
2636
2637         ret = btrfs_search_slot(NULL, fs_root, &key, path, 0, 0);
2638         if (ret < 0)
2639                 return ret;
2640
2641         eb = path->nodes[0];
2642         if (ret && path->slots[0] >= btrfs_header_nritems(eb)) {
2643                 ret = btrfs_next_leaf(fs_root, path);
2644                 if (ret)
2645                         return ret;
2646                 eb = path->nodes[0];
2647         }
2648
2649         btrfs_item_key_to_cpu(eb, found_key, path->slots[0]);
2650         if (found_key->type != key.type ||
2651                         found_key->objectid != key.objectid)
2652                 return 1;
2653
2654         return 0;
2655 }
2656
2657 /*
2658  * look for key in the tree.  path is filled in with nodes along the way
2659  * if key is found, we return zero and you can find the item in the leaf
2660  * level of the path (level 0)
2661  *
2662  * If the key isn't found, the path points to the slot where it should
2663  * be inserted, and 1 is returned.  If there are other errors during the
2664  * search a negative error number is returned.
2665  *
2666  * if ins_len > 0, nodes and leaves will be split as we walk down the
2667  * tree.  if ins_len < 0, nodes will be merged as we walk down the tree (if
2668  * possible)
2669  */
2670 int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root
2671                       *root, struct btrfs_key *key, struct btrfs_path *p, int
2672                       ins_len, int cow)
2673 {
2674         struct extent_buffer *b;
2675         int slot;
2676         int ret;
2677         int err;
2678         int level;
2679         int lowest_unlock = 1;
2680         int root_lock;
2681         /* everything at write_lock_level or lower must be write locked */
2682         int write_lock_level = 0;
2683         u8 lowest_level = 0;
2684         int min_write_lock_level;
2685         int prev_cmp;
2686
2687         lowest_level = p->lowest_level;
2688         WARN_ON(lowest_level && ins_len > 0);
2689         WARN_ON(p->nodes[0] != NULL);
2690         BUG_ON(!cow && ins_len);
2691
2692         if (ins_len < 0) {
2693                 lowest_unlock = 2;
2694
2695                 /* when we are removing items, we might have to go up to level
2696                  * two as we update tree pointers  Make sure we keep write
2697                  * for those levels as well
2698                  */
2699                 write_lock_level = 2;
2700         } else if (ins_len > 0) {
2701                 /*
2702                  * for inserting items, make sure we have a write lock on
2703                  * level 1 so we can update keys
2704                  */
2705                 write_lock_level = 1;
2706         }
2707
2708         if (!cow)
2709                 write_lock_level = -1;
2710
2711         if (cow && (p->keep_locks || p->lowest_level))
2712                 write_lock_level = BTRFS_MAX_LEVEL;
2713
2714         min_write_lock_level = write_lock_level;
2715
2716 again:
2717         prev_cmp = -1;
2718         /*
2719          * we try very hard to do read locks on the root
2720          */
2721         root_lock = BTRFS_READ_LOCK;
2722         level = 0;
2723         if (p->search_commit_root) {
2724                 /*
2725                  * the commit roots are read only
2726                  * so we always do read locks
2727                  */
2728                 if (p->need_commit_sem)
2729                         down_read(&root->fs_info->commit_root_sem);
2730                 b = root->commit_root;
2731                 extent_buffer_get(b);
2732                 level = btrfs_header_level(b);
2733                 if (p->need_commit_sem)
2734                         up_read(&root->fs_info->commit_root_sem);
2735                 if (!p->skip_locking)
2736                         btrfs_tree_read_lock(b);
2737         } else {
2738                 if (p->skip_locking) {
2739                         b = btrfs_root_node(root);
2740                         level = btrfs_header_level(b);
2741                 } else {
2742                         /* we don't know the level of the root node
2743                          * until we actually have it read locked
2744                          */
2745                         b = btrfs_read_lock_root_node(root);
2746                         level = btrfs_header_level(b);
2747                         if (level <= write_lock_level) {
2748                                 /* whoops, must trade for write lock */
2749                                 btrfs_tree_read_unlock(b);
2750                                 free_extent_buffer(b);
2751                                 b = btrfs_lock_root_node(root);
2752                                 root_lock = BTRFS_WRITE_LOCK;
2753
2754                                 /* the level might have changed, check again */
2755                                 level = btrfs_header_level(b);
2756                         }
2757                 }
2758         }
2759         p->nodes[level] = b;
2760         if (!p->skip_locking)
2761                 p->locks[level] = root_lock;
2762
2763         while (b) {
2764                 level = btrfs_header_level(b);
2765
2766                 /*
2767                  * setup the path here so we can release it under lock
2768                  * contention with the cow code
2769                  */
2770                 if (cow) {
2771                         /*
2772                          * if we don't really need to cow this block
2773                          * then we don't want to set the path blocking,
2774                          * so we test it here
2775                          */
2776                         if (!should_cow_block(trans, root, b))
2777                                 goto cow_done;
2778
2779                         /*
2780                          * must have write locks on this node and the
2781                          * parent
2782                          */
2783                         if (level > write_lock_level ||
2784                             (level + 1 > write_lock_level &&
2785                             level + 1 < BTRFS_MAX_LEVEL &&
2786                             p->nodes[level + 1])) {
2787                                 write_lock_level = level + 1;
2788                                 btrfs_release_path(p);
2789                                 goto again;
2790                         }
2791
2792                         btrfs_set_path_blocking(p);
2793                         err = btrfs_cow_block(trans, root, b,
2794                                               p->nodes[level + 1],
2795                                               p->slots[level + 1], &b);
2796                         if (err) {
2797                                 ret = err;
2798                                 goto done;
2799                         }
2800                 }
2801 cow_done:
2802                 p->nodes[level] = b;
2803                 btrfs_clear_path_blocking(p, NULL, 0);
2804
2805                 /*
2806                  * we have a lock on b and as long as we aren't changing
2807                  * the tree, there is no way to for the items in b to change.
2808                  * It is safe to drop the lock on our parent before we
2809                  * go through the expensive btree search on b.
2810                  *
2811                  * If we're inserting or deleting (ins_len != 0), then we might
2812                  * be changing slot zero, which may require changing the parent.
2813                  * So, we can't drop the lock until after we know which slot
2814                  * we're operating on.
2815                  */
2816                 if (!ins_len && !p->keep_locks) {
2817                         int u = level + 1;
2818
2819                         if (u < BTRFS_MAX_LEVEL && p->locks[u]) {
2820                                 btrfs_tree_unlock_rw(p->nodes[u], p->locks[u]);
2821                                 p->locks[u] = 0;
2822                         }
2823                 }
2824
2825                 ret = key_search(b, key, level, &prev_cmp, &slot);
2826
2827                 if (level != 0) {
2828                         int dec = 0;
2829                         if (ret && slot > 0) {
2830                                 dec = 1;
2831                                 slot -= 1;
2832                         }
2833                         p->slots[level] = slot;
2834                         err = setup_nodes_for_search(trans, root, p, b, level,
2835                                              ins_len, &write_lock_level);
2836                         if (err == -EAGAIN)
2837                                 goto again;
2838                         if (err) {
2839                                 ret = err;
2840                                 goto done;
2841                         }
2842                         b = p->nodes[level];
2843                         slot = p->slots[level];
2844
2845                         /*
2846                          * slot 0 is special, if we change the key
2847                          * we have to update the parent pointer
2848                          * which means we must have a write lock
2849                          * on the parent
2850                          */
2851                         if (slot == 0 && ins_len &&
2852                             write_lock_level < level + 1) {
2853                                 write_lock_level = level + 1;
2854                                 btrfs_release_path(p);
2855                                 goto again;
2856                         }
2857
2858                         unlock_up(p, level, lowest_unlock,
2859                                   min_write_lock_level, &write_lock_level);
2860
2861                         if (level == lowest_level) {
2862                                 if (dec)
2863                                         p->slots[level]++;
2864                                 goto done;
2865                         }
2866
2867                         err = read_block_for_search(trans, root, p,
2868                                                     &b, level, slot, key, 0);
2869                         if (err == -EAGAIN)
2870                                 goto again;
2871                         if (err) {
2872                                 ret = err;
2873                                 goto done;
2874                         }
2875
2876                         if (!p->skip_locking) {
2877                                 level = btrfs_header_level(b);
2878                                 if (level <= write_lock_level) {
2879                                         err = btrfs_try_tree_write_lock(b);
2880                                         if (!err) {
2881                                                 btrfs_set_path_blocking(p);
2882                                                 btrfs_tree_lock(b);
2883                                                 btrfs_clear_path_blocking(p, b,
2884                                                                   BTRFS_WRITE_LOCK);
2885                                         }
2886                                         p->locks[level] = BTRFS_WRITE_LOCK;
2887                                 } else {
2888                                         err = btrfs_tree_read_lock_atomic(b);
2889                                         if (!err) {
2890                                                 btrfs_set_path_blocking(p);
2891                                                 btrfs_tree_read_lock(b);
2892                                                 btrfs_clear_path_blocking(p, b,
2893                                                                   BTRFS_READ_LOCK);
2894                                         }
2895                                         p->locks[level] = BTRFS_READ_LOCK;
2896                                 }
2897                                 p->nodes[level] = b;
2898                         }
2899                 } else {
2900                         p->slots[level] = slot;
2901                         if (ins_len > 0 &&
2902                             btrfs_leaf_free_space(root, b) < ins_len) {
2903                                 if (write_lock_level < 1) {
2904                                         write_lock_level = 1;
2905                                         btrfs_release_path(p);
2906                                         goto again;
2907                                 }
2908
2909                                 btrfs_set_path_blocking(p);
2910                                 err = split_leaf(trans, root, key,
2911                                                  p, ins_len, ret == 0);
2912                                 btrfs_clear_path_blocking(p, NULL, 0);
2913
2914                                 BUG_ON(err > 0);
2915                                 if (err) {
2916                                         ret = err;
2917                                         goto done;
2918                                 }
2919                         }
2920                         if (!p->search_for_split)
2921                                 unlock_up(p, level, lowest_unlock,
2922                                           min_write_lock_level, &write_lock_level);
2923                         goto done;
2924                 }
2925         }
2926         ret = 1;
2927 done:
2928         /*
2929          * we don't really know what they plan on doing with the path
2930          * from here on, so for now just mark it as blocking
2931          */
2932         if (!p->leave_spinning)
2933                 btrfs_set_path_blocking(p);
2934         if (ret < 0 && !p->skip_release_on_error)
2935                 btrfs_release_path(p);
2936         return ret;
2937 }
2938
2939 /*
2940  * Like btrfs_search_slot, this looks for a key in the given tree. It uses the
2941  * current state of the tree together with the operations recorded in the tree
2942  * modification log to search for the key in a previous version of this tree, as
2943  * denoted by the time_seq parameter.
2944  *
2945  * Naturally, there is no support for insert, delete or cow operations.
2946  *
2947  * The resulting path and return value will be set up as if we called
2948  * btrfs_search_slot at that point in time with ins_len and cow both set to 0.
2949  */
2950 int btrfs_search_old_slot(struct btrfs_root *root, struct btrfs_key *key,
2951                           struct btrfs_path *p, u64 time_seq)
2952 {
2953         struct extent_buffer *b;
2954         int slot;
2955         int ret;
2956         int err;
2957         int level;
2958         int lowest_unlock = 1;
2959         u8 lowest_level = 0;
2960         int prev_cmp = -1;
2961
2962         lowest_level = p->lowest_level;
2963         WARN_ON(p->nodes[0] != NULL);
2964
2965         if (p->search_commit_root) {
2966                 BUG_ON(time_seq);
2967                 return btrfs_search_slot(NULL, root, key, p, 0, 0);
2968         }
2969
2970 again:
2971         b = get_old_root(root, time_seq);
2972         level = btrfs_header_level(b);
2973         p->locks[level] = BTRFS_READ_LOCK;
2974
2975         while (b) {
2976                 level = btrfs_header_level(b);
2977                 p->nodes[level] = b;
2978                 btrfs_clear_path_blocking(p, NULL, 0);
2979
2980                 /*
2981                  * we have a lock on b and as long as we aren't changing
2982                  * the tree, there is no way to for the items in b to change.
2983                  * It is safe to drop the lock on our parent before we
2984                  * go through the expensive btree search on b.
2985                  */
2986                 btrfs_unlock_up_safe(p, level + 1);
2987
2988                 /*
2989                  * Since we can unwind eb's we want to do a real search every
2990                  * time.
2991                  */
2992                 prev_cmp = -1;
2993                 ret = key_search(b, key, level, &prev_cmp, &slot);
2994
2995                 if (level != 0) {
2996                         int dec = 0;
2997                         if (ret && slot > 0) {
2998                                 dec = 1;
2999                                 slot -= 1;
3000                         }
3001                         p->slots[level] = slot;
3002                         unlock_up(p, level, lowest_unlock, 0, NULL);
3003
3004                         if (level == lowest_level) {
3005                                 if (dec)
3006                                         p->slots[level]++;
3007                                 goto done;
3008                         }
3009
3010                         err = read_block_for_search(NULL, root, p, &b, level,
3011                                                     slot, key, time_seq);
3012                         if (err == -EAGAIN)
3013                                 goto again;
3014                         if (err) {
3015                                 ret = err;
3016                                 goto done;
3017                         }
3018
3019                         level = btrfs_header_level(b);
3020                         err = btrfs_tree_read_lock_atomic(b);
3021                         if (!err) {
3022                                 btrfs_set_path_blocking(p);
3023                                 btrfs_tree_read_lock(b);
3024                                 btrfs_clear_path_blocking(p, b,
3025                                                           BTRFS_READ_LOCK);
3026                         }
3027                         b = tree_mod_log_rewind(root->fs_info, p, b, time_seq);
3028                         if (!b) {
3029                                 ret = -ENOMEM;
3030                                 goto done;
3031                         }
3032                         p->locks[level] = BTRFS_READ_LOCK;
3033                         p->nodes[level] = b;
3034                 } else {
3035                         p->slots[level] = slot;
3036                         unlock_up(p, level, lowest_unlock, 0, NULL);
3037                         goto done;
3038                 }
3039         }
3040         ret = 1;
3041 done:
3042         if (!p->leave_spinning)
3043                 btrfs_set_path_blocking(p);
3044         if (ret < 0)
3045                 btrfs_release_path(p);
3046
3047         return ret;
3048 }
3049
3050 /*
3051  * helper to use instead of search slot if no exact match is needed but
3052  * instead the next or previous item should be returned.
3053  * When find_higher is true, the next higher item is returned, the next lower
3054  * otherwise.
3055  * When return_any and find_higher are both true, and no higher item is found,
3056  * return the next lower instead.
3057  * When return_any is true and find_higher is false, and no lower item is found,
3058  * return the next higher instead.
3059  * It returns 0 if any item is found, 1 if none is found (tree empty), and
3060  * < 0 on error
3061  */
3062 int btrfs_search_slot_for_read(struct btrfs_root *root,
3063                                struct btrfs_key *key, struct btrfs_path *p,
3064                                int find_higher, int return_any)
3065 {
3066         int ret;
3067         struct extent_buffer *leaf;
3068
3069 again:
3070         ret = btrfs_search_slot(NULL, root, key, p, 0, 0);
3071         if (ret <= 0)
3072                 return ret;
3073         /*
3074          * a return value of 1 means the path is at the position where the
3075          * item should be inserted. Normally this is the next bigger item,
3076          * but in case the previous item is the last in a leaf, path points
3077          * to the first free slot in the previous leaf, i.e. at an invalid
3078          * item.
3079          */
3080         leaf = p->nodes[0];
3081
3082         if (find_higher) {
3083                 if (p->slots[0] >= btrfs_header_nritems(leaf)) {
3084                         ret = btrfs_next_leaf(root, p);
3085                         if (ret <= 0)
3086                                 return ret;
3087                         if (!return_any)
3088                                 return 1;
3089                         /*
3090                          * no higher item found, return the next
3091                          * lower instead
3092                          */
3093                         return_any = 0;
3094                         find_higher = 0;
3095                         btrfs_release_path(p);
3096                         goto again;
3097                 }
3098         } else {
3099                 if (p->slots[0] == 0) {
3100                         ret = btrfs_prev_leaf(root, p);
3101                         if (ret < 0)
3102                                 return ret;
3103                         if (!ret) {
3104                                 leaf = p->nodes[0];
3105                                 if (p->slots[0] == btrfs_header_nritems(leaf))
3106                                         p->slots[0]--;
3107                                 return 0;
3108                         }
3109                         if (!return_any)
3110                                 return 1;
3111                         /*
3112                          * no lower item found, return the next
3113                          * higher instead
3114                          */
3115                         return_any = 0;
3116                         find_higher = 1;
3117                         btrfs_release_path(p);
3118                         goto again;
3119                 } else {
3120                         --p->slots[0];
3121                 }
3122         }
3123         return 0;
3124 }
3125
3126 /*
3127  * adjust the pointers going up the tree, starting at level
3128  * making sure the right key of each node is points to 'key'.
3129  * This is used after shifting pointers to the left, so it stops
3130  * fixing up pointers when a given leaf/node is not in slot 0 of the
3131  * higher levels
3132  *
3133  */
3134 static void fixup_low_keys(struct btrfs_fs_info *fs_info,
3135                            struct btrfs_path *path,
3136                            struct btrfs_disk_key *key, int level)
3137 {
3138         int i;
3139         struct extent_buffer *t;
3140
3141         for (i = level; i < BTRFS_MAX_LEVEL; i++) {
3142                 int tslot = path->slots[i];
3143                 if (!path->nodes[i])
3144                         break;
3145                 t = path->nodes[i];
3146                 tree_mod_log_set_node_key(fs_info, t, tslot, 1);
3147                 btrfs_set_node_key(t, key, tslot);
3148                 btrfs_mark_buffer_dirty(path->nodes[i]);
3149                 if (tslot != 0)
3150                         break;
3151         }
3152 }
3153
3154 /*
3155  * update item key.
3156  *
3157  * This function isn't completely safe. It's the caller's responsibility
3158  * that the new key won't break the order
3159  */
3160 void btrfs_set_item_key_safe(struct btrfs_fs_info *fs_info,
3161                              struct btrfs_path *path,
3162                              struct btrfs_key *new_key)
3163 {
3164         struct btrfs_disk_key disk_key;
3165         struct extent_buffer *eb;
3166         int slot;
3167
3168         eb = path->nodes[0];
3169         slot = path->slots[0];
3170         if (slot > 0) {
3171                 btrfs_item_key(eb, &disk_key, slot - 1);
3172                 BUG_ON(comp_keys(&disk_key, new_key) >= 0);
3173         }
3174         if (slot < btrfs_header_nritems(eb) - 1) {
3175                 btrfs_item_key(eb, &disk_key, slot + 1);
3176                 BUG_ON(comp_keys(&disk_key, new_key) <= 0);
3177         }
3178
3179         btrfs_cpu_key_to_disk(&disk_key, new_key);
3180         btrfs_set_item_key(eb, &disk_key, slot);
3181         btrfs_mark_buffer_dirty(eb);
3182         if (slot == 0)
3183                 fixup_low_keys(fs_info, path, &disk_key, 1);
3184 }
3185
3186 /*
3187  * try to push data from one node into the next node left in the
3188  * tree.
3189  *
3190  * returns 0 if some ptrs were pushed left, < 0 if there was some horrible
3191  * error, and > 0 if there was no room in the left hand block.
3192  */
3193 static int push_node_left(struct btrfs_trans_handle *trans,
3194                           struct btrfs_root *root, struct extent_buffer *dst,
3195                           struct extent_buffer *src, int empty)
3196 {
3197         int push_items = 0;
3198         int src_nritems;
3199         int dst_nritems;
3200         int ret = 0;
3201
3202         src_nritems = btrfs_header_nritems(src);
3203         dst_nritems = btrfs_header_nritems(dst);
3204         push_items = BTRFS_NODEPTRS_PER_BLOCK(root) - dst_nritems;
3205         WARN_ON(btrfs_header_generation(src) != trans->transid);
3206         WARN_ON(btrfs_header_generation(dst) != trans->transid);
3207
3208         if (!empty && src_nritems <= 8)
3209                 return 1;
3210
3211         if (push_items <= 0)
3212                 return 1;
3213
3214         if (empty) {
3215                 push_items = min(src_nritems, push_items);
3216                 if (push_items < src_nritems) {
3217                         /* leave at least 8 pointers in the node if
3218                          * we aren't going to empty it
3219                          */
3220                         if (src_nritems - push_items < 8) {
3221                                 if (push_items <= 8)
3222                                         return 1;
3223                                 push_items -= 8;
3224                         }
3225                 }
3226         } else
3227                 push_items = min(src_nritems - 8, push_items);
3228
3229         ret = tree_mod_log_eb_copy(root->fs_info, dst, src, dst_nritems, 0,
3230                                    push_items);
3231         if (ret) {
3232                 btrfs_abort_transaction(trans, root, ret);
3233                 return ret;
3234         }
3235         copy_extent_buffer(dst, src,
3236                            btrfs_node_key_ptr_offset(dst_nritems),
3237                            btrfs_node_key_ptr_offset(0),
3238                            push_items * sizeof(struct btrfs_key_ptr));
3239
3240         if (push_items < src_nritems) {
3241                 /*
3242                  * don't call tree_mod_log_eb_move here, key removal was already
3243                  * fully logged by tree_mod_log_eb_copy above.
3244                  */
3245                 memmove_extent_buffer(src, btrfs_node_key_ptr_offset(0),
3246                                       btrfs_node_key_ptr_offset(push_items),
3247                                       (src_nritems - push_items) *
3248                                       sizeof(struct btrfs_key_ptr));
3249         }
3250         btrfs_set_header_nritems(src, src_nritems - push_items);
3251         btrfs_set_header_nritems(dst, dst_nritems + push_items);
3252         btrfs_mark_buffer_dirty(src);
3253         btrfs_mark_buffer_dirty(dst);
3254
3255         return ret;
3256 }
3257
3258 /*
3259  * try to push data from one node into the next node right in the
3260  * tree.
3261  *
3262  * returns 0 if some ptrs were pushed, < 0 if there was some horrible
3263  * error, and > 0 if there was no room in the right hand block.
3264  *
3265  * this will  only push up to 1/2 the contents of the left node over
3266  */
3267 static int balance_node_right(struct btrfs_trans_handle *trans,
3268                               struct btrfs_root *root,
3269                               struct extent_buffer *dst,
3270                               struct extent_buffer *src)
3271 {
3272         int push_items = 0;
3273         int max_push;
3274         int src_nritems;
3275         int dst_nritems;
3276         int ret = 0;
3277
3278         WARN_ON(btrfs_header_generation(src) != trans->transid);
3279         WARN_ON(btrfs_header_generation(dst) != trans->transid);
3280
3281         src_nritems = btrfs_header_nritems(src);
3282         dst_nritems = btrfs_header_nritems(dst);
3283         push_items = BTRFS_NODEPTRS_PER_BLOCK(root) - dst_nritems;
3284         if (push_items <= 0)
3285                 return 1;
3286
3287         if (src_nritems < 4)
3288                 return 1;
3289
3290         max_push = src_nritems / 2 + 1;
3291         /* don't try to empty the node */
3292         if (max_push >= src_nritems)
3293                 return 1;
3294
3295         if (max_push < push_items)
3296                 push_items = max_push;
3297
3298         tree_mod_log_eb_move(root->fs_info, dst, push_items, 0, dst_nritems);
3299         memmove_extent_buffer(dst, btrfs_node_key_ptr_offset(push_items),
3300                                       btrfs_node_key_ptr_offset(0),
3301                                       (dst_nritems) *
3302                                       sizeof(struct btrfs_key_ptr));
3303
3304         ret = tree_mod_log_eb_copy(root->fs_info, dst, src, 0,
3305                                    src_nritems - push_items, push_items);
3306         if (ret) {
3307                 btrfs_abort_transaction(trans, root, ret);
3308                 return ret;
3309         }
3310         copy_extent_buffer(dst, src,
3311                            btrfs_node_key_ptr_offset(0),
3312                            btrfs_node_key_ptr_offset(src_nritems - push_items),
3313                            push_items * sizeof(struct btrfs_key_ptr));
3314
3315         btrfs_set_header_nritems(src, src_nritems - push_items);
3316         btrfs_set_header_nritems(dst, dst_nritems + push_items);
3317
3318         btrfs_mark_buffer_dirty(src);
3319         btrfs_mark_buffer_dirty(dst);
3320
3321         return ret;
3322 }
3323
3324 /*
3325  * helper function to insert a new root level in the tree.
3326  * A new node is allocated, and a single item is inserted to
3327  * point to the existing root
3328  *
3329  * returns zero on success or < 0 on failure.
3330  */
3331 static noinline int insert_new_root(struct btrfs_trans_handle *trans,
3332                            struct btrfs_root *root,
3333                            struct btrfs_path *path, int level)
3334 {
3335         u64 lower_gen;
3336         struct extent_buffer *lower;
3337         struct extent_buffer *c;
3338         struct extent_buffer *old;
3339         struct btrfs_disk_key lower_key;
3340
3341         BUG_ON(path->nodes[level]);
3342         BUG_ON(path->nodes[level-1] != root->node);
3343
3344         lower = path->nodes[level-1];
3345         if (level == 1)
3346                 btrfs_item_key(lower, &lower_key, 0);
3347         else
3348                 btrfs_node_key(lower, &lower_key, 0);
3349
3350         c = btrfs_alloc_tree_block(trans, root, 0, root->root_key.objectid,
3351                                    &lower_key, level, root->node->start, 0);
3352         if (IS_ERR(c))
3353                 return PTR_ERR(c);
3354
3355         root_add_used(root, root->nodesize);
3356
3357         memset_extent_buffer(c, 0, 0, sizeof(struct btrfs_header));
3358         btrfs_set_header_nritems(c, 1);
3359         btrfs_set_header_level(c, level);
3360         btrfs_set_header_bytenr(c, c->start);
3361         btrfs_set_header_generation(c, trans->transid);
3362         btrfs_set_header_backref_rev(c, BTRFS_MIXED_BACKREF_REV);
3363         btrfs_set_header_owner(c, root->root_key.objectid);
3364
3365         write_extent_buffer(c, root->fs_info->fsid, btrfs_header_fsid(),
3366                             BTRFS_FSID_SIZE);
3367
3368         write_extent_buffer(c, root->fs_info->chunk_tree_uuid,
3369                             btrfs_header_chunk_tree_uuid(c), BTRFS_UUID_SIZE);
3370
3371         btrfs_set_node_key(c, &lower_key, 0);
3372         btrfs_set_node_blockptr(c, 0, lower->start);
3373         lower_gen = btrfs_header_generation(lower);
3374         WARN_ON(lower_gen != trans->transid);
3375
3376         btrfs_set_node_ptr_generation(c, 0, lower_gen);
3377
3378         btrfs_mark_buffer_dirty(c);
3379
3380         old = root->node;
3381         tree_mod_log_set_root_pointer(root, c, 0);
3382         rcu_assign_pointer(root->node, c);
3383
3384         /* the super has an extra ref to root->node */
3385         free_extent_buffer(old);
3386
3387         add_root_to_dirty_list(root);
3388         extent_buffer_get(c);
3389         path->nodes[level] = c;
3390         path->locks[level] = BTRFS_WRITE_LOCK_BLOCKING;
3391         path->slots[level] = 0;
3392         return 0;
3393 }
3394
3395 /*
3396  * worker function to insert a single pointer in a node.
3397  * the node should have enough room for the pointer already
3398  *
3399  * slot and level indicate where you want the key to go, and
3400  * blocknr is the block the key points to.
3401  */
3402 static void insert_ptr(struct btrfs_trans_handle *trans,
3403                        struct btrfs_root *root, struct btrfs_path *path,
3404                        struct btrfs_disk_key *key, u64 bytenr,
3405                        int slot, int level)
3406 {
3407         struct extent_buffer *lower;
3408         int nritems;
3409         int ret;
3410
3411         BUG_ON(!path->nodes[level]);
3412         btrfs_assert_tree_locked(path->nodes[level]);
3413         lower = path->nodes[level];
3414         nritems = btrfs_header_nritems(lower);
3415         BUG_ON(slot > nritems);
3416         BUG_ON(nritems == BTRFS_NODEPTRS_PER_BLOCK(root));
3417         if (slot != nritems) {
3418                 if (level)
3419                         tree_mod_log_eb_move(root->fs_info, lower, slot + 1,
3420                                              slot, nritems - slot);
3421                 memmove_extent_buffer(lower,
3422                               btrfs_node_key_ptr_offset(slot + 1),
3423                               btrfs_node_key_ptr_offset(slot),
3424                               (nritems - slot) * sizeof(struct btrfs_key_ptr));
3425         }
3426         if (level) {
3427                 ret = tree_mod_log_insert_key(root->fs_info, lower, slot,
3428                                               MOD_LOG_KEY_ADD, GFP_NOFS);
3429                 BUG_ON(ret < 0);
3430         }
3431         btrfs_set_node_key(lower, key, slot);
3432         btrfs_set_node_blockptr(lower, slot, bytenr);
3433         WARN_ON(trans->transid == 0);
3434         btrfs_set_node_ptr_generation(lower, slot, trans->transid);
3435         btrfs_set_header_nritems(lower, nritems + 1);
3436         btrfs_mark_buffer_dirty(lower);
3437 }
3438
3439 /*
3440  * split the node at the specified level in path in two.
3441  * The path is corrected to point to the appropriate node after the split
3442  *
3443  * Before splitting this tries to make some room in the node by pushing
3444  * left and right, if either one works, it returns right away.
3445  *
3446  * returns 0 on success and < 0 on failure
3447  */
3448 static noinline int split_node(struct btrfs_trans_handle *trans,
3449                                struct btrfs_root *root,
3450                                struct btrfs_path *path, int level)
3451 {
3452         struct extent_buffer *c;
3453         struct extent_buffer *split;
3454         struct btrfs_disk_key disk_key;
3455         int mid;
3456         int ret;
3457         u32 c_nritems;
3458
3459         c = path->nodes[level];
3460         WARN_ON(btrfs_header_generation(c) != trans->transid);
3461         if (c == root->node) {
3462                 /*
3463                  * trying to split the root, lets make a new one
3464                  *
3465                  * tree mod log: We don't log_removal old root in
3466                  * insert_new_root, because that root buffer will be kept as a
3467                  * normal node. We are going to log removal of half of the
3468                  * elements below with tree_mod_log_eb_copy. We're holding a
3469                  * tree lock on the buffer, which is why we cannot race with
3470                  * other tree_mod_log users.
3471                  */
3472                 ret = insert_new_root(trans, root, path, level + 1);
3473                 if (ret)
3474                         return ret;
3475         } else {
3476                 ret = push_nodes_for_insert(trans, root, path, level);
3477                 c = path->nodes[level];
3478                 if (!ret && btrfs_header_nritems(c) <
3479                     BTRFS_NODEPTRS_PER_BLOCK(root) - 3)
3480                         return 0;
3481                 if (ret < 0)
3482                         return ret;
3483         }
3484
3485         c_nritems = btrfs_header_nritems(c);
3486         mid = (c_nritems + 1) / 2;
3487         btrfs_node_key(c, &disk_key, mid);
3488
3489         split = btrfs_alloc_tree_block(trans, root, 0, root->root_key.objectid,
3490                         &disk_key, level, c->start, 0);
3491         if (IS_ERR(split))
3492                 return PTR_ERR(split);
3493
3494         root_add_used(root, root->nodesize);
3495
3496         memset_extent_buffer(split, 0, 0, sizeof(struct btrfs_header));
3497         btrfs_set_header_level(split, btrfs_header_level(c));
3498         btrfs_set_header_bytenr(split, split->start);
3499         btrfs_set_header_generation(split, trans->transid);
3500         btrfs_set_header_backref_rev(split, BTRFS_MIXED_BACKREF_REV);
3501         btrfs_set_header_owner(split, root->root_key.objectid);
3502         write_extent_buffer(split, root->fs_info->fsid,
3503                             btrfs_header_fsid(), BTRFS_FSID_SIZE);
3504         write_extent_buffer(split, root->fs_info->chunk_tree_uuid,
3505                             btrfs_header_chunk_tree_uuid(split),
3506                             BTRFS_UUID_SIZE);
3507
3508         ret = tree_mod_log_eb_copy(root->fs_info, split, c, 0,
3509                                    mid, c_nritems - mid);
3510         if (ret) {
3511                 btrfs_abort_transaction(trans, root, ret);
3512                 return ret;
3513         }
3514         copy_extent_buffer(split, c,
3515                            btrfs_node_key_ptr_offset(0),
3516                            btrfs_node_key_ptr_offset(mid),
3517                            (c_nritems - mid) * sizeof(struct btrfs_key_ptr));
3518         btrfs_set_header_nritems(split, c_nritems - mid);
3519         btrfs_set_header_nritems(c, mid);
3520         ret = 0;
3521
3522         btrfs_mark_buffer_dirty(c);
3523         btrfs_mark_buffer_dirty(split);
3524
3525         insert_ptr(trans, root, path, &disk_key, split->start,
3526                    path->slots[level + 1] + 1, level + 1);
3527
3528         if (path->slots[level] >= mid) {
3529                 path->slots[level] -= mid;
3530                 btrfs_tree_unlock(c);
3531                 free_extent_buffer(c);
3532                 path->nodes[level] = split;
3533                 path->slots[level + 1] += 1;
3534         } else {
3535                 btrfs_tree_unlock(split);
3536                 free_extent_buffer(split);
3537         }
3538         return ret;
3539 }
3540
3541 /*
3542  * how many bytes are required to store the items in a leaf.  start
3543  * and nr indicate which items in the leaf to check.  This totals up the
3544  * space used both by the item structs and the item data
3545  */
3546 static int leaf_space_used(struct extent_buffer *l, int start, int nr)
3547 {
3548         struct btrfs_item *start_item;
3549         struct btrfs_item *end_item;
3550         struct btrfs_map_token token;
3551         int data_len;
3552         int nritems = btrfs_header_nritems(l);
3553         int end = min(nritems, start + nr) - 1;
3554
3555         if (!nr)
3556                 return 0;
3557         btrfs_init_map_token(&token);
3558         start_item = btrfs_item_nr(start);
3559         end_item = btrfs_item_nr(end);
3560         data_len = btrfs_token_item_offset(l, start_item, &token) +
3561                 btrfs_token_item_size(l, start_item, &token);
3562         data_len = data_len - btrfs_token_item_offset(l, end_item, &token);
3563         data_len += sizeof(struct btrfs_item) * nr;
3564         WARN_ON(data_len < 0);
3565         return data_len;
3566 }
3567
3568 /*
3569  * The space between the end of the leaf items and
3570  * the start of the leaf data.  IOW, how much room
3571  * the leaf has left for both items and data
3572  */
3573 noinline int btrfs_leaf_free_space(struct btrfs_root *root,
3574                                    struct extent_buffer *leaf)
3575 {
3576         int nritems = btrfs_header_nritems(leaf);
3577         int ret;
3578         ret = BTRFS_LEAF_DATA_SIZE(root) - leaf_space_used(leaf, 0, nritems);
3579         if (ret < 0) {
3580                 btrfs_crit(root->fs_info,
3581                         "leaf free space ret %d, leaf data size %lu, used %d nritems %d",
3582                        ret, (unsigned long) BTRFS_LEAF_DATA_SIZE(root),
3583                        leaf_space_used(leaf, 0, nritems), nritems);
3584         }
3585         return ret;
3586 }
3587
3588 /*
3589  * min slot controls the lowest index we're willing to push to the
3590  * right.  We'll push up to and including min_slot, but no lower
3591  */
3592 static noinline int __push_leaf_right(struct btrfs_trans_handle *trans,
3593                                       struct btrfs_root *root,
3594                                       struct btrfs_path *path,
3595                                       int data_size, int empty,
3596                                       struct extent_buffer *right,
3597                                       int free_space, u32 left_nritems,
3598                                       u32 min_slot)
3599 {
3600         struct extent_buffer *left = path->nodes[0];
3601         struct extent_buffer *upper = path->nodes[1];
3602         struct btrfs_map_token token;
3603         struct btrfs_disk_key disk_key;
3604         int slot;
3605         u32 i;
3606         int push_space = 0;
3607         int push_items = 0;
3608         struct btrfs_item *item;
3609         u32 nr;
3610         u32 right_nritems;
3611         u32 data_end;
3612         u32 this_item_size;
3613
3614         btrfs_init_map_token(&token);
3615
3616         if (empty)
3617                 nr = 0;
3618         else
3619                 nr = max_t(u32, 1, min_slot);
3620
3621         if (path->slots[0] >= left_nritems)
3622                 push_space += data_size;
3623
3624         slot = path->slots[1];
3625         i = left_nritems - 1;
3626         while (i >= nr) {
3627                 item = btrfs_item_nr(i);
3628
3629                 if (!empty && push_items > 0) {
3630                         if (path->slots[0] > i)
3631                                 break;
3632                         if (path->slots[0] == i) {
3633                                 int space = btrfs_leaf_free_space(root, left);
3634                                 if (space + push_space * 2 > free_space)
3635                                         break;
3636                         }
3637                 }
3638
3639                 if (path->slots[0] == i)
3640                         push_space += data_size;
3641
3642                 this_item_size = btrfs_item_size(left, item);
3643                 if (this_item_size + sizeof(*item) + push_space > free_space)
3644                         break;
3645
3646                 push_items++;
3647                 push_space += this_item_size + sizeof(*item);
3648                 if (i == 0)
3649                         break;
3650                 i--;
3651         }
3652
3653         if (push_items == 0)
3654                 goto out_unlock;
3655
3656         WARN_ON(!empty && push_items == left_nritems);
3657
3658         /* push left to right */
3659         right_nritems = btrfs_header_nritems(right);
3660
3661         push_space = btrfs_item_end_nr(left, left_nritems - push_items);
3662         push_space -= leaf_data_end(root, left);
3663
3664         /* make room in the right data area */
3665         data_end = leaf_data_end(root, right);
3666         memmove_extent_buffer(right,
3667                               btrfs_leaf_data(right) + data_end - push_space,
3668                               btrfs_leaf_data(right) + data_end,
3669                               BTRFS_LEAF_DATA_SIZE(root) - data_end);
3670
3671         /* copy from the left data area */
3672         copy_extent_buffer(right, left, btrfs_leaf_data(right) +
3673                      BTRFS_LEAF_DATA_SIZE(root) - push_space,
3674                      btrfs_leaf_data(left) + leaf_data_end(root, left),
3675                      push_space);
3676
3677         memmove_extent_buffer(right, btrfs_item_nr_offset(push_items),
3678                               btrfs_item_nr_offset(0),
3679                               right_nritems * sizeof(struct btrfs_item));
3680
3681         /* copy the items from left to right */
3682         copy_extent_buffer(right, left, btrfs_item_nr_offset(0),
3683                    btrfs_item_nr_offset(left_nritems - push_items),
3684                    push_items * sizeof(struct btrfs_item));
3685
3686         /* update the item pointers */
3687         right_nritems += push_items;
3688         btrfs_set_header_nritems(right, right_nritems);
3689         push_space = BTRFS_LEAF_DATA_SIZE(root);
3690         for (i = 0; i < right_nritems; i++) {
3691                 item = btrfs_item_nr(i);
3692                 push_space -= btrfs_token_item_size(right, item, &token);
3693                 btrfs_set_token_item_offset(right, item, push_space, &token);
3694         }
3695
3696         left_nritems -= push_items;
3697         btrfs_set_header_nritems(left, left_nritems);
3698
3699         if (left_nritems)
3700                 btrfs_mark_buffer_dirty(left);
3701         else
3702                 clean_tree_block(trans, root->fs_info, left);
3703
3704         btrfs_mark_buffer_dirty(right);
3705
3706         btrfs_item_key(right, &disk_key, 0);
3707         btrfs_set_node_key(upper, &disk_key, slot + 1);
3708         btrfs_mark_buffer_dirty(upper);
3709
3710         /* then fixup the leaf pointer in the path */
3711         if (path->slots[0] >= left_nritems) {
3712                 path->slots[0] -= left_nritems;
3713                 if (btrfs_header_nritems(path->nodes[0]) == 0)
3714                         clean_tree_block(trans, root->fs_info, path->nodes[0]);
3715                 btrfs_tree_unlock(path->nodes[0]);
3716                 free_extent_buffer(path->nodes[0]);
3717                 path->nodes[0] = right;
3718                 path->slots[1] += 1;
3719         } else {
3720                 btrfs_tree_unlock(right);
3721                 free_extent_buffer(right);
3722         }
3723         return 0;
3724
3725 out_unlock:
3726         btrfs_tree_unlock(right);
3727         free_extent_buffer(right);
3728         return 1;
3729 }
3730
3731 /*
3732  * push some data in the path leaf to the right, trying to free up at
3733  * least data_size bytes.  returns zero if the push worked, nonzero otherwise
3734  *
3735  * returns 1 if the push failed because the other node didn't have enough
3736  * room, 0 if everything worked out and < 0 if there were major errors.
3737  *
3738  * this will push starting from min_slot to the end of the leaf.  It won't
3739  * push any slot lower than min_slot
3740  */
3741 static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root
3742                            *root, struct btrfs_path *path,
3743                            int min_data_size, int data_size,
3744                            int empty, u32 min_slot)
3745 {
3746         struct extent_buffer *left = path->nodes[0];
3747         struct extent_buffer *right;
3748         struct extent_buffer *upper;
3749         int slot;
3750         int free_space;
3751         u32 left_nritems;
3752         int ret;
3753
3754         if (!path->nodes[1])
3755                 return 1;
3756
3757         slot = path->slots[1];
3758         upper = path->nodes[1];
3759         if (slot >= btrfs_header_nritems(upper) - 1)
3760                 return 1;
3761
3762         btrfs_assert_tree_locked(path->nodes[1]);
3763
3764         right = read_node_slot(root, upper, slot + 1);
3765         if (right == NULL)
3766                 return 1;
3767
3768         btrfs_tree_lock(right);
3769         btrfs_set_lock_blocking(right);
3770
3771         free_space = btrfs_leaf_free_space(root, right);
3772         if (free_space < data_size)
3773                 goto out_unlock;
3774
3775         /* cow and double check */
3776         ret = btrfs_cow_block(trans, root, right, upper,
3777                               slot + 1, &right);
3778         if (ret)
3779                 goto out_unlock;
3780
3781         free_space = btrfs_leaf_free_space(root, right);
3782         if (free_space < data_size)
3783                 goto out_unlock;
3784
3785         left_nritems = btrfs_header_nritems(left);
3786         if (left_nritems == 0)
3787                 goto out_unlock;
3788
3789         if (path->slots[0] == left_nritems && !empty) {
3790                 /* Key greater than all keys in the leaf, right neighbor has
3791                  * enough room for it and we're not emptying our leaf to delete
3792                  * it, therefore use right neighbor to insert the new item and
3793                  * no need to touch/dirty our left leaft. */
3794                 btrfs_tree_unlock(left);
3795                 free_extent_buffer(left);
3796                 path->nodes[0] = right;
3797                 path->slots[0] = 0;
3798                 path->slots[1]++;
3799                 return 0;
3800         }
3801
3802         return __push_leaf_right(trans, root, path, min_data_size, empty,
3803                                 right, free_space, left_nritems, min_slot);
3804 out_unlock:
3805         btrfs_tree_unlock(right);
3806         free_extent_buffer(right);
3807         return 1;
3808 }
3809
3810 /*
3811  * push some data in the path leaf to the left, trying to free up at
3812  * least data_size bytes.  returns zero if the push worked, nonzero otherwise
3813  *
3814  * max_slot can put a limit on how far into the leaf we'll push items.  The
3815  * item at 'max_slot' won't be touched.  Use (u32)-1 to make us do all the
3816  * items
3817  */
3818 static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
3819                                      struct btrfs_root *root,
3820                                      struct btrfs_path *path, int data_size,
3821                                      int empty, struct extent_buffer *left,
3822                                      int free_space, u32 right_nritems,
3823                                      u32 max_slot)
3824 {
3825         struct btrfs_disk_key disk_key;
3826         struct extent_buffer *right = path->nodes[0];
3827         int i;
3828         int push_space = 0;
3829         int push_items = 0;
3830         struct btrfs_item *item;
3831         u32 old_left_nritems;
3832         u32 nr;
3833         int ret = 0;
3834         u32 this_item_size;
3835         u32 old_left_item_size;
3836         struct btrfs_map_token token;
3837
3838         btrfs_init_map_token(&token);
3839
3840         if (empty)
3841                 nr = min(right_nritems, max_slot);
3842         else
3843                 nr = min(right_nritems - 1, max_slot);
3844
3845         for (i = 0; i < nr; i++) {
3846                 item = btrfs_item_nr(i);
3847
3848                 if (!empty && push_items > 0) {
3849                         if (path->slots[0] < i)
3850                                 break;
3851                         if (path->slots[0] == i) {
3852                                 int space = btrfs_leaf_free_space(root, right);
3853                                 if (space + push_space * 2 > free_space)
3854                                         break;
3855                         }
3856                 }
3857
3858                 if (path->slots[0] == i)
3859                         push_space += data_size;
3860
3861                 this_item_size = btrfs_item_size(right, item);
3862                 if (this_item_size + sizeof(*item) + push_space > free_space)
3863                         break;
3864
3865                 push_items++;
3866                 push_space += this_item_size + sizeof(*item);
3867         }
3868
3869         if (push_items == 0) {
3870                 ret = 1;
3871                 goto out;
3872         }
3873         WARN_ON(!empty && push_items == btrfs_header_nritems(right));
3874
3875         /* push data from right to left */
3876         copy_extent_buffer(left, right,
3877                            btrfs_item_nr_offset(btrfs_header_nritems(left)),
3878                            btrfs_item_nr_offset(0),
3879                            push_items * sizeof(struct btrfs_item));
3880
3881         push_space = BTRFS_LEAF_DATA_SIZE(root) -
3882                      btrfs_item_offset_nr(right, push_items - 1);
3883
3884         copy_extent_buffer(left, right, btrfs_leaf_data(left) +
3885                      leaf_data_end(root, left) - push_space,
3886                      btrfs_leaf_data(right) +
3887                      btrfs_item_offset_nr(right, push_items - 1),
3888                      push_space);
3889         old_left_nritems = btrfs_header_nritems(left);
3890         BUG_ON(old_left_nritems <= 0);
3891
3892         old_left_item_size = btrfs_item_offset_nr(left, old_left_nritems - 1);
3893         for (i = old_left_nritems; i < old_left_nritems + push_items; i++) {
3894                 u32 ioff;
3895
3896                 item = btrfs_item_nr(i);
3897
3898                 ioff = btrfs_token_item_offset(left, item, &token);
3899                 btrfs_set_token_item_offset(left, item,
3900                       ioff - (BTRFS_LEAF_DATA_SIZE(root) - old_left_item_size),
3901                       &token);
3902         }
3903         btrfs_set_header_nritems(left, old_left_nritems + push_items);
3904
3905         /* fixup right node */
3906         if (push_items > right_nritems)
3907                 WARN(1, KERN_CRIT "push items %d nr %u\n", push_items,
3908                        right_nritems);
3909
3910         if (push_items < right_nritems) {
3911                 push_space = btrfs_item_offset_nr(right, push_items - 1) -
3912                                                   leaf_data_end(root, right);
3913                 memmove_extent_buffer(right, btrfs_leaf_data(right) +
3914                                       BTRFS_LEAF_DATA_SIZE(root) - push_space,
3915                                       btrfs_leaf_data(right) +
3916                                       leaf_data_end(root, right), push_space);
3917
3918                 memmove_extent_buffer(right, btrfs_item_nr_offset(0),
3919                               btrfs_item_nr_offset(push_items),
3920                              (btrfs_header_nritems(right) - push_items) *
3921                              sizeof(struct btrfs_item));
3922         }
3923         right_nritems -= push_items;
3924         btrfs_set_header_nritems(right, right_nritems);
3925         push_space = BTRFS_LEAF_DATA_SIZE(root);
3926         for (i = 0; i < right_nritems; i++) {
3927                 item = btrfs_item_nr(i);
3928
3929                 push_space = push_space - btrfs_token_item_size(right,
3930                                                                 item, &token);
3931                 btrfs_set_token_item_offset(right, item, push_space, &token);
3932         }
3933
3934         btrfs_mark_buffer_dirty(left);
3935         if (right_nritems)
3936                 btrfs_mark_buffer_dirty(right);
3937         else
3938                 clean_tree_block(trans, root->fs_info, right);
3939
3940         btrfs_item_key(right, &disk_key, 0);
3941         fixup_low_keys(root->fs_info, path, &disk_key, 1);
3942
3943         /* then fixup the leaf pointer in the path */
3944         if (path->slots[0] < push_items) {
3945                 path->slots[0] += old_left_nritems;
3946                 btrfs_tree_unlock(path->nodes[0]);
3947                 free_extent_buffer(path->nodes[0]);
3948                 path->nodes[0] = left;
3949                 path->slots[1] -= 1;
3950         } else {
3951                 btrfs_tree_unlock(left);
3952                 free_extent_buffer(left);
3953                 path->slots[0] -= push_items;
3954         }
3955         BUG_ON(path->slots[0] < 0);
3956         return ret;
3957 out:
3958         btrfs_tree_unlock(left);
3959         free_extent_buffer(left);
3960         return ret;
3961 }
3962
3963 /*
3964  * push some data in the path leaf to the left, trying to free up at
3965  * least data_size bytes.  returns zero if the push worked, nonzero otherwise
3966  *
3967  * max_slot can put a limit on how far into the leaf we'll push items.  The
3968  * item at 'max_slot' won't be touched.  Use (u32)-1 to make us push all the
3969  * items
3970  */
3971 static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root
3972                           *root, struct btrfs_path *path, int min_data_size,
3973                           int data_size, int empty, u32 max_slot)
3974 {
3975         struct extent_buffer *right = path->nodes[0];
3976         struct extent_buffer *left;
3977         int slot;
3978         int free_space;
3979         u32 right_nritems;
3980         int ret = 0;
3981
3982         slot = path->slots[1];
3983         if (slot == 0)
3984                 return 1;
3985         if (!path->nodes[1])
3986                 return 1;
3987
3988         right_nritems = btrfs_header_nritems(right);
3989         if (right_nritems == 0)
3990                 return 1;
3991
3992         btrfs_assert_tree_locked(path->nodes[1]);
3993
3994         left = read_node_slot(root, path->nodes[1], slot - 1);
3995         if (left == NULL)
3996                 return 1;
3997
3998         btrfs_tree_lock(left);
3999         btrfs_set_lock_blocking(left);
4000
4001         free_space = btrfs_leaf_free_space(root, left);
4002         if (free_space < data_size) {
4003                 ret = 1;
4004                 goto out;
4005         }
4006
4007         /* cow and double check */
4008         ret = btrfs_cow_block(trans, root, left,
4009                               path->nodes[1], slot - 1, &left);
4010         if (ret) {
4011                 /* we hit -ENOSPC, but it isn't fatal here */
4012                 if (ret == -ENOSPC)
4013                         ret = 1;
4014                 goto out;
4015         }
4016
4017         free_space = btrfs_leaf_free_space(root, left);
4018         if (free_space < data_size) {
4019                 ret = 1;
4020                 goto out;
4021         }
4022
4023         return __push_leaf_left(trans, root, path, min_data_size,
4024                                empty, left, free_space, right_nritems,
4025                                max_slot);
4026 out:
4027         btrfs_tree_unlock(left);
4028         free_extent_buffer(left);
4029         return ret;
4030 }
4031
4032 /*
4033  * split the path's leaf in two, making sure there is at least data_size
4034  * available for the resulting leaf level of the path.
4035  */
4036 static noinline void copy_for_split(struct btrfs_trans_handle *trans,
4037                                     struct btrfs_root *root,
4038                                     struct btrfs_path *path,
4039                                     struct extent_buffer *l,
4040                                     struct extent_buffer *right,
4041                                     int slot, int mid, int nritems)
4042 {
4043         int data_copy_size;
4044         int rt_data_off;
4045         int i;
4046         struct btrfs_disk_key disk_key;
4047         struct btrfs_map_token token;
4048
4049         btrfs_init_map_token(&token);
4050
4051         nritems = nritems - mid;
4052         btrfs_set_header_nritems(right, nritems);
4053         data_copy_size = btrfs_item_end_nr(l, mid) - leaf_data_end(root, l);
4054
4055         copy_extent_buffer(right, l, btrfs_item_nr_offset(0),
4056                            btrfs_item_nr_offset(mid),
4057                            nritems * sizeof(struct btrfs_item));
4058
4059         copy_extent_buffer(right, l,
4060                      btrfs_leaf_data(right) + BTRFS_LEAF_DATA_SIZE(root) -
4061                      data_copy_size, btrfs_leaf_data(l) +
4062                      leaf_data_end(root, l), data_copy_size);
4063
4064         rt_data_off = BTRFS_LEAF_DATA_SIZE(root) -
4065                       btrfs_item_end_nr(l, mid);
4066
4067         for (i = 0; i < nritems; i++) {
4068                 struct btrfs_item *item = btrfs_item_nr(i);
4069                 u32 ioff;
4070
4071                 ioff = btrfs_token_item_offset(right, item, &token);
4072                 btrfs_set_token_item_offset(right, item,
4073                                             ioff + rt_data_off, &token);
4074         }
4075
4076         btrfs_set_header_nritems(l, mid);
4077         btrfs_item_key(right, &disk_key, 0);
4078         insert_ptr(trans, root, path, &disk_key, right->start,
4079                    path->slots[1] + 1, 1);
4080
4081         btrfs_mark_buffer_dirty(right);
4082         btrfs_mark_buffer_dirty(l);
4083         BUG_ON(path->slots[0] != slot);
4084
4085         if (mid <= slot) {
4086                 btrfs_tree_unlock(path->nodes[0]);
4087                 free_extent_buffer(path->nodes[0]);
4088                 path->nodes[0] = right;
4089                 path->slots[0] -= mid;
4090                 path->slots[1] += 1;
4091         } else {
4092                 btrfs_tree_unlock(right);
4093                 free_extent_buffer(right);
4094         }
4095
4096         BUG_ON(path->slots[0] < 0);
4097 }
4098
4099 /*
4100  * double splits happen when we need to insert a big item in the middle
4101  * of a leaf.  A double split can leave us with 3 mostly empty leaves:
4102  * leaf: [ slots 0 - N] [ our target ] [ N + 1 - total in leaf ]
4103  *          A                 B                 C
4104  *
4105  * We avoid this by trying to push the items on either side of our target
4106  * into the adjacent leaves.  If all goes well we can avoid the double split
4107  * completely.
4108  */
4109 static noinline int push_for_double_split(struct btrfs_trans_handle *trans,
4110                                           struct btrfs_root *root,
4111                                           struct btrfs_path *path,
4112                                           int data_size)
4113 {
4114         int ret;
4115         int progress = 0;
4116         int slot;
4117         u32 nritems;
4118         int space_needed = data_size;
4119
4120         slot = path->slots[0];
4121         if (slot < btrfs_header_nritems(path->nodes[0]))
4122                 space_needed -= btrfs_leaf_free_space(root, path->nodes[0]);
4123
4124         /*
4125          * try to push all the items after our slot into the
4126          * right leaf
4127          */
4128         ret = push_leaf_right(trans, root, path, 1, space_needed, 0, slot);
4129         if (ret < 0)
4130                 return ret;
4131
4132         if (ret == 0)
4133                 progress++;
4134
4135         nritems = btrfs_header_nritems(path->nodes[0]);
4136         /*
4137          * our goal is to get our slot at the start or end of a leaf.  If
4138          * we've done so we're done
4139          */
4140         if (path->slots[0] == 0 || path->slots[0] == nritems)
4141                 return 0;
4142
4143         if (btrfs_leaf_free_space(root, path->nodes[0]) >= data_size)
4144                 return 0;
4145
4146         /* try to push all the items before our slot into the next leaf */
4147         slot = path->slots[0];
4148         ret = push_leaf_left(trans, root, path, 1, space_needed, 0, slot);
4149         if (ret < 0)
4150                 return ret;
4151
4152         if (ret == 0)
4153                 progress++;
4154
4155         if (progress)
4156                 return 0;
4157         return 1;
4158 }
4159
4160 /*
4161  * split the path's leaf in two, making sure there is at least data_size
4162  * available for the resulting leaf level of the path.
4163  *
4164  * returns 0 if all went well and < 0 on failure.
4165  */
4166 static noinline int split_leaf(struct btrfs_trans_handle *trans,
4167                                struct btrfs_root *root,
4168                                struct btrfs_key *ins_key,
4169                                struct btrfs_path *path, int data_size,
4170                                int extend)
4171 {
4172         struct btrfs_disk_key disk_key;
4173         struct extent_buffer *l;
4174         u32 nritems;
4175         int mid;
4176         int slot;
4177         struct extent_buffer *right;
4178         struct btrfs_fs_info *fs_info = root->fs_info;
4179         int ret = 0;
4180         int wret;
4181         int split;
4182         int num_doubles = 0;
4183         int tried_avoid_double = 0;
4184
4185         l = path->nodes[0];
4186         slot = path->slots[0];
4187         if (extend && data_size + btrfs_item_size_nr(l, slot) +
4188             sizeof(struct btrfs_item) > BTRFS_LEAF_DATA_SIZE(root))
4189                 return -EOVERFLOW;
4190
4191         /* first try to make some room by pushing left and right */
4192         if (data_size && path->nodes[1]) {
4193                 int space_needed = data_size;
4194
4195                 if (slot < btrfs_header_nritems(l))
4196                         space_needed -= btrfs_leaf_free_space(root, l);
4197
4198                 wret = push_leaf_right(trans, root, path, space_needed,
4199                                        space_needed, 0, 0);
4200                 if (wret < 0)
4201                         return wret;
4202                 if (wret) {
4203                         wret = push_leaf_left(trans, root, path, space_needed,
4204                                               space_needed, 0, (u32)-1);
4205                         if (wret < 0)
4206                                 return wret;
4207                 }
4208                 l = path->nodes[0];
4209
4210                 /* did the pushes work? */
4211                 if (btrfs_leaf_free_space(root, l) >= data_size)
4212                         return 0;
4213         }
4214
4215         if (!path->nodes[1]) {
4216                 ret = insert_new_root(trans, root, path, 1);
4217                 if (ret)
4218                         return ret;
4219         }
4220 again:
4221         split = 1;
4222         l = path->nodes[0];
4223         slot = path->slots[0];
4224         nritems = btrfs_header_nritems(l);
4225         mid = (nritems + 1) / 2;
4226
4227         if (mid <= slot) {
4228                 if (nritems == 1 ||
4229                     leaf_space_used(l, mid, nritems - mid) + data_size >
4230                         BTRFS_LEAF_DATA_SIZE(root)) {
4231                         if (slot >= nritems) {
4232                                 split = 0;
4233                         } else {
4234                                 mid = slot;
4235                                 if (mid != nritems &&
4236                                     leaf_space_used(l, mid, nritems - mid) +
4237                                     data_size > BTRFS_LEAF_DATA_SIZE(root)) {
4238                                         if (data_size && !tried_avoid_double)
4239                                                 goto push_for_double;
4240                                         split = 2;
4241                                 }
4242                         }
4243                 }
4244         } else {
4245                 if (leaf_space_used(l, 0, mid) + data_size >
4246                         BTRFS_LEAF_DATA_SIZE(root)) {
4247                         if (!extend && data_size && slot == 0) {
4248                                 split = 0;
4249                         } else if ((extend || !data_size) && slot == 0) {
4250                                 mid = 1;
4251                         } else {
4252                                 mid = slot;
4253                                 if (mid != nritems &&
4254                                     leaf_space_used(l, mid, nritems - mid) +
4255                                     data_size > BTRFS_LEAF_DATA_SIZE(root)) {
4256                                         if (data_size && !tried_avoid_double)
4257                                                 goto push_for_double;
4258                                         split = 2;
4259                                 }
4260                         }
4261                 }
4262         }
4263
4264         if (split == 0)
4265                 btrfs_cpu_key_to_disk(&disk_key, ins_key);
4266         else
4267                 btrfs_item_key(l, &disk_key, mid);
4268
4269         right = btrfs_alloc_tree_block(trans, root, 0, root->root_key.objectid,
4270                         &disk_key, 0, l->start, 0);
4271         if (IS_ERR(right))
4272                 return PTR_ERR(right);
4273
4274         root_add_used(root, root->nodesize);
4275
4276         memset_extent_buffer(right, 0, 0, sizeof(struct btrfs_header));
4277         btrfs_set_header_bytenr(right, right->start);
4278         btrfs_set_header_generation(right, trans->transid);
4279         btrfs_set_header_backref_rev(right, BTRFS_MIXED_BACKREF_REV);
4280         btrfs_set_header_owner(right, root->root_key.objectid);
4281         btrfs_set_header_level(right, 0);
4282         write_extent_buffer(right, fs_info->fsid,
4283                             btrfs_header_fsid(), BTRFS_FSID_SIZE);
4284
4285         write_extent_buffer(right, fs_info->chunk_tree_uuid,
4286                             btrfs_header_chunk_tree_uuid(right),
4287                             BTRFS_UUID_SIZE);
4288
4289         if (split == 0) {
4290                 if (mid <= slot) {
4291                         btrfs_set_header_nritems(right, 0);
4292                         insert_ptr(trans, root, path, &disk_key, right->start,
4293                                    path->slots[1] + 1, 1);
4294                         btrfs_tree_unlock(path->nodes[0]);
4295                         free_extent_buffer(path->nodes[0]);
4296                         path->nodes[0] = right;
4297                         path->slots[0] = 0;
4298                         path->slots[1] += 1;
4299                 } else {
4300                         btrfs_set_header_nritems(right, 0);
4301                         insert_ptr(trans, root, path, &disk_key, right->start,
4302                                           path->slots[1], 1);
4303                         btrfs_tree_unlock(path->nodes[0]);
4304                         free_extent_buffer(path->nodes[0]);
4305                         path->nodes[0] = right;
4306                         path->slots[0] = 0;
4307                         if (path->slots[1] == 0)
4308                                 fixup_low_keys(fs_info, path, &disk_key, 1);
4309                 }
4310                 btrfs_mark_buffer_dirty(right);
4311                 return ret;
4312         }
4313
4314         copy_for_split(trans, root, path, l, right, slot, mid, nritems);
4315
4316         if (split == 2) {
4317                 BUG_ON(num_doubles != 0);
4318                 num_doubles++;
4319                 goto again;
4320         }
4321
4322         return 0;
4323
4324 push_for_double:
4325         push_for_double_split(trans, root, path, data_size);
4326         tried_avoid_double = 1;
4327         if (btrfs_leaf_free_space(root, path->nodes[0]) >= data_size)
4328                 return 0;
4329         goto again;
4330 }
4331
4332 static noinline int setup_leaf_for_split(struct btrfs_trans_handle *trans,
4333                                          struct btrfs_root *root,
4334                                          struct btrfs_path *path, int ins_len)
4335 {
4336         struct btrfs_key key;
4337         struct extent_buffer *leaf;
4338         struct btrfs_file_extent_item *fi;
4339         u64 extent_len = 0;
4340         u32 item_size;
4341         int ret;
4342
4343         leaf = path->nodes[0];
4344         btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
4345
4346         BUG_ON(key.type != BTRFS_EXTENT_DATA_KEY &&
4347                key.type != BTRFS_EXTENT_CSUM_KEY);
4348
4349         if (btrfs_leaf_free_space(root, leaf) >= ins_len)
4350                 return 0;
4351
4352         item_size = btrfs_item_size_nr(leaf, path->slots[0]);
4353         if (key.type == BTRFS_EXTENT_DATA_KEY) {
4354                 fi = btrfs_item_ptr(leaf, path->slots[0],
4355                                     struct btrfs_file_extent_item);
4356                 extent_len = btrfs_file_extent_num_bytes(leaf, fi);
4357         }
4358         btrfs_release_path(path);
4359
4360         path->keep_locks = 1;
4361         path->search_for_split = 1;
4362         ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
4363         path->search_for_split = 0;
4364         if (ret > 0)
4365                 ret = -EAGAIN;
4366         if (ret < 0)
4367                 goto err;
4368
4369         ret = -EAGAIN;
4370         leaf = path->nodes[0];
4371         /* if our item isn't there, return now */
4372         if (item_size != btrfs_item_size_nr(leaf, path->slots[0]))
4373                 goto err;
4374
4375         /* the leaf has  changed, it now has room.  return now */
4376         if (btrfs_leaf_free_space(root, path->nodes[0]) >= ins_len)
4377                 goto err;
4378
4379         if (key.type == BTRFS_EXTENT_DATA_KEY) {
4380                 fi = btrfs_item_ptr(leaf, path->slots[0],
4381                                     struct btrfs_file_extent_item);
4382                 if (extent_len != btrfs_file_extent_num_bytes(leaf, fi))
4383                         goto err;
4384         }
4385
4386         btrfs_set_path_blocking(path);
4387         ret = split_leaf(trans, root, &key, path, ins_len, 1);
4388         if (ret)
4389                 goto err;
4390
4391         path->keep_locks = 0;
4392         btrfs_unlock_up_safe(path, 1);
4393         return 0;
4394 err:
4395         path->keep_locks = 0;
4396         return ret;
4397 }
4398
4399 static noinline int split_item(struct btrfs_trans_handle *trans,
4400                                struct btrfs_root *root,
4401                                struct btrfs_path *path,
4402                                struct btrfs_key *new_key,
4403                                unsigned long split_offset)
4404 {
4405         struct extent_buffer *leaf;
4406         struct btrfs_item *item;
4407         struct btrfs_item *new_item;
4408         int slot;
4409         char *buf;
4410         u32 nritems;
4411         u32 item_size;
4412         u32 orig_offset;
4413         struct btrfs_disk_key disk_key;
4414
4415         leaf = path->nodes[0];
4416         BUG_ON(btrfs_leaf_free_space(root, leaf) < sizeof(struct btrfs_item));
4417
4418         btrfs_set_path_blocking(path);
4419
4420         item = btrfs_item_nr(path->slots[0]);
4421         orig_offset = btrfs_item_offset(leaf, item);
4422         item_size = btrfs_item_size(leaf, item);
4423
4424         buf = kmalloc(item_size, GFP_NOFS);
4425         if (!buf)
4426                 return -ENOMEM;
4427
4428         read_extent_buffer(leaf, buf, btrfs_item_ptr_offset(leaf,
4429                             path->slots[0]), item_size);
4430
4431         slot = path->slots[0] + 1;
4432         nritems = btrfs_header_nritems(leaf);
4433         if (slot != nritems) {
4434                 /* shift the items */
4435                 memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot + 1),
4436                                 btrfs_item_nr_offset(slot),
4437                                 (nritems - slot) * sizeof(struct btrfs_item));
4438         }
4439
4440         btrfs_cpu_key_to_disk(&disk_key, new_key);
4441         btrfs_set_item_key(leaf, &disk_key, slot);
4442
4443         new_item = btrfs_item_nr(slot);
4444
4445         btrfs_set_item_offset(leaf, new_item, orig_offset);
4446         btrfs_set_item_size(leaf, new_item, item_size - split_offset);
4447
4448         btrfs_set_item_offset(leaf, item,
4449                               orig_offset + item_size - split_offset);
4450         btrfs_set_item_size(leaf, item, split_offset);
4451
4452         btrfs_set_header_nritems(leaf, nritems + 1);
4453
4454         /* write the data for the start of the original item */
4455         write_extent_buffer(leaf, buf,
4456                             btrfs_item_ptr_offset(leaf, path->slots[0]),
4457                             split_offset);
4458
4459         /* write the data for the new item */
4460         write_extent_buffer(leaf, buf + split_offset,
4461                             btrfs_item_ptr_offset(leaf, slot),
4462                             item_size - split_offset);
4463         btrfs_mark_buffer_dirty(leaf);
4464
4465         BUG_ON(btrfs_leaf_free_space(root, leaf) < 0);
4466         kfree(buf);
4467         return 0;
4468 }
4469
4470 /*
4471  * This function splits a single item into two items,
4472  * giving 'new_key' to the new item and splitting the
4473  * old one at split_offset (from the start of the item).
4474  *
4475  * The path may be released by this operation.  After
4476  * the split, the path is pointing to the old item.  The
4477  * new item is going to be in the same node as the old one.
4478  *
4479  * Note, the item being split must be smaller enough to live alone on
4480  * a tree block with room for one extra struct btrfs_item
4481  *
4482  * This allows us to split the item in place, keeping a lock on the
4483  * leaf the entire time.
4484  */
4485 int btrfs_split_item(struct btrfs_trans_handle *trans,
4486                      struct btrfs_root *root,
4487                      struct btrfs_path *path,
4488                      struct btrfs_key *new_key,
4489                      unsigned long split_offset)
4490 {
4491         int ret;
4492         ret = setup_leaf_for_split(trans, root, path,
4493                                    sizeof(struct btrfs_item));
4494         if (ret)
4495                 return ret;
4496
4497         ret = split_item(trans, root, path, new_key, split_offset);
4498         return ret;
4499 }
4500
4501 /*
4502  * This function duplicate a item, giving 'new_key' to the new item.
4503  * It guarantees both items live in the same tree leaf and the new item
4504  * is contiguous with the original item.
4505  *
4506  * This allows us to split file extent in place, keeping a lock on the
4507  * leaf the entire time.
4508  */
4509 int btrfs_duplicate_item(struct btrfs_trans_handle *trans,
4510                          struct btrfs_root *root,
4511                          struct btrfs_path *path,
4512                          struct btrfs_key *new_key)
4513 {
4514         struct extent_buffer *leaf;
4515         int ret;
4516         u32 item_size;
4517
4518         leaf = path->nodes[0];
4519         item_size = btrfs_item_size_nr(leaf, path->slots[0]);
4520         ret = setup_leaf_for_split(trans, root, path,
4521                                    item_size + sizeof(struct btrfs_item));
4522         if (ret)
4523                 return ret;
4524
4525         path->slots[0]++;
4526         setup_items_for_insert(root, path, new_key, &item_size,
4527                                item_size, item_size +
4528                                sizeof(struct btrfs_item), 1);
4529         leaf = path->nodes[0];
4530         memcpy_extent_buffer(leaf,
4531                              btrfs_item_ptr_offset(leaf, path->slots[0]),
4532                              btrfs_item_ptr_offset(leaf, path->slots[0] - 1),
4533                              item_size);
4534         return 0;
4535 }
4536
4537 /*
4538  * make the item pointed to by the path smaller.  new_size indicates
4539  * how small to make it, and from_end tells us if we just chop bytes
4540  * off the end of the item or if we shift the item to chop bytes off
4541  * the front.
4542  */
4543 void btrfs_truncate_item(struct btrfs_root *root, struct btrfs_path *path,
4544                          u32 new_size, int from_end)
4545 {
4546         int slot;
4547         struct extent_buffer *leaf;
4548         struct btrfs_item *item;
4549         u32 nritems;
4550         unsigned int data_end;
4551         unsigned int old_data_start;
4552         unsigned int old_size;
4553         unsigned int size_diff;
4554         int i;
4555         struct btrfs_map_token token;
4556
4557         btrfs_init_map_token(&token);
4558
4559         leaf = path->nodes[0];
4560         slot = path->slots[0];
4561
4562         old_size = btrfs_item_size_nr(leaf, slot);
4563         if (old_size == new_size)
4564                 return;
4565
4566         nritems = btrfs_header_nritems(leaf);
4567         data_end = leaf_data_end(root, leaf);
4568
4569         old_data_start = btrfs_item_offset_nr(leaf, slot);
4570
4571         size_diff = old_size - new_size;
4572
4573         BUG_ON(slot < 0);
4574         BUG_ON(slot >= nritems);
4575
4576         /*
4577          * item0..itemN ... dataN.offset..dataN.size .. data0.size
4578          */
4579         /* first correct the data pointers */
4580         for (i = slot; i < nritems; i++) {
4581                 u32 ioff;
4582                 item = btrfs_item_nr(i);
4583
4584                 ioff = btrfs_token_item_offset(leaf, item, &token);
4585                 btrfs_set_token_item_offset(leaf, item,
4586                                             ioff + size_diff, &token);
4587         }
4588
4589         /* shift the data */
4590         if (from_end) {
4591                 memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) +
4592                               data_end + size_diff, btrfs_leaf_data(leaf) +
4593                               data_end, old_data_start + new_size - data_end);
4594         } else {
4595                 struct btrfs_disk_key disk_key;
4596                 u64 offset;
4597
4598                 btrfs_item_key(leaf, &disk_key, slot);
4599
4600                 if (btrfs_disk_key_type(&disk_key) == BTRFS_EXTENT_DATA_KEY) {
4601                         unsigned long ptr;
4602                         struct btrfs_file_extent_item *fi;
4603
4604                         fi = btrfs_item_ptr(leaf, slot,
4605                                             struct btrfs_file_extent_item);
4606                         fi = (struct btrfs_file_extent_item *)(
4607                              (unsigned long)fi - size_diff);
4608
4609                         if (btrfs_file_extent_type(leaf, fi) ==
4610                             BTRFS_FILE_EXTENT_INLINE) {
4611                                 ptr = btrfs_item_ptr_offset(leaf, slot);
4612                                 memmove_extent_buffer(leaf, ptr,
4613                                       (unsigned long)fi,
4614                                       BTRFS_FILE_EXTENT_INLINE_DATA_START);
4615                         }
4616                 }
4617
4618                 memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) +
4619                               data_end + size_diff, btrfs_leaf_data(leaf) +
4620                               data_end, old_data_start - data_end);
4621
4622                 offset = btrfs_disk_key_offset(&disk_key);
4623                 btrfs_set_disk_key_offset(&disk_key, offset + size_diff);
4624                 btrfs_set_item_key(leaf, &disk_key, slot);
4625                 if (slot == 0)
4626                         fixup_low_keys(root->fs_info, path, &disk_key, 1);
4627         }
4628
4629         item = btrfs_item_nr(slot);
4630         btrfs_set_item_size(leaf, item, new_size);
4631         btrfs_mark_buffer_dirty(leaf);
4632
4633         if (btrfs_leaf_free_space(root, leaf) < 0) {
4634                 btrfs_print_leaf(root, leaf);
4635                 BUG();
4636         }
4637 }
4638
4639 /*
4640  * make the item pointed to by the path bigger, data_size is the added size.
4641  */
4642 void btrfs_extend_item(struct btrfs_root *root, struct btrfs_path *path,
4643                        u32 data_size)
4644 {
4645         int slot;
4646         struct extent_buffer *leaf;
4647         struct btrfs_item *item;
4648         u32 nritems;
4649         unsigned int data_end;
4650         unsigned int old_data;
4651         unsigned int old_size;
4652         int i;
4653         struct btrfs_map_token token;
4654
4655         btrfs_init_map_token(&token);
4656
4657         leaf = path->nodes[0];
4658
4659         nritems = btrfs_header_nritems(leaf);
4660         data_end = leaf_data_end(root, leaf);
4661
4662         if (btrfs_leaf_free_space(root, leaf) < data_size) {
4663                 btrfs_print_leaf(root, leaf);
4664                 BUG();
4665         }
4666         slot = path->slots[0];
4667         old_data = btrfs_item_end_nr(leaf, slot);
4668
4669         BUG_ON(slot < 0);
4670         if (slot >= nritems) {
4671                 btrfs_print_leaf(root, leaf);
4672                 btrfs_crit(root->fs_info, "slot %d too large, nritems %d",
4673                        slot, nritems);
4674                 BUG_ON(1);
4675         }
4676
4677         /*
4678          * item0..itemN ... dataN.offset..dataN.size .. data0.size
4679          */
4680         /* first correct the data pointers */
4681         for (i = slot; i < nritems; i++) {
4682                 u32 ioff;
4683                 item = btrfs_item_nr(i);
4684
4685                 ioff = btrfs_token_item_offset(leaf, item, &token);
4686                 btrfs_set_token_item_offset(leaf, item,
4687                                             ioff - data_size, &token);
4688         }
4689
4690         /* shift the data */
4691         memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) +
4692                       data_end - data_size, btrfs_leaf_data(leaf) +
4693                       data_end, old_data - data_end);
4694
4695         data_end = old_data;
4696         old_size = btrfs_item_size_nr(leaf, slot);
4697         item = btrfs_item_nr(slot);
4698         btrfs_set_item_size(leaf, item, old_size + data_size);
4699         btrfs_mark_buffer_dirty(leaf);
4700
4701         if (btrfs_leaf_free_space(root, leaf) < 0) {
4702                 btrfs_print_leaf(root, leaf);
4703                 BUG();
4704         }
4705 }
4706
4707 /*
4708  * this is a helper for btrfs_insert_empty_items, the main goal here is
4709  * to save stack depth by doing the bulk of the work in a function
4710  * that doesn't call btrfs_search_slot
4711  */
4712 void setup_items_for_insert(struct btrfs_root *root, struct btrfs_path *path,
4713                             struct btrfs_key *cpu_key, u32 *data_size,
4714                             u32 total_data, u32 total_size, int nr)
4715 {
4716         struct btrfs_item *item;
4717         int i;
4718         u32 nritems;
4719         unsigned int data_end;
4720         struct btrfs_disk_key disk_key;
4721         struct extent_buffer *leaf;
4722         int slot;
4723         struct btrfs_map_token token;
4724
4725         if (path->slots[0] == 0) {
4726                 btrfs_cpu_key_to_disk(&disk_key, cpu_key);
4727                 fixup_low_keys(root->fs_info, path, &disk_key, 1);
4728         }
4729         btrfs_unlock_up_safe(path, 1);
4730
4731         btrfs_init_map_token(&token);
4732
4733         leaf = path->nodes[0];
4734         slot = path->slots[0];
4735
4736         nritems = btrfs_header_nritems(leaf);
4737         data_end = leaf_data_end(root, leaf);
4738
4739         if (btrfs_leaf_free_space(root, leaf) < total_size) {
4740                 btrfs_print_leaf(root, leaf);
4741                 btrfs_crit(root->fs_info, "not enough freespace need %u have %d",
4742                        total_size, btrfs_leaf_free_space(root, leaf));
4743                 BUG();
4744         }
4745
4746         if (slot != nritems) {
4747                 unsigned int old_data = btrfs_item_end_nr(leaf, slot);
4748
4749                 if (old_data < data_end) {
4750                         btrfs_print_leaf(root, leaf);
4751                         btrfs_crit(root->fs_info, "slot %d old_data %d data_end %d",
4752                                slot, old_data, data_end);
4753                         BUG_ON(1);
4754                 }
4755                 /*
4756                  * item0..itemN ... dataN.offset..dataN.size .. data0.size
4757                  */
4758                 /* first correct the data pointers */
4759                 for (i = slot; i < nritems; i++) {
4760                         u32 ioff;
4761
4762                         item = btrfs_item_nr( i);
4763                         ioff = btrfs_token_item_offset(leaf, item, &token);
4764                         btrfs_set_token_item_offset(leaf, item,
4765                                                     ioff - total_data, &token);
4766                 }
4767                 /* shift the items */
4768                 memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot + nr),
4769                               btrfs_item_nr_offset(slot),
4770                               (nritems - slot) * sizeof(struct btrfs_item));
4771
4772                 /* shift the data */
4773                 memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) +
4774                               data_end - total_data, btrfs_leaf_data(leaf) +
4775                               data_end, old_data - data_end);
4776                 data_end = old_data;
4777         }
4778
4779         /* setup the item for the new data */
4780         for (i = 0; i < nr; i++) {
4781                 btrfs_cpu_key_to_disk(&disk_key, cpu_key + i);
4782                 btrfs_set_item_key(leaf, &disk_key, slot + i);
4783                 item = btrfs_item_nr(slot + i);
4784                 btrfs_set_token_item_offset(leaf, item,
4785                                             data_end - data_size[i], &token);
4786                 data_end -= data_size[i];
4787                 btrfs_set_token_item_size(leaf, item, data_size[i], &token);
4788         }
4789
4790         btrfs_set_header_nritems(leaf, nritems + nr);
4791         btrfs_mark_buffer_dirty(leaf);
4792
4793         if (btrfs_leaf_free_space(root, leaf) < 0) {
4794                 btrfs_print_leaf(root, leaf);
4795                 BUG();
4796         }
4797 }
4798
4799 /*
4800  * Given a key and some data, insert items into the tree.
4801  * This does all the path init required, making room in the tree if needed.
4802  */
4803 int btrfs_insert_empty_items(struct btrfs_trans_handle *trans,
4804                             struct btrfs_root *root,
4805                             struct btrfs_path *path,
4806                             struct btrfs_key *cpu_key, u32 *data_size,
4807                             int nr)
4808 {
4809         int ret = 0;
4810         int slot;
4811         int i;
4812         u32 total_size = 0;
4813         u32 total_data = 0;
4814
4815         for (i = 0; i < nr; i++)
4816                 total_data += data_size[i];
4817
4818         total_size = total_data + (nr * sizeof(struct btrfs_item));
4819         ret = btrfs_search_slot(trans, root, cpu_key, path, total_size, 1);
4820         if (ret == 0)
4821                 return -EEXIST;
4822         if (ret < 0)
4823                 return ret;
4824
4825         slot = path->slots[0];
4826         BUG_ON(slot < 0);
4827
4828         setup_items_for_insert(root, path, cpu_key, data_size,
4829                                total_data, total_size, nr);
4830         return 0;
4831 }
4832
4833 /*
4834  * Given a key and some data, insert an item into the tree.
4835  * This does all the path init required, making room in the tree if needed.
4836  */
4837 int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root
4838                       *root, struct btrfs_key *cpu_key, void *data, u32
4839                       data_size)
4840 {
4841         int ret = 0;
4842         struct btrfs_path *path;
4843         struct extent_buffer *leaf;
4844         unsigned long ptr;
4845
4846         path = btrfs_alloc_path();
4847         if (!path)
4848                 return -ENOMEM;
4849         ret = btrfs_insert_empty_item(trans, root, path, cpu_key, data_size);
4850         if (!ret) {
4851                 leaf = path->nodes[0];
4852                 ptr = btrfs_item_ptr_offset(leaf, path->slots[0]);
4853                 write_extent_buffer(leaf, data, ptr, data_size);
4854                 btrfs_mark_buffer_dirty(leaf);
4855         }
4856         btrfs_free_path(path);
4857         return ret;
4858 }
4859
4860 /*
4861  * delete the pointer from a given node.
4862  *
4863  * the tree should have been previously balanced so the deletion does not
4864  * empty a node.
4865  */
4866 static void del_ptr(struct btrfs_root *root, struct btrfs_path *path,
4867                     int level, int slot)
4868 {
4869         struct extent_buffer *parent = path->nodes[level];
4870         u32 nritems;
4871         int ret;
4872
4873         nritems = btrfs_header_nritems(parent);
4874         if (slot != nritems - 1) {
4875                 if (level)
4876                         tree_mod_log_eb_move(root->fs_info, parent, slot,
4877                                              slot + 1, nritems - slot - 1);
4878                 memmove_extent_buffer(parent,
4879                               btrfs_node_key_ptr_offset(slot),
4880                               btrfs_node_key_ptr_offset(slot + 1),
4881                               sizeof(struct btrfs_key_ptr) *
4882                               (nritems - slot - 1));
4883         } else if (level) {
4884                 ret = tree_mod_log_insert_key(root->fs_info, parent, slot,
4885                                               MOD_LOG_KEY_REMOVE, GFP_NOFS);
4886                 BUG_ON(ret < 0);
4887         }
4888
4889         nritems--;
4890         btrfs_set_header_nritems(parent, nritems);
4891         if (nritems == 0 && parent == root->node) {
4892                 BUG_ON(btrfs_header_level(root->node) != 1);
4893                 /* just turn the root into a leaf and break */
4894                 btrfs_set_header_level(root->node, 0);
4895         } else if (slot == 0) {
4896                 struct btrfs_disk_key disk_key;
4897
4898                 btrfs_node_key(parent, &disk_key, 0);
4899                 fixup_low_keys(root->fs_info, path, &disk_key, level + 1);
4900         }
4901         btrfs_mark_buffer_dirty(parent);
4902 }
4903
4904 /*
4905  * a helper function to delete the leaf pointed to by path->slots[1] and
4906  * path->nodes[1].
4907  *
4908  * This deletes the pointer in path->nodes[1] and frees the leaf
4909  * block extent.  zero is returned if it all worked out, < 0 otherwise.
4910  *
4911  * The path must have already been setup for deleting the leaf, including
4912  * all the proper balancing.  path->nodes[1] must be locked.
4913  */
4914 static noinline void btrfs_del_leaf(struct btrfs_trans_handle *trans,
4915                                     struct btrfs_root *root,
4916                                     struct btrfs_path *path,
4917                                     struct extent_buffer *leaf)
4918 {
4919         WARN_ON(btrfs_header_generation(leaf) != trans->transid);
4920         del_ptr(root, path, 1, path->slots[1]);
4921
4922         /*
4923          * btrfs_free_extent is expensive, we want to make sure we
4924          * aren't holding any locks when we call it
4925          */
4926         btrfs_unlock_up_safe(path, 0);
4927
4928         root_sub_used(root, leaf->len);
4929
4930         extent_buffer_get(leaf);
4931         btrfs_free_tree_block(trans, root, leaf, 0, 1);
4932         free_extent_buffer_stale(leaf);
4933 }
4934 /*
4935  * delete the item at the leaf level in path.  If that empties
4936  * the leaf, remove it from the tree
4937  */
4938 int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
4939                     struct btrfs_path *path, int slot, int nr)
4940 {
4941         struct extent_buffer *leaf;
4942         struct btrfs_item *item;
4943         u32 last_off;
4944         u32 dsize = 0;
4945         int ret = 0;
4946         int wret;
4947         int i;
4948         u32 nritems;
4949         struct btrfs_map_token token;
4950
4951         btrfs_init_map_token(&token);
4952
4953         leaf = path->nodes[0];
4954         last_off = btrfs_item_offset_nr(leaf, slot + nr - 1);
4955
4956         for (i = 0; i < nr; i++)
4957                 dsize += btrfs_item_size_nr(leaf, slot + i);
4958
4959         nritems = btrfs_header_nritems(leaf);
4960
4961         if (slot + nr != nritems) {
4962                 int data_end = leaf_data_end(root, leaf);
4963
4964                 memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) +
4965                               data_end + dsize,
4966                               btrfs_leaf_data(leaf) + data_end,
4967                               last_off - data_end);
4968
4969                 for (i = slot + nr; i < nritems; i++) {
4970                         u32 ioff;
4971
4972                         item = btrfs_item_nr(i);
4973                         ioff = btrfs_token_item_offset(leaf, item, &token);
4974                         btrfs_set_token_item_offset(leaf, item,
4975                                                     ioff + dsize, &token);
4976                 }
4977
4978                 memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot),
4979                               btrfs_item_nr_offset(slot + nr),
4980                               sizeof(struct btrfs_item) *
4981                               (nritems - slot - nr));
4982         }
4983         btrfs_set_header_nritems(leaf, nritems - nr);
4984         nritems -= nr;
4985
4986         /* delete the leaf if we've emptied it */
4987         if (nritems == 0) {
4988                 if (leaf == root->node) {
4989                         btrfs_set_header_level(leaf, 0);
4990                 } else {
4991                         btrfs_set_path_blocking(path);
4992                         clean_tree_block(trans, root->fs_info, leaf);
4993                         btrfs_del_leaf(trans, root, path, leaf);
4994                 }
4995         } else {
4996                 int used = leaf_space_used(leaf, 0, nritems);
4997                 if (slot == 0) {
4998                         struct btrfs_disk_key disk_key;
4999
5000                         btrfs_item_key(leaf, &disk_key, 0);
5001                         fixup_low_keys(root->fs_info, path, &disk_key, 1);
5002                 }
5003
5004                 /* delete the leaf if it is mostly empty */
5005                 if (used < BTRFS_LEAF_DATA_SIZE(root) / 3) {
5006                         /* push_leaf_left fixes the path.
5007                          * make sure the path still points to our leaf
5008                          * for possible call to del_ptr below
5009                          */
5010                         slot = path->slots[1];
5011                         extent_buffer_get(leaf);
5012
5013                         btrfs_set_path_blocking(path);
5014                         wret = push_leaf_left(trans, root, path, 1, 1,
5015                                               1, (u32)-1);
5016                         if (wret < 0 && wret != -ENOSPC)
5017                                 ret = wret;
5018
5019                         if (path->nodes[0] == leaf &&
5020                             btrfs_header_nritems(leaf)) {
5021                                 wret = push_leaf_right(trans, root, path, 1,
5022                                                        1, 1, 0);
5023                                 if (wret < 0 && wret != -ENOSPC)
5024                                         ret = wret;
5025                         }
5026
5027                         if (btrfs_header_nritems(leaf) == 0) {
5028                                 path->slots[1] = slot;
5029                                 btrfs_del_leaf(trans, root, path, leaf);
5030                                 free_extent_buffer(leaf);
5031                                 ret = 0;
5032                         } else {
5033                                 /* if we're still in the path, make sure
5034                                  * we're dirty.  Otherwise, one of the
5035                                  * push_leaf functions must have already
5036                                  * dirtied this buffer
5037                                  */
5038                                 if (path->nodes[0] == leaf)
5039                                         btrfs_mark_buffer_dirty(leaf);
5040                                 free_extent_buffer(leaf);
5041                         }
5042                 } else {
5043                         btrfs_mark_buffer_dirty(leaf);
5044                 }
5045         }
5046         return ret;
5047 }
5048
5049 /*
5050  * search the tree again to find a leaf with lesser keys
5051  * returns 0 if it found something or 1 if there are no lesser leaves.
5052  * returns < 0 on io errors.
5053  *
5054  * This may release the path, and so you may lose any locks held at the
5055  * time you call it.
5056  */
5057 int btrfs_prev_leaf(struct btrfs_root *root, struct btrfs_path *path)
5058 {
5059         struct btrfs_key key;
5060         struct btrfs_disk_key found_key;
5061         int ret;
5062
5063         btrfs_item_key_to_cpu(path->nodes[0], &key, 0);
5064
5065         if (key.offset > 0) {
5066                 key.offset--;
5067         } else if (key.type > 0) {
5068                 key.type--;
5069                 key.offset = (u64)-1;
5070         } else if (key.objectid > 0) {
5071                 key.objectid--;
5072                 key.type = (u8)-1;
5073                 key.offset = (u64)-1;
5074         } else {
5075                 return 1;
5076         }
5077
5078         btrfs_release_path(path);
5079         ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
5080         if (ret < 0)
5081                 return ret;
5082         btrfs_item_key(path->nodes[0], &found_key, 0);
5083         ret = comp_keys(&found_key, &key);
5084         /*
5085          * We might have had an item with the previous key in the tree right
5086          * before we released our path. And after we released our path, that
5087          * item might have been pushed to the first slot (0) of the leaf we
5088          * were holding due to a tree balance. Alternatively, an item with the
5089          * previous key can exist as the only element of a leaf (big fat item).
5090          * Therefore account for these 2 cases, so that our callers (like
5091          * btrfs_previous_item) don't miss an existing item with a key matching
5092          * the previous key we computed above.
5093          */
5094         if (ret <= 0)
5095                 return 0;
5096         return 1;
5097 }
5098
5099 /*
5100  * A helper function to walk down the tree starting at min_key, and looking
5101  * for nodes or leaves that are have a minimum transaction id.
5102  * This is used by the btree defrag code, and tree logging
5103  *
5104  * This does not cow, but it does stuff the starting key it finds back
5105  * into min_key, so you can call btrfs_search_slot with cow=1 on the
5106  * key and get a writable path.
5107  *
5108  * This does lock as it descends, and path->keep_locks should be set
5109  * to 1 by the caller.
5110  *
5111  * This honors path->lowest_level to prevent descent past a given level
5112  * of the tree.
5113  *
5114  * min_trans indicates the oldest transaction that you are interested
5115  * in walking through.  Any nodes or leaves older than min_trans are
5116  * skipped over (without reading them).
5117  *
5118  * returns zero if something useful was found, < 0 on error and 1 if there
5119  * was nothing in the tree that matched the search criteria.
5120  */
5121 int btrfs_search_forward(struct btrfs_root *root, struct btrfs_key *min_key,
5122                          struct btrfs_path *path,
5123                          u64 min_trans)
5124 {
5125         struct extent_buffer *cur;
5126         struct btrfs_key found_key;
5127         int slot;
5128         int sret;
5129         u32 nritems;
5130         int level;
5131         int ret = 1;
5132         int keep_locks = path->keep_locks;
5133
5134         path->keep_locks = 1;
5135 again:
5136         cur = btrfs_read_lock_root_node(root);
5137         level = btrfs_header_level(cur);
5138         WARN_ON(path->nodes[level]);
5139         path->nodes[level] = cur;
5140         path->locks[level] = BTRFS_READ_LOCK;
5141
5142         if (btrfs_header_generation(cur) < min_trans) {
5143                 ret = 1;
5144                 goto out;
5145         }
5146         while (1) {
5147                 nritems = btrfs_header_nritems(cur);
5148                 level = btrfs_header_level(cur);
5149                 sret = bin_search(cur, min_key, level, &slot);
5150
5151                 /* at the lowest level, we're done, setup the path and exit */
5152                 if (level == path->lowest_level) {
5153                         if (slot >= nritems)
5154                                 goto find_next_key;
5155                         ret = 0;
5156                         path->slots[level] = slot;
5157                         btrfs_item_key_to_cpu(cur, &found_key, slot);
5158                         goto out;
5159                 }
5160                 if (sret && slot > 0)
5161                         slot--;
5162                 /*
5163                  * check this node pointer against the min_trans parameters.
5164                  * If it is too old, old, skip to the next one.
5165                  */
5166                 while (slot < nritems) {
5167                         u64 gen;
5168
5169                         gen = btrfs_node_ptr_generation(cur, slot);
5170                         if (gen < min_trans) {
5171                                 slot++;
5172                                 continue;
5173                         }
5174                         break;
5175                 }
5176 find_next_key:
5177                 /*
5178                  * we didn't find a candidate key in this node, walk forward
5179                  * and find another one
5180                  */
5181                 if (slot >= nritems) {
5182                         path->slots[level] = slot;
5183                         btrfs_set_path_blocking(path);
5184                         sret = btrfs_find_next_key(root, path, min_key, level,
5185                                                   min_trans);
5186                         if (sret == 0) {
5187                                 btrfs_release_path(path);
5188                                 goto again;
5189                         } else {
5190                                 goto out;
5191                         }
5192                 }
5193                 /* save our key for returning back */
5194                 btrfs_node_key_to_cpu(cur, &found_key, slot);
5195                 path->slots[level] = slot;
5196                 if (level == path->lowest_level) {
5197                         ret = 0;
5198                         goto out;
5199                 }
5200                 btrfs_set_path_blocking(path);
5201                 cur = read_node_slot(root, cur, slot);
5202                 BUG_ON(!cur); /* -ENOMEM */
5203
5204                 btrfs_tree_read_lock(cur);
5205
5206                 path->locks[level - 1] = BTRFS_READ_LOCK;
5207                 path->nodes[level - 1] = cur;
5208                 unlock_up(path, level, 1, 0, NULL);
5209                 btrfs_clear_path_blocking(path, NULL, 0);
5210         }
5211 out:
5212         path->keep_locks = keep_locks;
5213         if (ret == 0) {
5214                 btrfs_unlock_up_safe(path, path->lowest_level + 1);
5215                 btrfs_set_path_blocking(path);
5216                 memcpy(min_key, &found_key, sizeof(found_key));
5217         }
5218         return ret;
5219 }
5220
5221 static void tree_move_down(struct btrfs_root *root,
5222                            struct btrfs_path *path,
5223                            int *level, int root_level)
5224 {
5225         BUG_ON(*level == 0);
5226         path->nodes[*level - 1] = read_node_slot(root, path->nodes[*level],
5227                                         path->slots[*level]);
5228         path->slots[*level - 1] = 0;
5229         (*level)--;
5230 }
5231
5232 static int tree_move_next_or_upnext(struct btrfs_root *root,
5233                                     struct btrfs_path *path,
5234                                     int *level, int root_level)
5235 {
5236         int ret = 0;
5237         int nritems;
5238         nritems = btrfs_header_nritems(path->nodes[*level]);
5239
5240         path->slots[*level]++;
5241
5242         while (path->slots[*level] >= nritems) {
5243                 if (*level == root_level)
5244                         return -1;
5245
5246                 /* move upnext */
5247                 path->slots[*level] = 0;
5248                 free_extent_buffer(path->nodes[*level]);
5249                 path->nodes[*level] = NULL;
5250                 (*level)++;
5251                 path->slots[*level]++;
5252
5253                 nritems = btrfs_header_nritems(path->nodes[*level]);
5254                 ret = 1;
5255         }
5256         return ret;
5257 }
5258
5259 /*
5260  * Returns 1 if it had to move up and next. 0 is returned if it moved only next
5261  * or down.
5262  */
5263 static int tree_advance(struct btrfs_root *root,
5264                         struct btrfs_path *path,
5265                         int *level, int root_level,
5266                         int allow_down,
5267                         struct btrfs_key *key)
5268 {
5269         int ret;
5270
5271         if (*level == 0 || !allow_down) {
5272                 ret = tree_move_next_or_upnext(root, path, level, root_level);
5273         } else {
5274                 tree_move_down(root, path, level, root_level);
5275                 ret = 0;
5276         }
5277         if (ret >= 0) {
5278                 if (*level == 0)
5279                         btrfs_item_key_to_cpu(path->nodes[*level], key,
5280                                         path->slots[*level]);
5281                 else
5282                         btrfs_node_key_to_cpu(path->nodes[*level], key,
5283                                         path->slots[*level]);
5284         }
5285         return ret;
5286 }
5287
5288 static int tree_compare_item(struct btrfs_root *left_root,
5289                              struct btrfs_path *left_path,
5290                              struct btrfs_path *right_path,
5291                              char *tmp_buf)
5292 {
5293         int cmp;
5294         int len1, len2;
5295         unsigned long off1, off2;
5296
5297         len1 = btrfs_item_size_nr(left_path->nodes[0], left_path->slots[0]);
5298         len2 = btrfs_item_size_nr(right_path->nodes[0], right_path->slots[0]);
5299         if (len1 != len2)
5300                 return 1;
5301
5302         off1 = btrfs_item_ptr_offset(left_path->nodes[0], left_path->slots[0]);
5303         off2 = btrfs_item_ptr_offset(right_path->nodes[0],
5304                                 right_path->slots[0]);
5305
5306         read_extent_buffer(left_path->nodes[0], tmp_buf, off1, len1);
5307
5308         cmp = memcmp_extent_buffer(right_path->nodes[0], tmp_buf, off2, len1);
5309         if (cmp)
5310                 return 1;
5311         return 0;
5312 }
5313
5314 #define ADVANCE 1
5315 #define ADVANCE_ONLY_NEXT -1
5316
5317 /*
5318  * This function compares two trees and calls the provided callback for
5319  * every changed/new/deleted item it finds.
5320  * If shared tree blocks are encountered, whole subtrees are skipped, making
5321  * the compare pretty fast on snapshotted subvolumes.
5322  *
5323  * This currently works on commit roots only. As commit roots are read only,
5324  * we don't do any locking. The commit roots are protected with transactions.
5325  * Transactions are ended and rejoined when a commit is tried in between.
5326  *
5327  * This function checks for modifications done to the trees while comparing.
5328  * If it detects a change, it aborts immediately.
5329  */
5330 int btrfs_compare_trees(struct btrfs_root *left_root,
5331                         struct btrfs_root *right_root,
5332                         btrfs_changed_cb_t changed_cb, void *ctx)
5333 {
5334         int ret;
5335         int cmp;
5336         struct btrfs_path *left_path = NULL;
5337         struct btrfs_path *right_path = NULL;
5338         struct btrfs_key left_key;
5339         struct btrfs_key right_key;
5340         char *tmp_buf = NULL;
5341         int left_root_level;
5342         int right_root_level;
5343         int left_level;
5344         int right_level;
5345         int left_end_reached;
5346         int right_end_reached;
5347         int advance_left;
5348         int advance_right;
5349         u64 left_blockptr;
5350         u64 right_blockptr;
5351         u64 left_gen;
5352         u64 right_gen;
5353
5354         left_path = btrfs_alloc_path();
5355         if (!left_path) {
5356                 ret = -ENOMEM;
5357                 goto out;
5358         }
5359         right_path = btrfs_alloc_path();
5360         if (!right_path) {
5361                 ret = -ENOMEM;
5362                 goto out;
5363         }
5364
5365         tmp_buf = kmalloc(left_root->nodesize, GFP_KERNEL | __GFP_NOWARN);
5366         if (!tmp_buf) {
5367                 tmp_buf = vmalloc(left_root->nodesize);
5368                 if (!tmp_buf) {
5369                         ret = -ENOMEM;
5370                         goto out;
5371                 }
5372         }
5373
5374         left_path->search_commit_root = 1;
5375         left_path->skip_locking = 1;
5376         right_path->search_commit_root = 1;
5377         right_path->skip_locking = 1;
5378
5379         /*
5380          * Strategy: Go to the first items of both trees. Then do
5381          *
5382          * If both trees are at level 0
5383          *   Compare keys of current items
5384          *     If left < right treat left item as new, advance left tree
5385          *       and repeat
5386          *     If left > right treat right item as deleted, advance right tree
5387          *       and repeat
5388          *     If left == right do deep compare of items, treat as changed if
5389          *       needed, advance both trees and repeat
5390          * If both trees are at the same level but not at level 0
5391          *   Compare keys of current nodes/leafs
5392          *     If left < right advance left tree and repeat
5393          *     If left > right advance right tree and repeat
5394          *     If left == right compare blockptrs of the next nodes/leafs
5395          *       If they match advance both trees but stay at the same level
5396          *         and repeat
5397          *       If they don't match advance both trees while allowing to go
5398          *         deeper and repeat
5399          * If tree levels are different
5400          *   Advance the tree that needs it and repeat
5401          *
5402          * Advancing a tree means:
5403          *   If we are at level 0, try to go to the next slot. If that's not
5404          *   possible, go one level up and repeat. Stop when we found a level
5405          *   where we could go to the next slot. We may at this point be on a
5406          *   node or a leaf.
5407          *
5408          *   If we are not at level 0 and not on shared tree blocks, go one
5409          *   level deeper.
5410          *
5411          *   If we are not at level 0 and on shared tree blocks, go one slot to
5412          *   the right if possible or go up and right.
5413          */
5414
5415         down_read(&left_root->fs_info->commit_root_sem);
5416         left_level = btrfs_header_level(left_root->commit_root);
5417         left_root_level = left_level;
5418         left_path->nodes[left_level] = left_root->commit_root;
5419         extent_buffer_get(left_path->nodes[left_level]);
5420
5421         right_level = btrfs_header_level(right_root->commit_root);
5422         right_root_level = right_level;
5423         right_path->nodes[right_level] = right_root->commit_root;
5424         extent_buffer_get(right_path->nodes[right_level]);
5425         up_read(&left_root->fs_info->commit_root_sem);
5426
5427         if (left_level == 0)
5428                 btrfs_item_key_to_cpu(left_path->nodes[left_level],
5429                                 &left_key, left_path->slots[left_level]);
5430         else
5431                 btrfs_node_key_to_cpu(left_path->nodes[left_level],
5432                                 &left_key, left_path->slots[left_level]);
5433         if (right_level == 0)
5434                 btrfs_item_key_to_cpu(right_path->nodes[right_level],
5435                                 &right_key, right_path->slots[right_level]);
5436         else
5437                 btrfs_node_key_to_cpu(right_path->nodes[right_level],
5438                                 &right_key, right_path->slots[right_level]);
5439
5440         left_end_reached = right_end_reached = 0;
5441         advance_left = advance_right = 0;
5442
5443         while (1) {
5444                 if (advance_left && !left_end_reached) {
5445                         ret = tree_advance(left_root, left_path, &left_level,
5446                                         left_root_level,
5447                                         advance_left != ADVANCE_ONLY_NEXT,
5448                                         &left_key);
5449                         if (ret < 0)
5450                                 left_end_reached = ADVANCE;
5451                         advance_left = 0;
5452                 }
5453                 if (advance_right && !right_end_reached) {
5454                         ret = tree_advance(right_root, right_path, &right_level,
5455                                         right_root_level,
5456                                         advance_right != ADVANCE_ONLY_NEXT,
5457                                         &right_key);
5458                         if (ret < 0)
5459                                 right_end_reached = ADVANCE;
5460                         advance_right = 0;
5461                 }
5462
5463                 if (left_end_reached && right_end_reached) {
5464                         ret = 0;
5465                         goto out;
5466                 } else if (left_end_reached) {
5467                         if (right_level == 0) {
5468                                 ret = changed_cb(left_root, right_root,
5469                                                 left_path, right_path,
5470                                                 &right_key,
5471                                                 BTRFS_COMPARE_TREE_DELETED,
5472                                                 ctx);
5473                                 if (ret < 0)
5474                                         goto out;
5475                         }
5476                         advance_right = ADVANCE;
5477                         continue;
5478                 } else if (right_end_reached) {
5479                         if (left_level == 0) {
5480                                 ret = changed_cb(left_root, right_root,
5481                                                 left_path, right_path,
5482                                                 &left_key,
5483                                                 BTRFS_COMPARE_TREE_NEW,
5484                                                 ctx);
5485                                 if (ret < 0)
5486                                         goto out;
5487                         }
5488                         advance_left = ADVANCE;
5489                         continue;
5490                 }
5491
5492                 if (left_level == 0 && right_level == 0) {
5493                         cmp = btrfs_comp_cpu_keys(&left_key, &right_key);
5494                         if (cmp < 0) {
5495                                 ret = changed_cb(left_root, right_root,
5496                                                 left_path, right_path,
5497                                                 &left_key,
5498                                                 BTRFS_COMPARE_TREE_NEW,
5499                                                 ctx);
5500                                 if (ret < 0)
5501                                         goto out;
5502                                 advance_left = ADVANCE;
5503                         } else if (cmp > 0) {
5504                                 ret = changed_cb(left_root, right_root,
5505                                                 left_path, right_path,
5506                                                 &right_key,
5507                                                 BTRFS_COMPARE_TREE_DELETED,
5508                                                 ctx);
5509                                 if (ret < 0)
5510                                         goto out;
5511                                 advance_right = ADVANCE;
5512                         } else {
5513                                 enum btrfs_compare_tree_result result;
5514
5515                                 WARN_ON(!extent_buffer_uptodate(left_path->nodes[0]));
5516                                 ret = tree_compare_item(left_root, left_path,
5517                                                 right_path, tmp_buf);
5518                                 if (ret)
5519                                         result = BTRFS_COMPARE_TREE_CHANGED;
5520                                 else
5521                                         result = BTRFS_COMPARE_TREE_SAME;
5522                                 ret = changed_cb(left_root, right_root,
5523                                                  left_path, right_path,
5524                                                  &left_key, result, ctx);
5525                                 if (ret < 0)
5526                                         goto out;
5527                                 advance_left = ADVANCE;
5528                                 advance_right = ADVANCE;
5529                         }
5530                 } else if (left_level == right_level) {
5531                         cmp = btrfs_comp_cpu_keys(&left_key, &right_key);
5532                         if (cmp < 0) {
5533                                 advance_left = ADVANCE;
5534                         } else if (cmp > 0) {
5535                                 advance_right = ADVANCE;
5536                         } else {
5537                                 left_blockptr = btrfs_node_blockptr(
5538                                                 left_path->nodes[left_level],
5539                                                 left_path->slots[left_level]);
5540                                 right_blockptr = btrfs_node_blockptr(
5541                                                 right_path->nodes[right_level],
5542                                                 right_path->slots[right_level]);
5543                                 left_gen = btrfs_node_ptr_generation(
5544                                                 left_path->nodes[left_level],
5545                                                 left_path->slots[left_level]);
5546                                 right_gen = btrfs_node_ptr_generation(
5547                                                 right_path->nodes[right_level],
5548                                                 right_path->slots[right_level]);
5549                                 if (left_blockptr == right_blockptr &&
5550                                     left_gen == right_gen) {
5551                                         /*
5552                                          * As we're on a shared block, don't
5553                                          * allow to go deeper.
5554                                          */
5555                                         advance_left = ADVANCE_ONLY_NEXT;
5556                                         advance_right = ADVANCE_ONLY_NEXT;
5557                                 } else {
5558                                         advance_left = ADVANCE;
5559                                         advance_right = ADVANCE;
5560                                 }
5561                         }
5562                 } else if (left_level < right_level) {
5563                         advance_right = ADVANCE;
5564                 } else {
5565                         advance_left = ADVANCE;
5566                 }
5567         }
5568
5569 out:
5570         btrfs_free_path(left_path);
5571         btrfs_free_path(right_path);
5572         kvfree(tmp_buf);
5573         return ret;
5574 }
5575
5576 /*
5577  * this is similar to btrfs_next_leaf, but does not try to preserve
5578  * and fixup the path.  It looks for and returns the next key in the
5579  * tree based on the current path and the min_trans parameters.
5580  *
5581  * 0 is returned if another key is found, < 0 if there are any errors
5582  * and 1 is returned if there are no higher keys in the tree
5583  *
5584  * path->keep_locks should be set to 1 on the search made before
5585  * calling this function.
5586  */
5587 int btrfs_find_next_key(struct btrfs_root *root, struct btrfs_path *path,
5588                         struct btrfs_key *key, int level, u64 min_trans)
5589 {
5590         int slot;
5591         struct extent_buffer *c;
5592
5593         WARN_ON(!path->keep_locks);
5594         while (level < BTRFS_MAX_LEVEL) {
5595                 if (!path->nodes[level])
5596                         return 1;
5597
5598                 slot = path->slots[level] + 1;
5599                 c = path->nodes[level];
5600 next:
5601                 if (slot >= btrfs_header_nritems(c)) {
5602                         int ret;
5603                         int orig_lowest;
5604                         struct btrfs_key cur_key;
5605                         if (level + 1 >= BTRFS_MAX_LEVEL ||
5606                             !path->nodes[level + 1])
5607                                 return 1;
5608
5609                         if (path->locks[level + 1]) {
5610                                 level++;
5611                                 continue;
5612                         }
5613
5614                         slot = btrfs_header_nritems(c) - 1;
5615                         if (level == 0)
5616                                 btrfs_item_key_to_cpu(c, &cur_key, slot);
5617                         else
5618                                 btrfs_node_key_to_cpu(c, &cur_key, slot);
5619
5620                         orig_lowest = path->lowest_level;
5621                         btrfs_release_path(path);
5622                         path->lowest_level = level;
5623                         ret = btrfs_search_slot(NULL, root, &cur_key, path,
5624                                                 0, 0);
5625                         path->lowest_level = orig_lowest;
5626                         if (ret < 0)
5627                                 return ret;
5628
5629                         c = path->nodes[level];
5630                         slot = path->slots[level];
5631                         if (ret == 0)
5632                                 slot++;
5633                         goto next;
5634                 }
5635
5636                 if (level == 0)
5637                         btrfs_item_key_to_cpu(c, key, slot);
5638                 else {
5639                         u64 gen = btrfs_node_ptr_generation(c, slot);
5640
5641                         if (gen < min_trans) {
5642                                 slot++;
5643                                 goto next;
5644                         }
5645                         btrfs_node_key_to_cpu(c, key, slot);
5646                 }
5647                 return 0;
5648         }
5649         return 1;
5650 }
5651
5652 /*
5653  * search the tree again to find a leaf with greater keys
5654  * returns 0 if it found something or 1 if there are no greater leaves.
5655  * returns < 0 on io errors.
5656  */
5657 int btrfs_next_leaf(struct btrfs_root *root, struct btrfs_path *path)
5658 {
5659         return btrfs_next_old_leaf(root, path, 0);
5660 }
5661
5662 int btrfs_next_old_leaf(struct btrfs_root *root, struct btrfs_path *path,
5663                         u64 time_seq)
5664 {
5665         int slot;
5666         int level;
5667         struct extent_buffer *c;
5668         struct extent_buffer *next;
5669         struct btrfs_key key;
5670         u32 nritems;
5671         int ret;
5672         int old_spinning = path->leave_spinning;
5673         int next_rw_lock = 0;
5674
5675         nritems = btrfs_header_nritems(path->nodes[0]);
5676         if (nritems == 0)
5677                 return 1;
5678
5679         btrfs_item_key_to_cpu(path->nodes[0], &key, nritems - 1);
5680 again:
5681         level = 1;
5682         next = NULL;
5683         next_rw_lock = 0;
5684         btrfs_release_path(path);
5685
5686         path->keep_locks = 1;
5687         path->leave_spinning = 1;
5688
5689         if (time_seq)
5690                 ret = btrfs_search_old_slot(root, &key, path, time_seq);
5691         else
5692                 ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
5693         path->keep_locks = 0;
5694
5695         if (ret < 0)
5696                 return ret;
5697
5698         nritems = btrfs_header_nritems(path->nodes[0]);
5699         /*
5700          * by releasing the path above we dropped all our locks.  A balance
5701          * could have added more items next to the key that used to be
5702          * at the very end of the block.  So, check again here and
5703          * advance the path if there are now more items available.
5704          */
5705         if (nritems > 0 && path->slots[0] < nritems - 1) {
5706                 if (ret == 0)
5707                         path->slots[0]++;
5708                 ret = 0;
5709                 goto done;
5710         }
5711         /*
5712          * So the above check misses one case:
5713          * - after releasing the path above, someone has removed the item that
5714          *   used to be at the very end of the block, and balance between leafs
5715          *   gets another one with bigger key.offset to replace it.
5716          *
5717          * This one should be returned as well, or we can get leaf corruption
5718          * later(esp. in __btrfs_drop_extents()).
5719          *
5720          * And a bit more explanation about this check,
5721          * with ret > 0, the key isn't found, the path points to the slot
5722          * where it should be inserted, so the path->slots[0] item must be the
5723          * bigger one.
5724          */
5725         if (nritems > 0 && ret > 0 && path->slots[0] == nritems - 1) {
5726                 ret = 0;
5727                 goto done;
5728         }
5729
5730         while (level < BTRFS_MAX_LEVEL) {
5731                 if (!path->nodes[level]) {
5732                         ret = 1;
5733                         goto done;
5734                 }
5735
5736                 slot = path->slots[level] + 1;
5737                 c = path->nodes[level];
5738                 if (slot >= btrfs_header_nritems(c)) {
5739                         level++;
5740                         if (level == BTRFS_MAX_LEVEL) {
5741                                 ret = 1;
5742                                 goto done;
5743                         }
5744                         continue;
5745                 }
5746
5747                 if (next) {
5748                         btrfs_tree_unlock_rw(next, next_rw_lock);
5749                         free_extent_buffer(next);
5750                 }
5751
5752                 next = c;
5753                 next_rw_lock = path->locks[level];
5754                 ret = read_block_for_search(NULL, root, path, &next, level,
5755                                             slot, &key, 0);
5756                 if (ret == -EAGAIN)
5757                         goto again;
5758
5759                 if (ret < 0) {
5760                         btrfs_release_path(path);
5761                         goto done;
5762                 }
5763
5764                 if (!path->skip_locking) {
5765                         ret = btrfs_try_tree_read_lock(next);
5766                         if (!ret && time_seq) {
5767                                 /*
5768                                  * If we don't get the lock, we may be racing
5769                                  * with push_leaf_left, holding that lock while
5770                                  * itself waiting for the leaf we've currently
5771                                  * locked. To solve this situation, we give up
5772                                  * on our lock and cycle.
5773                                  */
5774                                 free_extent_buffer(next);
5775                                 btrfs_release_path(path);
5776                                 cond_resched();
5777                                 goto again;
5778                         }
5779                         if (!ret) {
5780                                 btrfs_set_path_blocking(path);
5781                                 btrfs_tree_read_lock(next);
5782                                 btrfs_clear_path_blocking(path, next,
5783                                                           BTRFS_READ_LOCK);
5784                         }
5785                         next_rw_lock = BTRFS_READ_LOCK;
5786                 }
5787                 break;
5788         }
5789         path->slots[level] = slot;
5790         while (1) {
5791                 level--;
5792                 c = path->nodes[level];
5793                 if (path->locks[level])
5794                         btrfs_tree_unlock_rw(c, path->locks[level]);
5795
5796                 free_extent_buffer(c);
5797                 path->nodes[level] = next;
5798                 path->slots[level] = 0;
5799                 if (!path->skip_locking)
5800                         path->locks[level] = next_rw_lock;
5801                 if (!level)
5802                         break;
5803
5804                 ret = read_block_for_search(NULL, root, path, &next, level,
5805                                             0, &key, 0);
5806                 if (ret == -EAGAIN)
5807                         goto again;
5808
5809                 if (ret < 0) {
5810                         btrfs_release_path(path);
5811                         goto done;
5812                 }
5813
5814                 if (!path->skip_locking) {
5815                         ret = btrfs_try_tree_read_lock(next);
5816                         if (!ret) {
5817                                 btrfs_set_path_blocking(path);
5818                                 btrfs_tree_read_lock(next);
5819                                 btrfs_clear_path_blocking(path, next,
5820                                                           BTRFS_READ_LOCK);
5821                         }
5822                         next_rw_lock = BTRFS_READ_LOCK;
5823                 }
5824         }
5825         ret = 0;
5826 done:
5827         unlock_up(path, 0, 1, 0, NULL);
5828         path->leave_spinning = old_spinning;
5829         if (!old_spinning)
5830                 btrfs_set_path_blocking(path);
5831
5832         return ret;
5833 }
5834
5835 /*
5836  * this uses btrfs_prev_leaf to walk backwards in the tree, and keeps
5837  * searching until it gets past min_objectid or finds an item of 'type'
5838  *
5839  * returns 0 if something is found, 1 if nothing was found and < 0 on error
5840  */
5841 int btrfs_previous_item(struct btrfs_root *root,
5842                         struct btrfs_path *path, u64 min_objectid,
5843                         int type)
5844 {
5845         struct btrfs_key found_key;
5846         struct extent_buffer *leaf;
5847         u32 nritems;
5848         int ret;
5849
5850         while (1) {
5851                 if (path->slots[0] == 0) {
5852                         btrfs_set_path_blocking(path);
5853                         ret = btrfs_prev_leaf(root, path);
5854                         if (ret != 0)
5855                                 return ret;
5856                 } else {
5857                         path->slots[0]--;
5858                 }
5859                 leaf = path->nodes[0];
5860                 nritems = btrfs_header_nritems(leaf);
5861                 if (nritems == 0)
5862                         return 1;
5863                 if (path->slots[0] == nritems)
5864                         path->slots[0]--;
5865
5866                 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
5867                 if (found_key.objectid < min_objectid)
5868                         break;
5869                 if (found_key.type == type)
5870                         return 0;
5871                 if (found_key.objectid == min_objectid &&
5872                     found_key.type < type)
5873                         break;
5874         }
5875         return 1;
5876 }
5877
5878 /*
5879  * search in extent tree to find a previous Metadata/Data extent item with
5880  * min objecitd.
5881  *
5882  * returns 0 if something is found, 1 if nothing was found and < 0 on error
5883  */
5884 int btrfs_previous_extent_item(struct btrfs_root *root,
5885                         struct btrfs_path *path, u64 min_objectid)
5886 {
5887         struct btrfs_key found_key;
5888         struct extent_buffer *leaf;
5889         u32 nritems;
5890         int ret;
5891
5892         while (1) {
5893                 if (path->slots[0] == 0) {
5894                         btrfs_set_path_blocking(path);
5895                         ret = btrfs_prev_leaf(root, path);
5896                         if (ret != 0)
5897                                 return ret;
5898                 } else {
5899                         path->slots[0]--;
5900                 }
5901                 leaf = path->nodes[0];
5902                 nritems = btrfs_header_nritems(leaf);
5903                 if (nritems == 0)
5904                         return 1;
5905                 if (path->slots[0] == nritems)
5906                         path->slots[0]--;
5907
5908                 btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
5909                 if (found_key.objectid < min_objectid)
5910                         break;
5911                 if (found_key.type == BTRFS_EXTENT_ITEM_KEY ||
5912                     found_key.type == BTRFS_METADATA_ITEM_KEY)
5913                         return 0;
5914                 if (found_key.objectid == min_objectid &&
5915                     found_key.type < BTRFS_EXTENT_ITEM_KEY)
5916                         break;
5917         }
5918         return 1;
5919 }