delayed-inode.c 48.1 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84
/*
 * Copyright (C) 2011 Fujitsu.  All rights reserved.
 * Written by Miao Xie <miaox@cn.fujitsu.com>
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public
 * License v2 as published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * General Public License for more details.
 *
 * You should have received a copy of the GNU General Public
 * License along with this program; if not, write to the
 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
 * Boston, MA 021110-1307, USA.
 */

#include <linux/slab.h>
#include "delayed-inode.h"
#include "disk-io.h"
#include "transaction.h"

#define BTRFS_DELAYED_WRITEBACK		400
#define BTRFS_DELAYED_BACKGROUND	100

static struct kmem_cache *delayed_node_cache;

int __init btrfs_delayed_inode_init(void)
{
	delayed_node_cache = kmem_cache_create("delayed_node",
					sizeof(struct btrfs_delayed_node),
					0,
					SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD,
					NULL);
	if (!delayed_node_cache)
		return -ENOMEM;
	return 0;
}

void btrfs_delayed_inode_exit(void)
{
	if (delayed_node_cache)
		kmem_cache_destroy(delayed_node_cache);
}

static inline void btrfs_init_delayed_node(
				struct btrfs_delayed_node *delayed_node,
				struct btrfs_root *root, u64 inode_id)
{
	delayed_node->root = root;
	delayed_node->inode_id = inode_id;
	atomic_set(&delayed_node->refs, 0);
	delayed_node->count = 0;
	delayed_node->in_list = 0;
	delayed_node->inode_dirty = 0;
	delayed_node->ins_root = RB_ROOT;
	delayed_node->del_root = RB_ROOT;
	mutex_init(&delayed_node->mutex);
	delayed_node->index_cnt = 0;
	INIT_LIST_HEAD(&delayed_node->n_list);
	INIT_LIST_HEAD(&delayed_node->p_list);
	delayed_node->bytes_reserved = 0;
}

static inline int btrfs_is_continuous_delayed_item(
					struct btrfs_delayed_item *item1,
					struct btrfs_delayed_item *item2)
{
	if (item1->key.type == BTRFS_DIR_INDEX_KEY &&
	    item1->key.objectid == item2->key.objectid &&
	    item1->key.type == item2->key.type &&
	    item1->key.offset + 1 == item2->key.offset)
		return 1;
	return 0;
}

static inline struct btrfs_delayed_root *btrfs_get_delayed_root(
							struct btrfs_root *root)
{
	return root->fs_info->delayed_root;
}

85
static struct btrfs_delayed_node *btrfs_get_delayed_node(struct inode *inode)
86 87 88
{
	struct btrfs_inode *btrfs_inode = BTRFS_I(inode);
	struct btrfs_root *root = btrfs_inode->root;
89
	u64 ino = btrfs_ino(inode);
90
	struct btrfs_delayed_node *node;
91 92 93

	node = ACCESS_ONCE(btrfs_inode->delayed_node);
	if (node) {
94
		atomic_inc(&node->refs);
95 96 97 98
		return node;
	}

	spin_lock(&root->inode_lock);
99
	node = radix_tree_lookup(&root->delayed_nodes_tree, ino);
100 101
	if (node) {
		if (btrfs_inode->delayed_node) {
102 103
			atomic_inc(&node->refs);	/* can be accessed */
			BUG_ON(btrfs_inode->delayed_node != node);
104
			spin_unlock(&root->inode_lock);
105
			return node;
106 107 108 109 110 111 112 113 114
		}
		btrfs_inode->delayed_node = node;
		atomic_inc(&node->refs);	/* can be accessed */
		atomic_inc(&node->refs);	/* cached in the inode */
		spin_unlock(&root->inode_lock);
		return node;
	}
	spin_unlock(&root->inode_lock);

115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131
	return NULL;
}

static struct btrfs_delayed_node *btrfs_get_or_create_delayed_node(
							struct inode *inode)
{
	struct btrfs_delayed_node *node;
	struct btrfs_inode *btrfs_inode = BTRFS_I(inode);
	struct btrfs_root *root = btrfs_inode->root;
	u64 ino = btrfs_ino(inode);
	int ret;

again:
	node = btrfs_get_delayed_node(inode);
	if (node)
		return node;

132 133 134
	node = kmem_cache_alloc(delayed_node_cache, GFP_NOFS);
	if (!node)
		return ERR_PTR(-ENOMEM);
135
	btrfs_init_delayed_node(node, root, ino);
136 137 138 139 140 141 142 143 144 145 146

	atomic_inc(&node->refs);	/* cached in the btrfs inode */
	atomic_inc(&node->refs);	/* can be accessed */

	ret = radix_tree_preload(GFP_NOFS & ~__GFP_HIGHMEM);
	if (ret) {
		kmem_cache_free(delayed_node_cache, node);
		return ERR_PTR(ret);
	}

	spin_lock(&root->inode_lock);
147
	ret = radix_tree_insert(&root->delayed_nodes_tree, ino, node);
148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593
	if (ret == -EEXIST) {
		kmem_cache_free(delayed_node_cache, node);
		spin_unlock(&root->inode_lock);
		radix_tree_preload_end();
		goto again;
	}
	btrfs_inode->delayed_node = node;
	spin_unlock(&root->inode_lock);
	radix_tree_preload_end();

	return node;
}

/*
 * Call it when holding delayed_node->mutex
 *
 * If mod = 1, add this node into the prepared list.
 */
static void btrfs_queue_delayed_node(struct btrfs_delayed_root *root,
				     struct btrfs_delayed_node *node,
				     int mod)
{
	spin_lock(&root->lock);
	if (node->in_list) {
		if (!list_empty(&node->p_list))
			list_move_tail(&node->p_list, &root->prepare_list);
		else if (mod)
			list_add_tail(&node->p_list, &root->prepare_list);
	} else {
		list_add_tail(&node->n_list, &root->node_list);
		list_add_tail(&node->p_list, &root->prepare_list);
		atomic_inc(&node->refs);	/* inserted into list */
		root->nodes++;
		node->in_list = 1;
	}
	spin_unlock(&root->lock);
}

/* Call it when holding delayed_node->mutex */
static void btrfs_dequeue_delayed_node(struct btrfs_delayed_root *root,
				       struct btrfs_delayed_node *node)
{
	spin_lock(&root->lock);
	if (node->in_list) {
		root->nodes--;
		atomic_dec(&node->refs);	/* not in the list */
		list_del_init(&node->n_list);
		if (!list_empty(&node->p_list))
			list_del_init(&node->p_list);
		node->in_list = 0;
	}
	spin_unlock(&root->lock);
}

struct btrfs_delayed_node *btrfs_first_delayed_node(
			struct btrfs_delayed_root *delayed_root)
{
	struct list_head *p;
	struct btrfs_delayed_node *node = NULL;

	spin_lock(&delayed_root->lock);
	if (list_empty(&delayed_root->node_list))
		goto out;

	p = delayed_root->node_list.next;
	node = list_entry(p, struct btrfs_delayed_node, n_list);
	atomic_inc(&node->refs);
out:
	spin_unlock(&delayed_root->lock);

	return node;
}

struct btrfs_delayed_node *btrfs_next_delayed_node(
						struct btrfs_delayed_node *node)
{
	struct btrfs_delayed_root *delayed_root;
	struct list_head *p;
	struct btrfs_delayed_node *next = NULL;

	delayed_root = node->root->fs_info->delayed_root;
	spin_lock(&delayed_root->lock);
	if (!node->in_list) {	/* not in the list */
		if (list_empty(&delayed_root->node_list))
			goto out;
		p = delayed_root->node_list.next;
	} else if (list_is_last(&node->n_list, &delayed_root->node_list))
		goto out;
	else
		p = node->n_list.next;

	next = list_entry(p, struct btrfs_delayed_node, n_list);
	atomic_inc(&next->refs);
out:
	spin_unlock(&delayed_root->lock);

	return next;
}

static void __btrfs_release_delayed_node(
				struct btrfs_delayed_node *delayed_node,
				int mod)
{
	struct btrfs_delayed_root *delayed_root;

	if (!delayed_node)
		return;

	delayed_root = delayed_node->root->fs_info->delayed_root;

	mutex_lock(&delayed_node->mutex);
	if (delayed_node->count)
		btrfs_queue_delayed_node(delayed_root, delayed_node, mod);
	else
		btrfs_dequeue_delayed_node(delayed_root, delayed_node);
	mutex_unlock(&delayed_node->mutex);

	if (atomic_dec_and_test(&delayed_node->refs)) {
		struct btrfs_root *root = delayed_node->root;
		spin_lock(&root->inode_lock);
		if (atomic_read(&delayed_node->refs) == 0) {
			radix_tree_delete(&root->delayed_nodes_tree,
					  delayed_node->inode_id);
			kmem_cache_free(delayed_node_cache, delayed_node);
		}
		spin_unlock(&root->inode_lock);
	}
}

static inline void btrfs_release_delayed_node(struct btrfs_delayed_node *node)
{
	__btrfs_release_delayed_node(node, 0);
}

struct btrfs_delayed_node *btrfs_first_prepared_delayed_node(
					struct btrfs_delayed_root *delayed_root)
{
	struct list_head *p;
	struct btrfs_delayed_node *node = NULL;

	spin_lock(&delayed_root->lock);
	if (list_empty(&delayed_root->prepare_list))
		goto out;

	p = delayed_root->prepare_list.next;
	list_del_init(p);
	node = list_entry(p, struct btrfs_delayed_node, p_list);
	atomic_inc(&node->refs);
out:
	spin_unlock(&delayed_root->lock);

	return node;
}

static inline void btrfs_release_prepared_delayed_node(
					struct btrfs_delayed_node *node)
{
	__btrfs_release_delayed_node(node, 1);
}

struct btrfs_delayed_item *btrfs_alloc_delayed_item(u32 data_len)
{
	struct btrfs_delayed_item *item;
	item = kmalloc(sizeof(*item) + data_len, GFP_NOFS);
	if (item) {
		item->data_len = data_len;
		item->ins_or_del = 0;
		item->bytes_reserved = 0;
		item->delayed_node = NULL;
		atomic_set(&item->refs, 1);
	}
	return item;
}

/*
 * __btrfs_lookup_delayed_item - look up the delayed item by key
 * @delayed_node: pointer to the delayed node
 * @key:	  the key to look up
 * @prev:	  used to store the prev item if the right item isn't found
 * @next:	  used to store the next item if the right item isn't found
 *
 * Note: if we don't find the right item, we will return the prev item and
 * the next item.
 */
static struct btrfs_delayed_item *__btrfs_lookup_delayed_item(
				struct rb_root *root,
				struct btrfs_key *key,
				struct btrfs_delayed_item **prev,
				struct btrfs_delayed_item **next)
{
	struct rb_node *node, *prev_node = NULL;
	struct btrfs_delayed_item *delayed_item = NULL;
	int ret = 0;

	node = root->rb_node;

	while (node) {
		delayed_item = rb_entry(node, struct btrfs_delayed_item,
					rb_node);
		prev_node = node;
		ret = btrfs_comp_cpu_keys(&delayed_item->key, key);
		if (ret < 0)
			node = node->rb_right;
		else if (ret > 0)
			node = node->rb_left;
		else
			return delayed_item;
	}

	if (prev) {
		if (!prev_node)
			*prev = NULL;
		else if (ret < 0)
			*prev = delayed_item;
		else if ((node = rb_prev(prev_node)) != NULL) {
			*prev = rb_entry(node, struct btrfs_delayed_item,
					 rb_node);
		} else
			*prev = NULL;
	}

	if (next) {
		if (!prev_node)
			*next = NULL;
		else if (ret > 0)
			*next = delayed_item;
		else if ((node = rb_next(prev_node)) != NULL) {
			*next = rb_entry(node, struct btrfs_delayed_item,
					 rb_node);
		} else
			*next = NULL;
	}
	return NULL;
}

struct btrfs_delayed_item *__btrfs_lookup_delayed_insertion_item(
					struct btrfs_delayed_node *delayed_node,
					struct btrfs_key *key)
{
	struct btrfs_delayed_item *item;

	item = __btrfs_lookup_delayed_item(&delayed_node->ins_root, key,
					   NULL, NULL);
	return item;
}

struct btrfs_delayed_item *__btrfs_lookup_delayed_deletion_item(
					struct btrfs_delayed_node *delayed_node,
					struct btrfs_key *key)
{
	struct btrfs_delayed_item *item;

	item = __btrfs_lookup_delayed_item(&delayed_node->del_root, key,
					   NULL, NULL);
	return item;
}

struct btrfs_delayed_item *__btrfs_search_delayed_insertion_item(
					struct btrfs_delayed_node *delayed_node,
					struct btrfs_key *key)
{
	struct btrfs_delayed_item *item, *next;

	item = __btrfs_lookup_delayed_item(&delayed_node->ins_root, key,
					   NULL, &next);
	if (!item)
		item = next;

	return item;
}

struct btrfs_delayed_item *__btrfs_search_delayed_deletion_item(
					struct btrfs_delayed_node *delayed_node,
					struct btrfs_key *key)
{
	struct btrfs_delayed_item *item, *next;

	item = __btrfs_lookup_delayed_item(&delayed_node->del_root, key,
					   NULL, &next);
	if (!item)
		item = next;

	return item;
}

static int __btrfs_add_delayed_item(struct btrfs_delayed_node *delayed_node,
				    struct btrfs_delayed_item *ins,
				    int action)
{
	struct rb_node **p, *node;
	struct rb_node *parent_node = NULL;
	struct rb_root *root;
	struct btrfs_delayed_item *item;
	int cmp;

	if (action == BTRFS_DELAYED_INSERTION_ITEM)
		root = &delayed_node->ins_root;
	else if (action == BTRFS_DELAYED_DELETION_ITEM)
		root = &delayed_node->del_root;
	else
		BUG();
	p = &root->rb_node;
	node = &ins->rb_node;

	while (*p) {
		parent_node = *p;
		item = rb_entry(parent_node, struct btrfs_delayed_item,
				 rb_node);

		cmp = btrfs_comp_cpu_keys(&item->key, &ins->key);
		if (cmp < 0)
			p = &(*p)->rb_right;
		else if (cmp > 0)
			p = &(*p)->rb_left;
		else
			return -EEXIST;
	}

	rb_link_node(node, parent_node, p);
	rb_insert_color(node, root);
	ins->delayed_node = delayed_node;
	ins->ins_or_del = action;

	if (ins->key.type == BTRFS_DIR_INDEX_KEY &&
	    action == BTRFS_DELAYED_INSERTION_ITEM &&
	    ins->key.offset >= delayed_node->index_cnt)
			delayed_node->index_cnt = ins->key.offset + 1;

	delayed_node->count++;
	atomic_inc(&delayed_node->root->fs_info->delayed_root->items);
	return 0;
}

static int __btrfs_add_delayed_insertion_item(struct btrfs_delayed_node *node,
					      struct btrfs_delayed_item *item)
{
	return __btrfs_add_delayed_item(node, item,
					BTRFS_DELAYED_INSERTION_ITEM);
}

static int __btrfs_add_delayed_deletion_item(struct btrfs_delayed_node *node,
					     struct btrfs_delayed_item *item)
{
	return __btrfs_add_delayed_item(node, item,
					BTRFS_DELAYED_DELETION_ITEM);
}

static void __btrfs_remove_delayed_item(struct btrfs_delayed_item *delayed_item)
{
	struct rb_root *root;
	struct btrfs_delayed_root *delayed_root;

	delayed_root = delayed_item->delayed_node->root->fs_info->delayed_root;

	BUG_ON(!delayed_root);
	BUG_ON(delayed_item->ins_or_del != BTRFS_DELAYED_DELETION_ITEM &&
	       delayed_item->ins_or_del != BTRFS_DELAYED_INSERTION_ITEM);

	if (delayed_item->ins_or_del == BTRFS_DELAYED_INSERTION_ITEM)
		root = &delayed_item->delayed_node->ins_root;
	else
		root = &delayed_item->delayed_node->del_root;

	rb_erase(&delayed_item->rb_node, root);
	delayed_item->delayed_node->count--;
	atomic_dec(&delayed_root->items);
	if (atomic_read(&delayed_root->items) < BTRFS_DELAYED_BACKGROUND &&
	    waitqueue_active(&delayed_root->wait))
		wake_up(&delayed_root->wait);
}

static void btrfs_release_delayed_item(struct btrfs_delayed_item *item)
{
	if (item) {
		__btrfs_remove_delayed_item(item);
		if (atomic_dec_and_test(&item->refs))
			kfree(item);
	}
}

struct btrfs_delayed_item *__btrfs_first_delayed_insertion_item(
					struct btrfs_delayed_node *delayed_node)
{
	struct rb_node *p;
	struct btrfs_delayed_item *item = NULL;

	p = rb_first(&delayed_node->ins_root);
	if (p)
		item = rb_entry(p, struct btrfs_delayed_item, rb_node);

	return item;
}

struct btrfs_delayed_item *__btrfs_first_delayed_deletion_item(
					struct btrfs_delayed_node *delayed_node)
{
	struct rb_node *p;
	struct btrfs_delayed_item *item = NULL;

	p = rb_first(&delayed_node->del_root);
	if (p)
		item = rb_entry(p, struct btrfs_delayed_item, rb_node);

	return item;
}

struct btrfs_delayed_item *__btrfs_next_delayed_item(
						struct btrfs_delayed_item *item)
{
	struct rb_node *p;
	struct btrfs_delayed_item *next = NULL;

	p = rb_next(&item->rb_node);
	if (p)
		next = rb_entry(p, struct btrfs_delayed_item, rb_node);

	return next;
}

static inline struct btrfs_root *btrfs_get_fs_root(struct btrfs_root *root,
						   u64 root_id)
{
	struct btrfs_key root_key;

	if (root->objectid == root_id)
		return root;

	root_key.objectid = root_id;
	root_key.type = BTRFS_ROOT_ITEM_KEY;
	root_key.offset = (u64)-1;
	return btrfs_read_fs_root_no_name(root->fs_info, &root_key);
}

static int btrfs_delayed_item_reserve_metadata(struct btrfs_trans_handle *trans,
					       struct btrfs_root *root,
					       struct btrfs_delayed_item *item)
{
	struct btrfs_block_rsv *src_rsv;
	struct btrfs_block_rsv *dst_rsv;
	u64 num_bytes;
	int ret;

	if (!trans->bytes_reserved)
		return 0;

	src_rsv = trans->block_rsv;
594
	dst_rsv = &root->fs_info->delayed_block_rsv;
595 596 597

	num_bytes = btrfs_calc_trans_metadata_size(root, 1);
	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
598
	if (!ret)
599 600 601 602 603 604 605 606
		item->bytes_reserved = num_bytes;

	return ret;
}

static void btrfs_delayed_item_release_metadata(struct btrfs_root *root,
						struct btrfs_delayed_item *item)
{
607 608
	struct btrfs_block_rsv *rsv;

609 610 611
	if (!item->bytes_reserved)
		return;

612
	rsv = &root->fs_info->delayed_block_rsv;
613
	btrfs_block_rsv_release(root, rsv,
614 615 616 617 618 619
				item->bytes_reserved);
}

static int btrfs_delayed_inode_reserve_metadata(
					struct btrfs_trans_handle *trans,
					struct btrfs_root *root,
620
					struct inode *inode,
621 622 623 624 625 626
					struct btrfs_delayed_node *node)
{
	struct btrfs_block_rsv *src_rsv;
	struct btrfs_block_rsv *dst_rsv;
	u64 num_bytes;
	int ret;
627
	int release = false;
628 629

	src_rsv = trans->block_rsv;
630
	dst_rsv = &root->fs_info->delayed_block_rsv;
631 632

	num_bytes = btrfs_calc_trans_metadata_size(root, 1);
633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656

	/*
	 * btrfs_dirty_inode will update the inode under btrfs_join_transaction
	 * which doesn't reserve space for speed.  This is a problem since we
	 * still need to reserve space for this update, so try to reserve the
	 * space.
	 *
	 * Now if src_rsv == delalloc_block_rsv we'll let it just steal since
	 * we're accounted for.
	 */
	if (!trans->bytes_reserved &&
	    src_rsv != &root->fs_info->delalloc_block_rsv) {
		ret = btrfs_block_rsv_add_noflush(root, dst_rsv, num_bytes);
		/*
		 * Since we're under a transaction reserve_metadata_bytes could
		 * try to commit the transaction which will make it return
		 * EAGAIN to make us stop the transaction we have, so return
		 * ENOSPC instead so that btrfs_dirty_inode knows what to do.
		 */
		if (ret == -EAGAIN)
			ret = -ENOSPC;
		if (!ret)
			node->bytes_reserved = num_bytes;
		return ret;
657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690
	} else if (src_rsv == &root->fs_info->delalloc_block_rsv) {
		spin_lock(&BTRFS_I(inode)->lock);
		if (BTRFS_I(inode)->delalloc_meta_reserved) {
			BTRFS_I(inode)->delalloc_meta_reserved = 0;
			spin_unlock(&BTRFS_I(inode)->lock);
			release = true;
			goto migrate;
		}
		spin_unlock(&BTRFS_I(inode)->lock);

		/* Ok we didn't have space pre-reserved.  This shouldn't happen
		 * too often but it can happen if we do delalloc to an existing
		 * inode which gets dirtied because of the time update, and then
		 * isn't touched again until after the transaction commits and
		 * then we try to write out the data.  First try to be nice and
		 * reserve something strictly for us.  If not be a pain and try
		 * to steal from the delalloc block rsv.
		 */
		ret = btrfs_block_rsv_add_noflush(root, dst_rsv, num_bytes);
		if (!ret)
			goto out;

		ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
		if (!ret)
			goto out;

		/*
		 * Ok this is a problem, let's just steal from the global rsv
		 * since this really shouldn't happen that often.
		 */
		WARN_ON(1);
		ret = btrfs_block_rsv_migrate(&root->fs_info->global_block_rsv,
					      dst_rsv, num_bytes);
		goto out;
691 692
	}

693
migrate:
694
	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
695 696 697 698 699 700 701 702 703 704 705 706 707 708 709

out:
	/*
	 * Migrate only takes a reservation, it doesn't touch the size of the
	 * block_rsv.  This is to simplify people who don't normally have things
	 * migrated from their block rsv.  If they go to release their
	 * reservation, that will decrease the size as well, so if migrate
	 * reduced size we'd end up with a negative size.  But for the
	 * delalloc_meta_reserved stuff we will only know to drop 1 reservation,
	 * but we could in fact do this reserve/migrate dance several times
	 * between the time we did the original reservation and we'd clean it
	 * up.  So to take care of this, release the space for the meta
	 * reservation here.  I think it may be time for a documentation page on
	 * how block rsvs. work.
	 */
710 711 712
	if (!ret)
		node->bytes_reserved = num_bytes;

713 714
	if (release)
		btrfs_block_rsv_release(root, src_rsv, num_bytes);
715 716 717 718 719 720 721 722 723 724 725 726

	return ret;
}

static void btrfs_delayed_inode_release_metadata(struct btrfs_root *root,
						struct btrfs_delayed_node *node)
{
	struct btrfs_block_rsv *rsv;

	if (!node->bytes_reserved)
		return;

727
	rsv = &root->fs_info->delayed_block_rsv;
728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761
	btrfs_block_rsv_release(root, rsv,
				node->bytes_reserved);
	node->bytes_reserved = 0;
}

/*
 * This helper will insert some continuous items into the same leaf according
 * to the free space of the leaf.
 */
static int btrfs_batch_insert_items(struct btrfs_trans_handle *trans,
				struct btrfs_root *root,
				struct btrfs_path *path,
				struct btrfs_delayed_item *item)
{
	struct btrfs_delayed_item *curr, *next;
	int free_space;
	int total_data_size = 0, total_size = 0;
	struct extent_buffer *leaf;
	char *data_ptr;
	struct btrfs_key *keys;
	u32 *data_size;
	struct list_head head;
	int slot;
	int nitems;
	int i;
	int ret = 0;

	BUG_ON(!path->nodes[0]);

	leaf = path->nodes[0];
	free_space = btrfs_leaf_free_space(root, leaf);
	INIT_LIST_HEAD(&head);

	next = item;
762
	nitems = 0;
763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815

	/*
	 * count the number of the continuous items that we can insert in batch
	 */
	while (total_size + next->data_len + sizeof(struct btrfs_item) <=
	       free_space) {
		total_data_size += next->data_len;
		total_size += next->data_len + sizeof(struct btrfs_item);
		list_add_tail(&next->tree_list, &head);
		nitems++;

		curr = next;
		next = __btrfs_next_delayed_item(curr);
		if (!next)
			break;

		if (!btrfs_is_continuous_delayed_item(curr, next))
			break;
	}

	if (!nitems) {
		ret = 0;
		goto out;
	}

	/*
	 * we need allocate some memory space, but it might cause the task
	 * to sleep, so we set all locked nodes in the path to blocking locks
	 * first.
	 */
	btrfs_set_path_blocking(path);

	keys = kmalloc(sizeof(struct btrfs_key) * nitems, GFP_NOFS);
	if (!keys) {
		ret = -ENOMEM;
		goto out;
	}

	data_size = kmalloc(sizeof(u32) * nitems, GFP_NOFS);
	if (!data_size) {
		ret = -ENOMEM;
		goto error;
	}

	/* get keys of all the delayed items */
	i = 0;
	list_for_each_entry(next, &head, tree_list) {
		keys[i] = next->key;
		data_size[i] = next->data_len;
		i++;
	}

	/* reset all the locked nodes in the patch to spinning locks. */
816
	btrfs_clear_path_blocking(path, NULL, 0);
817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897

	/* insert the keys of the items */
	ret = setup_items_for_insert(trans, root, path, keys, data_size,
				     total_data_size, total_size, nitems);
	if (ret)
		goto error;

	/* insert the dir index items */
	slot = path->slots[0];
	list_for_each_entry_safe(curr, next, &head, tree_list) {
		data_ptr = btrfs_item_ptr(leaf, slot, char);
		write_extent_buffer(leaf, &curr->data,
				    (unsigned long)data_ptr,
				    curr->data_len);
		slot++;

		btrfs_delayed_item_release_metadata(root, curr);

		list_del(&curr->tree_list);
		btrfs_release_delayed_item(curr);
	}

error:
	kfree(data_size);
	kfree(keys);
out:
	return ret;
}

/*
 * This helper can just do simple insertion that needn't extend item for new
 * data, such as directory name index insertion, inode insertion.
 */
static int btrfs_insert_delayed_item(struct btrfs_trans_handle *trans,
				     struct btrfs_root *root,
				     struct btrfs_path *path,
				     struct btrfs_delayed_item *delayed_item)
{
	struct extent_buffer *leaf;
	struct btrfs_item *item;
	char *ptr;
	int ret;

	ret = btrfs_insert_empty_item(trans, root, path, &delayed_item->key,
				      delayed_item->data_len);
	if (ret < 0 && ret != -EEXIST)
		return ret;

	leaf = path->nodes[0];

	item = btrfs_item_nr(leaf, path->slots[0]);
	ptr = btrfs_item_ptr(leaf, path->slots[0], char);

	write_extent_buffer(leaf, delayed_item->data, (unsigned long)ptr,
			    delayed_item->data_len);
	btrfs_mark_buffer_dirty(leaf);

	btrfs_delayed_item_release_metadata(root, delayed_item);
	return 0;
}

/*
 * we insert an item first, then if there are some continuous items, we try
 * to insert those items into the same leaf.
 */
static int btrfs_insert_delayed_items(struct btrfs_trans_handle *trans,
				      struct btrfs_path *path,
				      struct btrfs_root *root,
				      struct btrfs_delayed_node *node)
{
	struct btrfs_delayed_item *curr, *prev;
	int ret = 0;

do_again:
	mutex_lock(&node->mutex);
	curr = __btrfs_first_delayed_insertion_item(node);
	if (!curr)
		goto insert_end;

	ret = btrfs_insert_delayed_item(trans, root, path, curr);
	if (ret < 0) {
898
		btrfs_release_path(path);
899 900 901 902 903 904 905 906 907 908 909 910 911
		goto insert_end;
	}

	prev = curr;
	curr = __btrfs_next_delayed_item(prev);
	if (curr && btrfs_is_continuous_delayed_item(prev, curr)) {
		/* insert the continuous items into the same leaf */
		path->slots[0]++;
		btrfs_batch_insert_items(trans, root, path, curr);
	}
	btrfs_release_delayed_item(prev);
	btrfs_mark_buffer_dirty(path->nodes[0]);

912
	btrfs_release_path(path);
913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009
	mutex_unlock(&node->mutex);
	goto do_again;

insert_end:
	mutex_unlock(&node->mutex);
	return ret;
}

static int btrfs_batch_delete_items(struct btrfs_trans_handle *trans,
				    struct btrfs_root *root,
				    struct btrfs_path *path,
				    struct btrfs_delayed_item *item)
{
	struct btrfs_delayed_item *curr, *next;
	struct extent_buffer *leaf;
	struct btrfs_key key;
	struct list_head head;
	int nitems, i, last_item;
	int ret = 0;

	BUG_ON(!path->nodes[0]);

	leaf = path->nodes[0];

	i = path->slots[0];
	last_item = btrfs_header_nritems(leaf) - 1;
	if (i > last_item)
		return -ENOENT;	/* FIXME: Is errno suitable? */

	next = item;
	INIT_LIST_HEAD(&head);
	btrfs_item_key_to_cpu(leaf, &key, i);
	nitems = 0;
	/*
	 * count the number of the dir index items that we can delete in batch
	 */
	while (btrfs_comp_cpu_keys(&next->key, &key) == 0) {
		list_add_tail(&next->tree_list, &head);
		nitems++;

		curr = next;
		next = __btrfs_next_delayed_item(curr);
		if (!next)
			break;

		if (!btrfs_is_continuous_delayed_item(curr, next))
			break;

		i++;
		if (i > last_item)
			break;
		btrfs_item_key_to_cpu(leaf, &key, i);
	}

	if (!nitems)
		return 0;

	ret = btrfs_del_items(trans, root, path, path->slots[0], nitems);
	if (ret)
		goto out;

	list_for_each_entry_safe(curr, next, &head, tree_list) {
		btrfs_delayed_item_release_metadata(root, curr);
		list_del(&curr->tree_list);
		btrfs_release_delayed_item(curr);
	}

out:
	return ret;
}

static int btrfs_delete_delayed_items(struct btrfs_trans_handle *trans,
				      struct btrfs_path *path,
				      struct btrfs_root *root,
				      struct btrfs_delayed_node *node)
{
	struct btrfs_delayed_item *curr, *prev;
	int ret = 0;

do_again:
	mutex_lock(&node->mutex);
	curr = __btrfs_first_delayed_deletion_item(node);
	if (!curr)
		goto delete_fail;

	ret = btrfs_search_slot(trans, root, &curr->key, path, -1, 1);
	if (ret < 0)
		goto delete_fail;
	else if (ret > 0) {
		/*
		 * can't find the item which the node points to, so this node
		 * is invalid, just drop it.
		 */
		prev = curr;
		curr = __btrfs_next_delayed_item(prev);
		btrfs_release_delayed_item(prev);
		ret = 0;
1010
		btrfs_release_path(path);
1011 1012 1013 1014 1015 1016 1017
		if (curr)
			goto do_again;
		else
			goto delete_fail;
	}

	btrfs_batch_delete_items(trans, root, path, curr);
1018
	btrfs_release_path(path);
1019 1020 1021 1022
	mutex_unlock(&node->mutex);
	goto do_again;

delete_fail:
1023
	btrfs_release_path(path);
1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066
	mutex_unlock(&node->mutex);
	return ret;
}

static void btrfs_release_delayed_inode(struct btrfs_delayed_node *delayed_node)
{
	struct btrfs_delayed_root *delayed_root;

	if (delayed_node && delayed_node->inode_dirty) {
		BUG_ON(!delayed_node->root);
		delayed_node->inode_dirty = 0;
		delayed_node->count--;

		delayed_root = delayed_node->root->fs_info->delayed_root;
		atomic_dec(&delayed_root->items);
		if (atomic_read(&delayed_root->items) <
		    BTRFS_DELAYED_BACKGROUND &&
		    waitqueue_active(&delayed_root->wait))
			wake_up(&delayed_root->wait);
	}
}

static int btrfs_update_delayed_inode(struct btrfs_trans_handle *trans,
				      struct btrfs_root *root,
				      struct btrfs_path *path,
				      struct btrfs_delayed_node *node)
{
	struct btrfs_key key;
	struct btrfs_inode_item *inode_item;
	struct extent_buffer *leaf;
	int ret;

	mutex_lock(&node->mutex);
	if (!node->inode_dirty) {
		mutex_unlock(&node->mutex);
		return 0;
	}

	key.objectid = node->inode_id;
	btrfs_set_key_type(&key, BTRFS_INODE_ITEM_KEY);
	key.offset = 0;
	ret = btrfs_lookup_inode(trans, root, path, &key, 1);
	if (ret > 0) {
1067
		btrfs_release_path(path);
1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081
		mutex_unlock(&node->mutex);
		return -ENOENT;
	} else if (ret < 0) {
		mutex_unlock(&node->mutex);
		return ret;
	}

	btrfs_unlock_up_safe(path, 1);
	leaf = path->nodes[0];
	inode_item = btrfs_item_ptr(leaf, path->slots[0],
				    struct btrfs_inode_item);
	write_extent_buffer(leaf, &node->inode_item, (unsigned long)inode_item,
			    sizeof(struct btrfs_inode_item));
	btrfs_mark_buffer_dirty(leaf);
1082
	btrfs_release_path(path);
1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097

	btrfs_delayed_inode_release_metadata(root, node);
	btrfs_release_delayed_inode(node);
	mutex_unlock(&node->mutex);

	return 0;
}

/* Called when committing the transaction. */
int btrfs_run_delayed_items(struct btrfs_trans_handle *trans,
			    struct btrfs_root *root)
{
	struct btrfs_delayed_root *delayed_root;
	struct btrfs_delayed_node *curr_node, *prev_node;
	struct btrfs_path *path;
1098
	struct btrfs_block_rsv *block_rsv;
1099 1100 1101 1102 1103 1104 1105
	int ret = 0;

	path = btrfs_alloc_path();
	if (!path)
		return -ENOMEM;
	path->leave_spinning = 1;

1106
	block_rsv = trans->block_rsv;
1107
	trans->block_rsv = &root->fs_info->delayed_block_rsv;
1108

1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132
	delayed_root = btrfs_get_delayed_root(root);

	curr_node = btrfs_first_delayed_node(delayed_root);
	while (curr_node) {
		root = curr_node->root;
		ret = btrfs_insert_delayed_items(trans, path, root,
						 curr_node);
		if (!ret)
			ret = btrfs_delete_delayed_items(trans, path, root,
							 curr_node);
		if (!ret)
			ret = btrfs_update_delayed_inode(trans, root, path,
							 curr_node);
		if (ret) {
			btrfs_release_delayed_node(curr_node);
			break;
		}

		prev_node = curr_node;
		curr_node = btrfs_next_delayed_node(curr_node);
		btrfs_release_delayed_node(prev_node);
	}

	btrfs_free_path(path);
1133
	trans->block_rsv = block_rsv;
1134 1135 1136 1137 1138 1139 1140
	return ret;
}

static int __btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
					      struct btrfs_delayed_node *node)
{
	struct btrfs_path *path;
1141
	struct btrfs_block_rsv *block_rsv;
1142 1143 1144 1145 1146 1147 1148
	int ret;

	path = btrfs_alloc_path();
	if (!path)
		return -ENOMEM;
	path->leave_spinning = 1;

1149
	block_rsv = trans->block_rsv;
1150
	trans->block_rsv = &node->root->fs_info->delayed_block_rsv;
1151

1152 1153 1154 1155 1156 1157 1158
	ret = btrfs_insert_delayed_items(trans, path, node->root, node);
	if (!ret)
		ret = btrfs_delete_delayed_items(trans, path, node->root, node);
	if (!ret)
		ret = btrfs_update_delayed_inode(trans, node->root, path, node);
	btrfs_free_path(path);

1159
	trans->block_rsv = block_rsv;
1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209
	return ret;
}

int btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
				     struct inode *inode)
{
	struct btrfs_delayed_node *delayed_node = btrfs_get_delayed_node(inode);
	int ret;

	if (!delayed_node)
		return 0;

	mutex_lock(&delayed_node->mutex);
	if (!delayed_node->count) {
		mutex_unlock(&delayed_node->mutex);
		btrfs_release_delayed_node(delayed_node);
		return 0;
	}
	mutex_unlock(&delayed_node->mutex);

	ret = __btrfs_commit_inode_delayed_items(trans, delayed_node);
	btrfs_release_delayed_node(delayed_node);
	return ret;
}

void btrfs_remove_delayed_node(struct inode *inode)
{
	struct btrfs_delayed_node *delayed_node;

	delayed_node = ACCESS_ONCE(BTRFS_I(inode)->delayed_node);
	if (!delayed_node)
		return;

	BTRFS_I(inode)->delayed_node = NULL;
	btrfs_release_delayed_node(delayed_node);
}

struct btrfs_async_delayed_node {
	struct btrfs_root *root;
	struct btrfs_delayed_node *delayed_node;
	struct btrfs_work work;
};

static void btrfs_async_run_delayed_node_done(struct btrfs_work *work)
{
	struct btrfs_async_delayed_node *async_node;
	struct btrfs_trans_handle *trans;
	struct btrfs_path *path;
	struct btrfs_delayed_node *delayed_node = NULL;
	struct btrfs_root *root;
1210
	struct btrfs_block_rsv *block_rsv;
1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224
	unsigned long nr = 0;
	int need_requeue = 0;
	int ret;

	async_node = container_of(work, struct btrfs_async_delayed_node, work);

	path = btrfs_alloc_path();
	if (!path)
		goto out;
	path->leave_spinning = 1;

	delayed_node = async_node->delayed_node;
	root = delayed_node->root;

C
Chris Mason 已提交
1225
	trans = btrfs_join_transaction(root);
1226 1227 1228
	if (IS_ERR(trans))
		goto free_path;

1229
	block_rsv = trans->block_rsv;
1230
	trans->block_rsv = &root->fs_info->delayed_block_rsv;
1231

1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273
	ret = btrfs_insert_delayed_items(trans, path, root, delayed_node);
	if (!ret)
		ret = btrfs_delete_delayed_items(trans, path, root,
						 delayed_node);

	if (!ret)
		btrfs_update_delayed_inode(trans, root, path, delayed_node);

	/*
	 * Maybe new delayed items have been inserted, so we need requeue
	 * the work. Besides that, we must dequeue the empty delayed nodes
	 * to avoid the race between delayed items balance and the worker.
	 * The race like this:
	 * 	Task1				Worker thread
	 * 					count == 0, needn't requeue
	 * 					  also needn't insert the
	 * 					  delayed node into prepare
	 * 					  list again.
	 * 	add lots of delayed items
	 * 	queue the delayed node
	 * 	  already in the list,
	 * 	  and not in the prepare
	 * 	  list, it means the delayed
	 * 	  node is being dealt with
	 * 	  by the worker.
	 * 	do delayed items balance
	 * 	  the delayed node is being
	 * 	  dealt with by the worker
	 * 	  now, just wait.
	 * 	  				the worker goto idle.
	 * Task1 will sleep until the transaction is commited.
	 */
	mutex_lock(&delayed_node->mutex);
	if (delayed_node->count)
		need_requeue = 1;
	else
		btrfs_dequeue_delayed_node(root->fs_info->delayed_root,
					   delayed_node);
	mutex_unlock(&delayed_node->mutex);

	nr = trans->blocks_used;

1274
	trans->block_rsv = block_rsv;
1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320
	btrfs_end_transaction_dmeta(trans, root);
	__btrfs_btree_balance_dirty(root, nr);
free_path:
	btrfs_free_path(path);
out:
	if (need_requeue)
		btrfs_requeue_work(&async_node->work);
	else {
		btrfs_release_prepared_delayed_node(delayed_node);
		kfree(async_node);
	}
}

static int btrfs_wq_run_delayed_node(struct btrfs_delayed_root *delayed_root,
				     struct btrfs_root *root, int all)
{
	struct btrfs_async_delayed_node *async_node;
	struct btrfs_delayed_node *curr;
	int count = 0;

again:
	curr = btrfs_first_prepared_delayed_node(delayed_root);
	if (!curr)
		return 0;

	async_node = kmalloc(sizeof(*async_node), GFP_NOFS);
	if (!async_node) {
		btrfs_release_prepared_delayed_node(curr);
		return -ENOMEM;
	}

	async_node->root = root;
	async_node->delayed_node = curr;

	async_node->work.func = btrfs_async_run_delayed_node_done;
	async_node->work.flags = 0;

	btrfs_queue_worker(&root->fs_info->delayed_workers, &async_node->work);
	count++;

	if (all || count < 4)
		goto again;

	return 0;
}

1321 1322 1323 1324 1325 1326 1327
void btrfs_assert_delayed_root_empty(struct btrfs_root *root)
{
	struct btrfs_delayed_root *delayed_root;
	delayed_root = btrfs_get_delayed_root(root);
	WARN_ON(btrfs_first_delayed_node(delayed_root));
}

1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381
void btrfs_balance_delayed_items(struct btrfs_root *root)
{
	struct btrfs_delayed_root *delayed_root;

	delayed_root = btrfs_get_delayed_root(root);

	if (atomic_read(&delayed_root->items) < BTRFS_DELAYED_BACKGROUND)
		return;

	if (atomic_read(&delayed_root->items) >= BTRFS_DELAYED_WRITEBACK) {
		int ret;
		ret = btrfs_wq_run_delayed_node(delayed_root, root, 1);
		if (ret)
			return;

		wait_event_interruptible_timeout(
				delayed_root->wait,
				(atomic_read(&delayed_root->items) <
				 BTRFS_DELAYED_BACKGROUND),
				HZ);
		return;
	}

	btrfs_wq_run_delayed_node(delayed_root, root, 0);
}

int btrfs_insert_delayed_dir_index(struct btrfs_trans_handle *trans,
				   struct btrfs_root *root, const char *name,
				   int name_len, struct inode *dir,
				   struct btrfs_disk_key *disk_key, u8 type,
				   u64 index)
{
	struct btrfs_delayed_node *delayed_node;
	struct btrfs_delayed_item *delayed_item;
	struct btrfs_dir_item *dir_item;
	int ret;

	delayed_node = btrfs_get_or_create_delayed_node(dir);
	if (IS_ERR(delayed_node))
		return PTR_ERR(delayed_node);

	delayed_item = btrfs_alloc_delayed_item(sizeof(*dir_item) + name_len);
	if (!delayed_item) {
		ret = -ENOMEM;
		goto release_node;
	}

	ret = btrfs_delayed_item_reserve_metadata(trans, root, delayed_item);
	/*
	 * we have reserved enough space when we start a new transaction,
	 * so reserving metadata failure is impossible
	 */
	BUG_ON(ret);

1382
	delayed_item->key.objectid = btrfs_ino(dir);
1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444
	btrfs_set_key_type(&delayed_item->key, BTRFS_DIR_INDEX_KEY);
	delayed_item->key.offset = index;

	dir_item = (struct btrfs_dir_item *)delayed_item->data;
	dir_item->location = *disk_key;
	dir_item->transid = cpu_to_le64(trans->transid);
	dir_item->data_len = 0;
	dir_item->name_len = cpu_to_le16(name_len);
	dir_item->type = type;
	memcpy((char *)(dir_item + 1), name, name_len);

	mutex_lock(&delayed_node->mutex);
	ret = __btrfs_add_delayed_insertion_item(delayed_node, delayed_item);
	if (unlikely(ret)) {
		printk(KERN_ERR "err add delayed dir index item(name: %s) into "
				"the insertion tree of the delayed node"
				"(root id: %llu, inode id: %llu, errno: %d)\n",
				name,
				(unsigned long long)delayed_node->root->objectid,
				(unsigned long long)delayed_node->inode_id,
				ret);
		BUG();
	}
	mutex_unlock(&delayed_node->mutex);

release_node:
	btrfs_release_delayed_node(delayed_node);
	return ret;
}

static int btrfs_delete_delayed_insertion_item(struct btrfs_root *root,
					       struct btrfs_delayed_node *node,
					       struct btrfs_key *key)
{
	struct btrfs_delayed_item *item;

	mutex_lock(&node->mutex);
	item = __btrfs_lookup_delayed_insertion_item(node, key);
	if (!item) {
		mutex_unlock(&node->mutex);
		return 1;
	}

	btrfs_delayed_item_release_metadata(root, item);
	btrfs_release_delayed_item(item);
	mutex_unlock(&node->mutex);
	return 0;
}

int btrfs_delete_delayed_dir_index(struct btrfs_trans_handle *trans,
				   struct btrfs_root *root, struct inode *dir,
				   u64 index)
{
	struct btrfs_delayed_node *node;
	struct btrfs_delayed_item *item;
	struct btrfs_key item_key;
	int ret;

	node = btrfs_get_or_create_delayed_node(dir);
	if (IS_ERR(node))
		return PTR_ERR(node);

1445
	item_key.objectid = btrfs_ino(dir);
1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487
	btrfs_set_key_type(&item_key, BTRFS_DIR_INDEX_KEY);
	item_key.offset = index;

	ret = btrfs_delete_delayed_insertion_item(root, node, &item_key);
	if (!ret)
		goto end;

	item = btrfs_alloc_delayed_item(0);
	if (!item) {
		ret = -ENOMEM;
		goto end;
	}

	item->key = item_key;

	ret = btrfs_delayed_item_reserve_metadata(trans, root, item);
	/*
	 * we have reserved enough space when we start a new transaction,
	 * so reserving metadata failure is impossible.
	 */
	BUG_ON(ret);

	mutex_lock(&node->mutex);
	ret = __btrfs_add_delayed_deletion_item(node, item);
	if (unlikely(ret)) {
		printk(KERN_ERR "err add delayed dir index item(index: %llu) "
				"into the deletion tree of the delayed node"
				"(root id: %llu, inode id: %llu, errno: %d)\n",
				(unsigned long long)index,
				(unsigned long long)node->root->objectid,
				(unsigned long long)node->inode_id,
				ret);
		BUG();
	}
	mutex_unlock(&node->mutex);
end:
	btrfs_release_delayed_node(node);
	return ret;
}

int btrfs_inode_delayed_dir_index_count(struct inode *inode)
{
1488
	struct btrfs_delayed_node *delayed_node = btrfs_get_delayed_node(inode);
1489 1490 1491 1492 1493 1494 1495 1496 1497

	if (!delayed_node)
		return -ENOENT;

	/*
	 * Since we have held i_mutex of this directory, it is impossible that
	 * a new directory index is added into the delayed node and index_cnt
	 * is updated now. So we needn't lock the delayed node.
	 */
1498 1499
	if (!delayed_node->index_cnt) {
		btrfs_release_delayed_node(delayed_node);
1500
		return -EINVAL;
1501
	}
1502 1503

	BTRFS_I(inode)->index_cnt = delayed_node->index_cnt;
1504 1505
	btrfs_release_delayed_node(delayed_node);
	return 0;
1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680
}

void btrfs_get_delayed_items(struct inode *inode, struct list_head *ins_list,
			     struct list_head *del_list)
{
	struct btrfs_delayed_node *delayed_node;
	struct btrfs_delayed_item *item;

	delayed_node = btrfs_get_delayed_node(inode);
	if (!delayed_node)
		return;

	mutex_lock(&delayed_node->mutex);
	item = __btrfs_first_delayed_insertion_item(delayed_node);
	while (item) {
		atomic_inc(&item->refs);
		list_add_tail(&item->readdir_list, ins_list);
		item = __btrfs_next_delayed_item(item);
	}

	item = __btrfs_first_delayed_deletion_item(delayed_node);
	while (item) {
		atomic_inc(&item->refs);
		list_add_tail(&item->readdir_list, del_list);
		item = __btrfs_next_delayed_item(item);
	}
	mutex_unlock(&delayed_node->mutex);
	/*
	 * This delayed node is still cached in the btrfs inode, so refs
	 * must be > 1 now, and we needn't check it is going to be freed
	 * or not.
	 *
	 * Besides that, this function is used to read dir, we do not
	 * insert/delete delayed items in this period. So we also needn't
	 * requeue or dequeue this delayed node.
	 */
	atomic_dec(&delayed_node->refs);
}

void btrfs_put_delayed_items(struct list_head *ins_list,
			     struct list_head *del_list)
{
	struct btrfs_delayed_item *curr, *next;

	list_for_each_entry_safe(curr, next, ins_list, readdir_list) {
		list_del(&curr->readdir_list);
		if (atomic_dec_and_test(&curr->refs))
			kfree(curr);
	}

	list_for_each_entry_safe(curr, next, del_list, readdir_list) {
		list_del(&curr->readdir_list);
		if (atomic_dec_and_test(&curr->refs))
			kfree(curr);
	}
}

int btrfs_should_delete_dir_index(struct list_head *del_list,
				  u64 index)
{
	struct btrfs_delayed_item *curr, *next;
	int ret;

	if (list_empty(del_list))
		return 0;

	list_for_each_entry_safe(curr, next, del_list, readdir_list) {
		if (curr->key.offset > index)
			break;

		list_del(&curr->readdir_list);
		ret = (curr->key.offset == index);

		if (atomic_dec_and_test(&curr->refs))
			kfree(curr);

		if (ret)
			return 1;
		else
			continue;
	}
	return 0;
}

/*
 * btrfs_readdir_delayed_dir_index - read dir info stored in the delayed tree
 *
 */
int btrfs_readdir_delayed_dir_index(struct file *filp, void *dirent,
				    filldir_t filldir,
				    struct list_head *ins_list)
{
	struct btrfs_dir_item *di;
	struct btrfs_delayed_item *curr, *next;
	struct btrfs_key location;
	char *name;
	int name_len;
	int over = 0;
	unsigned char d_type;

	if (list_empty(ins_list))
		return 0;

	/*
	 * Changing the data of the delayed item is impossible. So
	 * we needn't lock them. And we have held i_mutex of the
	 * directory, nobody can delete any directory indexes now.
	 */
	list_for_each_entry_safe(curr, next, ins_list, readdir_list) {
		list_del(&curr->readdir_list);

		if (curr->key.offset < filp->f_pos) {
			if (atomic_dec_and_test(&curr->refs))
				kfree(curr);
			continue;
		}

		filp->f_pos = curr->key.offset;

		di = (struct btrfs_dir_item *)curr->data;
		name = (char *)(di + 1);
		name_len = le16_to_cpu(di->name_len);

		d_type = btrfs_filetype_table[di->type];
		btrfs_disk_key_to_cpu(&location, &di->location);

		over = filldir(dirent, name, name_len, curr->key.offset,
			       location.objectid, d_type);

		if (atomic_dec_and_test(&curr->refs))
			kfree(curr);

		if (over)
			return 1;
	}
	return 0;
}

BTRFS_SETGET_STACK_FUNCS(stack_inode_generation, struct btrfs_inode_item,
			 generation, 64);
BTRFS_SETGET_STACK_FUNCS(stack_inode_sequence, struct btrfs_inode_item,
			 sequence, 64);
BTRFS_SETGET_STACK_FUNCS(stack_inode_transid, struct btrfs_inode_item,
			 transid, 64);
BTRFS_SETGET_STACK_FUNCS(stack_inode_size, struct btrfs_inode_item, size, 64);
BTRFS_SETGET_STACK_FUNCS(stack_inode_nbytes, struct btrfs_inode_item,
			 nbytes, 64);
BTRFS_SETGET_STACK_FUNCS(stack_inode_block_group, struct btrfs_inode_item,
			 block_group, 64);
BTRFS_SETGET_STACK_FUNCS(stack_inode_nlink, struct btrfs_inode_item, nlink, 32);
BTRFS_SETGET_STACK_FUNCS(stack_inode_uid, struct btrfs_inode_item, uid, 32);
BTRFS_SETGET_STACK_FUNCS(stack_inode_gid, struct btrfs_inode_item, gid, 32);
BTRFS_SETGET_STACK_FUNCS(stack_inode_mode, struct btrfs_inode_item, mode, 32);
BTRFS_SETGET_STACK_FUNCS(stack_inode_rdev, struct btrfs_inode_item, rdev, 64);
BTRFS_SETGET_STACK_FUNCS(stack_inode_flags, struct btrfs_inode_item, flags, 64);

BTRFS_SETGET_STACK_FUNCS(stack_timespec_sec, struct btrfs_timespec, sec, 64);
BTRFS_SETGET_STACK_FUNCS(stack_timespec_nsec, struct btrfs_timespec, nsec, 32);

static void fill_stack_inode_item(struct btrfs_trans_handle *trans,
				  struct btrfs_inode_item *inode_item,
				  struct inode *inode)
{
	btrfs_set_stack_inode_uid(inode_item, inode->i_uid);
	btrfs_set_stack_inode_gid(inode_item, inode->i_gid);
	btrfs_set_stack_inode_size(inode_item, BTRFS_I(inode)->disk_i_size);
	btrfs_set_stack_inode_mode(inode_item, inode->i_mode);
	btrfs_set_stack_inode_nlink(inode_item, inode->i_nlink);
	btrfs_set_stack_inode_nbytes(inode_item, inode_get_bytes(inode));
	btrfs_set_stack_inode_generation(inode_item,
					 BTRFS_I(inode)->generation);
	btrfs_set_stack_inode_sequence(inode_item, BTRFS_I(inode)->sequence);
	btrfs_set_stack_inode_transid(inode_item, trans->transid);
	btrfs_set_stack_inode_rdev(inode_item, inode->i_rdev);
	btrfs_set_stack_inode_flags(inode_item, BTRFS_I(inode)->flags);
C
Chris Mason 已提交
1681
	btrfs_set_stack_inode_block_group(inode_item, 0);
1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698

	btrfs_set_stack_timespec_sec(btrfs_inode_atime(inode_item),
				     inode->i_atime.tv_sec);
	btrfs_set_stack_timespec_nsec(btrfs_inode_atime(inode_item),
				      inode->i_atime.tv_nsec);

	btrfs_set_stack_timespec_sec(btrfs_inode_mtime(inode_item),
				     inode->i_mtime.tv_sec);
	btrfs_set_stack_timespec_nsec(btrfs_inode_mtime(inode_item),
				      inode->i_mtime.tv_nsec);

	btrfs_set_stack_timespec_sec(btrfs_inode_ctime(inode_item),
				     inode->i_ctime.tv_sec);
	btrfs_set_stack_timespec_nsec(btrfs_inode_ctime(inode_item),
				      inode->i_ctime.tv_nsec);
}

1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749
int btrfs_fill_inode(struct inode *inode, u32 *rdev)
{
	struct btrfs_delayed_node *delayed_node;
	struct btrfs_inode_item *inode_item;
	struct btrfs_timespec *tspec;

	delayed_node = btrfs_get_delayed_node(inode);
	if (!delayed_node)
		return -ENOENT;

	mutex_lock(&delayed_node->mutex);
	if (!delayed_node->inode_dirty) {
		mutex_unlock(&delayed_node->mutex);
		btrfs_release_delayed_node(delayed_node);
		return -ENOENT;
	}

	inode_item = &delayed_node->inode_item;

	inode->i_uid = btrfs_stack_inode_uid(inode_item);
	inode->i_gid = btrfs_stack_inode_gid(inode_item);
	btrfs_i_size_write(inode, btrfs_stack_inode_size(inode_item));
	inode->i_mode = btrfs_stack_inode_mode(inode_item);
	inode->i_nlink = btrfs_stack_inode_nlink(inode_item);
	inode_set_bytes(inode, btrfs_stack_inode_nbytes(inode_item));
	BTRFS_I(inode)->generation = btrfs_stack_inode_generation(inode_item);
	BTRFS_I(inode)->sequence = btrfs_stack_inode_sequence(inode_item);
	inode->i_rdev = 0;
	*rdev = btrfs_stack_inode_rdev(inode_item);
	BTRFS_I(inode)->flags = btrfs_stack_inode_flags(inode_item);

	tspec = btrfs_inode_atime(inode_item);
	inode->i_atime.tv_sec = btrfs_stack_timespec_sec(tspec);
	inode->i_atime.tv_nsec = btrfs_stack_timespec_nsec(tspec);

	tspec = btrfs_inode_mtime(inode_item);
	inode->i_mtime.tv_sec = btrfs_stack_timespec_sec(tspec);
	inode->i_mtime.tv_nsec = btrfs_stack_timespec_nsec(tspec);

	tspec = btrfs_inode_ctime(inode_item);
	inode->i_ctime.tv_sec = btrfs_stack_timespec_sec(tspec);
	inode->i_ctime.tv_nsec = btrfs_stack_timespec_nsec(tspec);

	inode->i_generation = BTRFS_I(inode)->generation;
	BTRFS_I(inode)->index_cnt = (u64)-1;

	mutex_unlock(&delayed_node->mutex);
	btrfs_release_delayed_node(delayed_node);
	return 0;
}

1750 1751 1752 1753
int btrfs_delayed_update_inode(struct btrfs_trans_handle *trans,
			       struct btrfs_root *root, struct inode *inode)
{
	struct btrfs_delayed_node *delayed_node;
1754
	int ret = 0;
1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765

	delayed_node = btrfs_get_or_create_delayed_node(inode);
	if (IS_ERR(delayed_node))
		return PTR_ERR(delayed_node);

	mutex_lock(&delayed_node->mutex);
	if (delayed_node->inode_dirty) {
		fill_stack_inode_item(trans, &delayed_node->inode_item, inode);
		goto release_node;
	}

1766 1767
	ret = btrfs_delayed_inode_reserve_metadata(trans, root, inode,
						   delayed_node);
1768 1769
	if (ret)
		goto release_node;
1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849

	fill_stack_inode_item(trans, &delayed_node->inode_item, inode);
	delayed_node->inode_dirty = 1;
	delayed_node->count++;
	atomic_inc(&root->fs_info->delayed_root->items);
release_node:
	mutex_unlock(&delayed_node->mutex);
	btrfs_release_delayed_node(delayed_node);
	return ret;
}

static void __btrfs_kill_delayed_node(struct btrfs_delayed_node *delayed_node)
{
	struct btrfs_root *root = delayed_node->root;
	struct btrfs_delayed_item *curr_item, *prev_item;

	mutex_lock(&delayed_node->mutex);
	curr_item = __btrfs_first_delayed_insertion_item(delayed_node);
	while (curr_item) {
		btrfs_delayed_item_release_metadata(root, curr_item);
		prev_item = curr_item;
		curr_item = __btrfs_next_delayed_item(prev_item);
		btrfs_release_delayed_item(prev_item);
	}

	curr_item = __btrfs_first_delayed_deletion_item(delayed_node);
	while (curr_item) {
		btrfs_delayed_item_release_metadata(root, curr_item);
		prev_item = curr_item;
		curr_item = __btrfs_next_delayed_item(prev_item);
		btrfs_release_delayed_item(prev_item);
	}

	if (delayed_node->inode_dirty) {
		btrfs_delayed_inode_release_metadata(root, delayed_node);
		btrfs_release_delayed_inode(delayed_node);
	}
	mutex_unlock(&delayed_node->mutex);
}

void btrfs_kill_delayed_inode_items(struct inode *inode)
{
	struct btrfs_delayed_node *delayed_node;

	delayed_node = btrfs_get_delayed_node(inode);
	if (!delayed_node)
		return;

	__btrfs_kill_delayed_node(delayed_node);
	btrfs_release_delayed_node(delayed_node);
}

void btrfs_kill_all_delayed_nodes(struct btrfs_root *root)
{
	u64 inode_id = 0;
	struct btrfs_delayed_node *delayed_nodes[8];
	int i, n;

	while (1) {
		spin_lock(&root->inode_lock);
		n = radix_tree_gang_lookup(&root->delayed_nodes_tree,
					   (void **)delayed_nodes, inode_id,
					   ARRAY_SIZE(delayed_nodes));
		if (!n) {
			spin_unlock(&root->inode_lock);
			break;
		}

		inode_id = delayed_nodes[n - 1]->inode_id + 1;

		for (i = 0; i < n; i++)
			atomic_inc(&delayed_nodes[i]->refs);
		spin_unlock(&root->inode_lock);

		for (i = 0; i < n; i++) {
			__btrfs_kill_delayed_node(delayed_nodes[i]);
			btrfs_release_delayed_node(delayed_nodes[i]);
		}
	}
}