inode.c 66.4 KB
Newer Older
C
Chris Mason 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
/*
 * Copyright (C) 2007 Oracle.  All rights reserved.
 *
 * 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.
 */

C
Chris Mason 已提交
19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
#include <linux/module.h>
#include <linux/buffer_head.h>
#include <linux/fs.h>
#include <linux/pagemap.h>
#include <linux/highmem.h>
#include <linux/time.h>
#include <linux/init.h>
#include <linux/string.h>
#include <linux/smp_lock.h>
#include <linux/backing-dev.h>
#include <linux/mpage.h>
#include <linux/swap.h>
#include <linux/writeback.h>
#include <linux/statfs.h>
#include <linux/compat.h>
C
Chris Mason 已提交
34
#include <linux/bit_spinlock.h>
C
Chris Mason 已提交
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 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
#include "btrfs_inode.h"
#include "ioctl.h"
#include "print-tree.h"

struct btrfs_iget_args {
	u64 ino;
	struct btrfs_root *root;
};

static struct inode_operations btrfs_dir_inode_operations;
static struct inode_operations btrfs_symlink_inode_operations;
static struct inode_operations btrfs_dir_ro_inode_operations;
static struct inode_operations btrfs_file_inode_operations;
static struct address_space_operations btrfs_aops;
static struct address_space_operations btrfs_symlink_aops;
static struct file_operations btrfs_dir_file_operations;

static struct kmem_cache *btrfs_inode_cachep;
struct kmem_cache *btrfs_trans_handle_cachep;
struct kmem_cache *btrfs_transaction_cachep;
struct kmem_cache *btrfs_bit_radix_cachep;
struct kmem_cache *btrfs_path_cachep;

#define S_SHIFT 12
static unsigned char btrfs_type_by_mode[S_IFMT >> S_SHIFT] = {
	[S_IFREG >> S_SHIFT]	= BTRFS_FT_REG_FILE,
	[S_IFDIR >> S_SHIFT]	= BTRFS_FT_DIR,
	[S_IFCHR >> S_SHIFT]	= BTRFS_FT_CHRDEV,
	[S_IFBLK >> S_SHIFT]	= BTRFS_FT_BLKDEV,
	[S_IFIFO >> S_SHIFT]	= BTRFS_FT_FIFO,
	[S_IFSOCK >> S_SHIFT]	= BTRFS_FT_SOCK,
	[S_IFLNK >> S_SHIFT]	= BTRFS_FT_SYMLINK,
};

void btrfs_read_locked_inode(struct inode *inode)
{
	struct btrfs_path *path;
	struct btrfs_inode_item *inode_item;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct btrfs_key location;
	u64 alloc_group_block;
	int ret;

	path = btrfs_alloc_path();
	BUG_ON(!path);
	mutex_lock(&root->fs_info->fs_mutex);

	memcpy(&location, &BTRFS_I(inode)->location, sizeof(location));
	ret = btrfs_lookup_inode(NULL, root, path, &location, 0);
	if (ret) {
		btrfs_free_path(path);
		goto make_bad;
	}
	inode_item = btrfs_item_ptr(btrfs_buffer_leaf(path->nodes[0]),
				  path->slots[0],
				  struct btrfs_inode_item);

	inode->i_mode = btrfs_inode_mode(inode_item);
	inode->i_nlink = btrfs_inode_nlink(inode_item);
	inode->i_uid = btrfs_inode_uid(inode_item);
	inode->i_gid = btrfs_inode_gid(inode_item);
	inode->i_size = btrfs_inode_size(inode_item);
	inode->i_atime.tv_sec = btrfs_timespec_sec(&inode_item->atime);
	inode->i_atime.tv_nsec = btrfs_timespec_nsec(&inode_item->atime);
	inode->i_mtime.tv_sec = btrfs_timespec_sec(&inode_item->mtime);
	inode->i_mtime.tv_nsec = btrfs_timespec_nsec(&inode_item->mtime);
	inode->i_ctime.tv_sec = btrfs_timespec_sec(&inode_item->ctime);
	inode->i_ctime.tv_nsec = btrfs_timespec_nsec(&inode_item->ctime);
	inode->i_blocks = btrfs_inode_nblocks(inode_item);
	inode->i_generation = btrfs_inode_generation(inode_item);
	alloc_group_block = btrfs_inode_block_group(inode_item);
	BTRFS_I(inode)->block_group = btrfs_lookup_block_group(root->fs_info,
						       alloc_group_block);

	btrfs_free_path(path);
	inode_item = NULL;

	mutex_unlock(&root->fs_info->fs_mutex);

	switch (inode->i_mode & S_IFMT) {
#if 0
	default:
		init_special_inode(inode, inode->i_mode,
				   btrfs_inode_rdev(inode_item));
		break;
#endif
	case S_IFREG:
		inode->i_mapping->a_ops = &btrfs_aops;
		inode->i_fop = &btrfs_file_operations;
		inode->i_op = &btrfs_file_inode_operations;
		break;
	case S_IFDIR:
		inode->i_fop = &btrfs_dir_file_operations;
		if (root == root->fs_info->tree_root)
			inode->i_op = &btrfs_dir_ro_inode_operations;
		else
			inode->i_op = &btrfs_dir_inode_operations;
		break;
	case S_IFLNK:
		inode->i_op = &btrfs_symlink_inode_operations;
		inode->i_mapping->a_ops = &btrfs_symlink_aops;
		break;
	}
	return;

make_bad:
	btrfs_release_path(root, path);
	btrfs_free_path(path);
	mutex_unlock(&root->fs_info->fs_mutex);
	make_bad_inode(inode);
}

static void fill_inode_item(struct btrfs_inode_item *item,
			    struct inode *inode)
{
	btrfs_set_inode_uid(item, inode->i_uid);
	btrfs_set_inode_gid(item, inode->i_gid);
	btrfs_set_inode_size(item, inode->i_size);
	btrfs_set_inode_mode(item, inode->i_mode);
	btrfs_set_inode_nlink(item, inode->i_nlink);
	btrfs_set_timespec_sec(&item->atime, inode->i_atime.tv_sec);
	btrfs_set_timespec_nsec(&item->atime, inode->i_atime.tv_nsec);
	btrfs_set_timespec_sec(&item->mtime, inode->i_mtime.tv_sec);
	btrfs_set_timespec_nsec(&item->mtime, inode->i_mtime.tv_nsec);
	btrfs_set_timespec_sec(&item->ctime, inode->i_ctime.tv_sec);
	btrfs_set_timespec_nsec(&item->ctime, inode->i_ctime.tv_nsec);
	btrfs_set_inode_nblocks(item, inode->i_blocks);
	btrfs_set_inode_generation(item, inode->i_generation);
	btrfs_set_inode_block_group(item,
				    BTRFS_I(inode)->block_group->key.objectid);
}

static int btrfs_update_inode(struct btrfs_trans_handle *trans,
			      struct btrfs_root *root,
			      struct inode *inode)
{
	struct btrfs_inode_item *inode_item;
	struct btrfs_path *path;
	int ret;

	path = btrfs_alloc_path();
	BUG_ON(!path);
	ret = btrfs_lookup_inode(trans, root, path,
				 &BTRFS_I(inode)->location, 1);
	if (ret) {
		if (ret > 0)
			ret = -ENOENT;
		goto failed;
	}

	inode_item = btrfs_item_ptr(btrfs_buffer_leaf(path->nodes[0]),
				  path->slots[0],
				  struct btrfs_inode_item);

	fill_inode_item(inode_item, inode);
	btrfs_mark_buffer_dirty(path->nodes[0]);
	ret = 0;
failed:
	btrfs_release_path(root, path);
	btrfs_free_path(path);
	return ret;
}


static int btrfs_unlink_trans(struct btrfs_trans_handle *trans,
			      struct btrfs_root *root,
			      struct inode *dir,
			      struct dentry *dentry)
{
	struct btrfs_path *path;
	const char *name = dentry->d_name.name;
	int name_len = dentry->d_name.len;
	int ret = 0;
	u64 objectid;
	struct btrfs_dir_item *di;

	path = btrfs_alloc_path();
215 216 217 218 219
	if (!path) {
		ret = -ENOMEM;
		goto err;
	}

C
Chris Mason 已提交
220 221 222 223 224 225 226 227 228 229 230 231
	di = btrfs_lookup_dir_item(trans, root, path, dir->i_ino,
				    name, name_len, -1);
	if (IS_ERR(di)) {
		ret = PTR_ERR(di);
		goto err;
	}
	if (!di) {
		ret = -ENOENT;
		goto err;
	}
	objectid = btrfs_disk_key_objectid(&di->location);
	ret = btrfs_delete_one_dir_name(trans, root, path, di);
232 233
	if (ret)
		goto err;
C
Chris Mason 已提交
234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254
	btrfs_release_path(root, path);

	di = btrfs_lookup_dir_index_item(trans, root, path, dir->i_ino,
					 objectid, name, name_len, -1);
	if (IS_ERR(di)) {
		ret = PTR_ERR(di);
		goto err;
	}
	if (!di) {
		ret = -ENOENT;
		goto err;
	}
	ret = btrfs_delete_one_dir_name(trans, root, path, di);

	dentry->d_inode->i_ctime = dir->i_ctime;
err:
	btrfs_free_path(path);
	if (!ret) {
		dir->i_size -= name_len * 2;
		btrfs_update_inode(trans, root, dir);
		drop_nlink(dentry->d_inode);
255
		ret = btrfs_update_inode(trans, root, dentry->d_inode);
C
Chris Mason 已提交
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
		dir->i_sb->s_dirt = 1;
	}
	return ret;
}

static int btrfs_unlink(struct inode *dir, struct dentry *dentry)
{
	struct btrfs_root *root;
	struct btrfs_trans_handle *trans;
	int ret;

	root = BTRFS_I(dir)->root;
	mutex_lock(&root->fs_info->fs_mutex);
	trans = btrfs_start_transaction(root, 1);
	btrfs_set_trans_block_group(trans, dir);
	ret = btrfs_unlink_trans(trans, root, dir, dentry);
	btrfs_end_transaction(trans, root);
	mutex_unlock(&root->fs_info->fs_mutex);
	btrfs_btree_balance_dirty(root);
	return ret;
}

static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
{
	struct inode *inode = dentry->d_inode;
	int err;
	int ret;
	struct btrfs_root *root = BTRFS_I(dir)->root;
	struct btrfs_path *path;
	struct btrfs_key key;
	struct btrfs_trans_handle *trans;
	struct btrfs_key found_key;
	int found_type;
	struct btrfs_leaf *leaf;
	char *goodnames = "..";

	path = btrfs_alloc_path();
	BUG_ON(!path);
	mutex_lock(&root->fs_info->fs_mutex);
	trans = btrfs_start_transaction(root, 1);
	btrfs_set_trans_block_group(trans, dir);
	key.objectid = inode->i_ino;
	key.offset = (u64)-1;
	key.flags = (u32)-1;
	while(1) {
		ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
		if (ret < 0) {
			err = ret;
			goto out;
		}
		BUG_ON(ret == 0);
		if (path->slots[0] == 0) {
			err = -ENOENT;
			goto out;
		}
		path->slots[0]--;
		leaf = btrfs_buffer_leaf(path->nodes[0]);
		btrfs_disk_key_to_cpu(&found_key,
				      &leaf->items[path->slots[0]].key);
		found_type = btrfs_key_type(&found_key);
		if (found_key.objectid != inode->i_ino) {
			err = -ENOENT;
			goto out;
		}
		if ((found_type != BTRFS_DIR_ITEM_KEY &&
		     found_type != BTRFS_DIR_INDEX_KEY) ||
	            (!btrfs_match_dir_item_name(root, path, goodnames, 2) &&
	            !btrfs_match_dir_item_name(root, path, goodnames, 1))) {
			err = -ENOTEMPTY;
			goto out;
		}
		ret = btrfs_del_item(trans, root, path);
		BUG_ON(ret);

		if (found_type == BTRFS_DIR_ITEM_KEY && found_key.offset == 1)
			break;
		btrfs_release_path(root, path);
	}
	ret = 0;
	btrfs_release_path(root, path);

	/* now the directory is empty */
	err = btrfs_unlink_trans(trans, root, dir, dentry);
	if (!err) {
		inode->i_size = 0;
	}
out:
	btrfs_release_path(root, path);
	btrfs_free_path(path);
	mutex_unlock(&root->fs_info->fs_mutex);
	ret = btrfs_end_transaction(trans, root);
	btrfs_btree_balance_dirty(root);
	if (ret && !err)
		err = ret;
	return err;
}

static int btrfs_free_inode(struct btrfs_trans_handle *trans,
			    struct btrfs_root *root,
			    struct inode *inode)
{
	struct btrfs_path *path;
	int ret;

	clear_inode(inode);

	path = btrfs_alloc_path();
	BUG_ON(!path);
	ret = btrfs_lookup_inode(trans, root, path,
				 &BTRFS_I(inode)->location, -1);
366 367 368 369
	if (ret > 0)
		ret = -ENOENT;
	if (!ret)
		ret = btrfs_del_item(trans, root, path);
C
Chris Mason 已提交
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
	btrfs_free_path(path);
	return ret;
}

/*
 * truncates go from a high offset to a low offset.  So, walk
 * from hi to lo in the node and issue readas.  Stop when you find
 * keys from a different objectid
 */
static void reada_truncate(struct btrfs_root *root, struct btrfs_path *path,
			   u64 objectid)
{
	struct btrfs_node *node;
	int i;
	int nritems;
	u64 item_objectid;
	u64 blocknr;
	int slot;
	int ret;

	if (!path->nodes[1])
		return;
	node = btrfs_buffer_node(path->nodes[1]);
	slot = path->slots[1];
	if (slot == 0)
		return;
	nritems = btrfs_header_nritems(&node->header);
	for (i = slot - 1; i >= 0; i--) {
		item_objectid = btrfs_disk_key_objectid(&node->ptrs[i].key);
		if (item_objectid != objectid)
			break;
		blocknr = btrfs_node_blockptr(node, i);
		ret = readahead_tree_block(root, blocknr);
		if (ret)
			break;
	}
}

/*
 * this can truncate away extent items, csum items and directory items.
 * It starts at a high offset and removes keys until it can't find
 * any higher than i_size.
 *
 * csum items that cross the new i_size are truncated to the new size
 * as well.
 */
static int btrfs_truncate_in_trans(struct btrfs_trans_handle *trans,
				   struct btrfs_root *root,
				   struct inode *inode)
{
	int ret;
	struct btrfs_path *path;
	struct btrfs_key key;
	struct btrfs_disk_key *found_key;
	u32 found_type;
	struct btrfs_leaf *leaf;
	struct btrfs_file_extent_item *fi;
	u64 extent_start = 0;
	u64 extent_num_blocks = 0;
	u64 item_end = 0;
	int found_extent;
	int del_item;

	path = btrfs_alloc_path();
	BUG_ON(!path);
	/* FIXME, add redo link to tree so we don't leak on crash */
	key.objectid = inode->i_ino;
	key.offset = (u64)-1;
	key.flags = (u32)-1;
	while(1) {
		btrfs_init_path(path);
		fi = NULL;
		ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
		if (ret < 0) {
			goto error;
		}
		if (ret > 0) {
			BUG_ON(path->slots[0] == 0);
			path->slots[0]--;
		}
		reada_truncate(root, path, inode->i_ino);
		leaf = btrfs_buffer_leaf(path->nodes[0]);
		found_key = &leaf->items[path->slots[0]].key;
		found_type = btrfs_disk_key_type(found_key);

		if (btrfs_disk_key_objectid(found_key) != inode->i_ino)
			break;
		if (found_type != BTRFS_CSUM_ITEM_KEY &&
		    found_type != BTRFS_DIR_ITEM_KEY &&
		    found_type != BTRFS_DIR_INDEX_KEY &&
		    found_type != BTRFS_EXTENT_DATA_KEY)
			break;

		item_end = btrfs_disk_key_offset(found_key);
		if (found_type == BTRFS_EXTENT_DATA_KEY) {
			fi = btrfs_item_ptr(btrfs_buffer_leaf(path->nodes[0]),
					    path->slots[0],
					    struct btrfs_file_extent_item);
			if (btrfs_file_extent_type(fi) !=
			    BTRFS_FILE_EXTENT_INLINE) {
				item_end += btrfs_file_extent_num_blocks(fi) <<
						inode->i_blkbits;
			}
		}
		if (found_type == BTRFS_CSUM_ITEM_KEY) {
			ret = btrfs_csum_truncate(trans, root, path,
						  inode->i_size);
			BUG_ON(ret);
		}
		if (item_end < inode->i_size) {
			if (found_type) {
				btrfs_set_key_type(&key, found_type - 1);
				continue;
			}
			break;
		}
		if (btrfs_disk_key_offset(found_key) >= inode->i_size)
			del_item = 1;
		else
			del_item = 0;
		found_extent = 0;

		/* FIXME, shrink the extent if the ref count is only 1 */
		if (found_type == BTRFS_EXTENT_DATA_KEY &&
			   btrfs_file_extent_type(fi) !=
			   BTRFS_FILE_EXTENT_INLINE) {
			u64 num_dec;
			if (!del_item) {
				u64 orig_num_blocks =
					btrfs_file_extent_num_blocks(fi);
				extent_num_blocks = inode->i_size -
					btrfs_disk_key_offset(found_key) +
					root->blocksize - 1;
				extent_num_blocks >>= inode->i_blkbits;
				btrfs_set_file_extent_num_blocks(fi,
							 extent_num_blocks);
				inode->i_blocks -= (orig_num_blocks -
					extent_num_blocks) << 3;
				mark_buffer_dirty(path->nodes[0]);
			} else {
				extent_start =
					btrfs_file_extent_disk_blocknr(fi);
				extent_num_blocks =
					btrfs_file_extent_disk_num_blocks(fi);
				/* FIXME blocksize != 4096 */
				num_dec = btrfs_file_extent_num_blocks(fi) << 3;
				if (extent_start != 0) {
					found_extent = 1;
					inode->i_blocks -= num_dec;
				}
			}
		}
		if (del_item) {
			ret = btrfs_del_item(trans, root, path);
524 525
			if (ret)
				goto error;
C
Chris Mason 已提交
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
		} else {
			break;
		}
		btrfs_release_path(root, path);
		if (found_extent) {
			ret = btrfs_free_extent(trans, root, extent_start,
						extent_num_blocks, 0);
			BUG_ON(ret);
		}
	}
	ret = 0;
error:
	btrfs_release_path(root, path);
	btrfs_free_path(path);
	inode->i_sb->s_dirt = 1;
	return ret;
}

/*
 * taken from block_truncate_page, but does cow as it zeros out
 * any bytes left in the last page in the file.
 */
static int btrfs_truncate_page(struct address_space *mapping, loff_t from)
{
	struct inode *inode = mapping->host;
	unsigned blocksize = 1 << inode->i_blkbits;
	pgoff_t index = from >> PAGE_CACHE_SHIFT;
	unsigned offset = from & (PAGE_CACHE_SIZE-1);
	struct page *page;
	char *kaddr;
	int ret = 0;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	u64 alloc_hint = 0;
	struct btrfs_key ins;
	struct btrfs_trans_handle *trans;

	if ((offset & (blocksize - 1)) == 0)
		goto out;

	ret = -ENOMEM;
	page = grab_cache_page(mapping, index);
	if (!page)
		goto out;

	if (!PageUptodate(page)) {
C
Chris Mason 已提交
571
		ret = btrfs_readpage(NULL, page);
C
Chris Mason 已提交
572 573 574 575 576 577 578 579 580 581 582 583 584 585
		lock_page(page);
		if (!PageUptodate(page)) {
			ret = -EIO;
			goto out;
		}
	}
	mutex_lock(&root->fs_info->fs_mutex);
	trans = btrfs_start_transaction(root, 1);
	btrfs_set_trans_block_group(trans, inode);

	ret = btrfs_drop_extents(trans, root, inode,
				 page->index << PAGE_CACHE_SHIFT,
				 (page->index + 1) << PAGE_CACHE_SHIFT,
				 &alloc_hint);
586 587
	if (ret)
		goto out;
C
Chris Mason 已提交
588 589
	ret = btrfs_alloc_extent(trans, root, inode->i_ino, 1,
				 alloc_hint, (u64)-1, &ins, 1);
590 591
	if (ret)
		goto out;
C
Chris Mason 已提交
592 593 594
	ret = btrfs_insert_file_extent(trans, root, inode->i_ino,
				       page->index << PAGE_CACHE_SHIFT,
				       ins.objectid, 1, 1);
595 596
	if (ret)
		goto out;
C
Chris Mason 已提交
597 598 599 600
	SetPageChecked(page);
	kaddr = kmap(page);
	memset(kaddr + offset, 0, PAGE_CACHE_SIZE - offset);
	flush_dcache_page(page);
601
	ret = btrfs_csum_file_block(trans, root, inode->i_ino,
C
Chris Mason 已提交
602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646
			      page->index << PAGE_CACHE_SHIFT,
			      kaddr, PAGE_CACHE_SIZE);
	kunmap(page);
	btrfs_end_transaction(trans, root);
	mutex_unlock(&root->fs_info->fs_mutex);

	set_page_dirty(page);
	unlock_page(page);
	page_cache_release(page);
out:
	return ret;
}

static int btrfs_setattr(struct dentry *dentry, struct iattr *attr)
{
	struct inode *inode = dentry->d_inode;
	int err;

	err = inode_change_ok(inode, attr);
	if (err)
		return err;

	if (S_ISREG(inode->i_mode) &&
	    attr->ia_valid & ATTR_SIZE && attr->ia_size > inode->i_size) {
		struct btrfs_trans_handle *trans;
		struct btrfs_root *root = BTRFS_I(inode)->root;
		u64 mask = root->blocksize - 1;
		u64 pos = (inode->i_size + mask) & ~mask;
		u64 hole_size;

		if (attr->ia_size <= pos)
			goto out;

		btrfs_truncate_page(inode->i_mapping, inode->i_size);

		hole_size = (attr->ia_size - pos + mask) & ~mask;
		hole_size >>= inode->i_blkbits;

		mutex_lock(&root->fs_info->fs_mutex);
		trans = btrfs_start_transaction(root, 1);
		btrfs_set_trans_block_group(trans, inode);
		err = btrfs_insert_file_extent(trans, root, inode->i_ino,
					       pos, 0, 0, hole_size);
		btrfs_end_transaction(trans, root);
		mutex_unlock(&root->fs_info->fs_mutex);
647 648
		if (err)
			return err;
C
Chris Mason 已提交
649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669
	}
out:
	err = inode_setattr(inode, attr);

	return err;
}
void btrfs_delete_inode(struct inode *inode)
{
	struct btrfs_trans_handle *trans;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	int ret;

	truncate_inode_pages(&inode->i_data, 0);
	if (is_bad_inode(inode)) {
		goto no_delete;
	}
	inode->i_size = 0;
	mutex_lock(&root->fs_info->fs_mutex);
	trans = btrfs_start_transaction(root, 1);
	btrfs_set_trans_block_group(trans, inode);
	ret = btrfs_truncate_in_trans(trans, root, inode);
670 671 672 673 674
	if (ret)
		goto no_delete_lock;
	ret = btrfs_free_inode(trans, root, inode);
	if (ret)
		goto no_delete_lock;
C
Chris Mason 已提交
675 676 677 678
	btrfs_end_transaction(trans, root);
	mutex_unlock(&root->fs_info->fs_mutex);
	btrfs_btree_balance_dirty(root);
	return;
679 680 681 682 683

no_delete_lock:
	btrfs_end_transaction(trans, root);
	mutex_unlock(&root->fs_info->fs_mutex);
	btrfs_btree_balance_dirty(root);
C
Chris Mason 已提交
684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966
no_delete:
	clear_inode(inode);
}

/*
 * this returns the key found in the dir entry in the location pointer.
 * If no dir entries were found, location->objectid is 0.
 */
static int btrfs_inode_by_name(struct inode *dir, struct dentry *dentry,
			       struct btrfs_key *location)
{
	const char *name = dentry->d_name.name;
	int namelen = dentry->d_name.len;
	struct btrfs_dir_item *di;
	struct btrfs_path *path;
	struct btrfs_root *root = BTRFS_I(dir)->root;
	int ret;

	path = btrfs_alloc_path();
	BUG_ON(!path);
	di = btrfs_lookup_dir_item(NULL, root, path, dir->i_ino, name,
				    namelen, 0);
	if (!di || IS_ERR(di)) {
		location->objectid = 0;
		ret = 0;
		goto out;
	}
	btrfs_disk_key_to_cpu(location, &di->location);
out:
	btrfs_release_path(root, path);
	btrfs_free_path(path);
	return ret;
}

/*
 * when we hit a tree root in a directory, the btrfs part of the inode
 * needs to be changed to reflect the root directory of the tree root.  This
 * is kind of like crossing a mount point.
 */
static int fixup_tree_root_location(struct btrfs_root *root,
			     struct btrfs_key *location,
			     struct btrfs_root **sub_root)
{
	struct btrfs_path *path;
	struct btrfs_root_item *ri;

	if (btrfs_key_type(location) != BTRFS_ROOT_ITEM_KEY)
		return 0;
	if (location->objectid == BTRFS_ROOT_TREE_OBJECTID)
		return 0;

	path = btrfs_alloc_path();
	BUG_ON(!path);
	mutex_lock(&root->fs_info->fs_mutex);

	*sub_root = btrfs_read_fs_root(root->fs_info, location);
	if (IS_ERR(*sub_root))
		return PTR_ERR(*sub_root);

	ri = &(*sub_root)->root_item;
	location->objectid = btrfs_root_dirid(ri);
	location->flags = 0;
	btrfs_set_key_type(location, BTRFS_INODE_ITEM_KEY);
	location->offset = 0;

	btrfs_free_path(path);
	mutex_unlock(&root->fs_info->fs_mutex);
	return 0;
}

static int btrfs_init_locked_inode(struct inode *inode, void *p)
{
	struct btrfs_iget_args *args = p;
	inode->i_ino = args->ino;
	BTRFS_I(inode)->root = args->root;
	return 0;
}

static int btrfs_find_actor(struct inode *inode, void *opaque)
{
	struct btrfs_iget_args *args = opaque;
	return (args->ino == inode->i_ino &&
		args->root == BTRFS_I(inode)->root);
}

struct inode *btrfs_iget_locked(struct super_block *s, u64 objectid,
				struct btrfs_root *root)
{
	struct inode *inode;
	struct btrfs_iget_args args;
	args.ino = objectid;
	args.root = root;

	inode = iget5_locked(s, objectid, btrfs_find_actor,
			     btrfs_init_locked_inode,
			     (void *)&args);
	return inode;
}

static struct dentry *btrfs_lookup(struct inode *dir, struct dentry *dentry,
				   struct nameidata *nd)
{
	struct inode * inode;
	struct btrfs_inode *bi = BTRFS_I(dir);
	struct btrfs_root *root = bi->root;
	struct btrfs_root *sub_root = root;
	struct btrfs_key location;
	int ret;

	if (dentry->d_name.len > BTRFS_NAME_LEN)
		return ERR_PTR(-ENAMETOOLONG);
	mutex_lock(&root->fs_info->fs_mutex);
	ret = btrfs_inode_by_name(dir, dentry, &location);
	mutex_unlock(&root->fs_info->fs_mutex);
	if (ret < 0)
		return ERR_PTR(ret);
	inode = NULL;
	if (location.objectid) {
		ret = fixup_tree_root_location(root, &location, &sub_root);
		if (ret < 0)
			return ERR_PTR(ret);
		if (ret > 0)
			return ERR_PTR(-ENOENT);
		inode = btrfs_iget_locked(dir->i_sb, location.objectid,
					  sub_root);
		if (!inode)
			return ERR_PTR(-EACCES);
		if (inode->i_state & I_NEW) {
			/* the inode and parent dir are two different roots */
			if (sub_root != root) {
				igrab(inode);
				sub_root->inode = inode;
			}
			BTRFS_I(inode)->root = sub_root;
			memcpy(&BTRFS_I(inode)->location, &location,
			       sizeof(location));
			btrfs_read_locked_inode(inode);
			unlock_new_inode(inode);
		}
	}
	return d_splice_alias(inode, dentry);
}

/*
 * readahead one full node of leaves as long as their keys include
 * the objectid supplied
 */
static void reada_leaves(struct btrfs_root *root, struct btrfs_path *path,
			 u64 objectid)
{
	struct btrfs_node *node;
	int i;
	u32 nritems;
	u64 item_objectid;
	u64 blocknr;
	int slot;
	int ret;

	if (!path->nodes[1])
		return;
	node = btrfs_buffer_node(path->nodes[1]);
	slot = path->slots[1];
	nritems = btrfs_header_nritems(&node->header);
	for (i = slot + 1; i < nritems; i++) {
		item_objectid = btrfs_disk_key_objectid(&node->ptrs[i].key);
		if (item_objectid != objectid)
			break;
		blocknr = btrfs_node_blockptr(node, i);
		ret = readahead_tree_block(root, blocknr);
		if (ret)
			break;
	}
}
static unsigned char btrfs_filetype_table[] = {
	DT_UNKNOWN, DT_REG, DT_DIR, DT_CHR, DT_BLK, DT_FIFO, DT_SOCK, DT_LNK
};

static int btrfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
{
	struct inode *inode = filp->f_path.dentry->d_inode;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct btrfs_item *item;
	struct btrfs_dir_item *di;
	struct btrfs_key key;
	struct btrfs_path *path;
	int ret;
	u32 nritems;
	struct btrfs_leaf *leaf;
	int slot;
	int advance;
	unsigned char d_type;
	int over = 0;
	u32 di_cur;
	u32 di_total;
	u32 di_len;
	int key_type = BTRFS_DIR_INDEX_KEY;

	/* FIXME, use a real flag for deciding about the key type */
	if (root->fs_info->tree_root == root)
		key_type = BTRFS_DIR_ITEM_KEY;
	mutex_lock(&root->fs_info->fs_mutex);
	key.objectid = inode->i_ino;
	key.flags = 0;
	btrfs_set_key_type(&key, key_type);
	key.offset = filp->f_pos;
	path = btrfs_alloc_path();
	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
	if (ret < 0)
		goto err;
	advance = 0;
	reada_leaves(root, path, inode->i_ino);
	while(1) {
		leaf = btrfs_buffer_leaf(path->nodes[0]);
		nritems = btrfs_header_nritems(&leaf->header);
		slot = path->slots[0];
		if (advance || slot >= nritems) {
			if (slot >= nritems -1) {
				reada_leaves(root, path, inode->i_ino);
				ret = btrfs_next_leaf(root, path);
				if (ret)
					break;
				leaf = btrfs_buffer_leaf(path->nodes[0]);
				nritems = btrfs_header_nritems(&leaf->header);
				slot = path->slots[0];
			} else {
				slot++;
				path->slots[0]++;
			}
		}
		advance = 1;
		item = leaf->items + slot;
		if (btrfs_disk_key_objectid(&item->key) != key.objectid)
			break;
		if (btrfs_disk_key_type(&item->key) != key_type)
			break;
		if (btrfs_disk_key_offset(&item->key) < filp->f_pos)
			continue;
		filp->f_pos = btrfs_disk_key_offset(&item->key);
		advance = 1;
		di = btrfs_item_ptr(leaf, slot, struct btrfs_dir_item);
		di_cur = 0;
		di_total = btrfs_item_size(leaf->items + slot);
		while(di_cur < di_total) {
			d_type = btrfs_filetype_table[btrfs_dir_type(di)];
			over = filldir(dirent, (const char *)(di + 1),
				       btrfs_dir_name_len(di),
				       btrfs_disk_key_offset(&item->key),
				       btrfs_disk_key_objectid(&di->location),
				       d_type);
			if (over)
				goto nopos;
			di_len = btrfs_dir_name_len(di) + sizeof(*di);
			di_cur += di_len;
			di = (struct btrfs_dir_item *)((char *)di + di_len);
		}
	}
	filp->f_pos++;
nopos:
	ret = 0;
err:
	btrfs_release_path(root, path);
	btrfs_free_path(path);
	mutex_unlock(&root->fs_info->fs_mutex);
	return ret;
}

int btrfs_write_inode(struct inode *inode, int wait)
{
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct btrfs_trans_handle *trans;
	int ret = 0;

	if (wait) {
		mutex_lock(&root->fs_info->fs_mutex);
		trans = btrfs_start_transaction(root, 1);
		btrfs_set_trans_block_group(trans, inode);
		ret = btrfs_commit_transaction(trans, root);
		mutex_unlock(&root->fs_info->fs_mutex);
	}
	return ret;
}

/*
967
 * This is somewhat expensive, updating the tree every time the
C
Chris Mason 已提交
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 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022
 * inode changes.  But, it is most likely to find the inode in cache.
 * FIXME, needs more benchmarking...there are no reasons other than performance
 * to keep or drop this code.
 */
void btrfs_dirty_inode(struct inode *inode)
{
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct btrfs_trans_handle *trans;

	mutex_lock(&root->fs_info->fs_mutex);
	trans = btrfs_start_transaction(root, 1);
	btrfs_set_trans_block_group(trans, inode);
	btrfs_update_inode(trans, root, inode);
	btrfs_end_transaction(trans, root);
	mutex_unlock(&root->fs_info->fs_mutex);
}

static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
				     struct btrfs_root *root,
				     u64 objectid,
				     struct btrfs_block_group_cache *group,
				     int mode)
{
	struct inode *inode;
	struct btrfs_inode_item inode_item;
	struct btrfs_key *location;
	int ret;
	int owner;

	inode = new_inode(root->fs_info->sb);
	if (!inode)
		return ERR_PTR(-ENOMEM);

	BTRFS_I(inode)->root = root;
	if (mode & S_IFDIR)
		owner = 0;
	else
		owner = 1;
	group = btrfs_find_block_group(root, group, 0, 0, owner);
	BTRFS_I(inode)->block_group = group;

	inode->i_uid = current->fsuid;
	inode->i_gid = current->fsgid;
	inode->i_mode = mode;
	inode->i_ino = objectid;
	inode->i_blocks = 0;
	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
	fill_inode_item(&inode_item, inode);
	location = &BTRFS_I(inode)->location;
	location->objectid = objectid;
	location->flags = 0;
	location->offset = 0;
	btrfs_set_key_type(location, BTRFS_INODE_ITEM_KEY);

	ret = btrfs_insert_inode(trans, root, objectid, &inode_item);
1023 1024
	if (ret)
		return ERR_PTR(ret);
C
Chris Mason 已提交
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 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 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 1133 1134 1135 1136 1137 1138 1139 1140 1141
	insert_inode_hash(inode);
	return inode;
}

static inline u8 btrfs_inode_type(struct inode *inode)
{
	return btrfs_type_by_mode[(inode->i_mode & S_IFMT) >> S_SHIFT];
}

static int btrfs_add_link(struct btrfs_trans_handle *trans,
			    struct dentry *dentry, struct inode *inode)
{
	int ret;
	struct btrfs_key key;
	struct btrfs_root *root = BTRFS_I(dentry->d_parent->d_inode)->root;
	key.objectid = inode->i_ino;
	key.flags = 0;
	btrfs_set_key_type(&key, BTRFS_INODE_ITEM_KEY);
	key.offset = 0;

	ret = btrfs_insert_dir_item(trans, root,
				    dentry->d_name.name, dentry->d_name.len,
				    dentry->d_parent->d_inode->i_ino,
				    &key, btrfs_inode_type(inode));
	if (ret == 0) {
		dentry->d_parent->d_inode->i_size += dentry->d_name.len * 2;
		ret = btrfs_update_inode(trans, root,
					 dentry->d_parent->d_inode);
	}
	return ret;
}

static int btrfs_add_nondir(struct btrfs_trans_handle *trans,
			    struct dentry *dentry, struct inode *inode)
{
	int err = btrfs_add_link(trans, dentry, inode);
	if (!err) {
		d_instantiate(dentry, inode);
		return 0;
	}
	if (err > 0)
		err = -EEXIST;
	return err;
}

static int btrfs_create(struct inode *dir, struct dentry *dentry,
			int mode, struct nameidata *nd)
{
	struct btrfs_trans_handle *trans;
	struct btrfs_root *root = BTRFS_I(dir)->root;
	struct inode *inode;
	int err;
	int drop_inode = 0;
	u64 objectid;

	mutex_lock(&root->fs_info->fs_mutex);
	trans = btrfs_start_transaction(root, 1);
	btrfs_set_trans_block_group(trans, dir);

	err = btrfs_find_free_objectid(trans, root, dir->i_ino, &objectid);
	if (err) {
		err = -ENOSPC;
		goto out_unlock;
	}

	inode = btrfs_new_inode(trans, root, objectid,
				BTRFS_I(dir)->block_group, mode);
	err = PTR_ERR(inode);
	if (IS_ERR(inode))
		goto out_unlock;

	btrfs_set_trans_block_group(trans, inode);
	err = btrfs_add_nondir(trans, dentry, inode);
	if (err)
		drop_inode = 1;
	else {
		inode->i_mapping->a_ops = &btrfs_aops;
		inode->i_fop = &btrfs_file_operations;
		inode->i_op = &btrfs_file_inode_operations;
	}
	dir->i_sb->s_dirt = 1;
	btrfs_update_inode_block_group(trans, inode);
	btrfs_update_inode_block_group(trans, dir);
out_unlock:
	btrfs_end_transaction(trans, root);
	mutex_unlock(&root->fs_info->fs_mutex);

	if (drop_inode) {
		inode_dec_link_count(inode);
		iput(inode);
	}
	btrfs_btree_balance_dirty(root);
	return err;
}

static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
		      struct dentry *dentry)
{
	struct btrfs_trans_handle *trans;
	struct btrfs_root *root = BTRFS_I(dir)->root;
	struct inode *inode = old_dentry->d_inode;
	int err;
	int drop_inode = 0;

	if (inode->i_nlink == 0)
		return -ENOENT;

	inc_nlink(inode);
	mutex_lock(&root->fs_info->fs_mutex);
	trans = btrfs_start_transaction(root, 1);
	btrfs_set_trans_block_group(trans, dir);
	atomic_inc(&inode->i_count);
	err = btrfs_add_nondir(trans, dentry, inode);
	if (err)
		drop_inode = 1;
	dir->i_sb->s_dirt = 1;
	btrfs_update_inode_block_group(trans, dir);
1142 1143 1144
	err = btrfs_update_inode(trans, root, inode);
	if (err)
		drop_inode = 1;
C
Chris Mason 已提交
1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 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 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 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 1274 1275 1276 1277 1278 1279

	btrfs_end_transaction(trans, root);
	mutex_unlock(&root->fs_info->fs_mutex);

	if (drop_inode) {
		inode_dec_link_count(inode);
		iput(inode);
	}
	btrfs_btree_balance_dirty(root);
	return err;
}

static int btrfs_make_empty_dir(struct btrfs_trans_handle *trans,
				struct btrfs_root *root,
				u64 objectid, u64 dirid)
{
	int ret;
	char buf[2];
	struct btrfs_key key;

	buf[0] = '.';
	buf[1] = '.';

	key.objectid = objectid;
	key.offset = 0;
	key.flags = 0;
	btrfs_set_key_type(&key, BTRFS_INODE_ITEM_KEY);

	ret = btrfs_insert_dir_item(trans, root, buf, 1, objectid,
				    &key, BTRFS_FT_DIR);
	if (ret)
		goto error;
	key.objectid = dirid;
	ret = btrfs_insert_dir_item(trans, root, buf, 2, objectid,
				    &key, BTRFS_FT_DIR);
	if (ret)
		goto error;
error:
	return ret;
}

static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
{
	struct inode *inode;
	struct btrfs_trans_handle *trans;
	struct btrfs_root *root = BTRFS_I(dir)->root;
	int err = 0;
	int drop_on_err = 0;
	u64 objectid;

	mutex_lock(&root->fs_info->fs_mutex);
	trans = btrfs_start_transaction(root, 1);
	btrfs_set_trans_block_group(trans, dir);
	if (IS_ERR(trans)) {
		err = PTR_ERR(trans);
		goto out_unlock;
	}

	err = btrfs_find_free_objectid(trans, root, dir->i_ino, &objectid);
	if (err) {
		err = -ENOSPC;
		goto out_unlock;
	}

	inode = btrfs_new_inode(trans, root, objectid,
				BTRFS_I(dir)->block_group, S_IFDIR | mode);
	if (IS_ERR(inode)) {
		err = PTR_ERR(inode);
		goto out_fail;
	}
	drop_on_err = 1;
	inode->i_op = &btrfs_dir_inode_operations;
	inode->i_fop = &btrfs_dir_file_operations;
	btrfs_set_trans_block_group(trans, inode);

	err = btrfs_make_empty_dir(trans, root, inode->i_ino, dir->i_ino);
	if (err)
		goto out_fail;

	inode->i_size = 6;
	err = btrfs_update_inode(trans, root, inode);
	if (err)
		goto out_fail;
	err = btrfs_add_link(trans, dentry, inode);
	if (err)
		goto out_fail;
	d_instantiate(dentry, inode);
	drop_on_err = 0;
	dir->i_sb->s_dirt = 1;
	btrfs_update_inode_block_group(trans, inode);
	btrfs_update_inode_block_group(trans, dir);

out_fail:
	btrfs_end_transaction(trans, root);
out_unlock:
	mutex_unlock(&root->fs_info->fs_mutex);
	if (drop_on_err)
		iput(inode);
	btrfs_btree_balance_dirty(root);
	return err;
}

/*
 * FIBMAP and others want to pass in a fake buffer head.  They need to
 * use BTRFS_GET_BLOCK_NO_DIRECT to make sure we don't try to memcpy
 * any packed file data into the fake bh
 */
#define BTRFS_GET_BLOCK_NO_CREATE 0
#define BTRFS_GET_BLOCK_CREATE 1
#define BTRFS_GET_BLOCK_NO_DIRECT 2

/*
 * FIXME create==1 doe not work.
 */
static int btrfs_get_block_lock(struct inode *inode, sector_t iblock,
				struct buffer_head *result, int create)
{
	int ret;
	int err = 0;
	u64 blocknr;
	u64 extent_start = 0;
	u64 extent_end = 0;
	u64 objectid = inode->i_ino;
	u32 found_type;
	u64 alloc_hint = 0;
	struct btrfs_path *path;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct btrfs_file_extent_item *item;
	struct btrfs_leaf *leaf;
	struct btrfs_disk_key *found_key;
	struct btrfs_trans_handle *trans = NULL;

	path = btrfs_alloc_path();
	BUG_ON(!path);
	if (create & BTRFS_GET_BLOCK_CREATE) {
C
Chris Mason 已提交
1280 1281 1282 1283
		/*
		 * danger!, this only works if the page is properly up
		 * to date somehow
		 */
C
Chris Mason 已提交
1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296
		trans = btrfs_start_transaction(root, 1);
		if (!trans) {
			err = -ENOMEM;
			goto out;
		}
		ret = btrfs_drop_extents(trans, root, inode,
					 iblock << inode->i_blkbits,
					 (iblock + 1) << inode->i_blkbits,
					 &alloc_hint);
		BUG_ON(ret);
	}

	ret = btrfs_lookup_file_extent(NULL, root, path,
A
Aneesh 已提交
1297
				       objectid,
C
Chris Mason 已提交
1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 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
				       iblock << inode->i_blkbits, 0);
	if (ret < 0) {
		err = ret;
		goto out;
	}

	if (ret != 0) {
		if (path->slots[0] == 0) {
			btrfs_release_path(root, path);
			goto not_found;
		}
		path->slots[0]--;
	}

	item = btrfs_item_ptr(btrfs_buffer_leaf(path->nodes[0]), path->slots[0],
			      struct btrfs_file_extent_item);
	leaf = btrfs_buffer_leaf(path->nodes[0]);
	blocknr = btrfs_file_extent_disk_blocknr(item);
	blocknr += btrfs_file_extent_offset(item);

	/* are we inside the extent that was found? */
	found_key = &leaf->items[path->slots[0]].key;
	found_type = btrfs_disk_key_type(found_key);
	if (btrfs_disk_key_objectid(found_key) != objectid ||
	    found_type != BTRFS_EXTENT_DATA_KEY) {
		extent_end = 0;
		extent_start = 0;
		goto not_found;
	}
	found_type = btrfs_file_extent_type(item);
	extent_start = btrfs_disk_key_offset(&leaf->items[path->slots[0]].key);
	if (found_type == BTRFS_FILE_EXTENT_REG) {
		extent_start = extent_start >> inode->i_blkbits;
		extent_end = extent_start + btrfs_file_extent_num_blocks(item);
		err = 0;
		if (btrfs_file_extent_disk_blocknr(item) == 0)
			goto out;
		if (iblock >= extent_start && iblock < extent_end) {
			btrfs_map_bh_to_logical(root, result, blocknr +
						iblock - extent_start);
			goto out;
		}
	} else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
		char *ptr;
		char *map;
		u32 size;

		if (create & BTRFS_GET_BLOCK_NO_DIRECT) {
			err = -EINVAL;
			goto out;
		}
		size = btrfs_file_extent_inline_len(leaf->items +
						    path->slots[0]);
		extent_end = (extent_start + size) >> inode->i_blkbits;
		extent_start >>= inode->i_blkbits;
		if (iblock < extent_start || iblock > extent_end) {
			goto not_found;
		}
		ptr = btrfs_file_extent_inline_start(item);
		map = kmap(result->b_page);
		memcpy(map, ptr, size);
		memset(map + size, 0, PAGE_CACHE_SIZE - size);
		flush_dcache_page(result->b_page);
		kunmap(result->b_page);
		set_buffer_uptodate(result);
		SetPageChecked(result->b_page);
		btrfs_map_bh_to_logical(root, result, 0);
	}
not_found:
	if (create & BTRFS_GET_BLOCK_CREATE) {
		struct btrfs_key ins;
		ret = btrfs_alloc_extent(trans, root, inode->i_ino,
					 1, alloc_hint, (u64)-1,
					 &ins, 1);
1372 1373 1374 1375
		if (ret) {
			err = ret;
			goto out;
		}
C
Chris Mason 已提交
1376 1377 1378 1379
		ret = btrfs_insert_file_extent(trans, root, inode->i_ino,
					       iblock << inode->i_blkbits,
					       ins.objectid, ins.offset,
					       ins.offset);
1380 1381 1382 1383
		if (ret) {
			err = ret;
			goto out;
		}
C
Chris Mason 已提交
1384 1385 1386
		btrfs_map_bh_to_logical(root, result, ins.objectid);
	}
out:
1387 1388 1389 1390 1391
	if (trans) {
		ret = btrfs_end_transaction(trans, root);
		if (!err)
			err = ret;
	}
C
Chris Mason 已提交
1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406
	btrfs_free_path(path);
	return err;
}

int btrfs_get_block(struct inode *inode, sector_t iblock,
		    struct buffer_head *result, int create)
{
	int err;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	mutex_lock(&root->fs_info->fs_mutex);
	err = btrfs_get_block_lock(inode, iblock, result, create);
	mutex_unlock(&root->fs_info->fs_mutex);
	return err;
}

1407 1408
static int btrfs_get_block_csum(struct inode *inode, sector_t iblock,
				struct buffer_head *result, int create)
C
Chris Mason 已提交
1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428
{
	int ret;
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct page *page = result->b_page;
	u64 offset = (page->index << PAGE_CACHE_SHIFT) + bh_offset(result);
	struct btrfs_csum_item *item;
	struct btrfs_path *path = NULL;

	mutex_lock(&root->fs_info->fs_mutex);
	ret = btrfs_get_block_lock(inode, iblock, result, create);
	if (ret)
		goto out;

	path = btrfs_alloc_path();
	item = btrfs_lookup_csum(NULL, root, path, inode->i_ino, offset, 0);
	if (IS_ERR(item)) {
		ret = PTR_ERR(item);
		/* a csum that isn't present is a preallocated region. */
		if (ret == -ENOENT || ret == -EFBIG)
			ret = 0;
1429
		result->b_private = NULL;
C
Chris Mason 已提交
1430 1431 1432 1433 1434 1435 1436 1437 1438 1439
		goto out;
	}
	memcpy((char *)&result->b_private, &item->csum, BTRFS_CRC32_SIZE);
out:
	if (path)
		btrfs_free_path(path);
	mutex_unlock(&root->fs_info->fs_mutex);
	return ret;
}

C
Chris Mason 已提交
1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460
static int btrfs_get_block_bmap(struct inode *inode, sector_t iblock,
			   struct buffer_head *result, int create)
{
	struct btrfs_root *root = BTRFS_I(inode)->root;
	mutex_lock(&root->fs_info->fs_mutex);
	btrfs_get_block_lock(inode, iblock, result, BTRFS_GET_BLOCK_NO_DIRECT);
	mutex_unlock(&root->fs_info->fs_mutex);
	return 0;
}

static sector_t btrfs_bmap(struct address_space *as, sector_t block)
{
	return generic_block_bmap(as, block, btrfs_get_block_bmap);
}

static int btrfs_prepare_write(struct file *file, struct page *page,
			       unsigned from, unsigned to)
{
	return block_prepare_write(page, from, to, btrfs_get_block);
}

C
Chris Mason 已提交
1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474
static void buffer_io_error(struct buffer_head *bh)
{
	char b[BDEVNAME_SIZE];

	printk(KERN_ERR "Buffer I/O error on device %s, logical block %Lu\n",
			bdevname(bh->b_bdev, b),
			(unsigned long long)bh->b_blocknr);
}

/*
 * I/O completion handler for block_read_full_page() - pages
 * which come unlocked at the end of I/O.
 */
static void btrfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
C
Chris Mason 已提交
1475
{
C
Chris Mason 已提交
1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 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
	unsigned long flags;
	struct buffer_head *first;
	struct buffer_head *tmp;
	struct page *page;
	int page_uptodate = 1;
	struct inode *inode;
	int ret;

	BUG_ON(!buffer_async_read(bh));

	page = bh->b_page;
	inode = page->mapping->host;
	if (uptodate) {
		void *kaddr;
		struct btrfs_root *root = BTRFS_I(page->mapping->host)->root;
		if (bh->b_private) {
			char csum[BTRFS_CRC32_SIZE];
			kaddr = kmap_atomic(page, KM_IRQ0);
			ret = btrfs_csum_data(root, kaddr + bh_offset(bh),
					      bh->b_size, csum);
			BUG_ON(ret);
			if (memcmp(csum, &bh->b_private, BTRFS_CRC32_SIZE)) {
				u64 offset;
				offset = (page->index << PAGE_CACHE_SHIFT) +
					bh_offset(bh);
				printk("btrfs csum failed ino %lu off %llu\n",
				       page->mapping->host->i_ino,
				       (unsigned long long)offset);
				memset(kaddr + bh_offset(bh), 1, bh->b_size);
				flush_dcache_page(page);
			}
			kunmap_atomic(kaddr, KM_IRQ0);
		}
		set_buffer_uptodate(bh);
	} else {
		clear_buffer_uptodate(bh);
		if (printk_ratelimit())
			buffer_io_error(bh);
		SetPageError(page);
	}

	/*
	 * Be _very_ careful from here on. Bad things can happen if
	 * two buffer heads end IO at almost the same time and both
	 * decide that the page is now completely done.
	 */
	first = page_buffers(page);
	local_irq_save(flags);
	bit_spin_lock(BH_Uptodate_Lock, &first->b_state);
	clear_buffer_async_read(bh);
	unlock_buffer(bh);
	tmp = bh;
	do {
		if (!buffer_uptodate(tmp))
			page_uptodate = 0;
		if (buffer_async_read(tmp)) {
			BUG_ON(!buffer_locked(tmp));
			goto still_busy;
		}
		tmp = tmp->b_this_page;
	} while (tmp != bh);
	bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
	local_irq_restore(flags);

	/*
	 * If none of the buffers had errors and they are all
	 * uptodate then we can set the page uptodate.
	 */
	if (page_uptodate && !PageError(page))
		SetPageUptodate(page);
	unlock_page(page);
	return;

still_busy:
	bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
	local_irq_restore(flags);
	return;
}

/*
 * Generic "read page" function for block devices that have the normal
 * get_block functionality. This is most of the block device filesystems.
 * Reads the page asynchronously --- the unlock_buffer() and
 * set/clear_buffer_uptodate() functions propagate buffer state into the
 * page struct once IO has completed.
 */
int btrfs_readpage(struct file *file, struct page *page)
{
	struct inode *inode = page->mapping->host;
	sector_t iblock, lblock;
	struct buffer_head *bh, *head, *arr[MAX_BUF_PER_PAGE];
	unsigned int blocksize;
	int nr, i;
	int fully_mapped = 1;

	BUG_ON(!PageLocked(page));
	blocksize = 1 << inode->i_blkbits;
	if (!page_has_buffers(page))
		create_empty_buffers(page, blocksize, 0);
	head = page_buffers(page);

	iblock = (sector_t)page->index << (PAGE_CACHE_SHIFT - inode->i_blkbits);
	lblock = (i_size_read(inode)+blocksize-1) >> inode->i_blkbits;
	bh = head;
	nr = 0;
	i = 0;

	do {
		if (buffer_uptodate(bh))
			continue;

		if (!buffer_mapped(bh)) {
			int err = 0;

			fully_mapped = 0;
			if (iblock < lblock) {
				WARN_ON(bh->b_size != blocksize);
				err = btrfs_get_block_csum(inode, iblock,
							   bh, 0);
				if (err)
					SetPageError(page);
			}
			if (!buffer_mapped(bh)) {
				void *kaddr = kmap_atomic(page, KM_USER0);
				memset(kaddr + i * blocksize, 0, blocksize);
				flush_dcache_page(page);
				kunmap_atomic(kaddr, KM_USER0);
				if (!err)
					set_buffer_uptodate(bh);
				continue;
			}
			/*
			 * get_block() might have updated the buffer
			 * synchronously
			 */
			if (buffer_uptodate(bh))
				continue;
		}
		arr[nr++] = bh;
	} while (i++, iblock++, (bh = bh->b_this_page) != head);

	if (fully_mapped)
		SetPageMappedToDisk(page);

	if (!nr) {
		/*
		 * All buffers are uptodate - we can set the page uptodate
		 * as well. But not if get_block() returned an error.
		 */
		if (!PageError(page))
			SetPageUptodate(page);
		unlock_page(page);
		return 0;
	}

	/* Stage two: lock the buffers */
	for (i = 0; i < nr; i++) {
		bh = arr[i];
		lock_buffer(bh);
		bh->b_end_io = btrfs_end_buffer_async_read;
		set_buffer_async_read(bh);
	}

	/*
	 * Stage 3: start the IO.  Check for uptodateness
	 * inside the buffer lock in case another process reading
	 * the underlying blockdev brought it uptodate (the sct fix).
	 */
	for (i = 0; i < nr; i++) {
		bh = arr[i];
		if (buffer_uptodate(bh))
			btrfs_end_buffer_async_read(bh, 1);
		else
			submit_bh(READ, bh);
	}
	return 0;
C
Chris Mason 已提交
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
}

/*
 * Aside from a tiny bit of packed file data handling, this is the
 * same as the generic code.
 *
 * While block_write_full_page is writing back the dirty buffers under
 * the page lock, whoever dirtied the buffers may decide to clean them
 * again at any time.  We handle that by only looking at the buffer
 * state inside lock_buffer().
 *
 * If block_write_full_page() is called for regular writeback
 * (wbc->sync_mode == WB_SYNC_NONE) then it will redirty a page which has a
 * locked buffer.   This only can happen if someone has written the buffer
 * directly, with submit_bh().  At the address_space level PageWriteback
 * prevents this contention from occurring.
 */
static int __btrfs_write_full_page(struct inode *inode, struct page *page,
				   struct writeback_control *wbc)
{
	int err;
	sector_t block;
	sector_t last_block;
	struct buffer_head *bh, *head;
	const unsigned blocksize = 1 << inode->i_blkbits;
	int nr_underway = 0;
C
Chris Mason 已提交
1678
	struct btrfs_root *root = BTRFS_I(inode)->root;
C
Chris Mason 已提交
1679 1680 1681 1682 1683

	BUG_ON(!PageLocked(page));

	last_block = (i_size_read(inode) - 1) >> inode->i_blkbits;

1684 1685 1686 1687 1688 1689 1690
	/* no csumming allowed when from PF_MEMALLOC */
	if (current->flags & PF_MEMALLOC) {
		redirty_page_for_writepage(wbc, page);
		unlock_page(page);
		return 0;
	}

C
Chris Mason 已提交
1691 1692 1693 1694 1695 1696 1697 1698 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 1750 1751 1752 1753 1754 1755 1756 1757
	if (!page_has_buffers(page)) {
		create_empty_buffers(page, blocksize,
					(1 << BH_Dirty)|(1 << BH_Uptodate));
	}

	/*
	 * Be very careful.  We have no exclusion from __set_page_dirty_buffers
	 * here, and the (potentially unmapped) buffers may become dirty at
	 * any time.  If a buffer becomes dirty here after we've inspected it
	 * then we just miss that fact, and the page stays dirty.
	 *
	 * Buffers outside i_size may be dirtied by __set_page_dirty_buffers;
	 * handle that here by just cleaning them.
	 */

	block = (sector_t)page->index << (PAGE_CACHE_SHIFT - inode->i_blkbits);
	head = page_buffers(page);
	bh = head;

	/*
	 * Get all the dirty buffers mapped to disk addresses and
	 * handle any aliases from the underlying blockdev's mapping.
	 */
	do {
		if (block > last_block) {
			/*
			 * mapped buffers outside i_size will occur, because
			 * this page can be outside i_size when there is a
			 * truncate in progress.
			 */
			/*
			 * The buffer was zeroed by block_write_full_page()
			 */
			clear_buffer_dirty(bh);
			set_buffer_uptodate(bh);
		} else if (!buffer_mapped(bh) && buffer_dirty(bh)) {
			WARN_ON(bh->b_size != blocksize);
			err = btrfs_get_block(inode, block, bh, 0);
			if (err) {
				goto recover;
			}
			if (buffer_new(bh)) {
				/* blockdev mappings never come here */
				clear_buffer_new(bh);
			}
		}
		bh = bh->b_this_page;
		block++;
	} while (bh != head);

	do {
		if (!buffer_mapped(bh))
			continue;
		/*
		 * If it's a fully non-blocking write attempt and we cannot
		 * lock the buffer then redirty the page.  Note that this can
		 * potentially cause a busy-wait loop from pdflush and kswapd
		 * activity, but those code paths have their own higher-level
		 * throttling.
		 */
		if (wbc->sync_mode != WB_SYNC_NONE || !wbc->nonblocking) {
			lock_buffer(bh);
		} else if (test_set_buffer_locked(bh)) {
			redirty_page_for_writepage(wbc, page);
			continue;
		}
		if (test_clear_buffer_dirty(bh) && bh->b_blocknr != 0) {
C
Chris Mason 已提交
1758 1759 1760 1761 1762 1763 1764 1765 1766 1767
			struct btrfs_trans_handle *trans;
			int ret;
			u64 off = page->index << PAGE_CACHE_SHIFT;
			char *kaddr;

			off += bh_offset(bh);
			mutex_lock(&root->fs_info->fs_mutex);
			trans = btrfs_start_transaction(root, 1);
			btrfs_set_trans_block_group(trans, inode);
			kaddr = kmap(page);
1768
			btrfs_csum_file_block(trans, root, inode->i_ino,
C
Chris Mason 已提交
1769 1770 1771 1772 1773 1774
						    off, kaddr + bh_offset(bh),
						    bh->b_size);
			kunmap(page);
			ret = btrfs_end_transaction(trans, root);
			BUG_ON(ret);
			mutex_unlock(&root->fs_info->fs_mutex);
C
Chris Mason 已提交
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 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895
			mark_buffer_async_write(bh);
		} else {
			unlock_buffer(bh);
		}
	} while ((bh = bh->b_this_page) != head);

	/*
	 * The page and its buffers are protected by PageWriteback(), so we can
	 * drop the bh refcounts early.
	 */
	BUG_ON(PageWriteback(page));
	set_page_writeback(page);

	do {
		struct buffer_head *next = bh->b_this_page;
		if (buffer_async_write(bh)) {
			submit_bh(WRITE, bh);
			nr_underway++;
		}
		bh = next;
	} while (bh != head);
	unlock_page(page);

	err = 0;
done:
	if (nr_underway == 0) {
		/*
		 * The page was marked dirty, but the buffers were
		 * clean.  Someone wrote them back by hand with
		 * ll_rw_block/submit_bh.  A rare case.
		 */
		int uptodate = 1;
		do {
			if (!buffer_uptodate(bh)) {
				uptodate = 0;
				break;
			}
			bh = bh->b_this_page;
		} while (bh != head);
		if (uptodate)
			SetPageUptodate(page);
		end_page_writeback(page);
	}
	return err;

recover:
	/*
	 * ENOSPC, or some other error.  We may already have added some
	 * blocks to the file, so we need to write these out to avoid
	 * exposing stale data.
	 * The page is currently locked and not marked for writeback
	 */
	bh = head;
	/* Recovery: lock and submit the mapped buffers */
	do {
		if (buffer_mapped(bh) && buffer_dirty(bh)) {
			lock_buffer(bh);
			mark_buffer_async_write(bh);
		} else {
			/*
			 * The buffer may have been set dirty during
			 * attachment to a dirty page.
			 */
			clear_buffer_dirty(bh);
		}
	} while ((bh = bh->b_this_page) != head);
	SetPageError(page);
	BUG_ON(PageWriteback(page));
	set_page_writeback(page);
	do {
		struct buffer_head *next = bh->b_this_page;
		if (buffer_async_write(bh)) {
			clear_buffer_dirty(bh);
			submit_bh(WRITE, bh);
			nr_underway++;
		}
		bh = next;
	} while (bh != head);
	unlock_page(page);
	goto done;
}

static int btrfs_writepage(struct page *page, struct writeback_control *wbc)
{
	struct inode * const inode = page->mapping->host;
	loff_t i_size = i_size_read(inode);
	const pgoff_t end_index = i_size >> PAGE_CACHE_SHIFT;
	unsigned offset;
	void *kaddr;

	/* Is the page fully inside i_size? */
	if (page->index < end_index)
		return __btrfs_write_full_page(inode, page, wbc);

	/* Is the page fully outside i_size? (truncate in progress) */
	offset = i_size & (PAGE_CACHE_SIZE-1);
	if (page->index >= end_index+1 || !offset) {
		/*
		 * The page may have dirty, unmapped buffers.  For example,
		 * they may have been added in ext3_writepage().  Make them
		 * freeable here, so the page does not leak.
		 */
		block_invalidatepage(page, 0);
		unlock_page(page);
		return 0; /* don't care */
	}

	/*
	 * The page straddles i_size.  It must be zeroed out on each and every
	 * writepage invokation because it may be mmapped.  "A file is mapped
	 * in multiples of the page size.  For a file that is not a multiple of
	 * the  page size, the remaining memory is zeroed when mapped, and
	 * writes to that region are not written out to the file."
	 */
	kaddr = kmap_atomic(page, KM_USER0);
	memset(kaddr + offset, 0, PAGE_CACHE_SIZE - offset);
	flush_dcache_page(page);
	kunmap_atomic(kaddr, KM_USER0);
	return __btrfs_write_full_page(inode, page, wbc);
}

C
Chris Mason 已提交
1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941
/*
 * btrfs_page_mkwrite() is not allowed to change the file size as it gets
 * called from a page fault handler when a page is first dirtied. Hence we must
 * be careful to check for EOF conditions here. We set the page up correctly
 * for a written page which means we get ENOSPC checking when writing into
 * holes and correct delalloc and unwritten extent mapping on filesystems that
 * support these features.
 *
 * We are not allowed to take the i_mutex here so we have to play games to
 * protect against truncate races as the page could now be beyond EOF.  Because
 * vmtruncate() writes the inode size before removing pages, once we have the
 * page lock we can determine safely if the page is beyond EOF. If it is not
 * beyond EOF, then the page is guaranteed safe against truncation until we
 * unlock the page.
 */
int btrfs_page_mkwrite(struct vm_area_struct *vma, struct page *page)
{
	struct inode *inode = vma->vm_file->f_path.dentry->d_inode;
	unsigned long end;
	loff_t size;
	int ret = -EINVAL;

	lock_page(page);
	wait_on_page_writeback(page);
	size = i_size_read(inode);
	if ((page->mapping != inode->i_mapping) ||
	    ((page->index << PAGE_CACHE_SHIFT) > size)) {
		/* page got truncated out from underneath us */
		goto out_unlock;
	}

	/* page is wholly or partially inside EOF */
	if (((page->index + 1) << PAGE_CACHE_SHIFT) > size)
		end = size & ~PAGE_CACHE_MASK;
	else
		end = PAGE_CACHE_SIZE;

	ret = btrfs_prepare_write(NULL, page, 0, end);
	if (!ret)
		ret = btrfs_commit_write(NULL, page, 0, end);

out_unlock:
	unlock_page(page);
	return ret;
}

C
Chris Mason 已提交
1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
static void btrfs_truncate(struct inode *inode)
{
	struct btrfs_root *root = BTRFS_I(inode)->root;
	int ret;
	struct btrfs_trans_handle *trans;

	if (!S_ISREG(inode->i_mode))
		return;
	if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
		return;

	btrfs_truncate_page(inode->i_mapping, inode->i_size);

	mutex_lock(&root->fs_info->fs_mutex);
	trans = btrfs_start_transaction(root, 1);
	btrfs_set_trans_block_group(trans, inode);

	/* FIXME, add redo link to tree so we don't leak on crash */
	ret = btrfs_truncate_in_trans(trans, root, inode);
	btrfs_update_inode(trans, root, inode);
	ret = btrfs_end_transaction(trans, root);
	BUG_ON(ret);
	mutex_unlock(&root->fs_info->fs_mutex);
	btrfs_btree_balance_dirty(root);
}

int btrfs_commit_write(struct file *file, struct page *page,
		       unsigned from, unsigned to)
{
	struct inode *inode = page->mapping->host;
	struct buffer_head *bh;
	loff_t pos = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to;

	SetPageUptodate(page);
	bh = page_buffers(page);
	set_buffer_uptodate(bh);
	if (buffer_mapped(bh) && bh->b_blocknr != 0) {
		set_page_dirty(page);
	}
	if (pos > inode->i_size) {
		i_size_write(inode, pos);
		mark_inode_dirty(inode);
	}
	return 0;
}

static int create_subvol(struct btrfs_root *root, char *name, int namelen)
{
	struct btrfs_trans_handle *trans;
	struct btrfs_key key;
	struct btrfs_root_item root_item;
	struct btrfs_inode_item *inode_item;
	struct buffer_head *subvol;
	struct btrfs_leaf *leaf;
	struct btrfs_root *new_root;
	struct inode *inode;
	struct inode *dir;
	int ret;
2000
	int err;
C
Chris Mason 已提交
2001 2002 2003 2004 2005 2006 2007 2008
	u64 objectid;
	u64 new_dirid = BTRFS_FIRST_FREE_OBJECTID;

	mutex_lock(&root->fs_info->fs_mutex);
	trans = btrfs_start_transaction(root, 1);
	BUG_ON(!trans);

	subvol = btrfs_alloc_free_block(trans, root, 0);
2009 2010
	if (IS_ERR(subvol))
		return PTR_ERR(subvol);
C
Chris Mason 已提交
2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030
	leaf = btrfs_buffer_leaf(subvol);
	btrfs_set_header_nritems(&leaf->header, 0);
	btrfs_set_header_level(&leaf->header, 0);
	btrfs_set_header_blocknr(&leaf->header, bh_blocknr(subvol));
	btrfs_set_header_generation(&leaf->header, trans->transid);
	btrfs_set_header_owner(&leaf->header, root->root_key.objectid);
	memcpy(leaf->header.fsid, root->fs_info->disk_super->fsid,
	       sizeof(leaf->header.fsid));
	mark_buffer_dirty(subvol);

	inode_item = &root_item.inode;
	memset(inode_item, 0, sizeof(*inode_item));
	btrfs_set_inode_generation(inode_item, 1);
	btrfs_set_inode_size(inode_item, 3);
	btrfs_set_inode_nlink(inode_item, 1);
	btrfs_set_inode_nblocks(inode_item, 1);
	btrfs_set_inode_mode(inode_item, S_IFDIR | 0755);

	btrfs_set_root_blocknr(&root_item, bh_blocknr(subvol));
	btrfs_set_root_refs(&root_item, 1);
2031 2032
	memset(&root_item.drop_progress, 0, sizeof(root_item.drop_progress));
	root_item.drop_level = 0;
C
Chris Mason 已提交
2033 2034 2035 2036 2037
	brelse(subvol);
	subvol = NULL;

	ret = btrfs_find_free_objectid(trans, root->fs_info->tree_root,
				       0, &objectid);
2038 2039
	if (ret)
		goto fail;
C
Chris Mason 已提交
2040 2041 2042 2043 2044 2045 2046 2047 2048

	btrfs_set_root_dirid(&root_item, new_dirid);

	key.objectid = objectid;
	key.offset = 1;
	key.flags = 0;
	btrfs_set_key_type(&key, BTRFS_ROOT_ITEM_KEY);
	ret = btrfs_insert_root(trans, root->fs_info->tree_root, &key,
				&root_item);
2049 2050
	if (ret)
		goto fail;
C
Chris Mason 已提交
2051 2052 2053 2054 2055 2056 2057 2058 2059

	/*
	 * insert the directory item
	 */
	key.offset = (u64)-1;
	dir = root->fs_info->sb->s_root->d_inode;
	ret = btrfs_insert_dir_item(trans, root->fs_info->tree_root,
				    name, namelen, dir->i_ino, &key,
				    BTRFS_FT_DIR);
2060 2061
	if (ret)
		goto fail;
C
Chris Mason 已提交
2062 2063

	ret = btrfs_commit_transaction(trans, root);
2064 2065
	if (ret)
		goto fail_commit;
C
Chris Mason 已提交
2066 2067 2068 2069 2070 2071 2072 2073 2074

	new_root = btrfs_read_fs_root(root->fs_info, &key);
	BUG_ON(!new_root);

	trans = btrfs_start_transaction(new_root, 1);
	BUG_ON(!trans);

	inode = btrfs_new_inode(trans, new_root, new_dirid,
				BTRFS_I(dir)->block_group, S_IFDIR | 0700);
2075 2076
	if (IS_ERR(inode))
		goto fail;
C
Chris Mason 已提交
2077 2078
	inode->i_op = &btrfs_dir_inode_operations;
	inode->i_fop = &btrfs_dir_file_operations;
C
Chris Mason 已提交
2079
	new_root->inode = inode;
C
Chris Mason 已提交
2080 2081

	ret = btrfs_make_empty_dir(trans, new_root, new_dirid, new_dirid);
2082 2083
	if (ret)
		goto fail;
C
Chris Mason 已提交
2084 2085 2086 2087

	inode->i_nlink = 1;
	inode->i_size = 6;
	ret = btrfs_update_inode(trans, new_root, inode);
2088 2089 2090 2091 2092 2093 2094
	if (ret)
		goto fail;
fail:
	err = btrfs_commit_transaction(trans, root);
	if (err && !ret)
		ret = err;
fail_commit:
C
Chris Mason 已提交
2095 2096
	mutex_unlock(&root->fs_info->fs_mutex);
	btrfs_btree_balance_dirty(root);
2097
	return ret;
C
Chris Mason 已提交
2098 2099 2100 2101 2102 2103 2104 2105
}

static int create_snapshot(struct btrfs_root *root, char *name, int namelen)
{
	struct btrfs_trans_handle *trans;
	struct btrfs_key key;
	struct btrfs_root_item new_root_item;
	int ret;
2106
	int err;
C
Chris Mason 已提交
2107 2108 2109 2110 2111 2112 2113 2114 2115 2116
	u64 objectid;

	if (!root->ref_cows)
		return -EINVAL;

	mutex_lock(&root->fs_info->fs_mutex);
	trans = btrfs_start_transaction(root, 1);
	BUG_ON(!trans);

	ret = btrfs_update_inode(trans, root, root->inode);
2117 2118
	if (ret)
		goto fail;
C
Chris Mason 已提交
2119 2120 2121

	ret = btrfs_find_free_objectid(trans, root->fs_info->tree_root,
				       0, &objectid);
2122 2123
	if (ret)
		goto fail;
C
Chris Mason 已提交
2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135

	memcpy(&new_root_item, &root->root_item,
	       sizeof(new_root_item));

	key.objectid = objectid;
	key.offset = 1;
	key.flags = 0;
	btrfs_set_key_type(&key, BTRFS_ROOT_ITEM_KEY);
	btrfs_set_root_blocknr(&new_root_item, bh_blocknr(root->node));

	ret = btrfs_insert_root(trans, root->fs_info->tree_root, &key,
				&new_root_item);
2136 2137
	if (ret)
		goto fail;
C
Chris Mason 已提交
2138 2139 2140 2141 2142 2143 2144 2145 2146 2147

	/*
	 * insert the directory item
	 */
	key.offset = (u64)-1;
	ret = btrfs_insert_dir_item(trans, root->fs_info->tree_root,
				    name, namelen,
				    root->fs_info->sb->s_root->d_inode->i_ino,
				    &key, BTRFS_FT_DIR);

2148 2149
	if (ret)
		goto fail;
C
Chris Mason 已提交
2150 2151

	ret = btrfs_inc_root_ref(trans, root);
2152 2153
	if (ret)
		goto fail;
C
Chris Mason 已提交
2154

2155 2156 2157 2158
fail:
	err = btrfs_commit_transaction(trans, root);
	if (err && !ret)
		ret = err;
C
Chris Mason 已提交
2159 2160
	mutex_unlock(&root->fs_info->fs_mutex);
	btrfs_btree_balance_dirty(root);
2161
	return ret;
C
Chris Mason 已提交
2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183
}

int btrfs_ioctl(struct inode *inode, struct file *filp, unsigned int
		cmd, unsigned long arg)
{
	struct btrfs_root *root = BTRFS_I(inode)->root;
	struct btrfs_ioctl_vol_args vol_args;
	int ret = 0;
	struct btrfs_dir_item *di;
	int namelen;
	struct btrfs_path *path;
	u64 root_dirid;

	switch (cmd) {
	case BTRFS_IOC_SNAP_CREATE:
		if (copy_from_user(&vol_args,
				   (struct btrfs_ioctl_vol_args __user *)arg,
				   sizeof(vol_args)))
			return -EFAULT;
		namelen = strlen(vol_args.name);
		if (namelen > BTRFS_VOL_NAME_MAX)
			return -EINVAL;
2184 2185
		if (strchr(vol_args.name, '/'))
			return -EINVAL;
C
Chris Mason 已提交
2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197
		path = btrfs_alloc_path();
		if (!path)
			return -ENOMEM;
		root_dirid = root->fs_info->sb->s_root->d_inode->i_ino,
		mutex_lock(&root->fs_info->fs_mutex);
		di = btrfs_lookup_dir_item(NULL, root->fs_info->tree_root,
				    path, root_dirid,
				    vol_args.name, namelen, 0);
		mutex_unlock(&root->fs_info->fs_mutex);
		btrfs_free_path(path);
		if (di && !IS_ERR(di))
			return -EEXIST;
2198 2199
		if (IS_ERR(di))
			return PTR_ERR(di);
C
Chris Mason 已提交
2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409

		if (root == root->fs_info->tree_root)
			ret = create_subvol(root, vol_args.name, namelen);
		else
			ret = create_snapshot(root, vol_args.name, namelen);
		break;
	default:
		return -ENOTTY;
	}
	return ret;
}

#ifdef CONFIG_COMPAT
long btrfs_compat_ioctl(struct file *file, unsigned int cmd,
			       unsigned long arg)
{
	struct inode *inode = file->f_path.dentry->d_inode;
	int ret;
	lock_kernel();
	ret = btrfs_ioctl(inode, file, cmd, (unsigned long) compat_ptr(arg));
	unlock_kernel();
	return ret;

}
#endif

/*
 * Called inside transaction, so use GFP_NOFS
 */
struct inode *btrfs_alloc_inode(struct super_block *sb)
{
	struct btrfs_inode *ei;

	ei = kmem_cache_alloc(btrfs_inode_cachep, GFP_NOFS);
	if (!ei)
		return NULL;
	return &ei->vfs_inode;
}

void btrfs_destroy_inode(struct inode *inode)
{
	WARN_ON(!list_empty(&inode->i_dentry));
	WARN_ON(inode->i_data.nrpages);

	kmem_cache_free(btrfs_inode_cachep, BTRFS_I(inode));
}

static void init_once(void * foo, struct kmem_cache * cachep,
		      unsigned long flags)
{
	struct btrfs_inode *ei = (struct btrfs_inode *) foo;

	inode_init_once(&ei->vfs_inode);
}

void btrfs_destroy_cachep(void)
{
	if (btrfs_inode_cachep)
		kmem_cache_destroy(btrfs_inode_cachep);
	if (btrfs_trans_handle_cachep)
		kmem_cache_destroy(btrfs_trans_handle_cachep);
	if (btrfs_transaction_cachep)
		kmem_cache_destroy(btrfs_transaction_cachep);
	if (btrfs_bit_radix_cachep)
		kmem_cache_destroy(btrfs_bit_radix_cachep);
	if (btrfs_path_cachep)
		kmem_cache_destroy(btrfs_path_cachep);
}

int btrfs_init_cachep(void)
{
	btrfs_inode_cachep = kmem_cache_create("btrfs_inode_cache",
					     sizeof(struct btrfs_inode),
					     0, (SLAB_RECLAIM_ACCOUNT|
						SLAB_MEM_SPREAD),
					     init_once, NULL);
	if (!btrfs_inode_cachep)
		goto fail;
	btrfs_trans_handle_cachep = kmem_cache_create("btrfs_trans_handle_cache",
					     sizeof(struct btrfs_trans_handle),
					     0, (SLAB_RECLAIM_ACCOUNT|
						SLAB_MEM_SPREAD),
					     NULL, NULL);
	if (!btrfs_trans_handle_cachep)
		goto fail;
	btrfs_transaction_cachep = kmem_cache_create("btrfs_transaction_cache",
					     sizeof(struct btrfs_transaction),
					     0, (SLAB_RECLAIM_ACCOUNT|
						SLAB_MEM_SPREAD),
					     NULL, NULL);
	if (!btrfs_transaction_cachep)
		goto fail;
	btrfs_path_cachep = kmem_cache_create("btrfs_path_cache",
					     sizeof(struct btrfs_transaction),
					     0, (SLAB_RECLAIM_ACCOUNT|
						SLAB_MEM_SPREAD),
					     NULL, NULL);
	if (!btrfs_path_cachep)
		goto fail;
	btrfs_bit_radix_cachep = kmem_cache_create("btrfs_radix",
					     256,
					     0, (SLAB_RECLAIM_ACCOUNT|
						SLAB_MEM_SPREAD |
						SLAB_DESTROY_BY_RCU),
					     NULL, NULL);
	if (!btrfs_bit_radix_cachep)
		goto fail;
	return 0;
fail:
	btrfs_destroy_cachep();
	return -ENOMEM;
}

static int btrfs_getattr(struct vfsmount *mnt,
			 struct dentry *dentry, struct kstat *stat)
{
	struct inode *inode = dentry->d_inode;
	generic_fillattr(inode, stat);
	stat->blksize = 256 * 1024;
	return 0;
}

static int btrfs_rename(struct inode * old_dir, struct dentry *old_dentry,
			   struct inode * new_dir,struct dentry *new_dentry)
{
	struct btrfs_trans_handle *trans;
	struct btrfs_root *root = BTRFS_I(old_dir)->root;
	struct inode *new_inode = new_dentry->d_inode;
	struct inode *old_inode = old_dentry->d_inode;
	struct timespec ctime = CURRENT_TIME;
	struct btrfs_path *path;
	struct btrfs_dir_item *di;
	int ret;

	if (S_ISDIR(old_inode->i_mode) && new_inode &&
	    new_inode->i_size > BTRFS_EMPTY_DIR_SIZE) {
		return -ENOTEMPTY;
	}
	mutex_lock(&root->fs_info->fs_mutex);
	trans = btrfs_start_transaction(root, 1);
	btrfs_set_trans_block_group(trans, new_dir);
	path = btrfs_alloc_path();
	if (!path) {
		ret = -ENOMEM;
		goto out_fail;
	}

	old_dentry->d_inode->i_nlink++;
	old_dir->i_ctime = old_dir->i_mtime = ctime;
	new_dir->i_ctime = new_dir->i_mtime = ctime;
	old_inode->i_ctime = ctime;
	if (S_ISDIR(old_inode->i_mode) && old_dir != new_dir) {
		struct btrfs_key *location = &BTRFS_I(new_dir)->location;
		u64 old_parent_oid;
		di = btrfs_lookup_dir_item(trans, root, path, old_inode->i_ino,
					   "..", 2, -1);
		if (IS_ERR(di)) {
			ret = PTR_ERR(di);
			goto out_fail;
		}
		if (!di) {
			ret = -ENOENT;
			goto out_fail;
		}
		old_parent_oid = btrfs_disk_key_objectid(&di->location);
		ret = btrfs_del_item(trans, root, path);
		if (ret) {
			goto out_fail;
		}
		btrfs_release_path(root, path);

		di = btrfs_lookup_dir_index_item(trans, root, path,
						 old_inode->i_ino,
						 old_parent_oid,
						 "..", 2, -1);
		if (IS_ERR(di)) {
			ret = PTR_ERR(di);
			goto out_fail;
		}
		if (!di) {
			ret = -ENOENT;
			goto out_fail;
		}
		ret = btrfs_del_item(trans, root, path);
		if (ret) {
			goto out_fail;
		}
		btrfs_release_path(root, path);

		ret = btrfs_insert_dir_item(trans, root, "..", 2,
					    old_inode->i_ino, location,
					    BTRFS_FT_DIR);
		if (ret)
			goto out_fail;
	}


	ret = btrfs_unlink_trans(trans, root, old_dir, old_dentry);
	if (ret)
		goto out_fail;

	if (new_inode) {
		new_inode->i_ctime = CURRENT_TIME;
		ret = btrfs_unlink_trans(trans, root, new_dir, new_dentry);
		if (ret)
			goto out_fail;
		if (S_ISDIR(new_inode->i_mode))
			clear_nlink(new_inode);
		else
			drop_nlink(new_inode);
2410 2411 2412
		ret = btrfs_update_inode(trans, root, new_inode);
		if (ret)
			goto out_fail;
C
Chris Mason 已提交
2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483
	}
	ret = btrfs_add_link(trans, new_dentry, old_inode);
	if (ret)
		goto out_fail;

out_fail:
	btrfs_free_path(path);
	btrfs_end_transaction(trans, root);
	mutex_unlock(&root->fs_info->fs_mutex);
	return ret;
}

static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
			 const char *symname)
{
	struct btrfs_trans_handle *trans;
	struct btrfs_root *root = BTRFS_I(dir)->root;
	struct btrfs_path *path;
	struct btrfs_key key;
	struct inode *inode;
	int err;
	int drop_inode = 0;
	u64 objectid;
	int name_len;
	int datasize;
	char *ptr;
	struct btrfs_file_extent_item *ei;

	name_len = strlen(symname) + 1;
	if (name_len > BTRFS_MAX_INLINE_DATA_SIZE(root))
		return -ENAMETOOLONG;
	mutex_lock(&root->fs_info->fs_mutex);
	trans = btrfs_start_transaction(root, 1);
	btrfs_set_trans_block_group(trans, dir);

	err = btrfs_find_free_objectid(trans, root, dir->i_ino, &objectid);
	if (err) {
		err = -ENOSPC;
		goto out_unlock;
	}

	inode = btrfs_new_inode(trans, root, objectid,
				BTRFS_I(dir)->block_group, S_IFLNK|S_IRWXUGO);
	err = PTR_ERR(inode);
	if (IS_ERR(inode))
		goto out_unlock;

	btrfs_set_trans_block_group(trans, inode);
	err = btrfs_add_nondir(trans, dentry, inode);
	if (err)
		drop_inode = 1;
	else {
		inode->i_mapping->a_ops = &btrfs_aops;
		inode->i_fop = &btrfs_file_operations;
		inode->i_op = &btrfs_file_inode_operations;
	}
	dir->i_sb->s_dirt = 1;
	btrfs_update_inode_block_group(trans, inode);
	btrfs_update_inode_block_group(trans, dir);
	if (drop_inode)
		goto out_unlock;

	path = btrfs_alloc_path();
	BUG_ON(!path);
	key.objectid = inode->i_ino;
	key.offset = 0;
	key.flags = 0;
	btrfs_set_key_type(&key, BTRFS_EXTENT_DATA_KEY);
	datasize = btrfs_file_extent_calc_inline_size(name_len);
	err = btrfs_insert_empty_item(trans, root, path, &key,
				      datasize);
2484 2485 2486 2487
	if (err) {
		drop_inode = 1;
		goto out_unlock;
	}
C
Chris Mason 已提交
2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500
	ei = btrfs_item_ptr(btrfs_buffer_leaf(path->nodes[0]),
	       path->slots[0], struct btrfs_file_extent_item);
	btrfs_set_file_extent_generation(ei, trans->transid);
	btrfs_set_file_extent_type(ei,
				   BTRFS_FILE_EXTENT_INLINE);
	ptr = btrfs_file_extent_inline_start(ei);
	btrfs_memcpy(root, path->nodes[0]->b_data,
		     ptr, symname, name_len);
	mark_buffer_dirty(path->nodes[0]);
	btrfs_free_path(path);
	inode->i_op = &btrfs_symlink_inode_operations;
	inode->i_mapping->a_ops = &btrfs_symlink_aops;
	inode->i_size = name_len - 1;
2501 2502 2503
	err = btrfs_update_inode(trans, root, inode);
	if (err)
		drop_inode = 1;
C
Chris Mason 已提交
2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566

out_unlock:
	btrfs_end_transaction(trans, root);
	mutex_unlock(&root->fs_info->fs_mutex);
	if (drop_inode) {
		inode_dec_link_count(inode);
		iput(inode);
	}
	btrfs_btree_balance_dirty(root);
	return err;
}

static struct inode_operations btrfs_dir_inode_operations = {
	.lookup		= btrfs_lookup,
	.create		= btrfs_create,
	.unlink		= btrfs_unlink,
	.link		= btrfs_link,
	.mkdir		= btrfs_mkdir,
	.rmdir		= btrfs_rmdir,
	.rename		= btrfs_rename,
	.symlink	= btrfs_symlink,
	.setattr	= btrfs_setattr,
};

static struct inode_operations btrfs_dir_ro_inode_operations = {
	.lookup		= btrfs_lookup,
};

static struct file_operations btrfs_dir_file_operations = {
	.llseek		= generic_file_llseek,
	.read		= generic_read_dir,
	.readdir	= btrfs_readdir,
	.ioctl		= btrfs_ioctl,
#ifdef CONFIG_COMPAT
	.compat_ioctl	= btrfs_compat_ioctl,
#endif
};

static struct address_space_operations btrfs_aops = {
	.readpage	= btrfs_readpage,
	.writepage	= btrfs_writepage,
	.sync_page	= block_sync_page,
	.prepare_write	= btrfs_prepare_write,
	.commit_write	= btrfs_commit_write,
	.bmap		= btrfs_bmap,
};

static struct address_space_operations btrfs_symlink_aops = {
	.readpage	= btrfs_readpage,
	.writepage	= btrfs_writepage,
};

static struct inode_operations btrfs_file_inode_operations = {
	.truncate	= btrfs_truncate,
	.getattr	= btrfs_getattr,
	.setattr	= btrfs_setattr,
};

static struct inode_operations btrfs_symlink_inode_operations = {
	.readlink	= generic_readlink,
	.follow_link	= page_follow_link_light,
	.put_link	= page_put_link,
};