disk-io.c

/*
 * 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.
 */

#include <linux/fs.h>
#include <linux/blkdev.h>
#include <linux/scatterlist.h>
#include <linux/swap.h>
#include <linux/radix-tree.h>
#include <linux/writeback.h>
#include <linux/buffer_head.h>
#include <linux/workqueue.h>
#include <linux/kthread.h>
#include <linux/freezer.h>
#include <linux/crc32c.h>
#include <linux/slab.h>
#include <linux/migrate.h>
#include <linux/ratelimit.h>
#include <asm/unaligned.h>
#include "compat.h"
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
#include "btrfs_inode.h"
#include "volumes.h"
#include "print-tree.h"
#include "async-thread.h"
#include "locking.h"
#include "tree-log.h"
#include "free-space-cache.h"
#include "inode-map.h"
#include "check-integrity.h"
#include "rcu-string.h"
#include "dev-replace.h"
#include "raid56.h"

#ifdef CONFIG_X86
#include <asm/cpufeature.h>
#endif

static struct extent_io_ops btree_extent_io_ops;
static void end_workqueue_fn(struct btrfs_work *work);
static void free_fs_root(struct btrfs_root *root);
static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
				    int read_only);
static void btrfs_destroy_ordered_operations(struct btrfs_transaction *t,
					     struct btrfs_root *root);
static void btrfs_destroy_ordered_extents(struct btrfs_root *root);
static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
				      struct btrfs_root *root);
static void btrfs_destroy_pending_snapshots(struct btrfs_transaction *t);
static void btrfs_destroy_delalloc_inodes(struct btrfs_root *root);
static int btrfs_destroy_marked_extents(struct btrfs_root *root,
					struct extent_io_tree *dirty_pages,
					int mark);
static int btrfs_destroy_pinned_extent(struct btrfs_root *root,
				       struct extent_io_tree *pinned_extents);

/*
 * end_io_wq structs are used to do processing in task context when an IO is
 * complete.  This is used during reads to verify checksums, and it is used
 * by writes to insert metadata for new file extents after IO is complete.
 */
struct end_io_wq {
	struct bio *bio;
	bio_end_io_t *end_io;
	void *private;
	struct btrfs_fs_info *info;
	int error;
	int metadata;
	struct list_head list;
	struct btrfs_work work;
};

/*
 * async submit bios are used to offload expensive checksumming
 * onto the worker threads.  They checksum file and metadata bios
 * just before they are sent down the IO stack.
 */
struct async_submit_bio {
	struct inode *inode;
	struct bio *bio;
	struct list_head list;
	extent_submit_bio_hook_t *submit_bio_start;
	extent_submit_bio_hook_t *submit_bio_done;
	int rw;
	int mirror_num;
	unsigned long bio_flags;
	/*
	 * bio_offset is optional, can be used if the pages in the bio
	 * can't tell us where in the file the bio should go
	 */
	u64 bio_offset;
	struct btrfs_work work;
	int error;
};

/*
 * Lockdep class keys for extent_buffer->lock's in this root.  For a given
 * eb, the lockdep key is determined by the btrfs_root it belongs to and
 * the level the eb occupies in the tree.
 *
 * Different roots are used for different purposes and may nest inside each
 * other and they require separate keysets.  As lockdep keys should be
 * static, assign keysets according to the purpose of the root as indicated
 * by btrfs_root->objectid.  This ensures that all special purpose roots
 * have separate keysets.
 *
 * Lock-nesting across peer nodes is always done with the immediate parent
 * node locked thus preventing deadlock.  As lockdep doesn't know this, use
 * subclass to avoid triggering lockdep warning in such cases.
 *
 * The key is set by the readpage_end_io_hook after the buffer has passed
 * csum validation but before the pages are unlocked.  It is also set by
 * btrfs_init_new_buffer on freshly allocated blocks.
 *
 * We also add a check to make sure the highest level of the tree is the
 * same as our lockdep setup here.  If BTRFS_MAX_LEVEL changes, this code
 * needs update as well.
 */
#ifdef CONFIG_DEBUG_LOCK_ALLOC
# if BTRFS_MAX_LEVEL != 8
#  error
# endif

static struct btrfs_lockdep_keyset {
	u64			id;		/* root objectid */
	const char		*name_stem;	/* lock name stem */
	char			names[BTRFS_MAX_LEVEL + 1][20];
	struct lock_class_key	keys[BTRFS_MAX_LEVEL + 1];
} btrfs_lockdep_keysets[] = {
	{ .id = BTRFS_ROOT_TREE_OBJECTID,	.name_stem = "root"	},
	{ .id = BTRFS_EXTENT_TREE_OBJECTID,	.name_stem = "extent"	},
	{ .id = BTRFS_CHUNK_TREE_OBJECTID,	.name_stem = "chunk"	},
	{ .id = BTRFS_DEV_TREE_OBJECTID,	.name_stem = "dev"	},
	{ .id = BTRFS_FS_TREE_OBJECTID,		.name_stem = "fs"	},
	{ .id = BTRFS_CSUM_TREE_OBJECTID,	.name_stem = "csum"	},
	{ .id = BTRFS_ORPHAN_OBJECTID,		.name_stem = "orphan"	},
	{ .id = BTRFS_TREE_LOG_OBJECTID,	.name_stem = "log"	},
	{ .id = BTRFS_TREE_RELOC_OBJECTID,	.name_stem = "treloc"	},
	{ .id = BTRFS_DATA_RELOC_TREE_OBJECTID,	.name_stem = "dreloc"	},
	{ .id = 0,				.name_stem = "tree"	},
};

void __init btrfs_init_lockdep(void)
{
	int i, j;

	/* initialize lockdep class names */
	for (i = 0; i < ARRAY_SIZE(btrfs_lockdep_keysets); i++) {
		struct btrfs_lockdep_keyset *ks = &btrfs_lockdep_keysets[i];

		for (j = 0; j < ARRAY_SIZE(ks->names); j++)
			snprintf(ks->names[j], sizeof(ks->names[j]),
				 "btrfs-%s-%02d", ks->name_stem, j);
	}
}

void btrfs_set_buffer_lockdep_class(u64 objectid, struct extent_buffer *eb,
				    int level)
{
	struct btrfs_lockdep_keyset *ks;

	BUG_ON(level >= ARRAY_SIZE(ks->keys));

	/* find the matching keyset, id 0 is the default entry */
	for (ks = btrfs_lockdep_keysets; ks->id; ks++)
		if (ks->id == objectid)
			break;

	lockdep_set_class_and_name(&eb->lock,
				   &ks->keys[level], ks->names[level]);
}

#endif

/*
 * extents on the btree inode are pretty simple, there's one extent
 * that covers the entire device
 */
static struct extent_map *btree_get_extent(struct inode *inode,
		struct page *page, size_t pg_offset, u64 start, u64 len,
		int create)
{
	struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
	struct extent_map *em;
	int ret;

	read_lock(&em_tree->lock);
	em = lookup_extent_mapping(em_tree, start, len);
	if (em) {
		em->bdev =
			BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev;
		read_unlock(&em_tree->lock);
		goto out;
	}
	read_unlock(&em_tree->lock);

	em = alloc_extent_map();
	if (!em) {
		em = ERR_PTR(-ENOMEM);
		goto out;
	}
	em->start = 0;
	em->len = (u64)-1;
	em->block_len = (u64)-1;
	em->block_start = 0;
	em->bdev = BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev;

	write_lock(&em_tree->lock);
	ret = add_extent_mapping(em_tree, em);
	if (ret == -EEXIST) {
		free_extent_map(em);
		em = lookup_extent_mapping(em_tree, start, len);
		if (!em)
			em = ERR_PTR(-EIO);
	} else if (ret) {
		free_extent_map(em);
		em = ERR_PTR(ret);
	}
	write_unlock(&em_tree->lock);

out:
	return em;
}

u32 btrfs_csum_data(struct btrfs_root *root, char *data, u32 seed, size_t len)
{
	return crc32c(seed, data, len);
}

void btrfs_csum_final(u32 crc, char *result)
{
	put_unaligned_le32(~crc, result);
}

/*
 * compute the csum for a btree block, and either verify it or write it
 * into the csum field of the block.
 */
static int csum_tree_block(struct btrfs_root *root, struct extent_buffer *buf,
			   int verify)
{
	u16 csum_size = btrfs_super_csum_size(root->fs_info->super_copy);
	char *result = NULL;
	unsigned long len;
	unsigned long cur_len;
	unsigned long offset = BTRFS_CSUM_SIZE;
	char *kaddr;
	unsigned long map_start;
	unsigned long map_len;
	int err;
	u32 crc = ~(u32)0;
	unsigned long inline_result;

	len = buf->len - offset;
	while (len > 0) {
		err = map_private_extent_buffer(buf, offset, 32,
					&kaddr, &map_start, &map_len);
		if (err)
			return 1;
		cur_len = min(len, map_len - (offset - map_start));
		crc = btrfs_csum_data(root, kaddr + offset - map_start,
				      crc, cur_len);
		len -= cur_len;
		offset += cur_len;
	}
	if (csum_size > sizeof(inline_result)) {
		result = kzalloc(csum_size * sizeof(char), GFP_NOFS);
		if (!result)
			return 1;
	} else {
		result = (char *)&inline_result;
	}

	btrfs_csum_final(crc, result);

	if (verify) {
		if (memcmp_extent_buffer(buf, result, 0, csum_size)) {
			u32 val;
			u32 found = 0;
			memcpy(&found, result, csum_size);

			read_extent_buffer(buf, &val, 0, csum_size);
			printk_ratelimited(KERN_INFO "btrfs: %s checksum verify "
				       "failed on %llu wanted %X found %X "
				       "level %d\n",
				       root->fs_info->sb->s_id,
				       (unsigned long long)buf->start, val, found,
				       btrfs_header_level(buf));
			if (result != (char *)&inline_result)
				kfree(result);
			return 1;
		}
	} else {
		write_extent_buffer(buf, result, 0, csum_size);
	}
	if (result != (char *)&inline_result)
		kfree(result);
	return 0;
}

/*
 * we can't consider a given block up to date unless the transid of the
 * block matches the transid in the parent node's pointer.  This is how we
 * detect blocks that either didn't get written at all or got written
 * in the wrong place.
 */
static int verify_parent_transid(struct extent_io_tree *io_tree,
				 struct extent_buffer *eb, u64 parent_transid,
				 int atomic)
{
	struct extent_state *cached_state = NULL;
	int ret;

	if (!parent_transid || btrfs_header_generation(eb) == parent_transid)
		return 0;

	if (atomic)
		return -EAGAIN;

	lock_extent_bits(io_tree, eb->start, eb->start + eb->len - 1,
			 0, &cached_state);
	if (extent_buffer_uptodate(eb) &&
	    btrfs_header_generation(eb) == parent_transid) {
		ret = 0;
		goto out;
	}
	printk_ratelimited("parent transid verify failed on %llu wanted %llu "
		       "found %llu\n",
		       (unsigned long long)eb->start,
		       (unsigned long long)parent_transid,
		       (unsigned long long)btrfs_header_generation(eb));
	ret = 1;
	clear_extent_buffer_uptodate(eb);
out:
	unlock_extent_cached(io_tree, eb->start, eb->start + eb->len - 1,
			     &cached_state, GFP_NOFS);
	return ret;
}

/*
 * helper to read a given tree block, doing retries as required when
 * the checksums don't match and we have alternate mirrors to try.
 */
static int btree_read_extent_buffer_pages(struct btrfs_root *root,
					  struct extent_buffer *eb,
					  u64 start, u64 parent_transid)
{
	struct extent_io_tree *io_tree;
	int failed = 0;
	int ret;
	int num_copies = 0;
	int mirror_num = 0;
	int failed_mirror = 0;

	clear_bit(EXTENT_BUFFER_CORRUPT, &eb->bflags);
	io_tree = &BTRFS_I(root->fs_info->btree_inode)->io_tree;
	while (1) {
		ret = read_extent_buffer_pages(io_tree, eb, start,
					       WAIT_COMPLETE,
					       btree_get_extent, mirror_num);
		if (!ret) {
			if (!verify_parent_transid(io_tree, eb,
						   parent_transid, 0))
				break;
			else
				ret = -EIO;
		}

		/*
		 * This buffer's crc is fine, but its contents are corrupted, so
		 * there is no reason to read the other copies, they won't be
		 * any less wrong.
		 */
		if (test_bit(EXTENT_BUFFER_CORRUPT, &eb->bflags))
			break;

		num_copies = btrfs_num_copies(root->fs_info,
					      eb->start, eb->len);
		if (num_copies == 1)
			break;

		if (!failed_mirror) {
			failed = 1;
			failed_mirror = eb->read_mirror;
		}

		mirror_num++;
		if (mirror_num == failed_mirror)
			mirror_num++;

		if (mirror_num > num_copies)
			break;
	}

	if (failed && !ret && failed_mirror)
		repair_eb_io_failure(root, eb, failed_mirror);

	return ret;
}

/*
 * checksum a dirty tree block before IO.  This has extra checks to make sure
 * we only fill in the checksum field in the first page of a multi-page block
 */

static int csum_dirty_buffer(struct btrfs_root *root, struct page *page)
{
	struct extent_io_tree *tree;
	u64 start = page_offset(page);
	u64 found_start;
	struct extent_buffer *eb;

	tree = &BTRFS_I(page->mapping->host)->io_tree;

	eb = (struct extent_buffer *)page->private;
	if (page != eb->pages[0])
		return 0;
	found_start = btrfs_header_bytenr(eb);
	if (found_start != start) {
		WARN_ON(1);
		return 0;
	}
	if (!PageUptodate(page)) {
		WARN_ON(1);
		return 0;
	}
	csum_tree_block(root, eb, 0);
	return 0;
}

static int check_tree_block_fsid(struct btrfs_root *root,
				 struct extent_buffer *eb)
{
	struct btrfs_fs_devices *fs_devices = root->fs_info->fs_devices;
	u8 fsid[BTRFS_UUID_SIZE];
	int ret = 1;

	read_extent_buffer(eb, fsid, (unsigned long)btrfs_header_fsid(eb),
			   BTRFS_FSID_SIZE);
	while (fs_devices) {
		if (!memcmp(fsid, fs_devices->fsid, BTRFS_FSID_SIZE)) {
			ret = 0;
			break;
		}
		fs_devices = fs_devices->seed;
	}
	return ret;
}

#define CORRUPT(reason, eb, root, slot)				\
	printk(KERN_CRIT "btrfs: corrupt leaf, %s: block=%llu,"	\
	       "root=%llu, slot=%d\n", reason,			\
	       (unsigned long long)btrfs_header_bytenr(eb),	\
	       (unsigned long long)root->objectid, slot)

static noinline int check_leaf(struct btrfs_root *root,
			       struct extent_buffer *leaf)
{
	struct btrfs_key key;
	struct btrfs_key leaf_key;
	u32 nritems = btrfs_header_nritems(leaf);
	int slot;

	if (nritems == 0)
		return 0;

	/* Check the 0 item */
	if (btrfs_item_offset_nr(leaf, 0) + btrfs_item_size_nr(leaf, 0) !=
	    BTRFS_LEAF_DATA_SIZE(root)) {
		CORRUPT("invalid item offset size pair", leaf, root, 0);
		return -EIO;
	}

	/*
	 * Check to make sure each items keys are in the correct order and their
	 * offsets make sense.  We only have to loop through nritems-1 because
	 * we check the current slot against the next slot, which verifies the
	 * next slot's offset+size makes sense and that the current's slot
	 * offset is correct.
	 */
	for (slot = 0; slot < nritems - 1; slot++) {
		btrfs_item_key_to_cpu(leaf, &leaf_key, slot);
		btrfs_item_key_to_cpu(leaf, &key, slot + 1);

		/* Make sure the keys are in the right order */
		if (btrfs_comp_cpu_keys(&leaf_key, &key) >= 0) {
			CORRUPT("bad key order", leaf, root, slot);
			return -EIO;
		}

		/*
		 * Make sure the offset and ends are right, remember that the
		 * item data starts at the end of the leaf and grows towards the
		 * front.
		 */
		if (btrfs_item_offset_nr(leaf, slot) !=
			btrfs_item_end_nr(leaf, slot + 1)) {
			CORRUPT("slot offset bad", leaf, root, slot);
			return -EIO;
		}

		/*
		 * Check to make sure that we don't point outside of the leaf,
		 * just incase all the items are consistent to eachother, but
		 * all point outside of the leaf.
		 */
		if (btrfs_item_end_nr(leaf, slot) >
		    BTRFS_LEAF_DATA_SIZE(root)) {
			CORRUPT("slot end outside of leaf", leaf, root, slot);
			return -EIO;
		}
	}

	return 0;
}

struct extent_buffer *find_eb_for_page(struct extent_io_tree *tree,
				       struct page *page, int max_walk)
{
	struct extent_buffer *eb;
	u64 start = page_offset(page);
	u64 target = start;
	u64 min_start;

	if (start < max_walk)
		min_start = 0;
	else
		min_start = start - max_walk;

	while (start >= min_start) {
		eb = find_extent_buffer(tree, start, 0);
		if (eb) {
			/*
			 * we found an extent buffer and it contains our page
			 * horray!
			 */
			if (eb->start <= target &&
			    eb->start + eb->len > target)
				return eb;

			/* we found an extent buffer that wasn't for us */
			free_extent_buffer(eb);
			return NULL;
		}
		if (start == 0)
			break;
		start -= PAGE_CACHE_SIZE;
	}
	return NULL;
}

static int btree_readpage_end_io_hook(struct page *page, u64 start, u64 end,
			       struct extent_state *state, int mirror)
{
	struct extent_io_tree *tree;
	u64 found_start;
	int found_level;
	struct extent_buffer *eb;
	struct btrfs_root *root = BTRFS_I(page->mapping->host)->root;
	int ret = 0;
	int reads_done;

	if (!page->private)
		goto out;

	tree = &BTRFS_I(page->mapping->host)->io_tree;
	eb = (struct extent_buffer *)page->private;

	/* the pending IO might have been the only thing that kept this buffer
	 * in memory.  Make sure we have a ref for all this other checks
	 */
	extent_buffer_get(eb);

	reads_done = atomic_dec_and_test(&eb->io_pages);
	if (!reads_done)
		goto err;

	eb->read_mirror = mirror;
	if (test_bit(EXTENT_BUFFER_IOERR, &eb->bflags)) {
		ret = -EIO;
		goto err;
	}

	found_start = btrfs_header_bytenr(eb);
	if (found_start != eb->start) {
		printk_ratelimited(KERN_INFO "btrfs bad tree block start "
			       "%llu %llu\n",
			       (unsigned long long)found_start,
			       (unsigned long long)eb->start);
		ret = -EIO;
		goto err;
	}
	if (check_tree_block_fsid(root, eb)) {
		printk_ratelimited(KERN_INFO "btrfs bad fsid on block %llu\n",
			       (unsigned long long)eb->start);
		ret = -EIO;
		goto err;
	}
	found_level = btrfs_header_level(eb);

	btrfs_set_buffer_lockdep_class(btrfs_header_owner(eb),
				       eb, found_level);

	ret = csum_tree_block(root, eb, 1);
	if (ret) {
		ret = -EIO;
		goto err;
	}

	/*
	 * If this is a leaf block and it is corrupt, set the corrupt bit so
	 * that we don't try and read the other copies of this block, just
	 * return -EIO.
	 */
	if (found_level == 0 && check_leaf(root, eb)) {
		set_bit(EXTENT_BUFFER_CORRUPT, &eb->bflags);
		ret = -EIO;
	}

	if (!ret)
		set_extent_buffer_uptodate(eb);
err:
	if (test_bit(EXTENT_BUFFER_READAHEAD, &eb->bflags)) {
		clear_bit(EXTENT_BUFFER_READAHEAD, &eb->bflags);
		btree_readahead_hook(root, eb, eb->start, ret);
	}

	if (ret) {
		/*
		 * our io error hook is going to dec the io pages
		 * again, we have to make sure it has something
		 * to decrement
		 */
		atomic_inc(&eb->io_pages);
		clear_extent_buffer_uptodate(eb);
	}
	free_extent_buffer(eb);
out:
	return ret;
}

static int btree_io_failed_hook(struct page *page, int failed_mirror)
{
	struct extent_buffer *eb;
	struct btrfs_root *root = BTRFS_I(page->mapping->host)->root;

	eb = (struct extent_buffer *)page->private;
	set_bit(EXTENT_BUFFER_IOERR, &eb->bflags);
	eb->read_mirror = failed_mirror;
	atomic_dec(&eb->io_pages);
	if (test_and_clear_bit(EXTENT_BUFFER_READAHEAD, &eb->bflags))
		btree_readahead_hook(root, eb, eb->start, -EIO);
	return -EIO;	/* we fixed nothing */
}

static void end_workqueue_bio(struct bio *bio, int err)
{
	struct end_io_wq *end_io_wq = bio->bi_private;
	struct btrfs_fs_info *fs_info;

	fs_info = end_io_wq->info;
	end_io_wq->error = err;
	end_io_wq->work.func = end_workqueue_fn;
	end_io_wq->work.flags = 0;

	if (bio->bi_rw & REQ_WRITE) {
		if (end_io_wq->metadata == BTRFS_WQ_ENDIO_METADATA)
			btrfs_queue_worker(&fs_info->endio_meta_write_workers,
					   &end_io_wq->work);
		else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_FREE_SPACE)
			btrfs_queue_worker(&fs_info->endio_freespace_worker,
					   &end_io_wq->work);
		else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
			btrfs_queue_worker(&fs_info->endio_raid56_workers,
					   &end_io_wq->work);
		else
			btrfs_queue_worker(&fs_info->endio_write_workers,
					   &end_io_wq->work);
	} else {
		if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
			btrfs_queue_worker(&fs_info->endio_raid56_workers,
					   &end_io_wq->work);
		else if (end_io_wq->metadata)
			btrfs_queue_worker(&fs_info->endio_meta_workers,
					   &end_io_wq->work);
		else
			btrfs_queue_worker(&fs_info->endio_workers,
					   &end_io_wq->work);
	}
}

/*
 * For the metadata arg you want
 *
 * 0 - if data
 * 1 - if normal metadta
 * 2 - if writing to the free space cache area
 * 3 - raid parity work
 */
int btrfs_bio_wq_end_io(struct btrfs_fs_info *info, struct bio *bio,
			int metadata)
{
	struct end_io_wq *end_io_wq;
	end_io_wq = kmalloc(sizeof(*end_io_wq), GFP_NOFS);
	if (!end_io_wq)
		return -ENOMEM;

	end_io_wq->private = bio->bi_private;
	end_io_wq->end_io = bio->bi_end_io;
	end_io_wq->info = info;
	end_io_wq->error = 0;
	end_io_wq->bio = bio;
	end_io_wq->metadata = metadata;

	bio->bi_private = end_io_wq;
	bio->bi_end_io = end_workqueue_bio;
	return 0;
}

unsigned long btrfs_async_submit_limit(struct btrfs_fs_info *info)
{
	unsigned long limit = min_t(unsigned long,
				    info->workers.max_workers,
				    info->fs_devices->open_devices);
	return 256 * limit;
}

static void run_one_async_start(struct btrfs_work *work)
{
	struct async_submit_bio *async;
	int ret;

	async = container_of(work, struct  async_submit_bio, work);
	ret = async->submit_bio_start(async->inode, async->rw, async->bio,
				      async->mirror_num, async->bio_flags,
				      async->bio_offset);
	if (ret)
		async->error = ret;
}

static void run_one_async_done(struct btrfs_work *work)
{
	struct btrfs_fs_info *fs_info;
	struct async_submit_bio *async;
	int limit;

	async = container_of(work, struct  async_submit_bio, work);
	fs_info = BTRFS_I(async->inode)->root->fs_info;

	limit = btrfs_async_submit_limit(fs_info);
	limit = limit * 2 / 3;

	if (atomic_dec_return(&fs_info->nr_async_submits) < limit &&
	    waitqueue_active(&fs_info->async_submit_wait))
		wake_up(&fs_info->async_submit_wait);

	/* If an error occured we just want to clean up the bio and move on */
	if (async->error) {
		bio_endio(async->bio, async->error);
		return;
	}

	async->submit_bio_done(async->inode, async->rw, async->bio,
			       async->mirror_num, async->bio_flags,
			       async->bio_offset);
}

static void run_one_async_free(struct btrfs_work *work)
{
	struct async_submit_bio *async;

	async = container_of(work, struct  async_submit_bio, work);
	kfree(async);
}

int btrfs_wq_submit_bio(struct btrfs_fs_info *fs_info, struct inode *inode,
			int rw, struct bio *bio, int mirror_num,
			unsigned long bio_flags,
			u64 bio_offset,
			extent_submit_bio_hook_t *submit_bio_start,
			extent_submit_bio_hook_t *submit_bio_done)
{
	struct async_submit_bio *async;

	async = kmalloc(sizeof(*async), GFP_NOFS);
	if (!async)
		return -ENOMEM;

	async->inode = inode;
	async->rw = rw;
	async->bio = bio;
	async->mirror_num = mirror_num;
	async->submit_bio_start = submit_bio_start;
	async->submit_bio_done = submit_bio_done;

	async->work.func = run_one_async_start;
	async->work.ordered_func = run_one_async_done;
	async->work.ordered_free = run_one_async_free;

	async->work.flags = 0;
	async->bio_flags = bio_flags;
	async->bio_offset = bio_offset;

	async->error = 0;

	atomic_inc(&fs_info->nr_async_submits);

	if (rw & REQ_SYNC)
		btrfs_set_work_high_prio(&async->work);

	btrfs_queue_worker(&fs_info->workers, &async->work);

	while (atomic_read(&fs_info->async_submit_draining) &&
	      atomic_read(&fs_info->nr_async_submits)) {
		wait_event(fs_info->async_submit_wait,
			   (atomic_read(&fs_info->nr_async_submits) == 0));
	}

	return 0;
}

static int btree_csum_one_bio(struct bio *bio)
{
	struct bio_vec *bvec = bio->bi_io_vec;
	int bio_index = 0;
	struct btrfs_root *root;
	int ret = 0;

	WARN_ON(bio->bi_vcnt <= 0);
	while (bio_index < bio->bi_vcnt) {
		root = BTRFS_I(bvec->bv_page->mapping->host)->root;
		ret = csum_dirty_buffer(root, bvec->bv_page);
		if (ret)
			break;
		bio_index++;
		bvec++;
	}
	return ret;
}

static int __btree_submit_bio_start(struct inode *inode, int rw,
				    struct bio *bio, int mirror_num,
				    unsigned long bio_flags,
				    u64 bio_offset)
{
	/*
	 * when we're called for a write, we're already in the async
	 * submission context.  Just jump into btrfs_map_bio
	 */
	return btree_csum_one_bio(bio);
}

static int __btree_submit_bio_done(struct inode *inode, int rw, struct bio *bio,
				 int mirror_num, unsigned long bio_flags,
				 u64 bio_offset)
{
	int ret;

	/*
	 * when we're called for a write, we're already in the async
	 * submission context.  Just jump into btrfs_map_bio
	 */
	ret = btrfs_map_bio(BTRFS_I(inode)->root, rw, bio, mirror_num, 1);
	if (ret)
		bio_endio(bio, ret);
	return ret;
}

static int check_async_write(struct inode *inode, unsigned long bio_flags)
{
	if (bio_flags & EXTENT_BIO_TREE_LOG)
		return 0;
#ifdef CONFIG_X86
	if (cpu_has_xmm4_2)
		return 0;
#endif
	return 1;
}

static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio,
				 int mirror_num, unsigned long bio_flags,
				 u64 bio_offset)
{
	int async = check_async_write(inode, bio_flags);
	int ret;

	if (!(rw & REQ_WRITE)) {
		/*
		 * called for a read, do the setup so that checksum validation
		 * can happen in the async kernel threads
		 */
		ret = btrfs_bio_wq_end_io(BTRFS_I(inode)->root->fs_info,
					  bio, 1);
		if (ret)
			goto out_w_error;
		ret = btrfs_map_bio(BTRFS_I(inode)->root, rw, bio,
				    mirror_num, 0);
	} else if (!async) {
		ret = btree_csum_one_bio(bio);
		if (ret)
			goto out_w_error;
		ret = btrfs_map_bio(BTRFS_I(inode)->root, rw, bio,
				    mirror_num, 0);
	} else {
		/*
		 * kthread helpers are used to submit writes so that
		 * checksumming can happen in parallel across all CPUs
		 */
		ret = btrfs_wq_submit_bio(BTRFS_I(inode)->root->fs_info,
					  inode, rw, bio, mirror_num, 0,
					  bio_offset,
					  __btree_submit_bio_start,
					  __btree_submit_bio_done);
	}

	if (ret) {
out_w_error:
		bio_endio(bio, ret);
	}
	return ret;
}

#ifdef CONFIG_MIGRATION
static int btree_migratepage(struct address_space *mapping,
			struct page *newpage, struct page *page,
			enum migrate_mode mode)
{
	/*
	 * we can't safely write a btree page from here,
	 * we haven't done the locking hook
	 */
	if (PageDirty(page))
		return -EAGAIN;
	/*
	 * Buffers may be managed in a filesystem specific way.
	 * We must have no buffers or drop them.
	 */
	if (page_has_private(page) &&
	    !try_to_release_page(page, GFP_KERNEL))
		return -EAGAIN;
	return migrate_page(mapping, newpage, page, mode);
}
#endif


static int btree_writepages(struct address_space *mapping,
			    struct writeback_control *wbc)
{
	struct extent_io_tree *tree;
	struct btrfs_fs_info *fs_info;
	int ret;

	tree = &BTRFS_I(mapping->host)->io_tree;
	if (wbc->sync_mode == WB_SYNC_NONE) {

		if (wbc->for_kupdate)
			return 0;

		fs_info = BTRFS_I(mapping->host)->root->fs_info;
		/* this is a bit racy, but that's ok */
		ret = percpu_counter_compare(&fs_info->dirty_metadata_bytes,
					     BTRFS_DIRTY_METADATA_THRESH);
		if (ret < 0)
			return 0;
	}
	return btree_write_cache_pages(mapping, wbc);
}

static int btree_readpage(struct file *file, struct page *page)
{
	struct extent_io_tree *tree;
	tree = &BTRFS_I(page->mapping->host)->io_tree;
	return extent_read_full_page(tree, page, btree_get_extent, 0);
}

static int btree_releasepage(struct page *page, gfp_t gfp_flags)
{
	if (PageWriteback(page) || PageDirty(page))
		return 0;
	/*
	 * We need to mask out eg. __GFP_HIGHMEM and __GFP_DMA32 as we're doing
	 * slab allocation from alloc_extent_state down the callchain where
	 * it'd hit a BUG_ON as those flags are not allowed.
	 */