linux-gtp/fs/btrfs/extent_io.c

#include <linux/bitops.h>
#include <linux/slab.h>
#include <linux/bio.h>
#include <linux/mm.h>
#include <linux/pagemap.h>
#include <linux/page-flags.h>
#include <linux/module.h>
#include <linux/spinlock.h>
#include <linux/blkdev.h>
#include <linux/swap.h>
#include <linux/writeback.h>
#include <linux/pagevec.h>
#include <linux/prefetch.h>
#include <linux/cleancache.h>
#include "extent_io.h"
#include "extent_map.h"
#include "compat.h"
#include "ctree.h"
#include "btrfs_inode.h"
#include "volumes.h"
#include "check-integrity.h"
#include "locking.h"
#include "rcu-string.h"

static struct kmem_cache *extent_state_cache;
static struct kmem_cache *extent_buffer_cache;

static LIST_HEAD(buffers);
static LIST_HEAD(states);

#define LEAK_DEBUG 0
#if LEAK_DEBUG
static DEFINE_SPINLOCK(leak_lock);
#endif

#define BUFFER_LRU_MAX 64

struct tree_entry {
	u64 start;
	u64 end;
	struct rb_node rb_node;
};

struct extent_page_data {
	struct bio *bio;
	struct extent_io_tree *tree;
	get_extent_t *get_extent;
	unsigned long bio_flags;

	/* tells writepage not to lock the state bits for this range
	 * it still does the unlocking
	 */
	unsigned int extent_locked:1;

	/* tells the submit_bio code to use a WRITE_SYNC */
	unsigned int sync_io:1;
};

static noinline void flush_write_bio(void *data);
static inline struct btrfs_fs_info *
tree_fs_info(struct extent_io_tree *tree)
{
	return btrfs_sb(tree->mapping->host->i_sb);
}

int __init extent_io_init(void)
{
	extent_state_cache = kmem_cache_create("btrfs_extent_state",
			sizeof(struct extent_state), 0,
			SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
	if (!extent_state_cache)
		return -ENOMEM;

	extent_buffer_cache = kmem_cache_create("btrfs_extent_buffer",
			sizeof(struct extent_buffer), 0,
			SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
	if (!extent_buffer_cache)
		goto free_state_cache;
	return 0;

free_state_cache:
	kmem_cache_destroy(extent_state_cache);
	return -ENOMEM;
}

void extent_io_exit(void)
{
	struct extent_state *state;
	struct extent_buffer *eb;

	while (!list_empty(&states)) {
		state = list_entry(states.next, struct extent_state, leak_list);
		printk(KERN_ERR "btrfs state leak: start %llu end %llu "
		       "state %lu in tree %p refs %d\n",
		       (unsigned long long)state->start,
		       (unsigned long long)state->end,
		       state->state, state->tree, atomic_read(&state->refs));
		list_del(&state->leak_list);
		kmem_cache_free(extent_state_cache, state);

	}

	while (!list_empty(&buffers)) {
		eb = list_entry(buffers.next, struct extent_buffer, leak_list);
		printk(KERN_ERR "btrfs buffer leak start %llu len %lu "
		       "refs %d\n", (unsigned long long)eb->start,
		       eb->len, atomic_read(&eb->refs));
		list_del(&eb->leak_list);
		kmem_cache_free(extent_buffer_cache, eb);
	}

	/*
	 * Make sure all delayed rcu free are flushed before we
	 * destroy caches.
	 */
	rcu_barrier();
	if (extent_state_cache)
		kmem_cache_destroy(extent_state_cache);
	if (extent_buffer_cache)
		kmem_cache_destroy(extent_buffer_cache);
}

void extent_io_tree_init(struct extent_io_tree *tree,
			 struct address_space *mapping)
{
	tree->state = RB_ROOT;
	INIT_RADIX_TREE(&tree->buffer, GFP_ATOMIC);
	tree->ops = NULL;
	tree->dirty_bytes = 0;
	spin_lock_init(&tree->lock);
	spin_lock_init(&tree->buffer_lock);
	tree->mapping = mapping;
}

static struct extent_state *alloc_extent_state(gfp_t mask)
{
	struct extent_state *state;
#if LEAK_DEBUG
	unsigned long flags;
#endif

	state = kmem_cache_alloc(extent_state_cache, mask);
	if (!state)
		return state;
	state->state = 0;
	state->private = 0;
	state->tree = NULL;
#if LEAK_DEBUG
	spin_lock_irqsave(&leak_lock, flags);
	list_add(&state->leak_list, &states);
	spin_unlock_irqrestore(&leak_lock, flags);
#endif
	atomic_set(&state->refs, 1);
	init_waitqueue_head(&state->wq);
	trace_alloc_extent_state(state, mask, _RET_IP_);
	return state;
}

void free_extent_state(struct extent_state *state)
{
	if (!state)
		return;
	if (atomic_dec_and_test(&state->refs)) {
#if LEAK_DEBUG
		unsigned long flags;
#endif
		WARN_ON(state->tree);
#if LEAK_DEBUG
		spin_lock_irqsave(&leak_lock, flags);
		list_del(&state->leak_list);
		spin_unlock_irqrestore(&leak_lock, flags);
#endif
		trace_free_extent_state(state, _RET_IP_);
		kmem_cache_free(extent_state_cache, state);
	}
}

static struct rb_node *tree_insert(struct rb_root *root, u64 offset,
				   struct rb_node *node)
{
	struct rb_node **p = &root->rb_node;
	struct rb_node *parent = NULL;
	struct tree_entry *entry;

	while (*p) {
		parent = *p;
		entry = rb_entry(parent, struct tree_entry, rb_node);

		if (offset < entry->start)
			p = &(*p)->rb_left;
		else if (offset > entry->end)
			p = &(*p)->rb_right;
		else
			return parent;
	}

	rb_link_node(node, parent, p);
	rb_insert_color(node, root);
	return NULL;
}

static struct rb_node *__etree_search(struct extent_io_tree *tree, u64 offset,
				     struct rb_node **prev_ret,
				     struct rb_node **next_ret)
{
	struct rb_root *root = &tree->state;
	struct rb_node *n = root->rb_node;
	struct rb_node *prev = NULL;
	struct rb_node *orig_prev = NULL;
	struct tree_entry *entry;
	struct tree_entry *prev_entry = NULL;

	while (n) {
		entry = rb_entry(n, struct tree_entry, rb_node);
		prev = n;
		prev_entry = entry;

		if (offset < entry->start)
			n = n->rb_left;
		else if (offset > entry->end)
			n = n->rb_right;
		else
			return n;
	}

	if (prev_ret) {
		orig_prev = prev;
		while (prev && offset > prev_entry->end) {
			prev = rb_next(prev);
			prev_entry = rb_entry(prev, struct tree_entry, rb_node);
		}
		*prev_ret = prev;
		prev = orig_prev;
	}

	if (next_ret) {
		prev_entry = rb_entry(prev, struct tree_entry, rb_node);
		while (prev && offset < prev_entry->start) {
			prev = rb_prev(prev);
			prev_entry = rb_entry(prev, struct tree_entry, rb_node);
		}
		*next_ret = prev;
	}
	return NULL;
}

static inline struct rb_node *tree_search(struct extent_io_tree *tree,
					  u64 offset)
{
	struct rb_node *prev = NULL;
	struct rb_node *ret;

	ret = __etree_search(tree, offset, &prev, NULL);
	if (!ret)
		return prev;
	return ret;
}

static void merge_cb(struct extent_io_tree *tree, struct extent_state *new,
		     struct extent_state *other)
{
	if (tree->ops && tree->ops->merge_extent_hook)
		tree->ops->merge_extent_hook(tree->mapping->host, new,
					     other);
}

/*
 * utility function to look for merge candidates inside a given range.
 * Any extents with matching state are merged together into a single
 * extent in the tree.  Extents with EXTENT_IO in their state field
 * are not merged because the end_io handlers need to be able to do
 * operations on them without sleeping (or doing allocations/splits).
 *
 * This should be called with the tree lock held.
 */
static void merge_state(struct extent_io_tree *tree,
		        struct extent_state *state)
{
	struct extent_state *other;
	struct rb_node *other_node;

	if (state->state & (EXTENT_IOBITS | EXTENT_BOUNDARY))
		return;

	other_node = rb_prev(&state->rb_node);
	if (other_node) {
		other = rb_entry(other_node, struct extent_state, rb_node);
		if (other->end == state->start - 1 &&
		    other->state == state->state) {
			merge_cb(tree, state, other);
			state->start = other->start;
			other->tree = NULL;
			rb_erase(&other->rb_node, &tree->state);
			free_extent_state(other);
		}
	}
	other_node = rb_next(&state->rb_node);
	if (other_node) {
		other = rb_entry(other_node, struct extent_state, rb_node);
		if (other->start == state->end + 1 &&
		    other->state == state->state) {
			merge_cb(tree, state, other);
			state->end = other->end;
			other->tree = NULL;
			rb_erase(&other->rb_node, &tree->state);
			free_extent_state(other);
		}
	}
}

static void set_state_cb(struct extent_io_tree *tree,
			 struct extent_state *state, int *bits)
{
	if (tree->ops && tree->ops->set_bit_hook)
		tree->ops->set_bit_hook(tree->mapping->host, state, bits);
}

static void clear_state_cb(struct extent_io_tree *tree,
			   struct extent_state *state, int *bits)
{
	if (tree->ops && tree->ops->clear_bit_hook)
		tree->ops->clear_bit_hook(tree->mapping->host, state, bits);
}

static void set_state_bits(struct extent_io_tree *tree,
			   struct extent_state *state, int *bits);

/*
 * insert an extent_state struct into the tree.  'bits' are set on the
 * struct before it is inserted.
 *
 * This may return -EEXIST if the extent is already there, in which case the
 * state struct is freed.
 *
 * The tree lock is not taken internally.  This is a utility function and
 * probably isn't what you want to call (see set/clear_extent_bit).
 */
static int insert_state(struct extent_io_tree *tree,
			struct extent_state *state, u64 start, u64 end,
			int *bits)
{
	struct rb_node *node;

	if (end < start)
		WARN(1, KERN_ERR "btrfs end < start %llu %llu\n",
		       (unsigned long long)end,
		       (unsigned long long)start);
	state->start = start;
	state->end = end;

	set_state_bits(tree, state, bits);

	node = tree_insert(&tree->state, end, &state->rb_node);
	if (node) {
		struct extent_state *found;
		found = rb_entry(node, struct extent_state, rb_node);
		printk(KERN_ERR "btrfs found node %llu %llu on insert of "
		       "%llu %llu\n", (unsigned long long)found->start,
		       (unsigned long long)found->end,
		       (unsigned long long)start, (unsigned long long)end);
		return -EEXIST;
	}
	state->tree = tree;
	merge_state(tree, state);
	return 0;
}

static void split_cb(struct extent_io_tree *tree, struct extent_state *orig,
		     u64 split)
{
	if (tree->ops && tree->ops->split_extent_hook)
		tree->ops->split_extent_hook(tree->mapping->host, orig, split);
}

/*
 * split a given extent state struct in two, inserting the preallocated
 * struct 'prealloc' as the newly created second half.  'split' indicates an
 * offset inside 'orig' where it should be split.
 *
 * Before calling,
 * the tree has 'orig' at [orig->start, orig->end].  After calling, there
 * are two extent state structs in the tree:
 * prealloc: [orig->start, split - 1]
 * orig: [ split, orig->end ]
 *
 * The tree locks are not taken by this function. They need to be held
 * by the caller.
 */
static int split_state(struct extent_io_tree *tree, struct extent_state *orig,
		       struct extent_state *prealloc, u64 split)
{
	struct rb_node *node;

	split_cb(tree, orig, split);

	prealloc->start = orig->start;
	prealloc->end = split - 1;
	prealloc->state = orig->state;
	orig->start = split;

	node = tree_insert(&tree->state, prealloc->end, &prealloc->rb_node);
	if (node) {
		free_extent_state(prealloc);
		return -EEXIST;
	}
	prealloc->tree = tree;
	return 0;
}

static struct extent_state *next_state(struct extent_state *state)
{
	struct rb_node *next = rb_next(&state->rb_node);
	if (next)
		return rb_entry(next, struct extent_state, rb_node);
	else
		return NULL;
}

/*
 * utility function to clear some bits in an extent state struct.
 * it will optionally wake up any one waiting on this state (wake == 1).
 *
 * If no bits are set on the state struct after clearing things, the
 * struct is freed and removed from the tree
 */
static struct extent_state *clear_state_bit(struct extent_io_tree *tree,
					    struct extent_state *state,
					    int *bits, int wake)
{
	struct extent_state *next;
	int bits_to_clear = *bits & ~EXTENT_CTLBITS;

	if ((bits_to_clear & EXTENT_DIRTY) && (state->state & EXTENT_DIRTY)) {
		u64 range = state->end - state->start + 1;
		WARN_ON(range > tree->dirty_bytes);
		tree->dirty_bytes -= range;
	}
	clear_state_cb(tree, state, bits);
	state->state &= ~bits_to_clear;
	if (wake)
		wake_up(&state->wq);
	if (state->state == 0) {
		next = next_state(state);
		if (state->tree) {
			rb_erase(&state->rb_node, &tree->state);
			state->tree = NULL;
			free_extent_state(state);
		} else {
			WARN_ON(1);
		}
	} else {
		merge_state(tree, state);
		next = next_state(state);
	}
	return next;
}

static struct extent_state *
alloc_extent_state_atomic(struct extent_state *prealloc)
{
	if (!prealloc)
		prealloc = alloc_extent_state(GFP_ATOMIC);

	return prealloc;
}

void extent_io_tree_panic(struct extent_io_tree *tree, int err)
{
	btrfs_panic(tree_fs_info(tree), err, "Locking error: "
		    "Extent tree was modified by another "
		    "thread while locked.");
}

/*
 * clear some bits on a range in the tree.  This may require splitting
 * or inserting elements in the tree, so the gfp mask is used to
 * indicate which allocations or sleeping are allowed.
 *
 * pass 'wake' == 1 to kick any sleepers, and 'delete' == 1 to remove
 * the given range from the tree regardless of state (ie for truncate).
 *
 * the range [start, end] is inclusive.
 *
 * This takes the tree lock, and returns 0 on success and < 0 on error.
 */
int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
		     int bits, int wake, int delete,
		     struct extent_state **cached_state,
		     gfp_t mask)
{
	struct extent_state *state;
	struct extent_state *cached;
	struct extent_state *prealloc = NULL;
	struct rb_node *node;
	u64 last_end;
	int err;
	int clear = 0;

	if (delete)
		bits |= ~EXTENT_CTLBITS;
	bits |= EXTENT_FIRST_DELALLOC;

	if (bits & (EXTENT_IOBITS | EXTENT_BOUNDARY))
		clear = 1;
again:
	if (!prealloc && (mask & __GFP_WAIT)) {
		prealloc = alloc_extent_state(mask);
		if (!prealloc)
			return -ENOMEM;
	}

	spin_lock(&tree->lock);
	if (cached_state) {
		cached = *cached_state;

		if (clear) {
			*cached_state = NULL;
			cached_state = NULL;
		}

		if (cached && cached->tree && cached->start <= start &&
		    cached->end > start) {
			if (clear)
				atomic_dec(&cached->refs);
			state = cached;
			goto hit_next;
		}
		if (clear)
			free_extent_state(cached);
	}
	/*
	 * this search will find the extents that end after
	 * our range starts
	 */
	node = tree_search(tree, start);
	if (!node)
		goto out;
	state = rb_entry(node, struct extent_state, rb_node);
hit_next:
	if (state->start > end)
		goto out;
	WARN_ON(state->end < start);
	last_end = state->end;

	/* the state doesn't have the wanted bits, go ahead */
	if (!(state->state & bits)) {
		state = next_state(state);
		goto next;
	}

	/*
	 *     | ---- desired range ---- |
	 *  | state | or
	 *  | ------------- state -------------- |
	 *
	 * We need to split the extent we found, and may flip
	 * bits on second half.
	 *
	 * If the extent we found extends past our range, we
	 * just split and search again.  It'll get split again
	 * the next time though.
	 *
	 * If the extent we found is inside our range, we clear
	 * the desired bit on it.
	 */

	if (state->start < start) {
		prealloc = alloc_extent_state_atomic(prealloc);
		BUG_ON(!prealloc);
		err = split_state(tree, state, prealloc, start);
		if (err)
			extent_io_tree_panic(tree, err);

		prealloc = NULL;
		if (err)
			goto out;
		if (state->end <= end) {
			state = clear_state_bit(tree, state, &bits, wake);
			goto next;
		}
		goto search_again;
	}
	/*
	 * | ---- desired range ---- |
	 *                        | state |
	 * We need to split the extent, and clear the bit
	 * on the first half
	 */
	if (state->start <= end && state->end > end) {
		prealloc = alloc_extent_state_atomic(prealloc);
		BUG_ON(!prealloc);
		err = split_state(tree, state, prealloc, end + 1);
		if (err)
			extent_io_tree_panic(tree, err);

		if (wake)
			wake_up(&state->wq);

		clear_state_bit(tree, prealloc, &bits, wake);

		prealloc = NULL;
		goto out;
	}

	state = clear_state_bit(tree, state, &bits, wake);
next:
	if (last_end == (u64)-1)
		goto out;
	start = last_end + 1;
	if (start <= end && state && !need_resched())
		goto hit_next;
	goto search_again;

out:
	spin_unlock(&tree->lock);
	if (prealloc)
		free_extent_state(prealloc);

	return 0;

search_again:
	if (start > end)
		goto out;
	spin_unlock(&tree->lock);
	if (mask & __GFP_WAIT)
		cond_resched();
	goto again;
}

static void wait_on_state(struct extent_io_tree *tree,
			  struct extent_state *state)
		__releases(tree->lock)
		__acquires(tree->lock)
{
	DEFINE_WAIT(wait);
	prepare_to_wait(&state->wq, &wait, TASK_UNINTERRUPTIBLE);
	spin_unlock(&tree->lock);
	schedule();
	spin_lock(&tree->lock);
	finish_wait(&state->wq, &wait);
}

/*
 * waits for one or more bits to clear on a range in the state tree.
 * The range [start, end] is inclusive.
 * The tree lock is taken by this function
 */
void wait_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, int bits)
{
	struct extent_state *state;
	struct rb_node *node;

	spin_lock(&tree->lock);
again:
	while (1) {
		/*
		 * this search will find all the extents that end after
		 * our range starts
		 */
		node = tree_search(tree, start);
		if (!node)
			break;

		state = rb_entry(node, struct extent_state, rb_node);

		if (state->start > end)
			goto out;

		if (state->state & bits) {
			start = state->start;
			atomic_inc(&state->refs);
			wait_on_state(tree, state);
			free_extent_state(state);
			goto again;
		}
		start = state->end + 1;

		if (start > end)
			break;

		cond_resched_lock(&tree->lock);
	}
out:
	spin_unlock(&tree->lock);
}

static void set_state_bits(struct extent_io_tree *tree,
			   struct extent_state *state,
			   int *bits)
{
	int bits_to_set = *bits & ~EXTENT_CTLBITS;

	set_state_cb(tree, state, bits);
	if ((bits_to_set & EXTENT_DIRTY) && !(state->state & EXTENT_DIRTY)) {
		u64 range = state->end - state->start + 1;
		tree->dirty_bytes += range;
	}
	state->state |= bits_to_set;
}

static void cache_state(struct extent_state *state,
			struct extent_state **cached_ptr)
{
	if (cached_ptr && !(*cached_ptr)) {
		if (state->state & (EXTENT_IOBITS | EXTENT_BOUNDARY)) {
			*cached_ptr = state;
			atomic_inc(&state->refs);
		}
	}
}

static void uncache_state(struct extent_state **cached_ptr)
{
	if (cached_ptr && (*cached_ptr)) {
		struct extent_state *state = *cached_ptr;
		*cached_ptr = NULL;
		free_extent_state(state);
	}
}

/*
 * set some bits on a range in the tree.  This may require allocations or
 * sleeping, so the gfp mask is used to indicate what is allowed.
 *
 * If any of the exclusive bits are set, this will fail with -EEXIST if some
 * part of the range already has the desired bits set.  The start of the
 * existing range is returned in failed_start in this case.
 *
 * [start, end] is inclusive This takes the tree lock.
 */

static int __must_check
__set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
		 int bits, int exclusive_bits, u64 *failed_start,
		 struct extent_state **cached_state, gfp_t mask)
{
	struct extent_state *state;
	struct extent_state *prealloc = NULL;
	struct rb_node *node;
	int err = 0;
	u64 last_start;
	u64 last_end;

	bits |= EXTENT_FIRST_DELALLOC;
again:
	if (!prealloc && (mask & __GFP_WAIT)) {
		prealloc = alloc_extent_state(mask);
		BUG_ON(!prealloc);
	}

	spin_lock(&tree->lock);
	if (cached_state && *cached_state) {
		state = *cached_state;
		if (state->start <= start && state->end > start &&
		    state->tree) {
			node = &state->rb_node;
			goto hit_next;
		}
	}
	/*
	 * this search will find all the extents that end after
	 * our range starts.
	 */
	node = tree_search(tree, start);
	if (!node) {
		prealloc = alloc_extent_state_atomic(prealloc);
		BUG_ON(!prealloc);
		err = insert_state(tree, prealloc, start, end, &bits);
		if (err)
			extent_io_tree_panic(tree, err);

		prealloc = NULL;
		goto out;
	}
	state = rb_entry(node, struct extent_state, rb_node);
hit_next:
	last_start = state->start;
	last_end = state->end;

	/*
	 * | ---- desired range ---- |
	 * | state |
	 *
	 * Just lock what we found and keep going
	 */
	if (state->start == start && state->end <= end) {
		if (state->state & exclusive_bits) {
			*failed_start = state->start;
			err = -EEXIST;
			goto out;
		}

		set_state_bits(tree, state, &bits);
		cache_state(state, cached_state);
		merge_state(tree, state);
		if (last_end == (u64)-1)
			goto out;
		start = last_end + 1;
		state = next_state(state);
		if (start < end && state && state->start == start &&
		    !need_resched())
			goto hit_next;
		goto search_again;
	}

	/*
	 *     | ---- desired range ---- |
	 * | state |
	 *   or
	 * | ------------- state -------------- |
	 *
	 * We need to split the extent we found, and may flip bits on
	 * second half.
	 *
	 * If the extent we found extends past our
	 * range, we just split and search again.  It'll get split
	 * again the next time though.
	 *
	 * If the extent we found is inside our range, we set the
	 * desired bit on it.
	 */
	if (state->start < start) {
		if (state->state & exclusive_bits) {
			*failed_start = start;
			err = -EEXIST;
			goto out;
		}

		prealloc = alloc_extent_state_atomic(prealloc);
		BUG_ON(!prealloc);
		err = split_state(tree, state, prealloc, start);
		if (err)
			extent_io_tree_panic(tree, err);

		prealloc = NULL;
		if (err)
			goto out;
		if (state->end <= end) {
			set_state_bits(tree, state, &bits);
			cache_state(state, cached_state);
			merge_state(tree, state);
			if (last_end == (u64)-1)
				goto out;
			start = last_end + 1;
			state = next_state(state);
			if (start < end && state && state->start == start &&
			    !need_resched())
				goto hit_next;
		}
		goto search_again;
	}
	/*
	 * | ---- desired range ---- |
	 *     | state | or               | state |
	 *
	 * There's a hole, we need to insert something in it and
	 * ignore the extent we found.
	 */
	if (state->start > start) {
		u64 this_end;
		if (end < last_start)
			this_end = end;
		else
			this_end = last_start - 1;

		prealloc = alloc_extent_state_atomic(prealloc);
		BUG_ON(!prealloc);

		/*
		 * Avoid to free 'prealloc' if it can be merged with
		 * the later extent.
		 */
		err = insert_state(tree, prealloc, start, this_end,
				   &bits);
		if (err)
			extent_io_tree_panic(tree, err);

		cache_state(prealloc, cached_state);
		prealloc = NULL;
		start = this_end + 1;
		goto search_again;
	}
	/*
	 * | ---- desired range ---- |
	 *                        | state |
	 * We need to split the extent, and set the bit
	 * on the first half
	 */
	if (state->start <= end && state->end > end) {
		if (state->state & exclusive_bits) {
			*failed_start = start;
			err = -EEXIST;
			goto out;
		}

		prealloc = alloc_extent_state_atomic(prealloc);
		BUG_ON(!prealloc);
		err = split_state(tree, state, prealloc, end + 1);
		if (err)
			extent_io_tree_panic(tree, err);

		set_state_bits(tree, prealloc, &bits);
		cache_state(prealloc, cached_state);
		merge_state(tree, prealloc);
		prealloc = NULL;
		goto out;
	}

	goto search_again;

out:
	spin_unlock(&tree->lock);
	if (prealloc)
		free_extent_state(prealloc);

	return err;

search_again:
	if (start > end)
		goto out;
	spin_unlock(&tree->lock);
	if (mask & __GFP_WAIT)
		cond_resched();
	goto again;
}

int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, int bits,
		   u64 *failed_start, struct extent_state **cached_state,
		   gfp_t mask)
{
	return __set_extent_bit(tree, start, end, bits, 0, failed_start,
				cached_state, mask);
}


/**
 * convert_extent_bit - convert all bits in a given range from one bit to
 * 			another
 * @tree:	the io tree to search
 * @start:	the start offset in bytes
 * @end:	the end offset in bytes (inclusive)
 * @bits:	the bits to set in this range
 * @clear_bits:	the bits to clear in this range
 * @cached_state:	state that we're going to cache
 * @mask:	the allocation mask
 *
 * This will go through and set bits for the given range.  If any states exist
 * already in this range they are set with the given bit and cleared of the
 * clear_bits.  This is only meant to be used by things that are mergeable, ie
 * converting from say DELALLOC to DIRTY.  This is not meant to be used with
 * boundary bits like LOCK.
 */
int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
		       int bits, int clear_bits,
		       struct extent_state **cached_state, gfp_t mask)
{
	struct extent_state *state;
	struct extent_state *prealloc = NULL;
	struct rb_node *node;
	int err = 0;
	u64 last_start;
	u64 last_end;

again:
	if (!prealloc && (mask & __GFP_WAIT)) {
		prealloc = alloc_extent_state(mask);
		if (!prealloc)
			return -ENOMEM;
	}

	spin_lock(&tree->lock);
	if (cached_state && *cached_state) {
		state = *cached_state;
		if (state->start <= start && state->end > start &&
		    state->tree) {
			node = &state->rb_node;
			goto hit_next;
		}
	}

	/*
	 * this search will find all the extents that end after
	 * our range starts.
	 */
	node = tree_search(tree, start);
	if (!node) {
		prealloc = alloc_extent_state_atomic(prealloc);
		if (!prealloc) {
			err = -ENOMEM;
			goto out;
		}
		err = insert_state(tree, prealloc, start, end, &bits);
		prealloc = NULL;
		if (err)
			extent_io_tree_panic(tree, err);
		goto out;
	}
	state = rb_entry(node, struct extent_state, rb_node);
hit_next:
	last_start = state->start;
	last_end = state->end;

	/*
	 * | ---- desired range ---- |
	 * | state |
	 *
	 * Just lock what we found and keep going
	 */
	if (state->start == start && state->end <= end) {
		set_state_bits(tree, state, &bits);
		cache_state(state, cached_state);
		state = clear_state_bit(tree, state, &clear_bits, 0);
		if (last_end == (u64)-1)
			goto out;
		start = last_end + 1;
		if (start < end && state && state->start == start &&
		    !need_resched())
			goto hit_next;
		goto search_again;
	}

	/*
	 *     | ---- desired range ---- |
	 * | state |
	 *   or
	 * | ------------- state -------------- |
	 *
	 * We need to split the extent we found, and may flip bits on
	 * second half.
	 *
	 * If the extent we found extends past our
	 * range, we just split and search again.  It'll get split
	 * again the next time though.
	 *
	 * If the extent we found is inside our range, we set the
	 * desired bit on it.
	 */
	if (state->start < start) {
		prealloc = alloc_extent_state_atomic(prealloc);
		if (!prealloc) {
			err = -ENOMEM;
			goto out;
		}
		err = split_state(tree, state, prealloc, start);
		if (err)
			extent_io_tree_panic(tree, err);
		prealloc = NULL;
		if (err)
			goto out;
		if (state->end <= end) {
			set_state_bits(tree, state, &bits);
			cache_state(state, cached_state);
			state = clear_state_bit(tree, state, &clear_bits, 0);
			if (last_end == (u64)-1)
				goto out;
			start = last_end + 1;
			if (start < end && state && state->start == start &&
			    !need_resched())
				goto hit_next;
		}
		goto search_again;
	}
	/*
	 * | ---- desired range ---- |
	 *     | state | or               | state |
	 *
	 * There's a hole, we need to insert something in it and
	 * ignore the extent we found.
	 */
	if (state->start > start) {
		u64 this_end;
		if (end < last_start)
			this_end = end;
		else
			this_end = last_start - 1;

		prealloc = alloc_extent_state_atomic(prealloc);
		if (!prealloc) {
			err = -ENOMEM;
			goto out;
		}

		/*
		 * Avoid to free 'prealloc' if it can be merged with
		 * the later extent.
		 */
		err = insert_state(tree, prealloc, start, this_end,
				   &bits);
		if (err)
			extent_io_tree_panic(tree, err);
		cache_state(prealloc, cached_state);
		prealloc = NULL;
		start = this_end + 1;
		goto search_again;
	}
	/*
	 * | ---- desired range ---- |
	 *                        | state |
	 * We need to split the extent, and set the bit
	 * on the first half
	 */
	if (state->start <= end && state->end > end) {
		prealloc = alloc_extent_state_atomic(prealloc);
		if (!prealloc) {
			err = -ENOMEM;
			goto out;
		}

		err = split_state(tree, state, prealloc, end + 1);
		if (err)
			extent_io_tree_panic(tree, err);

		set_state_bits(tree, prealloc, &bits);
		cache_state(prealloc, cached_state);
		clear_state_bit(tree, prealloc, &clear_bits, 0);
		prealloc = NULL;
		goto out;
	}

	goto search_again;

out:
	spin_unlock(&tree->lock);
	if (prealloc)
		free_extent_state(prealloc);

	return err;

search_again:
	if (start > end)
		goto out;
	spin_unlock(&tree->lock);
	if (mask & __GFP_WAIT)
		cond_resched();
	goto again;
}

/* wrappers around set/clear extent bit */
int set_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end,
		     gfp_t mask)
{
	return set_extent_bit(tree, start, end, EXTENT_DIRTY, NULL,
			      NULL, mask);
}

int set_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
		    int bits, gfp_t mask)
{
	return set_extent_bit(tree, start, end, bits, NULL,
			      NULL, mask);
}

int clear_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
		      int bits, gfp_t mask)
{
	return clear_extent_bit(tree, start, end, bits, 0, 0, NULL, mask);
}

int set_extent_delalloc(struct extent_io_tree *tree, u64 start, u64 end,
			struct extent_state **cached_state, gfp_t mask)
{
	return set_extent_bit(tree, start, end,
			      EXTENT_DELALLOC | EXTENT_UPTODATE,
			      NULL, cached_state, mask);
}

int set_extent_defrag(struct extent_io_tree *tree, u64 start, u64 end,
		      struct extent_state **cached_state, gfp_t mask)
{
	return set_extent_bit(tree, start, end,
			      EXTENT_DELALLOC | EXTENT_UPTODATE | EXTENT_DEFRAG,
			      NULL, cached_state, mask);
}

int clear_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end,
		       gfp_t mask)
{
	return clear_extent_bit(tree, start, end,
				EXTENT_DIRTY | EXTENT_DELALLOC |
				EXTENT_DO_ACCOUNTING, 0, 0, NULL, mask);
}

int set_extent_new(struct extent_io_tree *tree, u64 start, u64 end,
		     gfp_t mask)
{
	return set_extent_bit(tree, start, end, EXTENT_NEW, NULL,
			      NULL, mask);
}

int set_extent_uptodate(struct extent_io_tree *tree, u64 start, u64 end,
			struct extent_state **cached_state, gfp_t mask)
{
	return set_extent_bit(tree, start, end, EXTENT_UPTODATE, 0,
			      cached_state, mask);
}

int clear_extent_uptodate(struct extent_io_tree *tree, u64 start, u64 end,
			  struct extent_state **cached_state, gfp_t mask)
{
	return clear_extent_bit(tree, start, end, EXTENT_UPTODATE, 0, 0,
				cached_state, mask);
}

/*
 * either insert or lock state struct between start and end use mask to tell
 * us if waiting is desired.
 */
int lock_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
		     int bits, struct extent_state **cached_state)
{
	int err;
	u64 failed_start;
	while (1) {
		err = __set_extent_bit(tree, start, end, EXTENT_LOCKED | bits,
				       EXTENT_LOCKED, &failed_start,
				       cached_state, GFP_NOFS);
		if (err == -EEXIST) {
			wait_extent_bit(tree, failed_start, end, EXTENT_LOCKED);
			start = failed_start;
		} else
			break;
		WARN_ON(start > end);
	}
	return err;
}

int lock_extent(struct extent_io_tree *tree, u64 start, u64 end)
{
	return lock_extent_bits(tree, start, end, 0, NULL);
}

int try_lock_extent(struct extent_io_tree *tree, u64 start, u64 end)
{
	int err;
	u64 failed_start;

	err = __set_extent_bit(tree, start, end, EXTENT_LOCKED, EXTENT_LOCKED,
			       &failed_start, NULL, GFP_NOFS);
	if (err == -EEXIST) {
		if (failed_start > start)
			clear_extent_bit(tree, start, failed_start - 1,
					 EXTENT_LOCKED, 1, 0, NULL, GFP_NOFS);
		return 0;
	}
	return 1;
}

int unlock_extent_cached(struct extent_io_tree *tree, u64 start, u64 end,
			 struct extent_state **cached, gfp_t mask)
{
	return clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, cached,
				mask);
}

int unlock_extent(struct extent_io_tree *tree, u64 start, u64 end)
{
	return clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, NULL,
				GFP_NOFS);
}

/*
 * helper function to set both pages and extents in the tree writeback
 */
static int set_range_writeback(struct extent_io_tree *tree, u64 start, u64 end)
{
	unsigned long index = start >> PAGE_CACHE_SHIFT;
	unsigned long end_index = end >> PAGE_CACHE_SHIFT;
	struct page *page;

	while (index <= end_index) {
		page = find_get_page(tree->mapping, index);
		BUG_ON(!page); /* Pages should be in the extent_io_tree */
		set_page_writeback(page);
		page_cache_release(page);
		index++;
	}
	return 0;
}

/* find the first state struct with 'bits' set after 'start', and
 * return it.  tree->lock must be held.  NULL will returned if
 * nothing was found after 'start'
 */
struct extent_state *find_first_extent_bit_state(struct extent_io_tree *tree,
						 u64 start, int bits)
{
	struct rb_node *node;
	struct extent_state *state;

	/*
	 * this search will find all the extents that end after
	 * our range starts.
	 */
	node = tree_search(tree, start);
	if (!node)
		goto out;

	while (1) {
		state = rb_entry(node, struct extent_state, rb_node);
		if (state->end >= start && (state->state & bits))
			return state;

		node = rb_next(node);
		if (!node)
			break;
	}
out:
	return NULL;
}

/*
 * find the first offset in the io tree with 'bits' set. zero is
 * returned if we find something, and *start_ret and *end_ret are
 * set to reflect the state struct that was found.
 *
 * If nothing was found, 1 is returned. If found something, return 0.
 */
int find_first_extent_bit(struct extent_io_tree *tree, u64 start,
			  u64 *start_ret, u64 *end_ret, int bits,
			  struct extent_state **cached_state)
{
	struct extent_state *state;
	struct rb_node *n;
	int ret = 1;

	spin_lock(&tree->lock);
	if (cached_state && *cached_state) {
		state = *cached_state;
		if (state->end == start - 1 && state->tree) {
			n = rb_next(&state->rb_node);
			while (n) {
				state = rb_entry(n, struct extent_state,
						 rb_node);
				if (state->state & bits)
					goto got_it;
				n = rb_next(n);
			}
			free_extent_state(*cached_state);
			*cached_state = NULL;
			goto out;
		}
		free_extent_state(*cached_state);
		*cached_state = NULL;
	}

	state = find_first_extent_bit_state(tree, start, bits);
got_it:
	if (state) {
		cache_state(state, cached_state);
		*start_ret = state->start;
		*end_ret = state->end;
		ret = 0;
	}
out:
	spin_unlock(&tree->lock);
	return ret;
}

/*
 * find a contiguous range of bytes in the file marked as delalloc, not
 * more than 'max_bytes'.  start and end are used to return the range,
 *
 * 1 is returned if we find something, 0 if nothing was in the tree
 */
static noinline u64 find_delalloc_range(struct extent_io_tree *tree,
					u64 *start, u64 *end, u64 max_bytes,
					struct extent_state **cached_state)
{
	struct rb_node *node;
	struct extent_state *state;
	u64 cur_start = *start;
	u64 found = 0;
	u64 total_bytes = 0;

	spin_lock(&tree->lock);

	/*
	 * this search will find all the extents that end after
	 * our range starts.
	 */
	node = tree_search(tree, cur_start);
	if (!node) {
		if (!found)
			*end = (u64)-1;
		goto out;
	}

	while (1) {
		state = rb_entry(node, struct extent_state, rb_node);
		if (found && (state->start != cur_start ||
			      (state->state & EXTENT_BOUNDARY))) {
			goto out;
		}
		if (!(state->state & EXTENT_DELALLOC)) {
			if (!found)
				*end = state->end;
			goto out;
		}
		if (!found) {
			*start = state->start;
			*cached_state = state;
			atomic_inc(&state->refs);
		}
		found++;
		*end = state->end;
		cur_start = state->end + 1;
		node = rb_next(node);
		if (!node)
			break;
		total_bytes += state->end - state->start + 1;
		if (total_bytes >= max_bytes)
			break;
	}
out:
	spin_unlock(&tree->lock);
	return found;
}

static noinline void __unlock_for_delalloc(struct inode *inode,
					   struct page *locked_page,
					   u64 start, u64 end)
{
	int ret;
	struct page *pages[16];
	unsigned long index = start >> PAGE_CACHE_SHIFT;
	unsigned long end_index = end >> PAGE_CACHE_SHIFT;
	unsigned long nr_pages = end_index - index + 1;
	int i;

	if (index == locked_page->index && end_index == index)
		return;

	while (nr_pages > 0) {
		ret = find_get_pages_contig(inode->i_mapping, index,
				     min_t(unsigned long, nr_pages,
				     ARRAY_SIZE(pages)), pages);
		for (i = 0; i < ret; i++) {
			if (pages[i] != locked_page)
				unlock_page(pages[i]);
			page_cache_release(pages[i]);
		}
		nr_pages -= ret;
		index += ret;
		cond_resched();
	}
}

static noinline int lock_delalloc_pages(struct inode *inode,
					struct page *locked_page,
					u64 delalloc_start,
					u64 delalloc_end)
{
	unsigned long index = delalloc_start >> PAGE_CACHE_SHIFT;
	unsigned long start_index = index;
	unsigned long end_index = delalloc_end >> PAGE_CACHE_SHIFT;
	unsigned long pages_locked = 0;
	struct page *pages[16];
	unsigned long nrpages;
	int ret;
	int i;

	/* the caller is responsible for locking the start index */
	if (index == locked_page->index && index == end_index)
		return 0;

	/* skip the page at the start index */
	nrpages = end_index - index + 1;
	while (nrpages > 0) {
		ret = find_get_pages_contig(inode->i_mapping, index,
				     min_t(unsigned long,
				     nrpages, ARRAY_SIZE(pages)), pages);
		if (ret == 0) {
			ret = -EAGAIN;
			goto done;
		}
		/* now we have an array of pages, lock them all */
		for (i = 0; i < ret; i++) {
			/*
			 * the caller is taking responsibility for
			 * locked_page
			 */
			if (pages[i] != locked_page) {
				lock_page(pages[i]);
				if (!PageDirty(pages[i]) ||
				    pages[i]->mapping != inode->i_mapping) {
					ret = -EAGAIN;
					unlock_page(pages[i]);
					page_cache_release(pages[i]);
					goto done;
				}
			}
			page_cache_release(pages[i]);
			pages_locked++;
		}
		nrpages -= ret;
		index += ret;
		cond_resched();
	}
	ret = 0;
done:
	if (ret && pages_locked) {
		__unlock_for_delalloc(inode, locked_page,
			      delalloc_start,
			      ((u64)(start_index + pages_locked - 1)) <<
			      PAGE_CACHE_SHIFT);
	}
	return ret;
}

/*
 * find a contiguous range of bytes in the file marked as delalloc, not
 * more than 'max_bytes'.  start and end are used to return the range,
 *
 * 1 is returned if we find something, 0 if nothing was in the tree
 */
static noinline u64 find_lock_delalloc_range(struct inode *inode,
					     struct extent_io_tree *tree,
					     struct page *locked_page,
					     u64 *start, u64 *end,
					     u64 max_bytes)
{
	u64 delalloc_start;
	u64 delalloc_end;
	u64 found;
	struct extent_state *cached_state = NULL;
	int ret;
	int loops = 0;

again:
	/* step one, find a bunch of delalloc bytes starting at start */
	delalloc_start = *start;
	delalloc_end = 0;
	found = find_delalloc_range(tree, &delalloc_start, &delalloc_end,
				    max_bytes, &cached_state);
	if (!found || delalloc_end <= *start) {
		*start = delalloc_start;
		*end = delalloc_end;
		free_extent_state(cached_state);
		return found;
	}

	/*
	 * start comes from the offset of locked_page.  We have to lock
	 * pages in order, so we can't process delalloc bytes before
	 * locked_page
	 */
	if (delalloc_start < *start)
		delalloc_start = *start;

	/*
	 * make sure to limit the number of pages we try to lock down
	 * if we're looping.
	 */
	if (delalloc_end + 1 - delalloc_start > max_bytes && loops)
		delalloc_end = delalloc_start + PAGE_CACHE_SIZE - 1;

	/* step two, lock all the pages after the page that has start */
	ret = lock_delalloc_pages(inode, locked_page,
				  delalloc_start, delalloc_end);
	if (ret == -EAGAIN) {
		/* some of the pages are gone, lets avoid looping by
		 * shortening the size of the delalloc range we're searching
		 */
		free_extent_state(cached_state);
		if (!loops) {
			unsigned long offset = (*start) & (PAGE_CACHE_SIZE - 1);
			max_bytes = PAGE_CACHE_SIZE - offset;
			loops = 1;
			goto again;
		} else {
			found = 0;
			goto out_failed;
		}
	}
	BUG_ON(ret); /* Only valid values are 0 and -EAGAIN */

	/* step three, lock the state bits for the whole range */
	lock_extent_bits(tree, delalloc_start, delalloc_end, 0, &cached_state);

	/* then test to make sure it is all still delalloc */
	ret = test_range_bit(tree, delalloc_start, delalloc_end,
			     EXTENT_DELALLOC, 1, cached_state);
	if (!ret) {
		unlock_extent_cached(tree, delalloc_start, delalloc_end,
				     &cached_state, GFP_NOFS);
		__unlock_for_delalloc(inode, locked_page,
			      delalloc_start, delalloc_end);
		cond_resched();
		goto again;
	}
	free_extent_state(cached_state);
	*start = delalloc_start;
	*end = delalloc_end;
out_failed:
	return found;
}

int extent_clear_unlock_delalloc(struct inode *inode,
				struct extent_io_tree *tree,
				u64 start, u64 end, struct page *locked_page,
				unsigned long op)
{
	int ret;
	struct page *pages[16];
	unsigned long index = start >> PAGE_CACHE_SHIFT;
	unsigned long end_index = end >> PAGE_CACHE_SHIFT;
	unsigned long nr_pages = end_index - index + 1;
	int i;
	int clear_bits = 0;

	if (op & EXTENT_CLEAR_UNLOCK)
		clear_bits |= EXTENT_LOCKED;
	if (op & EXTENT_CLEAR_DIRTY)
		clear_bits |= EXTENT_DIRTY;

	if (op & EXTENT_CLEAR_DELALLOC)
		clear_bits |= EXTENT_DELALLOC;

	clear_extent_bit(tree, start, end, clear_bits, 1, 0, NULL, GFP_NOFS);
	if (!(op & (EXTENT_CLEAR_UNLOCK_PAGE | EXTENT_CLEAR_DIRTY |
		    EXTENT_SET_WRITEBACK | EXTENT_END_WRITEBACK |
		    EXTENT_SET_PRIVATE2)))
		return 0;

	while (nr_pages > 0) {
		ret = find_get_pages_contig(inode->i_mapping, index,
				     min_t(unsigned long,
				     nr_pages, ARRAY_SIZE(pages)), pages);
		for (i = 0; i < ret; i++) {

			if (op & EXTENT_SET_PRIVATE2)
				SetPagePrivate2(pages[i]);

			if (pages[i] == locked_page) {
				page_cache_release(pages[i]);
				continue;
			}
			if (op & EXTENT_CLEAR_DIRTY)
				clear_page_dirty_for_io(pages[i]);
			if (op & EXTENT_SET_WRITEBACK)
				set_page_writeback(pages[i]);
			if (op & EXTENT_END_WRITEBACK)
				end_page_writeback(pages[i]);
			if (op & EXTENT_CLEAR_UNLOCK_PAGE)
				unlock_page(pages[i]);
			page_cache_release(pages[i]);
		}
		nr_pages -= ret;
		index += ret;
		cond_resched();
	}
	return 0;
}

/*
 * count the number of bytes in the tree that have a given bit(s)
 * set.  This can be fairly slow, except for EXTENT_DIRTY which is
 * cached.  The total number found is returned.
 */
u64 count_range_bits(struct extent_io_tree *tree,
		     u64 *start, u64 search_end, u64 max_bytes,
		     unsigned long bits, int contig)
{
	struct rb_node *node;
	struct extent_state *state;
	u64 cur_start = *start;
	u64 total_bytes = 0;
	u64 last = 0;
	int found = 0;

	if (search_end <= cur_start) {
		WARN_ON(1);
		return 0;
	}

	spin_lock(&tree->lock);
	if (cur_start == 0 && bits == EXTENT_DIRTY) {
		total_bytes = tree->dirty_bytes;
		goto out;
	}
	/*
	 * this search will find all the extents that end after
	 * our range starts.
	 */
	node = tree_search(tree, cur_start);
	if (!node)
		goto out;

	while (1) {
		state = rb_entry(node, struct extent_state, rb_node);
		if (state->start > search_end)
			break;
		if (contig && found && state->start > last + 1)
			break;
		if (state->end >= cur_start && (state->state & bits) == bits) {
			total_bytes += min(search_end, state->end) + 1 -
				       max(cur_start, state->start);
			if (total_bytes >= max_bytes)
				break;
			if (!found) {
				*start = max(cur_start, state->start);
				found = 1;
			}
			last = state->end;
		} else if (contig && found) {
			break;
		}
		node = rb_next(node);
		if (!node)
			break;
	}
out:
	spin_unlock(&tree->lock);
	return total_bytes;
}

/*
 * set the private field for a given byte offset in the tree.  If there isn't
 * an extent_state there already, this does nothing.
 */
int set_state_private(struct extent_io_tree *tree, u64 start, u64 private)
{
	struct rb_node *node;
	struct extent_state *state;
	int ret = 0;

	spin_lock(&tree->lock);
	/*
	 * this search will find all the extents that end after
	 * our range starts.
	 */
	node = tree_search(tree, start);
	if (!node) {
		ret = -ENOENT;
		goto out;
	}
	state = rb_entry(node, struct extent_state, rb_node);
	if (state->start != start) {
		ret = -ENOENT;
		goto out;
	}
	state->private = private;
out:
	spin_unlock(&tree->lock);
	return ret;
}

int get_state_private(struct extent_io_tree *tree, u64 start, u64 *private)
{
	struct rb_node *node;
	struct extent_state *state;
	int ret = 0;

	spin_lock(&tree->lock);
	/*
	 * this search will find all the extents that end after
	 * our range starts.
	 */
	node = tree_search(tree, start);
	if (!node) {
		ret = -ENOENT;
		goto out;
	}
	state = rb_entry(node, struct extent_state, rb_node);
	if (state->start != start) {
		ret = -ENOENT;
		goto out;
	}
	*private = state->private;
out:
	spin_unlock(&tree->lock);
	return ret;
}

/*
 * searches a range in the state tree for a given mask.
 * If 'filled' == 1, this returns 1 only if every extent in the tree
 * has the bits set.  Otherwise, 1 is returned if any bit in the
 * range is found set.
 */
int test_range_bit(struct extent_io_tree *tree, u64 start, u64 end,
		   int bits, int filled, struct extent_state *cached)
{
	struct extent_state *state = NULL;
	struct rb_node *node;
	int bitset = 0;

	spin_lock(&tree->lock);
	if (cached && cached->tree && cached->start <= start &&
	    cached->end > start)
		node = &cached->rb_node;
	else
		node = tree_search(tree, start);
	while (node && start <= end) {
		state = rb_entry(node, struct extent_state, rb_node);

		if (filled && state->start > start) {
			bitset = 0;
			break;
		}

		if (state->start > end)
			break;

		if (state->state & bits) {
			bitset = 1;
			if (!filled)
				break;
		} else if (filled) {
			bitset = 0;
			break;
		}

		if (state->end == (u64)-1)
			break;

		start = state->end + 1;
		if (start > end)
			break;
		node = rb_next(node);
		if (!node) {
			if (filled)
				bitset = 0;
			break;
		}
	}
	spin_unlock(&tree->lock);
	return bitset;
}

/*
 * helper function to set a given page up to date if all the
 * extents in the tree for that page are up to date
 */
static void check_page_uptodate(struct extent_io_tree *tree, struct page *page)
{
	u64 start = (u64)page->index << PAGE_CACHE_SHIFT;
	u64 end = start + PAGE_CACHE_SIZE - 1;
	if (test_range_bit(tree, start, end, EXTENT_UPTODATE, 1, NULL))
		SetPageUptodate(page);
}

/*
 * helper function to unlock a page if all the extents in the tree
 * for that page are unlocked
 */
static void check_page_locked(struct extent_io_tree *tree, struct page *page)
{
	u64 start = (u64)page->index << PAGE_CACHE_SHIFT;
	u64 end = start + PAGE_CACHE_SIZE - 1;
	if (!test_range_bit(tree, start, end, EXTENT_LOCKED, 0, NULL))
		unlock_page(page);
}

/*
 * helper function to end page writeback if all the extents
 * in the tree for that page are done with writeback
 */
static void check_page_writeback(struct extent_io_tree *tree,
				 struct page *page)
{
	end_page_writeback(page);
}

/*
 * When IO fails, either with EIO or csum verification fails, we
 * try other mirrors that might have a good copy of the data.  This
 * io_failure_record is used to record state as we go through all the
 * mirrors.  If another mirror has good data, the page is set up to date
 * and things continue.  If a good mirror can't be found, the original
 * bio end_io callback is called to indicate things have failed.
 */
struct io_failure_record {
	struct page *page;
	u64 start;
	u64 len;
	u64 logical;
	unsigned long bio_flags;
	int this_mirror;
	int failed_mirror;
	int in_validation;
};

static int free_io_failure(struct inode *inode, struct io_failure_record *rec,
				int did_repair)
{
	int ret;
	int err = 0;
	struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;

	set_state_private(failure_tree, rec->start, 0);
	ret = clear_extent_bits(failure_tree, rec->start,
				rec->start + rec->len - 1,
				EXTENT_LOCKED | EXTENT_DIRTY, GFP_NOFS);
	if (ret)
		err = ret;

	if (did_repair) {
		ret = clear_extent_bits(&BTRFS_I(inode)->io_tree, rec->start,
					rec->start + rec->len - 1,
					EXTENT_DAMAGED, GFP_NOFS);
		if (ret && !err)
			err = ret;
	}

	kfree(rec);
	return err;
}

static void repair_io_failure_callback(struct bio *bio, int err)
{
	complete(bio->bi_private);
}

/*
 * this bypasses the standard btrfs submit functions deliberately, as
 * the standard behavior is to write all copies in a raid setup. here we only
 * want to write the one bad copy. so we do the mapping for ourselves and issue
 * submit_bio directly.
 * to avoid any synchonization issues, wait for the data after writing, which
 * actually prevents the read that triggered the error from finishing.
 * currently, there can be no more than two copies of every data bit. thus,
 * exactly one rewrite is required.
 */
int repair_io_failure(struct btrfs_mapping_tree *map_tree, u64 start,
			u64 length, u64 logical, struct page *page,
			int mirror_num)
{
	struct bio *bio;
	struct btrfs_device *dev;
	DECLARE_COMPLETION_ONSTACK(compl);
	u64 map_length = 0;
	u64 sector;
	struct btrfs_bio *bbio = NULL;
	int ret;

	BUG_ON(!mirror_num);

	bio = bio_alloc(GFP_NOFS, 1);
	if (!bio)
		return -EIO;
	bio->bi_private = &compl;
	bio->bi_end_io = repair_io_failure_callback;
	bio->bi_size = 0;
	map_length = length;

	ret = btrfs_map_block(map_tree, WRITE, logical,
			      &map_length, &bbio, mirror_num);
	if (ret) {
		bio_put(bio);
		return -EIO;
	}
	BUG_ON(mirror_num != bbio->mirror_num);
	sector = bbio->stripes[mirror_num-1].physical >> 9;
	bio->bi_sector = sector;
	dev = bbio->stripes[mirror_num-1].dev;
	kfree(bbio);
	if (!dev || !dev->bdev || !dev->writeable) {
		bio_put(bio);
		return -EIO;
	}
	bio->bi_bdev = dev->bdev;
	bio_add_page(bio, page, length, start-page_offset(page));
	btrfsic_submit_bio(WRITE_SYNC, bio);
	wait_for_completion(&compl);

	if (!test_bit(BIO_UPTODATE, &bio->bi_flags)) {
		/* try to remap that extent elsewhere? */
		bio_put(bio);
		btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_WRITE_ERRS);
		return -EIO;
	}

	printk_ratelimited_in_rcu(KERN_INFO "btrfs read error corrected: ino %lu off %llu "
		      "(dev %s sector %llu)\n", page->mapping->host->i_ino,
		      start, rcu_str_deref(dev->name), sector);

	bio_put(bio);
	return 0;
}

int repair_eb_io_failure(struct btrfs_root *root, struct extent_buffer *eb,
			 int mirror_num)
{
	struct btrfs_mapping_tree *map_tree = &root->fs_info->mapping_tree;
	u64 start = eb->start;
	unsigned long i, num_pages = num_extent_pages(eb->start, eb->len);
	int ret = 0;

	for (i = 0; i < num_pages; i++) {
		struct page *p = extent_buffer_page(eb, i);
		ret = repair_io_failure(map_tree, start, PAGE_CACHE_SIZE,
					start, p, mirror_num);
		if (ret)
			break;
		start += PAGE_CACHE_SIZE;
	}

	return ret;
}

/*
 * each time an IO finishes, we do a fast check in the IO failure tree
 * to see if we need to process or clean up an io_failure_record
 */
static int clean_io_failure(u64 start, struct page *page)
{
	u64 private;
	u64 private_failure;
	struct io_failure_record *failrec;
	struct btrfs_mapping_tree *map_tree;
	struct extent_state *state;
	int num_copies;
	int did_repair = 0;
	int ret;
	struct inode *inode = page->mapping->host;

	private = 0;
	ret = count_range_bits(&BTRFS_I(inode)->io_failure_tree, &private,
				(u64)-1, 1, EXTENT_DIRTY, 0);
	if (!ret)
		return 0;

	ret = get_state_private(&BTRFS_I(inode)->io_failure_tree, start,
				&private_failure);
	if (ret)
		return 0;

	failrec = (struct io_failure_record *)(unsigned long) private_failure;
	BUG_ON(!failrec->this_mirror);

	if (failrec->in_validation) {
		/* there was no real error, just free the record */
		pr_debug("clean_io_failure: freeing dummy error at %llu\n",
			 failrec->start);
		did_repair = 1;
		goto out;
	}

	spin_lock(&BTRFS_I(inode)->io_tree.lock);
	state = find_first_extent_bit_state(&BTRFS_I(inode)->io_tree,
					    failrec->start,
					    EXTENT_LOCKED);
	spin_unlock(&BTRFS_I(inode)->io_tree.lock);

	if (state && state->start == failrec->start) {
		num_copies = btrfs_num_copies(BTRFS_I(inode)->root->fs_info,
					      failrec->logical, failrec->len);
		if (num_copies > 1)  {
			map_tree = &BTRFS_I(inode)->root->fs_info->mapping_tree;
			ret = repair_io_failure(map_tree, start, failrec->len,
						failrec->logical, page,
						failrec->failed_mirror);
			did_repair = !ret;
		}
	}

out:
	if (!ret)
		ret = free_io_failure(inode, failrec, did_repair);

	return ret;
}

/*
 * this is a generic handler for readpage errors (default
 * readpage_io_failed_hook). if other copies exist, read those and write back
 * good data to the failed position. does not investigate in remapping the
 * failed extent elsewhere, hoping the device will be smart enough to do this as
 * needed
 */

static int bio_readpage_error(struct bio *failed_bio, struct page *page,
				u64 start, u64 end, int failed_mirror,
				struct extent_state *state)
{
	struct io_failure_record *failrec = NULL;
	u64 private;
	struct extent_map *em;
	struct inode *inode = page->mapping->host;
	struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;
	struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
	struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
	struct bio *bio;
	int num_copies;
	int ret;
	int read_mode;
	u64 logical;

	BUG_ON(failed_bio->bi_rw & REQ_WRITE);

	ret = get_state_private(failure_tree, start, &private);
	if (ret) {
		failrec = kzalloc(sizeof(*failrec), GFP_NOFS);
		if (!failrec)
			return -ENOMEM;
		failrec->start = start;
		failrec->len = end - start + 1;
		failrec->this_mirror = 0;
		failrec->bio_flags = 0;
		failrec->in_validation = 0;

		read_lock(&em_tree->lock);
		em = lookup_extent_mapping(em_tree, start, failrec->len);
		if (!em) {
			read_unlock(&em_tree->lock);
			kfree(failrec);
			return -EIO;
		}

		if (em->start > start || em->start + em->len < start) {
			free_extent_map(em);
			em = NULL;
		}
		read_unlock(&em_tree->lock);

		if (!em) {
			kfree(failrec);
			return -EIO;
		}
		logical = start - em->start;
		logical = em->block_start + logical;
		if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) {
			logical = em->block_start;
			failrec->bio_flags = EXTENT_BIO_COMPRESSED;
			extent_set_compress_type(&failrec->bio_flags,
						 em->compress_type);
		}
		pr_debug("bio_readpage_error: (new) logical=%llu, start=%llu, "
			 "len=%llu\n", logical, start, failrec->len);
		failrec->logical = logical;
		free_extent_map(em);

		/* set the bits in the private failure tree */
		ret = set_extent_bits(failure_tree, start, end,
					EXTENT_LOCKED | EXTENT_DIRTY, GFP_NOFS);
		if (ret >= 0)
			ret = set_state_private(failure_tree, start,
						(u64)(unsigned long)failrec);
		/* set the bits in the inode's tree */
		if (ret >= 0)
			ret = set_extent_bits(tree, start, end, EXTENT_DAMAGED,
						GFP_NOFS);
		if (ret < 0) {
			kfree(failrec);
			return ret;
		}
	} else {
		failrec = (struct io_failure_record *)(unsigned long)private;
		pr_debug("bio_readpage_error: (found) logical=%llu, "
			 "start=%llu, len=%llu, validation=%d\n",
			 failrec->logical, failrec->start, failrec->len,
			 failrec->in_validation);
		/*
		 * when data can be on disk more than twice, add to failrec here
		 * (e.g. with a list for failed_mirror) to make
		 * clean_io_failure() clean all those errors at once.
		 */
	}
	num_copies = btrfs_num_copies(BTRFS_I(inode)->root->fs_info,
				      failrec->logical, failrec->len);
	if (num_copies == 1) {
		/*
		 * we only have a single copy of the data, so don't bother with
		 * all the retry and error correction code that follows. no
		 * matter what the error is, it is very likely to persist.
		 */
		pr_debug("bio_readpage_error: cannot repair, num_copies == 1. "
			 "state=%p, num_copies=%d, next_mirror %d, "
			 "failed_mirror %d\n", state, num_copies,
			 failrec->this_mirror, failed_mirror);
		free_io_failure(inode, failrec, 0);
		return -EIO;
	}

	if (!state) {
		spin_lock(&tree->lock);
		state = find_first_extent_bit_state(tree, failrec->start,
						    EXTENT_LOCKED);
		if (state && state->start != failrec->start)
			state = NULL;
		spin_unlock(&tree->lock);
	}

	/*
	 * there are two premises:
	 *	a) deliver good data to the caller
	 *	b) correct the bad sectors on disk
	 */
	if (failed_bio->bi_vcnt > 1) {
		/*
		 * to fulfill b), we need to know the exact failing sectors, as
		 * we don't want to rewrite any more than the failed ones. thus,
		 * we need separate read requests for the failed bio
		 *
		 * if the following BUG_ON triggers, our validation request got
		 * merged. we need separate requests for our algorithm to work.
		 */
		BUG_ON(failrec->in_validation);
		failrec->in_validation = 1;
		failrec->this_mirror = failed_mirror;
		read_mode = READ_SYNC | REQ_FAILFAST_DEV;
	} else {
		/*
		 * we're ready to fulfill a) and b) alongside. get a good copy
		 * of the failed sector and if we succeed, we have setup
		 * everything for repair_io_failure to do the rest for us.
		 */
		if (failrec->in_validation) {
			BUG_ON(failrec->this_mirror != failed_mirror);
			failrec->in_validation = 0;
			failrec->this_mirror = 0;
		}
		failrec->failed_mirror = failed_mirror;
		failrec->this_mirror++;
		if (failrec->this_mirror == failed_mirror)
			failrec->this_mirror++;
		read_mode = READ_SYNC;
	}

	if (!state || failrec->this_mirror > num_copies) {
		pr_debug("bio_readpage_error: (fail) state=%p, num_copies=%d, "
			 "next_mirror %d, failed_mirror %d\n", state,
			 num_copies, failrec->this_mirror, failed_mirror);
		free_io_failure(inode, failrec, 0);
		return -EIO;
	}

	bio = bio_alloc(GFP_NOFS, 1);
	if (!bio) {
		free_io_failure(inode, failrec, 0);
		return -EIO;
	}
	bio->bi_private = state;
	bio->bi_end_io = failed_bio->bi_end_io;
	bio->bi_sector = failrec->logical >> 9;
	bio->bi_bdev = BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev;
	bio->bi_size = 0;

	bio_add_page(bio, page, failrec->len, start - page_offset(page));

	pr_debug("bio_readpage_error: submitting new read[%#x] to "
		 "this_mirror=%d, num_copies=%d, in_validation=%d\n", read_mode,
		 failrec->this_mirror, num_copies, failrec->in_validation);

	ret = tree->ops->submit_bio_hook(inode, read_mode, bio,
					 failrec->this_mirror,
					 failrec->bio_flags, 0);
	return ret;
}

/* lots and lots of room for performance fixes in the end_bio funcs */

int end_extent_writepage(struct page *page, int err, u64 start, u64 end)
{
	int uptodate = (err == 0);
	struct extent_io_tree *tree;
	int ret;

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

	if (tree->ops && tree->ops->writepage_end_io_hook) {
		ret = tree->ops->writepage_end_io_hook(page, start,
					       end, NULL, uptodate);
		if (ret)
			uptodate = 0;
	}

	if (!uptodate) {
		ClearPageUptodate(page);
		SetPageError(page);
	}
	return 0;
}

/*
 * after a writepage IO is done, we need to:
 * clear the uptodate bits on error
 * clear the writeback bits in the extent tree for this IO
 * end_page_writeback if the page has no more pending IO
 *
 * Scheduling is not allowed, so the extent state tree is expected
 * to have one and only one object corresponding to this IO.
 */
static void end_bio_extent_writepage(struct bio *bio, int err)
{
	struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1;
	struct extent_io_tree *tree;
	u64 start;
	u64 end;
	int whole_page;

	do {
		struct page *page = bvec->bv_page;
		tree = &BTRFS_I(page->mapping->host)->io_tree;

		start = ((u64)page->index << PAGE_CACHE_SHIFT) +
			 bvec->bv_offset;
		end = start + bvec->bv_len - 1;

		if (bvec->bv_offset == 0 && bvec->bv_len == PAGE_CACHE_SIZE)
			whole_page = 1;
		else
			whole_page = 0;

		if (--bvec >= bio->bi_io_vec)
			prefetchw(&bvec->bv_page->flags);

		if (end_extent_writepage(page, err, start, end))
			continue;

		if (whole_page)
			end_page_writeback(page);
		else
			check_page_writeback(tree, page);
	} while (bvec >= bio->bi_io_vec);

	bio_put(bio);
}

/*
 * after a readpage IO is done, we need to:
 * clear the uptodate bits on error
 * set the uptodate bits if things worked
 * set the page up to date if all extents in the tree are uptodate
 * clear the lock bit in the extent tree
 * unlock the page if there are no other extents locked for it
 *
 * Scheduling is not allowed, so the extent state tree is expected
 * to have one and only one object corresponding to this IO.
 */
static void end_bio_extent_readpage(struct bio *bio, int err)
{
	int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
	struct bio_vec *bvec_end = bio->bi_io_vec + bio->bi_vcnt - 1;
	struct bio_vec *bvec = bio->bi_io_vec;
	struct extent_io_tree *tree;
	u64 start;
	u64 end;
	int whole_page;
	int mirror;
	int ret;

	if (err)
		uptodate = 0;

	do {
		struct page *page = bvec->bv_page;
		struct extent_state *cached = NULL;
		struct extent_state *state;

		pr_debug("end_bio_extent_readpage: bi_sector=%llu, err=%d, "
			 "mirror=%ld\n", (u64)bio->bi_sector, err,
			 (long int)bio->bi_bdev);
		tree = &BTRFS_I(page->mapping->host)->io_tree;

		start = ((u64)page->index << PAGE_CACHE_SHIFT) +
			bvec->bv_offset;
		end = start + bvec->bv_len - 1;

		if (bvec->bv_offset == 0 && bvec->bv_len == PAGE_CACHE_SIZE)
			whole_page = 1;
		else
			whole_page = 0;

		if (++bvec <= bvec_end)
			prefetchw(&bvec->bv_page->flags);

		spin_lock(&tree->lock);
		state = find_first_extent_bit_state(tree, start, EXTENT_LOCKED);
		if (state && state->start == start) {
			/*
			 * take a reference on the state, unlock will drop
			 * the ref
			 */
			cache_state(state, &cached);
		}
		spin_unlock(&tree->lock);

		mirror = (int)(unsigned long)bio->bi_bdev;
		if (uptodate && tree->ops && tree->ops->readpage_end_io_hook) {
			ret = tree->ops->readpage_end_io_hook(page, start, end,
							      state, mirror);
			if (ret)
				uptodate = 0;
			else
				clean_io_failure(start, page);
		}

		if (!uptodate && tree->ops && tree->ops->readpage_io_failed_hook) {
			ret = tree->ops->readpage_io_failed_hook(page, mirror);
			if (!ret && !err &&
			    test_bit(BIO_UPTODATE, &bio->bi_flags))
				uptodate = 1;
		} else if (!uptodate) {
			/*
			 * The generic bio_readpage_error handles errors the
			 * following way: If possible, new read requests are
			 * created and submitted and will end up in
			 * end_bio_extent_readpage as well (if we're lucky, not
			 * in the !uptodate case). In that case it returns 0 and
			 * we just go on with the next page in our bio. If it
			 * can't handle the error it will return -EIO and we
			 * remain responsible for that page.
			 */