transaction.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/slab.h>
#include <linux/sched.h>
#include <linux/writeback.h>
#include <linux/pagemap.h>
#include <linux/blkdev.h>
#include <linux/uuid.h>
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
#include "locking.h"
#include "tree-log.h"
#include "inode-map.h"
#include "volumes.h"
#include "dev-replace.h"
#include "qgroup.h"

#define BTRFS_ROOT_TRANS_TAG 0

static const unsigned int btrfs_blocked_trans_types[TRANS_STATE_MAX] = {
	[TRANS_STATE_RUNNING]		= 0U,
	[TRANS_STATE_BLOCKED]		= (__TRANS_USERSPACE |
					   __TRANS_START),
	[TRANS_STATE_COMMIT_START]	= (__TRANS_USERSPACE |
					   __TRANS_START |
					   __TRANS_ATTACH),
	[TRANS_STATE_COMMIT_DOING]	= (__TRANS_USERSPACE |
					   __TRANS_START |
					   __TRANS_ATTACH |
					   __TRANS_JOIN),
	[TRANS_STATE_UNBLOCKED]		= (__TRANS_USERSPACE |
					   __TRANS_START |
					   __TRANS_ATTACH |
					   __TRANS_JOIN |
					   __TRANS_JOIN_NOLOCK),
	[TRANS_STATE_COMPLETED]		= (__TRANS_USERSPACE |
					   __TRANS_START |
					   __TRANS_ATTACH |
					   __TRANS_JOIN |
					   __TRANS_JOIN_NOLOCK),
};

void btrfs_put_transaction(struct btrfs_transaction *transaction)
{
	WARN_ON(atomic_read(&transaction->use_count) == 0);
	if (atomic_dec_and_test(&transaction->use_count)) {
		BUG_ON(!list_empty(&transaction->list));
		WARN_ON(!RB_EMPTY_ROOT(&transaction->delayed_refs.href_root));
		if (transaction->delayed_refs.pending_csums)
			printk(KERN_ERR "pending csums is %llu\n",
			       transaction->delayed_refs.pending_csums);
		while (!list_empty(&transaction->pending_chunks)) {
			struct extent_map *em;

			em = list_first_entry(&transaction->pending_chunks,
					      struct extent_map, list);
			list_del_init(&em->list);
			free_extent_map(em);
		}
		kmem_cache_free(btrfs_transaction_cachep, transaction);
	}
}

static void clear_btree_io_tree(struct extent_io_tree *tree)
{
	spin_lock(&tree->lock);
	/*
	 * Do a single barrier for the waitqueue_active check here, the state
	 * of the waitqueue should not change once clear_btree_io_tree is
	 * called.
	 */
	smp_mb();
	while (!RB_EMPTY_ROOT(&tree->state)) {
		struct rb_node *node;
		struct extent_state *state;

		node = rb_first(&tree->state);
		state = rb_entry(node, struct extent_state, rb_node);
		rb_erase(&state->rb_node, &tree->state);
		RB_CLEAR_NODE(&state->rb_node);
		/*
		 * btree io trees aren't supposed to have tasks waiting for
		 * changes in the flags of extent states ever.
		 */
		ASSERT(!waitqueue_active(&state->wq));
		free_extent_state(state);

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

static noinline void switch_commit_roots(struct btrfs_transaction *trans,
					 struct btrfs_fs_info *fs_info)
{
	struct btrfs_root *root, *tmp;

	down_write(&fs_info->commit_root_sem);
	list_for_each_entry_safe(root, tmp, &trans->switch_commits,
				 dirty_list) {
		list_del_init(&root->dirty_list);
		free_extent_buffer(root->commit_root);
		root->commit_root = btrfs_root_node(root);
		if (is_fstree(root->objectid))
			btrfs_unpin_free_ino(root);
		clear_btree_io_tree(&root->dirty_log_pages);
	}

	/* We can free old roots now. */
	spin_lock(&trans->dropped_roots_lock);
	while (!list_empty(&trans->dropped_roots)) {
		root = list_first_entry(&trans->dropped_roots,
					struct btrfs_root, root_list);
		list_del_init(&root->root_list);
		spin_unlock(&trans->dropped_roots_lock);
		btrfs_drop_and_free_fs_root(fs_info, root);
		spin_lock(&trans->dropped_roots_lock);
	}
	spin_unlock(&trans->dropped_roots_lock);
	up_write(&fs_info->commit_root_sem);
}

static inline void extwriter_counter_inc(struct btrfs_transaction *trans,
					 unsigned int type)
{
	if (type & TRANS_EXTWRITERS)
		atomic_inc(&trans->num_extwriters);
}

static inline void extwriter_counter_dec(struct btrfs_transaction *trans,
					 unsigned int type)
{
	if (type & TRANS_EXTWRITERS)
		atomic_dec(&trans->num_extwriters);
}

static inline void extwriter_counter_init(struct btrfs_transaction *trans,
					  unsigned int type)
{
	atomic_set(&trans->num_extwriters, ((type & TRANS_EXTWRITERS) ? 1 : 0));
}

static inline int extwriter_counter_read(struct btrfs_transaction *trans)
{
	return atomic_read(&trans->num_extwriters);
}

/*
 * either allocate a new transaction or hop into the existing one
 */
static noinline int join_transaction(struct btrfs_root *root, unsigned int type)
{
	struct btrfs_transaction *cur_trans;
	struct btrfs_fs_info *fs_info = root->fs_info;

	spin_lock(&fs_info->trans_lock);
loop:
	/* The file system has been taken offline. No new transactions. */
	if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
		spin_unlock(&fs_info->trans_lock);
		return -EROFS;
	}

	cur_trans = fs_info->running_transaction;
	if (cur_trans) {
		if (cur_trans->aborted) {
			spin_unlock(&fs_info->trans_lock);
			return cur_trans->aborted;
		}
		if (btrfs_blocked_trans_types[cur_trans->state] & type) {
			spin_unlock(&fs_info->trans_lock);
			return -EBUSY;
		}
		atomic_inc(&cur_trans->use_count);
		atomic_inc(&cur_trans->num_writers);
		extwriter_counter_inc(cur_trans, type);
		spin_unlock(&fs_info->trans_lock);
		return 0;
	}
	spin_unlock(&fs_info->trans_lock);

	/*
	 * If we are ATTACH, we just want to catch the current transaction,
	 * and commit it. If there is no transaction, just return ENOENT.
	 */
	if (type == TRANS_ATTACH)
		return -ENOENT;

	/*
	 * JOIN_NOLOCK only happens during the transaction commit, so
	 * it is impossible that ->running_transaction is NULL
	 */
	BUG_ON(type == TRANS_JOIN_NOLOCK);

	cur_trans = kmem_cache_alloc(btrfs_transaction_cachep, GFP_NOFS);
	if (!cur_trans)
		return -ENOMEM;

	spin_lock(&fs_info->trans_lock);
	if (fs_info->running_transaction) {
		/*
		 * someone started a transaction after we unlocked.  Make sure
		 * to redo the checks above
		 */
		kmem_cache_free(btrfs_transaction_cachep, cur_trans);
		goto loop;
	} else if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) {
		spin_unlock(&fs_info->trans_lock);
		kmem_cache_free(btrfs_transaction_cachep, cur_trans);
		return -EROFS;
	}

	atomic_set(&cur_trans->num_writers, 1);
	extwriter_counter_init(cur_trans, type);
	init_waitqueue_head(&cur_trans->writer_wait);
	init_waitqueue_head(&cur_trans->commit_wait);
	init_waitqueue_head(&cur_trans->pending_wait);
	cur_trans->state = TRANS_STATE_RUNNING;
	/*
	 * One for this trans handle, one so it will live on until we
	 * commit the transaction.
	 */
	atomic_set(&cur_trans->use_count, 2);
	cur_trans->have_free_bgs = 0;
	atomic_set(&cur_trans->pending_ordered, 0);
	cur_trans->start_time = get_seconds();
	cur_trans->dirty_bg_run = 0;

	memset(&cur_trans->delayed_refs, 0, sizeof(cur_trans->delayed_refs));

	cur_trans->delayed_refs.href_root = RB_ROOT;
	cur_trans->delayed_refs.dirty_extent_root = RB_ROOT;
	atomic_set(&cur_trans->delayed_refs.num_entries, 0);

	/*
	 * although the tree mod log is per file system and not per transaction,
	 * the log must never go across transaction boundaries.
	 */
	smp_mb();
	if (!list_empty(&fs_info->tree_mod_seq_list))
		WARN(1, KERN_ERR "BTRFS: tree_mod_seq_list not empty when "
			"creating a fresh transaction\n");
	if (!RB_EMPTY_ROOT(&fs_info->tree_mod_log))
		WARN(1, KERN_ERR "BTRFS: tree_mod_log rb tree not empty when "
			"creating a fresh transaction\n");
	atomic64_set(&fs_info->tree_mod_seq, 0);

	spin_lock_init(&cur_trans->delayed_refs.lock);

	INIT_LIST_HEAD(&cur_trans->pending_snapshots);
	INIT_LIST_HEAD(&cur_trans->pending_chunks);
	INIT_LIST_HEAD(&cur_trans->switch_commits);
	INIT_LIST_HEAD(&cur_trans->dirty_bgs);
	INIT_LIST_HEAD(&cur_trans->io_bgs);
	INIT_LIST_HEAD(&cur_trans->dropped_roots);
	mutex_init(&cur_trans->cache_write_mutex);
	cur_trans->num_dirty_bgs = 0;
	spin_lock_init(&cur_trans->dirty_bgs_lock);
	INIT_LIST_HEAD(&cur_trans->deleted_bgs);
	spin_lock_init(&cur_trans->deleted_bgs_lock);
	spin_lock_init(&cur_trans->dropped_roots_lock);
	list_add_tail(&cur_trans->list, &fs_info->trans_list);
	extent_io_tree_init(&cur_trans->dirty_pages,
			     fs_info->btree_inode->i_mapping);
	fs_info->generation++;
	cur_trans->transid = fs_info->generation;
	fs_info->running_transaction = cur_trans;
	cur_trans->aborted = 0;
	spin_unlock(&fs_info->trans_lock);

	return 0;
}

/*
 * this does all the record keeping required to make sure that a reference
 * counted root is properly recorded in a given transaction.  This is required
 * to make sure the old root from before we joined the transaction is deleted
 * when the transaction commits
 */
static int record_root_in_trans(struct btrfs_trans_handle *trans,
			       struct btrfs_root *root)
{
	if (test_bit(BTRFS_ROOT_REF_COWS, &root->state) &&
	    root->last_trans < trans->transid) {
		WARN_ON(root == root->fs_info->extent_root);
		WARN_ON(root->commit_root != root->node);

		/*
		 * see below for IN_TRANS_SETUP usage rules
		 * we have the reloc mutex held now, so there
		 * is only one writer in this function
		 */
		set_bit(BTRFS_ROOT_IN_TRANS_SETUP, &root->state);

		/* make sure readers find IN_TRANS_SETUP before
		 * they find our root->last_trans update
		 */
		smp_wmb();

		spin_lock(&root->fs_info->fs_roots_radix_lock);
		if (root->last_trans == trans->transid) {
			spin_unlock(&root->fs_info->fs_roots_radix_lock);
			return 0;
		}
		radix_tree_tag_set(&root->fs_info->fs_roots_radix,
			   (unsigned long)root->root_key.objectid,
			   BTRFS_ROOT_TRANS_TAG);
		spin_unlock(&root->fs_info->fs_roots_radix_lock);
		root->last_trans = trans->transid;

		/* this is pretty tricky.  We don't want to
		 * take the relocation lock in btrfs_record_root_in_trans
		 * unless we're really doing the first setup for this root in
		 * this transaction.
		 *
		 * Normally we'd use root->last_trans as a flag to decide
		 * if we want to take the expensive mutex.
		 *
		 * But, we have to set root->last_trans before we
		 * init the relocation root, otherwise, we trip over warnings
		 * in ctree.c.  The solution used here is to flag ourselves
		 * with root IN_TRANS_SETUP.  When this is 1, we're still
		 * fixing up the reloc trees and everyone must wait.
		 *
		 * When this is zero, they can trust root->last_trans and fly
		 * through btrfs_record_root_in_trans without having to take the
		 * lock.  smp_wmb() makes sure that all the writes above are
		 * done before we pop in the zero below
		 */
		btrfs_init_reloc_root(trans, root);
		smp_mb__before_atomic();
		clear_bit(BTRFS_ROOT_IN_TRANS_SETUP, &root->state);
	}
	return 0;
}


void btrfs_add_dropped_root(struct btrfs_trans_handle *trans,
			    struct btrfs_root *root)
{
	struct btrfs_transaction *cur_trans = trans->transaction;

	/* Add ourselves to the transaction dropped list */
	spin_lock(&cur_trans->dropped_roots_lock);
	list_add_tail(&root->root_list, &cur_trans->dropped_roots);
	spin_unlock(&cur_trans->dropped_roots_lock);

	/* Make sure we don't try to update the root at commit time */
	spin_lock(&root->fs_info->fs_roots_radix_lock);
	radix_tree_tag_clear(&root->fs_info->fs_roots_radix,
			     (unsigned long)root->root_key.objectid,
			     BTRFS_ROOT_TRANS_TAG);
	spin_unlock(&root->fs_info->fs_roots_radix_lock);
}

int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,
			       struct btrfs_root *root)
{
	if (!test_bit(BTRFS_ROOT_REF_COWS, &root->state))
		return 0;

	/*
	 * see record_root_in_trans for comments about IN_TRANS_SETUP usage
	 * and barriers
	 */
	smp_rmb();
	if (root->last_trans == trans->transid &&
	    !test_bit(BTRFS_ROOT_IN_TRANS_SETUP, &root->state))
		return 0;

	mutex_lock(&root->fs_info->reloc_mutex);
	record_root_in_trans(trans, root);
	mutex_unlock(&root->fs_info->reloc_mutex);

	return 0;
}

static inline int is_transaction_blocked(struct btrfs_transaction *trans)
{
	return (trans->state >= TRANS_STATE_BLOCKED &&
		trans->state < TRANS_STATE_UNBLOCKED &&
		!trans->aborted);
}

/* wait for commit against the current transaction to become unblocked
 * when this is done, it is safe to start a new transaction, but the current
 * transaction might not be fully on disk.
 */
static void wait_current_trans(struct btrfs_root *root)
{
	struct btrfs_transaction *cur_trans;

	spin_lock(&root->fs_info->trans_lock);
	cur_trans = root->fs_info->running_transaction;
	if (cur_trans && is_transaction_blocked(cur_trans)) {
		atomic_inc(&cur_trans->use_count);
		spin_unlock(&root->fs_info->trans_lock);

		wait_event(root->fs_info->transaction_wait,
			   cur_trans->state >= TRANS_STATE_UNBLOCKED ||
			   cur_trans->aborted);
		btrfs_put_transaction(cur_trans);
	} else {
		spin_unlock(&root->fs_info->trans_lock);
	}
}

static int may_wait_transaction(struct btrfs_root *root, int type)
{
	if (root->fs_info->log_root_recovering)
		return 0;

	if (type == TRANS_USERSPACE)
		return 1;

	if (type == TRANS_START &&
	    !atomic_read(&root->fs_info->open_ioctl_trans))
		return 1;

	return 0;
}

static inline bool need_reserve_reloc_root(struct btrfs_root *root)
{
	if (!root->fs_info->reloc_ctl ||
	    !test_bit(BTRFS_ROOT_REF_COWS, &root->state) ||
	    root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID ||
	    root->reloc_root)
		return false;

	return true;
}

static struct btrfs_trans_handle *
start_transaction(struct btrfs_root *root, unsigned int num_items,
		  unsigned int type, enum btrfs_reserve_flush_enum flush)
{
	struct btrfs_trans_handle *h;
	struct btrfs_transaction *cur_trans;
	u64 num_bytes = 0;
	u64 qgroup_reserved = 0;
	bool reloc_reserved = false;
	int ret;

	/* Send isn't supposed to start transactions. */
	ASSERT(current->journal_info != BTRFS_SEND_TRANS_STUB);

	if (test_bit(BTRFS_FS_STATE_ERROR, &root->fs_info->fs_state))
		return ERR_PTR(-EROFS);

	if (current->journal_info) {
		WARN_ON(type & TRANS_EXTWRITERS);
		h = current->journal_info;
		h->use_count++;
		WARN_ON(h->use_count > 2);
		h->orig_rsv = h->block_rsv;
		h->block_rsv = NULL;
		goto got_it;
	}

	/*
	 * Do the reservation before we join the transaction so we can do all
	 * the appropriate flushing if need be.
	 */
	if (num_items > 0 && root != root->fs_info->chunk_root) {
		if (root->fs_info->quota_enabled &&
		    is_fstree(root->root_key.objectid)) {
			qgroup_reserved = num_items * root->nodesize;
			ret = btrfs_qgroup_reserve(root, qgroup_reserved);
			if (ret)
				return ERR_PTR(ret);
		}

		num_bytes = btrfs_calc_trans_metadata_size(root, num_items);
		/*
		 * Do the reservation for the relocation root creation
		 */
		if (need_reserve_reloc_root(root)) {
			num_bytes += root->nodesize;
			reloc_reserved = true;
		}

		ret = btrfs_block_rsv_add(root,
					  &root->fs_info->trans_block_rsv,
					  num_bytes, flush);
		if (ret)
			goto reserve_fail;
	}
again:
	h = kmem_cache_zalloc(btrfs_trans_handle_cachep, GFP_NOFS);
	if (!h) {
		ret = -ENOMEM;
		goto alloc_fail;
	}

	/*
	 * If we are JOIN_NOLOCK we're already committing a transaction and
	 * waiting on this guy, so we don't need to do the sb_start_intwrite
	 * because we're already holding a ref.  We need this because we could
	 * have raced in and did an fsync() on a file which can kick a commit
	 * and then we deadlock with somebody doing a freeze.
	 *
	 * If we are ATTACH, it means we just want to catch the current
	 * transaction and commit it, so we needn't do sb_start_intwrite(). 
	 */
	if (type & __TRANS_FREEZABLE)
		sb_start_intwrite(root->fs_info->sb);

	if (may_wait_transaction(root, type))
		wait_current_trans(root);

	do {
		ret = join_transaction(root, type);
		if (ret == -EBUSY) {
			wait_current_trans(root);
			if (unlikely(type == TRANS_ATTACH))
				ret = -ENOENT;
		}
	} while (ret == -EBUSY);

	if (ret < 0) {
		/* We must get the transaction if we are JOIN_NOLOCK. */
		BUG_ON(type == TRANS_JOIN_NOLOCK);
		goto join_fail;
	}

	cur_trans = root->fs_info->running_transaction;

	h->transid = cur_trans->transid;
	h->transaction = cur_trans;
	h->root = root;
	h->use_count = 1;
	h->type = type;
	h->can_flush_pending_bgs = true;
	INIT_LIST_HEAD(&h->qgroup_ref_list);
	INIT_LIST_HEAD(&h->new_bgs);

	smp_mb();
	if (cur_trans->state >= TRANS_STATE_BLOCKED &&
	    may_wait_transaction(root, type)) {
		current->journal_info = h;
		btrfs_commit_transaction(h, root);
		goto again;
	}

	if (num_bytes) {
		trace_btrfs_space_reservation(root->fs_info, "transaction",
					      h->transid, num_bytes, 1);
		h->block_rsv = &root->fs_info->trans_block_rsv;
		h->bytes_reserved = num_bytes;
		h->reloc_reserved = reloc_reserved;
	}
	h->qgroup_reserved = qgroup_reserved;

got_it:
	btrfs_record_root_in_trans(h, root);

	if (!current->journal_info && type != TRANS_USERSPACE)
		current->journal_info = h;
	return h;

join_fail:
	if (type & __TRANS_FREEZABLE)
		sb_end_intwrite(root->fs_info->sb);
	kmem_cache_free(btrfs_trans_handle_cachep, h);
alloc_fail:
	if (num_bytes)
		btrfs_block_rsv_release(root, &root->fs_info->trans_block_rsv,
					num_bytes);
reserve_fail:
	if (qgroup_reserved)
		btrfs_qgroup_free(root, qgroup_reserved);
	return ERR_PTR(ret);
}

struct btrfs_trans_handle *btrfs_start_transaction(struct btrfs_root *root,
						   unsigned int num_items)
{
	return start_transaction(root, num_items, TRANS_START,
				 BTRFS_RESERVE_FLUSH_ALL);
}

struct btrfs_trans_handle *btrfs_start_transaction_lflush(
					struct btrfs_root *root,
					unsigned int num_items)
{
	return start_transaction(root, num_items, TRANS_START,
				 BTRFS_RESERVE_FLUSH_LIMIT);
}

struct btrfs_trans_handle *btrfs_join_transaction(struct btrfs_root *root)
{
	return start_transaction(root, 0, TRANS_JOIN, 0);
}

struct btrfs_trans_handle *btrfs_join_transaction_nolock(struct btrfs_root *root)
{
	return start_transaction(root, 0, TRANS_JOIN_NOLOCK, 0);
}

struct btrfs_trans_handle *btrfs_start_ioctl_transaction(struct btrfs_root *root)
{
	return start_transaction(root, 0, TRANS_USERSPACE, 0);
}

/*
 * btrfs_attach_transaction() - catch the running transaction
 *
 * It is used when we want to commit the current the transaction, but
 * don't want to start a new one.
 *
 * Note: If this function return -ENOENT, it just means there is no
 * running transaction. But it is possible that the inactive transaction
 * is still in the memory, not fully on disk. If you hope there is no
 * inactive transaction in the fs when -ENOENT is returned, you should
 * invoke
 *     btrfs_attach_transaction_barrier()
 */
struct btrfs_trans_handle *btrfs_attach_transaction(struct btrfs_root *root)
{
	return start_transaction(root, 0, TRANS_ATTACH, 0);
}

/*
 * btrfs_attach_transaction_barrier() - catch the running transaction
 *
 * It is similar to the above function, the differentia is this one
 * will wait for all the inactive transactions until they fully
 * complete.
 */
struct btrfs_trans_handle *
btrfs_attach_transaction_barrier(struct btrfs_root *root)
{
	struct btrfs_trans_handle *trans;

	trans = start_transaction(root, 0, TRANS_ATTACH, 0);
	if (IS_ERR(trans) && PTR_ERR(trans) == -ENOENT)
		btrfs_wait_for_commit(root, 0);

	return trans;
}

/* wait for a transaction commit to be fully complete */
static noinline void wait_for_commit(struct btrfs_root *root,
				    struct btrfs_transaction *commit)
{
	wait_event(commit->commit_wait, commit->state == TRANS_STATE_COMPLETED);
}

int btrfs_wait_for_commit(struct btrfs_root *root, u64 transid)
{
	struct btrfs_transaction *cur_trans = NULL, *t;
	int ret = 0;

	if (transid) {
		if (transid <= root->fs_info->last_trans_committed)
			goto out;

		/* find specified transaction */
		spin_lock(&root->fs_info->trans_lock);
		list_for_each_entry(t, &root->fs_info->trans_list, list) {
			if (t->transid == transid) {
				cur_trans = t;
				atomic_inc(&cur_trans->use_count);
				ret = 0;
				break;
			}
			if (t->transid > transid) {
				ret = 0;
				break;
			}
		}
		spin_unlock(&root->fs_info->trans_lock);

		/*
		 * The specified transaction doesn't exist, or we
		 * raced with btrfs_commit_transaction
		 */
		if (!cur_trans) {
			if (transid > root->fs_info->last_trans_committed)
				ret = -EINVAL;
			goto out;
		}
	} else {
		/* find newest transaction that is committing | committed */
		spin_lock(&root->fs_info->trans_lock);
		list_for_each_entry_reverse(t, &root->fs_info->trans_list,
					    list) {
			if (t->state >= TRANS_STATE_COMMIT_START) {
				if (t->state == TRANS_STATE_COMPLETED)
					break;
				cur_trans = t;
				atomic_inc(&cur_trans->use_count);
				break;
			}
		}
		spin_unlock(&root->fs_info->trans_lock);
		if (!cur_trans)
			goto out;  /* nothing committing|committed */
	}

	wait_for_commit(root, cur_trans);
	btrfs_put_transaction(cur_trans);
out:
	return ret;
}

void btrfs_throttle(struct btrfs_root *root)
{
	if (!atomic_read(&root->fs_info->open_ioctl_trans))
		wait_current_trans(root);
}

static int should_end_transaction(struct btrfs_trans_handle *trans,
				  struct btrfs_root *root)
{
	if (root->fs_info->global_block_rsv.space_info->full &&
	    btrfs_check_space_for_delayed_refs(trans, root))
		return 1;

	return !!btrfs_block_rsv_check(root, &root->fs_info->global_block_rsv, 5);
}

int btrfs_should_end_transaction(struct btrfs_trans_handle *trans,
				 struct btrfs_root *root)
{
	struct btrfs_transaction *cur_trans = trans->transaction;
	int updates;
	int err;

	smp_mb();
	if (cur_trans->state >= TRANS_STATE_BLOCKED ||
	    cur_trans->delayed_refs.flushing)
		return 1;

	updates = trans->delayed_ref_updates;
	trans->delayed_ref_updates = 0;
	if (updates) {
		err = btrfs_run_delayed_refs(trans, root, updates * 2);
		if (err) /* Error code will also eval true */
			return err;
	}

	return should_end_transaction(trans, root);
}

static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
			  struct btrfs_root *root, int throttle)
{
	struct btrfs_transaction *cur_trans = trans->transaction;
	struct btrfs_fs_info *info = root->fs_info;
	unsigned long cur = trans->delayed_ref_updates;
	int lock = (trans->type != TRANS_JOIN_NOLOCK);
	int err = 0;
	int must_run_delayed_refs = 0;

	if (trans->use_count > 1) {
		trans->use_count--;
		trans->block_rsv = trans->orig_rsv;
		return 0;
	}

	btrfs_trans_release_metadata(trans, root);
	trans->block_rsv = NULL;

	if (!list_empty(&trans->new_bgs))
		btrfs_create_pending_block_groups(trans, root);

	trans->delayed_ref_updates = 0;
	if (!trans->sync) {
		must_run_delayed_refs =
			btrfs_should_throttle_delayed_refs(trans, root);
		cur = max_t(unsigned long, cur, 32);

		/*
		 * don't make the caller wait if they are from a NOLOCK
		 * or ATTACH transaction, it will deadlock with commit
		 */
		if (must_run_delayed_refs == 1 &&
		    (trans->type & (__TRANS_JOIN_NOLOCK | __TRANS_ATTACH)))
			must_run_delayed_refs = 2;
	}

	if (trans->qgroup_reserved) {
		/*
		 * the same root has to be passed here between start_transaction
		 * and end_transaction. Subvolume quota depends on this.
		 */
		btrfs_qgroup_free(trans->root, trans->qgroup_reserved);
		trans->qgroup_reserved = 0;
	}

	btrfs_trans_release_metadata(trans, root);
	trans->block_rsv = NULL;

	if (!list_empty(&trans->new_bgs))
		btrfs_create_pending_block_groups(trans, root);

	btrfs_trans_release_chunk_metadata(trans);

	if (lock && !atomic_read(&root->fs_info->open_ioctl_trans) &&
	    should_end_transaction(trans, root) &&
	    ACCESS_ONCE(cur_trans->state) == TRANS_STATE_RUNNING) {
		spin_lock(&info->trans_lock);
		if (cur_trans->state == TRANS_STATE_RUNNING)
			cur_trans->state = TRANS_STATE_BLOCKED;
		spin_unlock(&info->trans_lock);
	}

	if (lock && ACCESS_ONCE(cur_trans->state) == TRANS_STATE_BLOCKED) {
		if (throttle)
			return btrfs_commit_transaction(trans, root);
		else
			wake_up_process(info->transaction_kthread);
	}

	if (trans->type & __TRANS_FREEZABLE)
		sb_end_intwrite(root->fs_info->sb);

	WARN_ON(cur_trans != info->running_transaction);
	WARN_ON(atomic_read(&cur_trans->num_writers) < 1);
	atomic_dec(&cur_trans->num_writers);
	extwriter_counter_dec(cur_trans, trans->type);

	/*
	 * Make sure counter is updated before we wake up waiters.
	 */
	smp_mb();
	if (waitqueue_active(&cur_trans->writer_wait))
		wake_up(&cur_trans->writer_wait);
	btrfs_put_transaction(cur_trans);

	if (current->journal_info == trans)
		current->journal_info = NULL;

	if (throttle)
		btrfs_run_delayed_iputs(root);

	if (trans->aborted ||
	    test_bit(BTRFS_FS_STATE_ERROR, &root->fs_info->fs_state)) {
		wake_up_process(info->transaction_kthread);
		err = -EIO;
	}
	assert_qgroups_uptodate(trans);

	kmem_cache_free(btrfs_trans_handle_cachep, trans);
	if (must_run_delayed_refs) {
		btrfs_async_run_delayed_refs(root, cur,
					     must_run_delayed_refs == 1);
	}
	return err;
}

int btrfs_end_transaction(struct btrfs_trans_handle *trans,
			  struct btrfs_root *root)
{
	return __btrfs_end_transaction(trans, root, 0);
}

int btrfs_end_transaction_throttle(struct btrfs_trans_handle *trans,
				   struct btrfs_root *root)
{
	return __btrfs_end_transaction(trans, root, 1);
}

/*
 * when btree blocks are allocated, they have some corresponding bits set for
 * them in one of two extent_io trees.  This is used to make sure all of
 * those extents are sent to disk but does not wait on them
 */
int btrfs_write_marked_extents(struct btrfs_root *root,
			       struct extent_io_tree *dirty_pages, int mark)
{
	int err = 0;
	int werr = 0;
	struct address_space *mapping = root->fs_info->btree_inode->i_mapping;
	struct extent_state *cached_state = NULL;
	u64 start = 0;
	u64 end;

	while (!find_first_extent_bit(dirty_pages, start, &start, &end,
				      mark, &cached_state)) {
		bool wait_writeback = false;

		err = convert_extent_bit(dirty_pages, start, end,
					 EXTENT_NEED_WAIT,
					 mark, &cached_state, GFP_NOFS);
		/*
		 * convert_extent_bit can return -ENOMEM, which is most of the
		 * time a temporary error. So when it happens, ignore the error
		 * and wait for writeback of this range to finish - because we
		 * failed to set the bit EXTENT_NEED_WAIT for the range, a call
		 * to btrfs_wait_marked_extents() would not know that writeback
		 * for this range started and therefore wouldn't wait for it to
		 * finish - we don't want to commit a superblock that points to
		 * btree nodes/leafs for which writeback hasn't finished yet
		 * (and without errors).
		 * We cleanup any entries left in the io tree when committing
		 * the transaction (through clear_btree_io_tree()).
		 */
		if (err == -ENOMEM) {
			err = 0;
			wait_writeback = true;
		}
		if (!err)
			err = filemap_fdatawrite_range(mapping, start, end);
		if (err)
			werr = err;
		else if (wait_writeback)
			werr = filemap_fdatawait_range(mapping, start, end);