1 files changed, 69 insertions, 147 deletions

diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index 69582d5b69d1..ac734ec4cc20 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -67,7 +67,7 @@ static void ordered_data_tree_panic(struct inode *inode, int errno,
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	btrfs_panic(fs_info, errno, "Inconsistency in ordered tree at offset "
-		    "%llu\n", offset);
+		    "%llu", offset);
 }
 
 /*
@@ -336,22 +336,26 @@ int btrfs_dec_test_first_ordered_pending(struct inode *inode,
 		      entry->len);
 	*file_offset = dec_end;
 	if (dec_start > dec_end) {
-		printk(KERN_CRIT "bad ordering dec_start %llu end %llu\n",
-		       dec_start, dec_end);
+		btrfs_crit(BTRFS_I(inode)->root->fs_info,
+			"bad ordering dec_start %llu end %llu", dec_start, dec_end);
 	}
 	to_dec = dec_end - dec_start;
 	if (to_dec > entry->bytes_left) {
-		printk(KERN_CRIT "bad ordered accounting left %llu size %llu\n",
-		       entry->bytes_left, to_dec);
+		btrfs_crit(BTRFS_I(inode)->root->fs_info,
+			"bad ordered accounting left %llu size %llu",
+			entry->bytes_left, to_dec);
 	}
 	entry->bytes_left -= to_dec;
 	if (!uptodate)
 		set_bit(BTRFS_ORDERED_IOERR, &entry->flags);
 
-	if (entry->bytes_left == 0)
+	if (entry->bytes_left == 0) {
 		ret = test_and_set_bit(BTRFS_ORDERED_IO_DONE, &entry->flags);
-	else
+		if (waitqueue_active(&entry->wait))
+			wake_up(&entry->wait);
+	} else {
 		ret = 1;
+	}
 out:
 	if (!ret && cached && entry) {
 		*cached = entry;
@@ -401,17 +405,21 @@ have_entry:
 	}
 
 	if (io_size > entry->bytes_left) {
-		printk(KERN_CRIT "bad ordered accounting left %llu size %llu\n",
+		btrfs_crit(BTRFS_I(inode)->root->fs_info,
+			   "bad ordered accounting left %llu size %llu",
 		       entry->bytes_left, io_size);
 	}
 	entry->bytes_left -= io_size;
 	if (!uptodate)
 		set_bit(BTRFS_ORDERED_IOERR, &entry->flags);
 
-	if (entry->bytes_left == 0)
+	if (entry->bytes_left == 0) {
 		ret = test_and_set_bit(BTRFS_ORDERED_IO_DONE, &entry->flags);
-	else
+		if (waitqueue_active(&entry->wait))
+			wake_up(&entry->wait);
+	} else {
 		ret = 1;
+	}
 out:
 	if (!ret && cached && entry) {
 		*cached = entry;
@@ -422,27 +430,48 @@ out:
 }
 
 /* Needs to either be called under a log transaction or the log_mutex */
-void btrfs_get_logged_extents(struct btrfs_root *log, struct inode *inode)
+void btrfs_get_logged_extents(struct inode *inode,
+			      struct list_head *logged_list)
 {
 	struct btrfs_ordered_inode_tree *tree;
 	struct btrfs_ordered_extent *ordered;
 	struct rb_node *n;
-	int index = log->log_transid % 2;
 
 	tree = &BTRFS_I(inode)->ordered_tree;
 	spin_lock_irq(&tree->lock);
 	for (n = rb_first(&tree->tree); n; n = rb_next(n)) {
 		ordered = rb_entry(n, struct btrfs_ordered_extent, rb_node);
-		spin_lock(&log->log_extents_lock[index]);
-		if (list_empty(&ordered->log_list)) {
-			list_add_tail(&ordered->log_list, &log->logged_list[index]);
-			atomic_inc(&ordered->refs);
-		}
-		spin_unlock(&log->log_extents_lock[index]);
+		if (!list_empty(&ordered->log_list))
+			continue;
+		list_add_tail(&ordered->log_list, logged_list);
+		atomic_inc(&ordered->refs);
 	}
 	spin_unlock_irq(&tree->lock);
 }
 
+void btrfs_put_logged_extents(struct list_head *logged_list)
+{
+	struct btrfs_ordered_extent *ordered;
+
+	while (!list_empty(logged_list)) {
+		ordered = list_first_entry(logged_list,
+					   struct btrfs_ordered_extent,
+					   log_list);
+		list_del_init(&ordered->log_list);
+		btrfs_put_ordered_extent(ordered);
+	}
+}
+
+void btrfs_submit_logged_extents(struct list_head *logged_list,
+				 struct btrfs_root *log)
+{
+	int index = log->log_transid % 2;
+
+	spin_lock_irq(&log->log_extents_lock[index]);
+	list_splice_tail(logged_list, &log->logged_list[index]);
+	spin_unlock_irq(&log->log_extents_lock[index]);
+}
+
 void btrfs_wait_logged_extents(struct btrfs_root *log, u64 transid)
 {
 	struct btrfs_ordered_extent *ordered;
@@ -455,8 +484,19 @@ void btrfs_wait_logged_extents(struct btrfs_root *log, u64 transid)
 					   log_list);
 		list_del_init(&ordered->log_list);
 		spin_unlock_irq(&log->log_extents_lock[index]);
+
+		if (!test_bit(BTRFS_ORDERED_IO_DONE, &ordered->flags) &&
+		    !test_bit(BTRFS_ORDERED_DIRECT, &ordered->flags)) {
+			struct inode *inode = ordered->inode;
+			u64 start = ordered->file_offset;
+			u64 end = ordered->file_offset + ordered->len - 1;
+
+			WARN_ON(!inode);
+			filemap_fdatawrite_range(inode->i_mapping, start, end);
+		}
 		wait_event(ordered->wait, test_bit(BTRFS_ORDERED_IO_DONE,
 						   &ordered->flags));
+
 		btrfs_put_ordered_extent(ordered);
 		spin_lock_irq(&log->log_extents_lock[index]);
 	}
@@ -520,7 +560,8 @@ void btrfs_remove_ordered_extent(struct inode *inode,
 	spin_lock_irq(&tree->lock);
 	node = &entry->rb_node;
 	rb_erase(node, &tree->tree);
-	tree->last = NULL;
+	if (tree->last == node)
+		tree->last = NULL;
 	set_bit(BTRFS_ORDERED_COMPLETE, &entry->flags);
 	spin_unlock_irq(&tree->lock);
 
@@ -530,18 +571,6 @@ void btrfs_remove_ordered_extent(struct inode *inode,
 
 	trace_btrfs_ordered_extent_remove(inode, entry);
 
-	/*
-	 * we have no more ordered extents for this inode and
-	 * no dirty pages.  We can safely remove it from the
-	 * list of ordered extents
-	 */
-	if (RB_EMPTY_ROOT(&tree->tree) &&
-	    !mapping_tagged(inode->i_mapping, PAGECACHE_TAG_DIRTY)) {
-		spin_lock(&root->fs_info->ordered_root_lock);
-		list_del_init(&BTRFS_I(inode)->ordered_operations);
-		spin_unlock(&root->fs_info->ordered_root_lock);
-	}
-
 	if (!root->nr_ordered_extents) {
 		spin_lock(&root->fs_info->ordered_root_lock);
 		BUG_ON(list_empty(&root->ordered_root));
@@ -574,7 +603,7 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr)
 	INIT_LIST_HEAD(&splice);
 	INIT_LIST_HEAD(&works);
 
-	mutex_lock(&root->fs_info->ordered_operations_mutex);
+	mutex_lock(&root->ordered_extent_mutex);
 	spin_lock(&root->ordered_extent_lock);
 	list_splice_init(&root->ordered_extents, &splice);
 	while (!list_empty(&splice) && nr) {
@@ -585,10 +614,12 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr)
 		atomic_inc(&ordered->refs);
 		spin_unlock(&root->ordered_extent_lock);
 
-		ordered->flush_work.func = btrfs_run_ordered_extent_work;
+		btrfs_init_work(&ordered->flush_work,
+				btrfs_flush_delalloc_helper,
+				btrfs_run_ordered_extent_work, NULL, NULL);
 		list_add_tail(&ordered->work_list, &works);
-		btrfs_queue_worker(&root->fs_info->flush_workers,
-				   &ordered->flush_work);
+		btrfs_queue_work(root->fs_info->flush_workers,
+				 &ordered->flush_work);
 
 		cond_resched();
 		spin_lock(&root->ordered_extent_lock);
@@ -605,7 +636,7 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr)
 		btrfs_put_ordered_extent(ordered);
 		cond_resched();
 	}
-	mutex_unlock(&root->fs_info->ordered_operations_mutex);
+	mutex_unlock(&root->ordered_extent_mutex);
 
 	return count;
 }
@@ -618,6 +649,7 @@ void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr)
 
 	INIT_LIST_HEAD(&splice);
 
+	mutex_lock(&fs_info->ordered_operations_mutex);
 	spin_lock(&fs_info->ordered_root_lock);
 	list_splice_init(&fs_info->ordered_roots, &splice);
 	while (!list_empty(&splice) && nr) {
@@ -640,81 +672,7 @@ void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr)
 	}
 	list_splice_tail(&splice, &fs_info->ordered_roots);
 	spin_unlock(&fs_info->ordered_root_lock);
-}
-
-/*
- * this is used during transaction commit to write all the inodes
- * added to the ordered operation list.  These files must be fully on
- * disk before the transaction commits.
- *
- * we have two modes here, one is to just start the IO via filemap_flush
- * and the other is to wait for all the io.  When we wait, we have an
- * extra check to make sure the ordered operation list really is empty
- * before we return
- */
-int btrfs_run_ordered_operations(struct btrfs_trans_handle *trans,
-				 struct btrfs_root *root, int wait)
-{
-	struct btrfs_inode *btrfs_inode;
-	struct inode *inode;
-	struct btrfs_transaction *cur_trans = trans->transaction;
-	struct list_head splice;
-	struct list_head works;
-	struct btrfs_delalloc_work *work, *next;
-	int ret = 0;
-
-	INIT_LIST_HEAD(&splice);
-	INIT_LIST_HEAD(&works);
-
-	mutex_lock(&root->fs_info->ordered_extent_flush_mutex);
-	spin_lock(&root->fs_info->ordered_root_lock);
-	list_splice_init(&cur_trans->ordered_operations, &splice);
-	while (!list_empty(&splice)) {
-		btrfs_inode = list_entry(splice.next, struct btrfs_inode,
-				   ordered_operations);
-		inode = &btrfs_inode->vfs_inode;
-
-		list_del_init(&btrfs_inode->ordered_operations);
-
-		/*
-		 * the inode may be getting freed (in sys_unlink path).
-		 */
-		inode = igrab(inode);
-		if (!inode)
-			continue;
-
-		if (!wait)
-			list_add_tail(&BTRFS_I(inode)->ordered_operations,
-				      &cur_trans->ordered_operations);
-		spin_unlock(&root->fs_info->ordered_root_lock);
-
-		work = btrfs_alloc_delalloc_work(inode, wait, 1);
-		if (!work) {
-			spin_lock(&root->fs_info->ordered_root_lock);
-			if (list_empty(&BTRFS_I(inode)->ordered_operations))
-				list_add_tail(&btrfs_inode->ordered_operations,
-					      &splice);
-			list_splice_tail(&splice,
-					 &cur_trans->ordered_operations);
-			spin_unlock(&root->fs_info->ordered_root_lock);
-			ret = -ENOMEM;
-			goto out;
-		}
-		list_add_tail(&work->list, &works);
-		btrfs_queue_worker(&root->fs_info->flush_workers,
-				   &work->work);
-
-		cond_resched();
-		spin_lock(&root->fs_info->ordered_root_lock);
-	}
-	spin_unlock(&root->fs_info->ordered_root_lock);
-out:
-	list_for_each_entry_safe(work, next, &works, list) {
-		list_del_init(&work->list);
-		btrfs_wait_and_free_delalloc_work(work);
-	}
-	mutex_unlock(&root->fs_info->ordered_extent_flush_mutex);
-	return ret;
+	mutex_unlock(&fs_info->ordered_operations_mutex);
 }
 
 /*
@@ -1076,42 +1034,6 @@ out:
 	return index;
 }
 
-
-/*
- * add a given inode to the list of inodes that must be fully on
- * disk before a transaction commit finishes.
- *
- * This basically gives us the ext3 style data=ordered mode, and it is mostly
- * used to make sure renamed files are fully on disk.
- *
- * It is a noop if the inode is already fully on disk.
- *
- * If trans is not null, we'll do a friendly check for a transaction that
- * is already flushing things and force the IO down ourselves.
- */
-void btrfs_add_ordered_operation(struct btrfs_trans_handle *trans,
-				 struct btrfs_root *root, struct inode *inode)
-{
-	struct btrfs_transaction *cur_trans = trans->transaction;
-	u64 last_mod;
-
-	last_mod = max(BTRFS_I(inode)->generation, BTRFS_I(inode)->last_trans);
-
-	/*
-	 * if this file hasn't been changed since the last transaction
-	 * commit, we can safely return without doing anything
-	 */
-	if (last_mod <= root->fs_info->last_trans_committed)
-		return;
-
-	spin_lock(&root->fs_info->ordered_root_lock);
-	if (list_empty(&BTRFS_I(inode)->ordered_operations)) {
-		list_add_tail(&BTRFS_I(inode)->ordered_operations,
-			      &cur_trans->ordered_operations);
-	}
-	spin_unlock(&root->fs_info->ordered_root_lock);
-}
-
 int __init ordered_data_init(void)
 {
 	btrfs_ordered_extent_cache = kmem_cache_create("btrfs_ordered_extent",