Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel…

…/git/mason/btrfs-unstable

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable:
  Btrfs: avoid delayed metadata items during commits
  btrfs: fix uninitialized return value
  btrfs: fix wrong reservation when doing delayed inode operations
  btrfs: Remove unused sysfs code
  btrfs: fix dereference of ERR_PTR value
  Btrfs: fix relocation races
  Btrfs: set no_trans_join after trying to expand the transaction
  Btrfs: protect the pending_snapshots list with trans_lock
  Btrfs: fix path leakage on subvol deletion
  Btrfs: drop the delalloc_bytes check in shrink_delalloc
  Btrfs: check the return value from set_anon_super

fs/btrfs/ctree.h

@@ -967,6 +967,12 @@ struct btrfs_fs_info {
 	struct srcu_struct subvol_srcu;
 
 	spinlock_t trans_lock;
+	/*
+	 * the reloc mutex goes with the trans lock, it is taken
+	 * during commit to protect us from the relocation code
+	 */
+	struct mutex reloc_mutex;
+
 	struct list_head trans_list;
 	struct list_head hashers;
 	struct list_head dead_roots;
@@ -1172,6 +1178,14 @@ struct btrfs_root {
 	u32 type;
 
 	u64 highest_objectid;
+
+	/* btrfs_record_root_in_trans is a multi-step process,
+	 * and it can race with the balancing code.   But the
+	 * race is very small, and only the first time the root
+	 * is added to each transaction.  So in_trans_setup
+	 * is used to tell us when more checks are required
+	 */
+	unsigned long in_trans_setup;
 	int ref_cows;
 	int track_dirty;
 	int in_radix;
@@ -1181,7 +1195,6 @@ struct btrfs_root {
 	struct btrfs_key defrag_max;
 	int defrag_running;
 	char *name;
-	int in_sysfs;
 
 	/* the dirty list is only used by non-reference counted roots */
 	struct list_head dirty_list;

fs/btrfs/delayed-inode.c

@@ -297,7 +297,6 @@ struct btrfs_delayed_item *btrfs_alloc_delayed_item(u32 data_len)
 		item->data_len = data_len;
 		item->ins_or_del = 0;
 		item->bytes_reserved = 0;
-		item->block_rsv = NULL;
 		item->delayed_node = NULL;
 		atomic_set(&item->refs, 1);
 	}
@@ -593,21 +592,22 @@ static int btrfs_delayed_item_reserve_metadata(struct btrfs_trans_handle *trans,
 
 	num_bytes = btrfs_calc_trans_metadata_size(root, 1);
 	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
-	if (!ret) {
+	if (!ret)
 		item->bytes_reserved = num_bytes;
-		item->block_rsv = dst_rsv;
-	}
 
 	return ret;
 }
 
 static void btrfs_delayed_item_release_metadata(struct btrfs_root *root,
 						struct btrfs_delayed_item *item)
 {
+	struct btrfs_block_rsv *rsv;
+
 	if (!item->bytes_reserved)
 		return;
 
-	btrfs_block_rsv_release(root, item->block_rsv,
+	rsv = &root->fs_info->global_block_rsv;
+	btrfs_block_rsv_release(root, rsv,
 				item->bytes_reserved);
 }
 
@@ -1014,13 +1014,17 @@ int btrfs_run_delayed_items(struct btrfs_trans_handle *trans,
 	struct btrfs_delayed_root *delayed_root;
 	struct btrfs_delayed_node *curr_node, *prev_node;
 	struct btrfs_path *path;
+	struct btrfs_block_rsv *block_rsv;
 	int ret = 0;
 
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
 	path->leave_spinning = 1;
 
+	block_rsv = trans->block_rsv;
+	trans->block_rsv = &root->fs_info->global_block_rsv;
+
 	delayed_root = btrfs_get_delayed_root(root);
 
 	curr_node = btrfs_first_delayed_node(delayed_root);
@@ -1045,27 +1049,33 @@ int btrfs_run_delayed_items(struct btrfs_trans_handle *trans,
 	}
 
 	btrfs_free_path(path);
+	trans->block_rsv = block_rsv;
 	return ret;
 }
 
 static int __btrfs_commit_inode_delayed_items(struct btrfs_trans_handle *trans,
 					      struct btrfs_delayed_node *node)
 {
 	struct btrfs_path *path;
+	struct btrfs_block_rsv *block_rsv;
 	int ret;
 
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
 	path->leave_spinning = 1;
 
+	block_rsv = trans->block_rsv;
+	trans->block_rsv = &node->root->fs_info->global_block_rsv;
+
 	ret = btrfs_insert_delayed_items(trans, path, node->root, node);
 	if (!ret)
 		ret = btrfs_delete_delayed_items(trans, path, node->root, node);
 	if (!ret)
 		ret = btrfs_update_delayed_inode(trans, node->root, path, node);
 	btrfs_free_path(path);
 
+	trans->block_rsv = block_rsv;
 	return ret;
 }
 
@@ -1116,6 +1126,7 @@ static void btrfs_async_run_delayed_node_done(struct btrfs_work *work)
 	struct btrfs_path *path;
 	struct btrfs_delayed_node *delayed_node = NULL;
 	struct btrfs_root *root;
+	struct btrfs_block_rsv *block_rsv;
 	unsigned long nr = 0;
 	int need_requeue = 0;
 	int ret;
@@ -1134,6 +1145,9 @@ static void btrfs_async_run_delayed_node_done(struct btrfs_work *work)
 	if (IS_ERR(trans))
 		goto free_path;
 
+	block_rsv = trans->block_rsv;
+	trans->block_rsv = &root->fs_info->global_block_rsv;
+
 	ret = btrfs_insert_delayed_items(trans, path, root, delayed_node);
 	if (!ret)
 		ret = btrfs_delete_delayed_items(trans, path, root,
@@ -1176,6 +1190,7 @@ static void btrfs_async_run_delayed_node_done(struct btrfs_work *work)
 
 	nr = trans->blocks_used;
 
+	trans->block_rsv = block_rsv;
 	btrfs_end_transaction_dmeta(trans, root);
 	__btrfs_btree_balance_dirty(root, nr);
 free_path:
@@ -1222,6 +1237,13 @@ static int btrfs_wq_run_delayed_node(struct btrfs_delayed_root *delayed_root,
 	return 0;
 }
 
+void btrfs_assert_delayed_root_empty(struct btrfs_root *root)
+{
+	struct btrfs_delayed_root *delayed_root;
+	delayed_root = btrfs_get_delayed_root(root);
+	WARN_ON(btrfs_first_delayed_node(delayed_root));
+}
+
 void btrfs_balance_delayed_items(struct btrfs_root *root)
 {
 	struct btrfs_delayed_root *delayed_root;

fs/btrfs/delayed-inode.h

@@ -75,7 +75,6 @@ struct btrfs_delayed_item {
 	struct list_head tree_list;	/* used for batch insert/delete items */
 	struct list_head readdir_list;	/* used for readdir items */
 	u64 bytes_reserved;
-	struct btrfs_block_rsv *block_rsv;
 	struct btrfs_delayed_node *delayed_node;
 	atomic_t refs;
 	int ins_or_del;
@@ -138,4 +137,8 @@ int btrfs_readdir_delayed_dir_index(struct file *filp, void *dirent,
 /* for init */
 int __init btrfs_delayed_inode_init(void);
 void btrfs_delayed_inode_exit(void);
+
+/* for debugging */
+void btrfs_assert_delayed_root_empty(struct btrfs_root *root);
+
 #endif

fs/btrfs/disk-io.c

@@ -1044,7 +1044,6 @@ static int __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
 	root->last_trans = 0;
 	root->highest_objectid = 0;
 	root->name = NULL;
-	root->in_sysfs = 0;
 	root->inode_tree = RB_ROOT;
 	INIT_RADIX_TREE(&root->delayed_nodes_tree, GFP_ATOMIC);
 	root->block_rsv = NULL;
@@ -1300,19 +1299,21 @@ struct btrfs_root *btrfs_read_fs_root_no_name(struct btrfs_fs_info *fs_info,
 		return root;
 
 	root->free_ino_ctl = kzalloc(sizeof(*root->free_ino_ctl), GFP_NOFS);
-	if (!root->free_ino_ctl)
-		goto fail;
 	root->free_ino_pinned = kzalloc(sizeof(*root->free_ino_pinned),
 					GFP_NOFS);
-	if (!root->free_ino_pinned)
+	if (!root->free_ino_pinned || !root->free_ino_ctl) {
+		ret = -ENOMEM;
 		goto fail;
+	}
 
 	btrfs_init_free_ino_ctl(root);
 	mutex_init(&root->fs_commit_mutex);
 	spin_lock_init(&root->cache_lock);
 	init_waitqueue_head(&root->cache_wait);
 
-	set_anon_super(&root->anon_super, NULL);
+	ret = set_anon_super(&root->anon_super, NULL);
+	if (ret)
+		goto fail;
 
 	if (btrfs_root_refs(&root->root_item) == 0) {
 		ret = -ENOENT;
@@ -1618,6 +1619,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	spin_lock_init(&fs_info->fs_roots_radix_lock);
 	spin_lock_init(&fs_info->delayed_iput_lock);
 	spin_lock_init(&fs_info->defrag_inodes_lock);
+	mutex_init(&fs_info->reloc_mutex);
 
 	init_completion(&fs_info->kobj_unregister);
 	fs_info->tree_root = tree_root;

fs/btrfs/extent-tree.c

@@ -3314,10 +3314,6 @@ static int shrink_delalloc(struct btrfs_trans_handle *trans,
 	if (reserved == 0)
 		return 0;
 
-	/* nothing to shrink - nothing to reclaim */
-	if (root->fs_info->delalloc_bytes == 0)
-		return 0;
-
 	max_reclaim = min(reserved, to_reclaim);
 
 	while (loops < 1024) {

fs/btrfs/inode.c

@@ -3076,6 +3076,7 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
 	ret = btrfs_update_inode(trans, root, dir);
 	BUG_ON(ret);
 
+	btrfs_free_path(path);
 	return 0;
 }
 

fs/btrfs/ioctl.c

@@ -482,8 +482,10 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
 	ret = btrfs_snap_reserve_metadata(trans, pending_snapshot);
 	BUG_ON(ret);
 
+	spin_lock(&root->fs_info->trans_lock);
 	list_add(&pending_snapshot->list,
 		 &trans->transaction->pending_snapshots);
+	spin_unlock(&root->fs_info->trans_lock);
 	if (async_transid) {
 		*async_transid = trans->transid;
 		ret = btrfs_commit_transaction_async(trans,

fs/btrfs/relocation.c

@@ -1368,7 +1368,7 @@ int btrfs_update_reloc_root(struct btrfs_trans_handle *trans,
 	int ret;
 
 	if (!root->reloc_root)
-		return 0;
+		goto out;
 
 	reloc_root = root->reloc_root;
 	root_item = &reloc_root->root_item;
@@ -1390,6 +1390,8 @@ int btrfs_update_reloc_root(struct btrfs_trans_handle *trans,
 	ret = btrfs_update_root(trans, root->fs_info->tree_root,
 				&reloc_root->root_key, root_item);
 	BUG_ON(ret);
+
+out:
 	return 0;
 }
 
@@ -2142,10 +2144,11 @@ int prepare_to_merge(struct reloc_control *rc, int err)
 	u64 num_bytes = 0;
 	int ret;
 
-	spin_lock(&root->fs_info->trans_lock);
+	mutex_lock(&root->fs_info->reloc_mutex);
 	rc->merging_rsv_size += root->nodesize * (BTRFS_MAX_LEVEL - 1) * 2;
 	rc->merging_rsv_size += rc->nodes_relocated * 2;
-	spin_unlock(&root->fs_info->trans_lock);
+	mutex_unlock(&root->fs_info->reloc_mutex);
+
 again:
 	if (!err) {
 		num_bytes = rc->merging_rsv_size;
@@ -2214,9 +2217,16 @@ int merge_reloc_roots(struct reloc_control *rc)
 	int ret;
 again:
 	root = rc->extent_root;
-	spin_lock(&root->fs_info->trans_lock);
+
+	/*
+	 * this serializes us with btrfs_record_root_in_transaction,
+	 * we have to make sure nobody is in the middle of
+	 * adding their roots to the list while we are
+	 * doing this splice
+	 */
+	mutex_lock(&root->fs_info->reloc_mutex);
 	list_splice_init(&rc->reloc_roots, &reloc_roots);
-	spin_unlock(&root->fs_info->trans_lock);
+	mutex_unlock(&root->fs_info->reloc_mutex);
 
 	while (!list_empty(&reloc_roots)) {
 		found = 1;
@@ -3590,17 +3600,19 @@ int find_next_extent(struct btrfs_trans_handle *trans,
 static void set_reloc_control(struct reloc_control *rc)
 {
 	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
-	spin_lock(&fs_info->trans_lock);
+
+	mutex_lock(&fs_info->reloc_mutex);
 	fs_info->reloc_ctl = rc;
-	spin_unlock(&fs_info->trans_lock);
+	mutex_unlock(&fs_info->reloc_mutex);
 }
 
 static void unset_reloc_control(struct reloc_control *rc)
 {
 	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
-	spin_lock(&fs_info->trans_lock);
+
+	mutex_lock(&fs_info->reloc_mutex);
 	fs_info->reloc_ctl = NULL;
-	spin_unlock(&fs_info->trans_lock);
+	mutex_unlock(&fs_info->reloc_mutex);
 }
 
 static int check_extent_flags(u64 flags)