[PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC

linux-btrfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed

* [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
@ 2011-03-31  4:00 Larry D'Anna
  2011-03-31  4:00 ` [PATCH 1/2] btrfs: remove unused argument 'root' from btrfs_release_path Larry D'Anna
                   ` (2 more replies)
  0 siblings, 3 replies; 19+ messages in thread
From: Larry D'Anna @ 2011-03-31  4:00 UTC (permalink / raw)
  To: linux-btrfs; +Cc: larry, Yan Zheng

This is a simple patch to allow reflinks to be made crossing subvolume
boundaries.  The only complication I found in implementing this is that
btrfs_ioctl_clone reuses a btrfs_path.  This is a slight problem because once we
allow the source and destination inodes to come from different btrfs_roots, we
need to know what root to pass into the various calls to btrfs_release_path, and
keeping track of this would further complicate the control flow of
btrfs_ioctl_clone.  Fortunately btrfs_release_path does not use it's first
argument, patch number 1 removes it.

Patch number 2 is then just a simple matter of distinguishing between the source
and destination btrfs_roots in btrfs_ioctl_clone.

If removing the first argument from btrfs_release_path is undesirable, then
btrfs_ioctl_clone could simply pass NULL.  There are a few other places in btrfs
that already do this.

I have tested this by cp -a --reflink=always all the files found in the UML
slackware root_fs, and also by cloning a test file with hundreds of
random extents and holes.  

Larry D'Anna (2):
  btrfs: remove unused argument 'root' from btrfs_release_path
  btrfs: allow cross-subvolume BTRFS_IOC_CLONE

 fs/btrfs/ctree.c            |   28 ++++++------
 fs/btrfs/ctree.h            |    2 +-
 fs/btrfs/dir-item.c         |    2 +-
 fs/btrfs/extent-tree.c      |   74 ++++++++++++++++----------------
 fs/btrfs/file-item.c        |   12 +++---
 fs/btrfs/file.c             |   12 +++---
 fs/btrfs/free-space-cache.c |   14 +++---
 fs/btrfs/inode.c            |   42 +++++++++---------
 fs/btrfs/ioctl.c            |   22 +++++----
 fs/btrfs/relocation.c       |   30 +++++++-------
 fs/btrfs/root-tree.c        |   10 ++--
 fs/btrfs/tree-defrag.c      |    2 +-
 fs/btrfs/tree-log.c         |   98 +++++++++++++++++++++---------------------
 fs/btrfs/volumes.c          |   16 ++++----
 fs/btrfs/xattr.c            |    4 +-
 15 files changed, 185 insertions(+), 183 deletions(-)

^ permalink raw reply	[flat|nested] 19+ messages in thread

* [PATCH 1/2] btrfs: remove unused argument 'root' from btrfs_release_path
  2011-03-31  4:00 [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE Larry D'Anna
@ 2011-03-31  4:00 ` Larry D'Anna
  2011-03-31  4:00 ` [PATCH 2/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE Larry D'Anna
  2011-03-31  6:36 ` [PATCH 0/2] " Christoph Hellwig
  2 siblings, 0 replies; 19+ messages in thread
From: Larry D'Anna @ 2011-03-31  4:00 UTC (permalink / raw)
  To: linux-btrfs; +Cc: larry, Yan Zheng


Signed-off-by: Larry D'Anna <larry@elder-gods.org>
---
 fs/btrfs/ctree.c            |   28 ++++++------
 fs/btrfs/ctree.h            |    2 +-
 fs/btrfs/dir-item.c         |    2 +-
 fs/btrfs/extent-tree.c      |   74 ++++++++++++++++----------------
 fs/btrfs/file-item.c        |   12 +++---
 fs/btrfs/file.c             |   12 +++---
 fs/btrfs/free-space-cache.c |   14 +++---
 fs/btrfs/inode.c            |   42 +++++++++---------
 fs/btrfs/ioctl.c            |   12 +++---
 fs/btrfs/relocation.c       |   30 +++++++-------
 fs/btrfs/root-tree.c        |   10 ++--
 fs/btrfs/tree-defrag.c      |    2 +-
 fs/btrfs/tree-log.c         |   98 +++++++++++++++++++++---------------------
 fs/btrfs/volumes.c          |   16 ++++----
 fs/btrfs/xattr.c            |    4 +-
 15 files changed, 179 insertions(+), 179 deletions(-)

diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 84d7ca1..70211e8 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -107,7 +107,7 @@ void btrfs_free_path(struct btrfs_path *p)
 {
 	if (!p)
 		return;
-	btrfs_release_path(NULL, p);
+	btrfs_release_path(p);
 	kmem_cache_free(btrfs_path_cachep, p);
 }
 
@@ -117,7 +117,7 @@ void btrfs_free_path(struct btrfs_path *p)
  *
  * It is safe to call this on paths that no locks or extent buffers held.
  */
-noinline void btrfs_release_path(struct btrfs_root *root, struct btrfs_path *p)
+noinline void btrfs_release_path(struct btrfs_path *p)
 {
 	int i;
 
@@ -1328,7 +1328,7 @@ static noinline int reada_for_balance(struct btrfs_root *root,
 		ret = -EAGAIN;
 
 		/* release the whole path */
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 
 		/* read the blocks */
 		if (block1)
@@ -1475,7 +1475,7 @@ read_block_for_search(struct btrfs_trans_handle *trans,
 				return 0;
 			}
 			free_extent_buffer(tmp);
-			btrfs_release_path(NULL, p);
+			btrfs_release_path(p);
 			return -EIO;
 		}
 	}
@@ -1494,7 +1494,7 @@ read_block_for_search(struct btrfs_trans_handle *trans,
 	if (p->reada)
 		reada_for_search(root, p, level, slot, key->objectid);
 
-	btrfs_release_path(NULL, p);
+	btrfs_release_path(p);
 
 	ret = -EAGAIN;
 	tmp = read_tree_block(root, blocknr, blocksize, 0);
@@ -1563,7 +1563,7 @@ setup_nodes_for_search(struct btrfs_trans_handle *trans,
 		}
 		b = p->nodes[level];
 		if (!b) {
-			btrfs_release_path(NULL, p);
+			btrfs_release_path(p);
 			goto again;
 		}
 		BUG_ON(btrfs_header_nritems(b) == 1);
@@ -1753,7 +1753,7 @@ done:
 	if (!p->leave_spinning)
 		btrfs_set_path_blocking(p);
 	if (ret < 0)
-		btrfs_release_path(root, p);
+		btrfs_release_path(p);
 	return ret;
 }
 
@@ -3026,7 +3026,7 @@ static noinline int setup_leaf_for_split(struct btrfs_trans_handle *trans,
 				    struct btrfs_file_extent_item);
 		extent_len = btrfs_file_extent_num_bytes(leaf, fi);
 	}
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	path->keep_locks = 1;
 	path->search_for_split = 1;
@@ -3949,7 +3949,7 @@ int btrfs_prev_leaf(struct btrfs_root *root, struct btrfs_path *path)
 	else
 		return 1;
 
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
 	if (ret < 0)
 		return ret;
@@ -4073,7 +4073,7 @@ find_next_key:
 			sret = btrfs_find_next_key(root, path, min_key, level,
 						  cache_only, min_trans);
 			if (sret == 0) {
-				btrfs_release_path(root, path);
+				btrfs_release_path(path);
 				goto again;
 			} else {
 				goto out;
@@ -4152,7 +4152,7 @@ next:
 				btrfs_node_key_to_cpu(c, &cur_key, slot);
 
 			orig_lowest = path->lowest_level;
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			path->lowest_level = level;
 			ret = btrfs_search_slot(NULL, root, &cur_key, path,
 						0, 0);
@@ -4229,7 +4229,7 @@ int btrfs_next_leaf(struct btrfs_root *root, struct btrfs_path *path)
 again:
 	level = 1;
 	next = NULL;
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	path->keep_locks = 1;
 
@@ -4285,7 +4285,7 @@ again:
 			goto again;
 
 		if (ret < 0) {
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			goto done;
 		}
 
@@ -4324,7 +4324,7 @@ again:
 			goto again;
 
 		if (ret < 0) {
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			goto done;
 		}
 
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index d47ce83..5270a37 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -2286,7 +2286,7 @@ int btrfs_realloc_node(struct btrfs_trans_handle *trans,
 		       struct btrfs_root *root, struct extent_buffer *parent,
 		       int start_slot, int cache_only, u64 *last_ret,
 		       struct btrfs_key *progress);
-void btrfs_release_path(struct btrfs_root *root, struct btrfs_path *p);
+void btrfs_release_path(struct btrfs_path *p);
 struct btrfs_path *btrfs_alloc_path(void);
 void btrfs_free_path(struct btrfs_path *p);
 void btrfs_set_path_blocking(struct btrfs_path *p);
diff --git a/fs/btrfs/dir-item.c b/fs/btrfs/dir-item.c
index c62f02f..ab8afed 100644
--- a/fs/btrfs/dir-item.c
+++ b/fs/btrfs/dir-item.c
@@ -172,7 +172,7 @@ second_insert:
 		ret = 0;
 		goto out_free;
 	}
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	btrfs_set_key_type(&key, BTRFS_DIR_INDEX_KEY);
 	key.offset = index;
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index f619c3c..e524a41 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -360,7 +360,7 @@ again:
 				break;
 
 			caching_ctl->progress = last;
-			btrfs_release_path(extent_root, path);
+			btrfs_release_path(path);
 			up_read(&fs_info->extent_commit_sem);
 			mutex_unlock(&caching_ctl->mutex);
 			if (btrfs_transaction_in_commit(fs_info))
@@ -735,7 +735,7 @@ again:
 			atomic_inc(&head->node.refs);
 			spin_unlock(&delayed_refs->lock);
 
-			btrfs_release_path(root->fs_info->extent_root, path);
+			btrfs_release_path(path);
 
 			mutex_lock(&head->mutex);
 			mutex_unlock(&head->mutex);
@@ -915,7 +915,7 @@ static int convert_extent_item_v0(struct btrfs_trans_handle *trans,
 			break;
 		}
 	}
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	if (owner < BTRFS_FIRST_FREE_OBJECTID)
 		new_size += sizeof(*bi);
@@ -1023,7 +1023,7 @@ again:
 			return 0;
 #ifdef BTRFS_COMPAT_EXTENT_TREE_V0
 		key.type = BTRFS_EXTENT_REF_V0_KEY;
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
 		if (ret < 0) {
 			err = ret;
@@ -1061,7 +1061,7 @@ again:
 		if (match_extent_data_ref(leaf, ref, root_objectid,
 					  owner, offset)) {
 			if (recow) {
-				btrfs_release_path(root, path);
+				btrfs_release_path(path);
 				goto again;
 			}
 			err = 0;
@@ -1122,7 +1122,7 @@ static noinline int insert_extent_data_ref(struct btrfs_trans_handle *trans,
 			if (match_extent_data_ref(leaf, ref, root_objectid,
 						  owner, offset))
 				break;
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			key.offset++;
 			ret = btrfs_insert_empty_item(trans, root, path, &key,
 						      size);
@@ -1148,7 +1148,7 @@ static noinline int insert_extent_data_ref(struct btrfs_trans_handle *trans,
 	btrfs_mark_buffer_dirty(leaf);
 	ret = 0;
 fail:
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	return ret;
 }
 
@@ -1274,7 +1274,7 @@ static noinline int lookup_tree_block_ref(struct btrfs_trans_handle *trans,
 		ret = -ENOENT;
 #ifdef BTRFS_COMPAT_EXTENT_TREE_V0
 	if (ret == -ENOENT && parent) {
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		key.type = BTRFS_EXTENT_REF_V0_KEY;
 		ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
 		if (ret > 0)
@@ -1303,7 +1303,7 @@ static noinline int insert_tree_block_ref(struct btrfs_trans_handle *trans,
 	}
 
 	ret = btrfs_insert_empty_item(trans, root, path, &key, 0);
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	return ret;
 }
 
@@ -1589,7 +1589,7 @@ static int lookup_extent_backref(struct btrfs_trans_handle *trans,
 	if (ret != -ENOENT)
 		return ret;
 
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	*ref_ret = NULL;
 
 	if (owner < BTRFS_FIRST_FREE_OBJECTID) {
@@ -1843,7 +1843,7 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
 		__run_delayed_extent_op(extent_op, leaf, item);
 
 	btrfs_mark_buffer_dirty(leaf);
-	btrfs_release_path(root->fs_info->extent_root, path);
+	btrfs_release_path(path);
 
 	path->reada = 1;
 	path->leave_spinning = 1;
@@ -2342,7 +2342,7 @@ static noinline int check_delayed_ref(struct btrfs_trans_handle *trans,
 		atomic_inc(&head->node.refs);
 		spin_unlock(&delayed_refs->lock);
 
-		btrfs_release_path(root->fs_info->extent_root, path);
+		btrfs_release_path(path);
 
 		mutex_lock(&head->mutex);
 		mutex_unlock(&head->mutex);
@@ -2713,7 +2713,7 @@ static int write_one_cache_group(struct btrfs_trans_handle *trans,
 	bi = btrfs_item_ptr_offset(leaf, path->slots[0]);
 	write_extent_buffer(leaf, &cache->item, bi, sizeof(cache->item));
 	btrfs_mark_buffer_dirty(leaf);
-	btrfs_release_path(extent_root, path);
+	btrfs_release_path(path);
 fail:
 	if (ret)
 		return ret;
@@ -2766,7 +2766,7 @@ again:
 	inode = lookup_free_space_inode(root, block_group, path);
 	if (IS_ERR(inode) && PTR_ERR(inode) != -ENOENT) {
 		ret = PTR_ERR(inode);
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		goto out;
 	}
 
@@ -2835,7 +2835,7 @@ again:
 out_put:
 	iput(inode);
 out_free:
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 out:
 	spin_lock(&block_group->lock);
 	block_group->disk_cache_state = dcs;
@@ -4476,7 +4476,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 						    NULL, refs_to_drop,
 						    is_data);
 			BUG_ON(ret);
-			btrfs_release_path(extent_root, path);
+			btrfs_release_path(path);
 			path->leave_spinning = 1;
 
 			key.objectid = bytenr;
@@ -4515,7 +4515,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 					     owner_objectid, 0);
 		BUG_ON(ret < 0);
 
-		btrfs_release_path(extent_root, path);
+		btrfs_release_path(path);
 		path->leave_spinning = 1;
 
 		key.objectid = bytenr;
@@ -4585,7 +4585,7 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans,
 		ret = btrfs_del_items(trans, extent_root, path, path->slots[0],
 				      num_to_del);
 		BUG_ON(ret);
-		btrfs_release_path(extent_root, path);
+		btrfs_release_path(path);
 
 		if (is_data) {
 			ret = btrfs_del_csums(trans, root, bytenr, num_bytes);
@@ -6412,7 +6412,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root,
 				trans->block_rsv = block_rsv;
 		}
 	}
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	BUG_ON(err);
 
 	ret = btrfs_del_root(trans, tree_root, &root->root_key);
@@ -6764,7 +6764,7 @@ walk_down:
 		}
 next:
 		level--;
-		btrfs_release_path(extent_root, path);
+		btrfs_release_path(path);
 		cond_resched();
 	}
 	/* reached lowest level */
@@ -6800,7 +6800,7 @@ walk_up:
 				/* the extent was freed by someone */
 				if (ref_path->lowest_level == level)
 					goto out;
-				btrfs_release_path(extent_root, path);
+				btrfs_release_path(path);
 				goto walk_down;
 			}
 			leaf = path->nodes[0];
@@ -6814,7 +6814,7 @@ walk_up:
 				ret = 1;
 				goto out;
 			}
-			btrfs_release_path(extent_root, path);
+			btrfs_release_path(path);
 			goto walk_down;
 		}
 found:
@@ -6876,7 +6876,7 @@ found:
 			goto out;
 		}
 
-		btrfs_release_path(extent_root, path);
+		btrfs_release_path(path);
 		cond_resched();
 	}
 	/* reached max tree level, but no tree root found. */
@@ -7112,7 +7112,7 @@ next:
 
 		if (inode && key.objectid != inode->i_ino) {
 			BUG_ON(extent_locked);
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			mutex_unlock(&inode->i_mutex);
 			iput(inode);
 			inode = NULL;
@@ -7157,7 +7157,7 @@ next:
 		}
 
 		if (!inode) {
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 
 			inode = btrfs_iget_locked(root->fs_info->sb,
 						  key.objectid, root);
@@ -7188,7 +7188,7 @@ next:
 		if (!extent_locked) {
 			struct btrfs_ordered_extent *ordered;
 
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 
 			lock_extent(&BTRFS_I(inode)->io_tree, lock_start,
 				    lock_end, GFP_NOFS);
@@ -7240,7 +7240,7 @@ next:
 						key.objectid, 0);
 			BUG_ON(ret);
 
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			key.offset += num_bytes;
 		} else {
 			BUG_ON(1);
@@ -7252,7 +7252,7 @@ next:
 			 * drop old extent pointer at first, then insert the
 			 * new pointers one bye one
 			 */
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			ret = btrfs_drop_extents(trans, root, inode, key.offset,
 						 key.offset + num_bytes,
 						 key.offset, &alloc_hint);
@@ -7309,7 +7309,7 @@ next:
 						root->root_key.objectid,
 						trans->transid, key.objectid);
 				BUG_ON(ret);
-				btrfs_release_path(root, path);
+				btrfs_release_path(path);
 
 				inode_add_bytes(inode, extent_len);
 
@@ -7338,7 +7338,7 @@ skip:
 	}
 	ret = 0;
 out:
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	if (inode) {
 		mutex_unlock(&inode->i_mutex);
 		if (extent_locked) {
@@ -7749,7 +7749,7 @@ static noinline int relocate_one_path(struct btrfs_trans_handle *trans,
 		ret = btrfs_search_slot(trans, root, first_key, path, 0, 1);
 		BUG_ON(ret < 0);
 		path->lowest_level = 0;
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		return 0;
 	}
 
@@ -7790,7 +7790,7 @@ static noinline int relocate_one_path(struct btrfs_trans_handle *trans,
 						      group, reloc_inode);
 			BUG_ON(ret);
 		}
-		btrfs_release_path(reloc_root, path);
+		btrfs_release_path(path);
 	} else {
 		ret = btrfs_merge_path(trans, reloc_root, keys, nodes,
 				       lowest_level);
@@ -7810,7 +7810,7 @@ static noinline int relocate_one_path(struct btrfs_trans_handle *trans,
 		BUG_ON(ret);
 		extent_buffer_get(path->nodes[0]);
 		eb = path->nodes[0];
-		btrfs_release_path(reloc_root, path);
+		btrfs_release_path(path);
 		ret = invalidate_extent_cache(reloc_root, eb, group, root);
 		BUG_ON(ret);
 		free_extent_buffer(eb);
@@ -7848,7 +7848,7 @@ static noinline int del_extent_zero(struct btrfs_trans_handle *trans,
 		goto out;
 	ret = btrfs_del_item(trans, extent_root, path);
 out:
-	btrfs_release_path(extent_root, path);
+	btrfs_release_path(path);
 	return ret;
 }
 
@@ -8505,7 +8505,7 @@ int btrfs_read_block_groups(struct btrfs_root *root)
 		memcpy(&cache->key, &found_key, sizeof(found_key));
 
 		key.objectid = found_key.objectid + found_key.offset;
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		cache->flags = btrfs_block_group_flags(&cache->item);
 		cache->sectorsize = root->sectorsize;
 
@@ -8727,12 +8727,12 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans,
 	if (ret < 0)
 		goto out;
 	if (ret > 0)
-		btrfs_release_path(tree_root, path);
+		btrfs_release_path(path);
 	if (ret == 0) {
 		ret = btrfs_del_item(trans, tree_root, path);
 		if (ret)
 			goto out;
-		btrfs_release_path(tree_root, path);
+		btrfs_release_path(path);
 	}
 
 	spin_lock(&root->fs_info->block_group_cache_lock);
diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c
index a6a9d4e..f47e43d 100644
--- a/fs/btrfs/file-item.c
+++ b/fs/btrfs/file-item.c
@@ -193,7 +193,7 @@ static int __btrfs_lookup_bio_sums(struct btrfs_root *root,
 			u32 item_size;
 
 			if (item)
-				btrfs_release_path(root, path);
+				btrfs_release_path(path);
 			item = btrfs_lookup_csum(NULL, root->fs_info->csum_root,
 						 path, disk_bytenr, 0);
 			if (IS_ERR(item)) {
@@ -213,7 +213,7 @@ static int __btrfs_lookup_bio_sums(struct btrfs_root *root,
 					       (unsigned long long)offset);
 				}
 				item = NULL;
-				btrfs_release_path(root, path);
+				btrfs_release_path(path);
 				goto found;
 			}
 			btrfs_item_key_to_cpu(path->nodes[0], &found_key,
@@ -631,7 +631,7 @@ int btrfs_del_csums(struct btrfs_trans_handle *trans,
 			if (key.offset < bytenr)
 				break;
 		}
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 	}
 out:
 	btrfs_free_path(path);
@@ -722,7 +722,7 @@ again:
 	 * at this point, we know the tree has an item, but it isn't big
 	 * enough yet to put our csum in.  Grow it
 	 */
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	ret = btrfs_search_slot(trans, root, &file_key, path,
 				csum_size, 1);
 	if (ret < 0)
@@ -766,7 +766,7 @@ again:
 	}
 
 insert:
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	csum_offset = 0;
 	if (found_next) {
 		u64 tmp = total_bytes + root->sectorsize;
@@ -850,7 +850,7 @@ next_sector:
 	}
 	btrfs_mark_buffer_dirty(path->nodes[0]);
 	if (total_bytes < sums->len) {
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		cond_resched();
 		goto again;
 	}
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 656bc0a..9b2f647 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -379,7 +379,7 @@ next_slot:
 
 		search_start = max(key.offset, start);
 		if (recow) {
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			continue;
 		}
 
@@ -396,7 +396,7 @@ next_slot:
 			ret = btrfs_duplicate_item(trans, root, path,
 						   &new_key);
 			if (ret == -EAGAIN) {
-				btrfs_release_path(root, path);
+				btrfs_release_path(path);
 				continue;
 			}
 			if (ret < 0)
@@ -519,7 +519,7 @@ next_slot:
 			del_nr = 0;
 			del_slot = 0;
 
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			continue;
 		}
 
@@ -684,7 +684,7 @@ again:
 		new_key.offset = split;
 		ret = btrfs_duplicate_item(trans, root, path, &new_key);
 		if (ret == -EAGAIN) {
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			goto again;
 		}
 		BUG_ON(ret < 0);
@@ -724,7 +724,7 @@ again:
 			     inode->i_ino, bytenr, orig_offset,
 			     &other_start, &other_end)) {
 		if (recow) {
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			goto again;
 		}
 		extent_end = other_end;
@@ -741,7 +741,7 @@ again:
 			     inode->i_ino, bytenr, orig_offset,
 			     &other_start, &other_end)) {
 		if (recow) {
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			goto again;
 		}
 		key.offset = other_start;
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index 0037427..e684f9f 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -60,7 +60,7 @@ struct inode *lookup_free_space_inode(struct btrfs_root *root,
 	if (ret < 0)
 		return ERR_PTR(ret);
 	if (ret > 0) {
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		return ERR_PTR(-ENOENT);
 	}
 
@@ -69,7 +69,7 @@ struct inode *lookup_free_space_inode(struct btrfs_root *root,
 				struct btrfs_free_space_header);
 	btrfs_free_space_key(leaf, header, &disk_key);
 	btrfs_disk_key_to_cpu(&location, &disk_key);
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	inode = btrfs_iget(root->fs_info->sb, &location, root, NULL);
 	if (!inode)
@@ -131,7 +131,7 @@ int create_free_space_inode(struct btrfs_root *root,
 	btrfs_set_inode_block_group(leaf, inode_item,
 				    block_group->key.objectid);
 	btrfs_mark_buffer_dirty(leaf);
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	key.objectid = BTRFS_FREE_SPACE_OBJECTID;
 	key.offset = block_group->key.objectid;
@@ -140,7 +140,7 @@ int create_free_space_inode(struct btrfs_root *root,
 	ret = btrfs_insert_empty_item(trans, root, path, &key,
 				      sizeof(struct btrfs_free_space_header));
 	if (ret < 0) {
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		return ret;
 	}
 	leaf = path->nodes[0];
@@ -149,7 +149,7 @@ int create_free_space_inode(struct btrfs_root *root,
 	memset_extent_buffer(leaf, 0, (unsigned long)header, sizeof(*header));
 	btrfs_set_free_space_key(leaf, header, &disk_key);
 	btrfs_mark_buffer_dirty(leaf);
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	return 0;
 }
@@ -754,7 +754,7 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 					 EXTENT_DIRTY | EXTENT_DELALLOC |
 					 EXTENT_DO_ACCOUNTING, 0, 0, NULL,
 					 GFP_NOFS);
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			goto out_free;
 		}
 	}
@@ -764,7 +764,7 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 	btrfs_set_free_space_bitmaps(leaf, header, bitmaps);
 	btrfs_set_free_space_generation(leaf, header, trans->transid);
 	btrfs_mark_buffer_dirty(leaf);
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	ret = 1;
 
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 93c28a1..dc204d1 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -1164,7 +1164,7 @@ out_check:
 			goto next_slot;
 		}
 
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		if (cow_start != (u64)-1) {
 			ret = cow_file_range(inode, locked_page, cow_start,
 					found_key.offset - 1, page_started,
@@ -1222,7 +1222,7 @@ out_check:
 		if (cur_offset > end)
 			break;
 	}
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	if (cur_offset <= end && cow_start == (u64)-1)
 		cow_start = cur_offset;
@@ -2345,7 +2345,7 @@ int btrfs_orphan_cleanup(struct btrfs_root *root)
 			break;
 
 		/* release the path since we're done with it */
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 
 		/*
 		 * this is where we are basically btrfs_lookup, without the
@@ -2701,7 +2701,7 @@ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans,
 	ret = btrfs_delete_one_dir_name(trans, root, path, di);
 	if (ret)
 		goto err;
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	ret = btrfs_del_inode_ref(trans, root, name, name_len,
 				  inode->i_ino,
@@ -2724,7 +2724,7 @@ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans,
 		goto err;
 	}
 	ret = btrfs_delete_one_dir_name(trans, root, path, di);
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	ret = btrfs_del_inode_ref_in_log(trans, root, name, name_len,
 					 inode, dir->i_ino);
@@ -2851,7 +2851,7 @@ static struct btrfs_trans_handle *__unlink_start_trans(struct inode *dir,
 	} else {
 		check_link = 0;
 	}
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	ret = btrfs_lookup_inode(trans, root, path,
 				&BTRFS_I(inode)->location, 0);
@@ -2865,7 +2865,7 @@ static struct btrfs_trans_handle *__unlink_start_trans(struct inode *dir,
 	} else {
 		check_link = 0;
 	}
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	if (ret == 0 && S_ISREG(inode->i_mode)) {
 		ret = btrfs_lookup_file_extent(trans, root, path,
@@ -2877,7 +2877,7 @@ static struct btrfs_trans_handle *__unlink_start_trans(struct inode *dir,
 		BUG_ON(ret == 0);
 		if (check_path_shared(root, path))
 			goto out;
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 	}
 
 	if (!check_link) {
@@ -2898,7 +2898,7 @@ static struct btrfs_trans_handle *__unlink_start_trans(struct inode *dir,
 		err = 0;
 		goto out;
 	}
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	ref = btrfs_lookup_inode_ref(trans, root, path,
 				dentry->d_name.name, dentry->d_name.len,
@@ -2911,7 +2911,7 @@ static struct btrfs_trans_handle *__unlink_start_trans(struct inode *dir,
 	if (check_path_shared(root, path))
 		goto out;
 	index = btrfs_inode_ref_index(path->nodes[0], ref);
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	di = btrfs_lookup_dir_index_item(trans, root, path, dir->i_ino, index,
 				dentry->d_name.name, dentry->d_name.len, 0);
@@ -3002,7 +3002,7 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
 	WARN_ON(key.type != BTRFS_ROOT_ITEM_KEY || key.objectid != objectid);
 	ret = btrfs_delete_one_dir_name(trans, root, path, di);
 	BUG_ON(ret);
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	ret = btrfs_del_root_ref(trans, root->fs_info->tree_root,
 				 objectid, root->root_key.objectid,
@@ -3015,7 +3015,7 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
 
 		leaf = path->nodes[0];
 		btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		index = key.offset;
 	}
 
@@ -3028,7 +3028,7 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans,
 	WARN_ON(key.type != BTRFS_ROOT_ITEM_KEY || key.objectid != objectid);
 	ret = btrfs_delete_one_dir_name(trans, root, path, di);
 	BUG_ON(ret);
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	btrfs_i_size_write(dir, dir->i_size - name_len * 2);
 	dir->i_mtime = dir->i_ctime = CURRENT_TIME;
@@ -3164,7 +3164,7 @@ next_node:
 		 * we could modify, but this drop is just an optimization
 		 * and is allowed to miss some leaves.
 		 */
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		found_key.offset++;
 
 		/* setup a max key for search_forward */
@@ -3183,7 +3183,7 @@ next_node:
 		}
 
 		key.offset = found_key.offset;
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		cond_resched();
 		goto again;
 	}
@@ -3235,7 +3235,7 @@ next_node:
 		}
 	}
 next_key:
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	if (other_key.objectid == inode->i_ino &&
 	    other_key.type == key.type && other_key.offset > key.offset) {
@@ -3248,7 +3248,7 @@ out:
 	/* fixup any changes we've made to the path */
 	path->lowest_level = 0;
 	path->keep_locks = 0;
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	return ret;
 }
 
@@ -3466,7 +3466,7 @@ delete:
 				BUG_ON(ret);
 				pending_del_nr = 0;
 			}
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			goto search_again;
 		} else {
 			path->slots[0]--;
@@ -3888,7 +3888,7 @@ static int fixup_tree_root_location(struct btrfs_root *root,
 	if (ret)
 		goto out;
 
-	btrfs_release_path(root->fs_info->tree_root, path);
+	btrfs_release_path(path);
 
 	new_root = btrfs_read_fs_root_no_name(root->fs_info, location);
 	if (IS_ERR(new_root)) {
@@ -5213,7 +5213,7 @@ again:
 				kunmap(page);
 				free_extent_map(em);
 				em = NULL;
-				btrfs_release_path(root, path);
+				btrfs_release_path(path);
 				trans = btrfs_join_transaction(root, 1);
 				if (IS_ERR(trans))
 					return ERR_CAST(trans);
@@ -5239,7 +5239,7 @@ not_found_em:
 	em->block_start = EXTENT_MAP_HOLE;
 	set_bit(EXTENT_FLAG_VACANCY, &em->flags);
 insert:
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	if (em->start > start || extent_map_end(em) <= start) {
 		printk(KERN_ERR "Btrfs: bad extent! em: [%llu %llu] passed "
 		       "[%llu %llu]\n", (unsigned long long)em->start,
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 7c07fe2..f9717b6 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -1392,7 +1392,7 @@ static noinline int search_ioctl(struct inode *inode,
 		}
 		ret = copy_to_sk(root, path, &key, sk, args->buf,
 				 &sk_offset, &num_found);
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		if (ret || num_found >= sk->nr_items)
 			break;
 
@@ -1499,7 +1499,7 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info,
 		if (key.offset == BTRFS_FIRST_FREE_OBJECTID)
 			break;
 
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		key.objectid = key.offset;
 		key.offset = (u64)-1;
 		dirid = key.objectid;
@@ -1978,7 +1978,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
 				datal = btrfs_file_extent_ram_bytes(leaf,
 								    extent);
 			}
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 
 			if (key.offset + datal <= off ||
 			    key.offset >= off+len)
@@ -2088,7 +2088,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
 			}
 
 			btrfs_mark_buffer_dirty(leaf);
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 
 			inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 
@@ -2109,12 +2109,12 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
 			btrfs_end_transaction(trans, root);
 		}
 next:
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		key.offset++;
 	}
 	ret = 0;
 out:
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	unlock_extent(&BTRFS_I(src)->io_tree, off, off+len, GFP_NOFS);
 out_unlock:
 	mutex_unlock(&src->i_mutex);
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 58250e0..b5d73ea 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -961,7 +961,7 @@ again:
 			lower = upper;
 			upper = NULL;
 		}
-		btrfs_release_path(root, path2);
+		btrfs_release_path(path2);
 next:
 		if (ptr < end) {
 			ptr += btrfs_extent_inline_ref_size(key.type);
@@ -974,7 +974,7 @@ next:
 		if (ptr >= end)
 			path1->slots[0]++;
 	}
-	btrfs_release_path(rc->extent_root, path1);
+	btrfs_release_path(path1);
 
 	cur->checked = 1;
 	WARN_ON(exist);
@@ -1749,7 +1749,7 @@ again:
 
 		btrfs_node_key_to_cpu(path->nodes[level], &key,
 				      path->slots[level]);
-		btrfs_release_path(src, path);
+		btrfs_release_path(path);
 
 		path->lowest_level = level;
 		ret = btrfs_search_slot(trans, src, &key, path, 0, 1);
@@ -2496,7 +2496,7 @@ static int do_relocation(struct btrfs_trans_handle *trans,
 			path->locks[upper->level] = 0;
 
 			slot = path->slots[upper->level];
-			btrfs_release_path(NULL, path);
+			btrfs_release_path(path);
 		} else {
 			ret = btrfs_bin_search(upper->eb, key, upper->level,
 					       &slot);
@@ -2737,7 +2737,7 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans,
 		} else {
 			path->lowest_level = node->level;
 			ret = btrfs_search_slot(trans, root, key, path, 0, 1);
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			if (ret > 0)
 				ret = 0;
 		}
@@ -3119,7 +3119,7 @@ static int add_tree_block(struct reloc_control *rc,
 #endif
 	}
 
-	btrfs_release_path(rc->extent_root, path);
+	btrfs_release_path(path);
 
 	BUG_ON(level == -1);
 
@@ -3505,7 +3505,7 @@ int add_data_references(struct reloc_control *rc,
 		}
 		path->slots[0]++;
 	}
-	btrfs_release_path(rc->extent_root, path);
+	btrfs_release_path(path);
 	if (err)
 		free_block_list(blocks);
 	return err;
@@ -3568,7 +3568,7 @@ next:
 					    EXTENT_DIRTY);
 
 		if (ret == 0 && start <= key.objectid) {
-			btrfs_release_path(rc->extent_root, path);
+			btrfs_release_path(path);
 			rc->search_start = end + 1;
 		} else {
 			rc->search_start = key.objectid + key.offset;
@@ -3576,7 +3576,7 @@ next:
 			return 0;
 		}
 	}
-	btrfs_release_path(rc->extent_root, path);
+	btrfs_release_path(path);
 	return ret;
 }
 
@@ -3713,7 +3713,7 @@ restart:
 				flags = BTRFS_EXTENT_FLAG_DATA;
 
 			if (path_change) {
-				btrfs_release_path(rc->extent_root, path);
+				btrfs_release_path(path);
 
 				path->search_commit_root = 1;
 				path->skip_locking = 1;
@@ -3736,7 +3736,7 @@ restart:
 			   (flags & BTRFS_EXTENT_FLAG_DATA)) {
 			ret = add_data_references(rc, &key, path, &blocks);
 		} else {
-			btrfs_release_path(rc->extent_root, path);
+			btrfs_release_path(path);
 			ret = 0;
 		}
 		if (ret < 0) {
@@ -3799,7 +3799,7 @@ restart:
 		}
 	}
 
-	btrfs_release_path(rc->extent_root, path);
+	btrfs_release_path(path);
 	clear_extent_bits(&rc->processed_blocks, 0, (u64)-1, EXTENT_DIRTY,
 			  GFP_NOFS);
 
@@ -3867,7 +3867,7 @@ static int __insert_orphan_inode(struct btrfs_trans_handle *trans,
 	btrfs_set_inode_flags(leaf, item, BTRFS_INODE_NOCOMPRESS |
 					  BTRFS_INODE_PREALLOC);
 	btrfs_mark_buffer_dirty(leaf);
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 out:
 	btrfs_free_path(path);
 	return ret;
@@ -4109,7 +4109,7 @@ int btrfs_recover_relocation(struct btrfs_root *root)
 		}
 		leaf = path->nodes[0];
 		btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
-		btrfs_release_path(root->fs_info->tree_root, path);
+		btrfs_release_path(path);
 
 		if (key.objectid != BTRFS_TREE_RELOC_OBJECTID ||
 		    key.type != BTRFS_ROOT_ITEM_KEY)
@@ -4141,7 +4141,7 @@ int btrfs_recover_relocation(struct btrfs_root *root)
 
 		key.offset--;
 	}
-	btrfs_release_path(root->fs_info->tree_root, path);
+	btrfs_release_path(path);
 
 	if (list_empty(&reloc_roots))
 		goto out;
diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c
index 29b2d7c..4ff3a6e 100644
--- a/fs/btrfs/root-tree.c
+++ b/fs/btrfs/root-tree.c
@@ -57,7 +57,7 @@ again:
 	btrfs_item_key_to_cpu(path->nodes[0], &search_key, path->slots[0]);
 	if (search_key.type != BTRFS_ROOT_ITEM_KEY) {
 		search_key.offset++;
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		goto again;
 	}
 	ret = 0;
@@ -230,7 +230,7 @@ again:
 
 		memcpy(&found_key, &key, sizeof(key));
 		key.offset++;
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		dead_root =
 			btrfs_read_fs_root_no_radix(root->fs_info->tree_root,
 						    &found_key);
@@ -292,7 +292,7 @@ int btrfs_find_orphan_roots(struct btrfs_root *tree_root)
 		}
 
 		btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
-		btrfs_release_path(tree_root, path);
+		btrfs_release_path(path);
 
 		if (key.objectid != BTRFS_ORPHAN_OBJECTID ||
 		    key.type != BTRFS_ORPHAN_ITEM_KEY)
@@ -390,7 +390,7 @@ again:
 		err = -ENOENT;
 
 	if (key.type == BTRFS_ROOT_BACKREF_KEY) {
-		btrfs_release_path(tree_root, path);
+		btrfs_release_path(path);
 		key.objectid = ref_id;
 		key.type = BTRFS_ROOT_REF_KEY;
 		key.offset = root_id;
@@ -463,7 +463,7 @@ again:
 	btrfs_mark_buffer_dirty(leaf);
 
 	if (key.type == BTRFS_ROOT_BACKREF_KEY) {
-		btrfs_release_path(tree_root, path);
+		btrfs_release_path(path);
 		key.objectid = ref_id;
 		key.type = BTRFS_ROOT_REF_KEY;
 		key.offset = root_id;
diff --git a/fs/btrfs/tree-defrag.c b/fs/btrfs/tree-defrag.c
index 992ab42..3b580ee 100644
--- a/fs/btrfs/tree-defrag.c
+++ b/fs/btrfs/tree-defrag.c
@@ -97,7 +97,7 @@ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
 		ret = 0;
 		goto out;
 	}
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	wret = btrfs_search_slot(trans, root, &key, path, 0, 1);
 
 	if (wret < 0) {
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index c50271a..cad926d 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -333,13 +333,13 @@ static noinline int overwrite_item(struct btrfs_trans_handle *trans,
 			goto insert;
 
 		if (item_size == 0) {
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			return 0;
 		}
 		dst_copy = kmalloc(item_size, GFP_NOFS);
 		src_copy = kmalloc(item_size, GFP_NOFS);
 		if (!dst_copy || !src_copy) {
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			kfree(dst_copy);
 			kfree(src_copy);
 			return -ENOMEM;
@@ -361,13 +361,13 @@ static noinline int overwrite_item(struct btrfs_trans_handle *trans,
 		 * sync
 		 */
 		if (ret == 0) {
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			return 0;
 		}
 
 	}
 insert:
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	/* try to insert the key into the destination tree */
 	ret = btrfs_insert_empty_item(trans, root, path,
 				      key, item_size);
@@ -438,7 +438,7 @@ insert:
 	}
 no_copy:
 	btrfs_mark_buffer_dirty(path->nodes[0]);
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	return 0;
 }
 
@@ -544,11 +544,11 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
 		 * we don't have to do anything
 		 */
 		if (memcmp(&cmp1, &cmp2, sizeof(cmp1)) == 0) {
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			goto out;
 		}
 	}
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	saved_nbytes = inode_get_bytes(inode);
 	/* drop any overlapping extents */
@@ -600,7 +600,7 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
 						key->objectid, offset, &ins);
 				BUG_ON(ret);
 			}
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 
 			if (btrfs_file_extent_compression(eb, item)) {
 				csum_start = ins.objectid;
@@ -629,7 +629,7 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
 				kfree(sums);
 			}
 		} else {
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 		}
 	} else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
 		/* inline extents are easy, we just overwrite them */
@@ -675,7 +675,7 @@ static noinline int drop_one_dir_item(struct btrfs_trans_handle *trans,
 		return -ENOMEM;
 
 	read_extent_buffer(leaf, name, (unsigned long)(di + 1), name_len);
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	inode = read_one_inode(root, location.objectid);
 	BUG_ON(!inode);
@@ -713,7 +713,7 @@ static noinline int inode_in_dir(struct btrfs_root *root,
 			goto out;
 	} else
 		goto out;
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	di = btrfs_lookup_dir_item(NULL, root, path, dirid, name, name_len, 0);
 	if (di && !IS_ERR(di)) {
@@ -724,7 +724,7 @@ static noinline int inode_in_dir(struct btrfs_root *root,
 		goto out;
 	match = 1;
 out:
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	return match;
 }
 
@@ -884,7 +884,7 @@ again:
 			if (!backref_in_log(log, key, victim_name,
 					    victim_name_len)) {
 				btrfs_inc_nlink(inode);
-				btrfs_release_path(root, path);
+				btrfs_release_path(path);
 
 				ret = btrfs_unlink_inode(trans, root, dir,
 							 inode, victim_name,
@@ -901,7 +901,7 @@ again:
 		 */
 		search_done = 1;
 	}
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 insert:
 	/* insert our name */
@@ -922,7 +922,7 @@ out:
 	BUG_ON(ret);
 
 out_nowrite:
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	iput(dir);
 	iput(inode);
 	return 0;
@@ -999,9 +999,9 @@ static noinline int fixup_inode_link_count(struct btrfs_trans_handle *trans,
 		if (key.offset == 0)
 			break;
 		key.offset--;
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 	}
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	if (nlink != inode->i_nlink) {
 		inode->i_nlink = nlink;
 		btrfs_update_inode(trans, root, inode);
@@ -1052,7 +1052,7 @@ static noinline int fixup_inode_link_counts(struct btrfs_trans_handle *trans,
 		ret = btrfs_del_item(trans, root, path);
 		BUG_ON(ret);
 
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		inode = read_one_inode(root, key.offset);
 		BUG_ON(!inode);
 
@@ -1068,7 +1068,7 @@ static noinline int fixup_inode_link_counts(struct btrfs_trans_handle *trans,
 		 */
 		key.offset = (u64)-1;
 	}
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	return 0;
 }
 
@@ -1096,7 +1096,7 @@ static noinline int link_to_fixup_dir(struct btrfs_trans_handle *trans,
 
 	ret = btrfs_insert_empty_item(trans, root, path, &key, 0);
 
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	if (ret == 0) {
 		btrfs_inc_nlink(inode);
 		btrfs_update_inode(trans, root, inode);
@@ -1192,7 +1192,7 @@ static noinline int replay_one_name(struct btrfs_trans_handle *trans,
 		exists = 1;
 	else
 		exists = 0;
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	if (key->type == BTRFS_DIR_ITEM_KEY) {
 		dst_di = btrfs_lookup_dir_item(trans, root, path, key->objectid,
@@ -1236,13 +1236,13 @@ static noinline int replay_one_name(struct btrfs_trans_handle *trans,
 	if (key->type == BTRFS_DIR_INDEX_KEY)
 		goto insert;
 out:
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	kfree(name);
 	iput(dir);
 	return 0;
 
 insert:
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	ret = insert_one_name(trans, root, path, key->objectid, key->offset,
 			      name, name_len, log_type, &log_key);
 
@@ -1363,7 +1363,7 @@ next:
 	*end_ret = found_end;
 	ret = 0;
 out:
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	return ret;
 }
 
@@ -1428,8 +1428,8 @@ again:
 		}
 		if (!log_di || IS_ERR(log_di)) {
 			btrfs_dir_item_key_to_cpu(eb, di, &location);
-			btrfs_release_path(root, path);
-			btrfs_release_path(log, log_path);
+			btrfs_release_path(path);
+			btrfs_release_path(log_path);
 			inode = read_one_inode(root, location.objectid);
 			BUG_ON(!inode);
 
@@ -1453,7 +1453,7 @@ again:
 			ret = 0;
 			goto out;
 		}
-		btrfs_release_path(log, log_path);
+		btrfs_release_path(log_path);
 		kfree(name);
 
 		ptr = (unsigned long)(di + 1);
@@ -1461,8 +1461,8 @@ again:
 	}
 	ret = 0;
 out:
-	btrfs_release_path(root, path);
-	btrfs_release_path(log, log_path);
+	btrfs_release_path(path);
+	btrfs_release_path(log_path);
 	return ret;
 }
 
@@ -1550,7 +1550,7 @@ again:
 				break;
 			dir_key.offset = found_key.offset + 1;
 		}
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		if (range_end == (u64)-1)
 			break;
 		range_start = range_end + 1;
@@ -1561,11 +1561,11 @@ next_type:
 	if (key_type == BTRFS_DIR_LOG_ITEM_KEY) {
 		key_type = BTRFS_DIR_LOG_INDEX_KEY;
 		dir_key.type = BTRFS_DIR_INDEX_KEY;
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		goto again;
 	}
 out:
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 	btrfs_free_path(log_path);
 	iput(dir);
 	return ret;
@@ -2223,7 +2223,7 @@ int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans,
 		bytes_del += name_len;
 		BUG_ON(ret);
 	}
-	btrfs_release_path(log, path);
+	btrfs_release_path(path);
 	di = btrfs_lookup_dir_index_item(trans, log, path, dir->i_ino,
 					 index, name, name_len, -1);
 	if (IS_ERR(di)) {
@@ -2245,7 +2245,7 @@ int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans,
 		key.objectid = dir->i_ino;
 		key.offset = 0;
 		key.type = BTRFS_INODE_ITEM_KEY;
-		btrfs_release_path(log, path);
+		btrfs_release_path(path);
 
 		ret = btrfs_search_slot(trans, log, &key, path, 0, 1);
 		if (ret < 0) {
@@ -2267,7 +2267,7 @@ int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans,
 			btrfs_mark_buffer_dirty(path->nodes[0]);
 		} else
 			ret = 0;
-		btrfs_release_path(log, path);
+		btrfs_release_path(path);
 	}
 fail:
 	btrfs_free_path(path);
@@ -2341,7 +2341,7 @@ static noinline int insert_dir_log_key(struct btrfs_trans_handle *trans,
 			      struct btrfs_dir_log_item);
 	btrfs_set_dir_log_end(path->nodes[0], item, last_offset);
 	btrfs_mark_buffer_dirty(path->nodes[0]);
-	btrfs_release_path(log, path);
+	btrfs_release_path(path);
 	return 0;
 }
 
@@ -2390,10 +2390,10 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 		min_key.objectid = inode->i_ino;
 		min_key.type = key_type;
 		min_key.offset = (u64)-1;
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		ret = btrfs_search_slot(NULL, root, &min_key, path, 0, 0);
 		if (ret < 0) {
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			return ret;
 		}
 		ret = btrfs_previous_item(root, path, inode->i_ino, key_type);
@@ -2429,7 +2429,7 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 			}
 		}
 	}
-	btrfs_release_path(root, path);
+	btrfs_release_path(path);
 
 	/* find the first key from this transaction again */
 	ret = btrfs_search_slot(NULL, root, &min_key, path, 0, 0);
@@ -2487,8 +2487,8 @@ static noinline int log_dir_items(struct btrfs_trans_handle *trans,
 		}
 	}
 done:
-	btrfs_release_path(root, path);
-	btrfs_release_path(log, dst_path);
+	btrfs_release_path(path);
+	btrfs_release_path(dst_path);
 
 	if (err == 0) {
 		*last_offset_ret = last_offset;
@@ -2585,9 +2585,9 @@ static int drop_objectid_items(struct btrfs_trans_handle *trans,
 
 		ret = btrfs_del_item(trans, log, path);
 		BUG_ON(ret);
-		btrfs_release_path(log, path);
+		btrfs_release_path(path);
 	}
-	btrfs_release_path(log, path);
+	btrfs_release_path(path);
 	return ret;
 }
 
@@ -2693,7 +2693,7 @@ static noinline int copy_items(struct btrfs_trans_handle *trans,
 	}
 
 	btrfs_mark_buffer_dirty(dst_path->nodes[0]);
-	btrfs_release_path(log, dst_path);
+	btrfs_release_path(dst_path);
 	kfree(ins_data);
 
 	/*
@@ -2842,7 +2842,7 @@ next_slot:
 			}
 			ins_nr = 0;
 		}
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 
 		if (min_key.offset < (u64)-1)
 			min_key.offset++;
@@ -2865,8 +2865,8 @@ next_slot:
 	}
 	WARN_ON(ins_nr);
 	if (inode_only == LOG_INODE_ALL && S_ISDIR(inode->i_mode)) {
-		btrfs_release_path(root, path);
-		btrfs_release_path(log, dst_path);
+		btrfs_release_path(path);
+		btrfs_release_path(dst_path);
 		ret = log_directory_changes(trans, root, inode, path, dst_path);
 		if (ret) {
 			err = ret;
@@ -3133,7 +3133,7 @@ again:
 		}
 		btrfs_item_key_to_cpu(path->nodes[0], &found_key,
 				      path->slots[0]);
-		btrfs_release_path(log_root_tree, path);
+		btrfs_release_path(path);
 		if (found_key.objectid != BTRFS_TREE_LOG_OBJECTID)
 			break;
 
@@ -3168,7 +3168,7 @@ again:
 		if (found_key.offset == 0)
 			break;
 	}
-	btrfs_release_path(log_root_tree, path);
+	btrfs_release_path(path);
 
 	/* step one is to pin it all, step two is to replay just inodes */
 	if (wc.pin) {
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 309a57b..694dccc 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -1465,7 +1465,7 @@ next_slot:
 				goto error;
 			leaf = path->nodes[0];
 			btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			continue;
 		}
 
@@ -1937,7 +1937,7 @@ again:
 		chunk = btrfs_item_ptr(leaf, path->slots[0],
 				       struct btrfs_chunk);
 		chunk_type = btrfs_chunk_type(leaf, chunk);
-		btrfs_release_path(chunk_root, path);
+		btrfs_release_path(path);
 
 		if (chunk_type & BTRFS_BLOCK_GROUP_SYSTEM) {
 			ret = btrfs_relocate_chunk(chunk_root, chunk_tree,
@@ -2055,7 +2055,7 @@ int btrfs_balance(struct btrfs_root *dev_root)
 		if (found_key.offset == 0)
 			break;
 
-		btrfs_release_path(chunk_root, path);
+		btrfs_release_path(path);
 		ret = btrfs_relocate_chunk(chunk_root,
 					   chunk_root->root_key.objectid,
 					   found_key.objectid,
@@ -2127,7 +2127,7 @@ again:
 			goto done;
 		if (ret) {
 			ret = 0;
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			break;
 		}
 
@@ -2136,7 +2136,7 @@ again:
 		btrfs_item_key_to_cpu(l, &key, path->slots[0]);
 
 		if (key.objectid != device->devid) {
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			break;
 		}
 
@@ -2144,14 +2144,14 @@ again:
 		length = btrfs_dev_extent_length(l, dev_extent);
 
 		if (key.offset + length <= new_size) {
-			btrfs_release_path(root, path);
+			btrfs_release_path(path);
 			break;
 		}
 
 		chunk_tree = btrfs_dev_extent_chunk_tree(l, dev_extent);
 		chunk_objectid = btrfs_dev_extent_chunk_objectid(l, dev_extent);
 		chunk_offset = btrfs_dev_extent_chunk_offset(l, dev_extent);
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 
 		ret = btrfs_relocate_chunk(root, chunk_tree, chunk_objectid,
 					   chunk_offset);
@@ -3803,7 +3803,7 @@ again:
 	}
 	if (key.objectid == BTRFS_DEV_ITEMS_OBJECTID) {
 		key.objectid = 0;
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 		goto again;
 	}
 	ret = 0;
diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c
index a5303b8..3dfd749 100644
--- a/fs/btrfs/xattr.c
+++ b/fs/btrfs/xattr.c
@@ -120,13 +120,13 @@ static int do_setxattr(struct btrfs_trans_handle *trans,
 
 		ret = btrfs_delete_one_dir_name(trans, root, path, di);
 		BUG_ON(ret);
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 
 		/* if we don't have a value then we are removing the xattr */
 		if (!value)
 			goto out;
 	} else {
-		btrfs_release_path(root, path);
+		btrfs_release_path(path);
 
 		if (flags & XATTR_REPLACE) {
 			/* we couldn't find the attr to replace */
-- 
1.7.0.rc2.40.g7d8aa


^ permalink raw reply related	[flat|nested] 19+ messages in thread

* [PATCH 2/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2011-03-31  4:00 [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE Larry D'Anna
  2011-03-31  4:00 ` [PATCH 1/2] btrfs: remove unused argument 'root' from btrfs_release_path Larry D'Anna
@ 2011-03-31  4:00 ` Larry D'Anna
  2011-03-31  6:36 ` [PATCH 0/2] " Christoph Hellwig
  2 siblings, 0 replies; 19+ messages in thread
From: Larry D'Anna @ 2011-03-31  4:00 UTC (permalink / raw)
  To: linux-btrfs; +Cc: larry, Yan Zheng


Signed-off-by: Larry D'Anna <larry@elder-gods.org>
---
 fs/btrfs/ioctl.c |   10 ++++++----
 1 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index f9717b6..10095c7 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -1804,6 +1804,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
 {
 	struct inode *inode = fdentry(file)->d_inode;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
+	struct btrfs_root *srcroot;
 	struct file *src_file;
 	struct inode *src;
 	struct btrfs_trans_handle *trans;
@@ -1846,6 +1847,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
 	}
 
 	src = src_file->f_dentry->d_inode;
+	srcroot = BTRFS_I(src)->root;
 
 	ret = -EINVAL;
 	if (src == inode)
@@ -1860,11 +1862,11 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
 		goto out_fput;
 
 	ret = -EXDEV;
-	if (src->i_sb != inode->i_sb || BTRFS_I(src)->root != root)
+	if (src->i_sb != inode->i_sb)
 		goto out_fput;
 
 	ret = -ENOMEM;
-	buf = vmalloc(btrfs_level_size(root, 0));
+	buf = vmalloc(btrfs_level_size(srcroot, 0));
 	if (!buf)
 		goto out_fput;
 
@@ -1924,13 +1926,13 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
 		 * note the key will change type as we walk through the
 		 * tree.
 		 */
-		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+		ret = btrfs_search_slot(NULL, srcroot, &key, path, 0, 0);
 		if (ret < 0)
 			goto out;
 
 		nritems = btrfs_header_nritems(path->nodes[0]);
 		if (path->slots[0] >= nritems) {
-			ret = btrfs_next_leaf(root, path);
+			ret = btrfs_next_leaf(srcroot, path);
 			if (ret < 0)
 				goto out;
 			if (ret > 0)
-- 
1.7.0.rc2.40.g7d8aa


^ permalink raw reply related	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2011-03-31  4:00 [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE Larry D'Anna
  2011-03-31  4:00 ` [PATCH 1/2] btrfs: remove unused argument 'root' from btrfs_release_path Larry D'Anna
  2011-03-31  4:00 ` [PATCH 2/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE Larry D'Anna
@ 2011-03-31  6:36 ` Christoph Hellwig
  2011-03-31  6:44   ` Arne Jansen
  2011-03-31 12:02   ` Chris Mason
  2 siblings, 2 replies; 19+ messages in thread
From: Christoph Hellwig @ 2011-03-31  6:36 UTC (permalink / raw)
  To: Larry D'Anna; +Cc: linux-btrfs, Yan Zheng

On Thu, Mar 31, 2011 at 12:00:11AM -0400, Larry D'Anna wrote:
> This is a simple patch to allow reflinks to be made crossing subvolume
> boundaries.

NAK.  subvolumes will have to become vfsmounts sooner or later, and we
really must not support any operations spanning mountpoints.


^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2011-03-31  6:36 ` [PATCH 0/2] " Christoph Hellwig
@ 2011-03-31  6:44   ` Arne Jansen
  2011-03-31 12:02   ` Chris Mason
  1 sibling, 0 replies; 19+ messages in thread
From: Arne Jansen @ 2011-03-31  6:44 UTC (permalink / raw)
  To: Christoph Hellwig; +Cc: Larry D'Anna, linux-btrfs, Yan Zheng

On 31.03.2011 08:36, Christoph Hellwig wrote:
> On Thu, Mar 31, 2011 at 12:00:11AM -0400, Larry D'Anna wrote:
>> This is a simple patch to allow reflinks to be made crossing subvolume
>> boundaries.
> 
> NAK.  subvolumes will have to become vfsmounts sooner or later, and we
> really must not support any operations spanning mountpoints.

In what way would this interfere? the reflinked file gets its own inum
in its own mountpoint. The two files only internally share some disk
space.

> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html


^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2011-03-31  6:36 ` [PATCH 0/2] " Christoph Hellwig
  2011-03-31  6:44   ` Arne Jansen
@ 2011-03-31 12:02   ` Chris Mason
  2011-04-01 13:34     ` Christoph Hellwig
  1 sibling, 1 reply; 19+ messages in thread
From: Chris Mason @ 2011-03-31 12:02 UTC (permalink / raw)
  To: Christoph Hellwig; +Cc: Larry D'Anna, linux-btrfs, Yan Zheng

Excerpts from Christoph Hellwig's message of 2011-03-31 02:36:36 -0400:
> On Thu, Mar 31, 2011 at 12:00:11AM -0400, Larry D'Anna wrote:
> > This is a simple patch to allow reflinks to be made crossing subvolume
> > boundaries.
> 
> NAK.  subvolumes will have to become vfsmounts sooner or later, and we
> really must not support any operations spanning mountpoints.
> 

Sorry, I disagree here.  reflinks were always intended to be able to
span subvolumes.  There's no conflict at all because they span different
inodes.

-chris

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2011-03-31 12:02   ` Chris Mason
@ 2011-04-01 13:34     ` Christoph Hellwig
  2011-04-01 13:40       ` Chris Mason
  2011-04-02 15:51       ` Ken Drummond
  0 siblings, 2 replies; 19+ messages in thread
From: Christoph Hellwig @ 2011-04-01 13:34 UTC (permalink / raw)
  To: Chris Mason; +Cc: Christoph Hellwig, Larry D'Anna, linux-btrfs, Yan Zheng

On Thu, Mar 31, 2011 at 08:02:22AM -0400, Chris Mason wrote:
> Excerpts from Christoph Hellwig's message of 2011-03-31 02:36:36 -0400:
> > On Thu, Mar 31, 2011 at 12:00:11AM -0400, Larry D'Anna wrote:
> > > This is a simple patch to allow reflinks to be made crossing subvolume
> > > boundaries.
> > 
> > NAK.  subvolumes will have to become vfsmounts sooner or later, and we
> > really must not support any operations spanning mountpoints.
> > 
> 
> Sorry, I disagree here.  reflinks were always intended to be able to
> span subvolumes.  There's no conflict at all because they span different
> inodes.

I don't think it's a good idea to introduce any user visible operations
over subvolume boundaries.  Currently we don't have any operations over
mount boundaries, which is pretty fumdamental to the unix filesystem
semantics.  If you want to change this please come up with a clear
description of the semantics and post it to linux-fsdevel for
discussion.  That of course requires a clear description of the
btrfs subvolumes, which is still completely missing.


^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2011-04-01 13:34     ` Christoph Hellwig
@ 2011-04-01 13:40       ` Chris Mason
  2011-04-02  1:59         ` Fajar A. Nugraha
  2011-12-22 12:24         ` Chris Samuel
  2011-04-02 15:51       ` Ken Drummond
  1 sibling, 2 replies; 19+ messages in thread
From: Chris Mason @ 2011-04-01 13:40 UTC (permalink / raw)
  To: Christoph Hellwig; +Cc: Larry D'Anna, linux-btrfs, Yan Zheng

Excerpts from Christoph Hellwig's message of 2011-04-01 09:34:05 -0400:
> On Thu, Mar 31, 2011 at 08:02:22AM -0400, Chris Mason wrote:
> > Excerpts from Christoph Hellwig's message of 2011-03-31 02:36:36 -0400:
> > > On Thu, Mar 31, 2011 at 12:00:11AM -0400, Larry D'Anna wrote:
> > > > This is a simple patch to allow reflinks to be made crossing subvolume
> > > > boundaries.
> > > 
> > > NAK.  subvolumes will have to become vfsmounts sooner or later, and we
> > > really must not support any operations spanning mountpoints.
> > > 
> > 
> > Sorry, I disagree here.  reflinks were always intended to be able to
> > span subvolumes.  There's no conflict at all because they span different
> > inodes.
> 
> I don't think it's a good idea to introduce any user visible operations
> over subvolume boundaries.  Currently we don't have any operations over
> mount boundaries, which is pretty fumdamental to the unix filesystem
> semantics.  If you want to change this please come up with a clear
> description of the semantics and post it to linux-fsdevel for
> discussion.  That of course requires a clear description of the
> btrfs subvolumes, which is still completely missing.
> 

The subvolume is just a directory tree that can be snapshotted, and has
it's own private inode number space. 

reflink across subvolumes is no different from copying a file from one
subvolume to another at the VFS level.  The src and destination are
different files and different inodes, they just happen to share data
extents.

-chris

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2011-04-01 13:40       ` Chris Mason
@ 2011-04-02  1:59         ` Fajar A. Nugraha
  2011-12-22 12:24         ` Chris Samuel
  1 sibling, 0 replies; 19+ messages in thread
From: Fajar A. Nugraha @ 2011-04-02  1:59 UTC (permalink / raw)
  To: linux-btrfs

On Fri, Apr 1, 2011 at 8:40 PM, Chris Mason <chris.mason@oracle.com> wr=
ote:
> Excerpts from Christoph Hellwig's message of 2011-04-01 09:34:05 -040=
0:
>> I don't think it's a good idea to introduce any user visible operati=
ons
>> over subvolume boundaries. =A0Currently we don't have any operations=
 over
>> mount boundaries, which is pretty fumdamental to the unix filesystem
>> semantics. =A0If you want to change this please come up with a clear
>> description of the semantics and post it to linux-fsdevel for
>> discussion. =A0That of course requires a clear description of the
>> btrfs subvolumes, which is still completely missing.
>>
>
> The subvolume is just a directory tree that can be snapshotted, and h=
as
> it's own private inode number space.
>
> reflink across subvolumes is no different from copying a file from on=
e
> subvolume to another at the VFS level. =A0The src and destination are
> different files and different inodes, they just happen to share data
> extents.

=2E.. and currently copying file from one subvolume to another requires
copying the data as well. It'd be great if you could only copy the
metadata.

A good possible use that comes to mind is to quickly merge several
subvolumes into a new one.

--=20
=46ajar
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" =
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2011-04-01 13:40       ` Chris Mason
  2011-04-02  1:59         ` Fajar A. Nugraha
@ 2011-12-22 12:24         ` Chris Samuel
  2012-01-06 12:04           ` Konstantinos Skarlatos
  1 sibling, 1 reply; 19+ messages in thread
From: Chris Samuel @ 2011-12-22 12:24 UTC (permalink / raw)
  To: linux-btrfs, Christoph Hellwig

[-- Attachment #1: Type: Text/Plain, Size: 1951 bytes --]

Christoph,

On Sat, 2 Apr 2011 12:40:11 AM Chris Mason wrote:

> Excerpts from Christoph Hellwig's message of 2011-04-01 09:34:05 
-0400:
>
> > I don't think it's a good idea to introduce any user visible
> > operations over subvolume boundaries.  Currently we don't have
> > any operations over mount boundaries, which is pretty
> > fumdamental to the unix filesystem semantics.  If you want to
> > change this please come up with a clear description of the
> > semantics and post it to linux-fsdevel for discussion.  That of
> > course requires a clear description of the btrfs subvolumes,
> > which is still completely missing.
> 
> The subvolume is just a directory tree that can be snapshotted, and
> has it's own private inode number space.
> 
> reflink across subvolumes is no different from copying a file from
> one subvolume to another at the VFS level.  The src and
> destination are different files and different inodes, they just
> happen to share data extents.

Were Chris Mason's points above enough to sway your opposition to this 
functionality/patch?

There is demand for the ability to move data between subvolumes 
without needing to copy the extents themselves, it's cropped up again 
on the list in recent days.

It seems a little hard (and counterintuitive) to enforce a wasteful 
use of resources to copy data between different parts of the same 
filesystem which happen to be a on a different subvolume when it's 
permitted & functional to the same filesystem on the same subvolume.

I don't dispute the comment about documentation on subvolumes though, 
there is a short discussion of them on the btrfs wiki in the sysadmins 
guide, but not really a lot of detail. :-)

All the best,
Chris
-- 
 Chris Samuel  :  http://www.csamuel.org/  :  Melbourne, VIC

This email may come with a PGP signature as a file. Do not panic.
For more info see: http://en.wikipedia.org/wiki/OpenPGP

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 482 bytes --]

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2011-12-22 12:24         ` Chris Samuel
@ 2012-01-06 12:04           ` Konstantinos Skarlatos
  2012-01-06 17:57             ` David Sterba
  0 siblings, 1 reply; 19+ messages in thread
From: Konstantinos Skarlatos @ 2012-01-06 12:04 UTC (permalink / raw)
  To: Chris Samuel; +Cc: linux-btrfs, Christoph Hellwig, Chris Mason

On 22/12/2011 2:24 =CE=BC=CE=BC, Chris Samuel wrote:
> Christoph,
>
> On Sat, 2 Apr 2011 12:40:11 AM Chris Mason wrote:
>
>> Excerpts from Christoph Hellwig's message of 2011-04-01 09:34:05
> -0400:
>>
>>> I don't think it's a good idea to introduce any user visible
>>> operations over subvolume boundaries.  Currently we don't have
>>> any operations over mount boundaries, which is pretty
>>> fumdamental to the unix filesystem semantics.  If you want to
>>> change this please come up with a clear description of the
>>> semantics and post it to linux-fsdevel for discussion.  That of
>>> course requires a clear description of the btrfs subvolumes,
>>> which is still completely missing.
>>
>> The subvolume is just a directory tree that can be snapshotted, and
>> has it's own private inode number space.
>>
>> reflink across subvolumes is no different from copying a file from
>> one subvolume to another at the VFS level.  The src and
>> destination are different files and different inodes, they just
>> happen to share data extents.
>
> Were Chris Mason's points above enough to sway your opposition to thi=
s
> functionality/patch?
>
> There is demand for the ability to move data between subvolumes
> without needing to copy the extents themselves, it's cropped up again
> on the list in recent days.
>
> It seems a little hard (and counterintuitive) to enforce a wasteful
> use of resources to copy data between different parts of the same
> filesystem which happen to be a on a different subvolume when it's
> permitted&  functional to the same filesystem on the same subvolume.
>
> I don't dispute the comment about documentation on subvolumes though,
> there is a short discussion of them on the btrfs wiki in the sysadmin=
s
> guide, but not really a lot of detail. :-)
>
> All the best,
> Chris

Me too wants cp --reflink across subvolumes. Please make this feature=20
available to us, as its a "poor man's dedupe" and would give big space=20
savings for many use cases.
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" =
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2012-01-06 12:04           ` Konstantinos Skarlatos
@ 2012-01-06 17:57             ` David Sterba
  2012-01-09  6:58               ` Marios Titas
  0 siblings, 1 reply; 19+ messages in thread
From: David Sterba @ 2012-01-06 17:57 UTC (permalink / raw)
  To: Konstantinos Skarlatos
  Cc: Chris Samuel, linux-btrfs, Christoph Hellwig, Chris Mason

On Fri, Jan 06, 2012 at 02:04:12PM +0200, Konstantinos Skarlatos wrote:
> Me too wants cp --reflink across subvolumes. Please make this feature 
> available to us, as its a "poor man's dedupe" and would give big space 
> savings for many use cases.

The simple case of 'cp --reflink' works fine, the only remaining case is
the unimplemented clone of part of a compressed inline extent. And it has
to be:
  * at least 2 blocks uncompressed
  * the cloned range does not span the whole extent

The ioctl needs ranges and length aligned up to blocksize, ie. 4096,
cloning of short inline extents work.

The missing case can be reproduced with this:

---
#!/bin/sh

# assume, that fs is mounted with compression on
src=test-clone-compressed-inline
dd if=/dev/zero of="$src" bs=1K count=3 oflag=sync
sync
filefrag -vbs "$src"
clone_range "$src" 4096 4096 subvol2/"$src"-dest0 0
clone_range "$src" 4096 4096 subvol2/"$src"-dest4096 4096
clone_range "$src" 4096 4096 subvol2/"$src"-dest8192 8192

---

$ ./test-clone-range-inline
3+0 records in
3+0 records out
3072 bytes (3.1 kB) copied, 0.0487466 s, 63.0 kB/s
Filesystem type is: 9123683e
File size of test-clone-compressed-inline is 3072 (3 blocks, blocksize 1024)
 ext logical physical expected length flags
   0       0        0            4096 not_aligned,inline,eof
test-clone-compressed-inline: 1 extent found
clone_range test-clone-compressed-inline 3 4096~4096 to subvol2/test-clone-compressed-inline-dest0 4 0 = -1 Invalid argument
clone_range test-clone-compressed-inline 3 4096~4096 to subvol2/test-clone-compressed-inline-dest4096 4 4096 = -1 Invalid argument
clone_range test-clone-compressed-inline 3 4096~4096 to subvol2/test-clone-compressed-inline-dest8192 4 8192 = -1 Invalid argument

---

This does not work on a single subvolume, so extending clone to span
subvolumes should not break anything that hasn't been broken already.


david

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2012-01-06 17:57             ` David Sterba
@ 2012-01-09  6:58               ` Marios Titas
  2012-01-09 13:31                 ` Jérôme Poulin
  0 siblings, 1 reply; 19+ messages in thread
From: Marios Titas @ 2012-01-09  6:58 UTC (permalink / raw)
  To: dave, Konstantinos Skarlatos, Chris Samuel, linux-btrfs,
	Christoph Hellwig, Chris Mason

On Fri, Jan 6, 2012 at 12:57, David Sterba <dave@jikos.cz> wrote:
> On Fri, Jan 06, 2012 at 02:04:12PM +0200, Konstantinos Skarlatos wrote:
>> Me too wants cp --reflink across subvolumes. Please make this feature
>> available to us, as its a "poor man's dedupe" and would give big space
>> savings for many use cases.
>
> The simple case of 'cp --reflink' works fine [...]

It doesn't work here:

cp: failed to clone `/tmp/test': Invalid cross-device link

That's with 3.1 + for-linus.

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2012-01-09  6:58               ` Marios Titas
@ 2012-01-09 13:31                 ` Jérôme Poulin
  2012-01-19 16:52                   ` Jérôme Poulin
  0 siblings, 1 reply; 19+ messages in thread
From: Jérôme Poulin @ 2012-01-09 13:31 UTC (permalink / raw)
  To: Marios Titas
  Cc: dave, Konstantinos Skarlatos, Chris Samuel, linux-btrfs,
	Christoph Hellwig, Chris Mason

On Mon, Jan 9, 2012 at 1:58 AM, Marios Titas <redneb8888@gmail.com> wrote:
>> The simple case of 'cp --reflink' works fine [...]
>
> It doesn't work here:
>
> cp: failed to clone `/tmp/test': Invalid cross-device link
>
> That's with 3.1 + for-linus.

This is the problem, it doesn't work because you have to apply the
patch at http://permalink.gmane.org/gmane.comp.file-systems.btrfs/9865
which is not mainlined yet. This patch dates back from March 31st, I
have been using it since it was released.

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2012-01-09 13:31                 ` Jérôme Poulin
@ 2012-01-19 16:52                   ` Jérôme Poulin
  0 siblings, 0 replies; 19+ messages in thread
From: Jérôme Poulin @ 2012-01-19 16:52 UTC (permalink / raw)
  To: linux-btrfs
  Cc: dave, Konstantinos Skarlatos, Chris Samuel, Christoph Hellwig,
	Chris Mason, Marios Titas

Is there any reason why this can't be applied in for-linus? Does it
need more testing? Still no corruption here after copying the whole
root FS in subvolume after btrfs-convert, this FS is used everyday
with big image files, clients backups of Windows directories (small
files) and 3 snapshots a day, using cp --reflink for whole image copy
to test recovery and more.

BTRFS is the only module I need to recompile on this machine after
each upgrading the kernel to apply this patch.

On Mon, Jan 9, 2012 at 8:31 AM, J=C3=A9r=C3=B4me Poulin <jeromepoulin@g=
mail.com> wrote:
> On Mon, Jan 9, 2012 at 1:58 AM, Marios Titas <redneb8888@gmail.com> w=
rote:
>>> The simple case of 'cp --reflink' works fine [...]
>>
>> It doesn't work here:
>>
>> cp: failed to clone `/tmp/test': Invalid cross-device link
>>
>> That's with 3.1 + for-linus.
>
> This is the problem, it doesn't work because you have to apply the
> patch at http://permalink.gmane.org/gmane.comp.file-systems.btrfs/986=
5
> which is not mainlined yet. This patch dates back from March 31st, I
> have been using it since it was released.
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" =
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2011-04-01 13:34     ` Christoph Hellwig
  2011-04-01 13:40       ` Chris Mason
@ 2011-04-02 15:51       ` Ken Drummond
  2011-04-02 16:56         ` Larry D'Anna
  1 sibling, 1 reply; 19+ messages in thread
From: Ken Drummond @ 2011-04-02 15:51 UTC (permalink / raw)
  To: linux-btrfs; +Cc: Christoph Hellwig

On Fri, 2011-04-01 at 09:34 -0400, Christoph Hellwig wrote:
> On Thu, Mar 31, 2011 at 08:02:22AM -0400, Chris Mason wrote:
> > Excerpts from Christoph Hellwig's message of 2011-03-31 02:36:36 -0400:
> > > On Thu, Mar 31, 2011 at 12:00:11AM -0400, Larry D'Anna wrote:
> > > > This is a simple patch to allow reflinks to be made crossing subvolume
> > > > boundaries.
> > > 
> > > NAK.  subvolumes will have to become vfsmounts sooner or later, and we
> > > really must not support any operations spanning mountpoints.
> > > 
> > 
> > Sorry, I disagree here.  reflinks were always intended to be able to
> > span subvolumes.  There's no conflict at all because they span different
> > inodes.
> 
> I don't think it's a good idea to introduce any user visible operations
> over subvolume boundaries.  Currently we don't have any operations over
> mount boundaries, which is pretty fumdamental to the unix filesystem
> semantics.  If you want to change this please come up with a clear
> description of the semantics and post it to linux-fsdevel for
> discussion.  That of course requires a clear description of the
> btrfs subvolumes, which is still completely missing.
> 

I don't really understand the details here, but doesn't the creation of
a snapshot already lead to data extents being shared between
sub-volumes?  From a simple user perspective this sounds like a very
useful capability.



^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2011-04-02 15:51       ` Ken Drummond
@ 2011-04-02 16:56         ` Larry D'Anna
  2011-04-02 20:01           ` Jérôme Poulin
  0 siblings, 1 reply; 19+ messages in thread
From: Larry D'Anna @ 2011-04-02 16:56 UTC (permalink / raw)
  To: Ken Drummond; +Cc: linux-btrfs, Christoph Hellwig

* Ken Drummond (btrfs@kendrummond.com) [110402 11:51]:
> I don't really understand the details here, but doesn't the creation of
> a snapshot already lead to data extents being shared between
> sub-volumes?  From a simple user perspective this sounds like a very
> useful capability.

I was surprised and frustrated to find it missing.  I had just copied a large
quantity of data into btrfs, realized i need to make a subvolume to try out
snapshotting, and then found out i had to copy all the data *again* to get it
into a subvolume.  There are tons of scenarios where users will want and expect
to be able to do this.

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
  2011-04-02 16:56         ` Larry D'Anna
@ 2011-04-02 20:01           ` Jérôme Poulin
  0 siblings, 0 replies; 19+ messages in thread
From: Jérôme Poulin @ 2011-04-02 20:01 UTC (permalink / raw)
  To: Larry D'Anna; +Cc: Ken Drummond, linux-btrfs, Christoph Hellwig

I am very happy to see this patch. It was one of the first thing I
tried after making a subvolume, cp --reflink, and it failed. I'll have
to try this out.

J=E9r=F4me Poulin

On 2011-04-02, at 12:56, Larry D'Anna <larry@elder-gods.org> wrote:

> * Ken Drummond (btrfs@kendrummond.com) [110402 11:51]:
>> I don't really understand the details here, but doesn't the creation=
 of
>> a snapshot already lead to data extents being shared between
>> sub-volumes?  From a simple user perspective this sounds like a very
>> useful capability.
>
> I was surprised and frustrated to find it missing.  I had just copied=
 a large
> quantity of data into btrfs, realized i need to make a subvolume to t=
ry out
> snapshotting, and then found out i had to copy all the data *again* t=
o get it
> into a subvolume.  There are tons of scenarios where users will want =
and expect
> to be able to do this.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs=
" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" =
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE
@ 2011-03-31 11:12 Tomasz Chmielewski
  0 siblings, 0 replies; 19+ messages in thread
From: Tomasz Chmielewski @ 2011-03-31 11:12 UTC (permalink / raw)
  To: linux-btrfs; +Cc: sensille, larry, hch

Arne Jansen wrote:

>> On Thu, Mar 31, 2011 at 12:00:11AM -0400, Larry D'Anna wrote:
>>> This is a simple patch to allow reflinks to be made crossing subvolume
>>> boundaries.
>>
>> NAK.  subvolumes will have to become vfsmounts sooner or later, and we
>> really must not support any operations spanning mountpoints.
>
> In what way would this interfere? the reflinked file gets its own inum
> in its own mountpoint. The two files only internally share some disk
> space.

Similarly, it would be pity if proposed deduplication couldn't cross 
subvolume boundaries...


-- 
Tomasz Chmielewski
http://wpkg.org


^ permalink raw reply	[flat|nested] 19+ messages in thread

end of thread, other threads:[~2012-01-19 16:52 UTC | newest]

Thread overview: 19+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-03-31  4:00 [PATCH 0/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE Larry D'Anna
2011-03-31  4:00 ` [PATCH 1/2] btrfs: remove unused argument 'root' from btrfs_release_path Larry D'Anna
2011-03-31  4:00 ` [PATCH 2/2] btrfs: allow cross-subvolume BTRFS_IOC_CLONE Larry D'Anna
2011-03-31  6:36 ` [PATCH 0/2] " Christoph Hellwig
2011-03-31  6:44   ` Arne Jansen
2011-03-31 12:02   ` Chris Mason
2011-04-01 13:34     ` Christoph Hellwig
2011-04-01 13:40       ` Chris Mason
2011-04-02  1:59         ` Fajar A. Nugraha
2011-12-22 12:24         ` Chris Samuel
2012-01-06 12:04           ` Konstantinos Skarlatos
2012-01-06 17:57             ` David Sterba
2012-01-09  6:58               ` Marios Titas
2012-01-09 13:31                 ` Jérôme Poulin
2012-01-19 16:52                   ` Jérôme Poulin
2011-04-02 15:51       ` Ken Drummond
2011-04-02 16:56         ` Larry D'Anna
2011-04-02 20:01           ` Jérôme Poulin
  -- strict thread matches above, loose matches on Subject: below --
2011-03-31 11:12 Tomasz Chmielewski

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).