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

…/git/mason/linux-btrfs

Pull btrfs fixes from Chris Mason:
 "This has our series of fixes for the next rc.  The biggest batch is
  from Jan Schmidt, fixing up some problems in our subvolume quota code
  and fixing btrfs send/receive to work with the new extended inode
  refs."

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
  Btrfs: do not bug when we fail to commit the transaction
  Btrfs: fix memory leak when cloning root's node
  Btrfs: Use btrfs_update_inode_fallback when creating a snapshot
  Btrfs: Send: preserve ownership (uid and gid) also for symlinks.
  Btrfs: fix deadlock caused by the nested chunk allocation
  btrfs: Return EINVAL when length to trim is less than FSB
  Btrfs: fix memory leak in btrfs_quota_enable()
  Btrfs: send correct rdev and mode in btrfs-send
  Btrfs: extended inode refs support for send mechanism
  Btrfs: Fix wrong error handling code
  Fix a sign bug causing invalid memory access in the ino_paths ioctl.
  Btrfs: comment for loop in tree_mod_log_insert_move
  Btrfs: fix extent buffer reference for tree mod log roots
  Btrfs: determine level of old roots
  Btrfs: tree mod log's old roots could still be part of the tree
  Btrfs: fix a tree mod logging issue for root replacement operations
  Btrfs: don't put removals from push_node_left into tree mod log twice

fs/btrfs/backref.c

@@ -283,9 +283,7 @@ static int __resolve_indirect_ref(struct btrfs_fs_info *fs_info,
 		goto out;
 	}
 
-	rcu_read_lock();
-	root_level = btrfs_header_level(root->node);
-	rcu_read_unlock();
+	root_level = btrfs_old_root_level(root, time_seq);
 
 	if (root_level + 1 == level)
 		goto out;
@@ -1177,16 +1175,15 @@ int btrfs_find_one_extref(struct btrfs_root *root, u64 inode_objectid,
 	return ret;
 }
 
-static char *ref_to_path(struct btrfs_root *fs_root,
-			 struct btrfs_path *path,
-			 u32 name_len, unsigned long name_off,
-			 struct extent_buffer *eb_in, u64 parent,
-			 char *dest, u32 size)
+char *btrfs_ref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path,
+			u32 name_len, unsigned long name_off,
+			struct extent_buffer *eb_in, u64 parent,
+			char *dest, u32 size)
 {
 	int slot;
 	u64 next_inum;
 	int ret;
-	s64 bytes_left = size - 1;
+	s64 bytes_left = ((s64)size) - 1;
 	struct extent_buffer *eb = eb_in;
 	struct btrfs_key found_key;
 	int leave_spinning = path->leave_spinning;
@@ -1266,10 +1263,10 @@ char *btrfs_iref_to_path(struct btrfs_root *fs_root,
 			 struct extent_buffer *eb_in, u64 parent,
 			 char *dest, u32 size)
 {
-	return ref_to_path(fs_root, path,
-			   btrfs_inode_ref_name_len(eb_in, iref),
-			   (unsigned long)(iref + 1),
-			   eb_in, parent, dest, size);
+	return btrfs_ref_to_path(fs_root, path,
+				 btrfs_inode_ref_name_len(eb_in, iref),
+				 (unsigned long)(iref + 1),
+				 eb_in, parent, dest, size);
 }
 
 /*
@@ -1715,9 +1712,8 @@ static int inode_to_path(u64 inum, u32 name_len, unsigned long name_off,
 					ipath->fspath->bytes_left - s_ptr : 0;
 
 	fspath_min = (char *)ipath->fspath->val + (i + 1) * s_ptr;
-	fspath = ref_to_path(ipath->fs_root, ipath->btrfs_path, name_len,
-			     name_off, eb, inum, fspath_min,
-			     bytes_left);
+	fspath = btrfs_ref_to_path(ipath->fs_root, ipath->btrfs_path, name_len,
+				   name_off, eb, inum, fspath_min, bytes_left);
 	if (IS_ERR(fspath))
 		return PTR_ERR(fspath);
 

fs/btrfs/backref.h

@@ -62,6 +62,10 @@ int btrfs_find_all_roots(struct btrfs_trans_handle *trans,
 char *btrfs_iref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path,
 			 struct btrfs_inode_ref *iref, struct extent_buffer *eb,
 			 u64 parent, char *dest, u32 size);
+char *btrfs_ref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path,
+			u32 name_len, unsigned long name_off,
+			struct extent_buffer *eb_in, u64 parent,
+			char *dest, u32 size);
 
 struct btrfs_data_container *init_data_container(u32 total_bytes);
 struct inode_fs_paths *init_ipath(s32 total_bytes, struct btrfs_root *fs_root,

fs/btrfs/ctree.c

@@ -596,6 +596,11 @@ tree_mod_log_insert_move(struct btrfs_fs_info *fs_info,
 	if (tree_mod_dont_log(fs_info, eb))
 		return 0;
 
+	/*
+	 * When we override something during the move, we log these removals.
+	 * This can only happen when we move towards the beginning of the
+	 * buffer, i.e. dst_slot < src_slot.
+	 */
 	for (i = 0; i + dst_slot < src_slot && i < nr_items; i++) {
 		ret = tree_mod_log_insert_key_locked(fs_info, eb, i + dst_slot,
 					      MOD_LOG_KEY_REMOVE_WHILE_MOVING);
@@ -647,8 +652,6 @@ tree_mod_log_insert_root(struct btrfs_fs_info *fs_info,
 	if (tree_mod_dont_log(fs_info, NULL))
 		return 0;
 
-	__tree_mod_log_free_eb(fs_info, old_root);
-
 	ret = tree_mod_alloc(fs_info, flags, &tm);
 	if (ret < 0)
 		goto out;
@@ -926,12 +929,7 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans,
 			ret = btrfs_dec_ref(trans, root, buf, 1, 1);
 			BUG_ON(ret); /* -ENOMEM */
 		}
-		/*
-		 * don't log freeing in case we're freeing the root node, this
-		 * is done by tree_mod_log_set_root_pointer later
-		 */
-		if (buf != root->node && btrfs_header_level(buf) != 0)
-			tree_mod_log_free_eb(root->fs_info, buf);
+		tree_mod_log_free_eb(root->fs_info, buf);
 		clean_tree_block(trans, root, buf);
 		*last_ref = 1;
 	}
@@ -1225,6 +1223,8 @@ tree_mod_log_rewind(struct btrfs_fs_info *fs_info, struct extent_buffer *eb,
 	free_extent_buffer(eb);
 
 	__tree_mod_log_rewind(eb_rewin, time_seq, tm);
+	WARN_ON(btrfs_header_nritems(eb_rewin) >
+		BTRFS_NODEPTRS_PER_BLOCK(fs_info->fs_root));
 
 	return eb_rewin;
 }
@@ -1241,9 +1241,11 @@ get_old_root(struct btrfs_root *root, u64 time_seq)
 {
 	struct tree_mod_elem *tm;
 	struct extent_buffer *eb;
+	struct extent_buffer *old;
 	struct tree_mod_root *old_root = NULL;
 	u64 old_generation = 0;
 	u64 logical;
+	u32 blocksize;
 
 	eb = btrfs_read_lock_root_node(root);
 	tm = __tree_mod_log_oldest_root(root->fs_info, root, time_seq);
@@ -1259,14 +1261,32 @@ get_old_root(struct btrfs_root *root, u64 time_seq)
 	}
 
 	tm = tree_mod_log_search(root->fs_info, logical, time_seq);
-	if (old_root)
+	if (old_root && tm && tm->op != MOD_LOG_KEY_REMOVE_WHILE_FREEING) {
+		btrfs_tree_read_unlock(root->node);
+		free_extent_buffer(root->node);
+		blocksize = btrfs_level_size(root, old_root->level);
+		old = read_tree_block(root, logical, blocksize, 0);
+		if (!old) {
+			pr_warn("btrfs: failed to read tree block %llu from get_old_root\n",
+				logical);
+			WARN_ON(1);
+		} else {
+			eb = btrfs_clone_extent_buffer(old);
+			free_extent_buffer(old);
+		}
+	} else if (old_root) {
+		btrfs_tree_read_unlock(root->node);
+		free_extent_buffer(root->node);
 		eb = alloc_dummy_extent_buffer(logical, root->nodesize);
-	else
+	} else {
 		eb = btrfs_clone_extent_buffer(root->node);
-	btrfs_tree_read_unlock(root->node);
-	free_extent_buffer(root->node);
+		btrfs_tree_read_unlock(root->node);
+		free_extent_buffer(root->node);
+	}
+
 	if (!eb)
 		return NULL;
+	extent_buffer_get(eb);
 	btrfs_tree_read_lock(eb);
 	if (old_root) {
 		btrfs_set_header_bytenr(eb, eb->start);
@@ -1279,11 +1299,28 @@ get_old_root(struct btrfs_root *root, u64 time_seq)
 		__tree_mod_log_rewind(eb, time_seq, tm);
 	else
 		WARN_ON(btrfs_header_level(eb) != 0);
-	extent_buffer_get(eb);
+	WARN_ON(btrfs_header_nritems(eb) > BTRFS_NODEPTRS_PER_BLOCK(root));
 
 	return eb;
 }
 
+int btrfs_old_root_level(struct btrfs_root *root, u64 time_seq)
+{
+	struct tree_mod_elem *tm;
+	int level;
+
+	tm = __tree_mod_log_oldest_root(root->fs_info, root, time_seq);
+	if (tm && tm->op == MOD_LOG_ROOT_REPLACE) {
+		level = tm->old_root.level;
+	} else {
+		rcu_read_lock();
+		level = btrfs_header_level(root->node);
+		rcu_read_unlock();
+	}
+
+	return level;
+}
+
 static inline int should_cow_block(struct btrfs_trans_handle *trans,
 				   struct btrfs_root *root,
 				   struct extent_buffer *buf)
@@ -1725,6 +1762,7 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 			goto enospc;
 		}
 
+		tree_mod_log_free_eb(root->fs_info, root->node);
 		tree_mod_log_set_root_pointer(root, child);
 		rcu_assign_pointer(root->node, child);
 
@@ -2970,8 +3008,10 @@ static int push_node_left(struct btrfs_trans_handle *trans,
 			   push_items * sizeof(struct btrfs_key_ptr));
 
 	if (push_items < src_nritems) {
-		tree_mod_log_eb_move(root->fs_info, src, 0, push_items,
-				     src_nritems - push_items);
+		/*
+		 * don't call tree_mod_log_eb_move here, key removal was already
+		 * fully logged by tree_mod_log_eb_copy above.
+		 */
 		memmove_extent_buffer(src, btrfs_node_key_ptr_offset(0),
 				      btrfs_node_key_ptr_offset(push_items),
 				      (src_nritems - push_items) *

fs/btrfs/ctree.h

@@ -3120,6 +3120,7 @@ static inline u64 btrfs_inc_tree_mod_seq(struct btrfs_fs_info *fs_info)
 {
 	return atomic_inc_return(&fs_info->tree_mod_seq);
 }
+int btrfs_old_root_level(struct btrfs_root *root, u64 time_seq);
 
 /* root-item.c */
 int btrfs_find_root_ref(struct btrfs_root *tree_root,
@@ -3338,6 +3339,8 @@ struct extent_map *btrfs_get_extent(struct inode *inode, struct page *page,
 int btrfs_update_inode(struct btrfs_trans_handle *trans,
 			      struct btrfs_root *root,
 			      struct inode *inode);
+int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans,
+				struct btrfs_root *root, struct inode *inode);
 int btrfs_orphan_add(struct btrfs_trans_handle *trans, struct inode *inode);
 int btrfs_orphan_del(struct btrfs_trans_handle *trans, struct inode *inode);
 int btrfs_orphan_cleanup(struct btrfs_root *root);

fs/btrfs/extent_io.c

@@ -4110,8 +4110,8 @@ struct extent_buffer *alloc_dummy_extent_buffer(u64 start, unsigned long len)
 
 	return eb;
 err:
-	for (i--; i >= 0; i--)
-		__free_page(eb->pages[i]);
+	for (; i > 0; i--)
+		__free_page(eb->pages[i - 1]);
 	__free_extent_buffer(eb);
 	return NULL;
 }

fs/btrfs/inode.c

@@ -94,8 +94,6 @@ static noinline int cow_file_range(struct inode *inode,
 				   struct page *locked_page,
 				   u64 start, u64 end, int *page_started,
 				   unsigned long *nr_written, int unlock);
-static noinline int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans,
-				struct btrfs_root *root, struct inode *inode);
 
 static int btrfs_init_inode_security(struct btrfs_trans_handle *trans,
 				     struct inode *inode,  struct inode *dir,
@@ -2746,8 +2744,9 @@ noinline int btrfs_update_inode(struct btrfs_trans_handle *trans,
 	return btrfs_update_inode_item(trans, root, inode);
 }
 
-static noinline int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans,
-				struct btrfs_root *root, struct inode *inode)
+noinline int btrfs_update_inode_fallback(struct btrfs_trans_handle *trans,
+					 struct btrfs_root *root,
+					 struct inode *inode)
 {
 	int ret;
 

fs/btrfs/ioctl.c

@@ -343,7 +343,8 @@ static noinline int btrfs_ioctl_fitrim(struct file *file, void __user *arg)
 		return -EOPNOTSUPP;
 	if (copy_from_user(&range, arg, sizeof(range)))
 		return -EFAULT;
-	if (range.start > total_bytes)
+	if (range.start > total_bytes ||
+	    range.len < fs_info->sb->s_blocksize)
 		return -EINVAL;
 
 	range.len = min(range.len, total_bytes - range.start);
@@ -570,7 +571,8 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
 		ret = btrfs_commit_transaction(trans,
 					       root->fs_info->extent_root);
 	}
-	BUG_ON(ret);
+	if (ret)
+		goto fail;
 
 	ret = pending_snapshot->error;
 	if (ret)

fs/btrfs/qgroup.c

@@ -790,8 +790,10 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
 	}
 
 	path = btrfs_alloc_path();
-	if (!path)
-		return -ENOMEM;
+	if (!path) {
+		ret = -ENOMEM;
+		goto out_free_root;
+	}
 
 	key.objectid = 0;
 	key.type = BTRFS_QGROUP_STATUS_KEY;
@@ -800,7 +802,7 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
 	ret = btrfs_insert_empty_item(trans, quota_root, path, &key,
 				      sizeof(*ptr));
 	if (ret)
-		goto out;
+		goto out_free_path;
 
 	leaf = path->nodes[0];
 	ptr = btrfs_item_ptr(leaf, path->slots[0],
@@ -818,8 +820,15 @@ int btrfs_quota_enable(struct btrfs_trans_handle *trans,
 	fs_info->quota_root = quota_root;
 	fs_info->pending_quota_state = 1;
 	spin_unlock(&fs_info->qgroup_lock);
-out:
+out_free_path:
 	btrfs_free_path(path);
+out_free_root:
+	if (ret) {
+		free_extent_buffer(quota_root->node);
+		free_extent_buffer(quota_root->commit_root);
+		kfree(quota_root);
+	}
+out:
 	return ret;
 }