ocfs2: manually d_move() during ocfs2_rename() (1ba9da2f) · Commits · 戴 / test

Make use of FS_RENAME_DOES_D_MOVE to avoid a race condition that can occur during ->rename() if we d_move() outside of the parent directory cluster locks, and another node discovers the new name (created during the rename) and unlinks it. d_move() will unconditionally rehash a dentry - which will leave stale data in the system. Signed-off-by:

Mark Fasheh <mark.fasheh@oracle.com>

fs/ocfs2/dcache.c

+4 −1

Original line number	Diff line number	Diff line
		@@ -414,7 +414,7 @@ void ocfs2_dentry_move(struct dentry dentry, struct dentry target,
		* XXX: Is there any advantage to dropping the lock here?
		*/
		if (old_dir == new_dir)
		return;
		goto out_move;

		ocfs2_dentry_lock_put(osb, dentry->d_fsdata);

		@@ -423,6 +423,9 @@ void ocfs2_dentry_move(struct dentry dentry, struct dentry target,
		OCFS2_I(new_dir)->ip_blkno, 0);
		if (ret)
		mlog_errno(ret);

		out_move:
		d_move(dentry, target);
		}

		struct dentry_operations ocfs2_dentry_ops = {

fs/ocfs2/super.c

+1 −1

Original line number	Diff line number	Diff line
		@@ -682,7 +682,7 @@ static struct file_system_type ocfs2_fs_type = {
		.kill_sb = kill_block_super, /* set to the generic one
		* right now, but do we
		* need to change that? */
		.fs_flags = FS_REQUIRES_DEV,
		.fs_flags = FS_REQUIRES_DEV\|FS_RENAME_DOES_D_MOVE,
		.next = NULL
		};

Admin message