unionfs, tmpfs, and xattrs

[James Morris]

On Tue, 7 Feb 2006, Bill Nottingham wrote:

> The way xattrs appear to work on tmpfs is that the VFS tries the getxattr
> op of tmpfs (which fails, as it doesn't exist), and then does an end-run
> around in the selinux code to get an attribute, as long as you're only
> looking for the security xattr.

What it's doing is checking if the fs can supply a security xattr, and if 
not, allows the kernel to supply one.

> This means that anything on tmpfs can have a xattr retrieved from userspace
> just fine with getxattr(2), but if you try and get it in the kernel via
> 'normal' means (such as the inode's getxattr method), it will fail. This
> breaks tmpfs as part of a unionfs branch pretty badly.
> 
> Why was xattrs-on-tmpfs done this way? It seems somewhat hackish.

So xattrs do not have to be implemented for every type of psudo fs.

What is the upstream status of unionfs?

> I could theoretically patch unionfs to call the vfs method, but... ew.


-- 
James Morris
<jmorris at namei.org>