Re: [PATCH 5/5] union: hybrid union filesystem prototype

[Valerie Aurora <vaurora@redhat.com>]

On Thu, Aug 26, 2010 at 08:33:45PM +0200, Miklos Szeredi wrote:
> From: Miklos Szeredi <mszeredi@suse.cz>
> 
> This union filesystem is a hybrid of entirely filesystem based
> (unionfs, aufs) and entierly VFS based (union mounts) solutions.
> 
> The dentry tree is duplicated from the underlying filesystems, this
> enables fast cached lookups without adding special support into the
> VFS.  This uses slightly more memory than union mounts, but dentries
> are relatively small.
> 
> Inode structures are only duplicated for directories.  Regular files,
> symlinks and special files each share a single inode.  This means that
> locking victim for unlink is a quasi-filesystem lock, which is
> suboptimal, but could be worked around in the VFS.
> 
> Opening non directories results in the open forwarded to the
> underlying filesystem.  This makes the behavior very similar to union
> mounts (with the same limitations vs. fchmod/fchown on O_RDONLY file
> descriptors).
> 
> Usage:
> 
>   mount -t union -olowerdir=/union/lower,upperdir=/union/upper union /mnt/union
> 
> Supported:
> 
>  - all operations
> 
> Missing:
> 
>  - upgrade credentials for copy-up
>  - ensure that filesystems part of the union are not modified outside
>    the union

Just a note that the infrastructure I wrote to do this last bullet
point for union mounts (hard read-only count plus mount checks) is
completely compatible with hybrid union fs.

-VAL