Regarding Wrapfs Code - function wrapfs_interpose

Regarding Wrapfs Code - function wrapfs_interpose

[UZAIR LAKHANI]

Hello All,

I want to find out the purpose of a part of Wrapfs
code.

The function name is wrapfs_interpose

The code is

/* check that the lower file system didn't cross a
mount point */
	if (lower_inode->i_sb != SUPERBLOCK_TO_LOWER(sb)) {
		err = -EXDEV;
		goto out;
	}

I want to find out what is the purpose of the above
check.

Thanks,

Uzair Lakhnai
Karachi, Pakistan


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

Re: Regarding Wrapfs Code - function wrapfs_interpose

[Avishay Traeger]

On Fri, 2006-04-07 at 21:51 -0700, UZAIR LAKHANI wrote:
> I want to find out the purpose of a part of Wrapfs
> code.
> 
> The function name is wrapfs_interpose
> 
> The code is
> 
> /* check that the lower file system didn't cross a
> mount point */
> 	if (lower_inode->i_sb != SUPERBLOCK_TO_LOWER(sb)) {
> 		err = -EXDEV;
> 		goto out;
> 	}
> 
> I want to find out what is the purpose of the above
> check.

Exactly what the comment says - it checks to make sure that we didn't
cross a mount point.  It does this by making sure that the lower-level
superblock of this lower-level inode is the same as the lower-level
superblock that we expect (the one used at mount-time).  So basically,
if we see a superblock that is different from the one we usually use, we
crossed a mount point and so we return an error.

Avishay Traeger
http://www.fsl.cs.sunysb.edu/~avishay/

Re: Regarding Wrapfs Code - function wrapfs_interpose

[Erez Zadok]

In message <1144501811.13374.13.camel@ool-44c32f98.dyn.optonline.net>, Avishay Traeger writes:
> On Fri, 2006-04-07 at 21:51 -0700, UZAIR LAKHANI wrote:
> > I want to find out the purpose of a part of Wrapfs
> > code.
> > 
> > The function name is wrapfs_interpose
> > 
> > The code is
> > 
> > /* check that the lower file system didn't cross a
> > mount point */
> > 	if (lower_inode->i_sb != SUPERBLOCK_TO_LOWER(sb)) {
> > 		err = -EXDEV;
> > 		goto out;
> > 	}
> > 
> > I want to find out what is the purpose of the above
> > check.
> 
> Exactly what the comment says - it checks to make sure that we didn't
> cross a mount point.  It does this by making sure that the lower-level
> superblock of this lower-level inode is the same as the lower-level
> superblock that we expect (the one used at mount-time).  So basically,
> if we see a superblock that is different from the one we usually use, we
> crossed a mount point and so we return an error.
> 
> Avishay Traeger
> http://www.fsl.cs.sunysb.edu/~avishay/

More detail.

In its default mode, Wrapfs generates inode numbers by inheriting the inum
of the lower inode: that way we are guaranteed unique inodes, and a fast
operation to generate inodes.  But if you cross into a new sb, you could
clash with other inodes that have the same inum (confusing the icache) -- so
the default alg doesn't work unless we prohibit you from crossing into lower
sb's.

Wrapfs supports therefore another mode in which we use iunique() to generate
unique inums.  In this mode you *can* cross into new sb's, but the
disadvantage now is that your inums are not persistent (this can mess up
certain tools like tar/find, and NFS).

In Unionfs we implemented yet another alternative (called "imap" or Inode
Map).  We generate unique inums, but we store them persistently on the lower
f/s, in a special mapping file.  That way we can map our own stacked inums
to the lower ones.  So this allows you to cross into any number of sb's, and
it works with NFS (client/server) and with tools like tar and find.  Problem
is, performance now suffers each time you have to consult the i-mapping file
to translate b/t upper and lower inums.

Erez.