XFS in mainline 2.4? (was Re: [PATCH] d_alloc_anon for 2.4.21-pre6)

[Nicholas Wourms <nwourms@xxxxxxxxxxxxx>]

Christoph Hellwig wrote:
> [$BIGNUM repost since August 2002, still zero feedback]
>
> The rewritten 2.5 nfsd export handling introduce a funcion,
> d_alloc_anon, to get an dentry for a given inode, either taking
> a well-connected one or allocating a new one.
>
> In 2.4 we have the functionality that it does in 2.4 duplicated over
> nfsd and all filesystems having their own fh_to_dentry method, and
> XFS needs even more instances of this for other handle to dentry
> conversations.

Speaking of XFS, would it be possible to merge it into the 
2.4.X mainline any time soon?  Despite what Alan says, XFS 
doesn't touch any more core code then his ide rewrite has. 
Also, the sendfile64 syscall modified core code as well. 
So, I don't think he or anyone else can get away with 
rejecting it based on it modifying core code.  JFS has been 
in there for a few releases now, and I'd argue that there 
are probably more XFS users then JFS users.  Since it was 
accepted into 2.5 awhile back, it seems appropriate (to me) 
that it eventually go into 2.4.  IT would also have the side 
benefit of keeping the XFS code more (or less) in sync with 
what's in bk.  Just a thought....

Cheers,
Nicholas