Re: [PATCH 2/7] nfsd: Ensure stateids remain unique until they are freed

[Jeff Layton <jeff.layton@primarydata.com>]

On Wed, 16 Jul 2014 06:49:45 -0700
Christoph Hellwig <hch@infradead.org> wrote:

> On Wed, Jul 16, 2014 at 08:17:38AM -0400, Jeff Layton wrote:
> > The problem with keeping them in the caller is that it defeats the main
> > purpose of the patch. The idea here is to make sure that we keep the
> > stateids in the IDR tree for as long as possible to help ensure
> > uniqueness. If we keep it in the caller, we're still removing the stateids
> > from the IDR hash earlier than we should, and that's a larger potential
> > for collisions.
> 
> I don't understand how that has anything to do with doing the call
> in either nfs4_free_stid or it's two caller.  For the open and lock
> stateids nothing changes in either case, and for delegations we
> move from destroy_(revoked_)delegation to the final put in
> nfs4_put_delegation for both variants.  While I'd normally prefer
> what you do in the patch it just seems like churn with the further
> changes.
> 

Ahh ok. I misunderstood what you were suggesting. Moving it into the
actual callers of nfs4_free_stid is fine.

Unfortunately, that creates a pile of merge conflicts in the later
patches. I'll just go ahead and change that and resend all of the
delegation patches that Bruce hasn't merged yet. Stay tuned...

-- 
Jeff Layton <jlayton@primarydata.com>