Re: possible module refcount leak with auth_gss

["J. Bruce Fields" <bfields@fieldses.org>]

On Wed, Dec 17, 2008 at 02:34:58PM -0500, Jeff Layton wrote:
> On Wed, 17 Dec 2008 14:20:47 -0500
> "J. Bruce Fields" <bfields@fieldses.org> wrote:
> > We certainly shouldn't be really refreshing the cred--that'd end up
> > creating a new gss context when what we're trying to do is destroy one.
> > So leaving cr_ops set to gss_credops() doesn't sound right.
> > 
> > Maybe that gss_refresh_null() should just return 0 and pretend the
> > cred's fine, instead of returning -EACCES?
> > 
> 
> Possibly -- it does look like this is the only place that those credops
> are used.
> 
> What's the reasoning behind clearing the RPCAUTH_CRED_UPTODATE bit
> here? If we don't want to refresh the cred, would it be better to just
> leave it set?

It looks like the CRED_UPTODATE is just being used as a signal to decide
whether the PROC_DESTROY has already been sent; the test_and_clear_bit
ensures that the null call is made just once.  Maybe a separate flag
would be simpler.

This is Trond's code, so he should probably comment.

--b.