Re: [kitten] New Version Notification for draft-kaduk-kitten-gss-loop-02.txt (fwd)

[Nico Williams <[e-mail]>]

On Fri, Jan 17, 2014 at 9:28 PM, Greg Hudson <[e-mail]> wrote:
> On 01/17/2014 08:01 PM, Martin Rex wrote:
>>>> /* It is safe to call gss_release_buffer twice on the same buffer. */
>
>> I hadn't notice this strange comment about gss_release_buffer() before,
>> and I'm voilently opposed, please remove that misleading comment.
>
> The comment in the draft is trying to say that it's okay to call
> gss_release_buffer twice with the same reference to a gss_buffer_desc
> structure, trusting that the first call will put the structure into a
> state where the second call is a no-op.  Unfortunately, RFC 2744 does
> not appear to guarantee this; 5.26 only requires that the length is set
> to 0 after the storage is freed (implementations are merely "encouraged"
> to zero the pointer field), and does not guarantee that a second call
> won't double-free the storage.

RFC2744, section 5.26 requires nothing.  It doesn't say "MUST", and so
on (because RFC2744 doesn't reference RFC2119, of course).  It doesn't
say that the value field will not be set.  Section 3.9.2 seems to
indicate that the value field ought to be set, but only if the buffer
is an output buffer.

> If it is okay for the sample code to rely on this behavior, in spite of
> RFC 2744 not specifying that it is okay, then perhaps the comment should
> be clarified or removed to avoid confusing people like it did Martin.

I would just note that RFC2744 can be interpreted as permitting
gss_release_buffer() implementations to set the length field but not
the value field.  But I'd rather just say that RFC2744 just has a
glaring omission... because that's what it is.

Nico
--