Re: [IPsec] Early code point assignment (was: I-D Action: draft-ietf-ipsecme-chacha20-poly1305-00.txt)

[Tero Kivinen <[e-mail]>]

Paul Wouters writes:
> On Tue, 31 Mar 2015, Yoav Nir wrote:
> >> Can we make it ENCR_CHACHA20_POLY1305 ?
> >> Currently IANA mostly has no CamelCase and it is kind of a C habbit to
> >> have defines in CAPS :)
> >
> > I guess, although some of the names don’t quite work as
> > identifiers: "AES-GCM with a 8 octet ICV” or "ENCR-AES-CCM_12”,
> > which is surprisingly different in terms of dashes vs underscores
> > than “ENCR_AES-CCM_8”. 

How has that happened. Hmmm... looking at the archives that was there
from the beginning. And those assignments which were done at the same
time as RFC4306 was published (i.e. RFC4309 and RFC4106) never went
through the IANA Expert review, hey were simply added to registry
without asking from anybody...

This was the time when we had some communications issues between IANA
and experts... 

> And I've poked both PaulH and Tero to never ever add more names with a
> "-" in it because that has to be replaced by "_" in C.

As an IANA expert, I try maintain the registry so it stays consistent,
but unfortunately when the registry already had those ENCR_AES-CCM_8,
ENCR-AES-CCM_* and AES-GCM with * octet ICV values already in before
expert was every consulted, those are still there.

The AEs-GCM values have annoyed me, as they are only ones without
ENCR_* prefix. I have never really noticed that ENCR-AES-CCM had - not
_ in between ENCR-AES, the one between AES-CCM I have noticed, but
haven't bothered to fix those yet.

On the other hand all of these are just strings so they do not affect
protocol. 

> My apologies I wasn't very active at the time this happened :)

If people feel it would be better to fix those, we can do that, i.e.
change:

14 	ENCR_AES-CCM_8
15 	ENCR-AES-CCM_12
16 	ENCR-AES-CCM_16
18 	AES-GCM with a 8 octet ICV
19 	AES-GCM with a 12 octet ICV
20 	AES-GCM with a 16 octet ICV
25 	ENCR_CAMELLIA_CCM with an 8-octet ICV
26 	ENCR_CAMELLIA_CCM with a 12-octet ICV
27 	ENCR_CAMELLIA_CCM with a 16-octet ICV

To:

14 	ENCR_AES_CCM_8
15 	ENCR_AES_CCM_12
16 	ENCR_AES_CCM_16
18 	ENCR_AES_GCM with a 8 octet ICV
19 	ENCR_AES_GCM with a 12 octet ICV
20 	ENCR_AES_GCM with a 16 octet ICV
25 	ENCR_CAMELLIA_CCM with an 8-octet ICV
26 	ENCR_CAMELLIA_CCM with a 12-octet ICV
27 	ENCR_CAMELLIA_CCM with a 16-octet ICV

Or even change the "n-octet" term to be consistent:

14 	ENCR_AES_CCM_8
15 	ENCR_AES_CCM_12
16 	ENCR_AES_CCM_16
18 	ENCR_AES_GCM with a 8-octet ICV
19 	ENCR_AES_GCM with a 12-octet ICV
20 	ENCR_AES_GCM with a 16-octet ICV
25 	ENCR_CAMELLIA_CCM with an 8-octet ICV
26 	ENCR_CAMELLIA_CCM with a 12-octet ICV
27 	ENCR_CAMELLIA_CCM with a 16-octet ICV

Or even go wild and change them:

14 	ENCR_AES_CCM_8
15 	ENCR_AES_CCM_12
16 	ENCR_AES_CCM_16
18 	ENCR_AES_GCM_8
19 	ENCR_AES_GCM 12
20 	ENCR_AES_GCM_16
25 	ENCR_CAMELLIA_CCM_8
26 	ENCR_CAMELLIA_CCM_12
27 	ENCR_CAMELLIA_CCM_16
-- 
[e-mail]