Re: [quicwg/base-drafts] Bring back AEAD_AES_128_CCM_8 now that we pad the plaintext (#2581)

[ekr <[e-mail]>]

@DavidSchinazi this doesn't seem like a very persuasive analogy, given that HTTP/2 has a long list of rules about disabling ciphers that were present in TLS 1.2:

https://tools.ietf.org/rfcmarkup?doc=7540#section-9.2.2

The reason that it doesn't have that in TLS 1.3 is because (1) TLS 1.3 post-dates HTTP/2 and (2) the suites were chosen according to the same principles

Second, as a practical matter, most browsers do not support CCM_8, and I suspect that will continue to be true for most stacks, partly because performance is not awesome (as it requires computing CBC). If, as @RyanAtGoogle implies, we want to have a cipher with a shorter authentication tag (I'm undecided on this topic myself at present) for general use then we should consult CFRG for one, rather than just use the one that happened to be the one that IoT devices supported.




-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/quicwg/base-drafts/issues/2581#issuecomment-493439412