A bunch of editorial cleanup. Short summary: #438
Conversation
- Remove a bunch of the "can be used" language because we are specifying how to use it. - Update the 2119 language. - Clean up the description of key derivation, also shorten it. - Misc other stuff.
draft-ietf-quic-tls.md
Outdated
| selected by TLS. The key length is the AEAD key size. As defined in Section | ||
| 5.3 of {{!I-D.ietf-tls-tls13}}, the IV length is the larger of 8 or N_MIN (see | ||
| Section 4 of {{!RFC5116}}). For any secret S, the corresponding key and | ||
| IV are derived as shown below: |
There was a problem hiding this comment.
There is an above, but not below.
I think that you mean to move the above code block to below this paragraph and remove the "as shown below:" from the previous one.
| NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", | ||
| "MAY", and "OPTIONAL" in this document are to be interpreted as | ||
| described in BCP 14 {{!RFC2119}}, {{!I-D.leiba-rfc2119-update}} when, and only when, they | ||
| appear in all capitals, as shown here. |
There was a problem hiding this comment.
Does this mean you don't want this change, or you are just annoyed that there is a draft on this?
| This arrangement means that some TLS messages receive redundant protection from | ||
| both the QUIC packet protection and the TLS record protection. These messages | ||
| are limited in number; the TLS connection is rarely needed once the handshake | ||
| completes. |
There was a problem hiding this comment.
I'd kinda like to keep something like this. We still double-encrypt NST and people will complain (or get it wrong) if we don't explain it somewhere.
There was a problem hiding this comment.
This point is made in #packet-protection
draft-ietf-quic-tls.md
Outdated
|
|
||
| ~~~ | ||
| info = (HashLen / 256) || (HashLen % 256) || 0x21 || | ||
| "TLS 1.3, QUIC client 1-RTT secret" || 0x00 | ||
| ~~~ | ||
|
|
||
|
|
There was a problem hiding this comment.
Please keep extra space between sections.
|
@martinthomson The reason that Circle is sad here is an overlong line that should be already fixed in trunk and so should go away on merge. |
|
Merged by hand. |
Remove a bunch of the "can be used" language because we are
specifying how to use it.
Update the 2119 language.
Clean up the description of key derivation, also shorten
it.
Misc other stuff.