On 11/30/15 2:54 AM, Short, Todd wrote: > This brings up an interesting point; having a record length that corresponds > to the TCP segment size can help hardware implementations such that they > don't need to deal with scatter/gather; i.e. one TCP segment corresponds to a > single TLS record. This goes along with 8 (or 4) byte record lengths for > hardware implementations.
This is wildly unsafe to assume or depend on for all sorts of reasons; TCP streams regularly get merged or re-segmented in the network in a variety of unpredictable ways (e.g., by middleboxes). See my earlier E-mail discussing that issue extensively. B
smime.p7s
Description: S/MIME Cryptographic Signature
_______________________________________________ TLS mailing list TLS@ietf.org https://www.ietf.org/mailman/listinfo/tls
