- From: Poul-Henning Kamp <phk@phk.freebsd.dk>
- Date: Tue, 02 Sep 2014 21:36:26 +0000
- To: Patrick McManus <mcmanus@ducksong.com>
- cc: "Roy T. Fielding" <fielding@gbiv.com>, HTTP Working Group <ietf-http-wg@w3.org>
-------- In message <CAOdDvNognsLEv_M0ngYz4jiNPWXXhyBPFRFLnthWeARG1t1zzQ@mail.gmail.com> , Patrick McManus writes: >But before anyone gets too excited, overall I personally don't think the >need to either align or pack the header into 64bits is very important given >the rather unaligned nature of the payloads and so I don't favor reopening >the issue. For what it is worth, I don't think trying to align the frames makes any sense, because there is no way we can guarantee that they will be aligned into underlying TCP and IP packets in the first place. Things like keeping the :headers first, maybe even in strictly defined order like HTTP/1.1 does it will be much more important for high performance implementations. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk@FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD committer | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence.
Received on Tuesday, 2 September 2014 21:36:51 UTC