W3C home > Mailing lists > Public > ietf-http-wg@w3.org > July to September 2014

Re: h2 frame layout

From: Patrick McManus <mcmanus@ducksong.com>
Date: Tue, 2 Sep 2014 17:25:31 -0400
Message-ID: <CAOdDvNognsLEv_M0ngYz4jiNPWXXhyBPFRFLnthWeARG1t1zzQ@mail.gmail.com>
To: "Roy T. Fielding" <fielding@gbiv.com>
Cc: HTTP Working Group <ietf-http-wg@w3.org>
Hi Roy,

On Fri, Aug 29, 2014 at 6:49 PM, Roy T. Fielding <fielding@gbiv.com> wrote:

> This comment is in reference to section 4.1 of
>   http://tools.ietf.org/id/draft-ietf-httpbis-http2-14.txt
> > 4.1.  Frame Format
> >
> >    All frames begin with a fixed 9-octet header followed by a variable-
> >    length payload.
> [..]
> I think this has become a bit moth-eaten over time as the WG has
> adjusted field lengths and perhaps added or removed a few.

That time was quite recent:  from -13 to -14 as the consensus was to
increase the payload length from 16 to 24 bits maintaining a requirement
for a fixed 10 0 bits unless otherwise negotiated via settings.

I was never a fan of that - if we were to decide that a concise 64 bits is
necessary then I think we should revert that length field increase rather
than giving up the type and flag space extensions may want to use. The
extended length users could operate via extension.

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. The 8 vs 9 byte header was part of the issue discussion already
- as can be seen here
http://lists.w3.org/Archives/Public/ietf-http-wg/2014JulSep/0706.html when
a proposal reallocating 8 bits from stream ID instead of type and flags so
I don't feel like this is unconsidered information when making that

Received on Tuesday, 2 September 2014 21:25:58 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 30 March 2016 09:57:10 UTC