Re: Large Frames, Continuations, Flow Control, and changing HPACK

• From: Jeff Pinner <jpinner@twitter.com>
• Date: Tue, 8 Jul 2014 09:29:38 -0700
• To: Tatsuhiro Tsujikawa <tatsuhiro.t@gmail.com>
• Cc: HTTP Working Group <ietf-http-wg@w3.org>
• Message-ID: <CA+pLO_h3ZtrOmBt+OtCU9Urykbwgyy9jNaVS0qb8eaam2R=3mw@mail.gmail.com>

> If I understand the proposal correctly, it still needs contagious
> HEADERS+CONTINUATION if >16K header field is sent (assuming the proposal
> sticks to 16K frame size).
> For flow control, we still have difficulty to encode (or split) headers in
> arbitrary length, dead locking issue still exist.

For a header set larger than 16K compressed, you can divide it into
multiple HEADERS frames and interleave them at will, the last HEADERS
frame needs to have the END_HEADERS flag set.

To send a single header field larger than 16K compressed, we would
need to increase the frame size. I suggest we handle this by
un-reserving the first two bits in the length field so that frames can
be 64K in size. This would put a "hard-limit" on the size of
individual header fields.

Received on Tuesday, 8 July 2014 17:28:44 UTC