[quicwg/base-drafts] Does HTTP spec mention all frames are flow controlled? (#4057)

Lucas Pardue <notifications@github.com> Wed, 26 August 2020 17:14 UTC

Date: Wed, 26 Aug 2020 10:14:34 -0700
From: Lucas Pardue <notifications@github.com>
Reply-To: quicwg/base-drafts <reply+AFTOJK2U2AA3MLA5BYGVQW55KJ4XVEVBNHHCR2XNIE@reply.github.com>
To: quicwg/base-drafts <base-drafts@noreply.github.com>
Cc: Subscribed <subscribed@noreply.github.com>
Message-ID: <quicwg/base-drafts/issues/4057@github.com>
Subject: [quicwg/base-drafts] Does HTTP spec mention all frames are flow controlled? (#4057)
Mime-Version: 1.0
Content-Type: multipart/alternative; boundary="--==_mimepart_5f46987aaac20_7cff196488141"; charset="UTF-8"
Content-Transfer-Encoding: 7bit
Precedence: list
Archived-At: <https://mailarchive.ietf.org/arch/msg/quic-issues/VIhk9JD0JvQoJGs4A_vfg_6QWhk>

In HTTP/2, only DATA frames are flow controlled. In HTTP/3, all frames are sent on STREAM frames, meaning that all frames are flow controlled. This is fine but it might bite people that make wrong assumptions, just like the differences in stream ordering would.

HTTP/3 appendix A provides Considerations for Transitioning from HTTP/2, and includes a section on Frame Types. So I expected to see some coverage about the flow control differences there but I did not.

I'm marking this as editorial because I don't want any change to HTTP/3. Just looking for some exposition.

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/quicwg/base-drafts/issues/4057

[quicwg/base-drafts] Does HTTP spec mention all f… Lucas Pardue
Re: [quicwg/base-drafts] Does HTTP spec mention a… Lars Eggert
Re: [quicwg/base-drafts] Does HTTP spec mention a… Lucas Pardue
Re: [quicwg/base-drafts] Does HTTP spec mention a… Mike Bishop