[quicwg/base-drafts] Loss-detection happens after CC-increase, leading to higher congestion window (#3495)

Christoph Paasch <notifications@github.com> Tue, 03 March 2020 22:10 UTC

Return-Path: <noreply@github.com>
X-Original-To: quic-issues@ietfa.amsl.com
Delivered-To: quic-issues@ietfa.amsl.com
Received: from localhost (localhost []) by ietfa.amsl.com (Postfix) with ESMTP id 75E3D3A0AFE for <quic-issues@ietfa.amsl.com>; Tue, 3 Mar 2020 14:10:57 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -1.696
X-Spam-Status: No, score=-1.696 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_IMAGE_ONLY_28=1.404, HTML_MESSAGE=0.001, MAILING_LIST_MULTI=-1, SPF_HELO_NONE=0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: ietfa.amsl.com (amavisd-new); dkim=pass (1024-bit key) header.d=github.com
Received: from mail.ietf.org ([]) by localhost (ietfa.amsl.com []) (amavisd-new, port 10024) with ESMTP id tDwNa-_bYKHk for <quic-issues@ietfa.amsl.com>; Tue, 3 Mar 2020 14:10:54 -0800 (PST)
Received: from out-27.smtp.github.com (out-27.smtp.github.com []) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id C5D553A0AEF for <quic-issues@ietf.org>; Tue, 3 Mar 2020 14:10:52 -0800 (PST)
Date: Tue, 03 Mar 2020 14:10:51 -0800
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=github.com; s=pf2014; t=1583273451; bh=qq5NXvWj7O7Oi1EbHUK9GmcnB9nJbHlYtFtX7xvq0qk=; h=Date:From:Reply-To:To:Cc:Subject:List-ID:List-Archive:List-Post: List-Unsubscribe:From; b=eUHlyHDR1cwDUAsnoLGWWHhMR5RbYqAH1+v+0qDy7VdzQuzXl0/NuvxKtxNr+KczR vYmzyTTcbdHQ/ItiXmH4tl6mrk+K0Ng7hw9AI4YIA6qYMvdBuR7jK3Dr2MEs4BYGg8 XUOwyYVKgsrpnHQqd3BUx7noWI75s5RxJmXBf4bQ=
From: Christoph Paasch <notifications@github.com>
Reply-To: quicwg/base-drafts <reply+AFTOJKYON4QE4N53WRFIVCN4NK3OXEVBNHHCERMSLI@reply.github.com>
To: quicwg/base-drafts <base-drafts@noreply.github.com>
Cc: Subscribed <subscribed@noreply.github.com>
Message-ID: <quicwg/base-drafts/issues/3495@github.com>
Subject: [quicwg/base-drafts] Loss-detection happens after CC-increase, leading to higher congestion window (#3495)
Mime-Version: 1.0
Content-Type: multipart/alternative; boundary="--==_mimepart_5e5ed5eba52bb_45283fea90ccd96c1896b"; charset="UTF-8"
Content-Transfer-Encoding: 7bit
Precedence: list
X-GitHub-Sender: cpaasch
X-GitHub-Recipient: quic-issues
X-GitHub-Reason: subscribed
X-Auto-Response-Suppress: All
X-GitHub-Recipient-Address: quic-issues@ietf.org
Archived-At: <https://mailarchive.ietf.org/arch/msg/quic-issues/mxM2_eXNzAsGKTBobHdd5QWVlWA>
X-BeenThere: quic-issues@ietf.org
X-Mailman-Version: 2.1.29
List-Id: Notification list for GitHub issues related to the QUIC WG <quic-issues.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/quic-issues>, <mailto:quic-issues-request@ietf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/quic-issues/>
List-Post: <mailto:quic-issues@ietf.org>
List-Help: <mailto:quic-issues-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/quic-issues>, <mailto:quic-issues-request@ietf.org?subject=subscribe>
X-List-Received-Date: Tue, 03 Mar 2020 22:10:58 -0000

In the pseudo-code in appendix A.6, the call to ```DetectLostPackets``` is after ```OnPacketAcked```, while the call to ```ProcessECN``` is before that:

  // Process ECN information if present.
  if (ACK frame contains ECN information):
      ProcessECN(ack, pn_space)

  for acked_packet in newly_acked_packets:
    OnPacketAcked(acked_packet.packet_number, pn_space)


This means that if I get an ACK while in slow-start that indicates a packet-loss, the stack will first call ```OnPacketAcked``` (which ends up increasing the congestion-window) and only later enter recovery through ```DetectLostPackets```. Meaning, the congestion-window will actually be bigger than the flight-size that caused the packet-loss.

E.g., if cwnd is 10 while in slow-start and 10 packets were sent. If I get an ACK for 9 packets, indicating a hole, the congestion-window will first increase to 19 and then enter congestion-avoidance and reduce the congestion-window to 13 (assuming Cubic). Thus, the congestion-window is effectively bigger than 10 while it should be smaller than 10 (the flight-size that caused the packet-loss).

Shouldn't we process the loss before ```OnPacketAcked```, the same way ECN is being processed before that?

You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub: