[tcpm] Re: [CCWG] CWND increase in CUBIC and ACK policy

Kazuho Oku <kazuhooku@gmail.com> Thu, 20 August 2026 21:17 UTC

Return-Path: <kazuhooku@gmail.com>
X-Original-To: tcpm@mail2.ietf.org
Delivered-To: tcpm@mail2.ietf.org
Received: from localhost (localhost [127.0.0.1]) by mail2.ietf.org (Postfix) with ESMTP id 2E6B612D109B0 for <tcpm@mail2.ietf.org>; Thu, 20 Aug 2026 14:17:27 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=ietf.org; s=ietf1; t=1787260647; bh=fsVLzf2tF1VLNNhSdJ5wL0kA5KnnQ9GOEqKfT6++PT4=; h=References:In-Reply-To:From:Date:Subject:To:Cc; b=R/B1xBNO6ASiqnjSfmEpYNH8nt8nsX9S6T7AJZ7uA2JHwL9KghNv3EFCQtzcPn2Ib Pq7b6C8xgDfpu/8jriu8LjMS9Zsz6x4MJdpcsu/J/2xpwfTN43eWFNdE7qJJP7CxB6 Zu9aX/M0e6pcNANqSWJkaSz2UxN6RciV7OF7eks8=
X-Virus-Scanned: amavisd-new at ietf.org
X-Spam-Flag: NO
X-Spam-Score: -2.098
X-Spam-Level:
X-Spam-Status: No, score=-2.098 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001] autolearn=unavailable autolearn_force=no
Authentication-Results: mail2.ietf.org (amavisd-new); dkim=pass (2048-bit key) header.d=gmail.com
Received: from mail2.ietf.org ([166.84.6.31]) by localhost (mail2.ietf.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id uLjuVUFp-Bt7 for <tcpm@mail2.ietf.org>; Thu, 20 Aug 2026 14:17:25 -0700 (PDT)
Received: from mail-pf1-x429.google.com (mail-pf1-x429.google.com [IPv6:2607:f8b0:4864:20::429]) (using TLSv1.3 with cipher TLS_AES_128_GCM_SHA256 (128/128 bits) key-exchange X25519 server-signature ECDSA (P-256) server-digest SHA256) (No client certificate requested) by mail2.ietf.org (Postfix) with ESMTPS id 34EE212D109A1 for <tcpm@ietf.org>; Thu, 20 Aug 2026 14:17:25 -0700 (PDT)
Received: by mail-pf1-x429.google.com with SMTP id d2e1a72fcca58-84f77cc7302so96678b3a.3 for <tcpm@ietf.org>; Thu, 20 Aug 2026 14:17:25 -0700 (PDT)
ARC-Seal: i=1; a=rsa-sha256; t=1787260644; cv=none; d=google.com; s=arc-20260327; b=ENB3AEDPBPUHexKPgRlzHZPlnYT5gxzVDVE6W1wNVspRdBy/N20m+MuGnVveCjAFI3 QD+Ge/t/HJ449GNA8Rs8GmBaL9g3XkX4nxflCptJl7jKwCOdgoERIckGUAkEXo3zdXmb V2x+QRNjZNa68NGFkmeoHZjGFlZD5s+avM5Mtx8qyZdFwmbOtuAFroWnkchCsb5Jbhor Px6foo3Dy922v55+BbRbCi9C7E793MJavmCmvye5+zfDp0isPr1Nytb0KkHzsbc5I/y9 LNWdIXdJlqZkr7sAnvRzf4hDMWzNHTQE+NrfvThxsxvYrqz0ps519eOWdI5bYG3N67zx H75Q==
ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20260327; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:dkim-signature; bh=px4R18aZVIGWniPXbp13AmWzPotJB/cwPWvq3oE3jJk=; fh=3EPXhmtnAYWzG7F+RmAbc5WEKwhZVpnbo2+XGJ+xLuE=; b=MM62bdSJVQIQw75T1Dd5M6Qs/eeF8GVBxFqaIEKwesc0VLqidrOWBdGsjzESy8GQRQ OU8KPqvzL6AHoF6dvbpLiTvVlD3gpxMqoAIRFflsILur0pv/VpoPF1/SgIfiYiCIZkxu Mvo+kmCR3aAjeemxfvP25EsorNfKqmNNgZRkK4NMLR2rgrFOVHx+oWkgu1z0Y3upEUBg k56cou1t0X35TMOgkEAEJylDS2Iv8jDbbG3UCgBjyw3FqzMHY0SSUcDeJPwnj4tmwzwb ZDznNhrgPMQ2ugmI0zLTa+nS7sJKkPa6fvvh0LKL99Q1+1OXmRfTf3o23LXqpQ371FGO ro0g==; darn=ietf.org
ARC-Authentication-Results: i=1; mx.google.com; arc=none
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20251104; t=1787260644; x=1787865444; darn=ietf.org; h=content-type:cc:to:subject:message-id:date:from:in-reply-to :references:mime-version:from:to:cc:subject:date:message-id:reply-to :content-type; bh=px4R18aZVIGWniPXbp13AmWzPotJB/cwPWvq3oE3jJk=; b=PgFTpaWJXeEu/9xj1RoX49mLxB3m7yfrTSAJTExvf5vRsPcSCnFp+ndfLXewnfNLG+ zDHt8yYkokydiLG808ReITzZYflKqJiTSaNIeiOORbEZ6ow7+1TrXB+Tpj1dWewyOrLF Tb2IoQXzyPm0Rhwdqd4hmCR++RHgixGaFcN8+4xOn6yL/tbYH7wJfGkvFbx1cYPoZ3H+ JU0eJBGzxEXNADez4awvZSGnRhz3Le1hc8PRI5VMirXrRz/0aHXsCflWx8nK9/KWE1Ey cab8nu32an5BvvSzqltV6+eVbwE+ZjXjBeqMzU+AQOxDem1ulqgc+unX2u+9WzuMsfCe ODIA==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1787260644; x=1787865444; h=content-type:cc:to:subject:message-id:date:from:in-reply-to :references:mime-version:x-gm-gg:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to:content-type; bh=px4R18aZVIGWniPXbp13AmWzPotJB/cwPWvq3oE3jJk=; b=QAc7b75D4lmGrqKZYrRQnheLhlIitjXV6qXJIkcX79jUh/rUWDePEHlMkvVRcK+lgD oSoHRVbb4vt52zS38poCC4CBg72Aq4DHL1WHwZcc+PVXCBUaQqLrjavBXk0XlsW/+5H9 zq8r+qtvalGL94B3233TUxfddyfJ9pbB0beIQIN+TiuCzM23j5bDT6bfbL8Vln5vf7bg AHbFtB2yOXWiwyyaYqjFzZ9jt6NT6aHXDeCsrLjhCUlmaxLRT9r0zGT/uaNJEAm5FMLk JhCo6aGoVmY+Ho4AbAAmxAjBGO5ANrSh0PgoC9NLiVBahA1hVocLgmkUqX2OThL+i+wZ Iz9g==
X-Forwarded-Encrypted: i=1; AHgh+RpzolX1lf+WRUbp1HhF8yWyRmsQggUoPzXSsxNneFMrBFA5mKdP8YUCX7mjQrggh0vExeky@ietf.org
X-Gm-Message-State: AOJu0Yz5gd1wDC71yd+4dM4ZDO2w12TbaNmWJSznnsaHPd+CS+BLGfwe kUVpB9Z3szLWBJQVPg6s7JyhAQNZ0VGa02F3i0FcfVv58pvs23tRUgD9EzVSlrAQA5N2owLTw0w 3kYzP0OiQDw7rIOHN98lWMBMzylFJhHg=
X-Gm-Gg: AR+sD13jNFhkvrfCp1TJZziGcImAz7eDvAxW+RQlDMwiBlVs5GU0nU8MokxRZBIoY0M h5kXPWcxKMWIRndzrupLuWu9vbOoOTCg0B7NoIQy+m5lMoxy8e1CTbSkrFmtCqJ4a9vIN5tqZ6/ 6FYmmwc8I1xMhB/Ej86IGTanLSg4z820cG/AuxsRboXSymlqCMPoSYdwGug4CQfgzjNF3NudQxi kMPkTm2cjTQIB7E15RPx8fNeF7UI3NBPlXUDF8zgha8FWBJLKqs2kEL6IlIdVHhtyugeotAg+WD Ys19v/hxE5M6Tf0sh4Cn8Uk3NVD+BW3hL3u7oz9Y20AcZuew6MFfW70RQ7ilKLka5ieZbzyYgxp V
X-Received: by 2002:a05:6a20:450c:b0:3bf:6d96:ac40 with SMTP id adf61e73a8af0-3cd3007ce9bmr2639682637.12.1787260644197; Thu, 20 Aug 2026 14:17:24 -0700 (PDT)
MIME-Version: 1.0
References: <CANatvzyZJxOLcwsrb52oyVm3T8ZhqbzPpOWDSvMvSo-b1hW+Ng@mail.gmail.com> <CADVnQymPjzRk-_ULFkz7F4ZkYD1dYizYd2BR_nwP7VbxunqgsA@mail.gmail.com> <CAAK044RXrUHBAZfjUFaKRtspWhOyi0hhXsyqaUt95fJJuvjcrQ@mail.gmail.com>
In-Reply-To: <CAAK044RXrUHBAZfjUFaKRtspWhOyi0hhXsyqaUt95fJJuvjcrQ@mail.gmail.com>
From: Kazuho Oku <kazuhooku@gmail.com>
Date: Fri, 21 Aug 2026 06:17:11 +0900
X-Gm-Features: AcwNN1WExV1-Meom_DQAp5heUCpC8LpvSlMMA4HU9aO0lIvIGY75mTeWBXlej_A
Message-ID: <CANatvzxU=AwSg5thk8GHLAJ0wh4N2RXdvTRs_PjhrcaV5z9NFQ@mail.gmail.com>
To: Yoshifumi Nishida <nsd.ietf@gmail.com>
Content-Type: multipart/alternative; boundary="000000000000d1877f0659810b0b"
Message-ID-Hash: 42XYX43OK7LNJWMYIPKR25Q3ABLRE3SD
X-Message-ID-Hash: 42XYX43OK7LNJWMYIPKR25Q3ABLRE3SD
X-MailFrom: kazuhooku@gmail.com
X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; emergency; loop; banned-address; member-moderation; header-match-tcpm.ietf.org-0; nonmember-moderation; administrivia; implicit-dest; max-recipients; max-size; news-moderation; no-subject; digests; suspicious-header
CC: Neal Cardwell <ncardwell=40google.com@dmarc.ietf.org>, ccwg@ietf.org, IETF QUIC WG <quic@ietf.org>, "tcpm@ietf.org Extensions" <tcpm@ietf.org>
X-Mailman-Version: 3.3.9rc6
Precedence: list
Subject: [tcpm] Re: [CCWG] CWND increase in CUBIC and ACK policy
List-Id: TCP Maintenance and Minor Extensions Working Group <tcpm.ietf.org>
Archived-At: <https://mailarchive.ietf.org/arch/msg/tcpm/J53sMLuJ8ZR7exVbD4y3SwilVBU>
List-Archive: <https://mailarchive.ietf.org/arch/browse/tcpm>
List-Help: <mailto:tcpm-request@ietf.org?subject=help>
List-Owner: <mailto:tcpm-owner@ietf.org>
List-Post: <mailto:tcpm@ietf.org>
List-Subscribe: <mailto:tcpm-join@ietf.org>
List-Unsubscribe: <mailto:tcpm-leave@ietf.org>

Neal, Yoshifumi, Thank you very much for your response!

I'm relieved that my evaluation was correct. I also agree that this
deserves an errata; I will submit one.

It was also great to learn that this bug existed inside Linux until 2015; I
had somehow assumed Linux always did the correct thing 😅

2026年8月18日(火) 18:02 Yoshifumi Nishida <nsd.ietf@gmail.com>:

> Hello,
>
> Added tcpm just in case as RFC9438 was standardized at tcpm wg.
>
> On Mon, Aug 17, 2026 at 6:31 AM Neal Cardwell <ncardwell=
> 40google.com@dmarc.ietf.org> wrote:
>
>>
>>
>> On Sun, Aug 16, 2026 at 9:14 PM Kazuho Oku <kazuhooku@gmail.com> wrote:
>>
>>> Dear CCWG and QUICWG folks,
>>>
>>> TL;DR: Sections 4.4 and 4.5 of RFC 9438 do not account for
>>> segments_acked when increasing CWND outside the Reno-friendly region. As a
>>> result, a literal implementation makes CUBIC's growth rate depend on the
>>> receiver's ACK ratio, and causes measurable unfairness. It gets
>>> increasingly worse as ACKs become less frequent.
>>>
>>> While testing a vibe-coded CUBIC implementation, I think I've found a
>>> peculiarity in RFC 9438.
>>>
>>> Section 4.1 defines a "new ACK" as
>>>
>>> > a "new
>>> > ACK" is received, i.e., an ACK that cumulatively acknowledges the
>>> > delivery of previously unacknowledged data.
>>>
>>> Then, figure 4 in section 4.3 uses segments_acked to determine the
>>> amount being acked, which is perfectly fine:
>>>
>>> > West = West + αcubic * segments_acked / cwnd
>>>
>>> It even notes that accounting for segments_acked is needed "for
>>> connections with enabled or disabled delayed ACKs".
>>>
>>> But in sections 4.4 and 4.5, the CWND increase is specified as:
>>>
>>> > (target - cwnd) / cwnd
>>>
>>> where the target is Wcubic(t + RTT). It does not take segments_acked
>>> into account. This update rule approximates the intended increase toward
>>> `target` over one RTT only when there are roughly `cwnd` ACK events per
>>> RTT, i.e., approximately one ACK per segment.
>>>
>>> Consequently, as ACKs become less frequent, the actual CWND increasingly
>>> trails the CUBIC window function.
>>>
>>> On a simulated 100 Mbps bottleneck with a full RTT of 80 ms, when
>>> competing against an otherwise identical CUBIC flow using ACK1 (i.e.,
>>> ack-every-packet), the bandwidth that a CUBIC flow doing ACKn (i.e., ack
>>> every n packet) obtains is:
>>>
>>> ACK1: 50.0%
>>> ACK2: 45.29%    <-- 👀
>>> ACK10: 10.07%
>>>
>>> ACK2 is particularly interesting because acknowledging at least every
>>> second packet is the conventional delayed-ACK behavior recommended by both
>>> TCP and QUIC v1.
>>>
>>> Changing the increase to:
>>>
>>> > (target - cwnd) / cwnd * segments_acked
>>>
>>> fixes the issue.
>>>
>>> Perhaps unsurprisingly, this issue does not exist in the Linux TCP stack
>>> or Google Quiche. Both account for the amount of newly acknowledged
>>> segments. I see some QUIC stacks also doing the correct thing, especially
>>> those that seem to have used CUBIC inside linux or Google Quiche as a
>>> reference.
>>>
>>> But there also seems to be multiple QUIC stacks that contain this bug.
>>>
>>> To paraphrase, under stable conditions and the receiver using the
>>> default ACK2 policy, a literal implementation of RFC 9438 gets about 17%
>>> less throughput than a stack that does not have this bug.
>>>
>>> Attempts to reduce ACKs (e.g., draft-ietf-quic-ack-frequency) are
>>> going to make the downside more prominent.
>>>
>>> Am I missing something? Please let me know what you think.
>>>
>>
>> Thanks for the post! Your analysis looks correct.
>>
>
> I agree. Thanks for the finding.
>
>
>> In the Linux TCP CUBIC implementation we fixed this "stretch ACKs" issue
>> in 2015 here:
>>
>> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=9cd981dcf174d26805a032aefa791436da709bee
>>
>> IMHO this is worth adding an errata, if we don't yet have one for this
>> issue.
>>
>
> Also, +1 for this.
> --
> Yoshi
>
>>

-- 
Kazuho Oku