From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <netdev-owner@vger.kernel.org>
Received: from vulcan.natalenko.name ([104.207.131.136]:41402 "EHLO
        vulcan.natalenko.name" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1751549AbeBRVtG (ORCPT
        <rfc822;netdev@vger.kernel.org>); Sun, 18 Feb 2018 16:49:06 -0500
From: Oleksandr Natalenko <oleksandr@natalenko.name>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Eric Dumazet <edumazet@google.com>,
        Neal Cardwell <ncardwell@google.com>,
        "David S. Miller" <davem@davemloft.net>,
        Netdev <netdev@vger.kernel.org>,
        Yuchung Cheng <ycheng@google.com>,
        Soheil Hassas Yeganeh <soheil@google.com>,
        Jerry Chu <hkchu@google.com>, Dave Taht <dave.taht@gmail.com>
Subject: Re: TCP and BBR: reproducibly low cwnd and bandwidth
Date: Sun, 18 Feb 2018 22:49:02 +0100
Message-ID: <2562547.S27nl9fb2E@natalenko.name>
In-Reply-To: <1518987867.55655.15.camel@gmail.com>
References: <1697118.nv5eASg0nx@natalenko.name> <1518893571.55655.12.camel@gmail.com> <1518987867.55655.15.camel@gmail.com>
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset="UTF-8"
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

Hi.

On ned=C4=9Ble 18. =C3=BAnora 2018 22:04:27 CET Eric Dumazet wrote:
> I was able to take a look today, and I believe this is the time to
> switch TCP to GSO being always on.
>=20
> As a bonus, we get speed boost for cubic as well.
>=20
> Todays high BDP and recent TCP improvements (rtx queue as rb-tree, sack
> coalescing, TCP pacing...) all were developed/tested/maintained with
> GSO/TSO being the norm.
>=20
> Can you please test the following patch ?

Yes, results below:

BBR+fq:
sg on:  6.02 Gbits/sec
sg off: 1.33 Gbits/sec

BBR+pfifo_fast:
sg on:  4.13 Gbits/sec
sg off: 1.34 Gbits/sec

BBR+fq_codel:
sg on:  4.16 Gbits/sec
sg off: 1.35 Gbits/sec

Reno+fq:
sg on:  6.44 Gbits/sec
sg off: 1.39 Gbits/sec

Reno+pfifo_fast:
sg on:  6.36 Gbits/sec
sg off: 1.39 Gbits/sec

Reno+fq_codel:
sg on:  6.41 Gbits/sec
sg off: 1.38 Gbits/sec

While BBR still suffers when fq is not used, disabling sg doesn't bring=20
drastic throughput drop anymore. So, looks good to me, eh?

> Note that some cleanups can be done later in TCP stack, removing lots
> of legacy stuff.
>=20
> Also TCP internal-pacing could benefit from something similar to this
> fq patch eventually, although there is no hurry.
> https://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next.git/commit=
/?i
> d=3Dfefa569a9d4bc4b7758c0fddd75bb0382c95da77 =20

=46eel free to ping me if you have something else to test then ;).

> Of course, you have to consider why SG was disabled on your device,
> this looks very pessimistic.

Dunno why that happens, but I've managed to just enable it automatically on=
=20
interface up.

Thanks.

Oleksandr