Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.32 as permitted sender) client-ip=23.128.96.32;
MIME-Version: 1.0
References: <1697056244-21888-1-git-send-email-haiyangz@microsoft.com>
In-Reply-To: <1697056244-21888-1-git-send-email-haiyangz@microsoft.com>
From:   Eric Dumazet <edumazet@google.com>
Date:   Mon, 16 Oct 2023 13:40:43 +0200
Message-ID: <CANn89iLth-thO7=V=b+3dbP=K-m+hbBk75FtM+7cFiUphGXwoA@mail.gmail.com>
Subject: Re: [PATCH net-next,v3] tcp: Set pingpong threshold via sysctl
To:     Haiyang Zhang <haiyangz@microsoft.com>
Cc:     linux-hyperv@vger.kernel.org, netdev@vger.kernel.org,
        kys@microsoft.com, davem@davemloft.net, kuba@kernel.org,
        pabeni@redhat.com, corbet@lwn.net, dsahern@kernel.org,
        ncardwell@google.com, ycheng@google.com, kuniyu@amazon.com,
        morleyd@google.com, mfreemon@cloudflare.com, mubashirq@google.com,
        linux-doc@vger.kernel.org, weiwan@google.com,
        linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Wed, Oct 11, 2023 at 10:31=E2=80=AFPM Haiyang Zhang <haiyangz@microsoft.=
com> wrote:
>
> TCP pingpong threshold is 1 by default. But some applications, like SQL D=
B
> may prefer a higher pingpong threshold to activate delayed acks in quick
> ack mode for better performance.
>

...

>
> diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
> index f207712eece1..7d0fe76d56ef 100644
> --- a/net/ipv4/tcp_output.c
> +++ b/net/ipv4/tcp_output.c
> @@ -170,10 +170,10 @@ static void tcp_event_data_sent(struct tcp_sock *tp=
,
>         tp->lsndtime =3D now;
>
>         /* If it is a reply for ato after last received
> -        * packet, enter pingpong mode.
> +        * packet, increase pingpong count.
>          */
>         if ((u32)(now - icsk->icsk_ack.lrcvtime) < icsk->icsk_ack.ato)
> -               inet_csk_enter_pingpong_mode(sk);
> +               inet_csk_inc_pingpong_cnt(sk);
>  }
>
>  /* Account for an ACK we sent. */

OK, but I do not think we solved the fundamental problem of using
jiffies for this heuristic,
especially for HZ=3D100 or HZ=3D250 builds.

Reviewed-by: Eric Dumazet <edumazet@google.com>