netdev - Re: [RFC net-next 4/4] tcp: defer regular ACK while processing socket backlog

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <CANn89iLZF5Me4isAqXe+cptAOwiunMyHVeOy7Xfz_FdhFoiaqw@mail.gmail.com>
Date: Thu, 7 Sep 2023 18:54:01 +0200
From: Eric Dumazet <edumazet@...gle.com>
To: Dave Taht <dave.taht@...il.com>
Cc: "David S . Miller" <davem@...emloft.net>, Jakub Kicinski <kuba@...nel.org>, 
	Paolo Abeni <pabeni@...hat.com>, netdev@...r.kernel.org, eric.dumazet@...il.com, 
	Soheil Hassas Yeganeh <soheil@...gle.com>, Neal Cardwell <ncardwell@...gle.com>, 
	Yuchung Cheng <ycheng@...gle.com>
Subject: Re: [RFC net-next 4/4] tcp: defer regular ACK while processing socket backlog

On Thu, Sep 7, 2023 at 6:50 PM Dave Taht <dave.taht@...il.com> wrote:

> When you once said you were going to find a way to reduce the number of acks, I did not expect this.
> Has this been tested on arm or at lower rates, such as a gigabit? Or against WiFi?
>

Not yet, but my guess is that it could benefit wifi drivers that are
not using GRO,
particularly if the TCP payload is in skb->head instead of a page fragment,
defeating tcp_try_coalesce() ability to 'pretend GRO was there in the
first place'.

I sent an RFC so that others could perform tests ;)

> Regardless:
>
> Acked-by: Dave Taht <dave.taht@...il.com>
>
> --
> Oct 30: https://netdevconf.info/0x17/news/the-maestro-and-the-music-bof.html
> Dave Täht CSO, LibreQos