Netdev Archive on lore.kernel.org
help / color / mirror / Atom feed
From: "Íñigo Huguet" <ihuguet@redhat.com>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Edward Cree <ecree.xilinx@gmail.com>,
	habetsm.xilinx@gmail.com, netdev@vger.kernel.org,
	Dinan Gunawardena <dinang@xilinx.com>,
	Pablo Cascon <pabloc@xilinx.com>
Subject: Re: Bad performance in RX with sfc 40G
Date: Thu, 2 Dec 2021 15:26:25 +0100	[thread overview]
Message-ID: <CACT4oufaWTsvHEwi4vxcPtJLocGD2HZAaZfEMSei4ESF1amsnQ@mail.gmail.com> (raw)
In-Reply-To: <beef3b28-6818-df7b-eaad-8569cac5d79b@gmail.com>

Hi,

On Thu, Nov 18, 2021 at 6:19 PM Eric Dumazet <eric.dumazet@gmail.com> wrote:
> You could try to :
>
> Make the RX ring buffers bigger (ethtool -G eth0 rx 8192)
>
> and/or
>
> Make sure your tcp socket receive buffer is smaller than number of frames in the ring buffer
>
> echo "4096 131072 2097152" >/proc/sys/net/ipv4/tcp_rmem

Thanks for the suggestion. Sadly, the results are almost the same.

> You can also try latest net-next, as TCP got something to help this case.
>
> f35f821935d8df76f9c92e2431a225bdff938169 tcp: defer skb freeing after socket lock is released

I'll try ASAP, but I will need some days. Thanks



-- 
Íñigo Huguet


  reply	other threads:[~2021-12-02 14:26 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-11-18 15:14 Íñigo Huguet
2021-11-18 17:19 ` Eric Dumazet
2021-12-02 14:26   ` Íñigo Huguet [this message]
2021-11-20  8:31 ` Martin Habets
2021-12-09 12:06   ` Íñigo Huguet
2021-12-23 13:18     ` Íñigo Huguet
2022-01-02  9:22       ` Martin Habets
2022-01-10  8:58         ` [PATCH net-next] sfc: The size of the RX recycle ring should be more flexible Martin Habets
2022-01-10  9:31           ` Íñigo Huguet
2022-01-12  9:05             ` Martin Habets
2022-01-31 11:08             ` Martin Habets
2022-01-10 17:22           ` Jakub Kicinski
2022-01-12  9:08             ` Martin Habets
2022-01-31 11:10           ` [PATCH V2 " Martin Habets
2022-02-02  5:10             ` patchwork-bot+netdevbpf

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CACT4oufaWTsvHEwi4vxcPtJLocGD2HZAaZfEMSei4ESF1amsnQ@mail.gmail.com \
    --to=ihuguet@redhat.com \
    --cc=dinang@xilinx.com \
    --cc=ecree.xilinx@gmail.com \
    --cc=eric.dumazet@gmail.com \
    --cc=habetsm.xilinx@gmail.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabloc@xilinx.com \
    --subject='Re: Bad performance in RX with sfc 40G' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).