DPDK usage discussions
 help / color / mirror / Atom feed
From: Pavel Vazharov <freakpv@gmail.com>
To: users <users@dpdk.org>
Subject: Bonding with net_af_xdp devices?
Date: Fri, 10 Nov 2023 15:46:32 +0200	[thread overview]
Message-ID: <CAK9EM19uu5eJHM-7wRrEreY5JdLDX44DPTaCJ+OQSoOXATqQcA@mail.gmail.com> (raw)

[-- Attachment #1: Type: text/plain, Size: 1289 bytes --]

Hi there,

We have an application running on top of DPDK. Now we need to change the
setup so that some traffic enters in the Linux stack while other continues
to be processed by the already existing DPDK application. For this reason
we decided to modify the DPDK application to run on top of XDP sockets with
applied BPF filters for the traffic separation. The scheme seems to work
fine in case of a single port. However, we have usage scenarios where the
application runs on top of several ports and creates bonding internally.
So, I tried to create a bonding of net_af_xdp devices but it doesn't seem
to work and I can't figure the reason why, looking at the code. It prints
the following errors internally:

xsk_configure(): Failed to create xsk socket.
eth_rx_queue_setup(): Failed to configure xdp socket
slave_start(1817) - rte_eth_rx_queue_setup: port=0 queue_id 0, err (-22)
bond_ethdev_start(2057) - bonded port (3) failed to start slave device (0)

So, I'd like to ask:
1. If bonding with net_af_xdp devices possible or not?
2. If the bonding is not possible are there other options to run DPDK on
top of XDP sockets when there are multiple physcial ports at the bottom?
Should I use "directly" the net_af_xdp devices and implement the bonding
logic in the application?

Thanks.

[-- Attachment #2: Type: text/html, Size: 1485 bytes --]

                 reply	other threads:[~2023-11-10 13:46 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=CAK9EM19uu5eJHM-7wRrEreY5JdLDX44DPTaCJ+OQSoOXATqQcA@mail.gmail.com \
    --to=freakpv@gmail.com \
    --cc=users@dpdk.org \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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).