DPDK usage discussions
 help / color / mirror / Atom feed
From: Stephen Hemminger <stephen@networkplumber.org>
To: "Lombardo, Ed" <Ed.Lombardo@netscout.com>
Cc: Ivan Malov <ivan.malov@arknetworks.am>, users <users@dpdk.org>
Subject: Re: dpdk Tx falling short
Date: Tue, 8 Jul 2025 06:47:07 -0700	[thread overview]
Message-ID: <20250708064707.583df905@hermes.local> (raw)
In-Reply-To: <CH3PR01MB84705D671EDC074FD3B1996D8F4EA@CH3PR01MB8470.prod.exchangelabs.com>

On Tue, 8 Jul 2025 04:10:05 +0000
"Lombardo, Ed" <Ed.Lombardo@netscout.com> wrote:

> Hi Stephen,
> I ensured that in every pipeline stage that enqueue or dequeues mbufs it uses the burst version, perf showed the repercussions of doing one mbuf dequeue and enqueue.
> For the receive stage rte_eth_rx_burst() is used and Tx stage we use rte_eth_tx_burst().  The burst size used in tx_thread for dequeue burst is 512 Mbufs.

You might try buffering like rte_eth_tx_buffer does.
Need to add an additional mechanism to ensure that buffer gets flushed
when you detect idle period.

  reply	other threads:[~2025-07-08 13:47 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-07-03 20:14 Lombardo, Ed
2025-07-03 21:49 ` Stephen Hemminger
2025-07-04  5:58   ` Rajesh Kumar
2025-07-04 11:44 ` Ivan Malov
2025-07-04 14:49   ` Stephen Hemminger
2025-07-05 17:36     ` Lombardo, Ed
2025-07-05 19:02       ` Stephen Hemminger
2025-07-05 19:08       ` Stephen Hemminger
2025-07-06  0:03         ` Lombardo, Ed
2025-07-06 16:02           ` Stephen Hemminger
2025-07-06 17:44             ` Lombardo, Ed
2025-07-07  3:30               ` Stephen Hemminger
2025-07-07 16:27               ` Lombardo, Ed
2025-07-07 21:00                 ` Lombardo, Ed
2025-07-07 21:49                 ` Ivan Malov
2025-07-07 23:04                   ` Stephen Hemminger
2025-07-08  4:10                     ` Lombardo, Ed
2025-07-08 13:47                       ` Stephen Hemminger [this message]
2025-07-08 14:03                         ` Lombardo, Ed
2025-07-08 14:18                           ` Ivan Malov
2025-07-08 14:29                             ` Lombardo, Ed
2025-07-08 14:49                               ` Ivan Malov
2025-07-08 16:31                                 ` Lombardo, Ed
2025-07-08 16:53                                   ` Ivan Malov
2025-07-09  1:09                                     ` Lombardo, Ed
2025-07-09 21:58                                       ` Lombardo, Ed
2025-07-10  6:45                                         ` Ivan Malov
2025-07-05 17:33   ` Lombardo, Ed

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=20250708064707.583df905@hermes.local \
    --to=stephen@networkplumber.org \
    --cc=Ed.Lombardo@netscout.com \
    --cc=ivan.malov@arknetworks.am \
    --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).