From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by smtp.lore.kernel.org (Postfix) with ESMTP id 01A21EF48ED for ; Fri, 13 Feb 2026 08:49:33 +0000 (UTC) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id F23A9402B2; Fri, 13 Feb 2026 09:49:32 +0100 (CET) Received: from authsmtp.register.it (authsmtp08.register.it [81.88.48.58]) by mails.dpdk.org (Postfix) with ESMTP id AF5F0402B2 for ; Fri, 13 Feb 2026 09:49:30 +0100 (CET) Received: from DESKTOPPE5KR91 ([95.231.224.119]) by cmsmtp with ESMTPSA id qorhvyYeREQSXqorhvwNbU; Fri, 13 Feb 2026 09:49:30 +0100 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=outsys.org; s=key_65m6du3g3t; t=1770972570; bh=UYKyg0BE7T6nnH+G/YxuWV49EFA4k3gQz8ng7QFj7Do=; h=From:To:Cc:References:In-Reply-To:Subject:Date; b=OAO0R8qmX96z0khewXfDWkrxYkLzC5/jn3sp/J2lv6F31eWjSw+K6aAnF4XUnbbKE qWPZh2NIcp72OmSljLpuZhTmlgI7Sv5BBLwpWVHLNP4byPTDW/kTi420y9t/xPGZq+ LIXQF34/aV4GSn/OVdly2U1WQtptNt/05hoHa6ZT6qid0egqnJLUpwWbhi+D+L9/Wm QnkvI2gJ9UvxrsFsnbw3Xu5k3xDeBtz1uLVfwORZmbAvaRyMhspub9kP1vtPXvtgK5 Z7M7LkrsXp5/uKKSu7FVdpX9JXy9qX/0D6tHOdyo3WrQnu8uZOOhpYQzBtTMuF6y/y +0d73l4zNULtg== X-Rid: smtp@outsys.eu@95.231.224.119 From: "Ernesto Ruffini" To: "'Ayshathul Thuhara'" , Cc: "'dev'" , References: In-Reply-To: Subject: RE: Guidance Required to Improve Throughput on ConnectX-7 (1x400G) with DPDK v23.07 Date: Fri, 13 Feb 2026 09:49:33 +0100 Message-ID: <039201dc9cc5$acd5dee0$06819ca0$@outsys.org> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0393_01DC9CCE.0E9BA670" X-Mailer: Microsoft Outlook 16.0 Thread-Index: AQHmp12vaPmWbH42Sqkb0yX68nXeOLVsOtMA Content-Language: en-us X-CMAE-Envelope: MS4xfJFirSFvJgncLKZWODCgbLmIU3IdoP1cK/Np43zZ9dwfyGVZ3GGJPdJ7yS4HwS3gtU9H3s1c4vjK7t9OScjlQ7d3KpW+3f5tZ46krrGA1vcPjJ6u/isN 4Lc6AfY2KG+m50d+SlO8qokPGUgvPCHH1sCHzDqccOSlEBzAJpGEBIFX8lZaTmWD+bqbHye+t9iWV5m9bSisrSGsqclVZqNc9nJFf6Go7nfnNilyPDP3BAG8 k8QLXHa5nh0eyuP2nENOYAQHwBmwTlQSF0PAm/Lfl2qSYIf9C0xG00RmtigO3qc1 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org This is a multipart message in MIME format. ------=_NextPart_000_0393_01DC9CCE.0E9BA670 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Hi Ayshathul, I think the best is to follow https://fast.dpdk.org/doc/perf/DPDK_23_11_NVIDIA_NIC_performance_report.pdf Did you check with ethtool or similar the actual connection speed? ethtool linux-nic-name | grep Speed Maybe your module limits it to 10Gbps. In case try to disable auto-negotiation Regards, Ernesto From: Ayshathul Thuhara Sent: Friday, February 13, 2026 07:36 To: users@dpdk.org Cc: dev ; yasu1976@gmail.com Subject: Guidance Required to Improve Throughput on ConnectX-7 (1x400G) with DPDK v23.07 Dear Team, I hope this message finds you well. I am currently performing testing on ConnectX-7 with 1x400G hardware using DPDK v23.07. However, I am unable to achieve data transfer rates beyond 10 Gbps. While reviewing the DPDK mailing list, I came across your previous responses in the related discussion (https://mails.dpdk.org/archives/users/2024-July/007767.html), where you mentioned achieving throughput of around 250 Gbps and above. Could you kindly advise on the configuration changes or optimizations required to improve throughput in this setup? For reference, I am using an OSFP loopback module for testing, where the same port is used for both transmission and reception (single-port loopback setup). CX-7 Firmware: 28.45.1200 CPU : Intel(R) Core(TM) i7-14700 Command used : sudo ./build/app/dpdk-testpmd -a 01:00.0 -- -i I would greatly appreciate your guidance. Thank you in advance, and I look forward to your response. Thanks and regards, Ayshathul Thuhara ------=_NextPart_000_0393_01DC9CCE.0E9BA670 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

Hi = Ayshathul,<= /span>

I think the = best is to follow https://fast.dpdk.org/doc/perf/DPDK_23_11_NVIDIA_NIC_performa= nce_report.pdf

Did you = check with ethtool or similar the actual connection speed? ethtool = linux-nic-name | grep Speed

Maybe your = module limits it to 10Gbps. In case try to disable = auto-negotiation

 

Regards,=

  &= nbsp;            = Ernesto

 

From:<= /b> = Ayshathul Thuhara <ayshathul.thuhara@iwave-global.com> =
Sent: Friday, February 13, 2026 07:36
To: = users@dpdk.org
Cc: dev <dev@dpdk.org>; = yasu1976@gmail.com
Subject: Guidance Required to Improve = Throughput on ConnectX-7 (1x400G) with DPDK = v23.07

 

Dear = Team,

I hope this message finds = you well.

I am currently performing = testing on ConnectX-7 with 1x400G hardware using DPDK v23.07. However, I = am unable to achieve data transfer rates beyond 10 Gbps. While reviewing = the DPDK mailing list, I came across your previous responses in the = related discussion (https://mails.dpdk.org/archi= ves/users/2024-July/007767.html), where you mentioned achieving = throughput of around 250 Gbps and above.

Could you kindly advise on = the configuration changes or optimizations required to improve = throughput in this setup?

For reference, I am using = an OSFP loopback module for testing, where the same port is used for = both transmission and reception (single-port loopback = setup).

CX-7 Firmware: 28.45.1200

CPU : Intel(R) Core(TM) = i7-14700

Command used : sudo = ./build/app/dpdk-testpmd -a 01:00.0 -- -i

I would greatly appreciate = your guidance.

Thank you in advance, and = I look forward to your response.

Thanks and = regards,
Ayshathul Thuhara

 

------=_NextPart_000_0393_01DC9CCE.0E9BA670--