From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 88C7241CCD; Sat, 18 Feb 2023 12:07:00 +0100 (CET) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 17505410EE; Sat, 18 Feb 2023 12:07:00 +0100 (CET) Received: from chinatelecom.cn (prt-mail.chinatelecom.cn [42.123.76.220]) by mails.dpdk.org (Postfix) with ESMTP id CCC1440223; Sat, 18 Feb 2023 12:06:55 +0100 (CET) HMM_SOURCE_IP: 172.18.0.48:45848.785969092 HMM_ATTACHE_NUM: 0000 HMM_SOURCE_TYPE: SMTP Received: from clientip-1.202.54.218 (unknown [172.18.0.48]) by chinatelecom.cn (HERMES) with SMTP id 4B495280090; Sat, 18 Feb 2023 19:06:18 +0800 (CST) X-189-SAVE-TO-SEND: wushaohua@chinatelecom.cn Received: from ([1.202.54.218]) by app0024 with ESMTP id b0006430b8c048a592cd819f0f77392f for ferruh.yigit@amd.com; Sat, 18 Feb 2023 19:06:27 CST X-Transaction-ID: b0006430b8c048a592cd819f0f77392f X-Real-From: wushaohua@chinatelecom.cn X-Receive-IP: 1.202.54.218 X-MEDUSA-Status: 0 Sender: wushaohua@chinatelecom.cn Content-Type: multipart/alternative; boundary="------------V0kQBYEMndhypkOZ00DKi4ot" Message-ID: <4d7f1677-b64e-add2-cc5f-fdc37467e1ab@chinatelecom.cn> Date: Sat, 18 Feb 2023 19:06:08 +0800 MIME-Version: 1.0 User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:102.0) Gecko/20100101 Thunderbird/102.8.0 From: wushaohua Subject: Re: [PATCH] app/testpmd:add vxlan txonly To: Ferruh Yigit , dev@dpdk.org Cc: aman.deep.singh@intel.com, yuying.zhang@intel.com, Keith Wiles , "techboard@dpdk.org" References: <20230103023027.1893801-1-wushaohua@chinatelecom.cn> In-Reply-To: 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 multi-part message in MIME format. --------------V0kQBYEMndhypkOZ00DKi4ot Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit 在 2023/2/18 4:43, Ferruh Yigit 写道: > On 1/3/2023 2:30 AM,wushaohua@chinatelecom.cn wrote: >> From: Shaohua Wu >> >> Add the vxlan packet sending module to actively send >> vxlan packets using a common network adapter. >> The default vni is 1000. >> example: >> ./dpdk-testpmd -l 32-47 -n 16 --file-prefix pg0 -- -i >> --rxq=16 --txq=16 --rxd=1024 --txd=1024 >> --txpkts=64 --burst=64 --mbuf-size=4096 >> --nb-cores=15 --underlay_tx_only >> --underlay-eth-peer=0,f0:00:00:00:00:66 >> --eth-peer=0,08:c0:eb:3e:87:af >> --utx-ip=11.0.0.1,11.0.0.2 >> --tx-ip=30.0.0.1,30.0.0.2 >> --forward-mode=tuntxonly >> --txonly-multi-flow >> >> Signed-off-by: Shaohua Wu > Hi Shaohua, > > This is a good feature but I am not sure if this is in the scope of > testpmd application. > > Testpmd has basic packet generation capabilities, and it is useful for > quick/dirty testing, but for more advanced packet generation perhaps > other tools like pktgen [1] can be used, what do you think? > > What is the justification to have this feature in testpmd, and what is > your usecase? > > > [1] > https://git.dpdk.org/apps/pktgen-dpdk/ Hi Ferruh， Thank you for your review.From what I know now pktgen doesn't seem to support the construction of vxlan packets. In a real test scenario, testpmd only sends bare packets and cannot cover test scenarios, such as ovs decap performance. Currently, testpmd of this patch is used, and the packet sending performance is 22+Mpps with a 25G network adapter. This is necessary for testing DPU devices or virtualization scenarios. Best regards.Shaohua --------------V0kQBYEMndhypkOZ00DKi4ot Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable


=E5=9C=A8 2023/2/18 4:43, Ferruh Yigit= =E5=86=99=E9=81=93:
On 1/3/2023 2:30 AM, wushaohua@chinatelecom.cn wrote:

From: Shaohua Wu <wus=
haohua@chinatelecom.cn>

Add the vxlan packet sending module to actively send
vxlan packets using a common network adapter.
The default vni is 1000.
example:
 ./dpdk-testpmd -l 32-47 -n 16 --file-prefix pg0 -- -i
	--rxq=3D16 --txq=3D16 --rxd=3D1024 --txd=3D1024
	--txpkts=3D64 --burst=3D64  --mbuf-size=3D4096
	--nb-cores=3D15  --underlay_tx_only
	--underlay-eth-peer=3D0,f0:00:00:00:00:66
	--eth-peer=3D0,08:c0:eb:3e:87:af
	--utx-ip=3D11.0.0.1,11.0.0.2
	--tx-ip=3D30.0.0.1,30.0.0.2
	--forward-mode=3Dtuntxonly
	--txonly-multi-flow

Signed-off-by: Shaohua Wu <wushaohua@chinatelecom.cn>

Hi Shaohua,

This is a good feature but I am not sure if this is in the scope of
testpmd application.

Testpmd has basic packet generation capabilities, and it is useful for
quick/dirty testing, but for more advanced packet generation perhaps
other tools like pktgen [1] can be used, what do you think?

What is the justification to have this feature in testpmd, and what is
your usecase?


[1]
https://git.dpdk.org/apps/pktgen-dpdk/

Hi Ferruh=EF=BC=8C

Thank=C2=A0you=C2=A0for=C2=A0your=C2=A0review.From=C2=A0what=C2=A0= I=C2=A0know=C2=A0now=C2=A0pktgen=C2=A0doesn't=C2=A0seem=C2=A0to=C2=A0supp= ort=C2=A0the=C2=A0construction=C2=A0of=C2=A0vxlan=C2=A0packets.

In=C2=A0a=C2=A0real=C2=A0test=C2=A0scenario= ,=C2=A0testpmd=C2=A0only=C2=A0sends=C2=A0bare=C2=A0packets=C2=A0and=C2=A0= cannot=C2=A0cover=C2=A0test=C2=A0scenarios,=C2=A0such=C2=A0as=C2=A0ovs decap=C2=A0performance.

Currently,=C2=A0testpmd=C2=A0of=C2=A0this=C2= =A0patch=C2=A0is=C2=A0used,=C2=A0and=C2=A0the=C2=A0packet=C2=A0sending=C2= =A0performance=C2=A0is=C2=A022+Mpps with=C2=A0a=C2=A025G=C2=A0network adapter.


This=C2=A0is=C2=A0necessary=C2=A0for=C2=A0t= esting=C2=A0DPU=C2=A0devices=C2=A0or=C2=A0virtualization=C2=A0scenarios.<= /p>



Best regards.Shaohua

--------------V0kQBYEMndhypkOZ00DKi4ot--