From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wg0-x232.google.com (mail-wg0-x232.google.com [IPv6:2a00:1450:400c:c00::232]) by dpdk.org (Postfix) with ESMTP id 95C61156 for ; Tue, 19 Nov 2013 17:41:52 +0100 (CET) Received: by mail-wg0-f50.google.com with SMTP id k14so7930080wgh.17 for ; Tue, 19 Nov 2013 08:42:50 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=LNlgguWIKEDhS+vvTpcqS1of0RAIFFbe/YnDl035+9A=; b=eVBrC2gcj1cbRkDXgiyRRr19r0Ri71V6JXc0P50P3J1+5fG0ZnidcP9zPaFiwtnNDV gbmnrc1fuNDVOb3Dd86CMr8s9dqv+NW5xxrDMgdl4eGrtlCvy6W32CiUUdXfOT/gpyUJ UqdWHAyv7F6vAXhnMPEtb+ZbG/V8dXqCpmGWG8/36+1Ra1UQUb0lXf/K5ywutkOAGoDo 0jEavchli/zxdHmK5dPgP0rRsoW4cNs1udP/tzDdVVjFfD5DiTxKUdOA+0JkNN//Cx4y W2dC/SZG5sY7OIaBhg+Z8uMCDj3vaWjLNX0UXk8i946owQA3guXiSQ8lUBJcMU+m0rDn AbGg== X-Received: by 10.180.101.230 with SMTP id fj6mr1544437wib.58.1384879370158; Tue, 19 Nov 2013 08:42:50 -0800 (PST) MIME-Version: 1.0 Received: by 10.194.94.100 with HTTP; Tue, 19 Nov 2013 08:42:30 -0800 (PST) In-Reply-To: <65281A9A-D192-4C9E-9EF8-A02E0A999862@windriver.com> References: <65281A9A-D192-4C9E-9EF8-A02E0A999862@windriver.com> From: jinho hwang Date: Tue, 19 Nov 2013 11:42:30 -0500 Message-ID: To: "Wiles, Roger Keith" Content-Type: text/plain; charset=ISO-8859-1 Cc: dev Subject: Re: [dpdk-dev] ways to generate 40Gbps with two NICs x two ports? X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Nov 2013 16:41:53 -0000 On Tue, Nov 19, 2013 at 11:31 AM, Wiles, Roger Keith wrote: > How do you have Pktgen configured in this case? > > On my westmere dual socket 3.4Ghz machine I can send 20G on a single NIC > 82599x two ports. My machine has a PCIe bug that does not allow me to send > on more then 3 ports at wire rate. I get close to 40G 64 byte packets, but > the forth port does is about 70% of wire rate because of the PCIe hardware > bottle neck problem. > > Keith Wiles, Principal Technologist for Networking member of the CTO office, > Wind River > direct 972.434.4136 mobile 940.213.5533 fax 000.000.0000 > > On Nov 19, 2013, at 10:09 AM, jinho hwang wrote: > > Hi All, > > I have two NICs (82599) x two ports that are used as packet generators. I > want to generate full line-rate packets (40Gbps), but Pktgen-DPDK does not > seem to be able to do it when two port in a NIC are used simultaneously. > Does anyone know how to generate 40Gbps without replicating packets in the > switch? > > Thank you, > > Jinho > > Hi Keith, Thank you for the e-mail. I am not sure how I figure out whether my PCIe also has any problems to prevent me from sending full line-rates. I use Intel(R) Xeon(R) CPU E5649 @ 2.53GHz. It is hard for me to figure out where is the bottleneck. My configuration is: sudo ./app/build/pktgen -c 1ff -n 3 $BLACK-LIST -- -p 0xf0 -P -m "[1:2].0, [3:4].1, [5:6].2, [7:8].3" -f test/forward.lua === port to lcore mapping table (# lcores 9) === lcore: 0 1 2 3 4 5 6 7 8 port 0: D: T 1: 0 0: 1 0: 0 0: 0 0: 0 0: 0 0: 0 0: 0 = 1: 1 port 1: D: T 0: 0 0: 0 1: 0 0: 1 0: 0 0: 0 0: 0 0: 0 = 1: 1 port 2: D: T 0: 0 0: 0 0: 0 0: 0 1: 0 0: 1 0: 0 0: 0 = 1: 1 port 3: D: T 0: 0 0: 0 0: 0 0: 0 0: 0 0: 0 1: 0 0: 1 = 1: 1 Total : 0: 0 1: 0 0: 1 1: 0 0: 1 1: 0 0: 1 1: 0 0: 1 Display and Timer on lcore 0, rx:tx counts per port/lcore Configuring 4 ports, MBUF Size 1984, MBUF Cache Size 128 Lcore: 1, type RX , rx_cnt 1, tx_cnt 0 private (nil), RX (pid:qid): ( 0: 0) , TX (pid:qid): 2, type TX , rx_cnt 0, tx_cnt 1 private (nil), RX (pid:qid): , TX (pid:qid): ( 0: 0) 3, type RX , rx_cnt 1, tx_cnt 0 private (nil), RX (pid:qid): ( 1: 0) , TX (pid:qid): 4, type TX , rx_cnt 0, tx_cnt 1 private (nil), RX (pid:qid): , TX (pid:qid): ( 1: 0) 5, type RX , rx_cnt 1, tx_cnt 0 private (nil), RX (pid:qid): ( 2: 0) , TX (pid:qid): 6, type TX , rx_cnt 0, tx_cnt 1 private (nil), RX (pid:qid): , TX (pid:qid): ( 2: 0) 7, type RX , rx_cnt 1, tx_cnt 0 private (nil), RX (pid:qid): ( 3: 0) , TX (pid:qid): 8, type TX , rx_cnt 0, tx_cnt 1 private (nil), RX (pid:qid): , TX (pid:qid): ( 3: 0) Port : 0, nb_lcores 2, private 0x6fd5a0, lcores: 1 2 1, nb_lcores 2, private 0x700208, lcores: 3 4 2, nb_lcores 2, private 0x702e70, lcores: 5 6 3, nb_lcores 2, private 0x705ad8, lcores: 7 8 Initialize Port 0 -- TxQ 1, RxQ 1, Src MAC 90:e2:ba:2f:f2:a4 Create: Default RX 0:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Default TX 0:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Range TX 0:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Sequence TX 0:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Special TX 0:0 - Memory used (MBUFs 64 x (size 1984 + Hdr 64)) + 395392 = 515 KB Port memory used = 10251 KB Initialize Port 1 -- TxQ 1, RxQ 1, Src MAC 90:e2:ba:2f:f2:a5 Create: Default RX 1:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Default TX 1:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Range TX 1:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Sequence TX 1:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Special TX 1:0 - Memory used (MBUFs 64 x (size 1984 + Hdr 64)) + 395392 = 515 KB Port memory used = 10251 KB Initialize Port 2 -- TxQ 1, RxQ 1, Src MAC 90:e2:ba:4a:e6:1c Create: Default RX 2:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Default TX 2:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Range TX 2:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Sequence TX 2:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Special TX 2:0 - Memory used (MBUFs 64 x (size 1984 + Hdr 64)) + 395392 = 515 KB Port memory used = 10251 KB Initialize Port 3 -- TxQ 1, RxQ 1, Src MAC 90:e2:ba:4a:e6:1d Create: Default RX 3:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Default TX 3:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Range TX 3:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Sequence TX 3:0 - Memory used (MBUFs 1024 x (size 1984 + Hdr 64)) + 395392 = 2435 KB Create: Special TX 3:0 - Memory used (MBUFs 64 x (size 1984 + Hdr 64)) + 395392 = 515 KB Port memory used = 10251 KB Total memory used = 41003 KB Port 0: Link Up - speed 10000 Mbps - full-duplex Port 1: Link Up - speed 10000 Mbps - full-duplex Port 2: Link Up - speed 10000 Mbps - full-duplex Port 3: Link Up - speed 10000 Mbps - full-duplex === Display processing on lcore 0 === RX processing on lcore 1, rxcnt 1, port/qid, 0/0 === TX processing on lcore 2, txcnt 1, port/qid, 0/0 === RX processing on lcore 3, rxcnt 1, port/qid, 1/0 === TX processing on lcore 4, txcnt 1, port/qid, 1/0 === RX processing on lcore 5, rxcnt 1, port/qid, 2/0 === TX processing on lcore 6, txcnt 1, port/qid, 2/0 === RX processing on lcore 7, rxcnt 1, port/qid, 3/0 === TX processing on lcore 8, txcnt 1, port/qid, 3/0 Please, advise me if you have time. Thank you always for your help! Jinho