From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from dpdk.org (dpdk.org [92.243.14.124]) by dpdk.space (Postfix) with ESMTP id 53FD7A05FE for ; Mon, 18 Mar 2019 16:31:02 +0100 (CET) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id 6A0AC324D; Mon, 18 Mar 2019 16:31:01 +0100 (CET) Received: from out20-1.mail.aliyun.com (out20-1.mail.aliyun.com [115.124.20.1]) by dpdk.org (Postfix) with ESMTP id 08DD62C28; Fri, 15 Mar 2019 02:40:22 +0100 (CET) X-Alimail-AntiSpam: AC=CONTINUE; BC=0.04614903|-1; CH=green; DM=CONTINUE|CONTINUE|true|0.136786-0.109141-0.754074; FP=0|0|0|0|0|-1|-1|-1; HT=e02c03275; MF=hfli@netitest.com; NM=1; PH=DS; RN=3; RT=3; SR=0; TI=SMTPD_---.E8IZFjf_1552614018; Received: from PC20180210SSYA(mailfrom:hfli@netitest.com fp:SMTPD_---.E8IZFjf_1552614018) by smtp.aliyun-inc.com(10.147.43.230); Fri, 15 Mar 2019 09:40:19 +0800 From: To: Cc: , References: In-Reply-To: Date: Fri, 15 Mar 2019 09:40:20 +0800 Message-ID: <006401d4dad0$0cf8c630$26ea5290$@netitest.com> MIME-Version: 1.0 X-Mailer: Microsoft Outlook 16.0 Thread-Index: AdTVm4GyxTIRR2M0SQ2FiVU1ERP1UgFM+pmQ Content-Language: zh-cn X-Mailman-Approved-At: Mon, 18 Mar 2019 16:31:00 +0100 Content-Type: text/plain; charset="gb2312" Content-Transfer-Encoding: quoted-printable X-Content-Filtered-By: Mailman/MimeDel 2.1.15 Subject: [dpdk-users] =?gb2312?b?tPC4tDogSG93IHRvIHNlbmQgcGFja2V0IHdpdGgg?= =?gb2312?b?bGluZSByYXRlIG9uIEkzNTAtREEyLCBidXQgbm90IGNhcmUgcngg?= =?gb2312?b?c3BlZWQ=?= X-BeenThere: users@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK usage discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: users-bounces@dpdk.org Sender: "users" Hi Wenzhuo, =20 I found the reason, it due to =A1=B0Flow Control=A1=B1 features of = network adapter, https://www.intel.com/content/www/us/en/support/articles/000005593/networ= k-a nd-i-o/ethernet-products.html, it will be send packet with line rate = however after call below function. =20 static void nova_disable_flow_control(int portid) { struct rte_eth_fc_conf fc_conf; memset(&fc_conf, 0, sizeof(struct rte_eth_fc_conf)); =20 if (rte_eth_dev_flow_ctrl_get(portid, &fc_conf) !=3D 0) { printf("Can't get flow ctrl config from port %d\n", portid); return; } =20 if (fc_conf.mode !=3D RTE_FC_NONE) { fc_conf.mode =3D RTE_FC_NONE; if (rte_eth_dev_flow_ctrl_set(portid, &fc_conf) !=3D 0) { printf("Can't set flow ctrl config from port %d\n", portid); } } =20 printf("Disable flow control on port %d ... Done\n", portid); } =20 =20 Thanks and Regards, Haifeng =20 =B7=A2=BC=FE=C8=CB: hfli@netitest.com =20 =B7=A2=CB=CD=CA=B1=BC=E4: 2019=C4=EA3=D4=C28=C8=D5 18:42 =CA=D5=BC=FE=C8=CB: 'wenzhuo.lu@intel.com' =B3=AD=CB=CD: 'dev@dpdk.org' ; 'users@dpdk.org' = =D6=F7=CC=E2: How to send packet with line rate on I350-DA2, but not = care rx speed =20 Wenzhuo, =20 >From =A1=B0MAINTAINERS=A1=B1 file in package of dpdk source code, I know = that you maintain e1000e driver now, could you help me for an question? =20 I use DPDK-pktgen based on Intel I350-DA2 network adapters, when I = loopback two ports on one NIC(I350_eth1 <-> I350_eth2), I got 1Gbps line rate TX throughput. =20 But when I link a Firewall between two ports, just like I350_eth1 -> Firewall_port1 -> Firewall_port2 -> I350_eth2, even I try to burst tx throughput as line rate by power CPU cores, it only get 600Mbps tx throughput which is firewall=A1=AFs rx speed, seem the tx speed from = I350 just depend on peer firewall rx capability. =20 I try this on others network adapters like X520-DA2 and X710DA2, both of them got different results, tx speed only depend on CPU power, but no related with rx capability of network devices. =20 Could you know why of this? Is there any = configuration/interface/register update this feature? Let us send packet with line rate, but not care rx speed. =20 =20 / # lspci | grep I350 0e:00.0 Ethernet controller: Intel Corporation I350 Gigabit Network Connection (rev 01) 0e:00.1 Ethernet controller: Intel Corporation I350 Gigabit Network Connection (rev 01) / # =20 Thanks and Regards, Haifeng