From patchwork Tue Mar 21 02:34:37 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Ling, WeiX" X-Patchwork-Id: 125348 Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 67500427EC; Tue, 21 Mar 2023 03:36:40 +0100 (CET) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 608D040C35; Tue, 21 Mar 2023 03:36:40 +0100 (CET) Received: from mga06.intel.com (mga06b.intel.com [134.134.136.31]) by mails.dpdk.org (Postfix) with ESMTP id 0707740C35 for ; Tue, 21 Mar 2023 03:36:37 +0100 (CET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1679366198; x=1710902198; h=from:to:cc:subject:date:message-id:mime-version: content-transfer-encoding; bh=jMWvBkmUidoIZLP+gL+XTiB9YZwmY6d5Vk+z5WJjcn8=; b=f8iUfpDP8HF1ojX5zhPf61jYzbIhB1EzwuSywFeHzcfrv6DQ1IMMFDbC vK85Wv2mtYmOxXndgtfELDIG8r/v6ONU0v2QkXGN4lmcyHWDh6OJ8OxCl YjUHAC9OuMotzpARZrdXeVnkRYbRS/Nt3w5/iZ8+ZtP8/2q4/tP5Evhg2 xHcEhIyjz4gwf7C/Cs2TZ3eXfO4z5ouL3qEKkO3yWtlyHdITpoICbbT9m LmrohORYBT2crjWRXvCCJa6cA8EPYe/q5G9H9Lzn6uejIJ7aarRxcBHbY TEJgvQ++8ZW40GGYx3AoVFu+NEdss+Ph8RwbvGIbqKWU/wPLr1I/h4uAr Q==; X-IronPort-AV: E=McAfee;i="6600,9927,10655"; a="401405433" X-IronPort-AV: E=Sophos;i="5.98,277,1673942400"; d="scan'208";a="401405433" Received: from orsmga003.jf.intel.com ([10.7.209.27]) by orsmga104.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 20 Mar 2023 19:36:36 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10655"; a="631380882" X-IronPort-AV: E=Sophos;i="5.98,277,1673942400"; d="scan'208";a="631380882" Received: from unknown (HELO localhost.localdomain) ([10.239.252.222]) by orsmga003-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 20 Mar 2023 19:36:35 -0700 From: Wei Ling To: dts@dpdk.org Cc: Wei Ling Subject: [dts][PATCH V1 1/3] test_plans/index: add vswitch_sample_dsa Date: Tue, 21 Mar 2023 10:34:37 +0800 Message-Id: <20230321023437.3854526-1-weix.ling@intel.com> X-Mailer: git-send-email 2.25.1 MIME-Version: 1.0 X-BeenThere: dts@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: test suite reviews and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dts-bounces@dpdk.org Add vswitch_sample_dsa_test_plan in test_plans/index.rst. Signed-off-by: Wei Ling --- test_plans/index.rst | 1 + 1 file changed, 1 insertion(+) diff --git a/test_plans/index.rst b/test_plans/index.rst index 0770a935..69a81697 100644 --- a/test_plans/index.rst +++ b/test_plans/index.rst @@ -243,6 +243,7 @@ The following are the test plans for the DPDK DTS automated test system. dpdk_gro_lib_cbdma_test_plan dpdk_gso_lib_test_plan vswitch_sample_cbdma_test_plan + vswitch_sample_dsa_test_plan vswitch_pvp_multi_paths_performance_with_cbdma_test_plan vxlan_gpe_support_in_i40e_test_plan pvp_diff_qemu_version_test_plan From patchwork Tue Mar 21 02:34:52 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: "Ling, WeiX" X-Patchwork-Id: 125349 Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id B5541427EC; Tue, 21 Mar 2023 03:36:56 +0100 (CET) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id ABCEE40A7A; Tue, 21 Mar 2023 03:36:56 +0100 (CET) Received: from mga06.intel.com (mga06b.intel.com [134.134.136.31]) by mails.dpdk.org (Postfix) with ESMTP id 0971A40693 for ; Tue, 21 Mar 2023 03:36:53 +0100 (CET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1679366214; x=1710902214; h=from:to:cc:subject:date:message-id:mime-version: content-transfer-encoding; bh=XonUe7DQjO4wTDmvVFb2c51Bqgbgmqlgjh/QaqWXHvk=; b=jbVtgiRscqDUth305F4rgkQSu0k6ujwGZQsFjfdqXvhlJ+FsCjJDOFyQ JFwvpdjaMYQruThITuMWrXmLPbaYpSKOBXjPAo0zwd0v71iO9pLaaQGe2 XlkeWPli6Z6bsLSURRmb/Wiq8DFfYuxS2Foi7/0Oi82/Q3ToUgrtVwu2Z k/nxTuNXe9mUlygRHxZpyHUJvAJKRwavzcYeRyynYadGTN8bdeCFgfqvV teIGneUbzvag017LtwWBileVbTjvu+JkSianipRmAaQQCOuqBz9Gfwq5X DfLrNdLPoSop0CDu37/Crow6Q8oAbVICd0+UT+gbM7nygb5IZZRWpHPuW g==; X-IronPort-AV: E=McAfee;i="6600,9927,10655"; a="401405461" X-IronPort-AV: E=Sophos;i="5.98,277,1673942400"; d="scan'208";a="401405461" Received: from orsmga003.jf.intel.com ([10.7.209.27]) by orsmga104.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 20 Mar 2023 19:36:53 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10655"; a="631380951" X-IronPort-AV: E=Sophos;i="5.98,277,1673942400"; d="scan'208";a="631380951" Received: from unknown (HELO localhost.localdomain) ([10.239.252.222]) by orsmga003-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 20 Mar 2023 19:36:51 -0700 From: Wei Ling To: dts@dpdk.org Cc: Wei Ling Subject: [dts][PATCH V1 2/3] test_plans/vswitch_sample_dsa: add new testplan Date: Tue, 21 Mar 2023 10:34:52 +0800 Message-Id: <20230321023452.3854588-1-weix.ling@intel.com> X-Mailer: git-send-email 2.25.1 MIME-Version: 1.0 X-BeenThere: dts@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: test suite reviews and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dts-bounces@dpdk.org Add vswitch_sample_dsa testplan to test dpdk-vhost APP with host DMA operation callbacks for DSA PMD and vhost async data-path. Signed-off-by: Wei Ling --- test_plans/vswitch_sample_dsa_test_plan.rst | 968 ++++++++++++++++++++ 1 file changed, 968 insertions(+) create mode 100644 test_plans/vswitch_sample_dsa_test_plan.rst diff --git a/test_plans/vswitch_sample_dsa_test_plan.rst b/test_plans/vswitch_sample_dsa_test_plan.rst new file mode 100644 index 00000000..677d0070 --- /dev/null +++ b/test_plans/vswitch_sample_dsa_test_plan.rst @@ -0,0 +1,968 @@ +.. SPDX-License-Identifier: BSD-3-Clause + Copyright(c) 2023 Intel Corporation + +======================================================== +Vswitch sample test with vhost async data path test plan +======================================================== + +Description +=========== + +Vswitch sample can leverage DMA to accelerate vhost async data-path from dpdk 20.11. This plan test +vhost DMA operation callbacks for DSA PMD and vhost async data-path in vhost sample. + +--dmas This parameter is used to specify the assigned DMA device of a vhost device. Async vhost-user +net driver will be used if --dmas is set. For example –dmas [txd0@wq0.0,txd1@wq0.1,rxd0@wq0.2,rxd1@wq0.3] +means use DMA channel wq0.0/wq0.2 for vhost device 0 enqueue/dequeue operation and use DMA channel +wq0.1/wq0.3 for vhost device 1 enqueue/dequeue operation. The index of the device corresponds to +the socket file in order, that means vhost device 0 is created through the first socket file, +vhost device 1 is created through the second socket file, and so on. + +For more about vswitch example, please refer to the DPDK docment: +http://doc.dpdk.org/guides/sample_app_ug/vhost.html + +Prerequisites +============= + +Hardware +-------- +Supportted NICs: NIC that supports VMDQ + +Software +-------- +Trex:http://trex-tgn.cisco.com/trex/release/v2.26.tar.gz + +General set up +-------------- +1. Compile DPDK and vhost example:: + + # CC=gcc meson --werror -Denable_kmods=True -Dlibdir=lib -Dexamples=all --default-library=static x86_64-native-linuxapp-gcc + # meson configure -Dexamples=vhost x86_64-native-linuxapp-gcc + # ninja -C x86_64-native-linuxapp-gcc -j 110 + +2. Get the PCI device of DUT, for example, 0000:6a:00.0 is NIC port, 0000:6a:01.0 - 0000:f6:01.0 are DSA devices:: + + # ./usertools/dpdk-devbind.py -s + + Network devices using kernel driver + =================================== + 0000:6a:00.0 'Ethernet Controller E810-C for QSFP 1592' drv=ice unused=vfio-pci + + DMA devices using kernel driver + =============================== + 0000:6a:01.0 'Device 0b25' drv=idxd unused=vfio-pci + 0000:6f:01.0 'Device 0b25' drv=idxd unused=vfio-pci + 0000:74:01.0 'Device 0b25' drv=idxd unused=vfio-pci + 0000:79:01.0 'Device 0b25' drv=idxd unused=vfio-pci + 0000:e7:01.0 'Device 0b25' drv=idxd unused=vfio-pci + 0000:ec:01.0 'Device 0b25' drv=idxd unused=vfio-pci + 0000:f1:01.0 'Device 0b25' drv=idxd unused=vfio-pci + 0000:f6:01.0 'Device 0b25' drv=idxd unused=vfio-pci + +Test case +========= + +Common steps +------------ +1. Bind 1 NIC port to vfio-pci:: + + # ./usertools/dpdk-devbind.py -b vfio-pci + + For example: + # ./usertools/dpdk-devbind.py -b vfio-pci 0000:6a:00.0 + +2. Bind DSA devices to DPDK vfio-pci driver:: + + # ./usertools/dpdk-devbind.py -b vfio-pci + + For example, bind 2 DSA devices to vfio-pci driver: + # ./usertools/dpdk-devbind.py -b vfio-pci 0000:e7:01.0 0000:ec:01.0 + +.. note:: + + One DPDK DSA device can create 8 WQ at most. Below is an example, where DPDK DSA device will create one and + eight WQ for DSA deivce 0000:e7:01.0 and 0000:ec:01.0. The value of “max_queues” is 1~8: + # ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 4 -n 4 -a 0000:e7:01.0,max_queues=1 -- -i + # ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 4 -n 4 -a 0000:ec:01.0,max_queues=8 -- -i + +3. Bind DSA devices to kernel idxd driver, and configure Work Queue (WQ):: + + # ./usertools/dpdk-devbind.py -b idxd + # ./drivers/dma/idxd/dpdk_idxd_cfg.py -q + +.. note:: + + Better to reset WQ when need operate DSA devices that bound to idxd drvier: + # ./drivers/dma/idxd/dpdk_idxd_cfg.py --reset + You can check it by 'ls /dev/dsa' + dsa_idx: Index of DSA devices, where 0<=dsa_idx<=7, corresponding to 0000:6a:01.0 - 0000:f6:01.0 + wq_num: Number of workqueues per DSA endpoint, where 1<=wq_num<=8 + + For example, bind 2 DSA devices to idxd driver and configure WQ: + + # ./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0 + # ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 1 0 + # ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 4 1 + Check WQ by 'ls /dev/dsa' and can find "wq0.0 wq1.0 wq1.1 wq1.2 wq1.3" + +Test Case 1: VM2VM virtio-user forwarding test when vhost async operation using DSA dpdk driver +----------------------------------------------------------------------------------------------- +This case tests that the imix packets can forward normally in VM2VM topology(virtio-user as front-end) when vhost async operation using DSA dpdk drvier, both split ring and packed ring have been covered. + +1.Bind 1 NIC port and 1 DSA device to vfio-pci. + +2. On host, launch dpdk-vhost by below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 26-28 -n 4 -a 0000:3a:00.0 -a 0000:6a:01.0,max_queues=4 \ + -- -p 0x1 --mergeable 1 --vm2vm 1 --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@0000:6a:01.0-q0,rxd0@0000:6a:01.0-q1,txd1@0000:6a:01.0-q2,rxd1@0000:6a:01.0-q3] \ + --client --total-num-mbufs 600000 + +3. Launch virtio-user:: + + ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 29-30 -n 4 --no-pci --file-prefix=testpmd0 \ + --vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=/root/dpdk/vhost-net0,queues=1,server=1,mrg_rxbuf=1,in_order=0,packed_vq=1 \ + -- -i --rxq=1 --txq=1 --txd=1024 --rxd=1024 --nb-cores=1 + + ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 31-32 -n 4 --no-pci --file-prefix=testpmd1 \ + --vdev=net_virtio_user0,mac=00:11:22:33:44:11,path=/root/dpdk/vhost-net1,queues=1,server=1,mrg_rxbuf=1,in_order=1,vectorized=1 \ + -- -i --rxq=1 --txq=1 --txd=1024 --rxd=1024 --nb-cores=1 + +4. Loop packets between two virtio-user sides, record performance number with 64b/2000b/8000b/IMIX pkts can get expected:: + + testpmd0>set fwd mac + testpmd0>start tx_first + testpmd0>stop + testpmd0>set eth-peer 0 00:11:22:33:44:11 + testpmd0>start + testpmd1>set fwd mac + testpmd1>set eth-peer 0 00:11:22:33:44:10 + testpmd1>set txpkts 64 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 2000 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 2000,2000,2000,2000 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 64,256,2000,64,256,2000 + testpmd1>start tx_first + testpmd1>show port stats all + +5. Stop dpdk-vhost side and relaunch it with below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 26-28 -n 4 -a 0000:3a:00.0 -a 0000:6a:01.0,max_queues=4 \ + -- -p 0x1 --mergeable 1 --vm2vm 1 --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@0000:6a:01.0-q0,rxd1@0000:6a:01.0-q1] \ + --client --total-num-mbufs 600000 + +6. Rerun step 4. + +Test Case 2: VM2VM virtio-pmd split ring test with DSA dpdk driver register/unregister stable check +--------------------------------------------------------------------------------------------------- +This case checks that the split ring with DSA dpdk driver can work stably when the virtio-pmd port is registed and unregisted for many times. + +1. Bind 1 NIC port and 1 DSA device to vfio-pci. + +2. On host, launch dpdk-vhost by below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 26-28 -n 4 -a 0000:3a:00.0 -a 0000:6a:01.0,max_queues=4 \ + -- -p 0x1 --mergeable 1 --vm2vm 1 --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@0000:6a:01.0-q0,rxd0@0000:6a:01.0-q1,txd1@0000:6a:01.0-q2,rxd1@0000:6a:01.0-q3] \ + --client --total-num-mbufs 600000 + +3. Start VM0 with qemu:: + + qemu-system-x86_64 -name vm0 -enable-kvm -pidfile /tmp/.vm0.pid \ + -daemonize -monitor unix:/tmp/vm0_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6000-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net0,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:01,disable-modern=true,mrg_rxbuf=on,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm0_qga0.sock,server,nowait,id=vm0_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm0_qga0,name=org.qemu.guest_agent.0 -vnc :4 -drive file=/home/image/ubuntu2004.img + +4. Start VM1 with qemu:: + + qemu-system-x86_64 -name vm1 -enable-kvm -pidfile /tmp/.vm1.pid \ + -daemonize -monitor unix:/tmp/vm1_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6001-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net1,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:02,disable-modern=false,mrg_rxbuf=on,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm1_qga0.sock,server,nowait,id=vm1_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm1_qga0,name=org.qemu.guest_agent.0 -vnc :5 -drive file=/home/image/ubuntu2004_2.img + +5. Bind virtio port to vfio-pci in both two VMs:: + + modprobe vfio enable_unsafe_noiommu_mode=1 + modprobe vfio-pci + echo 1 > /sys/module/vfio/parameters/enable_unsafe_noiommu_mode + ./usertools/dpdk-devbind.py --bind=vfio-pci 00:05.0 + +6. Start testpmd in VMs seperately:: + + ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 1-2 -n 4 -- -i --rxq=1 --txq=1 --nb-cores=1 --txd=1024 --rxd=1024 + +7. Loop pkts between two virtio-user sides, record performance number with 64b/2000b/8000b/IMIX pkts can get expected:: + + testpmd0>set fwd mac + testpmd0>start tx_first + testpmd0>stop + testpmd0>set eth-peer 0 52:54:00:00:00:02 + testpmd0>start + testpmd1>set fwd mac + testpmd1>set eth-peer 0 52:54:00:00:00:01 + testpmd1>set txpkts 64 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 2000 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 2000,2000,2000,2000 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 64,256,2000,64,256,2000 + testpmd1>start tx_first + testpmd1>show port stats all + +8. Quit two testpmd in two VMs, bind virtio-pmd port to virtio-pci,then bind port back to vfio-pci, rerun below cmd 50 times:: + + ./usertools/dpdk-devbind.py -u 00:05.0 + ./usertools/dpdk-devbind.py --bind=virtio-pci 00:05.0 + ./usertools/dpdk-devbind.py --bind=vfio-pci 00:05.0 + +9. Rerun step 6-7, check vhost can stable work and get expected throughput. + +10. Quit and relaunch dpdk-vhost with below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 26-28 -n 4 -a 0000:3a:00.0 -a 0000:6a:01.0,max_queues=4 -- -p 0x1 --mergeable 1 --vm2vm 1 \ + --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@0000:6a:01.0-q0,rxd1@0000:6a:01.0-q3] \ + --client --total-num-mbufs 600000 + +11. Rerun step 6-9, check vhost can stable work and get expected throughput. + +Test Case 3: VM2VM virtio-pmd packed ring test with DSA dpdk driver register/unregister stable check +---------------------------------------------------------------------------------------------------- +This case checks that the packed ring with DSA dpdk driver can work stably when the virtio-pmd port is registed and unregisted for many times. + +1. Bind 1 NIC port and 1 DSA devices to vfio-pci. + +2. On host, launch dpdk-vhost by below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 26-28 -n 4 -a 0000:3a:00.0 -a 0000:6a:01.0,max_queues=4 -a 0000:6f:01.0,max_queues=4 \ + -- -p 0x1 --mergeable 1 --vm2vm 1 --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@0000:6a:01.0-q0,rxd0@0000:6a:01.0-q1,txd1@0000:6f:01.0-q2,rxd1@0000:6f:01.0-q3] \ + --client --total-num-mbufs 600000 + +3. Start VM0 with qemu:: + + qemu-system-x86_64 -name vm0 -enable-kvm -pidfile /tmp/.vm0.pid \ + -daemonize -monitor unix:/tmp/vm0_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6000-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net0,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:01,disable-modern=false,mrg_rxbuf=on,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm0_qga0.sock,server,nowait,id=vm0_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm0_qga0,name=org.qemu.guest_agent.0 -vnc :4 -drive file=/home/image/ubuntu2004.img + +4. Start VM1 with qemu:: + + qemu-system-x86_64 -name vm1 -enable-kvm -pidfile /tmp/.vm1.pid \ + -daemonize -monitor unix:/tmp/vm1_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6001-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net1,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:02,disable-modern=false,mrg_rxbuf=on,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm1_qga0.sock,server,nowait,id=vm1_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm1_qga0,name=org.qemu.guest_agent.0 -vnc :5 -drive file=/home/image/ubuntu2004_2.img + +5. Bind virtio port to vfio-pci in both two VMs:: + + modprobe vfio enable_unsafe_noiommu_mode=1 + modprobe vfio-pci + echo 1 > /sys/module/vfio/parameters/enable_unsafe_noiommu_mode + ./usertools/dpdk-devbind.py --bind=vfio-pci 00:05.0 + +6. Start testpmd in VMs seperately:: + + ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 1-2 -n 4 -- -i --rxq=1 --txq=1 --nb-cores=1 --txd=1024 --rxd=1024 + +7. Loop packets between two virtio-user sides, record performance number with 64b/2000b/8000b/IMIX pkts can get expected:: + + testpmd0>set fwd mac + testpmd0>start tx_first + testpmd0>stop + testpmd0>set eth-peer 0 52:54:00:00:00:02 + testpmd0>start + testpmd1>set fwd mac + testpmd1>set eth-peer 0 52:54:00:00:00:01 + testpmd1>set txpkts 64 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 2000 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 2000,2000,2000,2000 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 64,256,2000,64,256,2000 + testpmd1>start tx_first + testpmd1>show port stats all + +8. Quit two testpmd in two VMs, bind virtio-pmd port to virtio-pci,then bind port back to vfio-pci, rerun below cmd 50 times:: + + ./usertools/dpdk-devbind.py -u 00:05.0 + ./usertools/dpdk-devbind.py --bind=virtio-pci 00:05.0 + ./usertools/dpdk-devbind.py --bind=vfio-pci 00:05.0 + +9. Rerun step 6-7，check vhost can stable work and get expected throughput. + +10. Quit VMs and rerun step 3-9. + +Test Case 4: VM2VM virtio-net split ring test with DSA dpdk driver and iperf stable check +----------------------------------------------------------------------------------------- +This case tests with split ring with DSA dpdk driver in two VMs, check that iperf/scp and reconnection can work stably between two virito-net. + +1. Bind 1 NIC port and 2 DSA devices to vfio-pci. + +2. On host, launch dpdk-vhost by below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 2-3 -n 4 -a 0000:3a:00.0 -a 0000:6a:01.0,max_queues=2 -a 0000:6f:01.0,max_queues=2 \ + -- -p 0x1 --mergeable 1 --vm2vm 1 --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@0000:6a:01.0-q0,rxd0@0000:6a:01.0-q1,txd1@0000:6f:01.0-q0,rxd1@0000:6f:01.0-q1] \ + --client --total-num-mbufs 600000 + +3. Start VM0 with qemu:: + + qemu-system-x86_64 -name vm0 -enable-kvm -pidfile /tmp/.vm0.pid \ + -daemonize -monitor unix:/tmp/vm0_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6000-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net0,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:01,disable-modern=true,mrg_rxbuf=on,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm0_qga0.sock,server,nowait,id=vm0_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm0_qga0,name=org.qemu.guest_agent.0 -vnc :4 -drive file=/home/image/ubuntu2004.img + +4. Start VM1 with qemu:: + + qemu-system-x86_64 -name vm1 -enable-kvm -pidfile /tmp/.vm1.pid \ + -daemonize -monitor unix:/tmp/vm1_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6001-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net1,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:02,disable-modern=false,mrg_rxbuf=on,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm1_qga0.sock,server,nowait,id=vm1_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm1_qga0,name=org.qemu.guest_agent.0 -vnc :5 -drive file=/home/image/ubuntu2004_2.img + +5. On VM1, set virtio device IP and run arp protocal:: + + ifconfig ens5 1.1.1.2 + arp -s 1.1.1.8 52:54:00:00:00:02 + +6. On VM2, set virtio device IP and run arp protocal:: + + ifconfig ens5 1.1.1.8 + arp -s 1.1.1.2 52:54:00:00:00:01 + +7. Check the iperf performance between two VMs by below commands:: + + Under VM1, run: `iperf -s -i 1` + Under VM2, run: `iperf -c 1.1.1.2 -i 1 -t 60` + +8. Check iperf throughput can get x Gbits/sec. + +9. Scp 1MB file form VM0 to VM1, check packets can be forwarding success by scp:: + + Under VM1, run: `scp [xxx] root@1.1.1.8:/` [xxx] is the file name + +10. Relaunch dpdk-vhost, then rerun step 7-9 five times. + +11. Relaunch dpdk-vhost by below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 2-3 -n 4 -a 0000:3a:00.0 -a 0000:6a:01.0,max_queues=2 -- -p 0x1 --mergeable 1 --vm2vm 1 \ + --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@0000:6a:01.0-q0,rxd1@0000:6a:01.0-q1] --client + +12. rerun step 7-9. + +Test Case 5: VM2VM virtio-net packed ring test with DSA dpdk driver and iperf stable check +------------------------------------------------------------------------------------------ +This case tests with packed ring with DSA dpdk driver in two VMs, check that iperf/scp and reconnection can work stably between two virito-net. + +1. Bind 1 NIC port to vfio-pci and 2 DSA devices to vfio-pci. + +2. On host, launch dpdk-vhost by below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 26-28 -n 4 -a 0000:3a:00.0 -a 0000:6a:01.0,max_queues=2 -a 0000:6f:01.0,max_queues=2 -- -p 0x1 --mergeable 1 --vm2vm 1 \ + --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@0000:6a:01.0-q0,rxd0@0000:6a:01.0-q1,txd1@0000:6f:01.0-q0,rxd1@0000:6f:01.0-q1] \ + --total-num-mbufs 600000 + +3. Start VM0 with qemu:: + + qemu-system-x86_64 -name vm0 -enable-kvm -pidfile /tmp/.vm0.pid \ + -daemonize -monitor unix:/tmp/vm0_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6000-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net0,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:01,disable-modern=false,mrg_rxbuf=off,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm0_qga0.sock,server,nowait,id=vm0_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm0_qga0,name=org.qemu.guest_agent.0 -vnc :4 -drive file=/home/image/ubuntu2004.img + +4. Start VM1 with qemu:: + + qemu-system-x86_64 -name vm1 -enable-kvm -pidfile /tmp/.vm1.pid \ + -daemonize -monitor unix:/tmp/vm1_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6001-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net1,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:02,disable-modern=false,mrg_rxbuf=off,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm1_qga0.sock,server,nowait,id=vm1_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm1_qga0,name=org.qemu.guest_agent.0 -vnc :5 -drive file=/home/image/ubuntu2004_2.img + +5. On VM1, set virtio device IP and run arp protocal:: + + ifconfig ens5 1.1.1.2 + arp -s 1.1.1.8 52:54:00:00:00:02 + +6. On VM2, set virtio device IP and run arp protocal:: + + ifconfig ens5 1.1.1.8 + arp -s 1.1.1.2 52:54:00:00:00:01 + +7. Check the iperf performance between two VMs by below commands:: + + Under VM1, run: `iperf -s -i 1` + Under VM2, run: `iperf -c 1.1.1.2 -i 1 -t 60` + +8. Check iperf throughput can get x Gbits/sec. + +9. Scp 1MB file form VM1 to VM2M, check packets can be forwarding success by scp:: + + Under VM1, run: `scp [xxx] root@1.1.1.8:/` [xxx] is the file name + +10. Rerun step 7-9 five times. + +11. Relaunch VMs with mrg_rxbuf=on. + +12. Rerun step 5-10. + +Test Case 6: VM2VM virtio-net packed ring test with 2 DSA WQ with dpdk driver and iperf stable check +---------------------------------------------------------------------------------------------------- +This case tests with packed ring with 2 DSA WQ with dpdk driver in two VMs, check that iperf/scp and reconnection can work stably between two virito-net. + +1. Bind 1 NIC port to vfio-pci and 2 DSA devices to vfio-pci. + +2. On host, launch dpdk-vhost by below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 26-28 -n 4 -a 0000:3a:00.0 -a 0000:6a:01.0,max_queues=2 \ + -- -p 0x1 --mergeable 1 --vm2vm 1 --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@0000:6a:01.0-q0,rxd1@0000:6a:01.0-q1] --total-num-mbufs 600000 + +3. Start VM0 with qemu:: + + qemu-system-x86_64 -name vm0 -enable-kvm -pidfile /tmp/.vm0.pid \ + -daemonize -monitor unix:/tmp/vm0_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6000-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net0,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:01,disable-modern=false,mrg_rxbuf=off,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm0_qga0.sock,server,nowait,id=vm0_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm0_qga0,name=org.qemu.guest_agent.0 -vnc :4 -drive file=/home/image/ubuntu2004.img + +4. Start VM1 with qemu:: + + qemu-system-x86_64 -name vm1 -enable-kvm -pidfile /tmp/.vm1.pid \ + -daemonize -monitor unix:/tmp/vm1_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6001-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net1,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:02,disable-modern=false,mrg_rxbuf=off,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm1_qga0.sock,server,nowait,id=vm1_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm1_qga0,name=org.qemu.guest_agent.0 -vnc :5 -drive file=/home/image/ubuntu2004_2.img + +5. On VM1, set virtio device IP and run arp protocal:: + + ifconfig ens5 1.1.1.2 + arp -s 1.1.1.8 52:54:00:00:00:02 + +6. On VM2, set virtio device IP and run arp protocal:: + + ifconfig ens5 1.1.1.8 + arp -s 1.1.1.2 52:54:00:00:00:01 + +7. Check the iperf performance between two VMs by below commands:: + + Under VM1, run: `iperf -s -i 1` + Under VM2, run: `iperf -c 1.1.1.2 -i 1 -t 60` + +8. Check iperf throughput can get x Gbits/sec. + +9. Scp 1MB file form VM1 to VM2M, check packets can be forwarding success by scp:: + + Under VM1, run: `scp [xxx] root@1.1.1.8:/` [xxx] is the file name + +10. Rerun step 7-9 five times. + +11. Relaunch VMs with mrg_rxbuf=on. + +12. Rerun step 5-10. + +Test Case 7: VM2VM virtio-user forwarding test when vhost async operation using DSA kernel driver +------------------------------------------------------------------------------------------------- +This case tests that the imix packets can forward normally in VM2VM topology(virtio-user as front-end) when vhost async operation using DSA kernel drvier, both split ring and packed ring have been covered. + +1.Bind 1 NIC port and 1 DSA device to idxd:: + + # ./usertools/dpdk-devbind.py -b vfio-pci 6a:00.0 + + ls /mnt/dsa #check wq configure, reset if exist + # ./usertools/dpdk-devbind.py -u 6a:01.0 + # ./usertools/dpdk-devbind.py -b idxd 6a:01.0 + ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 4 0 + ls /mnt/dsa #check wq configure success + +2. On host, launch dpdk-vhost by below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 26-28 -n 4 -a 0000:6a:01.0 \ + -- -p 0x1 --mergeable 1 --vm2vm 1 --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@wq0.0,rxd0@wq0.1,txd1@wq0.2,rxd1@wq0.3] \ + --client --total-num-mbufs 600000 + +3. Launch virtio-user:: + + ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 29-30 -n 4 --no-pci --file-prefix=testpmd0 \ + --vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=/root/dpdk/vhost-net0,queues=1,server=1,mrg_rxbuf=1,in_order=0,packed_vq=1 \ + -- -i --rxq=1 --txq=1 --txd=1024 --rxd=1024 --nb-cores=1 + + ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 31-32 -n 4 --no-pci --file-prefix=testpmd1 \ + --vdev=net_virtio_user0,mac=00:11:22:33:44:11,path=/root/dpdk/vhost-net1,queues=1,server=1,mrg_rxbuf=1,in_order=1,vectorized=1 \ + -- -i --rxq=1 --txq=1 --txd=1024 --rxd=1024 --nb-cores=1 + +4. Loop packets between two virtio-user sides, record performance number with 64b/2000b/8000b/IMIX pkts can get expected:: + + testpmd0>set fwd mac + testpmd0>start tx_first + testpmd0>stop + testpmd0>set eth-peer 0 00:11:22:33:44:11 + testpmd0>start + testpmd1>set fwd mac + testpmd1>set eth-peer 0 00:11:22:33:44:10 + testpmd1>set txpkts 64 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 2000 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 2000,2000,2000,2000 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 64,256,2000,64,256,2000 + testpmd1>start tx_first + testpmd1>show port stats all + +5. Stop dpdk-vhost side and relaunch it with below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 26-28 -n 4 -a 0000:6a:01.0 \ + -- -p 0x1 --mergeable 1 --vm2vm 1 --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@wq0.0,rxd1@wq0.1] \ + --client --total-num-mbufs 600000 + +6. Rerun step 4. + +Test Case 8: VM2VM virtio-pmd split ring test with DSA kernel driver register/unregister stable check +------------------------------------------------------------------------------------------------------ +This case checks that the split ring with DSA kernel driver can work stably when the virtio-pmd port is registed and unregisted for many times. + +1. Bind 1 NIC port and 1 DSA device to idxd:: + + # ./usertools/dpdk-devbind.py -b vfio-pci 6a:00.0 + + ls /mnt/dsa #check wq configure, reset if exist + # ./usertools/dpdk-devbind.py -u 6a:01.0 + # ./usertools/dpdk-devbind.py -b idxd 6a:01.0 + ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 4 0 + ls /mnt/dsa #check wq configure success + +2. On host, launch dpdk-vhost by below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 26-28 -n 4 -a 0000:6a:01.0 \ + -- -p 0x1 --mergeable 1 --vm2vm 1 --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@wq0.0,rxd0@wq0.1,txd1@wq0.2,rxd1@wq0.3] \ + --client --total-num-mbufs 600000 + +3. Start VM0 with qemu:: + + qemu-system-x86_64 -name vm0 -enable-kvm -pidfile /tmp/.vm0.pid \ + -daemonize -monitor unix:/tmp/vm0_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6000-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net0,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:01,disable-modern=false,mrg_rxbuf=on,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm0_qga0.sock,server,nowait,id=vm0_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm0_qga0,name=org.qemu.guest_agent.0 -vnc :4 -drive file=/home/image/ubuntu2004.img + +4. Start VM1 with qemu:: + + qemu-system-x86_64 -name vm1 -enable-kvm -pidfile /tmp/.vm1.pid \ + -daemonize -monitor unix:/tmp/vm1_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6001-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net1,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:02,disable-modern=true,mrg_rxbuf=on,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm1_qga0.sock,server,nowait,id=vm1_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm1_qga0,name=org.qemu.guest_agent.0 -vnc :5 -drive file=/home/image/ubuntu2004_2.img + +5. Bind virtio port to vfio-pci in both two VMs:: + + modprobe vfio enable_unsafe_noiommu_mode=1 + modprobe vfio-pci + echo 1 > /sys/module/vfio/parameters/enable_unsafe_noiommu_mode + ./usertools/dpdk-devbind.py --bind=vfio-pci 00:05.0 + +6. Start testpmd in VMs seperately:: + + ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 1-2 -n 4 -- -i --rxq=1 --txq=1 --nb-cores=1 --txd=1024 --rxd=1024 + +7. Loop pkts between two virtio-user sides, record performance number with 64b/2000b/8000b/IMIX pkts can get expected:: + + testpmd0>set fwd mac + testpmd0>start tx_first + testpmd0>stop + testpmd0>set eth-peer 0 52:54:00:00:00:02 + testpmd0>start + testpmd1>set fwd mac + testpmd1>set eth-peer 0 52:54:00:00:00:01 + testpmd1>set txpkts 64 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 2000 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 2000,2000,2000,2000 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 64,256,2000,64,256,2000 + testpmd1>start tx_first + testpmd1>show port stats all + +8. Quit two testpmd in two VMs, bind virtio-pmd port to virtio-pci,then bind port back to vfio-pci, rerun below cmd 50 times:: + + ./usertools/dpdk-devbind.py -u 00:05.0 + ./usertools/dpdk-devbind.py --bind=virtio-pci 00:05.0 + ./usertools/dpdk-devbind.py --bind=vfio-pci 00:05.0 + +9. Rerun step 6-7, check vhost can stable work and get expected throughput. + +10. Quit and relaunch dpdk-vhost with below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 26-28 -n 4 -a 0000:6a:01.0 \ + -- -p 0x1 --mergeable 1 --vm2vm 1 --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@wq0.0,rxd1@wq0.3] \ + --client --total-num-mbufs 600000 + +11. Rerun step 6-9, check vhost can stable work and get expected throughput. + +Test Case 9: VM2VM virtio-pmd packed ring test with DSA kernel driver register/unregister stable check +------------------------------------------------------------------------------------------------------- +This case checks that the packed ring with DSA kernel driver can work stably when the virtio-pmd port is registed and unregisted for many times. + +1. Bind 1 NIC port and 2 DSA devices to idxd:: + + # ./usertools/dpdk-devbind.py -b vfio-pci 6a:00.0 + + ls /mnt/dsa #check wq configure, reset if exist + # ./usertools/dpdk-devbind.py -u 6a:01.0 6f:01.0 + # ./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0 + ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 4 0 + ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 4 1 + ls /mnt/dsa #check wq configure success + +2. On host, launch dpdk-vhost by below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 26-28 -n 4 -a 0000:6a:01.0 \ + -- -p 0x1 --mergeable 1 --vm2vm 1 --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@wq0.0,rxd0@wq0.1,txd1@wq1.0,rxd1@wq1.1] \ + --client --total-num-mbufs 600000 + +3. Start VM0 with qemu:: + + qemu-system-x86_64 -name vm0 -enable-kvm -pidfile /tmp/.vm0.pid \ + -daemonize -monitor unix:/tmp/vm0_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6000-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net0,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:01,disable-modern=false,mrg_rxbuf=on,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm0_qga0.sock,server,nowait,id=vm0_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm0_qga0,name=org.qemu.guest_agent.0 -vnc :4 -drive file=/home/image/ubuntu2004.img + +4. Start VM1 with qemu:: + + qemu-system-x86_64 -name vm1 -enable-kvm -pidfile /tmp/.vm1.pid \ + -daemonize -monitor unix:/tmp/vm1_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6001-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net1,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:02,disable-modern=false,mrg_rxbuf=on,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm1_qga0.sock,server,nowait,id=vm1_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm1_qga0,name=org.qemu.guest_agent.0 -vnc :5 -drive file=/home/image/ubuntu2004_2.img + +5. Bind virtio port to vfio-pci in both two VMs:: + + modprobe vfio enable_unsafe_noiommu_mode=1 + modprobe vfio-pci + echo 1 > /sys/module/vfio/parameters/enable_unsafe_noiommu_mode + ./usertools/dpdk-devbind.py --bind=vfio-pci 00:05.0 + +6. Start testpmd in VMs seperately:: + + ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 1-2 -n 4 -- -i --rxq=1 --txq=1 --nb-cores=1 --txd=1024 --rxd=1024 + +7. Loop packets between two virtio-user sides, record performance number with 64b/2000b/8000b/IMIX pkts can get expected:: + + testpmd0>set fwd mac + testpmd0>start tx_first + testpmd0>stop + testpmd0>set eth-peer 0 52:54:00:00:00:02 + testpmd0>start + testpmd1>set fwd mac + testpmd1>set eth-peer 0 52:54:00:00:00:01 + testpmd1>set txpkts 64 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 2000 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 2000,2000,2000,2000 + testpmd1>start tx_first + testpmd1>show port stats all + testpmd1>stop + testpmd1>set txpkts 64,256,2000,64,256,2000 + testpmd1>start tx_first + testpmd1>show port stats all + +8. Quit two testpmd in two VMs, bind virtio-pmd port to virtio-pci,then bind port back to vfio-pci, rerun below cmd 50 times:: + + ./usertools/dpdk-devbind.py -u 00:05.0 + ./usertools/dpdk-devbind.py --bind=virtio-pci 00:05.0 + ./usertools/dpdk-devbind.py --bind=vfio-pci 00:05.0 + +9. Rerun step 6-7，check vhost can stable work and get expected throughput. + +10. Quit VMs and rerun step 3-9. + +Test Case 10: VM2VM virtio-net split ring test with DSA kernel driver and iperf stable check +-------------------------------------------------------------------------------------------- +This case tests with split ring with DSA kenrel driver in two VMs, check that iperf/scp and reconnection can work stably between two virito-net. + +1. Bind 1 NIC port and 2 DSA devices to idxd:: + + # ./usertools/dpdk-devbind.py -b vfio-pci 6a:00.0 + + ls /mnt/dsa #check wq configure, reset if exist + # ./usertools/dpdk-devbind.py -u 6a:01.0 6f:01.0 + # ./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0 + ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 2 0 + ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 2 1 + ls /mnt/dsa #check wq configure success + +2. On host, launch dpdk-vhost by below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 2-3 -n 4 -a 0000:6a:01.0 \ + -- -p 0x1 --tso 1 --vm2vm 1 --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@wq0.0,rxd0@wq0.1,txd1@wq1.0,rxd1@wq1.1] \ + --client --total-num-mbufs 600000 + +3. Start VM0 with qemu:: + + qemu-system-x86_64 -name vm0 -enable-kvm -pidfile /tmp/.vm0.pid \ + -daemonize -monitor unix:/tmp/vm0_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6000-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net0,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:01,disable-modern=true,mrg_rxbuf=off,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm0_qga0.sock,server,nowait,id=vm0_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm0_qga0,name=org.qemu.guest_agent.0 -vnc :4 -drive file=/home/image/ubuntu2004.img + +4. Start VM1 with qemu:: + + qemu-system-x86_64 -name vm1 -enable-kvm -pidfile /tmp/.vm1.pid \ + -daemonize -monitor unix:/tmp/vm1_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6001-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net1,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:02,disable-modern=false,mrg_rxbuf=off,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm1_qga0.sock,server,nowait,id=vm1_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm1_qga0,name=org.qemu.guest_agent.0 -vnc :5 -drive file=/home/image/ubuntu2004_2.img + +5. On VM1, set virtio device IP and run arp protocal:: + + ifconfig ens5 1.1.1.2 + arp -s 1.1.1.8 52:54:00:00:00:02 + +6. On VM2, set virtio device IP and run arp protocal:: + + ifconfig ens5 1.1.1.8 + arp -s 1.1.1.2 52:54:00:00:00:01 + +7. Check the iperf performance between two VMs by below commands:: + + Under VM1, run: `iperf -s -i 1` + Under VM2, run: `iperf -c 1.1.1.2 -i 1 -t 60` + +8. Check iperf throughput can get x Gbits/sec. + +9. Scp 1MB file form VM0 to VM1, check packets can be forwarding success by scp:: + + Under VM1, run: `scp [xxx] root@1.1.1.8:/` [xxx] is the file name + +10. Relaunch dpdk-vhost, then rerun step 7-9 five times. + +11. Relaunch dpdk-vhost by below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 2-3 -n 4 -a 0000:6a:01.0 -- -p 0x1 --tso 1 --vm2vm 1 \ + --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@wq0.0,rxd1@wq0.1] --client --total-num-mbufs 600000 + +12. rerun step 7-9. + +Test Case 11: VM2VM virtio-net packed ring test with DSA kenrel driver and iperf stable check +--------------------------------------------------------------------------------------------- +This case tests with packed ring with DSA kernel driver in two VMs, check that iperf/scp and reconnection can work stably between two virito-net. + +1. Bind 1 NIC port to vfio-pci and 2 DSA devices to idxd:: + + # ./usertools/dpdk-devbind.py -b vfio-pci 6a:00.0 + + ls /mnt/dsa #check wq configure, reset if exist + # ./usertools/dpdk-devbind.py -u 6a:01.0 6f:01.0 + # ./usertools/dpdk-devbind.py -b idxd 6a:01.0 6f:01.0 + ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 2 0 + ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 2 1 + ls /mnt/dsa #check wq configure success + +2. On host, launch dpdk-vhost by below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 26-28 -n 4 -a 0000:6a:01.0 \ + -- -p 0x1 --mergeable 1 --tso 1 --vm2vm 1 --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@wq0.0,rxd0@wq0.1,txd1@wq1.0,rxd1@wq1.1] \ + --client --total-num-mbufs 600000 + +3. Start VM0 with qemu:: + + qemu-system-x86_64 -name vm0 -enable-kvm -pidfile /tmp/.vm0.pid \ + -daemonize -monitor unix:/tmp/vm0_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6000-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net0,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:01,disable-modern=false,mrg_rxbuf=off,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm0_qga0.sock,server,nowait,id=vm0_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm0_qga0,name=org.qemu.guest_agent.0 -vnc :4 -drive file=/home/image/ubuntu2004.img + +4. Start VM1 with qemu:: + + qemu-system-x86_64 -name vm1 -enable-kvm -pidfile /tmp/.vm1.pid \ + -daemonize -monitor unix:/tmp/vm1_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6001-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net1,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:02,disable-modern=false,mrg_rxbuf=off,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm1_qga0.sock,server,nowait,id=vm1_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm1_qga0,name=org.qemu.guest_agent.0 -vnc :5 -drive file=/home/image/ubuntu2004_2.img + +5. On VM1, set virtio device IP and run arp protocal:: + + ifconfig ens5 1.1.1.2 + arp -s 1.1.1.8 52:54:00:00:00:02 + +6. On VM2, set virtio device IP and run arp protocal:: + + ifconfig ens5 1.1.1.8 + arp -s 1.1.1.2 52:54:00:00:00:01 + +7. Check the iperf performance between two VMs by below commands:: + + Under VM1, run: `iperf -s -i 1` + Under VM2, run: `iperf -c 1.1.1.2 -i 1 -t 60` + +8. Check iperf throughput can get x Gbits/sec. + +9. Scp 1MB file form VM1 to VM2M, check packets can be forwarding success by scp:: + + Under VM1, run: `scp [xxx] root@1.1.1.8:/` [xxx] is the file name + +10. Rerun step 7-9 five times. + +11. Relaunch VMs with mrg_rxbuf=on. + +12. Rerun step 5-10. + +Test Case 12: VM2VM virtio-net packed ring test with 2 DSA WQ with kernel driver and iperf stable check +------------------------------------------------------------------------------------------------------- +This case tests with packed ring with 2 DSA WQ with dpdk driver in two VMs, check that iperf/scp and reconnection can work stably between two virito-net. + +1. Bind 1 NIC port to vfio-pci and 1 DSA device to idxd:: + + # ./usertools/dpdk-devbind.py -b vfio-pci 6a:00.0 + + ls /mnt/dsa #check wq configure, reset if exist + # ./usertools/dpdk-devbind.py -u 6a:01.0 + # ./usertools/dpdk-devbind.py -b idxd 6a:01.0 + ./drivers/dma/idxd/dpdk_idxd_cfg.py -q 2 0 + ls /mnt/dsa #check wq configure success + +2. On host, launch dpdk-vhost by below command:: + + ./x86_64-native-linuxapp-gcc/examples/dpdk-vhost -l 26-28 -n 4 -a 0000:6a:01.0 \ + -- -p 0x1 --mergeable 1 --vm2vm 1 --socket-file /root/dpdk/vhost-net0 --socket-file /root/dpdk/vhost-net1 \ + --dmas [txd0@wq0.0,rxd1@wq0.1] \ + --client --total-num-mbufs 600000 + +3. Start VM0 with qemu:: + + qemu-system-x86_64 -name vm0 -enable-kvm -pidfile /tmp/.vm0.pid \ + -daemonize -monitor unix:/tmp/vm0_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6000-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net0,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:01,disable-modern=false,mrg_rxbuf=off,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm0_qga0.sock,server,nowait,id=vm0_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm0_qga0,name=org.qemu.guest_agent.0 -vnc :4 -drive file=/home/image/ubuntu2004.img + +4. Start VM1 with qemu:: + + qemu-system-x86_64 -name vm1 -enable-kvm -pidfile /tmp/.vm1.pid \ + -daemonize -monitor unix:/tmp/vm1_monitor.sock,server,nowait \ + -netdev user,id=nttsip1,hostfwd=tcp:127.0.0.1:6001-:22 -device e1000,netdev=nttsip1 \ + -chardev socket,id=char0,path=./vhost-net1,server -netdev type=vhost-user,id=netdev0,chardev=char0,vhostforce \ + -device virtio-net-pci,netdev=netdev0,mac=52:54:00:00:00:02,disable-modern=false,mrg_rxbuf=off,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on -cpu host -smp 8 -m 4096 \ + -object memory-backend-file,id=mem,size=4096M,mem-path=/mnt/huge,share=on -numa node,memdev=mem -mem-prealloc -chardev socket,path=/tmp/vm1_qga0.sock,server,nowait,id=vm1_qga0 \ + -device virtio-serial -device virtserialport,chardev=vm1_qga0,name=org.qemu.guest_agent.0 -vnc :5 -drive file=/home/image/ubuntu2004_2.img + +5. On VM1, set virtio device IP and run arp protocal:: + + ifconfig ens5 1.1.1.2 + arp -s 1.1.1.8 52:54:00:00:00:02 + +6. On VM2, set virtio device IP and run arp protocal:: + + ifconfig ens5 1.1.1.8 + arp -s 1.1.1.2 52:54:00:00:00:01 + +7. Check the iperf performance between two VMs by below commands:: + + Under VM1, run: `iperf -s -i 1` + Under VM2, run: `iperf -c 1.1.1.2 -i 1 -t 60` + +8. Check iperf throughput can get x Gbits/sec. + +9. Scp 1MB file form VM1 to VM2M, check packets can be forwarding success by scp:: + + Under VM1, run: `scp [xxx] root@1.1.1.8:/` [xxx] is the file name + +10. Rerun step 7-9 five times. + +11. Relaunch VMs with mrg_rxbuf=on. + +12. Rerun step 5-10. From patchwork Tue Mar 21 02:35:02 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Ling, WeiX" X-Patchwork-Id: 125350 Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id DD41E427EC; Tue, 21 Mar 2023 03:37:06 +0100 (CET) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id D63EB410EE; Tue, 21 Mar 2023 03:37:06 +0100 (CET) Received: from mga14.intel.com (mga14.intel.com [192.55.52.115]) by mails.dpdk.org (Postfix) with ESMTP id 7ACB240693 for ; Tue, 21 Mar 2023 03:37:04 +0100 (CET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1679366224; x=1710902224; h=from:to:cc:subject:date:message-id:mime-version: content-transfer-encoding; bh=CrrTHTee07nqpmJEVUX13zf69NThwSV7ivmiUR9SOtc=; b=TyLa7XF6bhy/sbEqCGG7BjdOr39+QfVMn79+rhs31ULTgyWvgBGfhM2r 9AImjNTdkwBRzPjLIkyAqcStAG5Fa9CfZ7elqn6Q9KDylJR+k5ew2x/dJ 9iThtlnhaST9ckdu2KoOduSFmTVGN3rLMyUjtFFEIrkrjXIwH/45tmE8Z 9XteBsTXXsJUHEQPBwaxSmq63j5WHrHYXI0YJdBAM0lep+l+cX5S5N3EM YQ16umgc6cgUo25mkjUh9gfRqvdVJQCZP+tK7p/pSb66TaIYph8AfamnD 0GqzwAE8c4tW7EbpajBvzOLzdqbwIATs2e7LWisJeKv6nr566q4oq4lWX g==; X-IronPort-AV: E=McAfee;i="6600,9927,10655"; a="338867373" X-IronPort-AV: E=Sophos;i="5.98,277,1673942400"; d="scan'208";a="338867373" Received: from orsmga002.jf.intel.com ([10.7.209.21]) by fmsmga103.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 20 Mar 2023 19:37:03 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10655"; a="681351356" X-IronPort-AV: E=Sophos;i="5.98,277,1673942400"; d="scan'208";a="681351356" Received: from unknown (HELO localhost.localdomain) ([10.239.252.222]) by orsmga002-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 20 Mar 2023 19:37:01 -0700 From: Wei Ling To: dts@dpdk.org Cc: Wei Ling Subject: [dts][PATCH V1 3/3] tests/vswitch_sample_dsa: add new testsuite Date: Tue, 21 Mar 2023 10:35:02 +0800 Message-Id: <20230321023502.3854650-1-weix.ling@intel.com> X-Mailer: git-send-email 2.25.1 MIME-Version: 1.0 X-BeenThere: dts@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: test suite reviews and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dts-bounces@dpdk.org Add vswitch_sample_dsa testsuite to test dpdk-vhost APP with host DMA operation callbacks for DSA PMD and vhost async data-path. Signed-off-by: Wei Ling --- tests/TestSuite_vswitch_sample_dsa.py | 1037 +++++++++++++++++++++++++ 1 file changed, 1037 insertions(+) create mode 100644 tests/TestSuite_vswitch_sample_dsa.py diff --git a/tests/TestSuite_vswitch_sample_dsa.py b/tests/TestSuite_vswitch_sample_dsa.py new file mode 100644 index 00000000..128c761e --- /dev/null +++ b/tests/TestSuite_vswitch_sample_dsa.py @@ -0,0 +1,1037 @@ +# SPDX-License-Identifier: BSD-3-Clause +# Copyright(c) 2023 Intel Corporation +# + +""" +DPDK Test suite. +""" + +import random +import re +import string +import time + +import framework.utils as utils +from framework.pmd_output import PmdOutput +from framework.test_case import TestCase +from framework.virt_common import VM + +from .virtio_common import dsa_common as DC + + +class TestVswitchSampleDSA(TestCase): + def set_up_all(self): + """ + Run at the start of each test suite. + """ + out = self.dut.build_dpdk_apps("./examples/vhost") + self.verify("Error" not in out, "compilation vhost error") + self.vhost_path = self.dut.apps_name["vhost"] + self.tester_tx_port_num = 1 + self.dut_ports = self.dut.get_ports() + self.ports_socket = self.dut.get_numa_id(self.dut_ports[0]) + self.cores = self.dut.get_core_list("all", socket=self.ports_socket) + self.vhost_core_list = self.cores[0:2] + self.vhost_core_range = "%s-%s" % ( + self.vhost_core_list[0], + self.vhost_core_list[-1], + ) + self.vuser0_core_list = self.cores[2:4] + self.vuser1_core_list = self.cores[4:6] + self.mem_channels = self.dut.get_memory_channels() + self.out_path = "/tmp" + out = self.tester.send_expect("ls -d %s" % self.out_path, "# ") + if "No such file or directory" in out: + self.tester.send_expect("mkdir -p %s" % self.out_path, "# ") + self.base_dir = self.dut.base_dir.replace("~", "/root") + self.virtio_dst_mac0 = "00:11:22:33:44:10" + self.virtio_dst_mac1 = "00:11:22:33:44:11" + self.vm_dst_mac0 = "52:54:00:00:00:01" + self.vm_dst_mac1 = "52:54:00:00:00:02" + self.vm_num = 2 + # create an instance to set stream field setting + self.vhost_user = self.dut.new_session(suite="vhost-user") + self.virtio_user0 = self.dut.new_session(suite="virtio-user0") + self.virtio_user1 = self.dut.new_session(suite="virtio-user1") + self.virtio_user0_pmd = PmdOutput(self.dut, self.virtio_user0) + self.virtio_user1_pmd = PmdOutput(self.dut, self.virtio_user1) + self.random_string = string.ascii_letters + string.digits + self.virtio_ip0 = "1.1.1.2" + self.virtio_ip1 = "1.1.1.3" + self.rerun_times = 5 + self.DC = DC(self) + + def set_up(self): + """ + Run before each test case. + """ + self.table_header = ["Frame Size(Byte)", "Mode", "Throughput(Mpps)"] + self.result_table_create(self.table_header) + self.dut.send_expect("rm -rf %s/vhost-net*" % self.base_dir, "#") + self.dut.send_expect("killall -I dpdk-vhost", "#", 20) + self.dut.send_expect("killall -I qemu-system-x86_64", "#", 20) + self.vm_dut = [] + self.vm = [] + self.DC.reset_all_work_queue() + self.DC.bind_all_dsa_to_kernel() + + @property + def check_2M_env(self): + out = self.dut.send_expect( + "cat /proc/meminfo |grep Hugepagesize|awk '{print($2)}'", "# " + ) + return True if out == "2048" else False + + def launch_vhost_app( + self, eal_params, vdev_num, dmas, mergeable=True, tso=False, client_mode=True + ): + """ + launch the vhost app on vhost side + """ + vdev_param = "" + for i in range(vdev_num): + vdev_param += "--socket-file ./vhost-net{} ".format(i) + mergeable_param = "--mergeable 1" if mergeable else "" + tso_param = "--tso 1" if tso else "" + client_param = "--client" if client_mode else "" + cmd = ( + "%s %s -- -p 0x1 %s %s --vm2vm 1 --stats 1 \ + %s --dmas [%s] \ + %s --total-num-mbufs 600000" + % ( + self.vhost_path, + eal_params, + tso_param, + mergeable_param, + vdev_param, + dmas, + client_param, + ) + ) + self.vhost_user.send_command(cmd) + time.sleep(3) + + def start_virtio_testpmd_with_vhost_net0(self, eal_param="", param=""): + """ + launch the testpmd as virtio with vhost_net0 + """ + if self.check_2M_env: + eal_param += " --single-file-segments" + self.virtio_user0_pmd.start_testpmd( + cores=self.vuser0_core_list, + eal_param=eal_param, + param=param, + no_pci=True, + prefix="virtio-user0", + fixed_prefix=True, + ) + + def start_virtio_testpmd_with_vhost_net1(self, eal_param="", param=""): + """ + launch the testpmd as virtio with vhost_net1 + """ + if self.check_2M_env: + eal_param += " --single-file-segments" + self.virtio_user1_pmd.start_testpmd( + cores=self.vuser1_core_list, + eal_param=eal_param, + param=param, + no_pci=True, + prefix="virtio-user1", + fixed_prefix=True, + ) + + def start_vms( + self, + moderns=["true", "true"], + mrg_rxbuf=True, + server_mode=False, + set_target=True, + bind_dev=True, + ): + """ + start two VM, each VM has one virtio device + """ + mrg_rxbuf = "on" if mrg_rxbuf else "off" + for i in range(self.vm_num): + vm_dut = None + setting_format = "disable-modern=%s,mrg_rxbuf=%s,csum=on,guest_csum=on,host_tso4=on,guest_tso4=on,guest_ecn=on" + setting_args = setting_format % (moderns[i], mrg_rxbuf) + vm_info = VM(self.dut, "vm%d" % i, "vhost_sample") + vm_params = {} + vm_params["driver"] = "vhost-user" + if server_mode: + vm_params["opt_path"] = "./vhost-net%d" % i + ",server" + else: + vm_params["opt_path"] = "./vhost-net%d" % i + vm_params["opt_mac"] = "52:54:00:00:00:0%d" % (i + 1) + vm_params["opt_settings"] = setting_args + vm_info.set_vm_device(**vm_params) + time.sleep(3) + try: + vm_dut = vm_info.start(set_target=set_target, bind_dev=bind_dev) + if vm_dut is None: + raise Exception("Set up VM ENV failed") + except Exception as e: + print((utils.RED("Failure for %s" % str(e)))) + raise e + self.vm_dut.append(vm_dut) + self.vm.append(vm_info) + + def start_vm_testpmd(self, pmd_session): + """ + launch the testpmd in vm + """ + self.vm_cores = [1, 2] + param = "--rxq=1 --txq=1 --nb-cores=1 --txd=1024 --rxd=1024" + pmd_session.start_testpmd(cores=self.vm_cores, param=param) + + def repeat_bind_driver(self, dut, repeat_times=50): + i = 0 + while i < repeat_times: + dut.unbind_interfaces_linux() + dut.bind_interfaces_linux(driver="virtio-pci") + dut.bind_interfaces_linux(driver="vfio-pci") + i += 1 + + def let_vswitch_know_mac(self, virtio_pmd, relaunch=False): + if not relaunch: + virtio_pmd.execute_cmd("set fwd mac") + virtio_pmd.execute_cmd("start tx_first") + else: + virtio_pmd.execute_cmd("stop") + virtio_pmd.execute_cmd("start tx_first") + + def get_receive_throughput(self, pmd_session, count=10): + i = 0 + while i < count: + pmd_session.execute_cmd("show port stats all") + i += 1 + else: + out = pmd_session.execute_cmd("show port stats all") + pmd_session.execute_cmd("stop") + rx_throughput = re.compile("Rx-pps: \s+(.*?)\s+?").findall(out, re.S) + return float(rx_throughput[0]) / 1000000.0 + + def set_testpmd0_param(self, pmd_session, eth_peer_mac): + pmd_session.execute_cmd("set fwd mac") + pmd_session.execute_cmd("start tx_first") + pmd_session.execute_cmd("stop") + pmd_session.execute_cmd("set eth-peer 0 %s" % eth_peer_mac) + pmd_session.execute_cmd("start") + + def set_testpmd1_param(self, pmd_session, eth_peer_mac): + pmd_session.execute_cmd("set fwd mac") + pmd_session.execute_cmd("set eth-peer 0 %s" % eth_peer_mac) + + def send_pkts_from_testpmd1(self, pmd_session, pkt_len): + if pkt_len in [64, 2000]: + pmd_session.execute_cmd("set txpkts %s" % pkt_len) + elif pkt_len == 8000: + pmd_session.execute_cmd("set txpkts 2000,2000,2000,2000") + elif pkt_len == "imix": + pmd_session.execute_cmd("set txpkts 64,256,2000,64,256,2000") + pmd_session.execute_cmd("start tx_first") + + def vm2vm_check_with_two_dsa(self): + frame_sizes = [64, 2000, 8000, "imix"] + self.set_testpmd0_param(self.virtio_user0_pmd, self.virtio_dst_mac1) + self.set_testpmd1_param(self.virtio_user1_pmd, self.virtio_dst_mac0) + + rx_throughput = {} + for frame_size in frame_sizes: + self.send_pkts_from_testpmd1( + pmd_session=self.virtio_user1_pmd, pkt_len=frame_size + ) + rx_pps = self.get_receive_throughput(pmd_session=self.virtio_user1_pmd) + rx_throughput[frame_size] = rx_pps + return rx_throughput + + def test_vm2vm_virtio_user_forwarding_test_using_dsa_dpdk_driver(self): + """ + Test Case 1: VM2VM virtio-user forwarding test when vhost async operation using DSA dpdk driver + """ + perf_result = [] + self.used_dsa = self.DC.bind_dsa_to_dpdk( + dsa_number=1, driver_name="vfio-pci", socket=self.ports_socket + ) + ports = self.used_dsa + ports.append(self.dut.ports_info[0]["pci"]) + port_options = {self.used_dsa[0]: "max_queues=4"} + eal_params = self.dut.create_eal_parameters( + cores=self.vhost_core_list, ports=ports, port_options=port_options + ) + dmas = "txd0@%s-q0,rxd0@%s-q1,txd1@%s-q2,rxd1@%s-q3" % ( + self.used_dsa[0], + self.used_dsa[0], + self.used_dsa[0], + self.used_dsa[0], + ) + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + virtio0_eal_param = "--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=1,server=1,mrg_rxbuf=1,in_order=0,packed_vq=1" + virtio0_param = "--rxq=1 --txq=1 --txd=1024 --rxd=1024 --nb-cores=1" + self.start_virtio_testpmd_with_vhost_net0( + eal_param=virtio0_eal_param, param=virtio0_param + ) + + virtio1_eal_param = "--vdev=net_virtio_user0,mac=00:11:22:33:44:11,path=./vhost-net1,queues=1,server=1,mrg_rxbuf=1,in_order=1,vectorized=1" + virtio1_param = "--rxq=1 --txq=1 --txd=1024 --rxd=1024 --nb-cores=1" + self.start_virtio_testpmd_with_vhost_net1( + eal_param=virtio1_eal_param, param=virtio1_param + ) + before_relunch_result = self.vm2vm_check_with_two_dsa() + + self.vhost_user.send_expect("^C", "# ", 20) + dmas = "txd0@%s-q0,rxd1@%s-q1" % (self.used_dsa[0], self.used_dsa[0]) + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + self.virtio_user0_pmd.execute_cmd("stop") + after_relunch_result = self.vm2vm_check_with_two_dsa() + + for key in before_relunch_result.keys(): + perf_result.append( + [key, "Before Re-launch vhost", before_relunch_result[key]] + ) + for key in after_relunch_result.keys(): + perf_result.append( + [key, "After Re-launch vhost ", after_relunch_result[key]] + ) + for table_row in perf_result: + self.result_table_add(table_row) + self.result_table_print() + + def vm2vm_check_with_two_vhost_device(self): + rx_throughput = {} + self.frame_sizes = [64, 2000, 8000, "imix"] + for frame_size in self.frame_sizes: + self.send_pkts_from_testpmd1(pmd_session=self.vm1_pmd, pkt_len=frame_size) + rx_pps = self.get_receive_throughput(pmd_session=self.vm1_pmd) + rx_throughput[frame_size] = rx_pps + return rx_throughput + + def start_vms_testpmd_and_test( + self, moderns=["true", "true"], mrg_rxbuf=True, need_start_vm=True, packed=False + ): + if need_start_vm: + self.start_vms( + moderns=moderns, + mrg_rxbuf=mrg_rxbuf, + server_mode=True, + set_target=True, + bind_dev=True, + ) + self.vm0_pmd = PmdOutput(self.vm_dut[0]) + self.vm1_pmd = PmdOutput(self.vm_dut[1]) + self.start_vm_testpmd(self.vm0_pmd) + self.start_vm_testpmd(self.vm1_pmd) + self.set_testpmd0_param(self.vm0_pmd, self.vm_dst_mac1) + self.set_testpmd1_param(self.vm1_pmd, self.vm_dst_mac0) + perf_result = self.vm2vm_check_with_two_vhost_device() + self.vm0_pmd.quit() + self.vm1_pmd.quit() + return perf_result + + def test_vm2vm_virtio_pmd_split_ring_test_with_dsa_dpdk_driver_register_and_unregister_stable_check( + self, + ): + """ + Test Case 2: VM2VM virtio-pmd split ring test with DSA dpdk driver register/unregister stable check + """ + perf_result = [] + self.used_dsa = self.DC.bind_dsa_to_dpdk( + dsa_number=1, driver_name="vfio-pci", socket=self.ports_socket + ) + ports = self.used_dsa + ports.append(self.dut.ports_info[0]["pci"]) + port_options = {self.used_dsa[0]: "max_queues=4"} + eal_params = self.dut.create_eal_parameters( + cores=self.vhost_core_list, ports=ports, port_options=port_options + ) + dmas = "txd0@%s-q0,rxd0@%s-q1,txd1@%s-q2,rxd1@%s-q3" % ( + self.used_dsa[0], + self.used_dsa[0], + self.used_dsa[0], + self.used_dsa[0], + ) + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + + before_rebind = self.start_vms_testpmd_and_test( + moderns=["true", "false"], mrg_rxbuf=True, need_start_vm=True, packed=False + ) + + # repeat bind 50 time from virtio-pci to vfio-pci + self.repeat_bind_driver(dut=self.vm_dut[0], repeat_times=50) + self.repeat_bind_driver(dut=self.vm_dut[1], repeat_times=50) + + self.vhost_user.send_expect("^C", "# ", 20) + dmas = "txd0@%s-q0,rxd1@%s-q3" % (self.used_dsa[0], self.used_dsa[0]) + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + + after_rebind = self.start_vms_testpmd_and_test(need_start_vm=False) + + for key in before_rebind.keys(): + perf_result.append([key, "Before rebind driver", before_rebind[key]]) + + for key in after_rebind.keys(): + perf_result.append([key, "After rebind driver", after_rebind[key]]) + + for table_row in perf_result: + self.result_table_add(table_row) + self.result_table_print() + + for i in perf_result: + self.verify(i[2] > 0, "%s Frame Size(Byte) is less than 0 Mpps" % i[0]) + + def test_vm2vm_virtio_pmd_packed_ring_test_with_dsa_dpdk_driver_register_and_unregister_stable_check( + self, + ): + """ + Test Case 3: VM2VM virtio-pmd packed ring test with DSA dpdk driver register/unregister stable check + """ + perf_result = [] + self.used_dsa = self.DC.bind_dsa_to_dpdk( + dsa_number=1, driver_name="vfio-pci", socket=self.ports_socket + ) + ports = self.used_dsa + ports.append(self.dut.ports_info[0]["pci"]) + port_options = {self.used_dsa[0]: "max_queues=4"} + eal_params = self.dut.create_eal_parameters( + cores=self.vhost_core_list, ports=ports, port_options=port_options + ) + dmas = "txd0@%s-q0,rxd0@%s-q1,txd1@%s-q2,rxd1@%s-q3" % ( + self.used_dsa[0], + self.used_dsa[0], + self.used_dsa[0], + self.used_dsa[0], + ) + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + + before_rebind = self.start_vms_testpmd_and_test( + moderns=["false", "false"], mrg_rxbuf=True, need_start_vm=True, packed=True + ) + + # repeat bind 50 time from virtio-pci to vfio-pci + self.repeat_bind_driver(dut=self.vm_dut[0], repeat_times=50) + self.repeat_bind_driver(dut=self.vm_dut[1], repeat_times=50) + + after_rebind = self.start_vms_testpmd_and_test(need_start_vm=False) + + for key in before_rebind.keys(): + perf_result.append([key, "Before rebind driver", before_rebind[key]]) + + for key in after_rebind.keys(): + perf_result.append([key, "After rebind driver", after_rebind[key]]) + + for table_row in perf_result: + self.result_table_add(table_row) + self.result_table_print() + + for i in perf_result: + self.verify(i[2] > 0, "%s Frame Size(Byte) is less than 0 Mpps" % i[0]) + + def config_vm_env(self): + """ + set virtio device IP and run arp protocal + """ + vm0_intf = self.vm_dut[0].ports_info[0]["intf"] + vm1_intf = self.vm_dut[1].ports_info[0]["intf"] + self.vm_dut[0].send_expect( + "ifconfig %s %s" % (vm0_intf, self.virtio_ip0), "#", 10 + ) + self.vm_dut[1].send_expect( + "ifconfig %s %s" % (vm1_intf, self.virtio_ip1), "#", 10 + ) + self.vm_dut[0].send_expect( + "arp -s %s %s" % (self.virtio_ip1, self.vm_dst_mac1), "#", 10 + ) + self.vm_dut[1].send_expect( + "arp -s %s %s" % (self.virtio_ip0, self.vm_dst_mac0), "#", 10 + ) + + def start_iperf_test(self): + """ + run perf command between to vms + """ + iperf_server = "iperf -f g -s -i 1" + iperf_client = "iperf -f g -c 1.1.1.2 -i 1 -t 60" + self.vm_dut[0].send_expect("%s > iperf_server.log &" % iperf_server, "", 10) + self.vm_dut[1].send_expect("%s > iperf_client.log &" % iperf_client, "", 60) + time.sleep(90) + + def get_iperf_result(self): + """ + get the iperf test result + """ + self.vm_dut[0].send_expect("pkill iperf", "# ") + self.vm_dut[1].session.copy_file_from("%s/iperf_client.log" % self.dut.base_dir) + fp = open("./iperf_client.log") + fmsg = fp.read() + fp.close() + # remove the server report info from msg + index = fmsg.find("Server Report") + if index != -1: + fmsg = fmsg[:index] + iperfdata = re.compile("\S*\s*[M|G]bits/sec").findall(fmsg) + # the last data of iperf is the ave data from 0-30 sec + self.verify(len(iperfdata) != 0, "The iperf data between to vms is 0") + self.logger.info("The iperf data between vms is %s" % iperfdata[-1]) + self.verify( + (iperfdata[-1].split()[1]) == "Gbits/sec" + and float(iperfdata[-1].split()[0]) >= 1, + "the throughput must be above 1Gbits/sec", + ) + # rm the iperf log file in vm + self.vm_dut[0].send_expect("rm iperf_server.log", "#", 10) + self.vm_dut[1].send_expect("rm iperf_client.log", "#", 10) + return float(iperfdata[-1].split()[0]) + + def check_scp_file_valid_between_vms(self, file_size=1024): + """ + scp file form VM1 to VM2, check the data is valid + """ + # default file_size=1024K + data = "" + for _ in range(file_size * 1024): + data += random.choice(self.random_string) + self.vm_dut[0].send_expect('echo "%s" > /tmp/payload' % data, "# ") + # scp this file to vm1 + out = self.vm_dut[1].send_command( + "scp root@%s:/tmp/payload /root" % self.virtio_ip0, timeout=5 + ) + if "Are you sure you want to continue connecting" in out: + self.vm_dut[1].send_command("yes", timeout=3) + self.vm_dut[1].send_command(self.vm[0].password, timeout=3) + # get the file info in vm1, and check it valid + md5_send = self.vm_dut[0].send_expect("md5sum /tmp/payload", "# ") + md5_revd = self.vm_dut[1].send_expect("md5sum /root/payload", "# ") + md5_send = md5_send[: md5_send.find(" ")] + md5_revd = md5_revd[: md5_revd.find(" ")] + self.verify( + md5_send == md5_revd, "the received file is different with send file" + ) + + def start_iperf_and_scp_test_in_vms( + self, + moderns=["true", "true"], + mrg_rxbuf=False, + need_start_vm=True, + packed=False, + server_mode=False, + ): + if need_start_vm: + self.start_vms( + moderns=moderns, + mrg_rxbuf=mrg_rxbuf, + server_mode=server_mode, + set_target=True, + bind_dev=False, + ) + self.vm0_pmd = PmdOutput(self.vm_dut[0]) + self.vm1_pmd = PmdOutput(self.vm_dut[1]) + self.config_vm_env() + self.check_scp_file_valid_between_vms() + self.start_iperf_test() + iperfdata = self.get_iperf_result() + return iperfdata + + def test_vm2vm_virtio_net_split_ring_test_with_dsa_dpdk_driver_and_iperf_stable_check( + self, + ): + """ + Test Case 4: VM2VM virtio-net split ring test with DSA dpdk driver and iperf stable check + """ + perf_result = [] + self.used_dsa = self.DC.bind_dsa_to_dpdk( + dsa_number=1, driver_name="vfio-pci", socket=self.ports_socket + ) + ports = self.used_dsa + ports.append(self.dut.ports_info[0]["pci"]) + port_options = {self.used_dsa[0]: "max_queues=2"} + eal_params = self.dut.create_eal_parameters( + cores=self.vhost_core_list, ports=ports, port_options=port_options + ) + dmas = "txd0@%s-q0,rxd0@%s-q1,txd1@%s-q0,rxd1@%s-q1" % ( + self.used_dsa[0], + self.used_dsa[0], + self.used_dsa[0], + self.used_dsa[0], + ) + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + + before_relaunch = self.start_iperf_and_scp_test_in_vms( + moderns=["true", "false"], + mrg_rxbuf=False, + need_start_vm=True, + packed=False, + server_mode=True, + ) + perf_result.append(["split ring", "Before relaunch test", before_relaunch]) + + self.vhost_user.send_expect("^C", "# ", 20) + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + + for _ in range(self.rerun_times): + rerun_result = self.start_iperf_and_scp_test_in_vms(need_start_vm=False) + perf_result.append(["split ring", "Before relaunch test", rerun_result]) + + self.vhost_user.send_expect("^C", "# ", 20) + dmas = "txd0@%s-q0,rxd1@%s-q1" % (self.used_dsa[0], self.used_dsa[0]) + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + + after_relaunch = self.start_iperf_and_scp_test_in_vms(need_start_vm=False) + + perf_result.append(["split ring", "After relaunch test", after_relaunch]) + + for table_row in perf_result: + self.result_table_add(table_row) + self.result_table_print() + + def test_vm2vm_virtio_net_packed_ring_test_with_dsa_dpdk_driver_and_iperf_stable_check( + self, + ): + """ + Test Case 5: VM2VM virtio-net packed ring test with DSA dpdk driver and iperf stable check + """ + perf_result = [] + self.used_dsa = self.DC.bind_dsa_to_dpdk( + dsa_number=1, driver_name="vfio-pci", socket=self.ports_socket + ) + ports = self.used_dsa + ports.append(self.dut.ports_info[0]["pci"]) + port_options = {self.used_dsa[0]: "max_queues=2"} + eal_params = self.dut.create_eal_parameters( + cores=self.vhost_core_list, ports=ports, port_options=port_options + ) + dmas = "txd0@%s-q0,rxd0@%s-q1,txd1@%s-q0,rxd1@%s-q1" % ( + self.used_dsa[0], + self.used_dsa[0], + self.used_dsa[0], + self.used_dsa[0], + ) + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + + before_relaunch = self.start_iperf_and_scp_test_in_vms( + moderns=["false", "false"], + mrg_rxbuf=False, + need_start_vm=True, + packed=True, + server_mode=True, + ) + perf_result.append(["packed ring", "Before relaunch test", before_relaunch]) + + for _ in range(self.rerun_times): + rerun_result = self.start_iperf_and_scp_test_in_vms(need_start_vm=False) + perf_result.append(["packed ring", "Before relaunch test", rerun_result]) + + self.stop_vms() + after_relaunch = self.start_iperf_and_scp_test_in_vms( + moderns=["false", "false"], + mrg_rxbuf=True, + need_start_vm=True, + packed=True, + server_mode=True, + ) + + perf_result.append(["packed ring", "After relaunch test", after_relaunch]) + + for table_row in perf_result: + self.result_table_add(table_row) + self.result_table_print() + + def test_vm2vm_virtio_net_packed_ring_test_with_2_dsa_wq_with_dpdk_driver_and_iperf_stable_check( + self, + ): + """ + Test Case 6: VM2VM virtio-net packed ring test with 2 DSA WQ with dpdk driver and iperf stable check + """ + perf_result = [] + self.used_dsa = self.DC.bind_dsa_to_dpdk( + dsa_number=1, driver_name="vfio-pci", socket=self.ports_socket + ) + ports = self.used_dsa + ports.append(self.dut.ports_info[0]["pci"]) + port_options = {self.used_dsa[0]: "max_queues=2"} + eal_params = self.dut.create_eal_parameters( + cores=self.vhost_core_list, ports=ports, port_options=port_options + ) + dmas = "txd0@%s-q0,rxd1@%s-q1" % (self.used_dsa[0], self.used_dsa[0]) + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + + before_relaunch = self.start_iperf_and_scp_test_in_vms( + moderns=["false", "false"], + mrg_rxbuf=False, + need_start_vm=True, + packed=True, + server_mode=True, + ) + perf_result.append(["packed ring", "Before relaunch test", before_relaunch]) + + for _ in range(self.rerun_times): + rerun_result = self.start_iperf_and_scp_test_in_vms(need_start_vm=False) + perf_result.append(["packed ring", "Before relaunch test", rerun_result]) + + self.stop_vms() + after_relaunch = self.start_iperf_and_scp_test_in_vms( + moderns=["false", "false"], + mrg_rxbuf=True, + need_start_vm=True, + packed=True, + server_mode=True, + ) + perf_result.append(["packed ring", "After relaunch test", after_relaunch]) + + for _ in range(self.rerun_times): + rerun_result = self.start_iperf_and_scp_test_in_vms(need_start_vm=False) + perf_result.append(["packed ring", "After relaunch test", rerun_result]) + + for table_row in perf_result: + self.result_table_add(table_row) + self.result_table_print() + + def test_vm2vm_virtio_user_forwarding_test_using_dsa_kernel_driver(self): + """ + Test Case 7: VM2VM virtio-user forwarding test when vhost async operation using DSA kernel driver + """ + perf_result = [] + self.DC.create_work_queue(work_queue_number=4, dsa_index=0) + ports = [self.dut.ports_info[0]["pci"]] + eal_params = self.dut.create_eal_parameters( + cores=self.vhost_core_list, ports=ports + ) + dmas = "txd0@wq0.0,rxd0@wq0.1,txd1@wq0.2,rxd1@wq0.3" + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + virtio0_eal_param = "--vdev=net_virtio_user0,mac=00:11:22:33:44:10,path=./vhost-net0,queues=1,server=1,mrg_rxbuf=1,in_order=0,packed_vq=1" + virtio0_param = "--rxq=1 --txq=1 --txd=1024 --rxd=1024 --nb-cores=1" + self.start_virtio_testpmd_with_vhost_net0( + eal_param=virtio0_eal_param, param=virtio0_param + ) + + virtio1_eal_param = "--vdev=net_virtio_user0,mac=00:11:22:33:44:11,path=./vhost-net1,queues=1,server=1,mrg_rxbuf=1,in_order=1,vectorized=1" + virtio1_param = "--rxq=1 --txq=1 --txd=1024 --rxd=1024 --nb-cores=1" + self.start_virtio_testpmd_with_vhost_net1( + eal_param=virtio1_eal_param, param=virtio1_param + ) + before_relunch_result = self.vm2vm_check_with_two_dsa() + + self.vhost_user.send_expect("^C", "# ", 20) + dmas = "txd0@wq0.0,rxd1@wq0.1" + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + self.virtio_user0_pmd.execute_cmd("stop") + after_relunch_result = self.vm2vm_check_with_two_dsa() + + for key in before_relunch_result.keys(): + perf_result.append( + [key, "Before Re-launch vhost", before_relunch_result[key]] + ) + for key in after_relunch_result.keys(): + perf_result.append( + [key, "After Re-launch vhost ", after_relunch_result[key]] + ) + for table_row in perf_result: + self.result_table_add(table_row) + self.result_table_print() + + def test_vm2vm_virtio_pmd_split_ring_test_with_dsa_kernle_driver_register_and_unregister_stable_check( + self, + ): + """ + Test Case 8: VM2VM virtio-pmd split ring test with DSA kernel driver register/unregister stable check + """ + perf_result = [] + self.DC.create_work_queue(work_queue_number=4, dsa_index=0) + ports = [self.dut.ports_info[0]["pci"]] + eal_params = self.dut.create_eal_parameters( + cores=self.vhost_core_list, ports=ports + ) + dmas = "txd0@wq0.0,rxd0@wq0.1,txd1@wq0.2,rxd1@wq0.3" + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + + before_rebind = self.start_vms_testpmd_and_test( + moderns=["false", "true"], mrg_rxbuf=True, need_start_vm=True, packed=False + ) + + # repeat bind 50 time from virtio-pci to vfio-pci + self.repeat_bind_driver(dut=self.vm_dut[0], repeat_times=50) + self.repeat_bind_driver(dut=self.vm_dut[1], repeat_times=50) + + self.vhost_user.send_expect("^C", "# ", 20) + dmas = "txd0@wq0.0,rxd1@wq0.3" + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + + after_rebind = self.start_vms_testpmd_and_test(need_start_vm=False) + + for key in before_rebind.keys(): + perf_result.append([key, "Before rebind driver", before_rebind[key]]) + + for key in after_rebind.keys(): + perf_result.append([key, "After rebind driver", after_rebind[key]]) + + for table_row in perf_result: + self.result_table_add(table_row) + self.result_table_print() + + for i in perf_result: + self.verify(i[2] > 0, "%s Frame Size(Byte) is less than 0 Mpps" % i[0]) + + def test_vm2vm_virtio_pmd_packed_ring_test_with_dsa_kernel_driver_register_and_unregister_stable_check( + self, + ): + """ + Test Case 9: VM2VM virtio-pmd packed ring test with DSA kernel driver register/unregister stable check + """ + perf_result = [] + self.DC.create_work_queue(work_queue_number=2, dsa_index=0) + self.DC.create_work_queue(work_queue_number=2, dsa_index=1) + ports = [self.dut.ports_info[0]["pci"]] + eal_params = self.dut.create_eal_parameters( + cores=self.vhost_core_list, ports=ports + ) + dmas = "txd0@wq0.0,rxd0@wq0.1,txd1@wq1.0,rxd1@wq1.1" + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + + before_rebind = self.start_vms_testpmd_and_test( + moderns=["false", "false"], mrg_rxbuf=True, need_start_vm=True, packed=True + ) + + # repeat bind 50 time from virtio-pci to vfio-pci + self.repeat_bind_driver(dut=self.vm_dut[0], repeat_times=50) + self.repeat_bind_driver(dut=self.vm_dut[1], repeat_times=50) + + after_rebind = self.start_vms_testpmd_and_test(need_start_vm=False) + + for key in before_rebind.keys(): + perf_result.append([key, "Before rebind driver", before_rebind[key]]) + + for key in after_rebind.keys(): + perf_result.append([key, "After rebind driver", after_rebind[key]]) + + for table_row in perf_result: + self.result_table_add(table_row) + self.result_table_print() + + for i in perf_result: + self.verify(i[2] > 0, "%s Frame Size(Byte) is less than 0 Mpps" % i[0]) + + def test_vm2vm_virtio_net_split_ring_test_with_dsa_kernel_driver_and_iperf_stable_check( + self, + ): + """ + Test Case 10: VM2VM virtio-net split ring test with DSA kernel driver and iperf stable check + """ + perf_result = [] + self.DC.create_work_queue(work_queue_number=2, dsa_index=0) + self.DC.create_work_queue(work_queue_number=2, dsa_index=1) + ports = [self.dut.ports_info[0]["pci"]] + eal_params = self.dut.create_eal_parameters( + cores=self.vhost_core_list, ports=ports + ) + dmas = "txd0@wq0.0,rxd0@wq0.1,txd1@wq1.0,rxd1@wq1.1" + self.launch_vhost_app( + eal_params=eal_params, + vdev_num=2, + dmas=dmas, + mergeable=False, + tso=True, + client_mode=True, + ) + + before_relaunch = self.start_iperf_and_scp_test_in_vms( + moderns=["true", "false"], + mrg_rxbuf=False, + need_start_vm=True, + packed=False, + server_mode=True, + ) + perf_result.append(["split ring", "Before relaunch test", before_relaunch]) + + self.vhost_user.send_expect("^C", "# ", 20) + self.launch_vhost_app( + eal_params=eal_params, + vdev_num=2, + dmas=dmas, + mergeable=False, + tso=True, + client_mode=True, + ) + + for _ in range(self.rerun_times): + rerun_result = self.start_iperf_and_scp_test_in_vms(need_start_vm=False) + perf_result.append(["split ring", "Before relaunch test", rerun_result]) + + self.vhost_user.send_expect("^C", "# ", 20) + dmas = "txd0@wq0.0,rxd1@wq0.1" + self.launch_vhost_app( + eal_params=eal_params, + vdev_num=2, + dmas=dmas, + mergeable=False, + tso=True, + client_mode=True, + ) + + after_relaunch = self.start_iperf_and_scp_test_in_vms( + moderns=["true", "false"], + mrg_rxbuf=False, + need_start_vm=False, + packed=False, + server_mode=True, + ) + + perf_result.append(["split ring", "After relaunch test", after_relaunch]) + + for table_row in perf_result: + self.result_table_add(table_row) + self.result_table_print() + + def test_vm2vm_virtio_net_packed_ring_test_with_dsa_kernel_driver_and_iperf_stable_check( + self, + ): + """ + Test Case 11: VM2VM virtio-net packed ring test with DSA kenrel driver and iperf stable check + """ + perf_result = [] + self.DC.create_work_queue(work_queue_number=2, dsa_index=0) + self.DC.create_work_queue(work_queue_number=2, dsa_index=1) + ports = [self.dut.ports_info[0]["pci"]] + eal_params = self.dut.create_eal_parameters( + cores=self.vhost_core_list, ports=ports + ) + dmas = "txd0@wq0.0,rxd0@wq0.1,txd1@wq1.0,rxd1@wq1.1" + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, tso=True, client_mode=True + ) + + before_relaunch = self.start_iperf_and_scp_test_in_vms( + moderns=["false", "false"], + mrg_rxbuf=False, + need_start_vm=True, + packed=True, + server_mode=True, + ) + perf_result.append(["packed ring", "Before relaunch test", before_relaunch]) + + for _ in range(self.rerun_times): + rerun_result = self.start_iperf_and_scp_test_in_vms(need_start_vm=False) + perf_result.append(["packed ring", "Before relaunch test", rerun_result]) + + self.stop_vms() + after_relaunch = self.start_iperf_and_scp_test_in_vms( + moderns=["false", "false"], + mrg_rxbuf=True, + need_start_vm=True, + packed=True, + server_mode=True, + ) + + perf_result.append(["packed ring", "After relaunch test", after_relaunch]) + + for table_row in perf_result: + self.result_table_add(table_row) + self.result_table_print() + + def test_vm2vm_virtio_net_packed_ring_test_with_2_dsa_wq_with_kernel_driver_and_iperf_stable_check( + self, + ): + """ + Test Case 12: VM2VM virtio-net packed ring test with 2 DSA WQ with kernel driver and iperf stable check + """ + perf_result = [] + self.DC.create_work_queue(work_queue_number=2, dsa_index=0) + ports = [self.dut.ports_info[0]["pci"]] + eal_params = self.dut.create_eal_parameters( + cores=self.vhost_core_list, ports=ports + ) + dmas = "txd0@wq0.0,rxd1@wq0.1" + self.launch_vhost_app( + eal_params=eal_params, vdev_num=2, dmas=dmas, client_mode=True + ) + + before_relaunch = self.start_iperf_and_scp_test_in_vms( + moderns=["false", "false"], + mrg_rxbuf=False, + need_start_vm=True, + packed=True, + server_mode=True, + ) + perf_result.append(["packed ring", "Before relaunch test", before_relaunch]) + + for _ in range(self.rerun_times): + rerun_result = self.start_iperf_and_scp_test_in_vms(need_start_vm=False) + perf_result.append(["packed ring", "Before relaunch test", rerun_result]) + + self.stop_vms() + after_relaunch = self.start_iperf_and_scp_test_in_vms( + moderns=["false", "false"], + mrg_rxbuf=True, + need_start_vm=True, + packed=True, + server_mode=True, + ) + perf_result.append(["packed ring", "After relaunch test", after_relaunch]) + + for _ in range(self.rerun_times): + rerun_result = self.start_iperf_and_scp_test_in_vms(need_start_vm=False) + perf_result.append(["packed ring", "After relaunch test", rerun_result]) + + for table_row in perf_result: + self.result_table_add(table_row) + self.result_table_print() + + def stop_vms(self): + for i in range(len(self.vm)): + self.vm[i].stop() + self.vm_dut = [] + self.vm = [] + + def close_all_session(self): + if getattr(self, "vhost_user", None): + self.dut.close_session(self.vhost_user) + if getattr(self, "virtio-user0", None): + self.dut.close_session(self.virtio_user0) + if getattr(self, "virtio-user1", None): + self.dut.close_session(self.virtio_user1) + + def tear_down(self): + """ + Run after each test case. + """ + self.stop_vms() + self.vhost_user.send_expect("^C", "# ", 20) + self.dut.kill_all() + self.DC.reset_all_work_queue() + self.DC.bind_all_dsa_to_kernel() + + def tear_down_all(self): + """ + Run after each test suite. + """ + self.close_all_session()