From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from dpdk.org (dpdk.org [92.243.14.124]) by inbox.dpdk.org (Postfix) with ESMTP id 3A04EA04EA; Fri, 29 Nov 2019 07:53:04 +0100 (CET) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id 2B7D12B88; Fri, 29 Nov 2019 07:53:04 +0100 (CET) Received: from mga17.intel.com (mga17.intel.com [192.55.52.151]) by dpdk.org (Postfix) with ESMTP id 390999E4 for ; Fri, 29 Nov 2019 07:53:02 +0100 (CET) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga007.jf.intel.com ([10.7.209.58]) by fmsmga107.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 28 Nov 2019 22:53:01 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.69,256,1571727600"; d="scan'208";a="199726530" Received: from dpdk-qifu-1.sh.intel.com ([10.67.119.88]) by orsmga007.jf.intel.com with ESMTP; 28 Nov 2019 22:53:00 -0800 From: qifu To: dts@dpdk.org Cc: qifu Date: Fri, 29 Nov 2019 15:41:52 +0000 Message-Id: <20191129154152.39796-1-qi.fu@intel.com> X-Mailer: git-send-email 2.17.1 Subject: [dts] [PATCH V1] test_plans: add test plan for cvl rteflow priority X-BeenThere: dts@dpdk.org X-Mailman-Version: 2.1.15 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 Sender: "dts" add test plan for cvl rteflow priority. Signed-off-by: qifu --- test_plans/rteflow_priority_test_plan.rst | 333 ++++++++++++++++++++++ 1 file changed, 333 insertions(+) create mode 100644 test_plans/rteflow_priority_test_plan.rst diff --git a/test_plans/rteflow_priority_test_plan.rst b/test_plans/rteflow_priority_test_plan.rst new file mode 100644 index 0000000..81d78d2 --- /dev/null +++ b/test_plans/rteflow_priority_test_plan.rst @@ -0,0 +1,333 @@ +.. Copyright (c) <2010-2017>, Intel Corporation + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions + are met: + + - Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + + - Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in + the documentation and/or other materials provided with the + distribution. + + - Neither the name of Intel Corporation nor the names of its + contributors may be used to endorse or promote products derived + from this software without specific prior written permission. + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS + FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE + COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, + INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR + SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, + STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED + OF THE POSSIBILITY OF SUCH DAMAGE. + + +=========================== +CVL Rte_flow Priority Tests +=========================== + + +Description +=========== + +This document provides the plan for testing the Rte_flow Priority feature. +this feature uses devargs as a hint to active flow priority or not. +when priority is not active, flows are created on fdir then switch/ACL. +when priority is active, flows are identified into 2 category: +High priority as permission stage that maps to switch/ACL, +Low priority as distribution stage that maps to fdir, +a no destination high priority rule is not acceptable, since it may be overwritten by a low priority rule due to cvl FXP behavior. + + +Prerequisites +============= + +If using vfio the kernel must be >= 3.6+ and VT-d must be enabled in bios.When +using vfio, use the following commands to load the vfio driver and bind it +to the device under test:: + + modprobe vfio + modprobe vfio-pci + usertools/dpdk-devbind.py --bind=vfio-pci device_bus_id + +bind the pf to dpdk driver:: + ./usertools/dpdk-devbind.py -b vfio-pci af:00.0 + + +Test case: Setting Priority in non-pipeline mode +=================================================== + +Priority is not active in non-pipeline mode. The default value of priority is 0 but it will be ignored + +Start the ``testpmd`` application as follows:: + + ./x86_64-native-linuxapp-gcc/app/testpmd -c 0xf -n 4 -w 0000:af:00.0 -- -i --txq=8 --rxq=8 + set fwd rxonly + set verbose 1 + +create a rule:: + + flow create 0 priority 0 ingress pattern eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 / end actions queue index 2 / end + +check the flow can be created but it will map to fdir filter. + +create a rule:: + + flow create 0 priority 1 ingress pattern eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 / end actions queue index 2 / end + +check the flow can not be created for the vallue of priority is 0 in non-pipeline mode. + +Start the ``testpmd`` application as follows:: + + ./x86_64-native-linuxapp-gcc/app/testpmd -c 0xf -n 4 -w 0000:af:00.0,pipeline-mode-support=0 -- -i --txq=8 --rxq=8 + set fwd rxonly + set verbose 1 + +create a rule:: + + flow create 0 priority 0 ingress pattern eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 / end actions queue index 2 / end + +check the flow can be created but it will map to fdir filter. + +create a rule:: + + flow create 0 priority 1 ingress pattern eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 / end actions queue index 2 / end + +check the flow can not be created. + + +Test case: Create flow rules with priority in pipeline mode +============================================================= + +Priority is active in pipeline mode. +create flow rules and set priority 0/1 to map switch/fdir filter separately. + +Start the ``testpmd`` application as follows:: + + ./x86_64-native-linuxapp-gcc/app/testpmd -c 0xf -n 4 -w 0000:af:00.0,pipeline-mode-support=1 -- -i --txq=8 --rxq=8 + set fwd rxonly + set verbose 1 + rx_vxlan_port add 4789 0 + +create switch filter rules:: + + MAC_IPV4_TCP + flow create 0 priority 0 ingress pattern eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 / tcp src is 25 dst is 23 / end actions queue index 1 / end + + MAC_IPV4_VXLAN_IPV4_UDP_PAY + flow create 0 priority 0 ingress pattern eth / ipv4 / udp / vxlan / eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 / udp src is 25 dst is 23 / end actions queue index 2 / end + +create fdir filter rules:: + + MAC_IPV4_TCP + flow create 0 priority 1 ingress pattern eth / ipv4 src is 192.168.0.4 dst is 192.168.0.7 tos is 4 ttl is 20 / tcp src is 25 dst is 23 / end actions queue index 3 / end + + MAC_IPV4_VXLAN_IPV4_UDP_PAY + flow create 0 priority 1 ingress pattern eth / ipv4 / udp / vxlan / eth / ipv4 src is 192.168.0.4 dst is 192.168.0.7 / udp src is 25 dst is 23 / end actions queue index 4 / end + +check flow list with commands "flow list 0", all flows are created correctly:: + + ID Group Prio Attr Rule + 0 0 0 i-- ETH IPV4 TCP => QUEUE + 1 ... + 2 ... + 3 ... + +send packets according to the created rules, check the packets are recieved in right queues:: + + sendp([Ether(dst="00:00:00:00:11:00",src="11:22:33:44:55:66")/IP(src="192.168.0.2",dst="192.168.0.3",tos=4)/TCP(sport=25,dport=23)/Raw('x'*80)],iface="enp134s0f0") + sendp([Ether(dst="00:00:00:00:11:00",src="11:22:33:44:55:66")/IP()/UDP()/VXLAN()/Ether()/IP(src="192.168.0.2",dst="192.168.0.3",tos=4)/UDP(sport=25,dport=23)/Raw('x'*80)],iface="enp134s0f0") + sendp([Ether(dst="00:00:00:00:11:00",src="11:22:33:44:55:66")/IP(src="192.168.0.4",dst="192.168.0.7",tos=4,ttl=20)/TCP(sport=25,dport=23)/Raw('x'*80)],iface="enp134s0f0") + sendp([Ether(dst="00:00:00:00:11:00",src="11:22:33:44:55:66")/IP()/UDP()/VXLAN()/Ether()/IP(src="192.168.0.4 ",dst="192.168.0.7")/UDP(sport=25,dport=23)/Raw('x'*80)],iface="enp134s0f0") + +create rules without priority:: + + flow create 0 ingress pattern eth / ipv4 src is 192.168.0.3 dst is 192.168.0.1 tos is 5 / tcp src is 25 dst is 23 / end actions queue index 1 / end + flow create 0 ingress pattern eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 ttl is 20 / sctp src is 25 dst is 23 / end actions queue index 1 / end + +check only patterns supported by switch can be created for the default priorty is 0. +So the first flow can be created and the second flow can not be created. + + +Test case: Create no destination high priority flow rule +============================================================= + +a no destination high priority rule is not acceptable. Destination here means exact actions. + +Start the ``testpmd`` application as follows:: + + ./x86_64-native-linuxapp-gcc/app/testpmd -c 0xf -n 4 -w 0000:af:00.0,pipeline-mode-support=1 -- -i --txq=8 --rxq=8 + set fwd rxonly + set verbose 1 + +create a rule:: + + flow create 0 priority 0 ingress pattern eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 / tcp src is 25 dst is 23 / end actions / end + flow create 0 priority 0 ingress pattern eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 / tcp src is 25 dst is 23 / end + +check the flows can not be created. + + +Test case: Create flow rules only supported by fdir filter with priority 0 +=========================================================================== + +create a rule only supported by fdir filter with priority 0, it is not acceptable. + +Start the ``testpmd`` application as follows:: + + ./x86_64-native-linuxapp-gcc/app/testpmd -c 0xf -n 4 -w 0000:af:00.0,pipeline-mode-support=1 -- -i --txq=8 --rxq=8 + set fwd rxonly + set verbose 1 + +create rules:: + + MAC_IPV6_SCTP + flow create 0 priority 0 ingress pattern eth / ipv6 src is 1111:2222:3333:4444:5555:6666:7777:8888 dst is 1111:2222:3333:4444:5555:6666:7777:9999 / sctp src is 25 dst is 23 / end actions queue index 1 / end + + MAC_IPV4_SCTP + flow create 0 priority 0 ingress pattern eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 ttl is 20 / sctp src is 25 dst is 23 / end actions queue index 1 / end + +check the flows can not be created. + + +Test case: Create flow rules only supported by switch filter with priority 1 +============================================================================= + +create a rule only supported by fdir switch with priority 1, it is not acceptable. + +Start the ``testpmd`` application as follows:: + + ./x86_64-native-linuxapp-gcc/app/testpmd -c 0xf -n 4 -w 0000:af:00.0,pipeline-mode-support=1 -- -i --txq=8 --rxq=8 + set fwd rxonly + set verbose 1 + +create rules:: + + MAC_IPV4_NVGRE_IPV4_FRAG + flow create 0 priority 1 ingress pattern eth / ipv4 / nvgre / eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 / end actions queue index 3 / end + + MAC_IPV4_NVGRE_IPV4_UDP_PAY + flow create 0 priority 1 ingress pattern eth / ipv4 / nvgre / eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 / udp src is 25 dst is 23 / end actions queue index 3 / end + +check the flows can not be created. + + +Test case: Create flow rules with same parameter but differenet actions +========================================================================== + +It is acceptable to create same rules with differenet filter in pipeline mode. +When fdir filter and switch filter has the same parameter rules, the flow will map to switch then fdir. + +Start the ``testpmd`` application as follows:: + + ./x86_64-native-linuxapp-gcc/app/testpmd -c 0xf -n 4 -w 0000:af:00.0,pipeline-mode-support=1 -- -i --txq=8 --rxq=8 + set fwd rxonly + set verbose 1 + +create switch rule:: + + flow create 0 priority 0 ingress pattern eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 / tcp src is 25 dst is 23 / end actions queue index 1 / end + +create fdir rule:: + + flow create 0 priority 1 ingress pattern eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 / tcp src is 25 dst is 23 / end actions queue index 3 / end + +check two flows can be created. + +send a pkt to check the packets are recieved in queue 1:: + + sendp([Ether(dst="00:00:00:00:11:00",src="11:22:33:44:55:66")/IP(src="192.168.0.2",dst="192.168.0.3",tos=4)/TCP(sport=25,dport=23)/Raw('x'*80)],iface="enp134s0f0") + +remove the switch rule and send a pkt to check the packets are recieved in queue 3:: + + flow destroy 0 rule 0 + + sendp([Ether(dst="00:00:00:00:11:00",src="11:22:33:44:55:66")/IP(src="192.168.0.2",dst="192.168.0.3",tos=4)/TCP(sport=25,dport=23)/Raw('x'*80)],iface="enp134s0f0") + +restart the ``testpmd`` application as follows:: + + ./x86_64-native-linuxapp-gcc/app/testpmd -c 0xf -n 4 -w 0000:af:00.0, pipeline-mode-support=1 -- -i --txq=8 --rxq=8 + set fwd rxonly + set verbose 1 + +create fdir rule:: + + flow create 0 priority 1 ingress pattern eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 / tcp src is 25 dst is 23 / end actions queue index 3 / end + +create switch rule:: + + flow create 0 priority 0 ingress pattern eth / ipv4 src is 192.168.0.2 dst is 192.168.0.3 tos is 4 / tcp src is 25 dst is 23 / end actions queue index 1 / end + +check two flows can be created. + +send a pkt to check the packets are recieved in queue 1:: + + sendp([Ether(dst="00:00:00:00:11:00",src="11:22:33:44:55:66")/IP(src="192.168.0.2",dst="192.168.0.3",tos=4)/TCP(sport=25,dport=23)/Raw('x'*80)],iface="enp134s0f2") + +remove the switch rule and send a pkt to check the packets are recieved in queue 3:: + + flow destroy 0 rule 1 + + sendp([Ether(dst="00:00:00:00:11:00",src="11:22:33:44:55:66")/IP(src="192.168.0.2",dst="192.168.0.3",tos=4)/TCP(sport=25,dport=23)/Raw('x'*80)],iface="enp134s0f2") + + +Test case: stress all fdir rules +=================================== + +the fdir tabel can contain 16K(16384) flow rules. Each port can use 14K rules, and 2K are shared equally to each port. +that means, 1 port of 2*25G/2*100G can use 14+2/2=15K fdir rules, 1 port of 4*25G can use 14+2/4=14.5K fdir rules. +in non-pipeline mode, we can create switch rules only after fill out fidr tabel. +in pipeline mode, these two tables are independent. + +Start the ``testpmd`` application as follows:: + + ./x86_64-native-linuxapp-gcc/app/testpmd -c 0xf -n 4 -w 0000:af:00.0,pipeline-mode-support=1 -- -i --txq=8 --rxq=8 + set fwd rxonly + set verbose 1 + +create 10K switch rules with priority 0:: + + flow create 0 priority 0 ingress pattern eth / ipv4 src is 192.168.0.0 dst is 192.1.0.0 tos is 4 / tcp src is 25 dst is 23 / end actions queue index 5 / end + ...... + +create 16K fdir rules with priority 1:: + + flow create 0 priority 1 ingress pattern eth / ipv4 src is 192.168.0.0 dst is 192.1.0.0 tos is 4 / tcp src is 25 dst is 23 / end actions queue index 3 / end + ...... + +check the flow rules can be created and sent packets which is match to these rules:: + + sendp([Ether(dst="00:00:00:00:01:00",src="11:22:33:44:55:66")/IP(src="192.168.0.0",dst="192.1.0.0",tos=4)/TCP(sport=25,dport=23)/Raw("x"*80)],iface="enp134s0f0") + ...... + +check the packets can be recieved by the right queues. + +restart the ``testpmd`` application as follows:: + + ./x86_64-native-linuxapp-gcc/app/testpmd -c 0xf -n 4 -w 0000:af:00.0 -- -i --txq=8 --rxq=8 + set fwd rxonly + set verbose 1 + +check the ports number of nic, create 15K/14.5K fdir rules:: + + flow create 0 ingress pattern eth / ipv4 src is 192.168.0.0 dst is 192.1.0.0 tos is 4 / tcp src is 25 dst is 23 / end actions queue index 3 / end + +create 10K switch rules:: + + flow create 0 ingress pattern eth / ipv4 src is 192.168.0.0 dst is 192.1.0.0 tos is 4 / tcp src is 14 dst is 17 / end actions queue index 5 / end + +check the flow rules can be created and sent packets which is match to these rules:: + + sendp([Ether(dst="00:00:00:00:01:00",src="11:22:33:44:55:66")/IP(src="192.168.0.0",dst="192.1.0.0",tos=4)/TCP(sport=25,dport=23)/Raw("x"*80)],iface="enp134s0f0") + ...... + +check the packets can be recieved by the right queues. -- 2.17.1