From: Peng Yuan <yuan.peng@intel.com>
To: dts@dpdk.org
Cc: Peng Yuan <yuan.peng@intel.com>
Subject: [v2 2/2]test_plans/cvl_advanced_rss_checksum: Enable CVL PF/IAVF RSS with checksum as inputset
Date: Wed, 1 Dec 2021 01:37:37 +0000 [thread overview]
Message-ID: <20211201013737.107397-1-yuan.peng@intel.com> (raw)
Add cvl_advanced_rss_checksum_test_plan.rst to test_plans
Signed-off-by: Peng Yuan <yuan.peng@intel.com>
---
| 593 ++++++++++++++++++
1 file changed, 593 insertions(+)
create mode 100644 test_plans/cvl_advanced_rss_checksum_test_plan.rst
--git a/test_plans/cvl_advanced_rss_checksum_test_plan.rst b/test_plans/cvl_advanced_rss_checksum_test_plan.rst
new file mode 100644
index 00000000..6b0a949c
--- /dev/null
+++ b/test_plans/cvl_advanced_rss_checksum_test_plan.rst
@@ -0,0 +1,593 @@
+.. Copyright (c) <2021>, 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: CVL enable RSS with checksum as inputset
+=============================================
+
+Description
+===========
+
+Enable CVL PF/IAVF RSS with checksum as inputset.
+1. distribute ipv4 packet into queues based on the hash value of checksum.
+2. distribute tcp / udp packet into queues based on hash value of l4 checksum.
+
+Pattern and input set
+---------------------
+.. table::
+
+ +-------------------------------+---------------------------+------------------------+
+ | Default hash function: Non Symmetric_toeplitz |
+ +-------------------------------+---------------------------+------------------------+
+ | Packet Type | Pattern | Input Set |
+ +===============================+===========================+========================+
+ | IPv4/IPv6 + TCP/UDP/SCTP | MAC_IPV4 | ipv4-chksum |
+ +-------------------------------+---------------------------+------------------------+
+ | | MAC_IPV4_UDP | ipv4-chksum, l4-chksum |
+ +-------------------------------+---------------------------+------------------------+
+ | | MAC_IPV4_TCP | ipv4-chksum, l4-chksum |
+ +-------------------------------+---------------------------+------------------------+
+ | | MAC_IPV4_SCTP | ipv4-chksum, l4-chksum |
+ +-------------------------------+---------------------------+------------------------+
+ | | MAC_IPV6_UDP | l4-chksum |
+ +-------------------------------+---------------------------+------------------------+
+ | | MAC_IPV6_TCP | l4-chksum |
+ +-------------------------------+---------------------------+------------------------+
+ | | MAC_IPV6_SCTP | l4-chksum |
+ +-------------------------------+---------------------------+------------------------+
+
+
+Prerequisites
+=============
+
+1. Hardware:
+
+ - Intel E810 series ethernet cards: columbiaville_25g/columbiaville_100g/
+
+2. Software:
+
+ - dpdk: http://dpdk.org/git/dpdk
+ - scapy: http://www.secdev.org/projects/scapy/
+
+.. note::
+
+ This RSS feature is designed for CVL NIC 25G and 100g, so below cases only support CVL nic.
+
+3. For PF, bind the CVL port to dpdk driver in DUT::
+
+ modprobe vfio-pci
+ ./usertools/dpdk-devbind.py --force --bind=vfio-pci 0000:3b:00.0
+
+4. For VF::
+
+ echo 1 > /sys/bus/pci/devices/0000\:3b\:00.0/sriov_numvfs
+ ip link set enp59s0f0 vf 0 mac 00:11:22:33:44:55
+ ./usertools/dpdk-devbind.py --force --bind=vfio-pci 0000:3b:01.0
+
+.. note::
+
+ The kernel must be >= 3.6+ and VT-d must be enabled in bios.
+
+4. Launch the testpmd to configuration queue of rx and tx number 16 in DUT::
+
+ ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -c f -n 6 -- -i --rxq=16 --txq=16 --disable-rss --rxd=384 --txd=384
+ testpmd> set fwd rxonly
+ testpmd> set verbose 1
+ testpmd> start
+
+Default parameters
+------------------
+
+ MAC::
+
+ [Dest MAC]: 00:11:22:33:44:55
+ [Src MAC]: 52:00:00:00:00:00
+
+ IPv4::
+
+ [Dest IP]: 192.168.0.2
+ [Source IP]: 192.168.0.1
+
+ IPv6::
+
+ [Source IPv6]: ABAB:910A:2222:5498:8475:1111:3900:1010
+ [Dest IPv6]: CDCD:910A:2222:5498:8475:1111:3900:2020
+
+ UDP/TCP/SCTP::
+
+ [Source Port]: 22
+ [Dest Port]: 23
+
+
+Common Test steps
+=================
+all the test cases run the same test steps as below:
+
+1. validate rule.
+2. create rule and list rule.
+3. send a basic hit pattern packet,record the hash value,
+ check the packet is distributed to queues by RSS.
+4. send hit pattern packet with changed input set in the rule.
+ check the received packet have different hash value with basic packet.
+ check the packet is distributed to queues by rss.
+5. send hit pattern packet with changed input set not in the rule.
+ check the received packet have same hash value with the basic packet.
+ check the packet is distributed to queues by rss.
+ note: if there is not this type packet in the case, omit this step.
+6. send not hit pattern packets with input set in the rule.
+ check the received packets have not hash value, and distributed to queue 0.
+ note: if there is not this type packet in the case, omit this step.
+7. distroy the rule and list rule.
+8. send same packet with step 3.
+ check the received packets have not hash value, and distributed to queue 0.
+
+Test case 1: PF_MAC_IPV4_IPV4_CHKSUM
+====================================
+1. create rss rule::
+
+ flow create 0 ingress pattern eth / ipv4 / end actions rss types ipv4-chksum end queues end / end
+
+2. basic hit pattern packet::
+
+ p1=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1",dst="192.168.0.2",chksum=0x1)/ ("X"*48)
+
+3. hit pattern/changed defined input set::
+
+ p2=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1",dst="192.168.0.2",chksum=0xffff)/ ("X"*48)
+
+4. hit pattern/changed not defined input set::
+
+ p3=Ether(dst="00:11:22:33:44:53", src="52:00:00:00:00:01")/IP(src="192.168.1.1",dst="192.168.1.2",chksum=0x1)/ ("X"*48)
+
+5. not hit pattern::
+
+ p4=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2020", src="ABAB:910A:2222:5498:8475:1111:3900:1010")/("X"*48)
+
+Test case 2: PF_MAC_IPV4_UDP
+============================
+Subcase 1: PF_MAC_IPV4_UDP_L4_CHKSUM
+------------------------------------
+1. create rss rule::
+
+ flow create 0 ingress pattern eth / ipv4 / udp / end actions rss types l4-chksum end queues end / end
+
+2. basic hit pattern packet::
+
+ p1=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1",dst="192.168.0.2")/UDP(sport=22, dport=23,chksum=0xffff)/("X"*48)
+
+3. hit pattern/changed defined input set::
+
+ p2=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1",dst="192.168.0.2")/UDP(sport=22, dport=23,chksum=0xfffe)/("X"*48)
+
+4. hit pattern/changed not defined input set::
+
+ p3=Ether(dst="00:11:22:33:44:53", src="52:00:00:00:00:01")/IP(src="192.168.1.1", dst="192.168.1.2",chksum=0x3)/UDP(sport=32, dport=33,chksum=0xffff)/("X"*48)
+
+5. not hit pattern::
+
+ p4=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1",dst="192.168.0.2")/SCTP(sport=22, dport=23,chksum=0xffff)/("X"*48)
+
+Subcase 2: PF_MAC_IPV4_UDP_IPV4_CHKSUM
+--------------------------------------
+1. create rss rule::
+
+ flow create 0 ingress pattern eth / ipv4 / udp / end actions rss types ipv4-chksum end queues end / end
+
+2. basic hit pattern packet::
+
+ p1=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1",dst="192.168.0.2",chksum=0xffff)/UDP(sport=22, dport=23)/("X"*48)
+
+3. hit pattern/changed defined input set::
+
+ p2=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1",dst="192.168.0.2",chksum=0xfffe)/UDP(sport=22, dport=23)/("X"*48)
+
+4. hit pattern/changed not defined input set::
+
+ p3=Ether(dst="00:11:22:33:44:53", src="52:00:00:00:00:01")/IP(src="192.168.1.1", dst="192.168.1.2",chksum=0xffff)/UDP(sport=32, dport=33,chksum=0xffff)/("X"*48)
+
+5. not hit pattern::
+
+ p4=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1",dst="192.168.0.2",chksum=0xffff)/SCTP(sport=22, dport=23)/("X"*48)
+
+Test case 3: PF_MAC_IPV4_TCP
+============================
+The rules and packets in this test case is similar to "Test case: PF_MAC_IPV4_UDP",
+just change some parts of rules and packets:
+
+ rule:
+ change udp to tcp
+ packets:
+ if the packet's L4 layer is UDP, change it to TCP;
+ if the packet's L4 layer is TCP, change it to UDP;
+
+Subcase 1: PF_MAC_IPV4_TCP_L4_CHKSUM
+------------------------------------
+
+Subcase 2: PF_MAC_IPV4_TCP_IPV4_CHKSUM
+--------------------------------------
+
+Test case 4: PF_MAC_IPV4_SCTP
+=============================
+The rules and packets in this test case is similar to "Test case: PF_MAC_IPV4_UDP",
+just change some parts of rules and packets:
+
+ rule:
+ change udp to sctp
+ packets:
+ if the packet's L4 layer is UDP, change it to SCTP;
+ if the packet's L4 layer is SCTP, change it to TCP;
+
+Subcase 1: PF_MAC_IPV4_SCTP_L4_CHKSUM
+-------------------------------------
+
+Subcase 2: PF_MAC_IPV4_SCTP_IPV4_CHKSUM
+---------------------------------------
+
+
+Test case 5: PF_MAC_IPV6_UDP_L4_CHKSUM
+======================================
+1. create rss rule::
+
+ flow create 0 ingress pattern eth / ipv6 / udp / end actions rss types l4-chksum end queues end / end
+
+2. basic hit pattern packet::
+
+ p1 = Ether(src="52:00:00:00:00:00", dst="00:11:22:33:44:55")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2020", src="ABAB:910A:2222:5498:8475:1111:3900:1010")/UDP(sport=22, dport=23, chksum=0x1)/("X"*48)
+
+3. hit pattern/changed defined input set::
+
+ p2 = Ether(src="52:00:00:00:00:00", dst="00:11:22:33:44:55")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2020", src="ABAB:910A:2222:5498:8475:1111:3900:1010")/UDP(sport=22, dport=23, chksum=0x2)/("X"*48)
+
+4. hit pattern/changed not defined input set::
+
+ p3 = Ether(src="52:00:00:00:00:01", dst="00:11:22:33:44:53")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2021", src="ABAB:910A:2222:5498:8475:1111:3900:1011")/UDP(sport=32, dport=33, chksum=0x1)/("X"*48)
+
+5. not hit pattern::
+
+ p4 = Ether(src="52:00:00:00:00:00", dst="00:11:22:33:44:55")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2020", src="ABAB:910A:2222:5498:8475:1111:3900:1010")/TCP(sport=22, dport=23, chksum=0x1)/("X"*49)
+
+Test case 6: PF_MAC_IPV6_TCP_L4_CHKSUM
+======================================
+1. create rss rule::
+
+ flow create 0 ingress pattern eth / ipv6 / tcp / end actions rss types l4-chksum end queues end / end
+
+2. basic hit pattern packet::
+
+ p1 = Ether(src="52:00:00:00:00:00", dst="00:11:22:33:44:55")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2020", src="ABAB:910A:2222:5498:8475:1111:3900:1010")/TCP(sport=22, dport=23, chksum=0x1)/("X"*48)
+
+3. hit pattern/changed defined input set::
+
+ p2 = Ether(src="52:00:00:00:00:00", dst="00:11:22:33:44:55")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2020", src="ABAB:910A:2222:5498:8475:1111:3900:1010")/TCP(sport=22, dport=23, chksum=0x2)/("X"*48)
+
+4. hit pattern/changed not defined input set::
+
+ p3 = Ether(src="52:00:00:00:00:01", dst="00:11:22:33:44:53")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2021", src="ABAB:910A:2222:5498:8475:1111:3900:1011")/TCP(sport=32, dport=33, chksum=0x1)/("X"*48)
+
+5. not hit pattern::
+
+ p4 = Ether(src="52:00:00:00:00:00", dst="00:11:22:33:44:55")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2020", src="ABAB:910A:2222:5498:8475:1111:3900:1010")/SCTP(sport=22, dport=23, chksum=0x1)/("X"*49)
+
+Test case 7: PF_MAC_IPV6_SCTP_L4_CHKSUM
+=======================================
+1. create rss rule::
+
+ flow create 0 ingress pattern eth / ipv6 / sctp / end actions rss types l4-chksum end queues end / end
+
+2. basic hit pattern packet::
+
+ p1 = Ether(src="52:00:00:00:00:00", dst="00:11:22:33:44:55")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2020", src="ABAB:910A:2222:5498:8475:1111:3900:1010")/SCTP(sport=22, dport=23, chksum=0xffffffff)/("X"*48)
+
+3. hit pattern/changed defined input set::
+
+ p2 = Ether(src="52:00:00:00:00:00", dst="00:11:22:33:44:55")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2020", src="ABAB:910A:2222:5498:8475:1111:3900:1010")/SCTP(sport=22, dport=23, chksum=0xfffffffe)/("X"*48)
+
+4. hit pattern/changed not defined input set::
+
+ p3 = Ether(src="52:00:00:00:00:01", dst="00:11:22:33:44:53")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2021", src="ABAB:910A:2222:5498:8475:1111:3900:1011")/SCTP(sport=32, dport=33, chksum=0xffffffff)/("X"*48)
+
+5. not hit pattern::
+
+ p4 = Ether(src="52:00:00:00:00:00", dst="00:11:22:33:44:55")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2020", src="ABAB:910A:2222:5498:8475:1111:3900:1010")/UDP(sport=22, dport=23, chksum=0x1)/("X"*49)
+
+Test case 8: PF Checksum for different payload length
+=====================================================
+1. launch testpmd without "--disable-rss"::
+
+ ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -c f -n 6 -- -i --rxq=16 --txq=16
+ testpmd> set fwd rxonly
+ testpmd> set verbose 1
+ testpmd> start
+
+2. Send packets with different payload length::
+
+ p1=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP()/("X"*48)
+ p2=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP()/("X"*64)
+ p3=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP()/UDP()/("X"*48)
+ p4=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP()/UDP()/("X"*64)
+ p5=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP()/TCP()/("X"*48)
+ p6=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP()/TCP()/("X"*64)
+ p7=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP()/SCTP()/("X"*48)
+ p8=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP()/SCTP()/("X"*64)
+
+ Check all the packets received have same hash value.
+
+3. create RSS rule of 5-tuple inputset for each packet type::
+
+ flow create 0 ingress pattern eth / ipv4 / udp / end actions rss types ipv4-udp end queues end / end
+ flow create 0 ingress pattern eth / ipv4 / tcp / end actions rss types ipv4-tcp end queues end / end
+ flow create 0 ingress pattern eth / ipv4 / sctp / end actions rss types ipv4-sctp end queues end / end
+
+4. Send packets of step 2.
+ Check the received packets with same packet type have same hash value.
+
+5. create RSS rule of l4-chksum inputset for each packet type::
+
+ flow create 0 ingress pattern eth / ipv4 / udp / end actions rss types l4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv4 / tcp / end actions rss types l4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv4 / sctp / end actions rss types l4-chksum end queues end / end
+
+6. Send packets of step 2.
+ Check the UDP/TCP/SCTP packets with different payload length have different hash value.
+
+7. create RSS rule of ipv4-chksum inputset for each packet type::
+
+ flow create 0 ingress pattern eth / ipv4 / end actions rss types ipv4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv4 / udp / end actions rss types ipv4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv4 / tcp / end actions rss types ipv4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv4 / sctp / end actions rss types ipv4-chksum end queues end / end
+
+8. Send packets of step 2.
+ Check the IPV4/UDP/TCP/SCTP packets with different payload length have different hash value.
+
+Test case 9: PF Set HW csum, flow rule doesn’t impact RX checksum and TX checksum
+=================================================================================
+1. launch testpmd without "--disable-rss"::
+
+ ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -c f -n 6 -- -i --rxq=16 --txq=16
+
+2. Set tx checksum::
+
+ port stop all
+ set fwd csum
+ csum set ip hw 0
+ csum set udp hw 0
+ csum set tcp hw 0
+ csum set sctp hw 0
+ port start all
+ set verbose 1
+ start
+
+3. Capture the tx packet at tester port::
+
+ tcpdump -i enp216s0f0 -Q in -e -n -v -x
+
+4. Send packets::
+
+ p1=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1", chksum=0xfff3)/("X"*48)
+ p2=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1")/TCP(sport=22, chksum=0xfff3)/("X"*48)
+ p3=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1")/UDP(sport=22, chksum=0x1)/("X"*48)
+ p4=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1")/SCTP(sport=22, chksum=0x1)/("X"*48)
+ p5=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IPv6()/TCP(sport=22, chksum=0xe38)/("X"*48)
+ p6=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IPv6()/UDP(sport=22, chksum=0xe38)/("X"*48)
+ p7=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IPv6()/SCTP(sport=22, chksum=0xf)/("X"*48)
+
+ Check rx checksum good or bad, check if the tx checksum correct.
+
+5. Create rss rules with chsum as inputset::
+
+ flow create 0 ingress pattern eth / ipv4 / end actions rss types ipv4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv4 / tcp / end actions rss types l4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv4 / udp / end actions rss types l4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv4 / sctp / end actions rss types l4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv6 / tcp / end actions rss types l4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv6 / udp / end actions rss types l4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv6 / sctp / end actions rss types l4-chksum end queues end / end
+
+6. Send the same packets, check the hash value changed, check rx and tx checksum, get the same result.
+
+Test case 10: PF Combined case with fdir queue group
+====================================================
+1. start testpmd without "--disable-rss"::
+
+ ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -c f -n 6 -- -i --rxq=16 --txq=16
+ testpmd> set fwd rxonly
+ testpmd> set verbose 1
+ testpmd> start
+
+2. Create fdir rules to queue group::
+
+ flow create 0 ingress pattern eth dst is 00:11:22:33:44:55 / ipv4 / tcp / end actions rss queues 4 5 end / end
+ flow create 0 ingress pattern eth dst is 00:11:22:33:44:55 / ipv4 / udp / end actions rss queues 6 7 end / end
+ flow create 0 ingress pattern eth dst is 00:11:22:33:44:55 / ipv4 / sctp / end actions rss queues 8 9 end / end
+ flow create 0 ingress pattern eth dst is 00:11:22:33:44:55 / ipv6 / tcp / end actions rss queues 10 11 end / end
+ flow create 0 ingress pattern eth dst is 00:11:22:33:44:55 / ipv6 / udp / end actions rss queues 12 13 end / end
+ flow create 0 ingress pattern eth dst is 00:11:22:33:44:55 / ipv6 / sctp / end actions rss queues 14 15 end / end
+
+3. Send packets::
+
+ p1=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1", chksum=0xfff3)/("X"*48)
+ p2=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1")/TCP(sport=22, chksum=0xfff3)/("X"*48)
+ p3=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1")/UDP(sport=22, chksum=0x1)/("X"*48)
+ p4=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IP(src="192.168.0.1")/SCTP(sport=22, chksum=0x1)/("X"*48)
+ p5=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IPv6()/TCP(sport=22, chksum=0xe38)/("X"*48)
+ p6=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IPv6()/UDP(sport=22, chksum=0xe38)/("X"*48)
+ p7=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:00")/IPv6(src="ABAB:910A:2222:5498:8475:1111:3900:1010")/SCTP(sport=22, chksum=0xf)/("X"*48)
+
+ Check p2-p7 are distributed to specified queue group,
+ p1 is distributed by RSS hash value.
+
+4. Create rss rule with inputset checksum::
+
+ flow create 0 ingress pattern eth / ipv4 / end actions rss types ipv4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv4 / tcp / end actions rss types l4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv4 / udp / end actions rss types l4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv4 / sctp / end actions rss types l4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv6 / tcp / end actions rss types l4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv6 / udp / end actions rss types l4-chksum end queues end / end
+ flow create 0 ingress pattern eth / ipv6 / sctp / end actions rss types l4-chksum end queues end / end
+
+ Send same packets again. Check hash values of p2-p6 are changed, but queue group not changed.
+ Check p1 hash value changed too, and distributed by RSS hash value.
+
+5. Create fdir rule to queue group for ipv4 pattern::
+
+ flow create 0 ingress pattern eth dst is 00:11:22:33:44:55 / ipv4 / end actions rss queues 0 1 2 3 end / end
+
+ Send p1, check the packet distributed to queue group without hash value changed.
+
+Test case 11: PF negative case
+==============================
+1. create a rule with invalid inputset::
+
+ flow create 0 ingress pattern eth / ipv4 / end actions rss types l4-chksum end queues end / end
+ ice_flow_create(): Failed to create flow
+ port_flow_complain(): Caught PMD error type 10 (item specification): cause: 0x7fff0a9b9f80, Invalid input set: Invalid argument
+
+2. create a rule with inputset not supported::
+
+ flow create 0 ingress pattern eth / ipv6 / end actions rss types ipv6-chksum end queues end / end
+ Bad arguments
+
+Test case 12: VF_MAC_IPV4_IPV4_CHKSUM
+=====================================
+Step1-3 and step5 are the same as Test case 1.
+Only the packet in step4 is replaced by::
+
+ p3=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:01")/IP(src="192.168.1.1",dst="192.168.1.2",chksum=0x1)/ ("X"*48)
+
+Test case 13: VF_MAC_IPV4_UDP
+=============================
+Subcase 1: VF_MAC_IPV4_UDP_L4_CHKSUM
+------------------------------------
+Step1-3 and step5 are the same as Test case 2, Subcase 1.
+Only the packet in step4 is replaced by::
+
+ p3=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:01")/IP(src="192.168.1.1", dst="192.168.1.2",chksum=0x3)/UDP(sport=32, dport=33,chksum=0xffff)/("X"*48)
+
+Subcase 2: VF_MAC_IPV4_UDP_IPV4_CHKSUM
+--------------------------------------
+Step1-3 and step5 are the same as Test case 2, Subcase 2.
+Only the packet in step4 is replaced by::
+
+ p3=Ether(dst="00:11:22:33:44:55", src="52:00:00:00:00:01")/IP(src="192.168.1.1", dst="192.168.1.2",chksum=0xffff)/UDP(sport=32, dport=33,chksum=0xffff)/("X"*48)
+
+Test case 14: VF_MAC_IPV4_TCP
+=============================
+The rules and packets in this test case is similar to "Test case: VF_MAC_IPV4_UDP"
+just change some parts of rules and packets:
+
+ rule:
+ change udp to tcp
+ packets:
+ if the packet's L4 layer is UDP, change it to TCP;
+ if the packet's L4 layer is TCP, change it to UDP;
+
+Subcase 1: VF_MAC_IPV4_TCP_L4_CHKSUM
+------------------------------------
+
+Subcase 2: VF_MAC_IPV4_TCP_IPV4_CHKSUM
+--------------------------------------
+
+Test case 15: VF_MAC_IPV4_SCTP
+==============================
+The rules and packets in this test case is similar to "Test case: VF_MAC_IPV4_UDP"
+just change some parts of rules and packets:
+
+ rule:
+ change udp to sctp
+ packets:
+ if the packet's L4 layer is UDP, change it to SCTP;
+ if the packet's L4 layer is SCTP, change it to TCP;
+
+Subcase 1: VF_MAC_IPV4_SCTP_L4_CHKSUM
+-------------------------------------
+
+Subcase 2: VF_MAC_IPV4_SCTP_IPV4_CHKSUM
+---------------------------------------
+
+Test case 16: VF_MAC_IPV6_UDP_L4_CHKSUM
+=======================================
+Step1-3 and step5 are the same as Test case 5.
+Only the packet in step4 is replaced by::
+
+ p3 = Ether(src="52:00:00:00:00:01", dst="00:11:22:33:44:55")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2021", src="ABAB:910A:2222:5498:8475:1111:3900:1011")/UDP(sport=32, dport=33, chksum=0x1)/("X"*48)
+
+Test case 17: VF_MAC_IPV6_TCP_L4_CHKSUM
+=======================================
+Step1-3 and step5 are the same as Test case 6.
+Only the packet in step4 is replaced by::
+
+ p3 = Ether(src="52:00:00:00:00:01", dst="00:11:22:33:44:55")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2021", src="ABAB:910A:2222:5498:8475:1111:3900:1011")/TCP(sport=32, dport=33, chksum=0x1)/("X"*48)
+
+Test case 18: VF_MAC_IPV6_SCTP_L4_CHKSUM
+========================================
+Step1-3 and step5 are the same as Test case 7.
+Only the packet in step4 is replaced by::
+
+ p3 = Ether(src="52:00:00:00:00:01", dst="00:11:22:33:44:55")/IPv6(dst="CDCD:910A:2222:5498:8475:1111:3900:2021", src="ABAB:910A:2222:5498:8475:1111:3900:1011")/SCTP(sport=32, dport=33, chksum=0xffffffff)/("X"*48)
+
+Test case 19: VF Checksum for different payload length
+======================================================
+The case steps are the same as Test case 8.
+
+Test case 20: VF Set HW csum, flow rule doesn’t impact RX checksum and TX checksum
+==================================================================================
+The case steps are the same as Test case 9.
+
+Test case 21: VF Combined case with fdir queue group
+====================================================
+Step1 is the same as Test case 10.
+
+2. Create fdir rules to queue group::
+
+ flow create 0 ingress pattern eth / ipv4 / tcp / end actions rss queues 4 5 end / end
+ flow create 0 ingress pattern eth / ipv4 / udp / end actions rss queues 6 7 end / end
+ flow create 0 ingress pattern eth / ipv4 src is 192.168.0.1 / sctp / end actions rss queues 8 9 end / end
+ flow create 0 ingress pattern eth / ipv6 / tcp / end actions rss queues 10 11 end / end
+ flow create 0 ingress pattern eth / ipv6 / udp / end actions rss queues 12 13 end / end
+ flow create 0 ingress pattern eth / ipv6 src is ABAB:910A:2222:5498:8475:1111:3900:1010 / sctp / end actions rss queues 14 15 end / end
+
+Step3 and step4 are the same as Test case 10.
+
+5. Create fdir rule to queue group for ipv4 pattern::
+
+ flow destroy 0 rule 0
+ flow destroy 0 rule 1
+ flow destroy 0 rule 2
+ flow create 0 ingress pattern eth / ipv4 src is 192.168.0.1 / end actions rss queues 0 1 2 3 end / end
+
+ Send p1, check the packet distributed to queue group without hash value changed.
+
+Test case 22: VF negative case
+==============================
+The case steps are the same as Test case 11.
--
2.17.1
reply other threads:[~2021-12-01 8:39 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20211201013737.107397-1-yuan.peng@intel.com \
--to=yuan.peng@intel.com \
--cc=dts@dpdk.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).