test suite reviews and discussions
 help / color / mirror / Atom feed
* [dts] [PATCH v1] test_plans: add test cases for af_xdp
@ 2019-12-19 17:39 Yinan
  2019-12-20  8:12 ` Tu, Lijuan
  0 siblings, 1 reply; 2+ messages in thread
From: Yinan @ 2019-12-19 17:39 UTC (permalink / raw)
  To: dts; +Cc: Wang Yinan

From: Wang Yinan <yinan.wang@intel.com>

Signed-off-by: Wang Yinan <yinan.wang@intel.com>
---
 test_plans/af_xdp_2_test_plan.rst | 217 ++++++++++++++++++++++++++++++
 1 file changed, 217 insertions(+)
 create mode 100644 test_plans/af_xdp_2_test_plan.rst

diff --git a/test_plans/af_xdp_2_test_plan.rst b/test_plans/af_xdp_2_test_plan.rst
new file mode 100644
index 0000000..01fae74
--- /dev/null
+++ b/test_plans/af_xdp_2_test_plan.rst
@@ -0,0 +1,217 @@
+.. Copyright (c) <2019>, 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.
+
+=========================
+DPDK PMD for AF_XDP Tests
+=========================
+
+Description
+===========
+
+AF_XDP is a proposed faster version of AF_PACKET interface in Linux.
+This test plan is to analysis the performance of DPDK PMD for AF_XDP.
+
+Prerequisites
+=============
+
+1. Hardware::
+
+    I40e 40G*1
+    enp26s0f1 <---> IXIA_port_0
+
+2. The NIC is located on the socket 1, so we define the cores of socket 1.
+
+3. Clone kernel branch master v5.4, make sure you turn on XDP socket/BPF/I40E before compiling kernel::
+
+    make menuconfig
+    Networking support -->
+         Networking options -->
+                 [ * ] XDP sockets
+
+4. Build kernel and replace your host kernel with it::
+
+    cd bpf-next
+    sh -c 'yes "" | make oldconfig'
+    make -j64
+    make modules_install install
+    make install
+    make headers_install
+    cd tools/lib/bpf && make clean && make install && make install_headers && cd -
+    make headers_install ARCH=x86_64 INSTALL_HDR_PATH=/usr
+    grub-mkconfig -o /boot/grub/grub.cfg
+    reboot
+
+5. Explicitly enable AF_XDP pmd by adding below line to config/common_linux, then build DPDK::
+
+    CONFIG_RTE_LIBRTE_PMD_AF_XDP=y
+    make -j 110 install T=x86_64-native-linuxapp-gcc
+
+6. Involve lib::
+
+    export LD_LIBRARY_PATH=/home/linux/tools/lib/bpf:$LD_LIBRARY_PATH    
+
+Test case 1: single port test with PMD core and IRQ core are pinned to separate cores
+=====================================================================================
+
+1. Start the testpmd::
+
+    ethtool -L enp26s0f1 combined 1
+    ./x86_64-native-linuxapp-gcc/app/testpmd -l 1-2 -n 4 --vdev net_af_xdp0,iface=enp26s0f1,start_queue=0,queue_count=1 --log-level=pmd.net.af_xdp:8  -- -i --nb-cores=1 --rxq=1 --txq=1 --port-topology=loop
+
+2. Assign the kernel core::
+
+    ./set_irq_affinity 3 enp26s0f1         #PMD and IRQs pinned to seperate cores
+    ./set_irq_affinity 2 enp26s0f1         #PMD and IRQs pinned to same cores
+
+3. Send packets by packet generator with different packet size, from 64 bytes to 1518 bytes, check the throughput.
+
+Test case 2: two ports test with PMD cores and IRQ cores are pinned to separate cores
+=====================================================================================
+
+1. Start the testpmd::
+
+    ethtool -L enp26s0f0 combined 1
+    ethtool -L enp26s0f1 combined 1
+    ./x86_64-native-linuxapp-gcc/app/testpmd -l 1-3 --no-pci -n 4 \
+    --vdev net_af_xdp0,iface=enp26s0f0 --vdev net_af_xdp1,iface=enp26s0f1 \
+    --log-level=pmd.net.af_xdp:8 -- -i --auto-start --nb-cores=2 --rxq=1 --txq=1 --port-topology=loop
+
+2. Assign the kernel cores::
+
+    ./set_irq_affinity 4 enp26s0f0
+    ./set_irq_affinity 5 enp26s0f1
+
+3. Send packets by packet generator to port0 and port1 with different packet size, from 64 bytes to 1518 bytes, check the throughput at port0 and port1.
+
+Test case 3: multi-queue test with PMD cores and IRQ cores are pinned to separate cores
+=======================================================================================
+
+1. Set hardware queues::
+
+      ethtool -L enp26s0f1 combined 2
+
+2. Start the testpmd with two queues::
+
+      ./x86_64-native-linuxapp-gcc/app/testpmd -l 1-3 -n 6 --no-pci \
+      --vdev net_af_xdp0,iface=enp26s0f1,start_queue=0,queue_count=2 \
+      -- -i --auto-start --nb-cores=2 --rxq=2 --txq=2 --port-topology=loop
+
+3. Assign the kernel cores::
+
+      ./set_irq_affinity 4-5 enp26s0f1
+
+4. Send packets with different dst IP address by packet generator with different packet size from 64 bytes to 1518 bytes, check the throughput and ensure the packets were distributed to the two queues.
+
+Test case 4: two ports test with PMD cores and IRQ cores pinned to same cores
+=============================================================================
+
+1. Start the testpmd::
+
+    ethtool -L enp26s0f0 combined 1
+    ethtool -L enp26s0f1 combined 1
+    ./x86_64-native-linuxapp-gcc/app/testpmd -l 29,30-31 --no-pci -n 4 \
+    --vdev net_af_xdp0,iface=enp26s0f0 --vdev net_af_xdp1,iface=enp26s0f1 \
+    -- -i --auto-start --nb-cores=2 --rxq=1 --txq=1 --port-topology=loop
+
+2. Assign the kernel cores::
+
+    ./set_irq_affinity 30 enp26s0f0
+    ./set_irq_affinity 31 enp26s0f1
+
+3. Send packets by packet generator to port0 and port1 with different packet size, from 64 bytes to 1518 bytes, check the throughput at port0 and port1.
+
+Test case 5: multi-queue test with PMD cores and IRQ cores pinned to same cores
+===============================================================================
+
+1. Set hardware queues::
+
+      ethtool -L enp26s0f1 combined 2
+
+2. Start the testpmd with two queues::
+
+      ./testpmd -l 29,30-31 -n 6 --no-pci \
+      --vdev net_af_xdp0,iface=enp26s0f1,start_queue=0,queue_count=2 \
+      -- -i --auto-start --nb-cores=2 --rxq=2 --txq=2 --port-topology=loop
+
+3. Assign the kernel cores::
+
+      ./set_irq_affinity 30-31 enp26s0f1
+
+4. Send packets with different dst IP address by packet generator with different packet size from 64 bytes to 1518 bytes, check the throughput and ensure packets were distributed to the two queues.
+
+Test case 6: one port with two vdev and single queue test
+=========================================================
+
+1. Set hardware queues::
+
+      ethtool -L enp26s0f1 combined 2
+
+2. Start the testpmd::
+
+    ./x86_64-native-linuxapp-gcc/app/testpmd -l 1-3 --no-pci -n 4 \
+    --vdev net_af_xdp0,iface=enp26s0f1,start_queue=0,queue_count=1 \
+    --vdev net_af_xdp1,iface=enp26s0f1,start_queue=1,queue_count=1 \
+    -- -i --nb-cores=2 --rxq=1 --txq=1 --port-topology=loop
+
+3. Assign the kernel core::
+
+    ./set_irq_affinity 4-5 enp26s0f1    #PMD and IRQs pinned to seperate cores
+    ./set_irq_affinity 2-3 enp26s0f1    #PMD and IRQs pinned to same cores
+
+4. Set flow director rules in kernel, mapping queue0 and queue1 of the port::
+
+    ethtool -N enp26s0f1 rx-flow-hash udp4 fn
+    ethtool -N enp26s0f1 flow-type udp4 src-port 4242 dst-port 4242 action 1
+    ethtool -N enp26s0f1 flow-type udp4 src-port 4243 dst-port 4243 action 0
+
+5. Send packets match the rules to port, check the throughput at queue0 and queue1.
+
+Test case 7: one port with two vdev and multi-queues test
+=========================================================
+
+1. Set hardware queues::
+
+      ethtool -L enp26s0f1 combined 8
+
+2. Start the testpmd::
+
+    ./x86_64-native-linuxapp-gcc/app/testpmd -l 1-9 --no-pci -n 6 \
+    --vdev net_af_xdp0,iface=enp26s0f1,start_queue=0,queue_count=4 \
+    --vdev net_af_xdp1,iface=enp26s0f1,start_queue=4,queue_count=4 --log-level=pmd.net.af_xdp:8 \
+    -- -i --rss-ip --nb-cores=8 --rxq=4 --txq=4 --port-topology=loop
+
+3. Assign the kernel core::
+
+    ./set_irq_affinity 10-17 enp26s0f1    #PMD and IRQs pinned to seperate cores
+    ./set_irq_affinity 2-9 enp26s0f1    #PMD and IRQs pinned to same cores
+
+4. Send random ip packets , check the packets were distributed to queue0 ~ queue7.
\ No newline at end of file
-- 
2.17.1


^ permalink raw reply	[flat|nested] 2+ messages in thread
end of thread, other threads:[~2019-12-20  8:13 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-12-19 17:39 [dts] [PATCH v1] test_plans: add test cases for af_xdp Yinan
2019-12-20  8:12 ` Tu, Lijuan

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).