test suite reviews and discussions
 help / color / mirror / Atom feed
From: Yu Jiang <yux.jiang@intel.com>
To: dts@dpdk.org
Cc: Yu Jiang <yux.jiang@intel.com>
Subject: [dts] [PATCH V1] test_plans/loadbalancer: remove example
Date: Sat, 18 Sep 2021 10:08:01 +0800	[thread overview]
Message-ID: <1631930882-12140-1-git-send-email-yux.jiang@intel.com> (raw)

According to DPDK commit id ab6ebd7020("examples/load_balancer: remove example"),
remove this plan and suite.

Signed-off-by: Yu Jiang <yux.jiang@intel.com>
---
 test_plans/index.rst                  |   1 -
 test_plans/loadbalancer_test_plan.rst | 194 ----------------------------------
 tests/TestSuite_loadbalancer.py       | 154 ---------------------------
 3 files changed, 349 deletions(-)
 delete mode 100644 test_plans/loadbalancer_test_plan.rst
 delete mode 100644 tests/TestSuite_loadbalancer.py

diff --git a/test_plans/index.rst b/test_plans/index.rst
index b5aaec5..d9dbc2a 100644
--- a/test_plans/index.rst
+++ b/test_plans/index.rst
@@ -179,7 +179,6 @@ The following are the test plans for the DPDK DTS automated test system.
     vxlan_test_plan
     af_xdp_test_plan
     l2fwd_jobstats_test_plan
-    loadbalancer_test_plan
     loopback_multi_queues_test_plan
     telemetry_test_plan
     compressdev_isal_pmd_test_plan
diff --git a/test_plans/loadbalancer_test_plan.rst b/test_plans/loadbalancer_test_plan.rst
deleted file mode 100644
index b5cc769..0000000
--- a/test_plans/loadbalancer_test_plan.rst
+++ /dev/null
@@ -1,194 +0,0 @@
-.. Copyright (c) <2011>, 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.
-
-=============
-Load Balancer
-=============
-
-This test case uses the Load Balancer sample application to benchmark the
-concept of isolating the packet I/O task from the application specific workload.
-A number of lcores are dedicated to handle the interaction with the NIC ports
-(I/O lcores), while the rest of the lcores are dedicated to performing the
-application processing (worker lcores). The worker lcores are totally oblivious
-to the intricacies of the packet I/O activity and use the NIC-agnostic interface
-provided by SW rings to exchange packets with the I/O cores.
-
-Prerequisites
-=============
-
-1. Hardware requirements
-
--  For each CPU socket, each memory channel should be populated with at least
-   1x DIMM.
--  If the PCIe controller is on the CPU, then each CPU socket is populated with
-   the same number of NICs (2x or 4x NICs per CPU socket);
--  Special PCIe restrictions may be required for performance. For example, the
-   following requirements should be met for 10GbE NICs:
-
-    -  NICs are plugged into PCIe Gen2 or Gen3 slots;
-    -  For PCIe Gen2 slots, the number of lanes should be 8x or higher;
-    -  A single port from each NIC card should be used, so for 4x ports, 4x NICs
-       should be connected to the traffic generator.
-
-2. BIOS requirements
-
--  Hyper-Threading Technology is ENABLED
--  Hardware Prefetcher is DISABLED
--  Adjacent Cache Line Prefetch is DISABLED
--  Direct Cache Access is DISABLED
-
-3. Linux kernel requirements
-
--  Linux kernel has the following features enabled: huge page support, UIO, HPET
--  Appropriate number of huge pages are reserved at kernel boot time
--  The IDs of the hardware threads (logical cores) per each CPU socket can be
-   determined by parsing the file /proc/cpuinfo. The naming convention for the
-   logical cores is: C{x.y.z} = hyper-thread z of physical core y of CPU socket
-   x, with typical values of x = 0 .. 3, y = 0 .. 7, z = 0 .. 1. Logical cores
-   C{0.0.1} and C{0.0.1} should be avoided while executing the test, as they
-   are used by the Linux kernel for running regular processes.
-
-4. Software application configuration
-
-The application configuration is done through the command line:
-
--  -c COREMASK: Reunion of all the cores specified by the --rx, --tx and --w
-   parameters.
--  --rx and --tx: These parameters are used to provide the list of lcores used
-   for packet I/O, plus the list of the RX ports & queues, as well as the list
-   of TX ports, that are handled by each of the packet I/O lcores;
--  The RX and TX of the NICs that are physically attached (through PCIe) to a
-   specific CPU socket should always be handled by lcores from the same socket;
--  The RX and TX of the same port can be handled by different lcores, depending
-   on the usecase, therefore the set of RX lcores can be different than the set
-   of TX lcores;
--  Typical configurations enabled for the I/O cores for each CPU socket (as long
-   as the conditions below are met, the actual lcore IDs are irrelevant):
-
-    -  Single lcore handling the RX and TX for all the NICs connected to its CPU
-       socket. Its sibling hyper-thread should not be used by the application;
-
--  One lcore handling the RX and TX for a single NIC port (with its sibling
-   hyper-thread not used by the application). For each CPU socket, there are N
-   physical cores used for packet I/O for N NIC ports;
--  One lcore handling the RX for all the NIC ports connected to its CPU socket
-   and another lcore handling the TX for the same NIC ports (with the sibling
-   hyper-threads not used by the application). For each CPU socket, there are 2
-   physical cores used for packet I/O for N NIC ports;
--  --w: This parameter specifies the list of worker lcores;
-
-    -  A worker lcore cannot be a packet I/O lcore;
-    -  Typical configurations enabled for each CPU socket: 1 / 2 / 4 / 8. Each
-       worker should be allocated on a different physical core. For 8 workers
-       (per CPU socket), if not enough physical cores, both hyper-threads of 4
-       physical cores can be used. As long as these conditions are met, the
-       actual lcore IDs are irrelevant.
-
--  --lpm: IPv4 routing table;
-    -  Typically, there is a single rule for each TX port and the address spaces
-       of the rules do not overlap, e.g. for each TX_PORT used by the
-       application, the rule "10.0.TX_PORT.0/24 => TX_PORT" is included in the
-       list.
--  --rsz: Ring sizes
-    -  Typically, the default values are used (parameter not present in the
-       command line).
--  --bsz: Burst sizes
-    -  Typically, the default values are used (parameter not present in the
-       command line).
--  --pos-lb: Position of the 1-byte header field within the input packet that is
-   used to determine the worker ID for each packet
-
-    -  Typically, the default value is used (parameter not present in the
-       command line).
-
-5. Traffic generator configuration
-
-The application is used to benchmark the penalty of packets going across
-different CPU sockets. In the general case, the input packets are RX-ed by NIC
-connected to CPU socket X, dispatched to worker running on CPU socket Y, TX-ed
-by NIC connected to CPU socket Z. The typical cases under test are: AAA, AAB,
-ABB, ABC (for 2-socket systems, ABC is actually ABA).
-
-The worker ID is determined by reading a 1-byte field from the input packet. Its
-position is specified using the --pos-lb command line argument. For convenience,
-the --pos-lb argument typically points to the last byte of the IPv4 source
-address, e.g. the IPv4 source address for a traffic flow that shoud be processed
-by WORKER_ID is: 0.0.0.WORKER_ID.
-
-The TX port is determined by the LPM rule that is hit by the IPv4 destination
-address field read from each packet. Therefore, the traffic generator
-configuration has to be in sync with the routing table of the application
-(provided using the --lpm parameter). Given the convention described above of
-LPM rules of: "10.0.TX_PORT.0/24 => TX_PORT", then packets with IPv4 destination
-address of 10.0.TX_PORT.1 will be sent out by TX_PORT, regardless of the worker
-lcore processing them.
-
-For a specific test case, the recommended flow configuration for each traffic
-generator port (connected to a NIC attached to CPU socket X) is to create a
-traffic flow for each pair of (worker on CPU socket Y, NIC TX port on CPU
-socket Z) and equally divide the TX rate amongst all the traffic flows on the
-same traffic generator port. This guarantees that all the workers on CPU
-socket Y will be hit evenly and none of the NIC TX ports on CPU socket Z will be
-oversubscribed.
-
-In this case, the same set of application command lines (testing different
-packet I/O and worker set configurations) can be applied with no modifications
-to test scenarios AAA, AAB, ABB, ABC/ABA by simply modifying two fields within
-each of the traffic flows sent by the traffic generator on each of its ports.
-
-Test Case: Load Balancer
-========================
-
-Assuming that Logical core 4, 5, 6, 7 are connected to a traffic generator,
-launch the ``load_balancer`` with the following arguments::
-
-  ./examples/load_balancer/build/load_balancer -l 3-7 -n 4 -- \
-  --rx "(0,0,3),(1,0,3),(2,0,3),(3,0,3)" \
-  --tx "(0,3),(1,3),(2,3),(3,3)" --w "4,5,6,7" \
-  --lpm "1.0.0.0/24=>0;1.0.1.0/24=>1;1.0.2.0/24=>2;1.0.3.0/24=>3; " \
-  --bsz "(10, 10), (10, 10), (10, 10)" --pos-lb 29
-
-If the app run successfully, it will be the same as the shown in the terminal. ::
-
-  ...
-  LPM rules:
-        0: 1.0.0.0/24 => 0;
-        1: 1.0.1.0/24 => 1;
-        2: 1.0.2.0/24 => 2;
-        3: 1.0.3.0/24 => 3;
-  Ring sizes: NIC RX = 1024; Worker in = 1024; Worker out = 1024; NIC TX = 1024;
-  Burst sizes: I/O RX (rd = 10, wr = 10); Worker (rd = 10, wr = 10); I/O TX (rd = 10, wr = 10)
-  Logical core 4 (worker 0) main loop.
-  Logical core 5 (worker 1) main loop.
-  Logical core 6 (worker 2) main loop.
-  Logical core 7 (worker 3) main loop.
-  Logical core 3 (I/O) main loop.
diff --git a/tests/TestSuite_loadbalancer.py b/tests/TestSuite_loadbalancer.py
deleted file mode 100644
index 84e534c..0000000
--- a/tests/TestSuite_loadbalancer.py
+++ /dev/null
@@ -1,154 +0,0 @@
-# BSD LICENSE
-#
-# Copyright(c) 2020 Intel Corporation. All rights reserved.
-# 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 Test suite.
-
-Test Load Balancer.
-
-"""
-
-import dts
-from packet import Packet
-from test_case import TestCase
-import utils
-import time
-
-
-class TestLoadbalancer(TestCase):
-
-    def set_up_all(self):
-        """
-        Run at the start of each test suite.
-
-        Load Balancer prerequisites.
-        """
-        # Verify that enough ports are available
-        global dutPorts
-        # Based on h/w type, choose how many ports to use
-        dutPorts = self.dut.get_ports(self.nic)
-
-        # Verify that enough ports are available
-        self.verify(len(dutPorts) >= 4, "Insufficient ports for testing")
-
-        cores = self.dut.get_core_list("all")
-        self.verify(len(cores) >= 5, "Insufficient cores for testing")
-        self.cores = self.dut.get_core_list("1S/5C/1T")
-        self.coremask = utils.create_mask(self.cores)
-
-        global rx_port0, rx_port1, rx_port2, rx_port3, trafficFlow
-        rx_port0 = self.tester.get_local_port(dutPorts[0])
-        rx_port1 = self.tester.get_local_port(dutPorts[1])
-        rx_port2 = self.tester.get_local_port(dutPorts[2])
-        rx_port3 = self.tester.get_local_port(dutPorts[3])
-
-        """
-        Designation the traffic flow is the same as LPM rules, send and receive packet verification:
-            0: 1.0.0.0/24 => 0;
-            1: 1.0.1.0/24 => 1;
-            2: 1.0.2.0/24 => 2;
-            3: 1.0.3.0/24 => 3;
-        """
-        trafficFlow = {
-            "Flow1": [rx_port0, "1.0.0.1"],
-            "Flow2": [rx_port1, "1.0.1.1"],
-            "Flow3": [rx_port2, "1.0.2.1"],
-            "Flow4": [rx_port3, "1.0.3.1"],
-        }
-
-        out = self.dut.build_dpdk_apps("examples/load_balancer")
-        self.verify("Error" not in out, "compilation error 1")
-        self.verify("No such file" not in out, "compilation error 2")
-
-    def set_up(self):
-        """
-        Run before each test case.
-        """
-        pass
-
-    def test_load_balancer(self):
-        """
-        --rx: Set the receive port, queue and main core;
-        --tx: Set the send port and main core;
-        --w: specify 4 workers lcores,
-        --lpm: IPv4 routing table,
-        --bsz: The number of packet is 10 for transceivers,
-        --pos-lb: Position of the 1-byte header field within the input packet that is used to
-        determine the worker ID for each packet
-        """
-
-        cmd = './examples/load_balancer/build/load_balancer -l {0}-{1} -n 4 -- --rx "(0,0,{2}),(1,0,{2}),(2,0,{2}),(3,0,{2})" '\
-              '--tx "(0,{2}),(1,{2}),(2,{2}),(3,{2})" --w "{3},{4},{5},{6}" '\
-              '--lpm "1.0.0.0/24=>0;1.0.1.0/24=>1;1.0.2.0/24=>2;1.0.3.0/24=>3;" '\
-              '--bsz "(10, 10), (10, 10), (10, 10)" --pos-lb 29'.format(self.cores[0], self.cores[4], self.cores[0], self.cores[1], self.cores[2], self.cores[3], self.cores[4])
-
-        self.dut.send_expect(cmd, 'main loop.')
-
-        # Verify the traffic flow according to Ipv4 route table
-        for flow in list(trafficFlow.keys()):
-            rx_port = trafficFlow[flow][0]
-
-            for i in range(len(dutPorts)):
-                dstport = self.tester.get_local_port(dutPorts[i])
-                pkt_count = 10
-                inst = self.tester.tcpdump_sniff_packets(intf=self.tester.get_interface(rx_port), count=pkt_count)
-
-                pkt = Packet(pkt_type='UDP', pkt_len=64)
-                dst_mac = self.dut.get_mac_address(dutPorts[i])
-                pkt.config_layer('ether', {'dst': '%s' % dst_mac})
-                pkt.config_layer('ipv4', {'src': "0.0.0.1", 'dst': '%s' % (trafficFlow[flow][1])})
-                pkt.send_pkt(self.tester, tx_port=self.tester.get_interface(dstport), count=10)
-                # Wait for the sniffer to finish.
-                time.sleep(5)
-
-                pkts = self.tester.load_tcpdump_sniff_packets(inst)
-                len_pkts = len(pkts)
-
-                self.verify(len_pkts == pkt_count, "Packet number is wrong")
-                for i in range(len_pkts):
-                    result = str(pkts[i].show)
-                    self.verify("Ether" in result, "No packet received")
-                    self.verify("src=0.0.0.1" + " dst=" + trafficFlow[flow][1] in result, "Wrong IP address")
-                    self.verify("dst=%s" % dst_mac in result, "No packet received or packet missed")
-
-        self.dut.send_expect("^C", "#")
-
-    def tear_down(self):
-        """
-        Run after each test case.
-        """
-        self.dut.kill_all()
-
-    def tear_down_all(self):
-        """
-        Run after each test suite.
-        """
-        pass
-- 
2.7.4


             reply	other threads:[~2021-09-18  2:08 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-18  2:08 Yu Jiang [this message]
2021-09-18  2:08 ` [dts] [PATCH V1] test_plans/netmap_compat: " Yu Jiang
2021-09-27  5:52   ` Tu, Lijuan
2021-09-27  5:58 ` [dts] [PATCH V1] test_plans/loadbalancer: " Tu, Lijuan

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=1631930882-12140-1-git-send-email-yux.jiang@intel.com \
    --to=yux.jiang@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).