* [dts][PATCH V3 0/8] split performance plan and suite
@ 2023-01-10 16:44 Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 1/8] tests/efd: " Hongbo Li
` (7 more replies)
0 siblings, 8 replies; 9+ messages in thread
From: Hongbo Li @ 2023-01-10 16:44 UTC (permalink / raw)
To: dts; +Cc: Hongbo Li
split performance plan and suite
Hongbo Li (8):
tests/efd: split performance plan and suite
tests/l2fwd: split performance plan and suite
tests/tso: split performance plan and suite
tests/vxlan: split performance plan and suite
tests/ipfrag: split performance plan and suite
tests/multiprocess: split performance plan and suite
tests/checksum_offload: split performance plan and suite
conf/test_case_checklist: split performance plan and suite
test_plans/efd_test_plan.rst | 77 +-
test_plans/index.rst | 6 +
test_plans/ipfrag_test_plan.rst | 33 +-
test_plans/l2fwd_test_plan.rst | 36 -
test_plans/multiprocess_test_plan.rst | 48 -
test_plans/perf_efd_test_plan.rst | 105 ++
test_plans/perf_ipfrag_test_plan.rst | 141 +++
test_plans/perf_l2fwd_test_plan.rst | 90 ++
test_plans/perf_multiprocess_test_plan.rst | 141 +++
test_plans/perf_tso_test_plan.rst | 87 ++
test_plans/perf_vxlan_test_plan.rst | 85 ++
test_plans/tso_test_plan.rst | 34 -
test_plans/vxlan_test_plan.rst | 60 +-
tests/TestSuite_checksum_offload.py | 54 --
tests/TestSuite_efd.py | 138 ---
tests/TestSuite_ipfrag.py | 21 -
tests/TestSuite_l2fwd.py | 106 --
tests/TestSuite_multiprocess.py | 210 ----
tests/TestSuite_perf_checksum_offload.py | 624 ++++++++++++
tests/TestSuite_perf_efd.py | 210 ++++
tests/TestSuite_perf_ipfrag.py | 386 ++++++++
tests/TestSuite_perf_l2fwd.py | 233 +++++
tests/TestSuite_perf_multiprocess.py | 994 +++++++++++++++++++
tests/TestSuite_perf_tso.py | 302 ++++++
tests/TestSuite_perf_vxlan.py | 1022 ++++++++++++++++++++
tests/TestSuite_tso.py | 113 ---
tests/TestSuite_vxlan.py | 213 ----
27 files changed, 4429 insertions(+), 1140 deletions(-)
create mode 100644 test_plans/perf_efd_test_plan.rst
create mode 100644 test_plans/perf_ipfrag_test_plan.rst
create mode 100644 test_plans/perf_l2fwd_test_plan.rst
create mode 100644 test_plans/perf_multiprocess_test_plan.rst
create mode 100644 test_plans/perf_tso_test_plan.rst
create mode 100644 test_plans/perf_vxlan_test_plan.rst
create mode 100644 tests/TestSuite_perf_checksum_offload.py
create mode 100644 tests/TestSuite_perf_efd.py
create mode 100644 tests/TestSuite_perf_ipfrag.py
create mode 100644 tests/TestSuite_perf_l2fwd.py
create mode 100644 tests/TestSuite_perf_multiprocess.py
create mode 100644 tests/TestSuite_perf_tso.py
create mode 100644 tests/TestSuite_perf_vxlan.py
--
2.17.1
^ permalink raw reply [flat|nested] 9+ messages in thread
* [dts][PATCH V3 1/8] tests/efd: split performance plan and suite
2023-01-10 16:44 [dts][PATCH V3 0/8] split performance plan and suite Hongbo Li
@ 2023-01-10 16:44 ` Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 2/8] tests/l2fwd: " Hongbo Li
` (6 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: Hongbo Li @ 2023-01-10 16:44 UTC (permalink / raw)
To: dts; +Cc: Hongbo Li
split performance plan and suite
Signed-off-by: Hongbo Li <hongbox.li@intel.com>
---
v2:
add patch 0
submit index.rst file separately
v3:
solve format problems
test_plans/efd_test_plan.rst | 77 +----------
test_plans/perf_efd_test_plan.rst | 105 +++++++++++++++
tests/TestSuite_efd.py | 138 --------------------
tests/TestSuite_perf_efd.py | 210 ++++++++++++++++++++++++++++++
4 files changed, 316 insertions(+), 214 deletions(-)
create mode 100644 test_plans/perf_efd_test_plan.rst
create mode 100644 tests/TestSuite_perf_efd.py
diff --git a/test_plans/efd_test_plan.rst b/test_plans/efd_test_plan.rst
index ee440c02..70dd3b53 100644
--- a/test_plans/efd_test_plan.rst
+++ b/test_plans/efd_test_plan.rst
@@ -49,79 +49,4 @@ Start test application and run EFD performance unit test::
Verify lookup and lookup bulk cpu cycles are reasonable.
Verify when key size increased, no significant increment in cpu cycles.
Verify when value bits increased, no significant increment in cpu cycles.
-Compare with cuckoo hash performance result, lookup cycles should be less.
-
-Performance test cases
-==============================
-In EFD sample, EFD work as a flow-level load balancer. Flows are received at
-a front end server before being forwarded to the target back end server for
-processing. This case will measure the performance of flow distribution with
-different parameters.
-
-Nodes: number of back end nodes
-Entries: number of flows to be added in EFD table
-Value bits: number of bits of value that be stored in EFD table
-
-Test Case: Load balancer performance based on node numbers
-----------------------------------------------------------------------
-This case will measure the performance based on node numbers.
-
-+--------------+-------+-----------+------------+
-| Value Bits | Nodes | Entries | Throughput |
-+--------------+-------+-----------+------------+
-| 8 | 1 | 2M | |
-+--------------+-------+-----------+------------+
-| 8 | 2 | 2M | |
-+--------------+-------+-----------+------------+
-| 8 | 3 | 2M | |
-+--------------+-------+-----------+------------+
-| 8 | 4 | 2M | |
-+--------------+-------+-----------+------------+
-| 8 | 5 | 2M | |
-+--------------+-------+-----------+------------+
-| 8 | 6 | 2M | |
-+--------------+-------+-----------+------------+
-| 8 | 7 | 2M | |
-+--------------+-------+-----------+------------+
-| 8 | 8 | 2M | |
-+--------------+-------+-----------+------------+
-
-Test Case: Load balancer performance based on flow numbers
------------------------------------------------------------------------
-This case will measure the performance based on flow numbers.
-
-+--------------+-------+-----------+------------+
-| Value Bits | Nodes | Entries | Throughput |
-+--------------+-------+-----------+------------+
-| 8 | 2 | 1M | |
-+--------------+-------+-----------+------------+
-| 8 | 2 | 2M | |
-+--------------+-------+-----------+------------+
-| 8 | 2 | 4M | |
-+--------------+-------+-----------+------------+
-| 8 | 2 | 8M | |
-+--------------+-------+-----------+------------+
-| 8 | 2 | 16M | |
-+--------------+-------+-----------+------------+
-| 8 | 2 | 32M | |
-+--------------+-------+-----------+------------+
-
-Test Case: Load balancer performance based on value bits
------------------------------------------------------------------------
-Modify different value size which must be between 1 and 32, and it need
-to be configured with the ``-Dc_args=-DRTE_EFD_VALUE_NUM_BITS`` option
-at compile time.
-
-This case will measure the performance based on value bits.
-
-+--------------+-------+-----------+------------+
-| Value Bits | Nodes | Entries | Throughput |
-+--------------+-------+-----------+------------+
-| 8 | 2 | 2M | |
-+--------------+-------+-----------+------------+
-| 16 | 2 | 2M | |
-+--------------+-------+-----------+------------+
-| 24 | 2 | 2M | |
-+--------------+-------+-----------+------------+
-| 32 | 2 | 2M | |
-+--------------+-------+-----------+------------+
+Compare with cuckoo hash performance result, lookup cycles should be less.
\ No newline at end of file
diff --git a/test_plans/perf_efd_test_plan.rst b/test_plans/perf_efd_test_plan.rst
new file mode 100644
index 00000000..fe3fedd8
--- /dev/null
+++ b/test_plans/perf_efd_test_plan.rst
@@ -0,0 +1,105 @@
+.. SPDX-License-Identifier: BSD-3-Clause
+ Copyright(c) 2010-2017 Intel Corporation
+
+==================================================
+Sample Application Tests: Elastic Flow Distributor
+==================================================
+
+Description
+-----------
+EFD is a distributor library that uses perfect hashing to determine a
+target/value for a given incoming flow key.
+It has the following advantages:
+1. It uses perfect hashing it does not store the key itself and hence
+lookup performance is not dependent on the key size.
+2. Target/value can be any arbitrary value hence the system designer
+and/or operator can better optimize service rates and inter-cluster
+network traffic locating.
+3. Since the storage requirement is much smaller than a hash-based flow
+table (i.e. better fit for CPU cache), EFD can scale to millions of flow
+keys.
+4. With the current optimized library implementation, performance is fully
+scalable with any number of CPU cores.
+
+For more details, please reference to dpdk online programming guide.
+
+Prerequisites
+=============
+Two ports connect to packet generator.
+
+DUT board must be two sockets system and each cpu have more than 16 lcores.
+
+Performance test cases
+==============================
+In EFD sample, EFD work as a flow-level load balancer. Flows are received at
+a front end server before being forwarded to the target back end server for
+processing. This case will measure the performance of flow distribution with
+different parameters.
+
+Nodes: number of back end nodes
+Entries: number of flows to be added in EFD table
+Value bits: number of bits of value that be stored in EFD table
+
+Test Case: Load balancer performance based on node numbers
+----------------------------------------------------------------------
+This case will measure the performance based on node numbers.
+
++--------------+-------+-----------+------------+
+| Value Bits | Nodes | Entries | Throughput |
++--------------+-------+-----------+------------+
+| 8 | 1 | 2M | |
++--------------+-------+-----------+------------+
+| 8 | 2 | 2M | |
++--------------+-------+-----------+------------+
+| 8 | 3 | 2M | |
++--------------+-------+-----------+------------+
+| 8 | 4 | 2M | |
++--------------+-------+-----------+------------+
+| 8 | 5 | 2M | |
++--------------+-------+-----------+------------+
+| 8 | 6 | 2M | |
++--------------+-------+-----------+------------+
+| 8 | 7 | 2M | |
++--------------+-------+-----------+------------+
+| 8 | 8 | 2M | |
++--------------+-------+-----------+------------+
+
+Test Case: Load balancer performance based on flow numbers
+-----------------------------------------------------------------------
+This case will measure the performance based on flow numbers.
+
++--------------+-------+-----------+------------+
+| Value Bits | Nodes | Entries | Throughput |
++--------------+-------+-----------+------------+
+| 8 | 2 | 1M | |
++--------------+-------+-----------+------------+
+| 8 | 2 | 2M | |
++--------------+-------+-----------+------------+
+| 8 | 2 | 4M | |
++--------------+-------+-----------+------------+
+| 8 | 2 | 8M | |
++--------------+-------+-----------+------------+
+| 8 | 2 | 16M | |
++--------------+-------+-----------+------------+
+| 8 | 2 | 32M | |
++--------------+-------+-----------+------------+
+
+Test Case: Load balancer performance based on value bits
+-----------------------------------------------------------------------
+Modify different value size which must be between 1 and 32, and it need
+to be configured with the ``-Dc_args=-DRTE_EFD_VALUE_NUM_BITS`` option
+at compile time.
+
+This case will measure the performance based on value bits.
+
++--------------+-------+-----------+------------+
+| Value Bits | Nodes | Entries | Throughput |
++--------------+-------+-----------+------------+
+| 8 | 2 | 2M | |
++--------------+-------+-----------+------------+
+| 16 | 2 | 2M | |
++--------------+-------+-----------+------------+
+| 24 | 2 | 2M | |
++--------------+-------+-----------+------------+
+| 32 | 2 | 2M | |
++--------------+-------+-----------+------------+
diff --git a/tests/TestSuite_efd.py b/tests/TestSuite_efd.py
index 9d94a3d4..071a3564 100644
--- a/tests/TestSuite_efd.py
+++ b/tests/TestSuite_efd.py
@@ -68,144 +68,6 @@ class TestEFD(TestCase):
self.dut.send_expect("quit", "# ")
self.verify("Test OK" in out, "Test failed")
- def test_perf_efd_nodenum(self):
- """
- Run EFD perf evaluation for number of nodes
- """
- self.verify(len(self.dut_ports) >= 2, "Not enough ports")
- node_nums = [1, 2, 3, 4, 5, 6, 7, 8]
-
- flow_num = 1024 * 1024 * 2
-
- table_header = ["Value Bits", "Nodes", "Flow Entries", "Throughput(mpps)"]
-
- self.result_table_create(table_header)
- # perf of different nodes
- for node_num in node_nums:
- pps = self._efd_perf_evaluate(node_num, flow_num)
-
- self.result_table_add([8, node_num, "2M", pps])
-
- self.result_table_print()
-
- def test_perf_efd_flownums(self):
- """
- Run EFD perf evaluation for millions of flows
- """
- self.logger.warning(
- "Millions of flow required huge memory, please allocate 16G hugepage"
- )
- self.dut.setup_memory_linux(hugepages=8192)
- self.verify(len(self.dut_ports) >= 2, "Not enough ports")
- flow_nums = [
- 1024 * 1024,
- 1024 * 1024 * 2,
- 1024 * 1024 * 4,
- 1024 * 1024 * 8,
- 1024 * 1024 * 16,
- 1024 * 1024 * 32,
- ]
-
- table_header = ["Value Bits", "Nodes", "Million Flows", "Throughput(mpps)"]
-
- self.result_table_create(table_header)
-
- # perf of different flow numbers
- for flow_num in flow_nums:
- pps = self._efd_perf_evaluate(2, flow_num)
-
- self.result_table_add([8, 2, flow_num / (1024 * 1024), pps])
-
- self.result_table_print()
-
- def test_perf_efd_valuesize(self):
- """
- Run EFD perf evaluation for different value size
- """
- self.verify(len(self.dut_ports) >= 2, "Not enough ports")
- val_bitnums = [8, 16, 24, 32]
- flow_num = 1024 * 1024 * 2
-
- table_header = ["Value Bits", "Nodes", "Flow Entries", "Throughput(mpps)"]
-
- self.result_table_create(table_header)
- # perf of different value bit lengths
- for val_bitnum in val_bitnums:
- # change value length and rebuild dpdk
- extra_options = "-Dc_args=-DRTE_EFD_VALUE_NUM_BITS=%d" % val_bitnum
- self.dut.build_install_dpdk(self.target, extra_options=extra_options)
- self.build_server_node_efd()
-
- pps = self._efd_perf_evaluate(2, flow_num)
- self.result_table_add([val_bitnum, 2, "2M", pps])
-
- self.result_table_print()
- extra_options = "-Dc_args=-DRTE_EFD_VALUE_NUM_BITS=8"
- self.dut.build_install_dpdk(self.target, extra_options=extra_options)
- self.build_server_node_efd()
-
- def _efd_perf_evaluate(self, node_num, flow_num):
- # extended flow number into etgen module
-
- # output port is calculated from overall ports number
- server_cmd_fmt = "%s %s -- -p 0x3 -n %d -f %s"
- node_cmd_fmt = "%s %s --proc-type=secondary -- -n %d"
- socket = self.dut.get_numa_id(self.dut_ports[0])
-
- pcap = os.sep.join([self.output_path, "efd.pcap"])
- self.tester.scapy_append(
- 'wrpcap("%s", [Ether()/IP(src="0.0.0.0", dst="0.0.0.0")/("X"*26)])' % pcap
- )
- self.tester.scapy_execute()
-
- tgen_input = []
- rx_port = self.tester.get_local_port(self.dut_ports[0])
- tx_port = self.tester.get_local_port(self.dut_ports[1])
-
- pcap = os.sep.join([self.output_path, "efd.pcap"])
- tgen_input.append((tx_port, rx_port, pcap))
- tgen_input.append((rx_port, tx_port, pcap))
-
- cores = self.dut.get_core_list("1S/%dC/1T" % (node_num + 2), socket)
-
- self.verify(len(cores), "Can't find enough cores")
-
- eal_para = self.dut.create_eal_parameters(cores=cores[0:2], ports=[0, 1])
- server_cmd = server_cmd_fmt % (
- self.server_app,
- eal_para,
- node_num,
- hex(flow_num),
- )
- # create table may need few minutes
- self.dut.send_expect(server_cmd, "Finished Process Init", timeout=240)
-
- node_sessions = []
- for node in range(node_num):
-
- eal_para = self.dut.create_eal_parameters(cores=[cores[2 + node]])
- node_cmd = node_cmd_fmt % (self.node_app, eal_para, node)
- node_session = self.dut.new_session(suite="node%d" % node)
- node_sessions.append(node_session)
- node_session.send_expect(node_cmd, "Finished Process Init", timeout=30)
-
- # clear streams before add new streams
- self.tester.pktgen.clear_streams()
- # run packet generator
- streams = self.pktgen_helper.prepare_stream_from_tginput(
- tgen_input, 100, None, self.tester.pktgen
- )
- _, pps = self.tester.pktgen.measure_throughput(stream_ids=streams)
-
- for node_session in node_sessions:
- node_session.send_expect("^C", "#")
- self.dut.close_session(node_session)
-
- self.dut.send_expect("^C", "#")
-
- pps /= 1000000.0
- return pps
-
def set_fields(self):
"""set ip protocol field behavior"""
fields_config = {
diff --git a/tests/TestSuite_perf_efd.py b/tests/TestSuite_perf_efd.py
new file mode 100644
index 00000000..5d638d2d
--- /dev/null
+++ b/tests/TestSuite_perf_efd.py
@@ -0,0 +1,210 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2010-2017 Intel Corporation
+#
+
+"""
+DPDK Test suite.
+"""
+import os
+import re
+
+import framework.utils as utils
+from framework.pktgen import PacketGeneratorHelper
+from framework.test_case import TestCase
+
+
+class TestEFD(TestCase):
+ def set_up_all(self):
+ """
+ Run at the start of each test suite.
+ """
+
+ self.build_server_node_efd()
+
+ self.dut_ports = self.dut.get_ports()
+ self.node_app = self.dut.apps_name["node"]
+ self.server_app = self.dut.apps_name["server"]
+ self.app_test_path = self.dut.apps_name["test"]
+ # get dts output path
+ if self.logger.log_path.startswith(os.sep):
+ self.output_path = self.logger.log_path
+ else:
+ cur_path = os.path.dirname(os.path.dirname(os.path.realpath(__file__)))
+ self.output_path = os.sep.join([cur_path, self.logger.log_path])
+ # create an instance to set stream field setting
+ self.pktgen_helper = PacketGeneratorHelper()
+
+ def build_server_node_efd(self):
+ apps = ["node", "server"]
+ for app in apps:
+ out = self.dut.build_dpdk_apps("./examples/server_node_efd/%s" % app)
+ self.verify("Error" not in out, "Compilation %s error" % app)
+ self.verify("No such" not in out, "Compilation %s error" % app)
+
+ def set_up(self):
+ """
+ Run before each test case.
+ """
+ pass
+
+ def test_perf_efd_nodenum(self):
+ """
+ Run EFD perf evaluation for number of nodes
+ """
+ self.verify(len(self.dut_ports) >= 2, "Not enough ports")
+ node_nums = [1, 2, 3, 4, 5, 6, 7, 8]
+
+ flow_num = 1024 * 1024 * 2
+
+ table_header = ["Value Bits", "Nodes", "Flow Entries", "Throughput(mpps)"]
+
+ self.result_table_create(table_header)
+ # perf of different nodes
+ for node_num in node_nums:
+ pps = self._efd_perf_evaluate(node_num, flow_num)
+
+ self.result_table_add([8, node_num, "2M", pps])
+
+ self.result_table_print()
+
+ def test_perf_efd_flownums(self):
+ """
+ Run EFD perf evaluation for millions of flows
+ """
+ self.logger.warning(
+ "Millions of flow required huge memory, please allocate 16G hugepage"
+ )
+ self.dut.setup_memory_linux(hugepages=8192)
+ self.verify(len(self.dut_ports) >= 2, "Not enough ports")
+ flow_nums = [
+ 1024 * 1024,
+ 1024 * 1024 * 2,
+ 1024 * 1024 * 4,
+ 1024 * 1024 * 8,
+ 1024 * 1024 * 16,
+ 1024 * 1024 * 32,
+ ]
+
+ table_header = ["Value Bits", "Nodes", "Million Flows", "Throughput(mpps)"]
+
+ self.result_table_create(table_header)
+
+ # perf of different flow numbers
+ for flow_num in flow_nums:
+ pps = self._efd_perf_evaluate(2, flow_num)
+
+ self.result_table_add([8, 2, flow_num / (1024 * 1024), pps])
+
+ self.result_table_print()
+
+ def test_perf_efd_valuesize(self):
+ """
+ Run EFD perf evaluation for different value size
+ """
+ self.verify(len(self.dut_ports) >= 2, "Not enough ports")
+ val_bitnums = [8, 16, 24, 32]
+ flow_num = 1024 * 1024 * 2
+
+ table_header = ["Value Bits", "Nodes", "Flow Entries", "Throughput(mpps)"]
+
+ self.result_table_create(table_header)
+ # perf of different value bit lengths
+ for val_bitnum in val_bitnums:
+ # change value length and rebuild dpdk
+ extra_options = "-Dc_args=-DRTE_EFD_VALUE_NUM_BITS=%d" % val_bitnum
+ self.dut.build_install_dpdk(self.target, extra_options=extra_options)
+ self.build_server_node_efd()
+
+ pps = self._efd_perf_evaluate(2, flow_num)
+ self.result_table_add([val_bitnum, 2, "2M", pps])
+
+ self.result_table_print()
+ extra_options = "-Dc_args=-DRTE_EFD_VALUE_NUM_BITS=8"
+ self.dut.build_install_dpdk(self.target, extra_options=extra_options)
+ self.build_server_node_efd()
+
+ def _efd_perf_evaluate(self, node_num, flow_num):
+ # extended flow number into etgen module
+
+ # output port is calculated from overall ports number
+ server_cmd_fmt = "%s %s -- -p 0x3 -n %d -f %s"
+ node_cmd_fmt = "%s %s --proc-type=secondary -- -n %d"
+ socket = self.dut.get_numa_id(self.dut_ports[0])
+
+ pcap = os.sep.join([self.output_path, "efd.pcap"])
+ self.tester.scapy_append(
+ 'wrpcap("%s", [Ether()/IP(src="0.0.0.0", dst="0.0.0.0")/("X"*26)])' % pcap
+ )
+ self.tester.scapy_execute()
+
+ tgen_input = []
+ rx_port = self.tester.get_local_port(self.dut_ports[0])
+ tx_port = self.tester.get_local_port(self.dut_ports[1])
+
+ pcap = os.sep.join([self.output_path, "efd.pcap"])
+ tgen_input.append((tx_port, rx_port, pcap))
+ tgen_input.append((rx_port, tx_port, pcap))
+
+ cores = self.dut.get_core_list("1S/%dC/1T" % (node_num + 2), socket)
+
+ self.verify(len(cores), "Can't find enough cores")
+
+ eal_para = self.dut.create_eal_parameters(cores=cores[0:2], ports=[0, 1])
+ server_cmd = server_cmd_fmt % (
+ self.server_app,
+ eal_para,
+ node_num,
+ hex(flow_num),
+ )
+ # create table may need few minutes
+ self.dut.send_expect(server_cmd, "Finished Process Init", timeout=240)
+
+ node_sessions = []
+ for node in range(node_num):
+
+ eal_para = self.dut.create_eal_parameters(cores=[cores[2 + node]])
+ node_cmd = node_cmd_fmt % (self.node_app, eal_para, node)
+ node_session = self.dut.new_session(suite="node%d" % node)
+ node_sessions.append(node_session)
+ node_session.send_expect(node_cmd, "Finished Process Init", timeout=30)
+
+ # clear streams before add new streams
+ self.tester.pktgen.clear_streams()
+ # run packet generator
+ streams = self.pktgen_helper.prepare_stream_from_tginput(
+ tgen_input, 100, None, self.tester.pktgen
+ )
+ _, pps = self.tester.pktgen.measure_throughput(stream_ids=streams)
+
+ for node_session in node_sessions:
+ node_session.send_expect("^C", "#")
+ self.dut.close_session(node_session)
+
+ self.dut.send_expect("^C", "#")
+
+ pps /= 1000000.0
+ return pps
+
+ def set_fields(self):
+ """set ip protocol field behavior"""
+ fields_config = {
+ "ip": {
+ # self.flow_num not used by this suite
+ # 'dst': {'range': self.flow_num, 'action': 'inc'}
+ "dst": {"range": 64, "action": "inc"}
+ },
+ }
+ return fields_config
+
+ def tear_down(self):
+ """
+ Run after each test case.
+ """
+ self.dut.kill_all()
+ pass
+
+ def tear_down_all(self):
+ """
+ Run after each test suite.
+ """
+ pass
--
2.17.1
^ permalink raw reply [flat|nested] 9+ messages in thread
* [dts][PATCH V3 2/8] tests/l2fwd: split performance plan and suite
2023-01-10 16:44 [dts][PATCH V3 0/8] split performance plan and suite Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 1/8] tests/efd: " Hongbo Li
@ 2023-01-10 16:44 ` Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 3/8] tests/tso: " Hongbo Li
` (5 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: Hongbo Li @ 2023-01-10 16:44 UTC (permalink / raw)
To: dts; +Cc: Hongbo Li
split performance plan and suite
Signed-off-by: Hongbo Li <hongbox.li@intel.com>
---
test_plans/l2fwd_test_plan.rst | 36 -----
test_plans/perf_l2fwd_test_plan.rst | 90 +++++++++++
tests/TestSuite_l2fwd.py | 106 -------------
tests/TestSuite_perf_l2fwd.py | 233 ++++++++++++++++++++++++++++
4 files changed, 323 insertions(+), 142 deletions(-)
create mode 100644 test_plans/perf_l2fwd_test_plan.rst
create mode 100644 tests/TestSuite_perf_l2fwd.py
diff --git a/test_plans/l2fwd_test_plan.rst b/test_plans/l2fwd_test_plan.rst
index d8803cb5..1bf9ad45 100644
--- a/test_plans/l2fwd_test_plan.rst
+++ b/test_plans/l2fwd_test_plan.rst
@@ -69,39 +69,3 @@ Trigger the packet generator of bursting packets from ``port A``, then check if
``port 0`` could receive them and ``port 1`` could forward them back. Stop it
and then trigger the packet generator of bursting packets from ``port B``, then
check if ``port 1`` could receive them and ``port 0`` could forward them back.
-
-Test Case: ``64/128/256/512/1024/1500`` bytes packet forwarding test
-====================================================================
-
-Set the packet stream to be sent out from packet generator before testing as below.
-
-+-------+---------+---------+---------+-----------+
-| Frame | 1q | 2q | 4q | 8 q |
-| Size | | | | |
-+-------+---------+---------+---------+-----------+
-| 64 | | | | |
-+-------+---------+---------+---------+-----------+
-| 65 | | | | |
-+-------+---------+---------+---------+-----------+
-| 128 | | | | |
-+-------+---------+---------+---------+-----------+
-| 256 | | | | |
-+-------+---------+---------+---------+-----------+
-| 512 | | | | |
-+-------+---------+---------+---------+-----------+
-| 1024 | | | | |
-+-------+---------+---------+---------+-----------+
-| 1280 | | | | |
-+-------+---------+---------+---------+-----------+
-| 1518 | | | | |
-+-------+---------+---------+---------+-----------+
-
-Then run the test application as below::
-
- $ ./x86_64-native-linuxapp-gcc/examples/dpdk-l2fwd -n 2 -c f -- -q 1 -p 0x3
-
-The -n command is used to select the number of memory channels. It should match the number of memory channels on that setup.
-
-Trigger the packet generator of bursting packets to the port 0 and 1 on the onboard
-NIC to be tested. Then measure the forwarding throughput for different packet sizes
-and different number of queues.
diff --git a/test_plans/perf_l2fwd_test_plan.rst b/test_plans/perf_l2fwd_test_plan.rst
new file mode 100644
index 00000000..e5fbaf30
--- /dev/null
+++ b/test_plans/perf_l2fwd_test_plan.rst
@@ -0,0 +1,90 @@
+.. SPDX-License-Identifier: BSD-3-Clause
+ Copyright(c) 2010-2017 Intel Corporation
+
+===================
+L2 Forwarding Tests
+===================
+
+This test application is a basic packet processing application using Intel®
+DPDK. It is a layer-2 (L2) forwarding application which takes traffic from
+a single RX port and transmits it with few modification on a single TX port.
+
+For a packet received on a RX port (RX_PORT), it would be transmitted from a
+TX port (TX_PORT=RX_PORT+1) if RX_PORT is even; otherwise from a TX port
+(TX_PORT=RX_PORT-1) if RX_PORT is odd. Before being transmitted, the source
+mac address of the packet would be replaced by the mac address of the TX port,
+while the destination mac address would be replaced by 00:09:c0:00:00:TX_PORT_ID.
+
+The test application should be run with the wanted paired ports configured using
+the coremask parameter via the command line. i.e. port 0 and 1 is a valid pair,
+while port 1 and 2 isn't. The test is performed by running the test application
+and using a traffic generator. Tests are run with receiving a variety of size of
+packets generated by the traffic generator and forwarding back to the traffic
+generator. The packet loss and the throughput are the right ones need to be
+measured.
+
+The ``l2fwd`` application is run with EAL parameters and parameters for
+the application itself. For details about the EAL parameters, see the relevant
+DPDK **Getting Started Guide**. This application supports two parameters for
+itself.
+
+- ``-p PORTMASK``: hexadecimal bitmask of ports to configure
+- ``-q NQ``: number of queue per lcore (default is 1)
+
+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
+
+Assume port 0 and 1 are connected to the traffic generator, to run the test
+application in linuxapp environment with 4 lcores, 2 ports and 8 RX queues
+per lcore::
+
+ $ ./x86_64-native-linuxapp-gcc/examples/dpdk-l2fwd -n 1 -c f -- -q 8 -p 0x3
+
+Also, if the ports to be tested are different, the port mask should be changed.
+The lcore used to run the test application and the number of queue used for a
+lcore could be changed. For benchmarking, the EAL parameters and the parameters
+for the application itself for different test cases should be the same.
+
+Test Case: ``64/128/256/512/1024/1500`` bytes packet forwarding test
+====================================================================
+
+Set the packet stream to be sent out from packet generator before testing as below.
+
++-------+---------+---------+---------+-----------+
+| Frame | 1q | 2q | 4q | 8 q |
+| Size | | | | |
++-------+---------+---------+---------+-----------+
+| 64 | | | | |
++-------+---------+---------+---------+-----------+
+| 65 | | | | |
++-------+---------+---------+---------+-----------+
+| 128 | | | | |
++-------+---------+---------+---------+-----------+
+| 256 | | | | |
++-------+---------+---------+---------+-----------+
+| 512 | | | | |
++-------+---------+---------+---------+-----------+
+| 1024 | | | | |
++-------+---------+---------+---------+-----------+
+| 1280 | | | | |
++-------+---------+---------+---------+-----------+
+| 1518 | | | | |
++-------+---------+---------+---------+-----------+
+
+Then run the test application as below::
+
+ $ ./x86_64-native-linuxapp-gcc/examples/dpdk-l2fwd -n 2 -c f -- -q 1 -p 0x3
+
+The -n command is used to select the number of memory channels. It should match the number of memory channels on that setup.
+
+Trigger the packet generator of bursting packets to the port 0 and 1 on the onboard
+NIC to be tested. Then measure the forwarding throughput for different packet sizes
+and different number of queues.
diff --git a/tests/TestSuite_l2fwd.py b/tests/TestSuite_l2fwd.py
index 9e8fd2d5..7594958b 100644
--- a/tests/TestSuite_l2fwd.py
+++ b/tests/TestSuite_l2fwd.py
@@ -150,112 +150,6 @@ class TestL2fwd(TestCase):
self.quit_l2fwd()
- def test_perf_l2fwd_performance(self):
- """
- Benchmark performance for frame_sizes.
- """
- ports = []
- for port in range(self.number_of_ports):
- ports.append(self.dut_ports[port])
-
- port_mask = utils.create_mask(ports)
- cores = self.dut.get_core_list(self.core_config, socket=self.ports_socket)
-
- eal_params = self.dut.create_eal_parameters(cores=cores)
- eal_param = ""
- for i in ports:
- eal_param += " -a %s" % self.dut.ports_info[i]["pci"]
-
- for frame_size in self.frame_sizes:
-
- payload_size = frame_size - self.headers_size
-
- tgen_input = []
- cnt = 1
- for port in range(self.number_of_ports):
- rx_port = self.tester.get_local_port(
- self.dut_ports[port % self.number_of_ports]
- )
- tx_port = self.tester.get_local_port(
- self.dut_ports[(port + 1) % self.number_of_ports]
- )
- destination_mac = self.dut.get_mac_address(
- self.dut_ports[(port + 1) % self.number_of_ports]
- )
- pcap = os.sep.join(
- [self.output_path, "l2fwd_{0}_{1}.pcap".format(port, cnt)]
- )
- self.tester.scapy_append(
- 'wrpcap("%s", [Ether(dst="%s")/IP()/UDP()/("X"*%d)])'
- % (pcap, destination_mac, payload_size)
- )
- tgen_input.append((tx_port, rx_port, pcap))
- time.sleep(3)
- self.tester.scapy_execute()
- cnt += 1
-
- for queues in self.test_queues:
-
- command_line = "./%s %s %s -- -q %s -p %s &" % (
- self.app_l2fwd_path,
- eal_params,
- eal_param,
- str(queues["queues"]),
- port_mask,
- )
-
- # self.dut.send_expect(command_line, "memory mapped", 60)
- self.dut.send_expect(command_line, "L2FWD: entering main loop", 60)
- # wait 5 second after l2fwd boot up.
- # It is aimed to make sure trex detect link up status.
- if self.tester.is_pktgen:
- time.sleep(5)
- info = (
- "Executing l2fwd using %s queues, frame size %d and %s setup.\n"
- % (queues["queues"], frame_size, self.core_config)
- )
-
- self.logger.info(info)
- self.rst_report(info, annex=True)
- self.rst_report(command_line + "\n\n", frame=True, annex=True)
-
- # clear streams before add new streams
- self.tester.pktgen.clear_streams()
- # run packet generator
- streams = self.pktgen_helper.prepare_stream_from_tginput(
- tgen_input, 100, None, self.tester.pktgen
- )
- _, pps = self.tester.pktgen.measure_throughput(stream_ids=streams)
-
- Mpps = pps / 1000000.0
- queues["Mpps"][frame_size] = Mpps
- queues["pct"][frame_size] = (
- Mpps
- * 100
- / float(self.wirespeed(self.nic, frame_size, self.number_of_ports))
- )
-
- self.quit_l2fwd()
-
- # Look for transmission error in the results
- for frame_size in self.frame_sizes:
- for n in range(len(self.test_queues)):
- self.verify(
- self.test_queues[n]["Mpps"][frame_size] > 0, "No traffic detected"
- )
-
- # Prepare the results for table
- for frame_size in self.frame_sizes:
- results_row = []
- results_row.append(frame_size)
- for queue in self.test_queues:
- results_row.append(queue["Mpps"][frame_size])
- results_row.append(queue["pct"][frame_size])
-
- self.result_table_add(results_row)
-
- self.result_table_print()
-
def tear_down(self):
"""
Run after each test case.
diff --git a/tests/TestSuite_perf_l2fwd.py b/tests/TestSuite_perf_l2fwd.py
new file mode 100644
index 00000000..2cbf8899
--- /dev/null
+++ b/tests/TestSuite_perf_l2fwd.py
@@ -0,0 +1,233 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2010-2019 Intel Corporation
+#
+
+"""
+DPDK Test suite.
+Test Layer-2 Forwarding support
+"""
+import os
+import time
+
+import framework.utils as utils
+from framework.pktgen import PacketGeneratorHelper
+from framework.settings import HEADER_SIZE
+from framework.test_case import TestCase
+
+
+class TestL2fwd(TestCase):
+ def set_up_all(self):
+ """
+ Run at the start of each test suite.
+
+ L2fwd prerequisites.
+ """
+ self.frame_sizes = [64, 65, 128, 256, 512, 1024, 1280, 1518]
+
+ self.test_queues = [
+ {"queues": 1, "Mpps": {}, "pct": {}},
+ {"queues": 2, "Mpps": {}, "pct": {}},
+ {"queues": 4, "Mpps": {}, "pct": {}},
+ {"queues": 8, "Mpps": {}, "pct": {}},
+ ]
+
+ self.core_config = "1S/4C/1T"
+ self.number_of_ports = 2
+ self.headers_size = HEADER_SIZE["eth"] + HEADER_SIZE["ip"] + HEADER_SIZE["udp"]
+
+ self.dut_ports = self.dut.get_ports_performance(force_different_nic=False)
+
+ self.verify(
+ len(self.dut_ports) >= self.number_of_ports,
+ "Not enough ports for " + self.nic,
+ )
+
+ self.ports_socket = self.dut.get_numa_id(self.dut_ports[0])
+
+ # compile
+ out = self.dut.build_dpdk_apps("./examples/l2fwd")
+ self.app_l2fwd_path = self.dut.apps_name["l2fwd"]
+ self.verify("Error" not in out, "Compilation error")
+ self.verify("No such" not in out, "Compilation error")
+
+ self.table_header = ["Frame"]
+ for queue in self.test_queues:
+ self.table_header.append("%d queues Mpps" % queue["queues"])
+ self.table_header.append("% linerate")
+
+ self.result_table_create(self.table_header)
+
+ # get dts output path
+ if self.logger.log_path.startswith(os.sep):
+ self.output_path = self.logger.log_path
+ else:
+ cur_path = os.path.dirname(os.path.dirname(os.path.realpath(__file__)))
+ self.output_path = os.sep.join([cur_path, self.logger.log_path])
+ # create an instance to set stream field setting
+ self.pktgen_helper = PacketGeneratorHelper()
+
+ def set_up(self):
+ """
+ Run before each test case.
+ """
+ pass
+
+ def quit_l2fwd(self):
+ self.dut.send_expect("fg", "l2fwd ", 5)
+ self.dut.send_expect("^C", "# ", 5)
+
+ def notest_port_testing(self):
+ """
+ Check port forwarding.
+ """
+ # the cases use the first two ports
+ port_mask = utils.create_mask([self.dut_ports[0], self.dut_ports[1]])
+ eal_params = self.dut.create_eal_parameters()
+
+ self.dut.send_expect(
+ "./%s %s -- -q 8 -p %s &" % (self.app_l2fwd_path, eal_params, port_mask),
+ "L2FWD: entering main loop",
+ 60,
+ )
+
+ for i in [0, 1]:
+ tx_port = self.tester.get_local_port(self.dut_ports[i])
+ rx_port = self.tester.get_local_port(self.dut_ports[1 - i])
+
+ tx_interface = self.tester.get_interface(tx_port)
+ rx_interface = self.tester.get_interface(rx_port)
+
+ self.tester.scapy_background()
+ self.tester.scapy_append('p = sniff(iface="%s", count=1)' % rx_interface)
+ self.tester.scapy_append("number_packets=len(p)")
+ self.tester.scapy_append("RESULT = str(number_packets)")
+
+ self.tester.scapy_foreground()
+ self.tester.scapy_append(
+ 'sendp([Ether()/IP()/UDP()/("X"*46)], iface="%s")' % tx_interface
+ )
+
+ self.tester.scapy_execute()
+ number_packets = self.tester.scapy_get_result()
+ self.verify(number_packets == "1", "Failed to switch L2 frame")
+
+ self.quit_l2fwd()
+
+ def test_perf_l2fwd_performance(self):
+ """
+ Benchmark performance for frame_sizes.
+ """
+ ports = []
+ for port in range(self.number_of_ports):
+ ports.append(self.dut_ports[port])
+
+ port_mask = utils.create_mask(ports)
+ cores = self.dut.get_core_list(self.core_config, socket=self.ports_socket)
+
+ eal_params = self.dut.create_eal_parameters(cores=cores)
+ eal_param = ""
+ for i in ports:
+ eal_param += " -a %s" % self.dut.ports_info[i]["pci"]
+
+ for frame_size in self.frame_sizes:
+
+ payload_size = frame_size - self.headers_size
+
+ tgen_input = []
+ cnt = 1
+ for port in range(self.number_of_ports):
+ rx_port = self.tester.get_local_port(
+ self.dut_ports[port % self.number_of_ports]
+ )
+ tx_port = self.tester.get_local_port(
+ self.dut_ports[(port + 1) % self.number_of_ports]
+ )
+ destination_mac = self.dut.get_mac_address(
+ self.dut_ports[(port + 1) % self.number_of_ports]
+ )
+ pcap = os.sep.join(
+ [self.output_path, "l2fwd_{0}_{1}.pcap".format(port, cnt)]
+ )
+ self.tester.scapy_append(
+ 'wrpcap("%s", [Ether(dst="%s")/IP()/UDP()/("X"*%d)])'
+ % (pcap, destination_mac, payload_size)
+ )
+ tgen_input.append((tx_port, rx_port, pcap))
+ time.sleep(3)
+ self.tester.scapy_execute()
+ cnt += 1
+
+ for queues in self.test_queues:
+
+ command_line = "./%s %s %s -- -q %s -p %s &" % (
+ self.app_l2fwd_path,
+ eal_params,
+ eal_param,
+ str(queues["queues"]),
+ port_mask,
+ )
+
+ # self.dut.send_expect(command_line, "memory mapped", 60)
+ self.dut.send_expect(command_line, "L2FWD: entering main loop", 60)
+ # wait 5 second after l2fwd boot up.
+ # It is aimed to make sure trex detect link up status.
+ if self.tester.is_pktgen:
+ time.sleep(5)
+ info = (
+ "Executing l2fwd using %s queues, frame size %d and %s setup.\n"
+ % (queues["queues"], frame_size, self.core_config)
+ )
+
+ self.logger.info(info)
+ self.rst_report(info, annex=True)
+ self.rst_report(command_line + "\n\n", frame=True, annex=True)
+
+ # clear streams before add new streams
+ self.tester.pktgen.clear_streams()
+ # run packet generator
+ streams = self.pktgen_helper.prepare_stream_from_tginput(
+ tgen_input, 100, None, self.tester.pktgen
+ )
+ _, pps = self.tester.pktgen.measure_throughput(stream_ids=streams)
+
+ Mpps = pps / 1000000.0
+ queues["Mpps"][frame_size] = Mpps
+ queues["pct"][frame_size] = (
+ Mpps
+ * 100
+ / float(self.wirespeed(self.nic, frame_size, self.number_of_ports))
+ )
+
+ self.quit_l2fwd()
+
+ # Look for transmission error in the results
+ for frame_size in self.frame_sizes:
+ for n in range(len(self.test_queues)):
+ self.verify(
+ self.test_queues[n]["Mpps"][frame_size] > 0, "No traffic detected"
+ )
+
+ # Prepare the results for table
+ for frame_size in self.frame_sizes:
+ results_row = []
+ results_row.append(frame_size)
+ for queue in self.test_queues:
+ results_row.append(queue["Mpps"][frame_size])
+ results_row.append(queue["pct"][frame_size])
+
+ self.result_table_add(results_row)
+
+ self.result_table_print()
+
+ def tear_down(self):
+ """
+ Run after each test case.
+ """
+ self.dut.send_expect("fg", "l2fwd|# ", 5)
+ self.dut.send_expect("^C", "# ", 5)
+
+ def tear_down_all(self):
+ """
+ Run after each test suite.
+ """
+ pass
--
2.17.1
^ permalink raw reply [flat|nested] 9+ messages in thread
* [dts][PATCH V3 3/8] tests/tso: split performance plan and suite
2023-01-10 16:44 [dts][PATCH V3 0/8] split performance plan and suite Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 1/8] tests/efd: " Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 2/8] tests/l2fwd: " Hongbo Li
@ 2023-01-10 16:44 ` Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 4/8] tests/vxlan: " Hongbo Li
` (4 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: Hongbo Li @ 2023-01-10 16:44 UTC (permalink / raw)
To: dts; +Cc: Hongbo Li
split performance plan and suite
Signed-off-by: Hongbo Li <hongbox.li@intel.com>
---
test_plans/perf_tso_test_plan.rst | 87 +++++++++
test_plans/tso_test_plan.rst | 34 ----
tests/TestSuite_perf_tso.py | 302 ++++++++++++++++++++++++++++++
tests/TestSuite_tso.py | 113 -----------
4 files changed, 389 insertions(+), 147 deletions(-)
create mode 100644 test_plans/perf_tso_test_plan.rst
create mode 100644 tests/TestSuite_perf_tso.py
diff --git a/test_plans/perf_tso_test_plan.rst b/test_plans/perf_tso_test_plan.rst
new file mode 100644
index 00000000..026291ca
--- /dev/null
+++ b/test_plans/perf_tso_test_plan.rst
@@ -0,0 +1,87 @@
+.. SPDX-License-Identifier: BSD-3-Clause
+ Copyright(c) 2015-2017 Intel Corporation
+
+=========================================
+Transmit Segmentation Offload (TSO) Tests
+=========================================
+
+Description
+===========
+
+This document provides the plan for testing the TSO (Transmit Segmentation
+Offload, also called Large Send offload - LSO) feature of
+Intel Ethernet Controller, including Intel 82599 10GbE Ethernet Controller and
+Intel® Ethernet Converged Network Adapter XL710-QDA2. TSO enables the TCP/IP stack to
+pass to the network device a larger ULP datagram than the Maximum Transmit
+Unit Size (MTU). NIC divides the large ULP datagram to multiple segments
+according to the MTU size.
+
+
+Prerequisites
+=============
+
+Hardware:
+ Intel® Ethernet 700 Series, Intel® Ethernet 800 Series and 82599/500 Series
+
+The DUT must take one of the Ethernet controller ports connected to a port on another
+device that is controlled by the Scapy packet generator.
+
+The Ethernet interface identifier of the port that Scapy will use must be known.
+On tester, all offload feature should be disabled on tx port, and start rx port capture::
+
+ ifconfig <tx port> mtu 9000
+ ethtool -K <tx port> rx off tx off tso off gso off gro off lro off
+ ip l set <tx port> up
+ tcpdump -n -e -i <rx port> -s 0 -w /tmp/cap
+
+
+On DUT, run pmd with parameter "--enable-rx-cksum". Then enable TSO on tx port
+and checksum on rx port. The test commands is below::
+
+ #enable hw checksum on rx port
+ csum set ip hw 0
+ csum set udp hw 0
+ csum set tcp hw 0
+ csum set sctp hw 0
+ set fwd csum
+
+ # enable TSO on tx port
+ *tso set 800 1
+
+
+
+Test case: TSO performance
+==========================
+
+Set the packet stream to be sent out from packet generator before testing as
+below.
+
++-------+---------+---------+---------+----------+----------+
+| Frame | 1S/1C/1T| 1S/1C/1T| 1S/2C/1T| 1S/2C/2T | 1S/2C/2T |
+| Size | | | | | |
++-------+---------+---------+---------+----------+----------+
+| 64 | | | | | |
++-------+---------+---------+---------+----------+----------+
+| 65 | | | | | |
++-------+---------+---------+---------+----------+----------+
+| 128 | | | | | |
++-------+---------+---------+---------+----------+----------+
+| 256 | | | | | |
++-------+---------+---------+---------+----------+----------+
+| 512 | | | | | |
++-------+---------+---------+---------+----------+----------+
+| 1024 | | | | | |
++-------+---------+---------+---------+----------+----------+
+| 1280 | | | | | |
++-------+---------+---------+---------+----------+----------+
+| 1518 | | | | | |
++-------+---------+---------+---------+----------+----------+
+
+Then run the test application as below::
+
+ ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -c 0xffffffff -n 2 -- -i --rxd=512 --txd=512
+ --burst=32 --rxfreet=64 --mbcache=128 --portmask=0x3 --txpt=36 --txht=0 --txwt=0
+ --txfreet=32 --txrst=32 --enable-rx-cksum
+
+The -n command is used to select the number of memory channels. It should match the
+number of memory channels on that setup.
diff --git a/test_plans/tso_test_plan.rst b/test_plans/tso_test_plan.rst
index e5122bb9..a44d6eba 100644
--- a/test_plans/tso_test_plan.rst
+++ b/test_plans/tso_test_plan.rst
@@ -161,38 +161,4 @@ Test nvgre() in scapy::
sendp([Ether(dst="%s",src="52:00:00:00:00:00")/IP(src="192.168.1.1",dst="192.168.1.2",proto=47)/GRE(key_present=1,proto=0x6558,key=0x00001000)/Ether(dst="%s",src="52:00:00:00:00:00")/IP(src="192.168.1.1",dst="192.168.1.2")/TCP(sport=1021,dport=1021)/("X"*%s)], iface="%s")
-Test case: TSO performance
-==========================
-
-Set the packet stream to be sent out from packet generator before testing as
-below.
-
-+-------+---------+---------+---------+----------+----------+
-| Frame | 1S/1C/1T| 1S/1C/1T| 1S/2C/1T| 1S/2C/2T | 1S/2C/2T |
-| Size | | | | | |
-+-------+---------+---------+---------+----------+----------+
-| 64 | | | | | |
-+-------+---------+---------+---------+----------+----------+
-| 65 | | | | | |
-+-------+---------+---------+---------+----------+----------+
-| 128 | | | | | |
-+-------+---------+---------+---------+----------+----------+
-| 256 | | | | | |
-+-------+---------+---------+---------+----------+----------+
-| 512 | | | | | |
-+-------+---------+---------+---------+----------+----------+
-| 1024 | | | | | |
-+-------+---------+---------+---------+----------+----------+
-| 1280 | | | | | |
-+-------+---------+---------+---------+----------+----------+
-| 1518 | | | | | |
-+-------+---------+---------+---------+----------+----------+
-
-Then run the test application as below::
- ./x86_64-native-linuxapp-gcc/app/dpdk-testpmd -c 0xffffffff -n 2 -- -i --rxd=512 --txd=512
- --burst=32 --rxfreet=64 --mbcache=128 --portmask=0x3 --txpt=36 --txht=0 --txwt=0
- --txfreet=32 --txrst=32 --enable-rx-cksum
-
-The -n command is used to select the number of memory channels. It should match the
-number of memory channels on that setup.
diff --git a/tests/TestSuite_perf_tso.py b/tests/TestSuite_perf_tso.py
new file mode 100644
index 00000000..95933307
--- /dev/null
+++ b/tests/TestSuite_perf_tso.py
@@ -0,0 +1,302 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2010-2014 Intel Corporation
+#
+
+"""
+DPDK Test suite.
+
+Tests for TSO.
+
+"""
+import os
+import re
+import time
+
+import framework.utils as utils
+from framework.packet import Packet
+from framework.pktgen import PacketGeneratorHelper
+from framework.settings import HEADER_SIZE
+from framework.test_case import TestCase
+
+DEFAULT_MUT = 1500
+TSO_MTU = 9000
+
+
+class TestTSO(TestCase):
+ dut_ports = []
+
+ def set_up_all(self):
+ """
+ Run at the start of each test suite.
+ """
+ # Based on h/w type, choose how many ports to use
+ self.dut_ports = self.dut.get_ports(self.nic)
+
+ # Verify that enough ports are available
+ self.verify(len(self.dut_ports) >= 2, "Insufficient ports for testing")
+
+ # Verify that enough threads are available
+ self.portMask = utils.create_mask([self.dut_ports[0], self.dut_ports[1]])
+ self.ports_socket = self.dut.get_numa_id(self.dut_ports[0])
+ core_config = "1S/2C/1T"
+ cores = self.dut.get_core_list(core_config, socket=self.ports_socket)
+ self.verify(cores is not None, "Insufficient cores for speed testing")
+
+ self.loading_sizes = [128, 800, 801, 1700, 2500]
+
+ self.test_result = {"header": [], "data": []}
+
+ self.eal_param = self.dut.create_eal_parameters(
+ cores=core_config, socket=self.ports_socket, ports=self.dut_ports
+ )
+ self.headers_size = HEADER_SIZE["eth"] + HEADER_SIZE["ip"] + HEADER_SIZE["tcp"]
+
+ self.tester.send_expect(
+ "ifconfig %s mtu %s"
+ % (
+ self.tester.get_interface(
+ self.tester.get_local_port(self.dut_ports[0])
+ ),
+ TSO_MTU,
+ ),
+ "# ",
+ )
+ # get dts output path
+ if self.logger.log_path.startswith(os.sep):
+ self.output_path = self.logger.log_path
+ else:
+ cur_path = os.path.dirname(os.path.dirname(os.path.realpath(__file__)))
+ self.output_path = os.sep.join([cur_path, self.logger.log_path])
+ # create an instance to set stream field setting
+ self.pktgen_helper = PacketGeneratorHelper()
+ self.path = self.dut.apps_name["test-pmd"]
+
+ def set_up(self):
+ """
+ Run before each test case.
+ """
+ pass
+
+ def tcpdump_start_sniffing(self, ifaces=[]):
+ """
+ Starts tcpdump in the background to sniff the tester interface where
+ the packets are transmitted to and from the self.dut.
+ All the captured packets are going to be stored in a file for a
+ post-analysis.
+ """
+
+ for iface in ifaces:
+ command = (
+ "tcpdump -w /tmp/tcpdump_{0}.pcap -i {0} 2>tcpdump_{0}.out &"
+ ).format(iface)
+ del_cmd = ("rm -f /tmp/tcpdump_{0}.pcap").format(iface)
+ self.tester.send_expect(del_cmd, "#")
+ self.tester.send_expect(command, "#")
+
+ def tcpdump_stop_sniff(self):
+ """
+ Stops the tcpdump process running in the background.
+ """
+ self.tester.send_expect("killall tcpdump", "#")
+ time.sleep(1)
+ self.tester.send_expect('echo "Cleaning buffer"', "#")
+ time.sleep(1)
+
+ def tcpdump_command(self, command):
+ """
+ Sends a tcpdump related command and returns an integer from the output
+ """
+
+ result = self.tester.send_expect(command, "#")
+ return int(result.strip())
+
+ def number_of_packets(self, iface):
+ """
+ By reading the file generated by tcpdump it counts how many packets were
+ forwarded by the sample app and received in the self.tester. The sample app
+ will add a known MAC address for the test to look for.
+ """
+
+ command = (
+ "tcpdump -A -nn -e -v -r /tmp/tcpdump_{iface}.pcap 2>/dev/null | "
+ + 'grep -c "seq"'
+ )
+ return self.tcpdump_command(command.format(**locals()))
+
+ def tcpdump_scanner(self, scanner):
+ """
+ Execute scanner to return results
+ """
+ scanner_result = self.tester.send_expect(scanner, "#", 60)
+ fially_result = re.findall(r"length( \d+)", scanner_result)
+ return list(fially_result)
+
+ def number_of_bytes(self, iface):
+ """
+ Get the length of loading_sizes
+ """
+ scanner = 'tcpdump -vv -r /tmp/tcpdump_{iface}.pcap 2>/dev/null | grep "seq" | grep "length"'
+ return self.tcpdump_scanner(scanner.format(**locals()))
+
+ def get_chksum_value_and_verify(self, dump_pcap, Nic_list):
+ packet = Packet()
+ pkts = packet.read_pcapfile(dump_pcap, self.tester)
+ for pkt in pkts:
+ chksum_list_rx = re.findall(r"chksum\s*=\s*(0x\w+)", pkt.show(dump=True))
+ pkt["IP"].chksum = None
+ if "VXLAN" in pkt:
+ pkt["UDP"].chksum = None
+ pkt["VXLAN"]["IP"].chksum = None
+ pkt["VXLAN"]["TCP"].chksum = None
+ elif "GRE" in pkt:
+ pkt["GRE"]["IP"].chksum = None
+ pkt["GRE"]["TCP"].chksum = None
+ chksum_list_good = re.findall(r"chksum\s*=\s*(0x\w+)", pkt.show2(dump=True))
+ if self.nic in Nic_list and "VXLAN" in pkt:
+ self.verify(
+ chksum_list_rx[0] == chksum_list_good[0]
+ and chksum_list_rx[2] == chksum_list_good[2]
+ and chksum_list_rx[3] == chksum_list_good[3],
+ "The obtained chksum value is incorrect.",
+ )
+ else:
+ self.verify(
+ chksum_list_rx == chksum_list_good,
+ "The obtained chksum value is incorrect.",
+ )
+
+ def test_perf_TSO_2ports(self):
+ """
+ TSO Performance Benchmarking with 2 ports.
+ """
+
+ # set header table
+ header_row = ["Fwd Core", "Frame Size", "Throughput", "Rate"]
+ self.test_result["header"] = header_row
+ self.result_table_create(header_row)
+ self.test_result["data"] = []
+
+ test_configs = ["1S/1C/1T", "1S/1C/2T", "1S/2C/2T"]
+ core_offset = 3
+ # prepare traffic generator input
+ tgen_input = []
+
+ # run testpmd for each core config
+ for configs in test_configs:
+ cores = configs.split("/")[1]
+ thread = configs.split("/")[-1]
+ thread_num = int(int(thread[:-1]) // int(cores[:-1]))
+ _cores = str(core_offset + int(cores[0])) + "C"
+ core_config = "/".join(["1S", _cores, str(thread_num) + "T"])
+ corelist = self.dut.get_core_list(core_config, self.ports_socket)
+ core_list = corelist[(core_offset - 1) * thread_num :]
+ if "2T" in core_config:
+ core_list = core_list[1:2] + core_list[0::2] + core_list[1::2][1:]
+ _core_list = core_list[thread_num - 1 :]
+ self.eal_param = self.dut.create_eal_parameters(
+ cores=_core_list, socket=self.ports_socket, ports=self.dut_ports
+ )
+ command_line = (
+ "%s %s -- -i --rxd=512 --txd=512 --burst=32 --rxfreet=64 --mbcache=128 --portmask=%s --max-pkt-len=%s --txpt=36 --txht=0 --txwt=0 --txfreet=32 --txrst=32 "
+ % (self.path, self.eal_param, self.portMask, TSO_MTU)
+ )
+ info = "Executing PMD using cores: {0} of config {1}".format(
+ _core_list, configs
+ )
+ self.logger.info(info)
+ self.dut.send_expect(command_line, "testpmd> ", 120)
+ self.dut.send_expect("port stop all", "testpmd> ", 120)
+ for i in range(2):
+ self.dut.send_expect(
+ "csum set ip hw %d" % self.dut_ports[i], "testpmd> ", 120
+ )
+ self.dut.send_expect(
+ "csum set udp hw %d" % self.dut_ports[i], "testpmd> ", 120
+ )
+ self.dut.send_expect(
+ "csum set tcp hw %d" % self.dut_ports[i], "testpmd> ", 120
+ )
+ self.dut.send_expect(
+ "csum set sctp hw %d" % self.dut_ports[i], "testpmd> ", 120
+ )
+ self.dut.send_expect(
+ "csum set outer-ip hw %d" % self.dut_ports[i], "testpmd> ", 120
+ )
+ self.dut.send_expect(
+ "csum parse-tunnel on %d" % self.dut_ports[i], "testpmd> ", 120
+ )
+ self.dut.send_expect("tso set 800 %d" % self.dut_ports[1], "testpmd> ", 120)
+ self.dut.send_expect("set fwd csum", "testpmd> ", 120)
+ self.dut.send_expect("port start all", "testpmd> ", 120)
+ self.dut.send_expect("set promisc all off", "testpmd> ", 120)
+ self.dut.send_expect("start", "testpmd> ")
+ for loading_size in self.loading_sizes:
+ frame_size = loading_size + self.headers_size
+ wirespeed = self.wirespeed(self.nic, frame_size, 2)
+
+ # create pcap file
+ self.logger.info("Running with frame size %d " % frame_size)
+ payload_size = frame_size - self.headers_size
+ for _port in range(2):
+ mac = self.dut.get_mac_address(self.dut_ports[_port])
+
+ pcap = os.sep.join([self.output_path, "dts{0}.pcap".format(_port)])
+ self.tester.scapy_append(
+ 'wrpcap("%s", [Ether(dst="%s",src="52:00:00:00:00:01")/IP(src="192.168.1.1",dst="192.168.1.2")/TCP(sport=1021,dport=1021)/("X"*%d)])'
+ % (pcap, mac, payload_size)
+ )
+ tgen_input.append(
+ (
+ self.tester.get_local_port(self.dut_ports[_port]),
+ self.tester.get_local_port(self.dut_ports[1 - _port]),
+ "%s" % pcap,
+ )
+ )
+ self.tester.scapy_execute()
+
+ # clear streams before add new streams
+ self.tester.pktgen.clear_streams()
+ # run packet generator
+ streams = self.pktgen_helper.prepare_stream_from_tginput(
+ tgen_input, 100, None, self.tester.pktgen
+ )
+ _, pps = self.tester.pktgen.measure_throughput(stream_ids=streams)
+ self.verify(pps > 0, "No traffic detected")
+ pps /= 1000000.0
+ percentage = pps * 100 / wirespeed
+ data_row = [
+ configs,
+ frame_size,
+ "{:.3f} Mpps".format(pps),
+ "{:.3f}%".format(percentage),
+ ]
+ self.result_table_add(data_row)
+ self.dut.send_expect("stop", "testpmd> ")
+ self.dut.send_expect("quit", "# ", 30)
+ time.sleep(5)
+
+ # Print results
+ self.result_table_print()
+
+ def tear_down(self):
+ """
+ Run after each test case.
+ """
+ self.dut.send_expect("quit", "# ")
+ self.dut.kill_all()
+ time.sleep(2)
+
+ def tear_down_all(self):
+ """
+ Run after each test suite.
+ """
+ self.tester.send_expect(
+ "ifconfig %s mtu %s"
+ % (
+ self.tester.get_interface(
+ self.tester.get_local_port(self.dut_ports[0])
+ ),
+ DEFAULT_MUT,
+ ),
+ "# ",
+ )
diff --git a/tests/TestSuite_tso.py b/tests/TestSuite_tso.py
index 8cbd7a70..acefbec1 100755
--- a/tests/TestSuite_tso.py
+++ b/tests/TestSuite_tso.py
@@ -454,119 +454,6 @@ class TestTSO(TestCase):
)
self.get_chksum_value_and_verify(dump_pcap, Nic_list)
- def test_perf_TSO_2ports(self):
- """
- TSO Performance Benchmarking with 2 ports.
- """
-
- # set header table
- header_row = ["Fwd Core", "Frame Size", "Throughput", "Rate"]
- self.test_result["header"] = header_row
- self.result_table_create(header_row)
- self.test_result["data"] = []
-
- test_configs = ["1S/1C/1T", "1S/1C/2T", "1S/2C/2T"]
- core_offset = 3
- # prepare traffic generator input
- tgen_input = []
-
- # run testpmd for each core config
- for configs in test_configs:
- cores = configs.split("/")[1]
- thread = configs.split("/")[-1]
- thread_num = int(int(thread[:-1]) // int(cores[:-1]))
- _cores = str(core_offset + int(cores[0])) + "C"
- core_config = "/".join(["1S", _cores, str(thread_num) + "T"])
- corelist = self.dut.get_core_list(core_config, self.ports_socket)
- core_list = corelist[(core_offset - 1) * thread_num :]
- if "2T" in core_config:
- core_list = core_list[1:2] + core_list[0::2] + core_list[1::2][1:]
- _core_list = core_list[thread_num - 1 :]
- self.eal_param = self.dut.create_eal_parameters(
- cores=_core_list, socket=self.ports_socket, ports=self.dut_ports
- )
- command_line = (
- "%s %s -- -i --rxd=512 --txd=512 --burst=32 --rxfreet=64 --mbcache=128 --portmask=%s --max-pkt-len=%s --txpt=36 --txht=0 --txwt=0 --txfreet=32 --txrst=32 "
- % (self.path, self.eal_param, self.portMask, TSO_MTU)
- )
- info = "Executing PMD using cores: {0} of config {1}".format(
- _core_list, configs
- )
- self.logger.info(info)
- self.dut.send_expect(command_line, "testpmd> ", 120)
- self.dut.send_expect("port stop all", "testpmd> ", 120)
- for i in range(2):
- self.dut.send_expect(
- "csum set ip hw %d" % self.dut_ports[i], "testpmd> ", 120
- )
- self.dut.send_expect(
- "csum set udp hw %d" % self.dut_ports[i], "testpmd> ", 120
- )
- self.dut.send_expect(
- "csum set tcp hw %d" % self.dut_ports[i], "testpmd> ", 120
- )
- self.dut.send_expect(
- "csum set sctp hw %d" % self.dut_ports[i], "testpmd> ", 120
- )
- self.dut.send_expect(
- "csum set outer-ip hw %d" % self.dut_ports[i], "testpmd> ", 120
- )
- self.dut.send_expect(
- "csum parse-tunnel on %d" % self.dut_ports[i], "testpmd> ", 120
- )
- self.dut.send_expect("tso set 800 %d" % self.dut_ports[1], "testpmd> ", 120)
- self.dut.send_expect("set fwd csum", "testpmd> ", 120)
- self.dut.send_expect("port start all", "testpmd> ", 120)
- self.dut.send_expect("set promisc all off", "testpmd> ", 120)
- self.dut.send_expect("start", "testpmd> ")
- for loading_size in self.loading_sizes:
- frame_size = loading_size + self.headers_size
- wirespeed = self.wirespeed(self.nic, frame_size, 2)
-
- # create pcap file
- self.logger.info("Running with frame size %d " % frame_size)
- payload_size = frame_size - self.headers_size
- for _port in range(2):
- mac = self.dut.get_mac_address(self.dut_ports[_port])
-
- pcap = os.sep.join([self.output_path, "dts{0}.pcap".format(_port)])
- self.tester.scapy_append(
- 'wrpcap("%s", [Ether(dst="%s",src="52:00:00:00:00:01")/IP(src="192.168.1.1",dst="192.168.1.2")/TCP(sport=1021,dport=1021)/("X"*%d)])'
- % (pcap, mac, payload_size)
- )
- tgen_input.append(
- (
- self.tester.get_local_port(self.dut_ports[_port]),
- self.tester.get_local_port(self.dut_ports[1 - _port]),
- "%s" % pcap,
- )
- )
- self.tester.scapy_execute()
-
- # clear streams before add new streams
- self.tester.pktgen.clear_streams()
- # run packet generator
- streams = self.pktgen_helper.prepare_stream_from_tginput(
- tgen_input, 100, None, self.tester.pktgen
- )
- _, pps = self.tester.pktgen.measure_throughput(stream_ids=streams)
- self.verify(pps > 0, "No traffic detected")
- pps /= 1000000.0
- percentage = pps * 100 / wirespeed
- data_row = [
- configs,
- frame_size,
- "{:.3f} Mpps".format(pps),
- "{:.3f}%".format(percentage),
- ]
- self.result_table_add(data_row)
- self.dut.send_expect("stop", "testpmd> ")
- self.dut.send_expect("quit", "# ", 30)
- time.sleep(5)
-
- # Print results
- self.result_table_print()
-
def tear_down(self):
"""
Run after each test case.
--
2.17.1
^ permalink raw reply [flat|nested] 9+ messages in thread
* [dts][PATCH V3 4/8] tests/vxlan: split performance plan and suite
2023-01-10 16:44 [dts][PATCH V3 0/8] split performance plan and suite Hongbo Li
` (2 preceding siblings ...)
2023-01-10 16:44 ` [dts][PATCH V3 3/8] tests/tso: " Hongbo Li
@ 2023-01-10 16:44 ` Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 5/8] tests/ipfrag: " Hongbo Li
` (3 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: Hongbo Li @ 2023-01-10 16:44 UTC (permalink / raw)
To: dts; +Cc: Hongbo Li
split performance plan and suite
Signed-off-by: Hongbo Li <hongbox.li@intel.com>
---
test_plans/perf_vxlan_test_plan.rst | 85 +++
test_plans/vxlan_test_plan.rst | 60 +-
tests/TestSuite_perf_vxlan.py | 1022 +++++++++++++++++++++++++++
tests/TestSuite_vxlan.py | 213 ------
4 files changed, 1108 insertions(+), 272 deletions(-)
create mode 100644 test_plans/perf_vxlan_test_plan.rst
create mode 100644 tests/TestSuite_perf_vxlan.py
diff --git a/test_plans/perf_vxlan_test_plan.rst b/test_plans/perf_vxlan_test_plan.rst
new file mode 100644
index 00000000..a330657a
--- /dev/null
+++ b/test_plans/perf_vxlan_test_plan.rst
@@ -0,0 +1,85 @@
+.. SPDX-License-Identifier: BSD-3-Clause
+ Copyright(c) 2014-2017 Intel Corporation
+
+==========================================
+Intel® Ethernet 700/800 Series Vxlan Tests
+==========================================
+Cloud providers build virtual network overlays over existing network
+infrastructure that provide tenant isolation and scaling. Tunneling
+layers added to the packets carry the virtual networking frames over
+existing Layer 2 and IP networks. Conceptually, this is similar to
+creating virtual private networks over the Internet. Intel® Ethernet
+700 Series will process these tunneling layers by the hardware.
+
+The suit support NIC: Intel® Ethernet 700 Series, Intel® Ethernet 800 Series.
+
+This document provides test plan for Intel® Ethernet 700 Series vxlan
+packet detecting, checksum computing and filtering.
+
+Prerequisites
+=============
+1x Intel® X710 (Intel® Ethernet 700 Series) NICs (2x 40GbE full duplex
+optical ports per NIC) plugged into the available PCIe Gen3 8-lane slot.
+
+1x Intel® XL710-DA4 (Eagle Fountain) (1x 10GbE full duplex optical ports per NIC)
+plugged into the available PCIe Gen3 8-lane slot.
+
+DUT board must be two sockets system and each cpu have more than 8 lcores.
+
+Test Case: Vxlan Checksum Offload Performance Benchmarking
+==========================================================
+The throughput is measured for each of these cases for vxlan tx checksum
+offload of "all by software", "L3 offload by hardware", "L4 offload by
+hardware", "l3&l4 offload by hardware".
+
+The results are printed in the following table:
+
++----------------+--------+--------+------------+
+| Calculate Type | Queues | Mpps | % linerate |
++================+========+========+============+
+| SOFTWARE ALL | Single | | |
++----------------+--------+--------+------------+
+| HW L4 | Single | | |
++----------------+--------+--------+------------+
+| HW L3&L4 | Single | | |
++----------------+--------+--------+------------+
+| SOFTWARE ALL | Multi | | |
++----------------+--------+--------+------------+
+| HW L4 | Multi | | |
++----------------+--------+--------+------------+
+| HW L3&L4 | Multi | | |
++----------------+--------+--------+------------+
+
+Test Case: Vxlan Tunnel filter Performance Benchmarking
+=======================================================
+The throughput is measured for different Vxlan tunnel filter types.
+Queue single mean there's only one flow and forwarded to the first queue.
+Queue multi mean there are two flows and configure to different queues.
+
++--------+------------------+--------+--------+------------+
+| Packet | Filter | Queue | Mpps | % linerate |
++========+==================+========+========+============+
+| Normal | None | Single | | |
++--------+------------------+--------+--------+------------+
+| Vxlan | None | Single | | |
++--------+------------------+--------+--------+------------+
+| Vxlan | imac-ivlan | Single | | |
++--------+------------------+--------+--------+------------+
+| Vxlan | imac-ivlan-tenid | Single | | |
++--------+------------------+--------+--------+------------+
+| Vxlan | imac-tenid | Single | | |
++--------+------------------+--------+--------+------------+
+| Vxlan | imac | Single | | |
++--------+------------------+--------+--------+------------+
+| Vxlan | omac-imac-tenid | Single | | |
++--------+------------------+--------+--------+------------+
+| Vxlan | imac-ivlan | Multi | | |
++--------+------------------+--------+--------+------------+
+| Vxlan | imac-ivlan-tenid | Multi | | |
++--------+------------------+--------+--------+------------+
+| Vxlan | imac-tenid | Multi | | |
++--------+------------------+--------+--------+------------+
+| Vxlan | imac | Multi | | |
++--------+------------------+--------+--------+------------+
+| Vxlan | omac-imac-tenid | Multi | | |
++--------+------------------+--------+--------+------------+
diff --git a/test_plans/vxlan_test_plan.rst b/test_plans/vxlan_test_plan.rst
index e65b18d6..d46c550f 100644
--- a/test_plans/vxlan_test_plan.rst
+++ b/test_plans/vxlan_test_plan.rst
@@ -307,62 +307,4 @@ Add Cloud filter with invalid vlan "4097" will be failed.
Add Cloud filter with invalid vni "16777216" will be failed.
-Add Cloud filter with invalid queue id "64" will be failed.
-
-Test Case: Vxlan Checksum Offload Performance Benchmarking
-==========================================================
-The throughput is measured for each of these cases for vxlan tx checksum
-offload of "all by software", "L3 offload by hardware", "L4 offload by
-hardware", "l3&l4 offload by hardware".
-
-The results are printed in the following table:
-
-+----------------+--------+--------+------------+
-| Calculate Type | Queues | Mpps | % linerate |
-+================+========+========+============+
-| SOFTWARE ALL | Single | | |
-+----------------+--------+--------+------------+
-| HW L4 | Single | | |
-+----------------+--------+--------+------------+
-| HW L3&L4 | Single | | |
-+----------------+--------+--------+------------+
-| SOFTWARE ALL | Multi | | |
-+----------------+--------+--------+------------+
-| HW L4 | Multi | | |
-+----------------+--------+--------+------------+
-| HW L3&L4 | Multi | | |
-+----------------+--------+--------+------------+
-
-Test Case: Vxlan Tunnel filter Performance Benchmarking
-=======================================================
-The throughput is measured for different Vxlan tunnel filter types.
-Queue single mean there's only one flow and forwarded to the first queue.
-Queue multi mean there are two flows and configure to different queues.
-
-+--------+------------------+--------+--------+------------+
-| Packet | Filter | Queue | Mpps | % linerate |
-+========+==================+========+========+============+
-| Normal | None | Single | | |
-+--------+------------------+--------+--------+------------+
-| Vxlan | None | Single | | |
-+--------+------------------+--------+--------+------------+
-| Vxlan | imac-ivlan | Single | | |
-+--------+------------------+--------+--------+------------+
-| Vxlan | imac-ivlan-tenid | Single | | |
-+--------+------------------+--------+--------+------------+
-| Vxlan | imac-tenid | Single | | |
-+--------+------------------+--------+--------+------------+
-| Vxlan | imac | Single | | |
-+--------+------------------+--------+--------+------------+
-| Vxlan | omac-imac-tenid | Single | | |
-+--------+------------------+--------+--------+------------+
-| Vxlan | imac-ivlan | Multi | | |
-+--------+------------------+--------+--------+------------+
-| Vxlan | imac-ivlan-tenid | Multi | | |
-+--------+------------------+--------+--------+------------+
-| Vxlan | imac-tenid | Multi | | |
-+--------+------------------+--------+--------+------------+
-| Vxlan | imac | Multi | | |
-+--------+------------------+--------+--------+------------+
-| Vxlan | omac-imac-tenid | Multi | | |
-+--------+------------------+--------+--------+------------+
+Add Cloud filter with invalid queue id "64" will be failed.
\ No newline at end of file
diff --git a/tests/TestSuite_perf_vxlan.py b/tests/TestSuite_perf_vxlan.py
new file mode 100644
index 00000000..69e5603e
--- /dev/null
+++ b/tests/TestSuite_perf_vxlan.py
@@ -0,0 +1,1022 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2020 Intel Corporation
+#
+
+"""
+DPDK Test suite.
+
+Test VXLAN behaviour in DPDK.
+
+"""
+
+import os
+import re
+import string
+import time
+from random import randint
+
+from scapy.config import conf
+from scapy.layers.inet import IP, TCP, UDP, Ether
+from scapy.layers.inet6 import IPv6
+from scapy.layers.l2 import Dot1Q
+from scapy.layers.sctp import SCTP, SCTPChunkData
+from scapy.layers.vxlan import VXLAN
+from scapy.route import *
+from scapy.sendrecv import sniff
+from scapy.utils import rdpcap, wrpcap
+
+import framework.packet as packet
+import framework.utils as utils
+from framework.packet import IncreaseIP, IncreaseIPv6
+from framework.pktgen import PacketGeneratorHelper
+from framework.pmd_output import PmdOutput
+from framework.settings import FOLDERS, HEADER_SIZE
+from framework.test_case import TestCase
+
+#
+#
+# Test class.
+#
+
+VXLAN_PORT = 4789
+PACKET_LEN = 128
+MAX_TXQ_RXQ = 4
+BIDIRECT = True
+
+
+class VxlanTestConfig(object):
+
+ """
+ Module for config/create/transmit vxlan packet
+ """
+
+ def __init__(self, test_case, **kwargs):
+ self.test_case = test_case
+ self.init()
+ for name in kwargs:
+ setattr(self, name, kwargs[name])
+ self.pkt_obj = packet.Packet()
+
+ def init(self):
+ self.packets_config()
+
+ def packets_config(self):
+ """
+ Default vxlan packet format
+ """
+ self.pcap_file = packet.TMP_PATH + "vxlan.pcap"
+ self.capture_file = packet.TMP_PATH + "vxlan_capture.pcap"
+ self.outer_mac_src = "00:00:10:00:00:00"
+ self.outer_mac_dst = "11:22:33:44:55:66"
+ self.outer_vlan = "N/A"
+ self.outer_ip_src = "192.168.1.1"
+ self.outer_ip_dst = "192.168.1.2"
+ self.outer_ip_invalid = 0
+ self.outer_ip6_src = "N/A"
+ self.outer_ip6_dst = "N/A"
+ self.outer_ip6_invalid = 0
+ self.outer_udp_src = 63
+ self.outer_udp_dst = VXLAN_PORT
+ self.outer_udp_invalid = 0
+ self.vni = 1
+ self.inner_mac_src = "00:00:20:00:00:00"
+ self.inner_mac_dst = "00:00:20:00:00:01"
+ self.inner_vlan = "N/A"
+ self.inner_ip_src = "192.168.2.1"
+ self.inner_ip_dst = "192.168.2.2"
+ self.inner_ip_invalid = 0
+ self.inner_ip6_src = "N/A"
+ self.inner_ip6_dst = "N/A"
+ self.inner_ip6_invalid = 0
+ self.payload_size = 18
+ self.inner_l4_type = "UDP"
+ self.inner_l4_invalid = 0
+
+ def packet_type(self):
+ """
+ Return vxlan packet type
+ """
+ if self.outer_udp_dst != VXLAN_PORT:
+ if self.outer_ip6_src != "N/A":
+ return "L3_IPV6_EXT_UNKNOWN"
+ else:
+ return "L3_IPV4_EXT_UNKNOWN"
+ else:
+ if self.inner_ip6_src != "N/A":
+ return "L3_IPV6_EXT_UNKNOWN"
+ else:
+ return "L3_IPV4_EXT_UNKNOWN"
+
+ def create_pcap(self):
+ """
+ Create pcap file and copy it to tester if configured
+ Return scapy packet object for later usage
+ """
+ if self.inner_l4_type == "SCTP":
+ self.inner_payload = SCTPChunkData(data="X" * 16)
+ else:
+ self.inner_payload = "X" * self.payload_size
+
+ if self.inner_l4_type == "TCP":
+ l4_pro = TCP()
+ elif self.inner_l4_type == "SCTP":
+ l4_pro = SCTP()
+ else:
+ l4_pro = UDP()
+
+ if self.inner_ip6_src != "N/A":
+ inner_l3 = IPv6()
+ else:
+ inner_l3 = IP()
+
+ if self.inner_vlan != "N/A":
+ inner = Ether() / Dot1Q() / inner_l3 / l4_pro / self.inner_payload
+ inner[Dot1Q].vlan = self.inner_vlan
+ else:
+ inner = Ether() / inner_l3 / l4_pro / self.inner_payload
+
+ if self.inner_ip6_src != "N/A":
+ inner[inner_l3.name].src = self.inner_ip6_src
+ inner[inner_l3.name].dst = self.inner_ip6_dst
+ else:
+ inner[inner_l3.name].src = self.inner_ip_src
+ inner[inner_l3.name].dst = self.inner_ip_dst
+
+ if self.inner_ip_invalid == 1:
+ inner[inner_l3.name].chksum = 0
+
+ # when udp checksum is 0, will skip checksum
+ if self.inner_l4_invalid == 1:
+ if self.inner_l4_type == "SCTP":
+ inner[SCTP].chksum = 0
+ else:
+ inner[self.inner_l4_type].chksum = 1
+
+ inner[Ether].src = self.inner_mac_src
+ inner[Ether].dst = self.inner_mac_dst
+
+ if self.outer_ip6_src != "N/A":
+ outer_l3 = IPv6()
+ else:
+ outer_l3 = IP()
+
+ if self.outer_vlan != "N/A":
+ outer = Ether() / Dot1Q() / outer_l3 / UDP()
+ outer[Dot1Q].vlan = self.outer_vlan
+ else:
+ outer = Ether() / outer_l3 / UDP()
+
+ outer[Ether].src = self.outer_mac_src
+ outer[Ether].dst = self.outer_mac_dst
+
+ if self.outer_ip6_src != "N/A":
+ outer[outer_l3.name].src = self.outer_ip6_src
+ outer[outer_l3.name].dst = self.outer_ip6_dst
+ else:
+ outer[outer_l3.name].src = self.outer_ip_src
+ outer[outer_l3.name].dst = self.outer_ip_dst
+
+ outer[UDP].sport = self.outer_udp_src
+ outer[UDP].dport = self.outer_udp_dst
+
+ if self.outer_ip_invalid == 1:
+ outer[outer_l3.name].chksum = 0
+ # when udp checksum is 0, will skip checksum
+ if self.outer_udp_invalid == 1:
+ outer[UDP].chksum = 1
+
+ if self.outer_udp_dst == VXLAN_PORT:
+ self.pkt = outer / VXLAN(vni=self.vni) / inner
+ else:
+ self.pkt = outer / ("X" * self.payload_size)
+
+ wrpcap(self.pcap_file, self.pkt)
+
+ return self.pkt
+
+ def get_chksums(self, pkt=None):
+ """
+ get chksum values of Outer and Inner packet L3&L4
+ Skip outer udp for it will be calculated by software
+ """
+ chk_sums = {}
+ if pkt is None:
+ pkt = rdpcap(self.pcap_file)
+ else:
+ pkt = pkt.pktgen.pkt
+
+ time.sleep(1)
+ if pkt[0].guess_payload_class(pkt[0]).name == "802.1Q":
+ payload = pkt[0][Dot1Q]
+ else:
+ payload = pkt[0]
+
+ if payload.guess_payload_class(payload).name == "IP":
+ chk_sums["outer_ip"] = hex(payload[IP].chksum)
+
+ if pkt[0].haslayer("VXLAN") == 1:
+ inner = pkt[0]["VXLAN"]
+ if inner.haslayer(IP) == 1:
+ chk_sums["inner_ip"] = hex(inner[IP].chksum)
+ if inner[IP].proto == 6:
+ chk_sums["inner_tcp"] = hex(inner[TCP].chksum)
+ if inner[IP].proto == 17:
+ chk_sums["inner_udp"] = hex(inner[UDP].chksum)
+ if inner[IP].proto == 132:
+ chk_sums["inner_sctp"] = hex(inner[SCTP].chksum)
+ elif inner.haslayer(IPv6) == 1:
+ if inner[IPv6].nh == 6:
+ chk_sums["inner_tcp"] = hex(inner[TCP].chksum)
+ if inner[IPv6].nh == 17:
+ chk_sums["inner_udp"] = hex(inner[UDP].chksum)
+ # scapy can not get sctp checksum, so extracted manually
+ if inner[IPv6].nh == 59:
+ load = str(inner[IPv6].payload)
+ chk_sums["inner_sctp"] = hex(
+ (ord(load[8]) << 24)
+ | (ord(load[9]) << 16)
+ | (ord(load[10]) << 8)
+ | (ord(load[11]))
+ )
+
+ return chk_sums
+
+ def send_pcap(self, iface=""):
+ """
+ Send vxlan pcap file by iface
+ """
+ del self.pkt_obj.pktgen.pkts[:]
+ self.pkt_obj.pktgen.assign_pkt(self.pkt)
+ self.pkt_obj.pktgen.update_pkts()
+ self.pkt_obj.send_pkt(crb=self.test_case.tester, tx_port=iface)
+
+ def pcap_len(self):
+ """
+ Return length of pcap packet, will plus 4 bytes crc
+ """
+ # add four bytes crc
+ return len(self.pkt) + 4
+
+
+class TestVxlan(TestCase):
+ def set_up_all(self):
+ """
+ vxlan Prerequisites
+ """
+ # this feature only enable in Intel® Ethernet 700 Series now
+ if self.nic in [
+ "I40E_10G-SFP_XL710",
+ "I40E_40G-QSFP_A",
+ "I40E_40G-QSFP_B",
+ "I40E_25G-25G_SFP28",
+ "I40E_10G-SFP_X722",
+ "I40E_10G-10G_BASE_T_X722",
+ "I40E_10G-10G_BASE_T_BC",
+ ]:
+ self.compile_switch = "CONFIG_RTE_LIBRTE_I40E_INC_VECTOR"
+ elif self.nic in ["IXGBE_10G-X550T", "IXGBE_10G-X550EM_X_10G_T"]:
+ self.compile_switch = "CONFIG_RTE_IXGBE_INC_VECTOR"
+ elif self.nic in [
+ "ICE_25G-E810C_SFP",
+ "ICE_100G-E810C_QSFP",
+ "ICE_25G-E823C_QSFP",
+ ]:
+ print("Intel® Ethernet 700 Series support default none VECTOR")
+ else:
+ self.verify(False, "%s not support this vxlan" % self.nic)
+ # Based on h/w type, choose how many ports to use
+ ports = self.dut.get_ports()
+
+ # Verify that enough ports are available
+ self.verify(len(ports) >= 2, "Insufficient ports for testing")
+ global valports
+ valports = [_ for _ in ports if self.tester.get_local_port(_) != -1]
+
+ self.portMask = utils.create_mask(valports[:2])
+
+ # Verify that enough threads are available
+ netdev = self.dut.ports_info[ports[0]]["port"]
+ self.ports_socket = netdev.socket
+
+ # start testpmd
+ self.pmdout = PmdOutput(self.dut)
+
+ # init port config
+ self.dut_port = valports[0]
+ self.dut_port_mac = self.dut.get_mac_address(self.dut_port)
+ tester_port = self.tester.get_local_port(self.dut_port)
+ self.tester_iface = self.tester.get_interface(tester_port)
+ self.recv_port = valports[1]
+ tester_recv_port = self.tester.get_local_port(self.recv_port)
+ self.recv_iface = self.tester.get_interface(tester_recv_port)
+
+ # invalid parameter
+ self.invalid_mac = "00:00:00:00:01"
+ self.invalid_ip = "192.168.1.256"
+ self.invalid_vlan = 4097
+ self.invalid_queue = 64
+ self.path = self.dut.apps_name["test-pmd"]
+
+ # vxlan payload length for performance test
+ # inner packet not contain crc, should need add four
+ self.vxlan_payload = (
+ PACKET_LEN
+ - HEADER_SIZE["eth"]
+ - HEADER_SIZE["ip"]
+ - HEADER_SIZE["udp"]
+ - HEADER_SIZE["vxlan"]
+ - HEADER_SIZE["eth"]
+ - HEADER_SIZE["ip"]
+ - HEADER_SIZE["udp"]
+ + 4
+ )
+
+ self.cal_type = [
+ {
+ "Type": "SOFTWARE ALL",
+ "csum": [],
+ "recvqueue": "Single",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Type": "HW L4",
+ "csum": ["udp"],
+ "recvqueue": "Single",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Type": "HW L3&L4",
+ "csum": ["ip", "udp", "outer-ip"],
+ "recvqueue": "Single",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Type": "SOFTWARE ALL",
+ "csum": [],
+ "recvqueue": "Multi",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Type": "HW L4",
+ "csum": ["udp"],
+ "recvqueue": "Multi",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Type": "HW L3&L4",
+ "csum": ["ip", "udp", "outer-ip"],
+ "recvqueue": "Multi",
+ "Mpps": {},
+ "pct": {},
+ },
+ ]
+
+ self.chksum_header = ["Calculate Type"]
+ self.chksum_header.append("Queues")
+ self.chksum_header.append("Mpps")
+ self.chksum_header.append("% linerate")
+
+ # tunnel filter performance test
+ self.default_vlan = 1
+ self.tunnel_multiqueue = 2
+ self.tunnel_header = ["Packet", "Filter", "Queue", "Mpps", "% linerate"]
+ self.tunnel_perf = [
+ {
+ "Packet": "Normal",
+ "tunnel_filter": "None",
+ "recvqueue": "Single",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Packet": "VXLAN",
+ "tunnel_filter": "None",
+ "recvqueue": "Single",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Packet": "VXLAN",
+ "tunnel_filter": "imac-ivlan",
+ "recvqueue": "Single",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Packet": "VXLAN",
+ "tunnel_filter": "imac-ivlan-tenid",
+ "recvqueue": "Single",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Packet": "VXLAN",
+ "tunnel_filter": "imac-tenid",
+ "recvqueue": "Single",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Packet": "VXLAN",
+ "tunnel_filter": "imac",
+ "recvqueue": "Single",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Packet": "VXLAN",
+ "tunnel_filter": "omac-imac-tenid",
+ "recvqueue": "Single",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Packet": "VXLAN",
+ "tunnel_filter": "None",
+ "recvqueue": "Multi",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Packet": "VXLAN",
+ "tunnel_filter": "imac-ivlan",
+ "recvqueue": "Multi",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Packet": "VXLAN",
+ "tunnel_filter": "imac-ivlan-tenid",
+ "recvqueue": "Multi",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Packet": "VXLAN",
+ "tunnel_filter": "imac-tenid",
+ "recvqueue": "Multi",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Packet": "VXLAN",
+ "tunnel_filter": "imac",
+ "recvqueue": "Multi",
+ "Mpps": {},
+ "pct": {},
+ },
+ {
+ "Packet": "VXLAN",
+ "tunnel_filter": "omac-imac-tenid",
+ "recvqueue": "Multi",
+ },
+ ]
+
+ self.pktgen_helper = PacketGeneratorHelper()
+
+ def set_fields(self):
+ fields_config = {
+ "ip": {
+ "src": {"action": "random"},
+ "dst": {"action": "random"},
+ },
+ }
+ return fields_config
+
+ def suite_measure_throughput(self, tgen_input, use_vm=False):
+ vm_config = self.set_fields()
+ self.tester.pktgen.clear_streams()
+ streams = self.pktgen_helper.prepare_stream_from_tginput(
+ tgen_input, 100, vm_config if use_vm else None, self.tester.pktgen
+ )
+ result = self.tester.pktgen.measure_throughput(stream_ids=streams)
+
+ return result
+
+ def perf_tunnel_filter_set_rule(self, rule_config):
+ rule_list = {
+ # check inner mac + inner vlan filter can work
+ "imac-ivlan": f'flow create {rule_config.get("dut_port")} ingress pattern eth / '
+ f'ipv4 / udp / vxlan / eth dst is {rule_config.get("inner_mac_dst")} / '
+ f'vlan tci is {rule_config.get("inner_vlan")} / end actions pf / '
+ f'queue index {rule_config.get("queue")} / end',
+ # check inner mac + inner vlan + tunnel id filter can work
+ "imac-ivlan-tenid": f'flow create {rule_config.get("dut_port")} ingress pattern eth / '
+ f'ipv4 / udp / vxlan vni is {rule_config.get("vni")} / '
+ f'eth dst is {rule_config.get("inner_mac_dst")} / '
+ f'vlan tci is {rule_config.get("inner_vlan")} / '
+ f'end actions pf / queue index {rule_config.get("queue")} / end',
+ # check inner mac + tunnel id filter can work
+ "imac-tenid": f'flow create {rule_config.get("dut_port")} ingress pattern eth / '
+ f'ipv4 / udp / vxlan vni is {rule_config.get("vni")} / '
+ f'eth dst is {rule_config.get("inner_mac_dst")} / end actions pf / '
+ f'queue index {rule_config.get("queue")} / end',
+ # check inner mac filter can work
+ "imac": f'flow create {rule_config.get("dut_port")} ingress pattern eth / '
+ f'ipv4 / udp / vxlan / eth dst is {rule_config.get("inner_mac_dst")} / end actions pf / '
+ f'queue index {rule_config.get("queue")} / end',
+ # check outer mac + inner mac + tunnel id filter can work
+ "omac-imac-tenid": f'flow create {rule_config.get("dut_port")} ingress pattern '
+ f'eth dst is {rule_config.get("outer_mac_dst")} / '
+ f'ipv4 / udp / vxlan vni is {rule_config.get("vni")} / '
+ f'eth dst is {rule_config.get("inner_mac_dst")} / '
+ f'end actions pf / queue index {rule_config.get("queue")} / end',
+ }
+ rule = rule_list.get(rule_config.get("tun_filter"))
+ if not rule:
+ msg = "not support format"
+ self.logger.error(msg)
+ return
+ out = self.dut.send_expect(rule, "testpmd>", 3)
+ pat = "Flow rule #\d+ created"
+ self.verify(re.findall(pat, out, re.M), "Flow rule create failed")
+
+ def send_and_detect(self, **kwargs):
+ """
+ send vxlan packet and check whether testpmd detect the correct
+ packet type
+ """
+ arg_str = ""
+ for arg in kwargs:
+ arg_str += "[%s = %s]" % (arg, kwargs[arg])
+
+ # create pcap file with supplied arguments
+ self.logger.info("send vxlan pkts %s" % arg_str)
+ config = VxlanTestConfig(self, **kwargs)
+ # now cloud filter will default enable L2 mac filter, so dst mac must
+ # be same
+ config.outer_mac_dst = self.dut_port_mac
+ config.create_pcap()
+ self.dut.send_expect("start", "testpmd>", 10)
+ self.pmdout.wait_link_status_up(self.dut_port)
+ config.send_pcap(self.tester_iface)
+ # check whether detect vxlan type
+ out = self.dut.get_session_output(timeout=2)
+ print(out)
+ self.verify(config.packet_type() in out, "Vxlan Packet not detected")
+
+ def send_and_check(self, **kwargs):
+ """
+ send vxlan packet and check whether receive packet with correct
+ checksum
+ """
+ # create pcap file with supplied arguments
+ outer_ipv6 = False
+ args = {}
+ for arg in kwargs:
+ if "invalid" not in arg:
+ args[arg] = kwargs[arg]
+ if "outer_ip6" in arg:
+ outer_ipv6 = True
+
+ # if packet outer L3 is ipv6, should not enable hardware checksum
+ if outer_ipv6:
+ self.csum_set_sw("outer-ip", self.dut_port)
+ self.csum_set_sw("outer-ip", self.recv_port)
+
+ config = VxlanTestConfig(self, **args)
+ # now cloud filter will default enable L2 mac filter, so dst mac must
+ # be same
+ config.outer_mac_dst = self.dut_port_mac
+ # csum function will not auto add outer ip src address already, so update send packet src ip address
+ if config.outer_ip6_src != "N/A":
+ config.outer_ip6_src = config.outer_ip6_src
+ else:
+ config.outer_ip_src = config.outer_ip_src
+
+ # csum function will not auto add outer ip src address already, so update send packet src ip address
+ if config.outer_udp_dst == VXLAN_PORT:
+ if config.inner_ip6_src != "N/A":
+ config.inner_ip6_src = config.inner_ip6_src
+ else:
+ config.inner_ip_src = config.inner_ip_src
+
+ # extract the checksum value of vxlan packet
+ config.create_pcap()
+ chksums_ref = config.get_chksums()
+ self.logger.info("chksums_ref" + str(chksums_ref))
+
+ # log the vxlan format
+ arg_str = ""
+ for arg in kwargs:
+ arg_str += "[%s = %s]" % (arg, kwargs[arg])
+
+ self.logger.info("vxlan packet %s" % arg_str)
+
+ out = self.dut.send_expect("start", "testpmd>", 10)
+
+ # create pcap file with supplied arguments
+ config = VxlanTestConfig(self, **kwargs)
+ config.outer_mac_dst = self.dut_port_mac
+ config.create_pcap()
+
+ self.pmdout.wait_link_status_up(self.dut_port)
+ # save the capture packet into pcap format
+ inst = self.tester.tcpdump_sniff_packets(self.recv_iface)
+ config.send_pcap(self.tester_iface)
+ pkt = self.tester.load_tcpdump_sniff_packets(inst, timeout=3)
+
+ # extract the checksum offload from saved pcap file
+ chksums = config.get_chksums(pkt=pkt)
+ self.logger.info("chksums" + str(chksums))
+
+ out = self.dut.send_expect("stop", "testpmd>", 10)
+ print(out)
+
+ # verify detected l4 invalid checksum
+ if "inner_l4_invalid" in kwargs:
+ self.verify(
+ self.pmdout.get_pmd_value("Bad-l4csum:", out) == 1,
+ "Failed to count inner l4 chksum error",
+ )
+
+ # verify detected l3 invalid checksum
+ if "ip_invalid" in kwargs:
+ self.verify(
+ self.pmdout.get_pmd_value("Bad-ipcsum:", out) == self.iperr_num + 1,
+ "Failed to count inner ip chksum error",
+ )
+ self.iperr_num += 1
+
+ # verify saved pcap checksum same to expected checksum
+ for key in chksums_ref:
+ self.verify(
+ chksums[key] == chksums_ref[key],
+ "%s not matched to %s" % (key, chksums_ref[key]),
+ )
+
+ def filter_and_check(self, rule, config, queue_id):
+ """
+ send vxlan packet and check whether receive packet in assigned queue
+ """
+ # create rule
+ self.tunnel_filter_add(rule)
+
+ # send vxlan packet
+ config.create_pcap()
+ self.dut.send_expect("start", "testpmd>", 10)
+ self.pmdout.wait_link_status_up(self.dut_port)
+ config.send_pcap(self.tester_iface)
+ out = self.dut.get_session_output(timeout=2)
+ print(out)
+
+ queue = -1
+ pattern = re.compile("- Receive queue=0x(\d)")
+ m = pattern.search(out)
+ if m is not None:
+ queue = m.group(1)
+
+ # verify received in expected queue
+ self.verify(queue_id == int(queue), "invalid receive queue")
+
+ # del rule
+ args = [self.dut_port]
+ self.tunnel_filter_del(*args)
+ self.dut.send_expect("stop", "testpmd>", 10)
+
+ def config_tunnelfilter(self, dut_port, recv_port, perf_config, pcapfile):
+ pkts = []
+ config = VxlanTestConfig(self, payload_size=self.vxlan_payload - 4)
+ config.inner_vlan = self.default_vlan
+ config.outer_mac_dst = self.dut.get_mac_address(dut_port)
+ config.pcap_file = pcapfile
+
+ tun_filter = perf_config["tunnel_filter"]
+ recv_queue = perf_config["recvqueue"]
+ # there's known bug that if enable vxlan, rss will be disabled
+ if tun_filter == "None" and recv_queue == "Multi":
+ print((utils.RED("RSS and Tunel filter can't enable in the same time")))
+ else:
+ self.enable_vxlan(dut_port)
+
+ if tun_filter != "None":
+ rule_config = {
+ "dut_port": dut_port,
+ "outer_mac_dst": config.outer_mac_dst,
+ "inner_mac_dst": config.inner_mac_dst,
+ "inner_ip_dst": config.inner_ip_dst,
+ "inner_vlan": config.inner_vlan,
+ "tun_filter": tun_filter,
+ "vni": config.vni,
+ "queue": 0,
+ }
+ self.perf_tunnel_filter_set_rule(rule_config)
+
+ if perf_config["Packet"] == "Normal":
+ config.outer_udp_dst = 63
+ config.outer_mac_dst = self.dut.get_mac_address(dut_port)
+ config.payload_size = (
+ PACKET_LEN - HEADER_SIZE["eth"] - HEADER_SIZE["ip"] - HEADER_SIZE["udp"]
+ )
+
+ # add default pkt into pkt list
+ pkt = config.create_pcap()
+ pkts.append(pkt)
+
+ # add other pkts into pkt list when enable multi receive queues
+ if recv_queue == "Multi" and tun_filter != "None":
+ for queue in range(self.tunnel_multiqueue - 1):
+ if "imac" in tun_filter:
+ config.inner_mac_dst = "00:00:20:00:00:0%d" % (queue + 2)
+ if "ivlan" in tun_filter:
+ config.inner_vlan = queue + 2
+ if "tenid" in tun_filter:
+ config.vni = queue + 2
+
+ # add tunnel filter the same as pkt
+ pkt = config.create_pcap()
+ pkts.append(pkt)
+
+ rule_config = {
+ "dut_port": dut_port,
+ "outer_mac_dst": config.outer_mac_dst,
+ "inner_mac_dst": config.inner_mac_dst,
+ "inner_ip_dst": config.inner_ip_dst,
+ "inner_vlan": config.inner_vlan,
+ "tun_filter": tun_filter,
+ "vni": config.vni,
+ "queue": (queue + 1),
+ }
+ self.perf_tunnel_filter_set_rule(rule_config)
+
+ # save pkt list into pcap file
+ wrpcap(config.pcap_file, pkts)
+ self.tester.session.copy_file_to(config.pcap_file)
+
+ def combine_pcap(self, dest_pcap, src_pcap):
+ pkts = rdpcap(dest_pcap)
+ if len(pkts) != 1:
+ return
+
+ pkts_src = rdpcap(src_pcap)
+ pkts += pkts_src
+
+ wrpcap(dest_pcap, pkts)
+
+ def test_perf_vxlan_tunnelfilter_performance_2ports(self):
+ self.result_table_create(self.tunnel_header)
+ core_list = self.dut.get_core_list(
+ "1S/%dC/1T" % (self.tunnel_multiqueue * 2 + 1), socket=self.ports_socket
+ )
+
+ pmd_temp = (
+ "./%s %s -- -i --disable-rss --rxq=2 --txq=2 --nb-cores=4 --portmask=%s"
+ )
+
+ for perf_config in self.tunnel_perf:
+ tun_filter = perf_config["tunnel_filter"]
+ recv_queue = perf_config["recvqueue"]
+ print(
+ (
+ utils.GREEN(
+ "Measure tunnel performance of [%s %s %s]"
+ % (perf_config["Packet"], tun_filter, recv_queue)
+ )
+ )
+ )
+
+ if tun_filter == "None" and recv_queue == "Multi":
+ pmd_temp = (
+ "./%s %s -- -i --rss-udp --rxq=2 --txq=2 --nb-cores=4 --portmask=%s"
+ )
+
+ self.eal_para = self.dut.create_eal_parameters(cores=core_list)
+ pmd_cmd = pmd_temp % (self.path, self.eal_para, self.portMask)
+ self.dut.send_expect(pmd_cmd, "testpmd> ", 100)
+
+ # config flow
+ self.config_tunnelfilter(
+ self.dut_port, self.recv_port, perf_config, "flow1.pcap"
+ )
+ # config the flows
+ tgen_input = []
+ tgen_input.append(
+ (
+ self.tester.get_local_port(self.dut_port),
+ self.tester.get_local_port(self.recv_port),
+ "flow1.pcap",
+ )
+ )
+
+ if BIDIRECT:
+ self.config_tunnelfilter(
+ self.recv_port, self.dut_port, perf_config, "flow2.pcap"
+ )
+ tgen_input.append(
+ (
+ self.tester.get_local_port(self.recv_port),
+ self.tester.get_local_port(self.dut_port),
+ "flow2.pcap",
+ )
+ )
+
+ self.dut.send_expect("set fwd io", "testpmd>", 10)
+ self.dut.send_expect("start", "testpmd>", 10)
+ self.pmdout.wait_link_status_up(self.dut_port)
+ if BIDIRECT:
+ wirespeed = self.wirespeed(self.nic, PACKET_LEN, 2)
+ else:
+ wirespeed = self.wirespeed(self.nic, PACKET_LEN, 1)
+
+ # run traffic generator
+ use_vm = True if recv_queue == "Multi" and tun_filter == "None" else False
+ _, pps = self.suite_measure_throughput(tgen_input, use_vm=use_vm)
+
+ pps /= 1000000.0
+ perf_config["Mpps"] = pps
+ perf_config["pct"] = pps * 100 / wirespeed
+
+ out = self.dut.send_expect("stop", "testpmd>", 10)
+ self.dut.send_expect("quit", "# ", 10)
+
+ # verify every queue work fine
+ check_queue = 0
+ if recv_queue == "Multi":
+ for queue in range(check_queue):
+ self.verify(
+ "Queue= %d -> TX Port" % (queue) in out,
+ "Queue %d no traffic" % queue,
+ )
+
+ table_row = [
+ perf_config["Packet"],
+ tun_filter,
+ recv_queue,
+ perf_config["Mpps"],
+ perf_config["pct"],
+ ]
+
+ self.result_table_add(table_row)
+
+ self.result_table_print()
+
+ def test_perf_vxlan_checksum_performance_2ports(self):
+ self.result_table_create(self.chksum_header)
+ vxlan = VxlanTestConfig(self, payload_size=self.vxlan_payload)
+ vxlan.outer_mac_dst = self.dut.get_mac_address(self.dut_port)
+ vxlan.pcap_file = "vxlan1.pcap"
+ vxlan.inner_mac_dst = "00:00:20:00:00:01"
+ vxlan.create_pcap()
+
+ vxlan_queue = VxlanTestConfig(self, payload_size=self.vxlan_payload)
+ vxlan_queue.outer_mac_dst = self.dut.get_mac_address(self.dut_port)
+ vxlan_queue.pcap_file = "vxlan1_1.pcap"
+ vxlan_queue.inner_mac_dst = "00:00:20:00:00:02"
+ vxlan_queue.create_pcap()
+
+ # socket/core/thread
+ core_list = self.dut.get_core_list(
+ "1S/%dC/1T" % (self.tunnel_multiqueue * 2 + 1), socket=self.ports_socket
+ )
+ core_mask = utils.create_mask(core_list)
+
+ self.dut_ports = self.dut.get_ports_performance(force_different_nic=False)
+ tx_port = self.tester.get_local_port(self.dut_ports[0])
+ rx_port = self.tester.get_local_port(self.dut_ports[1])
+
+ for cal in self.cal_type:
+ recv_queue = cal["recvqueue"]
+ print(
+ (
+ utils.GREEN(
+ "Measure checksum performance of [%s %s %s]"
+ % (cal["Type"], recv_queue, cal["csum"])
+ )
+ )
+ )
+
+ # configure flows
+ tgen_input = []
+ if recv_queue == "Multi":
+ tgen_input.append((tx_port, rx_port, "vxlan1.pcap"))
+ tgen_input.append((tx_port, rx_port, "vxlan1_1.pcap"))
+ else:
+ tgen_input.append((tx_port, rx_port, "vxlan1.pcap"))
+
+ # multi queue and signle queue commands
+ if recv_queue == "Multi":
+ pmd_temp = "./%s %s -- -i --disable-rss --rxq=2 --txq=2 --nb-cores=4 --portmask=%s"
+ else:
+ pmd_temp = "./%s %s -- -i --nb-cores=2 --portmask=%s"
+
+ self.eal_para = self.dut.create_eal_parameters(cores=core_list)
+ pmd_cmd = pmd_temp % (self.path, self.eal_para, self.portMask)
+
+ self.dut.send_expect(pmd_cmd, "testpmd> ", 100)
+ self.dut.send_expect("set fwd csum", "testpmd>", 10)
+ self.enable_vxlan(self.dut_port)
+ self.enable_vxlan(self.recv_port)
+ self.pmdout.wait_link_status_up(self.dut_port)
+
+ # redirect flow to another queue by tunnel filter
+ rule_config = {
+ "dut_port": self.dut_port,
+ "outer_mac_dst": vxlan.outer_mac_dst,
+ "inner_mac_dst": vxlan.inner_mac_dst,
+ "inner_ip_dst": vxlan.inner_ip_dst,
+ "inner_vlan": 0,
+ "tun_filter": "imac",
+ "vni": vxlan.vni,
+ "queue": 0,
+ }
+ self.perf_tunnel_filter_set_rule(rule_config)
+
+ if recv_queue == "Multi":
+ rule_config = {
+ "dut_port": self.dut_port,
+ "outer_mac_dst": vxlan_queue.outer_mac_dst,
+ "inner_mac_dst": vxlan_queue.inner_mac_dst,
+ "inner_ip_dst": vxlan_queue.inner_ip_dst,
+ "inner_vlan": 0,
+ "tun_filter": "imac",
+ "vni": vxlan.vni,
+ "queue": 1,
+ }
+ self.perf_tunnel_filter_set_rule(rule_config)
+
+ for pro in cal["csum"]:
+ self.csum_set_type(pro, self.dut_port)
+ self.csum_set_type(pro, self.recv_port)
+
+ self.dut.send_expect("start", "testpmd>", 10)
+
+ wirespeed = self.wirespeed(self.nic, PACKET_LEN, 1)
+
+ # run traffic generator
+ _, pps = self.suite_measure_throughput(tgen_input)
+
+ pps /= 1000000.0
+ cal["Mpps"] = pps
+ cal["pct"] = pps * 100 / wirespeed
+
+ out = self.dut.send_expect("stop", "testpmd>", 10)
+ self.dut.send_expect("quit", "# ", 10)
+
+ # verify every queue work fine
+ check_queue = 1
+ if recv_queue == "Multi":
+ for queue in range(check_queue):
+ self.verify(
+ "Queue= %d -> TX Port" % (queue) in out,
+ "Queue %d no traffic" % queue,
+ )
+
+ table_row = [cal["Type"], recv_queue, cal["Mpps"], cal["pct"]]
+ self.result_table_add(table_row)
+
+ self.result_table_print()
+
+ def enable_vxlan(self, port):
+ self.dut.send_expect(
+ "rx_vxlan_port add %d %d" % (VXLAN_PORT, port), "testpmd>", 10
+ )
+
+ def csum_set_type(self, proto, port):
+ self.dut.send_expect("port stop all", "testpmd>")
+ out = self.dut.send_expect("csum set %s hw %d" % (proto, port), "testpmd>", 10)
+ self.dut.send_expect("port start all", "testpmd>")
+ self.verify("Bad arguments" not in out, "Failed to set vxlan csum")
+ self.verify("error" not in out, "Failed to set vxlan csum")
+
+ def csum_set_sw(self, proto, port):
+ self.dut.send_expect("port stop all", "testpmd>")
+ out = self.dut.send_expect("csum set %s sw %d" % (proto, port), "testpmd>", 10)
+ self.dut.send_expect("port start all", "testpmd>")
+ self.verify("Bad arguments" not in out, "Failed to set vxlan csum")
+ self.verify("error" not in out, "Failed to set vxlan csum")
+
+ def tunnel_filter_add(self, rule):
+ out = self.dut.send_expect(rule, "testpmd>", 3)
+ self.verify("Flow rule #0 created" in out, "Flow rule create failed")
+ return out
+
+ def tunnel_filter_add_nocheck(self, rule):
+ out = self.dut.send_expect(rule, "testpmd>", 3)
+ return out
+
+ def tunnel_filter_del(self, *args):
+ out = self.dut.send_expect("flow flush 0", "testpmd>", 10)
+ return out
+
+ def set_up(self):
+ """
+ Run before each test case.
+ """
+ pass
+
+ def tear_down(self):
+ """
+ Run after each test case.
+ """
+ self.dut.kill_all()
+
+ def tear_down_all(self):
+ """
+ Run after each test suite.
+ """
+ pass
diff --git a/tests/TestSuite_vxlan.py b/tests/TestSuite_vxlan.py
index 1bf12743..7c309a01 100644
--- a/tests/TestSuite_vxlan.py
+++ b/tests/TestSuite_vxlan.py
@@ -1175,219 +1175,6 @@ class TestVxlan(TestCase):
wrpcap(dest_pcap, pkts)
- def test_perf_vxlan_tunnelfilter_performance_2ports(self):
- self.result_table_create(self.tunnel_header)
- core_list = self.dut.get_core_list(
- "1S/%dC/1T" % (self.tunnel_multiqueue * 2 + 1), socket=self.ports_socket
- )
-
- pmd_temp = (
- "./%s %s -- -i --disable-rss --rxq=2 --txq=2 --nb-cores=4 --portmask=%s"
- )
-
- for perf_config in self.tunnel_perf:
- tun_filter = perf_config["tunnel_filter"]
- recv_queue = perf_config["recvqueue"]
- print(
- (
- utils.GREEN(
- "Measure tunnel performance of [%s %s %s]"
- % (perf_config["Packet"], tun_filter, recv_queue)
- )
- )
- )
-
- if tun_filter == "None" and recv_queue == "Multi":
- pmd_temp = (
- "./%s %s -- -i --rss-udp --rxq=2 --txq=2 --nb-cores=4 --portmask=%s"
- )
-
- self.eal_para = self.dut.create_eal_parameters(cores=core_list)
- pmd_cmd = pmd_temp % (self.path, self.eal_para, self.portMask)
- self.dut.send_expect(pmd_cmd, "testpmd> ", 100)
-
- # config flow
- self.config_tunnelfilter(
- self.dut_port, self.recv_port, perf_config, "flow1.pcap"
- )
- # config the flows
- tgen_input = []
- tgen_input.append(
- (
- self.tester.get_local_port(self.dut_port),
- self.tester.get_local_port(self.recv_port),
- "flow1.pcap",
- )
- )
-
- if BIDIRECT:
- self.config_tunnelfilter(
- self.recv_port, self.dut_port, perf_config, "flow2.pcap"
- )
- tgen_input.append(
- (
- self.tester.get_local_port(self.recv_port),
- self.tester.get_local_port(self.dut_port),
- "flow2.pcap",
- )
- )
-
- self.dut.send_expect("set fwd io", "testpmd>", 10)
- self.dut.send_expect("start", "testpmd>", 10)
- self.pmdout.wait_link_status_up(self.dut_port)
- if BIDIRECT:
- wirespeed = self.wirespeed(self.nic, PACKET_LEN, 2)
- else:
- wirespeed = self.wirespeed(self.nic, PACKET_LEN, 1)
-
- # run traffic generator
- use_vm = True if recv_queue == "Multi" and tun_filter == "None" else False
- _, pps = self.suite_measure_throughput(tgen_input, use_vm=use_vm)
-
- pps /= 1000000.0
- perf_config["Mpps"] = pps
- perf_config["pct"] = pps * 100 / wirespeed
-
- out = self.dut.send_expect("stop", "testpmd>", 10)
- self.dut.send_expect("quit", "# ", 10)
-
- # verify every queue work fine
- check_queue = 0
- if recv_queue == "Multi":
- for queue in range(check_queue):
- self.verify(
- "Queue= %d -> TX Port" % (queue) in out,
- "Queue %d no traffic" % queue,
- )
-
- table_row = [
- perf_config["Packet"],
- tun_filter,
- recv_queue,
- perf_config["Mpps"],
- perf_config["pct"],
- ]
-
- self.result_table_add(table_row)
-
- self.result_table_print()
-
- def test_perf_vxlan_checksum_performance_2ports(self):
- self.result_table_create(self.chksum_header)
- vxlan = VxlanTestConfig(self, payload_size=self.vxlan_payload)
- vxlan.outer_mac_dst = self.dut.get_mac_address(self.dut_port)
- vxlan.pcap_file = "vxlan1.pcap"
- vxlan.inner_mac_dst = "00:00:20:00:00:01"
- vxlan.create_pcap()
-
- vxlan_queue = VxlanTestConfig(self, payload_size=self.vxlan_payload)
- vxlan_queue.outer_mac_dst = self.dut.get_mac_address(self.dut_port)
- vxlan_queue.pcap_file = "vxlan1_1.pcap"
- vxlan_queue.inner_mac_dst = "00:00:20:00:00:02"
- vxlan_queue.create_pcap()
-
- # socket/core/thread
- core_list = self.dut.get_core_list(
- "1S/%dC/1T" % (self.tunnel_multiqueue * 2 + 1), socket=self.ports_socket
- )
- core_mask = utils.create_mask(core_list)
-
- self.dut_ports = self.dut.get_ports_performance(force_different_nic=False)
- tx_port = self.tester.get_local_port(self.dut_ports[0])
- rx_port = self.tester.get_local_port(self.dut_ports[1])
-
- for cal in self.cal_type:
- recv_queue = cal["recvqueue"]
- print(
- (
- utils.GREEN(
- "Measure checksum performance of [%s %s %s]"
- % (cal["Type"], recv_queue, cal["csum"])
- )
- )
- )
-
- # configure flows
- tgen_input = []
- if recv_queue == "Multi":
- tgen_input.append((tx_port, rx_port, "vxlan1.pcap"))
- tgen_input.append((tx_port, rx_port, "vxlan1_1.pcap"))
- else:
- tgen_input.append((tx_port, rx_port, "vxlan1.pcap"))
-
- # multi queue and signle queue commands
- if recv_queue == "Multi":
- pmd_temp = "./%s %s -- -i --disable-rss --rxq=2 --txq=2 --nb-cores=4 --portmask=%s"
- else:
- pmd_temp = "./%s %s -- -i --nb-cores=2 --portmask=%s"
-
- self.eal_para = self.dut.create_eal_parameters(cores=core_list)
- pmd_cmd = pmd_temp % (self.path, self.eal_para, self.portMask)
-
- self.dut.send_expect(pmd_cmd, "testpmd> ", 100)
- self.dut.send_expect("set fwd csum", "testpmd>", 10)
- self.enable_vxlan(self.dut_port)
- self.enable_vxlan(self.recv_port)
- self.pmdout.wait_link_status_up(self.dut_port)
-
- # redirect flow to another queue by tunnel filter
- rule_config = {
- "dut_port": self.dut_port,
- "outer_mac_dst": vxlan.outer_mac_dst,
- "inner_mac_dst": vxlan.inner_mac_dst,
- "inner_ip_dst": vxlan.inner_ip_dst,
- "inner_vlan": 0,
- "tun_filter": "imac",
- "vni": vxlan.vni,
- "queue": 0,
- }
- self.perf_tunnel_filter_set_rule(rule_config)
-
- if recv_queue == "Multi":
- rule_config = {
- "dut_port": self.dut_port,
- "outer_mac_dst": vxlan_queue.outer_mac_dst,
- "inner_mac_dst": vxlan_queue.inner_mac_dst,
- "inner_ip_dst": vxlan_queue.inner_ip_dst,
- "inner_vlan": 0,
- "tun_filter": "imac",
- "vni": vxlan.vni,
- "queue": 1,
- }
- self.perf_tunnel_filter_set_rule(rule_config)
-
- for pro in cal["csum"]:
- self.csum_set_type(pro, self.dut_port)
- self.csum_set_type(pro, self.recv_port)
-
- self.dut.send_expect("start", "testpmd>", 10)
-
- wirespeed = self.wirespeed(self.nic, PACKET_LEN, 1)
-
- # run traffic generator
- _, pps = self.suite_measure_throughput(tgen_input)
-
- pps /= 1000000.0
- cal["Mpps"] = pps
- cal["pct"] = pps * 100 / wirespeed
-
- out = self.dut.send_expect("stop", "testpmd>", 10)
- self.dut.send_expect("quit", "# ", 10)
-
- # verify every queue work fine
- check_queue = 1
- if recv_queue == "Multi":
- for queue in range(check_queue):
- self.verify(
- "Queue= %d -> TX Port" % (queue) in out,
- "Queue %d no traffic" % queue,
- )
-
- table_row = [cal["Type"], recv_queue, cal["Mpps"], cal["pct"]]
- self.result_table_add(table_row)
-
- self.result_table_print()
-
def enable_vxlan(self, port):
self.dut.send_expect(
"rx_vxlan_port add %d %d" % (VXLAN_PORT, port), "testpmd>", 10
--
2.17.1
^ permalink raw reply [flat|nested] 9+ messages in thread
* [dts][PATCH V3 5/8] tests/ipfrag: split performance plan and suite
2023-01-10 16:44 [dts][PATCH V3 0/8] split performance plan and suite Hongbo Li
` (3 preceding siblings ...)
2023-01-10 16:44 ` [dts][PATCH V3 4/8] tests/vxlan: " Hongbo Li
@ 2023-01-10 16:44 ` Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 6/8] tests/multiprocess: " Hongbo Li
` (2 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: Hongbo Li @ 2023-01-10 16:44 UTC (permalink / raw)
To: dts; +Cc: Hongbo Li
split performance plan and suite
Signed-off-by: Hongbo Li <hongbox.li@intel.com>
---
test_plans/ipfrag_test_plan.rst | 33 +--
test_plans/perf_ipfrag_test_plan.rst | 141 ++++++++++
tests/TestSuite_ipfrag.py | 21 --
tests/TestSuite_perf_ipfrag.py | 386 +++++++++++++++++++++++++++
4 files changed, 528 insertions(+), 53 deletions(-)
create mode 100644 test_plans/perf_ipfrag_test_plan.rst
create mode 100644 tests/TestSuite_perf_ipfrag.py
diff --git a/test_plans/ipfrag_test_plan.rst b/test_plans/ipfrag_test_plan.rst
index e30a5961..17ea5c3c 100644
--- a/test_plans/ipfrag_test_plan.rst
+++ b/test_plans/ipfrag_test_plan.rst
@@ -129,35 +129,4 @@ For each of them check that:
#. Check number of output packets.
#. Check header of each output packet: length, ID, fragment offset, flags.
-#. Check payload: size and contents as expected, not corrupted.
-
-
-
-Test Case 4: Throughput test
-============================
-
-The test report should provide the throughput rate measurements (in mpps and % of the line rate for 2x NIC ports)
-for the following input frame sizes: 64 bytes, 1518 bytes, 1519 bytes, 2K, 9k.
-
-The following configurations should be tested:
-
-|
-
-+----------+-------------------------+----------------------+
-|# of ports| Socket/Core/HyperThread|Total # of sw threads |
-+----------+-------------------------+----------------------+
-| 2 | 1S/1C/1T | 1 |
-+----------+-------------------------+----------------------+
-| 2 | 1S/1C/2T | 2 |
-+----------+-------------------------+----------------------+
-| 2 | 1S/2C/1T | 2 |
-+----------+-------------------------+----------------------+
-| 2 | 2S/1C/1T | 2 |
-+----------+-------------------------+----------------------+
-
-|
-
-Command line::
-
- ./x86_64-native-linuxapp-gcc/examples/dpdk-ip_fragmentation -c <LCOREMASK> -n 4 -- [-P] -p PORTMASK
- -q <NUM_OF_PORTS_PER_THREAD>
+#. Check payload: size and contents as expected, not corrupted.
\ No newline at end of file
diff --git a/test_plans/perf_ipfrag_test_plan.rst b/test_plans/perf_ipfrag_test_plan.rst
new file mode 100644
index 00000000..96dca62d
--- /dev/null
+++ b/test_plans/perf_ipfrag_test_plan.rst
@@ -0,0 +1,141 @@
+.. SPDX-License-Identifier: BSD-3-Clause
+ Copyright(c) 2011-2017 Intel Corporation
+
+======================
+IP fragmentation Tests
+======================
+
+The IP fragmentation results are produced using ''ip_fragmentation'' application.
+The test application should run with both IPv4 and IPv6 fragmentation.
+
+The suit support NIC: Intel® Ethernet 700 Series, Intel® Ethernet 800 Series, 82599 and igc driver NIC.
+
+Prerequisites
+=============
+
+1. Hardware requirements:
+
+ - For each CPU socket, each memory channel should be populated with at least 1x DIMM
+ - Board is populated with at least 2x 1GbE or 10GbE ports. Special PCIe restrictions may
+ be required for performance. For example, the following requirements should be
+ met for Intel 82599 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 should be used, so for 2x ports, 2x NICs should
+ be used
+
+ - NIC ports connected to traffic generator. It is assumed that the NIC ports
+ P0, P1, P2, P3 (as identified by the DPDK application) are connected to the
+ traffic generator ports TG0, TG1, TG2, TG3. The application-side port mask of
+ NIC ports P0, P1, P2, P3 is noted as PORTMASK in this section.
+ Traffic generator should support sending jumbo frames with size up to 9K.
+
+2. BIOS requirements:
+
+ - Intel 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 requirements
+
+5. 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
+
+ - The test can be run with IPv4 package. The LPM table used for IPv4 packet routing is:
+
+ +-------+-------------------------------------+-----------+
+ |Entry #|LPM prefix (IP/length) |Output port|
+ +-------+-------------------------------------+-----------+
+ | 0 | 100.10.0.0/16 | P2 |
+ +-------+-------------------------------------+-----------+
+ | 1 | 100.20.0.0/16 | P2 |
+ +-------+-------------------------------------+-----------+
+ | 2 | 100.30.0.0/16 | P0 |
+ +-------+-------------------------------------+-----------+
+ | 3 | 100.40.0.0/16 | P0 |
+ +-------+-------------------------------------+-----------+
+
+
+ - The test can be run with IPv6 package, which follows rules below.
+
+ - There is no support for Hop-by-Hop or Routing extension headers in the packet
+ to be fragmented. All other optional headers, which are not part of the
+ unfragmentable part of the IPv6 packet are supported.
+
+ - When a fragment is generated, its identification field in the IPv6
+ fragmentation extension header is set to 0. This is not RFC compliant, but
+ proper identification number generation is out of the scope of the application
+ and routers in an IPv6 path are not allowed to fragment in the first place...
+ Generating that identification number is the job of a proper IP stack.
+
+ - The LPM table used for IPv6 packet routing is:
+
+ +-------+-------------------------------------+-----------+
+ |Entry #|LPM prefix (IP/length) |Output port|
+ +-------+-------------------------------------+-----------+
+ | 0 | 101:101:101:101:101:101:101:101/48| P2 |
+ +-------+-------------------------------------+-----------+
+ | 1 | 201:101:101:101:101:101:101:101/48| P2 |
+ +-------+-------------------------------------+-----------+
+ | 2 | 301:101:101:101:101:101:101:101/48| P0 |
+ +-------+-------------------------------------+-----------+
+ | 3 | 401:101:101:101:101:101:101:101/48| P0 |
+ +-------+-------------------------------------+-----------+
+
+ The following items are configured through the command line interface of the application:
+
+ - The set of one or several RX queues to be enabled for each NIC port
+ - The set of logical cores to execute the packet forwarding task
+ - Mapping of the NIC RX queues to logical cores handling them.
+
+6. Compile examples/ip_fragmentation::
+
+ meson configure -Dexamples=ip_fragmentation x86_64-native-linuxapp-gcc
+ ninja -C x86_64-native-linuxapp-gcc
+
+Test Case 1: Throughput test
+============================
+
+The test report should provide the throughput rate measurements (in mpps and % of the line rate for 2x NIC ports)
+for the following input frame sizes: 64 bytes, 1518 bytes, 1519 bytes, 2K, 9k.
+
+The following configurations should be tested:
+
+|
+
++----------+-------------------------+----------------------+
+|# of ports| Socket/Core/HyperThread|Total # of sw threads |
++----------+-------------------------+----------------------+
+| 2 | 1S/1C/1T | 1 |
++----------+-------------------------+----------------------+
+| 2 | 1S/1C/2T | 2 |
++----------+-------------------------+----------------------+
+| 2 | 1S/2C/1T | 2 |
++----------+-------------------------+----------------------+
+| 2 | 2S/1C/1T | 2 |
++----------+-------------------------+----------------------+
+
+|
+
+Command line::
+
+ ./x86_64-native-linuxapp-gcc/examples/dpdk-ip_fragmentation -c <LCOREMASK> -n 4 -- [-P] -p PORTMASK
+ -q <NUM_OF_PORTS_PER_THREAD>
diff --git a/tests/TestSuite_ipfrag.py b/tests/TestSuite_ipfrag.py
index 95f99281..169df06b 100644
--- a/tests/TestSuite_ipfrag.py
+++ b/tests/TestSuite_ipfrag.py
@@ -372,27 +372,6 @@ class TestIpfrag(TestCase):
self.dut.send_expect("^C", "#")
- def test_perf_ipfrag_throughtput(self):
- """
- Performance test for 64, 1518, 1519, 2k and 9k.
- """
- sizes = [64, 1518, 1519, 2000, 9000]
-
- tblheader = ["Ports", "S/C/T", "SW threads"]
- for size in sizes:
- tblheader.append("%dB Mpps" % size)
- tblheader.append("%d" % size)
-
- self.result_table_create(tblheader)
-
- lcores = [("1S/1C/1T", 2), ("1S/1C/2T", 2), ("1S/2C/1T", 2), ("2S/1C/1T", 2)]
- index = 1
- for (lcore, numThr) in lcores:
- self.benchmark(index, lcore, numThr, sizes)
- index += 1
-
- self.result_table_print()
-
def tear_down(self):
"""
Run after each test case.
diff --git a/tests/TestSuite_perf_ipfrag.py b/tests/TestSuite_perf_ipfrag.py
new file mode 100644
index 00000000..030aa378
--- /dev/null
+++ b/tests/TestSuite_perf_ipfrag.py
@@ -0,0 +1,386 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2010-2014 Intel Corporation
+#
+
+"""
+DPDK Test suite.
+Test IPv4 fragmentation features in DPDK.
+"""
+
+import os
+import re
+import string
+import time
+
+import framework.utils as utils
+from framework.packet import Packet
+from framework.pktgen import PacketGeneratorHelper
+from framework.settings import HEADER_SIZE
+from framework.test_case import TestCase
+
+lpm_table_ipv6 = [
+ "{{1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}, 48, P1}",
+ "{{2,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}, 48, P1}",
+ "{{3,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}, 48, P0}",
+ "{{4,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}, 48, P0}",
+ "{{5,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}, 48, P1}",
+ "{{6,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}, 48, P1}",
+ "{{7,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}, 48, P0}",
+ "{{8,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}, 48, P0}",
+]
+
+
+class TestIpfrag(TestCase):
+ def portRepl(self, match):
+ """
+ Function to replace P([0123]) pattern in tables
+ """
+
+ portid = match.group(1)
+ self.verify(int(portid) in range(4), "invalid port id")
+ return "%s" % eval("P" + str(portid))
+
+ def set_up_all(self):
+ """
+ ip_fragmentation Prerequisites
+ """
+
+ # Based on h/w type, choose how many ports to use
+ self.ports = self.dut.get_ports()
+
+ # Verify that enough ports are available
+ self.verify(len(self.ports) >= 2, "Insufficient ports for testing")
+
+ self.ports_socket = self.dut.get_numa_id(self.ports[0])
+
+ # Verify that enough threads are available
+ cores = self.dut.get_core_list("1S/1C/1T")
+ self.verify(cores is not None, "Insufficient cores for speed testing")
+
+ global P0, P1
+ P0 = self.ports[0]
+ P1 = self.ports[1]
+
+ # make application
+ out = self.dut.build_dpdk_apps("examples/ip_fragmentation")
+ self.verify("Error" not in out, "compilation error 1")
+ self.verify("No such file" not in out, "compilation error 2")
+
+ self.eal_para = self.dut.create_eal_parameters(
+ socket=self.ports_socket, ports=self.ports
+ )
+ portmask = utils.create_mask([P0, P1])
+ numPortThread = len([P0, P1]) / len(cores)
+
+ # run ipv4_frag
+ self.app_ip_fragmentation_path = self.dut.apps_name["ip_fragmentation"]
+ self.dut.send_expect(
+ "%s %s -- -p %s -q %s"
+ % (
+ self.app_ip_fragmentation_path,
+ self.eal_para,
+ portmask,
+ int(numPortThread),
+ ),
+ "Link [Uu]p",
+ 120,
+ )
+
+ time.sleep(2)
+ self.txItf = self.tester.get_interface(self.tester.get_local_port(P0))
+ self.rxItf = self.tester.get_interface(self.tester.get_local_port(P1))
+ self.dmac = self.dut.get_mac_address(P0)
+
+ # get dts output path
+ if self.logger.log_path.startswith(os.sep):
+ self.output_path = self.logger.log_path
+ else:
+ cur_path = os.path.dirname(os.path.dirname(os.path.realpath(__file__)))
+ self.output_path = os.sep.join([cur_path, self.logger.log_path])
+ # create an instance to set stream field setting
+ self.pktgen_helper = PacketGeneratorHelper()
+
+ def functional_check_ipv4(self, pkt_sizes, burst=1, flag=None):
+ """
+ Perform functional fragmentation checks.
+ """
+ for size in pkt_sizes[::burst]:
+ # simulate to set TG properties
+ if flag == "frag":
+ # do fragment, each packet max length 1518 - 18 - 20 = 1480
+ expPkts = int((size - HEADER_SIZE["eth"] - HEADER_SIZE["ip"]) / 1480)
+ if (size - HEADER_SIZE["eth"] - HEADER_SIZE["ip"]) % 1480:
+ expPkts += 1
+ val = 0
+ elif flag == "nofrag":
+ expPkts = 0
+ val = 2
+ else:
+ expPkts = 1
+ val = 2
+
+ inst = self.tester.tcpdump_sniff_packets(intf=self.rxItf)
+ # send packet
+ for times in range(burst):
+ pkt_size = pkt_sizes[pkt_sizes.index(size) + times]
+ pkt = Packet(pkt_type="UDP", pkt_len=pkt_size)
+ pkt.config_layer("ether", {"dst": "%s" % self.dmac})
+ pkt.config_layer(
+ "ipv4", {"dst": "100.20.0.1", "src": "1.2.3.4", "flags": val}
+ )
+ pkt.send_pkt(self.tester, tx_port=self.txItf)
+
+ # verify normal packet just by number, verify fragment packet by all elements
+ pkts = self.tester.load_tcpdump_sniff_packets(inst)
+ self.verify(
+ len(pkts) == expPkts,
+ "in functional_check_ipv4(): failed on forward packet size "
+ + str(size),
+ )
+ if flag == "frag":
+ idx = 1
+ for i in range(len(pkts)):
+ pkt_id = pkts.strip_element_layer3("id", p_index=i)
+ if idx == 1:
+ prev_idx = pkt_id
+ self.verify(
+ prev_idx == pkt_id, "Fragmented packets index not match"
+ )
+ prev_idx = pkt_id
+
+ # last flags should be 0
+ flags = pkts.strip_element_layer3("flags", p_index=i)
+ if idx == expPkts:
+ self.verify(
+ flags == 0, "Fragmented last packet flags not match"
+ )
+ else:
+ self.verify(flags == 1, "Fragmented packets flags not match")
+
+ # fragment offset should be correct
+ frag = pkts.strip_element_layer3("frag", p_index=i)
+ self.verify(
+ (frag == ((idx - 1) * 185)), "Fragment packet frag not match"
+ )
+ idx += 1
+
+ def functional_check_ipv6(self, pkt_sizes, burst=1, flag=None, funtion=None):
+ """
+ Perform functional fragmentation checks.
+ """
+ for size in pkt_sizes[::burst]:
+ # simulate to set TG properties
+ if flag == "frag":
+ # each packet max len: 1518 - 18 (eth) - 40 (ipv6) - 8 (ipv6 ext hdr) = 1452
+ expPkts = int((size - HEADER_SIZE["eth"] - HEADER_SIZE["ipv6"]) / 1452)
+ if (size - HEADER_SIZE["eth"] - HEADER_SIZE["ipv6"]) % 1452:
+ expPkts += 1
+ val = 0
+ else:
+ expPkts = 1
+ val = 2
+
+ inst = self.tester.tcpdump_sniff_packets(intf=self.rxItf)
+ # send packet
+ for times in range(burst):
+ pkt_size = pkt_sizes[pkt_sizes.index(size) + times]
+ pkt = Packet(pkt_type="IPv6_UDP", pkt_len=pkt_size)
+ pkt.config_layer("ether", {"dst": "%s" % self.dmac})
+ pkt.config_layer(
+ "ipv6",
+ {
+ "dst": "201:101:101:101:101:101:101:101",
+ "src": "ee80:ee80:ee80:ee80:ee80:ee80:ee80:ee80",
+ },
+ )
+ pkt.send_pkt(self.tester, tx_port=self.txItf)
+
+ # verify normal packet just by number, verify fragment packet by all elements
+ pkts = self.tester.load_tcpdump_sniff_packets(inst)
+ self.verify(
+ len(pkts) == expPkts,
+ "In functional_check_ipv6(): failed on forward packet size "
+ + str(size),
+ )
+ if flag == "frag":
+ idx = 1
+ for i in range(len(pkts)):
+ pkt_id = pkts.strip_element_layer4("id", p_index=i)
+ if idx == 1:
+ prev_idx = pkt_id
+ self.verify(
+ prev_idx == pkt_id, "Fragmented packets index not match"
+ )
+ prev_idx = pkt_id
+
+ # last flags should be 0
+ flags = pkts.strip_element_layer4("m", p_index=i)
+ if idx == expPkts:
+ self.verify(
+ flags == 0, "Fragmented last packet flags not match"
+ )
+ else:
+ self.verify(flags == 1, "Fragmented packets flags not match")
+
+ # fragment offset should be correct
+ frag = pkts.strip_element_layer4("offset", p_index=i)
+ self.verify(
+ (frag == int((idx - 1) * 181)), "Fragment packet frag not match"
+ )
+ idx += 1
+
+ def set_up(self):
+ """
+ Run before each test case.
+ """
+ self.tester.send_expect(
+ "ifconfig %s mtu 9200"
+ % self.tester.get_interface(self.tester.get_local_port(P0)),
+ "#",
+ )
+ self.tester.send_expect(
+ "ifconfig %s mtu 9200"
+ % self.tester.get_interface(self.tester.get_local_port(P1)),
+ "#",
+ )
+
+ def benchmark(self, index, lcore, num_pthreads, size_list):
+ """
+ Just Test IPv4 Throughput for selected parameters.
+ """
+
+ Bps = dict()
+ Pps = dict()
+ Pct = dict()
+
+ if int(lcore[0]) == 1:
+ eal_param = self.dut.create_eal_parameters(
+ cores=lcore, socket=self.ports_socket, ports=self.ports
+ )
+ else:
+ eal_param = self.dut.create_eal_parameters(cores=lcore, ports=self.ports)
+ portmask = utils.create_mask([P0, P1])
+ self.dut.send_expect("^c", "# ", 120)
+ self.dut.send_expect(
+ "%s %s -- -p %s -q %s"
+ % (self.app_ip_fragmentation_path, eal_param, portmask, num_pthreads),
+ "IP_FRAG:",
+ 120,
+ )
+ result = [2, lcore, num_pthreads]
+ for size in size_list:
+ dmac = self.dut.get_mac_address(P0)
+ flows_p0 = [
+ 'Ether(dst="%s")/IP(src="1.2.3.4", dst="100.10.0.1", flags=0)/("X"*%d)'
+ % (dmac, size - 38),
+ 'Ether(dst="%s")/IP(src="1.2.3.4", dst="100.20.0.1", flags=0)/("X"*%d)'
+ % (dmac, size - 38),
+ 'Ether(dst="%s")/IPv6(dst="101:101:101:101:101:101:101:101",src="ee80:ee80:ee80:ee80:ee80:ee80:ee80:ee80")/Raw(load="X"*%d)'
+ % (dmac, size - 58),
+ 'Ether(dst="%s")/IPv6(dst="201:101:101:101:101:101:101:101",src="ee80:ee80:ee80:ee80:ee80:ee80:ee80:ee80")/Raw(load="X"*%d)'
+ % (dmac, size - 58),
+ ]
+
+ # reserved for rx/tx bidirection test
+ dmac = self.dut.get_mac_address(P1)
+ flows_p1 = [
+ 'Ether(dst="%s")/IP(src="1.2.3.4", dst="100.30.0.1", flags=0)/("X"*%d)'
+ % (dmac, size - 38),
+ 'Ether(dst="%s")/IP(src="1.2.3.4", dst="100.40.0.1", flags=0)/("X"*%d)'
+ % (dmac, size - 38),
+ 'Ether(dst="%s")/IPv6(dst="301:101:101:101:101:101:101:101",src="ee80:ee80:ee80:ee80:ee80:ee80:ee80:ee80")/Raw(load="X"*%d)'
+ % (dmac, size - 58),
+ 'Ether(dst="%s")/IPv6(dst="401:101:101:101:101:101:101:101",src="ee80:ee80:ee80:ee80:ee80:ee80:ee80:ee80")/Raw(load="X"*%d)'
+ % (dmac, size - 58),
+ ]
+ flow_len = len(flows_p0)
+ tgenInput = []
+ for i in range(flow_len):
+
+ pcap0 = os.sep.join([self.output_path, "p0_{}.pcap".format(i)])
+ self.tester.scapy_append('wrpcap("%s", [%s])' % (pcap0, flows_p0[i]))
+ pcap1 = os.sep.join([self.output_path, "p1_{}.pcap".format(i)])
+ self.tester.scapy_append('wrpcap("%s", [%s])' % (pcap1, flows_p1[i]))
+ self.tester.scapy_execute()
+
+ tgenInput.append(
+ (
+ self.tester.get_local_port(P0),
+ self.tester.get_local_port(P1),
+ pcap0,
+ )
+ )
+ tgenInput.append(
+ (
+ self.tester.get_local_port(P1),
+ self.tester.get_local_port(P0),
+ pcap1,
+ )
+ )
+
+ factor = (size + 1517) / 1518
+ # wireSpd = 2 * 10000.0 / ((20 + size) * 8)
+
+ # clear streams before add new streams
+ self.tester.pktgen.clear_streams()
+ # run packet generator
+ streams = self.pktgen_helper.prepare_stream_from_tginput(
+ tgenInput, 100, None, self.tester.pktgen
+ )
+ Bps[str(size)], Pps[str(size)] = self.tester.pktgen.measure_throughput(
+ stream_ids=streams
+ )
+
+ self.verify(Pps[str(size)] > 0, "No traffic detected")
+ Pps[str(size)] *= 1.0 / factor / 1000000
+ Pct[str(size)] = (1.0 * Bps[str(size)] * 100) / (2 * 10000000000)
+
+ result.append(Pps[str(size)])
+ result.append(Pct[str(size)])
+
+ self.result_table_add(result)
+
+ self.dut.send_expect("^C", "#")
+
+ def test_perf_ipfrag_throughtput(self):
+ """
+ Performance test for 64, 1518, 1519, 2k and 9k.
+ """
+ sizes = [64, 1518, 1519, 2000, 9000]
+
+ tblheader = ["Ports", "S/C/T", "SW threads"]
+ for size in sizes:
+ tblheader.append("%dB Mpps" % size)
+ tblheader.append("%d" % size)
+
+ self.result_table_create(tblheader)
+
+ lcores = [("1S/1C/1T", 2), ("1S/1C/2T", 2), ("1S/2C/1T", 2), ("2S/1C/1T", 2)]
+ index = 1
+ for (lcore, numThr) in lcores:
+ self.benchmark(index, lcore, numThr, sizes)
+ index += 1
+
+ self.result_table_print()
+
+ def tear_down(self):
+ """
+ Run after each test case.
+ """
+ self.tester.send_expect(
+ "ifconfig %s mtu 1500"
+ % self.tester.get_interface(self.tester.get_local_port(P0)),
+ "#",
+ )
+ self.tester.send_expect(
+ "ifconfig %s mtu 1500"
+ % self.tester.get_interface(self.tester.get_local_port(P1)),
+ "#",
+ )
+
+ def tear_down_all(self):
+ """
+ Run after each test suite.
+ """
+ self.dut.send_expect("^C", "#")
+ pass
--
2.17.1
^ permalink raw reply [flat|nested] 9+ messages in thread
* [dts][PATCH V3 6/8] tests/multiprocess: split performance plan and suite
2023-01-10 16:44 [dts][PATCH V3 0/8] split performance plan and suite Hongbo Li
` (4 preceding siblings ...)
2023-01-10 16:44 ` [dts][PATCH V3 5/8] tests/ipfrag: " Hongbo Li
@ 2023-01-10 16:44 ` Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 7/8] tests/checksum_offload: " Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 8/8] conf/test_case_checklist: " Hongbo Li
7 siblings, 0 replies; 9+ messages in thread
From: Hongbo Li @ 2023-01-10 16:44 UTC (permalink / raw)
To: dts; +Cc: Hongbo Li
split performance plan and suite
Signed-off-by: Hongbo Li <hongbox.li@intel.com>
---
test_plans/multiprocess_test_plan.rst | 48 -
test_plans/perf_multiprocess_test_plan.rst | 141 +++
tests/TestSuite_multiprocess.py | 210 -----
tests/TestSuite_perf_multiprocess.py | 994 +++++++++++++++++++++
4 files changed, 1135 insertions(+), 258 deletions(-)
create mode 100644 test_plans/perf_multiprocess_test_plan.rst
create mode 100644 tests/TestSuite_perf_multiprocess.py
diff --git a/test_plans/multiprocess_test_plan.rst b/test_plans/multiprocess_test_plan.rst
index c7aae44b..9f5ef8fa 100644
--- a/test_plans/multiprocess_test_plan.rst
+++ b/test_plans/multiprocess_test_plan.rst
@@ -196,27 +196,6 @@ run should remain the same, except for the ``num-procs`` value, which should be
adjusted appropriately.
-Test Case: Performance Tests
-----------------------------
-
-Run the multiprocess application using standard IP traffic - varying source
-and destination address information to allow RSS to evenly distribute packets
-among RX queues. Record traffic throughput results as below.
-
-+-------------------+-----+-----+-----+-----+-----+-----+
-| Num-procs | 1 | 2 | 2 | 4 | 4 | 8 |
-+-------------------+-----+-----+-----+-----+-----+-----+
-| Cores/Threads | 1/1 | 1/2 | 2/1 | 2/2 | 4/1 | 4/2 |
-+-------------------+-----+-----+-----+-----+-----+-----+
-| Num Ports | 2 | 2 | 2 | 2 | 2 | 2 |
-+-------------------+-----+-----+-----+-----+-----+-----+
-| Packet Size | 64 | 64 | 64 | 64 | 64 | 64 |
-+-------------------+-----+-----+-----+-----+-----+-----+
-| %-age Line Rate | X | X | X | X | X | X |
-+-------------------+-----+-----+-----+-----+-----+-----+
-| Packet Rate(mpps) | X | X | X | X | X | X |
-+-------------------+-----+-----+-----+-----+-----+-----+
-
Test Case: Function Tests
-------------------------
start 2 symmetric_mp process, send some packets, the number of packets is a random value between 20 and 256.
@@ -294,33 +273,6 @@ An example commands to run 8 client processes is as follows::
root@host:mp_client# ./x86_64-native-linuxapp-gcc/examples/dpdk-mp_client -c 40000 --proc-type=secondary -- -n 6 &
root@host:mp_client# ./x86_64-native-linuxapp-gcc/examples/dpdk-mp_client -c 100000 --proc-type=secondary -- -n 7 &
-Test Case: Performance Measurement
-----------------------------------
-
-- On the traffic generator set up a traffic flow in both directions specifying
- IP traffic.
-- Run the server and client applications as above.
-- Start the traffic and record the throughput for transmitted and received packets.
-
-An example set of results is shown below.
-
-+----------------------+-----+-----+-----+-----+-----+-----+
-| Server threads | 1 | 1 | 1 | 1 | 1 | 1 |
-+----------------------+-----+-----+-----+-----+-----+-----+
-| Server Cores/Threads | 1/1 | 1/1 | 1/1 | 1/1 | 1/1 | 1/1 |
-+----------------------+-----+-----+-----+-----+-----+-----+
-| Num-clients | 1 | 2 | 2 | 4 | 4 | 8 |
-+----------------------+-----+-----+-----+-----+-----+-----+
-| Client Cores/Threads | 1/1 | 1/2 | 2/1 | 2/2 | 4/1 | 4/2 |
-+----------------------+-----+-----+-----+-----+-----+-----+
-| Num Ports | 2 | 2 | 2 | 2 | 2 | 2 |
-+----------------------+-----+-----+-----+-----+-----+-----+
-| Packet Size | 64 | 64 | 64 | 64 | 64 | 64 |
-+----------------------+-----+-----+-----+-----+-----+-----+
-| %-age Line Rate | X | X | X | X | X | X |
-+----------------------+-----+-----+-----+-----+-----+-----+
-| Packet Rate(mpps) | X | X | X | X | X | X |
-+----------------------+-----+-----+-----+-----+-----+-----+
Test Case: Function Tests
-------------------------
diff --git a/test_plans/perf_multiprocess_test_plan.rst b/test_plans/perf_multiprocess_test_plan.rst
new file mode 100644
index 00000000..c1e7ff87
--- /dev/null
+++ b/test_plans/perf_multiprocess_test_plan.rst
@@ -0,0 +1,141 @@
+.. SPDX-License-Identifier: BSD-3-Clause
+ Copyright(c) 2010-2017 Intel Corporation
+
+=======================================
+Sample Application Tests: Multi-Process
+=======================================
+
+Simple MP Application Test
+==========================
+
+Description
+-----------
+
+This test is a basic multi-process test which demonstrates the basics of sharing
+information between DPDK processes. The same application binary is run
+twice - once as a primary instance, and once as a secondary instance. Messages
+are sent from primary to secondary and vice versa, demonstrating the processes
+are sharing memory and can communicate using rte_ring structures.
+
+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
+
+Assuming that a DPDK build has been set up and the multi-process sample
+applications have been built.
+
+
+Test Case: Performance Tests
+----------------------------
+
+Run the multiprocess application using standard IP traffic - varying source
+and destination address information to allow RSS to evenly distribute packets
+among RX queues. Record traffic throughput results as below.
+
++-------------------+-----+-----+-----+-----+-----+-----+
+| Num-procs | 1 | 2 | 2 | 4 | 4 | 8 |
++-------------------+-----+-----+-----+-----+-----+-----+
+| Cores/Threads | 1/1 | 1/2 | 2/1 | 2/2 | 4/1 | 4/2 |
++-------------------+-----+-----+-----+-----+-----+-----+
+| Num Ports | 2 | 2 | 2 | 2 | 2 | 2 |
++-------------------+-----+-----+-----+-----+-----+-----+
+| Packet Size | 64 | 64 | 64 | 64 | 64 | 64 |
++-------------------+-----+-----+-----+-----+-----+-----+
+| %-age Line Rate | X | X | X | X | X | X |
++-------------------+-----+-----+-----+-----+-----+-----+
+| Packet Rate(mpps) | X | X | X | X | X | X |
++-------------------+-----+-----+-----+-----+-----+-----+
+
+
+Client Server Multiprocess Tests
+================================
+
+Description
+-----------
+
+The client-server sample application demonstrates the ability of Intel� DPDK
+to use multiple processes in which a server process performs packet I/O and one
+or multiple client processes perform packet processing. The server process
+controls load balancing on the traffic received from a number of input ports to
+a user-specified number of clients. The client processes forward the received
+traffic, outputting the packets directly by writing them to the TX rings of the
+outgoing ports.
+
+Prerequisites
+-------------
+
+Assuming that an Intel� DPDK build has been set up and the multi-process
+sample application has been built.
+Also assuming a traffic generator is connected to the ports "0" and "1".
+
+It is important to run the server application before the client application,
+as the server application manages both the NIC ports with packet transmission
+and reception, as well as shared memory areas and client queues.
+
+Run the Server Application:
+
+- Provide the core mask on which the server process is to run using -c, e.g. -c 3 (bitmask number).
+- Set the number of ports to be engaged using -p, e.g. -p 3 refers to ports 0 & 1.
+- Define the maximum number of clients using -n, e.g. -n 8.
+
+The command line below is an example on how to start the server process on
+logical core 2 to handle a maximum of 8 client processes configured to
+run on socket 0 to handle traffic from NIC ports 0 and 1::
+
+ root@host:mp_server# ./x86_64-native-linuxapp-gcc/examples/dpdk-mp_server -c 2 -- -p 3 -n 8
+
+NOTE: If an additional second core is given in the coremask to the server process
+that second core will be used to print statistics. When benchmarking, only a
+single lcore is needed for the server process
+
+Run the Client application:
+
+- In another terminal run the client application.
+- Give each client a distinct core mask with -c.
+- Give each client a unique client-id with -n.
+
+An example commands to run 8 client processes is as follows::
+
+ root@host:mp_client# ./x86_64-native-linuxapp-gcc/examples/dpdk-mp_client -c 40 --proc-type=secondary -- -n 0 &
+ root@host:mp_client# ./x86_64-native-linuxapp-gcc/examples/dpdk-mp_client -c 100 --proc-type=secondary -- -n 1 &
+ root@host:mp_client# ./x86_64-native-linuxapp-gcc/examples/dpdk-mp_client -c 400 --proc-type=secondary -- -n 2 &
+ root@host:mp_client# ./x86_64-native-linuxapp-gcc/examples/dpdk-mp_client -c 1000 --proc-type=secondary -- -n 3 &
+ root@host:mp_client# ./x86_64-native-linuxapp-gcc/examples/dpdk-mp_client -c 4000 --proc-type=secondary -- -n 4 &
+ root@host:mp_client# ./x86_64-native-linuxapp-gcc/examples/dpdk-mp_client -c 10000 --proc-type=secondary -- -n 5 &
+ root@host:mp_client# ./x86_64-native-linuxapp-gcc/examples/dpdk-mp_client -c 40000 --proc-type=secondary -- -n 6 &
+ root@host:mp_client# ./x86_64-native-linuxapp-gcc/examples/dpdk-mp_client -c 100000 --proc-type=secondary -- -n 7 &
+
+Test Case: Performance Measurement
+----------------------------------
+
+- On the traffic generator set up a traffic flow in both directions specifying
+ IP traffic.
+- Run the server and client applications as above.
+- Start the traffic and record the throughput for transmitted and received packets.
+
+An example set of results is shown below.
+
++----------------------+-----+-----+-----+-----+-----+-----+
+| Server threads | 1 | 1 | 1 | 1 | 1 | 1 |
++----------------------+-----+-----+-----+-----+-----+-----+
+| Server Cores/Threads | 1/1 | 1/1 | 1/1 | 1/1 | 1/1 | 1/1 |
++----------------------+-----+-----+-----+-----+-----+-----+
+| Num-clients | 1 | 2 | 2 | 4 | 4 | 8 |
++----------------------+-----+-----+-----+-----+-----+-----+
+| Client Cores/Threads | 1/1 | 1/2 | 2/1 | 2/2 | 4/1 | 4/2 |
++----------------------+-----+-----+-----+-----+-----+-----+
+| Num Ports | 2 | 2 | 2 | 2 | 2 | 2 |
++----------------------+-----+-----+-----+-----+-----+-----+
+| Packet Size | 64 | 64 | 64 | 64 | 64 | 64 |
++----------------------+-----+-----+-----+-----+-----+-----+
+| %-age Line Rate | X | X | X | X | X | X |
++----------------------+-----+-----+-----+-----+-----+-----+
+| Packet Rate(mpps) | X | X | X | X | X | X |
++----------------------+-----+-----+-----+-----+-----+-----+
\ No newline at end of file
diff --git a/tests/TestSuite_multiprocess.py b/tests/TestSuite_multiprocess.py
index 099ce6e7..a52622c9 100644
--- a/tests/TestSuite_multiprocess.py
+++ b/tests/TestSuite_multiprocess.py
@@ -1714,216 +1714,6 @@ class TestMultiprocess(TestCase):
"core dump" not in out, "Core dump occurred in the secondary process!!!"
)
- def test_perf_multiprocess_performance(self):
- """
- Benchmark Multiprocess performance.
- #"""
- packet_count = 16
- self.dut.send_expect("fg", "# ")
- txPort = self.tester.get_local_port(self.dut_ports[0])
- rxPort = self.tester.get_local_port(self.dut_ports[1])
- mac = self.tester.get_mac(txPort)
- dmac = self.dut.get_mac_address(self.dut_ports[0])
- tgenInput = []
-
- # create mutative src_ip+dst_ip package
- for i in range(packet_count):
- package = (
- r'flows = [Ether(src="%s", dst="%s")/IP(src="192.168.1.%d", dst="192.168.1.%d")/("X"*26)]'
- % (mac, dmac, i + 1, i + 2)
- )
- self.tester.scapy_append(package)
- pcap = os.sep.join([self.output_path, "test_%d.pcap" % i])
- self.tester.scapy_append('wrpcap("%s", flows)' % pcap)
- tgenInput.append([txPort, rxPort, pcap])
- self.tester.scapy_execute()
-
- # run multiple symmetric_mp process
- validExecutions = []
- for execution in executions:
- if len(self.dut.get_core_list(execution["cores"])) == execution["nprocs"]:
- validExecutions.append(execution)
-
- portMask = utils.create_mask(self.dut_ports)
-
- for n in range(len(validExecutions)):
- execution = validExecutions[n]
- # get coreList form execution['cores']
- coreList = self.dut.get_core_list(execution["cores"], socket=self.socket)
- # to run a set of symmetric_mp instances, like test plan
- dutSessionList = []
- for index in range(len(coreList)):
- dut_new_session = self.dut.new_session()
- dutSessionList.append(dut_new_session)
- # add -a option when tester and dut in same server
- dut_new_session.send_expect(
- self.app_symmetric_mp
- + " -c %s --proc-type=auto %s -- -p %s --num-procs=%d --proc-id=%d"
- % (
- utils.create_mask([coreList[index]]),
- self.eal_param,
- portMask,
- execution["nprocs"],
- index,
- ),
- "Finished Process Init",
- )
-
- # clear streams before add new streams
- self.tester.pktgen.clear_streams()
- # run packet generator
- streams = self.pktgen_helper.prepare_stream_from_tginput(
- tgenInput, 100, None, self.tester.pktgen
- )
- _, pps = self.tester.pktgen.measure_throughput(stream_ids=streams)
-
- execution["pps"] = pps
-
- # close all symmetric_mp process
- self.dut.send_expect("killall symmetric_mp", "# ")
- # close all dut sessions
- for dut_session in dutSessionList:
- self.dut.close_session(dut_session)
-
- # get rate and mpps data
- for n in range(len(executions)):
- self.verify(executions[n]["pps"] is not 0, "No traffic detected")
- self.result_table_create(
- [
- "Num-procs",
- "Sockets/Cores/Threads",
- "Num Ports",
- "Frame Size",
- "%-age Line Rate",
- "Packet Rate(mpps)",
- ]
- )
-
- for execution in validExecutions:
- self.result_table_add(
- [
- execution["nprocs"],
- execution["cores"],
- 2,
- 64,
- execution["pps"] / float(100000000 / (8 * 84)),
- execution["pps"] / float(1000000),
- ]
- )
-
- self.result_table_print()
-
- def test_perf_multiprocess_client_serverperformance(self):
- """
- Benchmark Multiprocess client-server performance.
- """
- self.dut.kill_all()
- self.dut.send_expect("fg", "# ")
- txPort = self.tester.get_local_port(self.dut_ports[0])
- rxPort = self.tester.get_local_port(self.dut_ports[1])
- mac = self.tester.get_mac(txPort)
-
- self.tester.scapy_append(
- 'dmac="%s"' % self.dut.get_mac_address(self.dut_ports[0])
- )
- self.tester.scapy_append('smac="%s"' % mac)
- self.tester.scapy_append(
- 'flows = [Ether(src=smac, dst=dmac)/IP(src="192.168.1.1", dst="192.168.1.1")/("X"*26)]'
- )
-
- pcap = os.sep.join([self.output_path, "test.pcap"])
- self.tester.scapy_append('wrpcap("%s", flows)' % pcap)
- self.tester.scapy_execute()
-
- validExecutions = []
- for execution in executions:
- if len(self.dut.get_core_list(execution["cores"])) == execution["nprocs"]:
- validExecutions.append(execution)
-
- for execution in validExecutions:
- coreList = self.dut.get_core_list(execution["cores"], socket=self.socket)
- # get core with socket parameter to specified which core dut used when tester and dut in same server
- coreMask = utils.create_mask(
- self.dut.get_core_list("1S/1C/1T", socket=self.socket)
- )
- portMask = utils.create_mask(self.dut_ports)
- # specified mp_server core and add -a option when tester and dut in same server
- self.dut.send_expect(
- self.app_mp_server
- + " -n %d -c %s %s -- -p %s -n %d"
- % (
- self.dut.get_memory_channels(),
- coreMask,
- self.eal_param,
- portMask,
- execution["nprocs"],
- ),
- "Finished Process Init",
- 20,
- )
- self.dut.send_expect("^Z", "\r\n")
- self.dut.send_expect("bg", "# ")
-
- for n in range(execution["nprocs"]):
- time.sleep(5)
- # use next core as mp_client core, different from mp_server
- coreMask = utils.create_mask([str(int(coreList[n]) + 1)])
- self.dut.send_expect(
- self.app_mp_client
- + " -n %d -c %s --proc-type=secondary %s -- -n %d"
- % (self.dut.get_memory_channels(), coreMask, self.eal_param, n),
- "Finished Process Init",
- )
- self.dut.send_expect("^Z", "\r\n")
- self.dut.send_expect("bg", "# ")
-
- tgenInput = []
- tgenInput.append([txPort, rxPort, pcap])
-
- # clear streams before add new streams
- self.tester.pktgen.clear_streams()
- # run packet generator
- streams = self.pktgen_helper.prepare_stream_from_tginput(
- tgenInput, 100, None, self.tester.pktgen
- )
- _, pps = self.tester.pktgen.measure_throughput(stream_ids=streams)
-
- execution["pps"] = pps
- self.dut.kill_all()
- time.sleep(5)
-
- for n in range(len(executions)):
- self.verify(executions[n]["pps"] is not 0, "No traffic detected")
-
- self.result_table_create(
- [
- "Server threads",
- "Server Cores/Threads",
- "Num-procs",
- "Sockets/Cores/Threads",
- "Num Ports",
- "Frame Size",
- "%-age Line Rate",
- "Packet Rate(mpps)",
- ]
- )
-
- for execution in validExecutions:
- self.result_table_add(
- [
- 1,
- "1S/1C/1T",
- execution["nprocs"],
- execution["cores"],
- 2,
- 64,
- execution["pps"] / float(100000000 / (8 * 84)),
- execution["pps"] / float(1000000),
- ]
- )
-
- self.result_table_print()
-
def set_fields(self):
"""set ip protocol field behavior"""
fields_config = {
diff --git a/tests/TestSuite_perf_multiprocess.py b/tests/TestSuite_perf_multiprocess.py
new file mode 100644
index 00000000..179574a5
--- /dev/null
+++ b/tests/TestSuite_perf_multiprocess.py
@@ -0,0 +1,994 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2010-2014 Intel Corporation
+#
+
+"""
+DPDK Test suite.
+Multi-process Test.
+"""
+
+import copy
+import os
+import random
+import re
+import time
+import traceback
+from collections import OrderedDict
+
+import framework.utils as utils
+from framework.exception import VerifyFailure
+from framework.packet import Packet
+from framework.pktgen import PacketGeneratorHelper
+from framework.pmd_output import PmdOutput
+from framework.test_case import TestCase, check_supported_nic
+from framework.utils import GREEN, RED
+
+from .rte_flow_common import FdirProcessing as fdirprocess
+from .rte_flow_common import RssProcessing as rssprocess
+
+executions = []
+
+
+class TestMultiprocess(TestCase):
+
+ support_nic = ["ICE_100G-E810C_QSFP", "ICE_25G-E810C_SFP", "ICE_25G-E810_XXV_SFP"]
+
+ def set_up_all(self):
+ """
+ Run at the start of each test suite.
+
+ Multiprocess prerequisites.
+ Requirements:
+ OS is not freeBSD
+ DUT core number >= 4
+ multi_process build pass
+ """
+ # self.verify('bsdapp' not in self.target, "Multiprocess not support freebsd")
+
+ self.verify(len(self.dut.get_all_cores()) >= 4, "Not enough Cores")
+ self.pkt = Packet()
+ self.dut_ports = self.dut.get_ports()
+ self.socket = self.dut.get_numa_id(self.dut_ports[0])
+ extra_option = "-Dexamples='multi_process/client_server_mp/mp_server,multi_process/client_server_mp/mp_client,multi_process/simple_mp,multi_process/symmetric_mp'"
+ self.dut.build_install_dpdk(target=self.target, extra_options=extra_option)
+ self.app_mp_client = self.dut.apps_name["mp_client"]
+ self.app_mp_server = self.dut.apps_name["mp_server"]
+ self.app_simple_mp = self.dut.apps_name["simple_mp"]
+ self.app_symmetric_mp = self.dut.apps_name["symmetric_mp"]
+
+ executions.append({"nprocs": 1, "cores": "1S/1C/1T", "pps": 0})
+ executions.append({"nprocs": 2, "cores": "1S/1C/2T", "pps": 0})
+ executions.append({"nprocs": 2, "cores": "1S/2C/1T", "pps": 0})
+ executions.append({"nprocs": 4, "cores": "1S/2C/2T", "pps": 0})
+ executions.append({"nprocs": 4, "cores": "1S/4C/1T", "pps": 0})
+ executions.append({"nprocs": 8, "cores": "1S/4C/2T", "pps": 0})
+
+ self.eal_param = ""
+ for i in self.dut_ports:
+ self.eal_param += " -a %s" % self.dut.ports_info[i]["pci"]
+
+ self.eal_para = self.dut.create_eal_parameters(cores="1S/2C/1T")
+ # start new session to run secondary
+ self.session_secondary = self.dut.new_session()
+
+ # get dts output path
+ if self.logger.log_path.startswith(os.sep):
+ self.output_path = self.logger.log_path
+ else:
+ cur_path = os.path.dirname(os.path.dirname(os.path.realpath(__file__)))
+ self.output_path = os.sep.join([cur_path, self.logger.log_path])
+ # create an instance to set stream field setting
+ self.pktgen_helper = PacketGeneratorHelper()
+ self.dport_info0 = self.dut.ports_info[self.dut_ports[0]]
+ self.pci0 = self.dport_info0["pci"]
+ self.tester_ifaces = [
+ self.tester.get_interface(self.dut.ports_map[port])
+ for port in self.dut_ports
+ ]
+ rxq = 1
+ self.session_list = []
+ self.logfmt = "*" * 20
+
+ def set_up(self):
+ """
+ Run before each test case.
+ """
+ pass
+
+ def launch_multi_testpmd(self, proc_type, queue_num, process_num, **kwargs):
+ self.session_list = [
+ self.dut.new_session("process_{}".format(i)) for i in range(process_num)
+ ]
+ self.pmd_output_list = [
+ PmdOutput(self.dut, self.session_list[i]) for i in range(process_num)
+ ]
+ self.dut.init_reserved_core()
+ proc_type_list = []
+ self.out_list = []
+ if isinstance(proc_type, list):
+ proc_type_list = copy.deepcopy(proc_type)
+ proc_type = proc_type_list[0]
+ for i in range(process_num):
+ cores = self.dut.get_reserved_core("2C", socket=0)
+ if i != 0 and proc_type_list:
+ proc_type = proc_type_list[1]
+ eal_param = "--proc-type={} -a {} --log-level=ice,7".format(
+ proc_type, self.pci0
+ )
+ param = "--rxq={0} --txq={0} --num-procs={1} --proc-id={2}".format(
+ queue_num, process_num, i
+ )
+ if kwargs.get("options") is not None:
+ param = "".join([param, kwargs.get("options")])
+ out = self.pmd_output_list[i].start_testpmd(
+ cores=cores,
+ eal_param=eal_param,
+ param=param,
+ timeout=kwargs.get("timeout", 20),
+ )
+ self.out_list.append(out)
+ self.pmd_output_list[i].execute_cmd("set fwd rxonly")
+ self.pmd_output_list[i].execute_cmd("set verbose 1")
+ self.pmd_output_list[i].execute_cmd("start")
+ self.pmd_output_list[i].execute_cmd("clear port stats all")
+
+ def get_pkt_statistic_process(self, out, **kwargs):
+ """
+ :param out: information received by testpmd process after sending packets and port statistics
+ :return: forward statistic dict, eg: {'rx-packets':1, 'tx-packets:0, 'tx-dropped':1}
+ """
+ p = re.compile(
+ r"Forward\s+Stats\s+for\s+RX\s+Port=\s+{}/Queue=([\s\d+]\d+)\s+.*\n.*RX-packets:\s+(\d+)\s+TX-packets:\s+(\d+)\s+TX-dropped:\s+(\d+)\s".format(
+ kwargs.get("port_id")
+ )
+ )
+ item_name = ["rx-packets", "tx-packets", "tx-dropped"]
+ statistic = p.findall(out)
+ if statistic:
+ rx_pkt_total, tx_pkt_total, tx_drop_total = 0, 0, 0
+ queue_set = set()
+ for item in statistic:
+ queue, rx_pkt, tx_pkt, tx_drop = map(int, item)
+ queue_set.add(queue)
+ rx_pkt_total += rx_pkt
+ tx_pkt_total += tx_pkt
+ tx_drop_total += tx_drop
+ static_dict = {
+ k: v
+ for k, v in zip(item_name, [rx_pkt_total, tx_pkt_total, tx_drop_total])
+ }
+ static_dict["queue"] = queue_set
+ return static_dict
+ else:
+ raise Exception("got wrong output, not match pattern {}".format(p.pattern))
+
+ def random_packet(self, pkt_num):
+ pkt = Packet()
+ if self.kdriver == "i40e":
+ pkt.generate_random_pkts(
+ pktnum=pkt_num,
+ dstmac="00:11:22:33:44:55",
+ random_type=["IP_RAW", "IPv6_RAW"],
+ )
+ else:
+ pkt.generate_random_pkts(
+ pktnum=pkt_num,
+ dstmac="00:11:22:33:44:55",
+ )
+ pkt.send_pkt(crb=self.tester, tx_port=self.tester_ifaces[0], count=1)
+
+ def specify_packet(self, que_num):
+ # create rule to set queue as one of each process queues
+ rule_str = "flow create 0 ingress pattern eth / ipv4 src is 192.168.{0}.3 / end actions queue index {0} / end"
+ rules = [rule_str.format(i) for i in range(que_num)]
+ fdirprocess(
+ self, self.pmd_output_list[0], self.tester_ifaces, rxq=que_num
+ ).create_rule(rules)
+ # send 1 packet for each queue,the number of packets should be received by each process is (queue_num/proc_num)
+ pkt = Packet()
+ pkt_num = que_num
+ self.logger.info("packet num:{}".format(pkt_num))
+ packets = [
+ 'Ether(dst="00:11:22:33:44:55") / IP(src="192.168.{0}.3", dst="192.168.0.21") / Raw("x" * 80)'.format(
+ i
+ )
+ for i in range(pkt_num)
+ ]
+ pkt.update_pkt(packets)
+ pkt.send_pkt(crb=self.tester, tx_port=self.tester_ifaces[0], count=1)
+
+ def _multiprocess_data_pass(self, case):
+ que_num, proc_num = case.get("queue_num"), case.get("proc_num")
+ pkt_num = case.setdefault("pkt_num", que_num)
+ step = int(que_num / proc_num)
+ proc_queue = [set(range(i, i + step)) for i in range(0, que_num, step)]
+ queue_dict = {
+ k: v
+ for k, v in zip(
+ ["process_{}".format(i) for i in range(que_num)], proc_queue
+ )
+ }
+ # start testpmd multi-process
+ self.launch_multi_testpmd(
+ proc_type=case.get("proc_type"), queue_num=que_num, process_num=proc_num
+ )
+ # send random or specify packets
+ packet_func = getattr(self, case.get("packet_type") + "_packet")
+ packet_func(pkt_num)
+ # get output for each process
+ process_static = {}
+ for i in range(len(self.pmd_output_list)):
+ out = self.pmd_output_list[i].execute_cmd("stop")
+ static = self.get_pkt_statistic_process(out, port_id=0)
+ process_static["process_{}".format(i)] = static
+ self.logger.info("process output static:{}".format(process_static))
+ # check whether each process receives packet, and ecah process receives packets with the corresponding queue
+ for k, v in process_static.items():
+ self.verify(
+ v.get("rx-packets") > 0,
+ "fail:process:{} does not receive packet".format(k),
+ )
+ self.verify(
+ v.get("queue").issubset(queue_dict.get(k)),
+ "fail: {} is not a subset of {}, "
+ "process should use its own queues".format(
+ v.get("queue"), queue_dict.get(k)
+ ),
+ )
+ self.logger.info("pass:each process receives packets and uses its own queue")
+ # check whether the sum of packets received by all processes is equal to the number of packets sent
+ received_pkts = sum(
+ int(v.get("rx-packets", 0)) for v in process_static.values()
+ )
+ self.verify(
+ received_pkts == pkt_num,
+ "the number of packets received is not equal to packets sent,"
+ "send packet:{}, received packet:{}".format(pkt_num, received_pkts),
+ )
+ self.logger.info(
+ "pass:the number of packets received is {}, equal to packets sent".format(
+ received_pkts
+ )
+ )
+
+ def check_rss(self, out, **kwargs):
+ """
+ check whether the packet directed by rss or not according to the specified parameters
+ :param out: information received by testpmd after sending packets and port statistics
+ :param kwargs: some specified parameters, such as: rxq, stats
+ :return: queue value list
+ usage:
+ check_rss(out, rxq=rxq, stats=stats)
+ """
+ self.logger.info("{0} check rss {0}".format(self.logfmt))
+ rxq = kwargs.get("rxq")
+ p = re.compile("RSS\shash=(\w+)\s-\sRSS\squeue=(\w+)")
+ pkt_info = p.findall(out)
+ self.verify(
+ pkt_info,
+ "no information matching the pattern was found,pattern:{}".format(
+ p.pattern
+ ),
+ )
+ pkt_queue = set([int(i[1], 16) for i in pkt_info])
+ if kwargs.get("stats"):
+ self.verify(
+ all([int(i[0], 16) % rxq == int(i[1], 16) for i in pkt_info]),
+ "some pkt not directed by rss.",
+ )
+ self.logger.info((GREEN("pass: all pkts directed by rss")))
+ else:
+ self.verify(
+ not any([int(i[0], 16) % rxq == int(i[1], 16) for i in pkt_info]),
+ "some pkt directed by rss, expect not directed by rss",
+ )
+ self.logger.info((GREEN("pass: no pkt directed by rss")))
+ return pkt_queue
+
+ def check_queue(self, out, check_param, **kwargs):
+ """
+ verify that queue value matches the expected value
+ :param out: information received by testpmd after sending packets and port statistics
+ :param check_param: check item name and value, eg
+ "check_param": {"port_id": 0, "queue": 2}
+ :param kwargs: some specified parameters, such as: pkt_num, port_id, stats
+ :return:
+ """
+ self.logger.info("{0} check queue {0}".format(self.logfmt))
+ queue = check_param["queue"]
+ if isinstance(check_param["queue"], int):
+ queue = [queue]
+ patt = re.compile(
+ r"port\s+{}/queue(.+?):\s+received\s+(\d+)\s+packets".format(
+ kwargs.get("port_id")
+ )
+ )
+ res = patt.findall(out)
+ if res:
+ pkt_queue = set([int(i[0]) for i in res])
+ if kwargs.get("stats"):
+ self.verify(
+ all(q in queue for q in pkt_queue),
+ "fail: queue id not matched, expect queue {}, got {}".format(
+ queue, pkt_queue
+ ),
+ )
+ self.logger.info((GREEN("pass: queue id {} matched".format(pkt_queue))))
+ else:
+ try:
+ self.verify(
+ not any(q in queue for q in pkt_queue),
+ "fail: queue id should not matched, {} should not in {}".format(
+ pkt_queue, queue
+ ),
+ )
+ self.logger.info(
+ (GREEN("pass: queue id {} not matched".format(pkt_queue)))
+ )
+ except VerifyFailure:
+ self.logger.info(
+ "queue id {} contains the queue {} specified in rule, so need to check"
+ " whether the packet directed by rss or not".format(
+ pkt_queue, queue
+ )
+ )
+ # for mismatch packet the 'stats' parameter is False, need to change to True
+ kwargs["stats"] = True
+ self.check_rss(out, **kwargs)
+
+ else:
+ raise Exception("got wrong output, not match pattern")
+
+ def check_mark_id(self, out, check_param, **kwargs):
+ """
+ verify that the mark ID matches the expected value
+ :param out: information received by testpmd after sending packets
+ :param check_param: check item name and value, eg
+ "check_param": {"port_id": 0, "mark_id": 1}
+ :param kwargs: some specified parameters,eg: stats
+ :return: None
+ usage:
+ check_mark_id(out, check_param, stats=stats)
+ """
+ self.logger.info("{0} check mark id {0}".format(self.logfmt))
+ fdir_scanner = re.compile("FDIR matched ID=(0x\w+)")
+ all_mark = fdir_scanner.findall(out)
+ stats = kwargs.get("stats")
+ if stats:
+ mark_list = set(int(i, 16) for i in all_mark)
+ self.verify(
+ all([i == check_param["mark_id"] for i in mark_list]) and mark_list,
+ "failed: some packet mark id of {} not match expect {}".format(
+ mark_list, check_param["mark_id"]
+ ),
+ )
+ self.logger.info((GREEN("pass: all packets mark id are matched ")))
+ else:
+ # for mismatch packet,verify no mark id in output of received packet
+ self.verify(
+ not all_mark, "mark id {} in output, expect no mark id".format(all_mark)
+ )
+ self.logger.info((GREEN("pass: no mark id in output")))
+
+ def check_drop(self, out, **kwargs):
+ """
+ check the drop number of packets according to the specified parameters
+ :param out: information received by testpmd after sending packets and port statistics
+ :param kwargs: some specified parameters, such as: pkt_num, port_id, stats
+ :return: None
+ usage:
+ chek_drop(out, pkt_num=pkt_num, port_id=portid, stats=stats)
+ """
+ self.logger.info("{0} check drop {0}".format(self.logfmt))
+ pkt_num = kwargs.get("pkt_num")
+ stats = kwargs.get("stats")
+ res = self.get_pkt_statistic(out, **kwargs)
+ self.verify(
+ pkt_num == res["rx-total"],
+ "failed: get wrong amount of packet {}, expected {}".format(
+ res["rx-total"], pkt_num
+ ),
+ )
+ drop_packet_num = res["rx-dropped"]
+ if stats:
+ self.verify(
+ drop_packet_num == pkt_num,
+ "failed: {} packet dropped,expect {} dropped".format(
+ drop_packet_num, pkt_num
+ ),
+ )
+ self.logger.info(
+ (
+ GREEN(
+ "pass: drop packet number {} is matched".format(drop_packet_num)
+ )
+ )
+ )
+ else:
+ self.verify(
+ drop_packet_num == 0 and res["rx-packets"] == pkt_num,
+ "failed: {} packet dropped, expect 0 packet dropped".format(
+ drop_packet_num
+ ),
+ )
+ self.logger.info(
+ (
+ GREEN(
+ "pass: drop packet number {} is matched".format(drop_packet_num)
+ )
+ )
+ )
+
+ @staticmethod
+ def get_pkt_statistic(out, **kwargs):
+ """
+ :param out: information received by testpmd after sending packets and port statistics
+ :return: rx statistic dict, eg: {'rx-packets':1, 'rx-dropped':0, 'rx-total':1}
+ """
+ p = re.compile(
+ r"Forward\sstatistics\s+for\s+port\s+{}\s+.*\n.*RX-packets:\s(\d+)\s+RX-dropped:\s(\d+)\s+RX-total:\s(\d+)\s".format(
+ kwargs.get("port_id")
+ )
+ )
+ item_name = ["rx-packets", "rx-dropped", "rx-total"]
+ statistic = p.findall(out)
+ if statistic:
+ static_dict = {
+ k: v for k, v in zip(item_name, list(map(int, list(statistic[0]))))
+ }
+ return static_dict
+ else:
+ raise Exception(
+ "got wrong output, not match pattern {}".format(p.pattern).replace(
+ "\\\\", "\\"
+ )
+ )
+
+ def send_pkt_get_output(
+ self, instance_obj, pkts, port_id=0, count=1, interval=0, get_stats=False
+ ):
+ instance_obj.pmd_output.execute_cmd("clear port stats all")
+ tx_port = self.tester_ifaces[port_id]
+ self.logger.info("----------send packet-------------")
+ self.logger.info("{}".format(pkts))
+ if not isinstance(pkts, list):
+ pkts = [pkts]
+ self.pkt.update_pkt(pkts)
+ self.pkt.send_pkt(
+ crb=self.tester,
+ tx_port=tx_port,
+ count=count,
+ interval=interval,
+ )
+ out1 = instance_obj.pmd_output.get_output(timeout=1)
+ if get_stats:
+ out2 = instance_obj.pmd_output.execute_cmd("show port stats all")
+ instance_obj.pmd_output.execute_cmd("stop")
+ else:
+ out2 = instance_obj.pmd_output.execute_cmd("stop")
+ instance_obj.pmd_output.execute_cmd("start")
+ return "".join([out1, out2])
+
+ def check_pkt_num(self, out, **kwargs):
+ """
+ check number of received packets matches the expected value
+ :param out: information received by testpmd after sending packets and port statistics
+ :param kwargs: some specified parameters, such as: pkt_num, port_id
+ :return: rx statistic dict
+ """
+ self.logger.info(
+ "{0} check pkt num for port:{1} {0}".format(
+ self.logfmt, kwargs.get("port_id")
+ )
+ )
+ pkt_num = kwargs.get("pkt_num")
+ res = self.get_pkt_statistic(out, **kwargs)
+ res_num = res["rx-total"]
+ self.verify(
+ res_num == pkt_num,
+ "fail: got wrong number of packets, expect pakcet number {}, got {}".format(
+ pkt_num, res_num
+ ),
+ )
+ self.logger.info(
+ (GREEN("pass: pkt num is {} same as expected".format(pkt_num)))
+ )
+ return res
+
+ def check_with_param(self, out, pkt_num, check_param, stats=True):
+ """
+ according to the key and value of the check parameter,
+ perform the corresponding verification in the out information
+ :param out: information received by testpmd after sending packets and port statistics
+ :param pkt_num: number of packets sent
+ :param check_param: check item name and value, eg:
+ "check_param": {"port_id": 0, "mark_id": 1, "queue": 1}
+ "check_param": {"port_id": 0, "drop": 1}
+ :param stats: effective status of rule, True or False, default is True
+ :return:
+ usage:
+ check_with_param(out, pkt_num, check_param, stats)
+ check_with_param(out, pkt_num, check_param=check_param)
+ """
+ rxq = check_param.get("rxq")
+ port_id = (
+ check_param["port_id"] if check_param.get("port_id") is not None else 0
+ )
+ match_flag = True
+ """
+ check_dict shows the supported check items,the key is item name and value represent the check priority,
+ the smaller the value, the higher the priority, priority default value is 999. if need to add new check item,
+ please add it to the dict and implement the corresponding method and named as 'check_itemname',eg: check_queue
+ """
+ self.matched_queue = []
+ default_pri = 999
+ check_dict = {
+ "queue": default_pri,
+ "drop": default_pri,
+ "mark_id": 1,
+ "rss": default_pri,
+ }
+ params = {"port_id": port_id, "rxq": rxq, "pkt_num": pkt_num, "stats": stats}
+ # sort check_param order by priority, from high to low, set priority as 999 if key not in check_dict
+ check_param = OrderedDict(
+ sorted(
+ check_param.items(),
+ key=lambda item: check_dict.get(item[0], default_pri),
+ )
+ )
+ if not check_param.get("drop"):
+ self.check_pkt_num(out, **params)
+ for k in check_param:
+ parameter = copy.deepcopy(params)
+ if k not in check_dict:
+ continue
+ func_name = "check_{}".format(k)
+ try:
+ func = getattr(self, func_name)
+ except AttributeError:
+ emsg = "{},this func is not implemented, please check!".format(
+ traceback.format_exc()
+ )
+ raise Exception(emsg)
+ else:
+ # for mismatch packet, if the check item is 'rss',should also verify the packets are distributed by rss
+ if k == "rss" and not stats:
+ parameter["stats"] = True
+ match_flag = False
+ res = func(out=out, check_param=check_param, **parameter)
+ if k == "rss" and match_flag:
+ self.matched_queue.append(res)
+
+ def destroy_rule(self, instance_obj, port_id=0, rule_id=None):
+ rule_id = 0 if rule_id is None else rule_id
+ if not isinstance(rule_id, list):
+ rule_id = [rule_id]
+ for i in rule_id:
+ out = instance_obj.pmd_output.execute_cmd(
+ "flow destroy {} rule {}".format(port_id, i)
+ )
+ p = re.compile(r"Flow rule #(\d+) destroyed")
+ m = p.search(out)
+ self.verify(m, "flow rule {} delete failed".format(rule_id))
+
+ def multiprocess_flow_data(self, case, **pmd_param):
+ que_num, proc_num = pmd_param.get("queue_num"), pmd_param.get("proc_num")
+ # start testpmd multi-process
+ self.launch_multi_testpmd(
+ proc_type=pmd_param.get("proc_type"),
+ queue_num=que_num,
+ process_num=proc_num,
+ )
+ self.pmd_output_list[0].execute_cmd("flow flush 0")
+ check_param = case["check_param"]
+ check_param["rxq"] = pmd_param.get("queue_num")
+ if check_param.get("rss"):
+ [pmd.execute_cmd("port config all rss all") for pmd in self.pmd_output_list]
+ fdir_pro = fdirprocess(
+ self,
+ self.pmd_output_list[0],
+ self.tester_ifaces,
+ rxq=pmd_param.get("queue_num"),
+ )
+ fdir_pro.create_rule(case.get("rule"))
+ # send match and mismatch packet
+ packets = [case.get("packet")["match"], case.get("packet")["mismatch"]]
+ for i in range(2):
+ out1 = self.send_pkt_get_output(fdir_pro, packets[i])
+ patt = re.compile(
+ r"port\s+{}/queue(.+?):\s+received\s+(\d+)\s+packets".format(
+ check_param.get("port_id")
+ )
+ )
+ if patt.findall(out1) and check_param.get("rss"):
+ self.logger.info(
+ "check whether the packets received by the primary process are distributed by RSS"
+ )
+ self.check_rss(out1, stats=True, **check_param)
+ for proc_pmd in self.pmd_output_list[1:]:
+ out2 = proc_pmd.get_output(timeout=1)
+ out3 = proc_pmd.execute_cmd("stop")
+ out1 = "".join([out1, out2, out3])
+ proc_pmd.execute_cmd("start")
+ if patt.findall(out2) and check_param.get("rss"):
+ self.logger.info(
+ "check whether the packets received by the secondary process are distributed by RSS"
+ )
+ self.check_rss(out2, stats=True, **check_param)
+ pkt_num = len(packets[i])
+ self.check_with_param(
+ out1,
+ pkt_num=pkt_num,
+ check_param=check_param,
+ stats=True if i == 0 else False,
+ )
+
+ def _handle_test(self, tests, instance_obj, port_id=0):
+ instance_obj.pmd_output.wait_link_status_up(port_id)
+ for test in tests:
+ if "send_packet" in test:
+ out = self.send_pkt_get_output(
+ instance_obj, test["send_packet"], port_id
+ )
+ for proc_pmd in self.pmd_output_list[1:]:
+ out1 = proc_pmd.get_output(timeout=1)
+ out = "".join([out, out1])
+ if "action" in test:
+ instance_obj.handle_actions(out, test["action"])
+
+ def multiprocess_rss_data(self, case, **pmd_param):
+ que_num, proc_num = pmd_param.get("queue_num"), pmd_param.get("proc_num")
+ # start testpmd multi-process
+ self.launch_multi_testpmd(
+ proc_type=pmd_param.get("proc_type"),
+ queue_num=que_num,
+ process_num=proc_num,
+ options=pmd_param.get("options", None),
+ )
+ self.pmd_output_list[0].execute_cmd("flow flush 0")
+ rss_pro = rssprocess(
+ self,
+ self.pmd_output_list[0],
+ self.tester_ifaces,
+ rxq=pmd_param.get("queue_num"),
+ )
+ rss_pro.error_msgs = []
+ # handle tests
+ tests = case["test"]
+ port_id = case["port_id"]
+ self.logger.info("------------handle test--------------")
+ # validate rule
+ rule = case.get("rule", None)
+ if rule:
+ rss_pro.validate_rule(rule=rule)
+ rule_ids = rss_pro.create_rule(rule=rule)
+ rss_pro.check_rule(rule_list=rule_ids)
+ self._handle_test(tests, rss_pro, port_id)
+ # handle post-test
+ if "post-test" in case:
+ self.logger.info("------------handle post-test--------------")
+ self.destroy_rule(rss_pro, port_id=port_id, rule_id=rule_ids)
+ rss_pro.check_rule(port_id=port_id, stats=False)
+ self._handle_test(case["post-test"], rss_pro, port_id)
+ if rss_pro.error_msgs:
+ self.verify(
+ False,
+ " ".join([errs.replace("'", " ") for errs in rss_pro.error_msgs[:500]]),
+ )
+
+ def rte_flow(self, case_list, func_name, **kwargs):
+ """
+ main flow of case:
+ 1. iterate the case list and do the below steps:
+ a. get the subcase name and init dict to save result
+ b. call method by func name to execute case step
+ c. record case result and err msg if case failed
+ d. clear flow rule
+ 2. calculate the case passing rate according to the result dict
+ 3. record case result and pass rate in the case log file
+ 4. verify whether the case pass rate is equal to 100, if not, mark the case as failed and raise the err msg
+ :param case_list: case list, each item is a subcase of case
+ :param func_name: hadle case method name, eg:
+ 'flow_rule_operate': a method of 'FlowRuleProcessing' class,
+ used to handle flow rule related suites,such as fdir and switch_filter
+ 'handle_rss_distribute_cases': a method of 'RssProcessing' class,
+ used to handle rss related suites
+ :return:
+ usage:
+ for flow rule related:
+ rte_flow(caselist, flow_rule_operate)
+ for rss related:
+ rte_flow(caselist, handle_rss_distribute_cases)
+ """
+ if not isinstance(case_list, list):
+ case_list = [case_list]
+ test_results = dict()
+ for case in case_list:
+ case_name = case.get("sub_casename")
+ test_results[case_name] = {}
+ try:
+ self.logger.info("{0} case_name:{1} {0}".format("*" * 20, case_name))
+ func_name(case, **kwargs)
+ except Exception:
+ test_results[case_name]["result"] = "failed"
+ test_results[case_name]["err"] = re.sub(
+ r"['\r\n]", "", str(traceback.format_exc(limit=1))
+ ).replace("\\\\", "\\")
+ self.logger.info(
+ (
+ RED(
+ "case failed:{}, err:{}".format(
+ case_name, traceback.format_exc()
+ )
+ )
+ )
+ )
+ else:
+ test_results[case_name]["result"] = "passed"
+ self.logger.info((GREEN("case passed: {}".format(case_name))))
+ finally:
+ self.session_list[0].send_command("flow flush 0", timeout=1)
+ for sess in self.session_list:
+ self.dut.close_session(sess)
+ pass_rate = (
+ round(
+ sum(1 for k in test_results if "passed" in test_results[k]["result"])
+ / len(test_results),
+ 4,
+ )
+ * 100
+ )
+ self.logger.info(
+ [
+ "{}:{}".format(sub_name, test_results[sub_name]["result"])
+ for sub_name in test_results
+ ]
+ )
+ self.logger.info("pass rate is: {}".format(pass_rate))
+ msg = [
+ "subcase_name:{}:{},err:{}".format(
+ name, test_results[name].get("result"), test_results[name].get("err")
+ )
+ for name in test_results.keys()
+ if "failed" in test_results[name]["result"]
+ ]
+ self.verify(
+ int(pass_rate) == 100,
+ "some subcases failed, detail as below:{}".format(msg),
+ )
+
+ def test_perf_multiprocess_performance(self):
+ """
+ Benchmark Multiprocess performance.
+ #"""
+ packet_count = 16
+ self.dut.send_expect("fg", "# ")
+ txPort = self.tester.get_local_port(self.dut_ports[0])
+ rxPort = self.tester.get_local_port(self.dut_ports[1])
+ mac = self.tester.get_mac(txPort)
+ dmac = self.dut.get_mac_address(self.dut_ports[0])
+ tgenInput = []
+
+ # create mutative src_ip+dst_ip package
+ for i in range(packet_count):
+ package = (
+ r'flows = [Ether(src="%s", dst="%s")/IP(src="192.168.1.%d", dst="192.168.1.%d")/("X"*26)]'
+ % (mac, dmac, i + 1, i + 2)
+ )
+ self.tester.scapy_append(package)
+ pcap = os.sep.join([self.output_path, "test_%d.pcap" % i])
+ self.tester.scapy_append('wrpcap("%s", flows)' % pcap)
+ tgenInput.append([txPort, rxPort, pcap])
+ self.tester.scapy_execute()
+
+ # run multiple symmetric_mp process
+ validExecutions = []
+ for execution in executions:
+ if len(self.dut.get_core_list(execution["cores"])) == execution["nprocs"]:
+ validExecutions.append(execution)
+
+ portMask = utils.create_mask(self.dut_ports)
+
+ for n in range(len(validExecutions)):
+ execution = validExecutions[n]
+ # get coreList form execution['cores']
+ coreList = self.dut.get_core_list(execution["cores"], socket=self.socket)
+ # to run a set of symmetric_mp instances, like test plan
+ dutSessionList = []
+ for index in range(len(coreList)):
+ dut_new_session = self.dut.new_session()
+ dutSessionList.append(dut_new_session)
+ # add -a option when tester and dut in same server
+ dut_new_session.send_expect(
+ self.app_symmetric_mp
+ + " -c %s --proc-type=auto %s -- -p %s --num-procs=%d --proc-id=%d"
+ % (
+ utils.create_mask([coreList[index]]),
+ self.eal_param,
+ portMask,
+ execution["nprocs"],
+ index,
+ ),
+ "Finished Process Init",
+ )
+
+ # clear streams before add new streams
+ self.tester.pktgen.clear_streams()
+ # run packet generator
+ streams = self.pktgen_helper.prepare_stream_from_tginput(
+ tgenInput, 100, None, self.tester.pktgen
+ )
+ _, pps = self.tester.pktgen.measure_throughput(stream_ids=streams)
+
+ execution["pps"] = pps
+
+ # close all symmetric_mp process
+ self.dut.send_expect("killall symmetric_mp", "# ")
+ # close all dut sessions
+ for dut_session in dutSessionList:
+ self.dut.close_session(dut_session)
+
+ # get rate and mpps data
+ for n in range(len(executions)):
+ self.verify(executions[n]["pps"] is not 0, "No traffic detected")
+ self.result_table_create(
+ [
+ "Num-procs",
+ "Sockets/Cores/Threads",
+ "Num Ports",
+ "Frame Size",
+ "%-age Line Rate",
+ "Packet Rate(mpps)",
+ ]
+ )
+
+ for execution in validExecutions:
+ self.result_table_add(
+ [
+ execution["nprocs"],
+ execution["cores"],
+ 2,
+ 64,
+ execution["pps"] / float(100000000 / (8 * 84)),
+ execution["pps"] / float(1000000),
+ ]
+ )
+
+ self.result_table_print()
+
+ def test_perf_multiprocess_client_serverperformance(self):
+ """
+ Benchmark Multiprocess client-server performance.
+ """
+ self.dut.kill_all()
+ self.dut.send_expect("fg", "# ")
+ txPort = self.tester.get_local_port(self.dut_ports[0])
+ rxPort = self.tester.get_local_port(self.dut_ports[1])
+ mac = self.tester.get_mac(txPort)
+
+ self.tester.scapy_append(
+ 'dmac="%s"' % self.dut.get_mac_address(self.dut_ports[0])
+ )
+ self.tester.scapy_append('smac="%s"' % mac)
+ self.tester.scapy_append(
+ 'flows = [Ether(src=smac, dst=dmac)/IP(src="192.168.1.1", dst="192.168.1.1")/("X"*26)]'
+ )
+
+ pcap = os.sep.join([self.output_path, "test.pcap"])
+ self.tester.scapy_append('wrpcap("%s", flows)' % pcap)
+ self.tester.scapy_execute()
+
+ validExecutions = []
+ for execution in executions:
+ if len(self.dut.get_core_list(execution["cores"])) == execution["nprocs"]:
+ validExecutions.append(execution)
+
+ for execution in validExecutions:
+ coreList = self.dut.get_core_list(execution["cores"], socket=self.socket)
+ # get core with socket parameter to specified which core dut used when tester and dut in same server
+ coreMask = utils.create_mask(
+ self.dut.get_core_list("1S/1C/1T", socket=self.socket)
+ )
+ portMask = utils.create_mask(self.dut_ports)
+ # specified mp_server core and add -a option when tester and dut in same server
+ self.dut.send_expect(
+ self.app_mp_server
+ + " -n %d -c %s %s -- -p %s -n %d"
+ % (
+ self.dut.get_memory_channels(),
+ coreMask,
+ self.eal_param,
+ portMask,
+ execution["nprocs"],
+ ),
+ "Finished Process Init",
+ 20,
+ )
+ self.dut.send_expect("^Z", "\r\n")
+ self.dut.send_expect("bg", "# ")
+
+ for n in range(execution["nprocs"]):
+ time.sleep(5)
+ # use next core as mp_client core, different from mp_server
+ coreMask = utils.create_mask([str(int(coreList[n]) + 1)])
+ self.dut.send_expect(
+ self.app_mp_client
+ + " -n %d -c %s --proc-type=secondary %s -- -n %d"
+ % (self.dut.get_memory_channels(), coreMask, self.eal_param, n),
+ "Finished Process Init",
+ )
+ self.dut.send_expect("^Z", "\r\n")
+ self.dut.send_expect("bg", "# ")
+
+ tgenInput = []
+ tgenInput.append([txPort, rxPort, pcap])
+
+ # clear streams before add new streams
+ self.tester.pktgen.clear_streams()
+ # run packet generator
+ streams = self.pktgen_helper.prepare_stream_from_tginput(
+ tgenInput, 100, None, self.tester.pktgen
+ )
+ _, pps = self.tester.pktgen.measure_throughput(stream_ids=streams)
+
+ execution["pps"] = pps
+ self.dut.kill_all()
+ time.sleep(5)
+
+ for n in range(len(executions)):
+ self.verify(executions[n]["pps"] is not 0, "No traffic detected")
+
+ self.result_table_create(
+ [
+ "Server threads",
+ "Server Cores/Threads",
+ "Num-procs",
+ "Sockets/Cores/Threads",
+ "Num Ports",
+ "Frame Size",
+ "%-age Line Rate",
+ "Packet Rate(mpps)",
+ ]
+ )
+
+ for execution in validExecutions:
+ self.result_table_add(
+ [
+ 1,
+ "1S/1C/1T",
+ execution["nprocs"],
+ execution["cores"],
+ 2,
+ 64,
+ execution["pps"] / float(100000000 / (8 * 84)),
+ execution["pps"] / float(1000000),
+ ]
+ )
+
+ self.result_table_print()
+
+ def set_fields(self):
+ """set ip protocol field behavior"""
+ fields_config = {
+ "ip": {
+ "src": {"range": 64, "action": "inc"},
+ "dst": {"range": 64, "action": "inc"},
+ },
+ }
+
+ return fields_config
+
+ def tear_down(self):
+ """
+ Run after each test case.
+ """
+ if self.session_list:
+ for sess in self.session_list:
+ self.dut.close_session(sess)
+ self.dut.kill_all()
+
+ def tear_down_all(self):
+ """
+ Run after each test suite.
+ """
+ self.dut.kill_all()
--
2.17.1
^ permalink raw reply [flat|nested] 9+ messages in thread
* [dts][PATCH V3 7/8] tests/checksum_offload: split performance plan and suite
2023-01-10 16:44 [dts][PATCH V3 0/8] split performance plan and suite Hongbo Li
` (5 preceding siblings ...)
2023-01-10 16:44 ` [dts][PATCH V3 6/8] tests/multiprocess: " Hongbo Li
@ 2023-01-10 16:44 ` Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 8/8] conf/test_case_checklist: " Hongbo Li
7 siblings, 0 replies; 9+ messages in thread
From: Hongbo Li @ 2023-01-10 16:44 UTC (permalink / raw)
To: dts; +Cc: Hongbo Li
split performance plan and suite
Signed-off-by: Hongbo Li <hongbox.li@intel.com>
---
tests/TestSuite_checksum_offload.py | 54 --
tests/TestSuite_perf_checksum_offload.py | 624 +++++++++++++++++++++++
2 files changed, 624 insertions(+), 54 deletions(-)
create mode 100644 tests/TestSuite_perf_checksum_offload.py
diff --git a/tests/TestSuite_checksum_offload.py b/tests/TestSuite_checksum_offload.py
index 0214231c..9ba06a19 100644
--- a/tests/TestSuite_checksum_offload.py
+++ b/tests/TestSuite_checksum_offload.py
@@ -756,60 +756,6 @@ class TestChecksumOffload(TestCase):
self.result_table_add(result)
- def test_perf_checksum_throughtput(self):
- """
- Test checksum offload performance.
- """
- # Verify that enough ports are available
- self.verify(len(self.dut_ports) >= 2, "Insufficient ports for testing")
- self.dut.send_expect("quit", "#")
-
- # sizes = [64, 128, 256, 512, 1024]
- sizes = [64, 128]
- pkts = {
- "IP/UDP": 'Ether(dst="%s", src="52:00:00:00:00:00")/IP()/UDP()/("X"*(%d-46))',
- "IP/TCP": 'Ether(dst="%s", src="52:00:00:00:00:00")/IP()/TCP()/("X"*(%d-58))',
- "IP/SCTP": 'Ether(dst="%s", src="52:00:00:00:00:00")/IP()/SCTP()/("X"*(%d-50+2))',
- "IPv6/UDP": 'Ether(dst="%s", src="52:00:00:00:00:00")/IPv6()/UDP()/("X"* (lambda x: x - 66 if x > 66 else 0)(%d))',
- "IPv6/TCP": 'Ether(dst="%s", src="52:00:00:00:00:00")/IPv6()/TCP()/("X"* (lambda x: x - 78 if x > 78 else 0)(%d))',
- }
-
- if self.kdriver in DRIVER_TEST_LACK_CAPA["sctp_tx_offload"]:
- del pkts["IP/SCTP"]
-
- lcore = "1S/2C/1T"
- portMask = utils.create_mask([self.dut_ports[0], self.dut_ports[1]])
- for mode in ["sw", "hw"]:
- self.logger.info("%s performance" % mode)
- tblheader = ["Ports", "S/C/T", "Packet Type", "Mode"]
- for size in sizes:
- tblheader.append("%sB mpps" % str(size))
- tblheader.append("%sB %% " % str(size))
- self.result_table_create(tblheader)
- self.pmdout.start_testpmd(
- lcore,
- "--portmask=%s" % self.portMask
- + " --enable-rx-cksum "
- + "--port-topology=loop",
- socket=self.ports_socket,
- )
-
- self.dut.send_expect("set fwd csum", "testpmd> ")
- if mode == "hw":
- self.checksum_enablehw(self.dut_ports[0])
- self.checksum_enablehw(self.dut_ports[1])
- else:
- self.checksum_enablesw(self.dut_ports[0])
- self.checksum_enablesw(self.dut_ports[1])
-
- self.dut.send_expect("start", "testpmd> ", 3)
- for ptype in list(pkts.keys()):
- self.benchmark(lcore, ptype, mode, pkts[ptype], sizes, self.nic)
-
- self.dut.send_expect("stop", "testpmd> ")
- self.dut.send_expect("quit", "#", 10)
- self.result_table_print()
-
def test_hardware_checksum_check_ip_rx(self):
self.tester.send_expect("scapy", ">>>")
self.checksum_enablehw(self.dut_ports[0])
diff --git a/tests/TestSuite_perf_checksum_offload.py b/tests/TestSuite_perf_checksum_offload.py
new file mode 100644
index 00000000..a2288613
--- /dev/null
+++ b/tests/TestSuite_perf_checksum_offload.py
@@ -0,0 +1,624 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2010-2014 Intel Corporation
+# Copyright(c) 2018-2019 The University of New Hampshire
+#
+
+"""
+DPDK Test suite.
+
+Test support of RX/TX Checksum Offload Features by Poll Mode Drivers.
+
+"""
+
+import os
+import re
+import subprocess
+import time
+from typing import List, Pattern, Tuple, Union
+
+from scapy.layers.inet import IP, TCP, UDP
+from scapy.layers.inet6 import IPv6
+from scapy.layers.l2 import GRE, Ether
+from scapy.layers.sctp import SCTP
+from scapy.layers.vxlan import VXLAN
+from scapy.packet import Raw
+from scapy.utils import rdpcap, wrpcap
+
+import framework.packet as packet
+import framework.utils as utils
+from framework.exception import VerifyFailure
+from framework.pktgen import PacketGeneratorHelper
+from framework.pmd_output import PmdOutput
+from framework.rst import RstReport
+from framework.settings import FOLDERS
+from framework.test_capabilities import DRIVER_TEST_LACK_CAPA
+from framework.test_case import TestCase
+
+l3_proto_classes = [IP, IPv6]
+
+l4_proto_classes = [
+ UDP,
+ TCP,
+]
+
+tunnelling_proto_classes = [
+ VXLAN,
+ GRE,
+]
+
+l3_protos = ["IP", "IPv6"]
+
+l4_protos = [
+ "UDP",
+ "TCP",
+ "SCTP",
+]
+
+
+class TestChecksumOffload(TestCase):
+ def set_up_all(self):
+ """
+ Run at the start of each test suite.
+ Checksum offload prerequisites.
+ """
+ # Based on h/w type, choose how many ports to use
+ self.dut_ports = self.dut.get_ports(self.nic)
+ # Verify that enough ports are available
+ self.verify(len(self.dut_ports) >= 1, "Insufficient ports for testing")
+ self.pmdout: PmdOutput = PmdOutput(self.dut)
+ self.portMask = utils.create_mask([self.dut_ports[0]])
+ self.ports_socket = self.dut.get_numa_id(self.dut_ports[0])
+ # get dts output path
+ if self.logger.log_path.startswith(os.sep):
+ self.output_path = self.logger.log_path
+ else:
+ cur_path = os.path.dirname(os.path.dirname(os.path.realpath(__file__)))
+ self.output_path = os.sep.join([cur_path, self.logger.log_path])
+ # log debug used
+ self.count = 0
+
+ def set_up(self):
+ """
+ Run before each test case.
+ """
+ self.pmdout.start_testpmd(
+ "Default",
+ "--portmask=%s " % (self.portMask)
+ + " --enable-rx-cksum "
+ + "--port-topology=loop",
+ socket=self.ports_socket,
+ )
+ self.dut.send_expect("set verbose 1", "testpmd>")
+ self.dut.send_expect("set fwd csum", "testpmd>")
+ self.dut.send_expect("csum mac-swap off 0", "testpmd>")
+
+ def checksum_enablehw(self, port):
+ self.dut.send_expect("port stop all", "testpmd>")
+ self.dut.send_expect("rx_vxlan_port add 4789 0 ", "testpmd>")
+ self.dut.send_expect("csum set ip hw %d" % port, "testpmd>")
+ self.dut.send_expect("csum set udp hw %d" % port, "testpmd>")
+ self.dut.send_expect("csum set tcp hw %d" % port, "testpmd>")
+ self.dut.send_expect("csum set sctp hw %d" % port, "testpmd>")
+ self.dut.send_expect("csum set outer-ip hw %d" % port, "testpmd>")
+ self.dut.send_expect("csum set outer-udp hw %d" % port, "testpmd>")
+ self.dut.send_expect("csum parse-tunnel on %d" % port, "testpmd>")
+ self.dut.send_expect("port start all", "testpmd>")
+
+ def checksum_enablesw(self, port):
+ self.dut.send_expect("port stop all", "testpmd>")
+ self.dut.send_expect("csum set ip sw %d" % port, "testpmd>")
+ self.dut.send_expect("csum set udp sw %d" % port, "testpmd>")
+ self.dut.send_expect("csum set tcp sw %d" % port, "testpmd>")
+ self.dut.send_expect("csum set sctp sw %d" % port, "testpmd>")
+ self.dut.send_expect("port start all", "testpmd>")
+
+ def get_chksum_values(self, packets_expected):
+ """
+ Validate the checksum flags.
+ """
+ checksum_pattern = re.compile("chksum.*=.*(0x[0-9a-z]+)")
+
+ chksum = dict()
+
+ self.tester.send_expect("scapy", ">>> ")
+
+ for packet_type in list(packets_expected.keys()):
+ self.tester.send_expect("p = %s" % packets_expected[packet_type], ">>>")
+ out = self.tester.send_command("p.show2()", timeout=1)
+ chksums = checksum_pattern.findall(out)
+ chksum[packet_type] = chksums
+
+ self.tester.send_expect("exit()", "#")
+
+ return chksum
+
+ def checksum_valid_flags(self, packets_sent, flag):
+ """
+ Sends packets and check the checksum valid-flags.
+ """
+ self.dut.send_expect("start", "testpmd>")
+ tx_interface = self.tester.get_interface(
+ self.tester.get_local_port(self.dut_ports[0])
+ )
+ for packet_type in list(packets_sent.keys()):
+ self.pkt = packet.Packet(pkt_str=packets_sent[packet_type])
+ self.pkt.send_pkt(self.tester, tx_interface, count=4)
+ out = self.dut.get_session_output(timeout=1)
+ lines = out.split("\r\n")
+
+ # collect the checksum result
+ for line in lines:
+ line = line.strip()
+ if len(line) != 0 and line.startswith("rx"):
+ # IPv6 don't be checksum, so always show "GOOD"
+ if packet_type.startswith("IPv6"):
+ if "RTE_MBUF_F_RX_L4_CKSUM" not in line:
+ self.verify(0, "There is no checksum flags appeared!")
+ else:
+ if flag == 1:
+ self.verify(
+ "RTE_MBUF_F_RX_L4_CKSUM_GOOD" in line,
+ "Packet Rx L4 checksum valid-flags error!",
+ )
+ elif flag == 0:
+ self.verify(
+ "RTE_MBUF_F_RX_L4_CKSUM_BAD" in line
+ or "RTE_MBUF_F_RX_L4_CKSUM_UNKNOWN" in line,
+ "Packet Rx L4 checksum valid-flags error!",
+ )
+ else:
+ if "RTE_MBUF_F_RX_L4_CKSUM" not in line:
+ self.verify(0, "There is no L4 checksum flags appeared!")
+ elif "RTE_MBUF_F_RX_IP_CKSUM" not in line:
+ self.verify(0, "There is no IP checksum flags appeared!")
+ else:
+ if flag == 1:
+ self.verify(
+ "RTE_MBUF_F_RX_L4_CKSUM_GOOD" in line,
+ "Packet Rx L4 checksum valid-flags error!",
+ )
+ self.verify(
+ "RTE_MBUF_F_RX_IP_CKSUM_GOOD" in line,
+ "Packet Rx IP checksum valid-flags error!",
+ )
+ elif flag == 0:
+ self.verify(
+ "RTE_MBUF_F_RX_L4_CKSUM_BAD" in line
+ or "RTE_MBUF_F_RX_L4_CKSUM_UNKNOWN" in line,
+ "Packet Rx L4 checksum valid-flags error!",
+ )
+ self.verify(
+ "RTE_MBUF_F_RX_IP_CKSUM_BAD" in line,
+ "Packet Rx IP checksum valid-flags error!",
+ )
+
+ self.dut.send_expect("stop", "testpmd>")
+
+ def checksum_validate(self, packets_sent, packets_expected):
+ """
+ Validate the checksum.
+ """
+ tx_interface = self.tester.get_interface(
+ self.tester.get_local_port(self.dut_ports[0])
+ )
+ rx_interface = self.tester.get_interface(
+ self.tester.get_local_port(self.dut_ports[0])
+ )
+
+ sniff_src = "52:00:00:00:00:00"
+ result = dict()
+
+ chksum = self.get_chksum_values(packets_expected)
+ inst = self.tester.tcpdump_sniff_packets(
+ intf=rx_interface,
+ count=len(packets_sent) * 4,
+ filters=[{"layer": "ether", "config": {"src": sniff_src}}],
+ )
+
+ self.pkt = packet.Packet()
+ for packet_type in list(packets_sent.keys()):
+ self.pkt.append_pkt(packets_sent[packet_type])
+ self.pkt.send_pkt(crb=self.tester, tx_port=tx_interface, count=4)
+
+ p = self.tester.load_tcpdump_sniff_packets(inst)
+ nr_packets = len(p)
+ print(p)
+ packets_received = [
+ p[i].sprintf("%IP.chksum%;%TCP.chksum%;%UDP.chksum%;%SCTP.chksum%")
+ for i in range(nr_packets)
+ ]
+ print(len(packets_sent), len(packets_received))
+ self.verify(
+ len(packets_sent) * 4 == len(packets_received), "Unexpected Packets Drop"
+ )
+
+ for packet_received in packets_received:
+ (
+ ip_checksum,
+ tcp_checksum,
+ udp_checksum,
+ sctp_checksum,
+ ) = packet_received.split(";")
+
+ packet_type = ""
+ l4_checksum = ""
+ if tcp_checksum != "??":
+ packet_type = "TCP"
+ l4_checksum = tcp_checksum
+ elif udp_checksum != "??":
+ packet_type = "UDP"
+ l4_checksum = udp_checksum
+ elif sctp_checksum != "??":
+ packet_type = "SCTP"
+ l4_checksum = sctp_checksum
+
+ if ip_checksum != "??":
+ packet_type = "IP/" + packet_type
+ if chksum[packet_type] != [ip_checksum, l4_checksum]:
+ result[packet_type] = packet_type + " checksum error"
+ else:
+ packet_type = "IPv6/" + packet_type
+ if chksum[packet_type] != [l4_checksum]:
+ result[packet_type] = packet_type + " checksum error"
+
+ return result
+
+ def send_scapy_packet(self, packet: str):
+ itf = self.tester.get_interface(self.tester.get_local_port(self.dut_ports[0]))
+
+ self.tester.scapy_foreground()
+ self.tester.scapy_append(f'sendp({packet}, iface="{itf}")')
+ return self.tester.scapy_execute()
+
+ def get_pkt_rx_l4_cksum(self, testpmd_output: str) -> bool:
+ return self.checksum_flags_are_good("RTE_MBUF_F_RX_L4_CKSUM_", testpmd_output)
+
+ def get_pkt_rx_ip_cksum(self, testpmd_output: str) -> bool:
+ return self.checksum_flags_are_good("RTE_MBUF_F_RX_IP_CKSUM_", testpmd_output)
+
+ def send_pkt_expect_good_bad_from_flag(
+ self, pkt_str: str, flag: str, test_name: str, should_pass: bool = True
+ ):
+ self.pmdout.get_output(timeout=1) # Remove any old output
+ self.scapy_exec(f"sendp({pkt_str}, iface=iface)")
+ testpmd_output: str = self.pmdout.get_output(timeout=1)
+ self.verify(
+ flag in testpmd_output,
+ f"Flag {flag[:-1]} not found for test {test_name}, please run test_rx_checksum_valid_flags.",
+ )
+ self.verify(
+ (flag + "UNKNOWN") not in testpmd_output,
+ f"Flag {flag[:-1]} was found to be unknown for test {test_name}, indicating a possible lack of support",
+ )
+ if should_pass:
+ if flag + "GOOD" in testpmd_output:
+ return
+ else: # flag + "BAD" in testpmd_output
+ self.verify(
+ False, f"{flag}BAD was found in output, expecting {flag}GOOD."
+ )
+ else:
+ if flag + "BAD" in testpmd_output:
+ return
+ else: # flag + "GOOD" in testpmd_output
+ self.verify(
+ False, f"{flag}GOOD was found in output, expecting {flag}BAD."
+ )
+
+ def send_pkt_expect_good_bad_from_flag_catch_failure(
+ self, pkt_str: str, flag: str, test_name: str, should_pass: bool = True
+ ) -> Union[VerifyFailure, None]:
+ try:
+ self.send_pkt_expect_good_bad_from_flag(
+ pkt_str, flag, test_name, should_pass=should_pass
+ )
+ except VerifyFailure as vf:
+ return vf
+
+ return None
+
+ def validate_checksum(self, pkt, pkt_type, inner_flag=False) -> bool:
+ """
+ @param pkt: The packet to validate the checksum of.
+ @return: Whether the checksum was valid.
+ """
+ if pkt is None:
+ return False
+ for i in range(0, len(l3_protos)):
+ if l3_protos[i] in pkt:
+ l3 = l3_protos[i]
+ for j in range(0, len(l4_protos)):
+ if l4_protos[j] in pkt:
+ layer = l4_proto_classes[j]
+ csum = pkt[layer].chksum
+ if csum is None:
+ csum = 0
+ del pkt[layer].chksum
+ # Converting it to raw will calculate the checksum
+ correct_csum = layer(bytes(Raw(pkt[layer]))).chksum
+ if correct_csum == csum:
+ # checksum value is correct
+ return False
+ else:
+ if inner_flag:
+ print(
+ "{} pkg[{}] VXLAN/{}/{} inner checksum {} is not correct {}".format(
+ pkt_type,
+ self.count,
+ l3,
+ l4_protos[j],
+ hex(correct_csum),
+ hex(csum),
+ )
+ )
+ else:
+ print(
+ "{} pkg[{}] {}/{} outer checksum {} is not correct {}".format(
+ pkt_type,
+ self.count,
+ l3,
+ l4_protos[j],
+ hex(correct_csum),
+ hex(csum),
+ )
+ )
+ return True
+ return False
+
+ def scapy_exec(self, cmd: str, timeout=1) -> str:
+ return self.tester.send_expect(cmd, ">>>", timeout=timeout)
+
+ def get_packets(self, dut_mac, tester_mac):
+ eth = Ether(dst=dut_mac, src=tester_mac)
+ packets = []
+ checksum_options = (
+ {},
+ {"chksum": 0xF},
+ )
+ # Untunneled
+ for l3 in l3_proto_classes:
+ for l4 in l4_proto_classes:
+ for chksum in checksum_options:
+ # The packet's data can be used to identify how the packet was constructed so avoid any issues with
+ # ordering
+ pkt = eth / l3() / l4(**chksum) / (b"X" * 48)
+ # Prevents the default behavior which adds DNS headers
+ if l4 == UDP:
+ pkt[UDP].dport, pkt[UDP].sport = 1001, 1001
+ packets.append(pkt)
+
+ # Tunneled
+ # VXLAN
+ for l3 in l3_proto_classes:
+ for l4 in l4_proto_classes:
+ for outer_arg in checksum_options:
+ for inner_arg in checksum_options:
+ pkt = (
+ eth
+ / l3()
+ / UDP(**outer_arg)
+ / VXLAN()
+ / Ether()
+ / l3()
+ / l4(**inner_arg)
+ / (b"Y" * 48)
+ )
+ # Prevents the default behavior which adds DNS headers
+ if l4 == UDP:
+ pkt[VXLAN][UDP].dport, pkt[VXLAN][UDP].sport = 1001, 1001
+ packets.append(pkt)
+ # GRE
+ for l3 in l3_proto_classes:
+ for l4 in l4_proto_classes:
+ for chksum in checksum_options:
+ pkt = eth / l3() / GRE() / l3() / l4(**chksum) / (b"Z" * 48)
+ # Prevents the default behavior which adds DNS headers
+ if l4 == UDP:
+ pkt[GRE][UDP].dport, pkt[GRE][UDP].sport = 1001, 1001
+ packets.append(pkt)
+
+ return packets
+
+ def send_tx_package(
+ self, packet_file_path, capture_file_path, packets, iface, dut_mac
+ ):
+ if os.path.isfile(capture_file_path):
+ os.remove(capture_file_path)
+ src_mac = "52:00:00:00:00:00"
+ self.tester.send_expect(
+ f"tcpdump -i '{iface}' ether src {src_mac} -s 0 -w {capture_file_path} -Q in &",
+ "# ",
+ )
+
+ if os.path.isfile(packet_file_path):
+ os.remove(packet_file_path)
+ wrpcap(packet_file_path, packets)
+ self.tester.session.copy_file_to(packet_file_path, packet_file_path)
+
+ # send packet
+ self.tester.send_expect("scapy", ">>>")
+ self.scapy_exec(f"packets = rdpcap('{packet_file_path}')")
+ for i in range(0, len(packets)):
+ self.scapy_exec(f"packets[{i}].show")
+ self.scapy_exec(f"sendp(packets[{i}], iface='{iface}')")
+ self.pmdout.get_output(timeout=0.5)
+ self.dut.send_expect(
+ "show port stats {}".format(self.dut_ports[0]), "testpmd>"
+ )
+ self.tester.send_expect("quit()", "# ")
+
+ time.sleep(1)
+ self.tester.send_expect("killall tcpdump", "#")
+ time.sleep(1)
+ self.tester.send_expect('echo "Cleaning buffer"', "#")
+ time.sleep(1)
+ return
+
+ def validate_packet_list_checksums(self, packets):
+ error_messages = []
+ untunnelled_error_message = (
+ f"un-tunneled checksum state for pkg[%s] with a invalid checksum."
+ )
+ vxlan_error_message = (
+ f"VXLAN tunnelled checksum state for pkg[%s] with a invalid checksum."
+ )
+ gre_error_message = (
+ f"GRE tunnelled checksum state for pkg[%s] with a invalid checksum."
+ )
+
+ for packet in packets:
+ self.count = self.count + 1
+ payload: str
+ # try:
+ payload = packet[Raw].load.decode("utf-8")
+ # # This error usually happens with tunneling protocols, and means that an additional cast is needed
+ # except UnicodeDecodeError:
+ # for proto in tunnelling_proto_classes:
+ if "X" in payload:
+ if self.validate_checksum(packet, "un-tunneled"):
+ error_messages.append(untunnelled_error_message % self.count)
+ elif "Y" in payload:
+ if self.validate_checksum(
+ packet[VXLAN][Ether], "VXLAN", inner_flag=True
+ ):
+ error_messages.append(vxlan_error_message % self.count)
+ # Intel® Ethernet 700 Series not support outer udp checksum
+ if self.is_eth_series_nic(700):
+ continue
+ if self.validate_checksum(packet, "VXLAN"):
+ error_messages.append(vxlan_error_message % self.count)
+ elif "Z" in payload:
+ if self.validate_checksum(packet, "GRE"):
+ error_messages.append(gre_error_message % self.count)
+ return error_messages
+
+ #
+ #
+ #
+ # Test Cases
+ #
+ def benchmark(self, lcore, ptype, mode, flow_format, size_list, nic):
+ """
+ Test ans report checksum offload performance for given parameters.
+ """
+ Bps = dict()
+ Pps = dict()
+ Pct = dict()
+ dmac = self.dut.get_mac_address(self.dut_ports[0])
+ dmac1 = self.dut.get_mac_address(self.dut_ports[1])
+
+ result = [2, lcore, ptype, mode]
+ for size in size_list:
+ flow = flow_format % (dmac, size)
+ pcap = os.sep.join([self.output_path, "test.pcap"])
+ self.tester.scapy_append('wrpcap("%s", [%s])' % (pcap, flow))
+ self.tester.scapy_execute()
+ flow = flow_format % (dmac1, size)
+ pcap = os.sep.join([self.output_path, "test1.pcap"])
+ self.tester.scapy_append('wrpcap("%s", [%s])' % (pcap, flow))
+ self.tester.scapy_execute()
+
+ tgenInput = []
+ pcap = os.sep.join([self.output_path, "test.pcap"])
+ tgenInput.append(
+ (
+ self.tester.get_local_port(self.dut_ports[0]),
+ self.tester.get_local_port(self.dut_ports[1]),
+ pcap,
+ )
+ )
+ pcap = os.sep.join([self.output_path, "test1.pcap"])
+ tgenInput.append(
+ (
+ self.tester.get_local_port(self.dut_ports[1]),
+ self.tester.get_local_port(self.dut_ports[0]),
+ pcap,
+ )
+ )
+
+ # clear streams before add new streams
+ self.tester.pktgen.clear_streams()
+ # create an instance to set stream field setting
+ # Moved here because it messes with the ability of the functional tests to use scapy.
+ self.pktgen_helper = PacketGeneratorHelper()
+ # run packet generator
+ streams = self.pktgen_helper.prepare_stream_from_tginput(
+ tgenInput, 100, None, self.tester.pktgen
+ )
+ Bps[str(size)], Pps[str(size)] = self.tester.pktgen.measure_throughput(
+ stream_ids=streams
+ )
+ self.verify(Pps[str(size)] > 0, "No traffic detected")
+ Pps[str(size)] /= 1e6
+ Pct[str(size)] = (Pps[str(size)] * 100) / self.wirespeed(self.nic, size, 2)
+
+ result.append(Pps[str(size)])
+ result.append(Pct[str(size)])
+
+ self.result_table_add(result)
+
+ def test_perf_checksum_throughtput(self):
+ """
+ Test checksum offload performance.
+ """
+ # Verify that enough ports are available
+ self.verify(len(self.dut_ports) >= 2, "Insufficient ports for testing")
+ self.dut.send_expect("quit", "#")
+
+ # sizes = [64, 128, 256, 512, 1024]
+ sizes = [64, 128]
+ pkts = {
+ "IP/UDP": 'Ether(dst="%s", src="52:00:00:00:00:00")/IP()/UDP()/("X"*(%d-46))',
+ "IP/TCP": 'Ether(dst="%s", src="52:00:00:00:00:00")/IP()/TCP()/("X"*(%d-58))',
+ "IP/SCTP": 'Ether(dst="%s", src="52:00:00:00:00:00")/IP()/SCTP()/("X"*(%d-50+2))',
+ "IPv6/UDP": 'Ether(dst="%s", src="52:00:00:00:00:00")/IPv6()/UDP()/("X"* (lambda x: x - 66 if x > 66 else 0)(%d))',
+ "IPv6/TCP": 'Ether(dst="%s", src="52:00:00:00:00:00")/IPv6()/TCP()/("X"* (lambda x: x - 78 if x > 78 else 0)(%d))',
+ }
+
+ if self.kdriver in DRIVER_TEST_LACK_CAPA["sctp_tx_offload"]:
+ del pkts["IP/SCTP"]
+
+ lcore = "1S/2C/1T"
+ portMask = utils.create_mask([self.dut_ports[0], self.dut_ports[1]])
+ for mode in ["sw", "hw"]:
+ self.logger.info("%s performance" % mode)
+ tblheader = ["Ports", "S/C/T", "Packet Type", "Mode"]
+ for size in sizes:
+ tblheader.append("%sB mpps" % str(size))
+ tblheader.append("%sB %% " % str(size))
+ self.result_table_create(tblheader)
+ self.pmdout.start_testpmd(
+ lcore,
+ "--portmask=%s" % self.portMask
+ + " --enable-rx-cksum "
+ + "--port-topology=loop",
+ socket=self.ports_socket,
+ )
+
+ self.dut.send_expect("set fwd csum", "testpmd> ")
+ if mode == "hw":
+ self.checksum_enablehw(self.dut_ports[0])
+ self.checksum_enablehw(self.dut_ports[1])
+ else:
+ self.checksum_enablesw(self.dut_ports[0])
+ self.checksum_enablesw(self.dut_ports[1])
+
+ self.dut.send_expect("start", "testpmd> ", 3)
+ for ptype in list(pkts.keys()):
+ self.benchmark(lcore, ptype, mode, pkts[ptype], sizes, self.nic)
+
+ self.dut.send_expect("stop", "testpmd> ")
+ self.dut.send_expect("quit", "#", 10)
+ self.result_table_print()
+
+ def tear_down(self):
+ """
+ Run after each test case.
+ """
+ self.dut.send_expect("quit", "#")
+
+ def tear_down_all(self):
+ """
+ Run after each test suite.
+ """
+ pass
--
2.17.1
^ permalink raw reply [flat|nested] 9+ messages in thread
* [dts][PATCH V3 8/8] conf/test_case_checklist: split performance plan and suite
2023-01-10 16:44 [dts][PATCH V3 0/8] split performance plan and suite Hongbo Li
` (6 preceding siblings ...)
2023-01-10 16:44 ` [dts][PATCH V3 7/8] tests/checksum_offload: " Hongbo Li
@ 2023-01-10 16:44 ` Hongbo Li
7 siblings, 0 replies; 9+ messages in thread
From: Hongbo Li @ 2023-01-10 16:44 UTC (permalink / raw)
To: dts; +Cc: Hongbo Li
add performance plan in index.rst
Signed-off-by: Hongbo Li <hongbox.li@intel.com>
---
test_plans/index.rst | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/test_plans/index.rst b/test_plans/index.rst
index f56f99a8..b7d42895 100644
--- a/test_plans/index.rst
+++ b/test_plans/index.rst
@@ -116,6 +116,12 @@ The following are the test plans for the DPDK DTS automated test system.
ntb_test_plan
nvgre_test_plan
perf_virtio_user_loopback_test_plan
+ perf_efd_test_plan
+ perf_ipfrag_test_plan
+ perf_l2fwd_test_plan
+ perf_multiprocess_test_plan
+ perf_tso_test_plan
+ perf_vxlan_test_plan
pf_smoke_test_plan
pipeline_test_plan
pvp_virtio_user_multi_queues_port_restart_test_plan
--
2.17.1
^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2023-01-10 8:25 UTC | newest]
Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-01-10 16:44 [dts][PATCH V3 0/8] split performance plan and suite Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 1/8] tests/efd: " Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 2/8] tests/l2fwd: " Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 3/8] tests/tso: " Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 4/8] tests/vxlan: " Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 5/8] tests/ipfrag: " Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 6/8] tests/multiprocess: " Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 7/8] tests/checksum_offload: " Hongbo Li
2023-01-10 16:44 ` [dts][PATCH V3 8/8] conf/test_case_checklist: " Hongbo Li
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).