From: Xinfeng Zhao <xinfengx.zhao@intel.com>
To: dts@dpdk.org
Cc: Xinfeng Zhao <xinfengx.zhao@intel.com>
Subject: [dts] [PATCH V1] tests: add the suite of cryptodev performance test
Date: Wed, 16 Jan 2019 14:35:49 +0800 [thread overview]
Message-ID: <1547620549-4538-1-git-send-email-xinfengx.zhao@intel.com> (raw)
add the script of cryptodev performance test to tests
Signed-off-by: Xinfeng Zhao <xinfengx.zhao@intel.com>
---
tests/TestSuite_cryptodev_perf_crypto-perf.py | 484 ++++++++++++++++++++++++++
1 file changed, 484 insertions(+)
create mode 100644 tests/TestSuite_cryptodev_perf_crypto-perf.py
diff --git a/tests/TestSuite_cryptodev_perf_crypto-perf.py b/tests/TestSuite_cryptodev_perf_crypto-perf.py
new file mode 100644
index 0000000..5a50747
--- /dev/null
+++ b/tests/TestSuite_cryptodev_perf_crypto-perf.py
@@ -0,0 +1,484 @@
+# BSD LICENSE
+#
+# Copyright(c) 2016-2017 Intel Corporation. All rights reserved.
+# All rights reserved.
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions
+# are met:
+#
+# * Redistributions of source code must retain the above copyright
+# notice, this list of conditions and the following disclaimer.
+# * Redistributions in binary form must reproduce the above copyright
+# notice, this list of conditions and the following disclaimer in
+# the documentation and/or other materials provided with the
+# distribution.
+# * Neither the name of Intel Corporation nor the names of its
+# contributors may be used to endorse or promote products derived
+# from this software without specific prior written permission.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+import re
+import json
+from test_case import TestCase
+import TestSuite_cryptodev_common as cc
+
+
+class PerfTestsCryptodev(TestCase):
+
+ def set_up_all(self):
+ self._perf_result = {}
+
+ self._default_crypto_perf_opts = {
+ "ptest": "throughput",
+ "silent": "",
+ "pool-sz": None,
+ "total-ops": "1000000",
+ "burst-sz": "32",
+ "buffer-sz": "64",
+ "devtype": None,
+ "segments-nb": None,
+ "optype": "cipher-then-auth",
+ "sessionless": None,
+ "out-of-place": None,
+ "test-file": None,
+ "test-name": None,
+ "cipher-algo": None,
+ "cipher-op": None,
+ "cipher-key-sz": None,
+ "cipher-iv-sz": None,
+ "auth-algo": None,
+ "auth-op": None,
+ "auth-key-sz": None,
+ "auth-iv-sz": None,
+ "aead-algo": None,
+ "aead-op": None,
+ "aead-key-sz": None,
+ "aead-iv-sz": None,
+ "aead-aad-sz": None,
+ "digest-sz": None,
+ "csv-friendly": None
+ }
+ self._app_path = "./x86_64-native-linuxapp-gcc/build/app/test-crypto-perf/dpdk-test-crypto-perf"
+
+ if not cc.is_build_skip(self):
+ cc.build_dpdk_with_cryptodev(self)
+
+ self.vf_driver = self.get_suite_cfg()['vf_driver']
+ cc.bind_qat_device(self, self.vf_driver)
+
+ def tear_down_all(self):
+ cc.clear_dpdk_config(self)
+
+ if not self._perf_result:
+ return
+
+ with open(self.logger.log_path + "/" + "perf_cryptodev_result.json", "w") as fv:
+ json.dump(self._perf_result, fv, indent=4, encoding="utf-8")
+
+ def set_up(self):
+ pass
+
+ def tear_down(self):
+ self.dut.kill_all()
+
+ def test_qat_aes_cbc_sha1_hmac(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self, {"vdev":None})
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "qat_aes_cbc_sha1_hmac")
+ self.verify(result, "Test failed")
+
+ def test_sw_aes_cbc_sha1_hmac(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self, {"w":None})
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "sw_aes_cbc_sha1_hmac")
+ self.verify(result, "Test failed")
+
+ def test_qat_aes_cbc_sha2_hmac(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self, {"vdev":None})
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "qat_aes_cbc_sha2_hmac")
+ self.verify(result, "Test failed")
+
+ def test_sw_aes_cbc_sha2_hmac(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self, {"w":None})
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "sw_aes_cbc_sha2_hmac")
+ self.verify(result, "Test failed")
+
+ def test_qat_aes_gcm(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self, {"vdev":None})
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "qat_aes_gcm")
+ self.verify(result, "Test failed")
+
+ def test_sw_aes_gcm(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self, {"w":None})
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "sw_aes_gcm")
+ self.verify(result, "Test failed")
+
+ def test_qat_kasumi(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self, {"vdev":None})
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "qat_kasumi")
+ self.verify(result, "Test failed")
+
+ def test_sw_kasumi(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self, {"w":None})
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "sw_kasumi")
+ self.verify(result, "Test failed")
+
+ def test_qat_snow3g(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self, {"vdev":None})
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "qat_snow3g")
+ self.verify(result, "Test failed")
+
+ def test_sw_snow3g(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self, {"w":None})
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "sw_snow3g")
+ self.verify(result, "Test failed")
+
+ def test_qat_zuc(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self, {"vdev":None})
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "qat_zuc")
+ self.verify(result, "Test failed")
+
+ def test_sw_zuc(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self, {"w":None})
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "sw_zuc")
+ self.verify(result, "Test failed")
+
+ def test_scheduler_aes_cbc_sha1_hmac(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self)
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "scheduler_aes_cbc_sha1_hmac")
+ self.verify(result, "Test failed")
+
+ def test_scheduler_aes_cbc_sha2_hmac(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self)
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "scheduler_aes_cbc_sha2_hmac")
+ self.verify(result, "Test failed")
+
+ def test_scheduler_aes_gcm(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self)
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "scheduler_aes_gcm")
+ self.verify(result, "Test failed")
+
+ def test_scheduler_kasumi(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self)
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "scheduler_kasumi")
+ self.verify(result, "Test failed")
+
+ def test_scheduler_snow3g(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self)
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "scheduler_snow3g")
+ self.verify(result, "Test failed")
+
+ def test_scheduler_zuc(self):
+ if cc.is_test_skip(self):
+ return
+
+ eal_opt_str = cc.get_eal_opt_str(self)
+ crypto_perf_opt_str = self._get_crypto_perf_opt_str()
+ result = self._run_crypto_perf_throughput(eal_opt_str,
+ crypto_perf_opt_str,
+ "scheduler_zuc")
+ self.verify(result, "Test failed")
+
+ # Private functions
+ def _get_crypto_perf_opt_str(self, override_crypto_perf_opts={}):
+ return cc.get_opt_str(self, self._default_crypto_perf_opts,
+ override_crypto_perf_opts)
+
+ def _parse_output(self, output):
+ try:
+ lines = output.split("\r\n")
+ line_nb = len(lines)
+ self.logger.debug("Total output lines: " + str(line_nb))
+
+ # Find data line
+ for line_index in range(line_nb):
+ if lines[line_index].startswith(" lcore id"):
+ self.logger.debug("data output line from: " + str(line_index))
+ break
+ data_line = line_index + 2
+
+ results = []
+ pattern = re.compile(r'\s+')
+ for line in lines[data_line:-1]:
+ result = {}
+ result_list = pattern.split(line.strip(" "))
+ if result_list[0] == "lcore id" or result_list[0] == "lcore" or not result_list[0]:
+ continue
+ result["lcore_id"] = int(result_list[0])
+ result["buf_size"] = int(result_list[1])
+ result["burst_size"] = int(result_list[2])
+ result["enqueue"] = int(result_list[3])
+ result["dequeue"] = int(result_list[4])
+ result["enqueue_failures"] = int(result_list[5])
+ result["dequeue_failures"] = int(result_list[6])
+ result["mops"] = float(result_list[7])
+ result["gbps"] = float(result_list[8])
+ result["cycle_buf"] = float(result_list[9])
+ results.append(result)
+
+ self.logger.debug(results)
+ return results
+ except Exception, ex:
+ self.logger.error(ex)
+ raise ex
+
+ def _stat_results_by_buf_size(self, results):
+ stats_results = {}
+ for result in results:
+ buf_size = result["buf_size"]
+ if buf_size in stats_results:
+ stats_results[buf_size]["lcore_id"] = \
+ str(stats_results[buf_size]["lcore_id"]) \
+ + ":" + str(result["lcore_id"])
+ stats_results[buf_size]["enqueue"] = \
+ stats_results[buf_size]["enqueue"] + \
+ result["enqueue"]
+ stats_results[buf_size]["enqueue_failures"] = \
+ stats_results[buf_size]["enqueue_failures"] + \
+ result["enqueue_failures"]
+ stats_results[buf_size]["dequeue_failures"] = \
+ stats_results[buf_size]["dequeue_failures"] + \
+ result["dequeue_failures"]
+ stats_results[buf_size]["mops"] = \
+ stats_results[buf_size]["mops"] + \
+ result["mops"]
+ stats_results[buf_size]["gbps"] = \
+ stats_results[buf_size]["gbps"] + \
+ result["gbps"]
+ stats_results[buf_size]["cycle_buf"] = \
+ stats_results[buf_size]["cycle_buf"] + \
+ result["cycle_buf"]
+ stats_results[buf_size]["nr"] = stats_results[buf_size]["nr"] + 1
+ else:
+ stats_results[buf_size] = result
+ stats_results[buf_size]["nr"] = 1
+ self.logger.debug(stats_results)
+ return stats_results
+
+ def _run_crypto_perf(self, eal_opt_str, crypto_perf_opt_str):
+ cmd_str = cc.get_dpdk_app_cmd_str(self._app_path,
+ eal_opt_str,
+ crypto_perf_opt_str)
+ self.logger.info(cmd_str)
+ try:
+ out = self.dut.send_expect(cmd_str, "#", 600)
+ except Exception, ex:
+ self.logger.error(ex)
+ raise ex
+
+ results = self._parse_output(out)
+
+ return results
+
+ def _get_core_and_thread_num(self):
+ cpu_info ={}
+ out = self.dut.send_expect("lscpu", "#")
+ for each_line in out.split('\n'):
+ key, value = each_line.split(':')
+ cpu_info[key] = value.strip()
+
+ core, thread = 0, 0
+ lcores = self.get_case_cfg()["l"].split(",")
+ for lcore in lcores:
+ if int(lcore) < int(cpu_info['Core(s) per socket']) * int(cpu_info['Socket(s)']):
+ core =+1
+ thread =+1
+ elif int(lcore) < int(cpu_info['CPU(s)']):
+ thread =+1
+ return core, thread
+
+ def _run_crypto_perf_throughput(self, eal_opt_str, crypto_perf_opt_str,
+ case_name):
+ c_num, t_num = self._get_core_and_thread_num()
+ try:
+ results = self._run_crypto_perf(eal_opt_str, crypto_perf_opt_str)
+ if not results:
+ return False
+ stats_results = self._stat_results_by_buf_size(results)
+ json_result = []
+ for buf_size, values in stats_results.items():
+ status = "PASS"
+ # delta, status
+ self.accepted_gap = self.get_suite_cfg()['accepted_tolerance']
+ self.expected_throughput =\
+ self.get_suite_cfg()['expected_throughput'][case_name][buf_size]
+ delta = (values["gbps"] - self.expected_throughput)/self.expected_throughput
+ if abs(delta) > self.accepted_gap:
+ status = "FAIL"
+
+ perf_info={
+ "status": status,
+ "performance":
+ [
+ {
+ "name": "throughput",
+ "value": values["gbps"],
+ "unit": "Gbps",
+ "delta": delta
+ },
+ {
+ "name":"failed_enq",
+ "unit": "ops",
+ "value": values["enqueue_failures"]
+ },
+ {
+ "name":"failed_deq",
+ "unit": "ops",
+ "value": values["dequeue_failures"]
+ },
+ {
+ "name":"throughput_mops",
+ "unit": "Mops",
+ "value": values["mops"]
+ },
+ {
+ "name":"cycle_buf",
+ "unit": "Cycles",
+ "value": values["cycle_buf"]/values["nr"]
+ },
+ ],
+ "parameters":
+ [
+ {
+ "name": "core_num/thread_num",
+ "unit": "num/num",
+ "value": "{}/{}".format(c_num, t_num)
+ },
+ {
+ "name":"frame_size",
+ "unit": "bytes",
+ "value": buf_size
+ },
+ {
+ "name":"burst_size",
+ "unit": "bytes",
+ "value": values["burst_size"]
+ },
+ {
+ "name":"total_ops",
+ "unit": "ops",
+ "value": values["enqueue"]
+ },
+ ]
+ }
+
+ self.logger.debug(perf_info)
+ json_result.append(perf_info)
+ self._perf_result[case_name] = json_result
+ self.logger.debug(self._perf_result)
+ return True
+ except Exception, ex:
+ self.logger.error(ex)
+ return False
--
2.7.4
next reply other threads:[~2019-01-16 6:32 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-16 6:35 Xinfeng Zhao [this message]
2019-01-17 8:45 ` Zhao, XinfengX
2019-01-21 2:54 ` Chen, Zhaoyan
2019-01-21 3:26 ` Tu, Lijuan
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1547620549-4538-1-git-send-email-xinfengx.zhao@intel.com \
--to=xinfengx.zhao@intel.com \
--cc=dts@dpdk.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).