test suite reviews and discussions
 help / color / mirror / Atom feed
From: "Lu, Nannan" <nannan.lu@intel.com>
To: "Zeng, XiaoxiaoX" <xiaoxiaox.zeng@intel.com>,
	"dts@dpdk.org" <dts@dpdk.org>
Cc: "Zeng, XiaoxiaoX" <xiaoxiaox.zeng@intel.com>
Subject: Re: [dts] [PATCH	V1]tests/cvl_switch_filter:optimization_cvl_switch_filter
Date: Mon, 8 Jun 2020 03:00:45 +0000	[thread overview]
Message-ID: <69fa936677e34758a218eafa6c9321ff@intel.com> (raw)
In-Reply-To: <20200605144711.10375-1-xiaoxiaox.zeng@intel.com>

Comments in line.

> -----Original Message-----
> From: dts [mailto:dts-bounces@dpdk.org] On Behalf Of Zeng Xiaoxiao
> Sent: Friday, June 5, 2020 10:47 PM
> To: dts@dpdk.org
> Cc: Zeng, XiaoxiaoX <xiaoxiaox.zeng@intel.com>
> Subject: [dts] [PATCH V1]tests/cvl_switch_filter:optimization_cvl_switch_filter
> 
> *.Avoid that with different dpdk code, packet default queue same as queue that
> rule set, realize dynamic set rule in testpmd.
> 
> Signed-off-by: Zeng Xiaoxiao <xiaoxiaox.zeng@intel.com>
> ---
>  tests/TestSuite_cvl_switch_filter.py | 68 +++++++++++++++++++++++++++-
>  1 file changed, 67 insertions(+), 1 deletion(-)
> 
> diff --git a/tests/TestSuite_cvl_switch_filter.py
> b/tests/TestSuite_cvl_switch_filter.py
> index ffd9af4..9731c37 100644
> --- a/tests/TestSuite_cvl_switch_filter.py
> +++ b/tests/TestSuite_cvl_switch_filter.py
> @@ -34,6 +34,7 @@ import json
>  import time
>  import re
>  import copy
> +import random
> 
>  from test_case import TestCase
>  from pmd_output import PmdOutput
> @@ -2571,6 +2572,35 @@ class SwitchFilterTest(TestCase):
>          result_flag, log_msg = dic["check_func"]["func"](out,
> dic["check_func"]["param"], dic["expect_results"])
>          return result_flag, log_msg
> 
> +    def send_packet_get_queue(self, dic):
> +        """
> +        general packets processing workflow.
> +        """
> +        self.dut.send_expect("start", "testpmd> ")
> +        # send packets
> +        for per_packet in dic["scapy_str"]:
> +            pkt = Packet(pkt_str=per_packet)
> +            pkt.send_pkt(self.tester, tx_port=self.__tx_iface, count=1)
> +        out = self.dut.send_expect("stop", "testpmd> ")
> +        p = re.compile(r"Forward Stats for RX Port= \d+/Queue=(\s?\d+)")
> +        res = p.findall(out)
> +        default_queue = [int(i) for i in res]
> +        return default_queue
> +
> +    def get_available_queue_num(self, default_queue):
> +        """
> +        general packets processing workflow.
> +        """
> +        queue_list = [1, 2, 3, 4, 5, 6]
> +        for q in default_queue:
> +            if q in queue_list:
> +                queue_list.remove(q)
> +            if (q - 1) in queue_list:
> +                queue_list.remove(q - 1)
> +
> +        set_queue = random.choice(queue_list)
> +        return int(set_queue)
> +

Need a more flexible way to get available queues, need to support queue number 4, 8, etc.

>      def save_results(self, pattern_name, flag, result_flag, log_msg,
> overall_result):
>          """
>          save results to dictionary: test_results.
> @@ -2598,10 +2628,46 @@ class SwitchFilterTest(TestCase):
>          overall_result = True
>          test_results.clear()
>          for tv in test_vectors:
> +            # get packet default_queue number
> +            mismatched_dic = tv["mismatched"]
> +            default_queue = self.send_packet_get_queue(mismatched_dic)
> +
> +            # check if default_queue same with expect_queue
> +            expect_queue =
> tv["mismatched"]["check_func"]["param"]["expect_queues"]
> +            if expect_queue != "null":
> +                if isinstance(expect_queue, int):
> +                    eq_list = []
> +                    eq_list.append(expect_queue)
> +                elif isinstance(expect_queue, list):
> +                    eq_list = expect_queue
> +                recover_flag = list(set(eq_list) & set(default_queue))
> +            else:
> +                recover_flag = None
> +
> +            # if default_queue has same one with expect_queue, recover rule
> +            if recover_flag:
> +                # exclude defult_queue number and get set_queue
> +                set_queue_num = self.get_available_queue_num(default_queue)
> +                # recover rule command and check queue
> +                if isinstance(expect_queue, int):
> +                    rule_command = tv["rte_flow_pattern"].replace("/ end actions
> queue index %s" % str(expect_queue),
> +                                                                  "/ end actions queue index %s" %
> str(set_queue_num))
> +                    tv["matched"]["check_func"]["param"]["expect_queues"] =
> set_queue_num
> +                    tv["mismatched"]["check_func"]["param"]["expect_queues"] =
> set_queue_num
> +                elif isinstance(expect_queue, list):
> +                    q = [str(i) for i in expect_queue]
> +                    expect_queue_str = " ".join(q)
> +                    rule_command = tv["rte_flow_pattern"].replace("/ end actions rss
> queues %s" % expect_queue_str,
> +                                                                  "/ end actions rss queues %s" %
> (str(set_queue_num) + " " + str(set_queue_num + 1)))

Not only need to support 2 queues, but also need to be able to support more queue numbers in rule.

> +                    tv["matched"]["check_func"]["param"]["expect_queues"] =
> [set_queue_num, set_queue_num + 1]
> +                    tv["mismatched"]["check_func"]["param"]["expect_queues"] =
> [set_queue_num, set_queue_num + 1]
> +            else:
> +                rule_command = tv["rte_flow_pattern"]
> +
>              pattern_name = tv["name"]
>              test_results[pattern_name] = OrderedDict()
> 
> -            out = self.dut.send_expect(tv["rte_flow_pattern"], "testpmd> ", 15)
> #create a rule
> +            out = self.dut.send_expect(rule_command, "testpmd> ", 15)
> + #create a rule
>              #get the rule number
>              rule_num = self.get_rule_number(out)
> 
> --
> 2.17.1


      parent reply	other threads:[~2020-06-08  3:00 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-05 14:47 Zeng Xiaoxiao
2020-06-05  6:39 ` Zeng, XiaoxiaoX
2020-06-08  3:00 ` Lu, Nannan [this message]

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=69fa936677e34758a218eafa6c9321ff@intel.com \
    --to=nannan.lu@intel.com \
    --cc=dts@dpdk.org \
    --cc=xiaoxiaox.zeng@intel.com \
    /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).