All of lore.kernel.org
 help / color / mirror / Atom feed
From: bugzilla@dpdk.org
To: dev@dpdk.org
Subject: [dpdk-dev] [Bug 674] [dpdk-21.05] nic_single_core_perf/test_perf_nic_single_core_avx2: Throughput dropped about 8% when ues send 64 byte packets on 1C2T cores config.
Date: Wed, 14 Apr 2021 02:12:06 +0000	[thread overview]
Message-ID: <bug-674-3@http.bugs.dpdk.org/> (raw)

https://bugs.dpdk.org/show_bug.cgi?id=674

            Bug ID: 674
           Summary: [dpdk-21.05]
                    nic_single_core_perf/test_perf_nic_single_core_avx2:
                    Throughput dropped about 8% when ues send 64 byte
                    packets on 1C2T cores config.
           Product: DPDK
           Version: unspecified
          Hardware: All
                OS: All
            Status: UNCONFIRMED
          Severity: normal
          Priority: Normal
         Component: testpmd
          Assignee: dev@dpdk.org
          Reporter: weix.ling@intel.com
  Target Milestone: ---

Description

DPDK version: 
 21.05.0-rc0 4113ddd45293d7b26ff4033bfd86cef03d29124f
Other software versions: N/A.
OS: Ubuntu 20.04.2 LTS/Linux 5.8.0-48-generic
Compiler: gcc version 10.2.0 (Ubuntu 10.2.0-5ubuntu1~20.04)
Hardware platform: Intel(R) Xeon(R) Platinum 8280M CPU @ 2.70GHz.
NIC hardware: Ethernet Controller E810-C for QSFP 1592.
NIC firmware: 2.30 0x80004dd0 1.2839.0
NIC driver: ice-0.8.2-k

Test Setup
Steps to reproduce
List the steps to reproduce the issue.

1.Bind NIC to DPDK
usertools/dpdk-devbind.py --force --bind=vfio-pci 0000:18:00.0

2.Compile DPDK
rm -rf x86_64-native-linuxapp-gcc
CC=gcc meson -Denable_kmods=True -Dlibdir=lib  --default-library=static
x86_64-native-linuxapp-gcc
ninja -C x86_64-native-linuxapp-gcc -j 32

3.Start testpmd
x86_64-native-linuxapp-gcc/app/dpdk-testpmd -l 5,6,62 -n 6 
--file-prefix=dpdk_2066596_20210413175519  --force-max-simd-bitwidth=256  -a
0000:18:00.0 -- -i --portmask=0x1 --rxq=2 --txq=2 --txd=1024 --rxd=1024
--nb-cores=2
show port info all
start

4.Use IXIA to sent IP packets
[Ether(src="52:00:00:00:00:00")/IP(src="1.2.3.4",dst="192.18.1.0")/("X"*26)]

5.Get the Throughput on 1C/2T cores, 64Byte framesize 1024 TXD/RXD config.
Show the output from the previous commands.

+----------+------------+---------+-------------+---------+---------------------+-----------------------+
| Fwd_core | Frame Size | TXD/RXD | Throughput  |  Rate   | Expected Throughput
| Throughput Difference |
+==========+============+=========+=============+=========+=====================+=======================+
| 1C/1T    | 64         | 1024    | 77.595 Mpps | 52.144% | 76.069 Mpps        
| 1.526 Mpps            |
+----------+------------+---------+-------------+---------+---------------------+-----------------------+
| 1C/1T    | 128        | 1024    | 76.071 Mpps | 90.068% | 75.420 Mpps        
| 0.651 Mpps            |
+----------+------------+---------+-------------+---------+---------------------+-----------------------+
| 1C/2T    | 64         | 1024    | 88.346 Mpps | 59.369% | 90.638 Mpps        
| -2.292 Mpps           |
+----------+------------+---------+-------------+---------+---------------------+-----------------------+
| 1C/2T    | 128        | 1024    | 74.002 Mpps | 87.619% | 73.972 Mpps        
| 0.030 Mpps            |
+----------+------------+---------+-------------+---------+---------------------+-----------------------+
Expected Result
Explain what is the expected result in text or as an example output:

+----------+------------+---------+-------------+---------+---------------------+-----------------------+
| Fwd_core | Frame Size | TXD/RXD | Throughput  |  Rate   | Expected Throughput
| Throughput Difference |
+==========+============+=========+=============+=========+=====================+=======================+
| 1C/1T    | 64         | 1024    | 75.834 Mpps | 50.961% | 76.069 Mpps        
| -0.235 Mpps           |
+----------+------------+---------+-------------+---------+---------------------+-----------------------+
| 1C/1T    | 128        | 1024    | 74.528 Mpps | 88.241% | 75.420 Mpps        
| -0.892 Mpps           |
+----------+------------+---------+-------------+---------+---------------------+-----------------------+
| 1C/2T    | 64         | 1024    | 96.322 Mpps | 64.728% | 90.638 Mpps        
| 5.684 Mpps            |
+----------+------------+---------+-------------+---------+---------------------+-----------------------+
| 1C/2T    | 128        | 1024    | 74.178 Mpps | 87.827% | 73.972 Mpps        
| 0.206 Mpps            |
+----------+------------+---------+-------------+---------+---------------------+-----------------------+
Regression
Is this issue a regression: (Y/N) Y

Version the regression was introduced: 

Bad commit:

commit 3d2913c67c83cca05a009ebbca3d11a5586d6117
Author: Tal Shnaiderman <talshn@nvidia.com>
Date: Tue Mar 16 15:28:18 2021 +0200

eal: add error numbers in thread TLS API

Add error number reporting to rte_errno in all
functions in the rte_thread_tls_* API.

Suggested-by: Anatoly Burakov <anatoly.burakov@intel.com>
Signed-off-by: Tal Shnaiderman <talshn@nvidia.com>
Acked-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>

Stack Trace or Log
CPU info:
root@dpdk-xuemin4x-cvl100g:~/dpdk# usertools/cpu_layout.py
======================================================================
Core and Socket Information (as reported by '/sys/devices/system/cpu')
======================================================================cores = 
[0, 1, 2, 3, 4, 5, 6, 8, 9, 10, 11, 12, 13, 14, 16, 17, 18, 19, 20, 21, 22, 24,
25, 26, 27, 28, 29, 30]
sockets =  [0, 1]        Socket 0          Socket 1
        --------          --------
Core 0  [0, 56]           [28, 84]
Core 1  [1, 57]           [29, 85]
Core 2  [2, 58]           [30, 86]
Core 3  [3, 59]           [31, 87]
Core 4  [4, 60]           [32, 88]
Core 5  [5, 61]           [33, 89]
Core 6  [6, 62]           [34, 90]
Core 8  [7, 63]           [35, 91]
Core 9  [8, 64]           [36, 92]
Core 10 [9, 65]           [37, 93]
Core 11 [10, 66]          [38, 94]
Core 12 [11, 67]          [39, 95]
Core 13 [12, 68]          [40, 96]
Core 14 [13, 69]          [41, 97]
Core 16 [14, 70]          [42, 98]
Core 17 [15, 71]          [43, 99]
Core 18 [16, 72]          [44, 100]
Core 19 [17, 73]          [45, 101]
Core 20 [18, 74]          [46, 102]
Core 21 [19, 75]          [47, 103]
Core 22 [20, 76]          [48, 104]
Core 24 [21, 77]          [49, 105]
Core 25 [22, 78]          [50, 106]
Core 26 [23, 79]          [51, 107]
Core 27 [24, 80]          [52, 108]
Core 28 [25, 81]          [53, 109]
Core 29 [26, 82]          [54, 110]
Core 30 [27, 83]          [55, 111]

-- 
You are receiving this mail because:
You are the assignee for the bug.

                 reply	other threads:[~2021-04-14  2:16 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=bug-674-3@http.bugs.dpdk.org/ \
    --to=bugzilla@dpdk.org \
    --cc=dev@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.