Bug 1571545

Summary: NFP nic 1500 byte rfc 2544 throughput test shows very low results
Product: Red Hat Enterprise Linux 7 Reporter: liting <tli>
Component: openvswitchAssignee: Open vSwitch development team <ovs-team>
Status: CLOSED NOTABUG QA Contact: ovs-qe
Severity: medium Docs Contact:
Priority: medium    
Version: 7.5CC: atragler, ctrautma
Target Milestone: rc   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2018-05-09 03:44:01 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:

Description liting 2018-04-25 06:10:12 UTC
Description of problem:

Version-Release number of selected component (if applicable):

How reproducible:

Steps to Reproduce:
1. Run vsperf pvp 1q/2pmd case with 1500bytes packet, loss rate configured 0 
2. Used 25g xena sender do the rfc 2544 throughput test

Actual results:
1q/2pmd 1500bytes case got 0.04-0.1mpps not was 1.6mpps, it was very low.
1q/2pmd 64bytes case can got 3mpps. but 1500 bytes always drops 1 packet occasionally.

Expected results:
1500 bytes case should not drop packet and got about 1.6mpps.

Additional info:
The 25g NFP nic detail info
[root@netqe14 ~]# ethtool -i enp130s0np0
driver: nfp
version: 3.10.0-862.el7.x86_64 SMP mod_u
firmware-version: 0.22 nic-2.0.4 nic
bus-info: 0000:82:00.0
supports-statistics: yes
supports-test: no
supports-eeprom-access: no
supports-register-dump: yes
supports-priv-flags: no

[root@netqe14 ~]# lspci -s 0000:82:00.0
82:00.0 Ethernet controller: Netronome Systems, Inc. Device 4000

Comment 2 Christian Trautman 2018-04-25 18:56:22 UTC
This could be a cable problem or a problem with our layer 1 switch. I need time to investigate this further.