I have the performance issue of ConnectX-4 Lx on ubuntu 20.04LTS, kernel 5.4.0-84-generic. Could you please help us for fix the issue?.

I have the performance issue of ConnectX-4 Lx on ubuntu 20.04LTS, kernel 5.4.0-84-generic. The scenario is Server-01 direct connect to Server-02 single port 25G via Juniper Switch (Layer 2). Then we test the performance with iperf3 get the result is bandwidth not stable (Throughput not maximum 25G). We checked the configure on Juniper switch is normally. Which we upgrade the kernel from 5.4 to 5.11 and try to test again have the same issue. Then we try test as same as the Server-01 and Server-02 with the Intel Card (XXV710) to get the result is stable (24.7G - 24.8G). Could you please help us for fix the issue?.

More Informantion :

Server-01:~# modinfo mlx5_core

filename: /lib/modules/5.4.0-84-generic/updates/dkms/mlx5_core.ko

alias: auxiliary:mlx5_core.eth-rep

alias: auxiliary:mlx5_core.eth

version: 5.4-1.0.3

license: Dual BSD/GPL

description: Mellanox 5th generation network adapters (ConnectX series) core driver

author: Eli Cohen eli@mellanox.com

srcversion: 7B67674E4C53E7F9FEBDFFF

alias: pci:v000015B3d0000A2DCsvsdbcsci*

alias: pci:v000015B3d0000A2D6svsdbcsci*

alias: pci:v000015B3d0000A2D3svsdbcsci*

alias: pci:v000015B3d0000A2D2svsdbcsci*

alias: pci:v000015B3d00001021svsdbcsci*

alias: pci:v000015B3d0000101Fsvsdbcsci*

alias: pci:v000015B3d0000101Esvsdbcsci*

alias: pci:v000015B3d0000101Dsvsdbcsci*

alias: pci:v000015B3d0000101Csvsdbcsci*

alias: pci:v000015B3d0000101Bsvsdbcsci*

alias: pci:v000015B3d0000101Asvsdbcsci*

alias: pci:v000015B3d00001019svsdbcsci*

alias: pci:v000015B3d00001018svsdbcsci*

alias: pci:v000015B3d00001017svsdbcsci*

alias: pci:v000015B3d00001016svsdbcsci*

alias: pci:v000015B3d00001015svsdbcsci*

alias: pci:v000015B3d00001014svsdbcsci*

alias: pci:v000015B3d00001013svsdbcsci*

depends: mlx_compat,tls,auxiliary,mlxdevm,pci-hyperv-intf,mlxfw,psample

retpoline: Y

name: mlx5_core

vermagic: 5.4.0-84-generic SMP mod_unload modversions

parm: guids:charp

parm: node_guid:guids configuration. This module parameter will be obsolete!

parm: num_of_groups:Eswitch offloads number of big groups in FDB table. Valid range 1 - 1024. Default 15 (uint)

parm: debug_mask:debug mask: 1 = dump cmd data, 2 = dump cmd exec time, 3 = both. Default=0 (uint)

parm: prof_sel:profile selector. Valid range 0 - 3 (uint)

parm: probe_vf:probe VFs or not, 0 = not probe, 1 = probe. Default = 1 (bool)

Server-01:~# uname -r


Server-01:~# lscpu

Architecture: x86_64

CPU op-mode(s): 32-bit, 64-bit

Byte Order: Little Endian

Address sizes: 43 bits physical, 48 bits virtual

CPU(s): 192

On-line CPU(s) list: 0-191

Thread(s) per core: 2

Core(s) per socket: 48

Socket(s): 2

NUMA node(s): 2

Vendor ID: AuthenticAMD

CPU family: 23

Model: 49

Model name: AMD EPYC 7642 48-Core Processor

Stepping: 0

Frequency boost: enabled

CPU MHz: 1499.440

CPU max MHz: 2300.0000

CPU min MHz: 1500.0000

BogoMIPS: 4599.98

Virtualization: AMD-V

L1d cache: 3 MiB

L1i cache: 3 MiB

L2 cache: 48 MiB

L3 cache: 512 MiB

NUMA node0 CPU(s): 0-47,96-143

NUMA node1 CPU(s): 48-95,144-191

Vulnerability Itlb multihit: Not affected

Vulnerability L1tf: Not affected

Vulnerability Mds: Not affected

Vulnerability Meltdown: Not affected

Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp

Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization

Vulnerability Spectre v2: Mitigation; Full AMD retpoline, IBPB conditional, IBRS_FW, STIBP conditional, RSB filling

Vulnerability Srbds: Not affected

Vulnerability Tsx async abort: Not affected

Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nop

l nonstop_tsc cpuid extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_leg

acy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 hw_pstate sme ssbd mba sev ibrs

ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm

_local clzero irperf xsaveerptr wbnoinvd arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif

umip rdpid overflow_recov succor smca


I suggest to apply tuning on the Server/BIOS/Kernel/OS level following our guide


If this won’t help , feel free to open support case by sending email to Networking-support@nvidia.com




Thanks you for your response. So i think driver is not compatibility with Ubuntu 20.04. Because I try to test on same hardware with CentOS 7 haven’t the issue. So i will try again follow as your recommend.