How to flash a generic firmware to ConnectX6-DX board?

Hi,

I’m having trouble in enabling a couple of features(Tx Offloads capabilities) from “5t for 5g” on a Connext6-DX from a DELL server(RHEL 8.8).
It was started in this forum(see the topic below), and ended up pointing to each other between DELL and RedHat.
Anyway, I’d like to see the feature(Tx Offloads Capability, TX_ON_TIMESTAMP) is working on my environment.
Would you mind telling me how to flash the ‘generic’ firmware into ConnectX6-DX?
Q1. What is the procedure for changing a specific PSID, for example, DELL???, into a generic one, MT??? ?
Q2. Which generic firmware version should I use?
Q3. Which PSID should I use?

Here is my environment.

Linux OS :
"Linux MELB-TEST-D18.ad.analog.com 4.18.0-477.10.1.rt7.274.el8_8.x86_64 #1 SMP PREEMPT_RT Wed Apr 5 13:20:38 EDT 2023 x86_64 x86_64 x86_64 GNU/Linux
"

Mellanox Firmware version :
"
[root@MELB-TEST-D18 generator]# mlxfwmanager
Querying Mellanox devices firmware …

Device #1:

Device Type: ConnectX6DX
Part Number: 0F6FXM_08P2T2_Ax
Description: Mellanox ConnectX-6 Dx Dual Port 100 GbE QSFP56 Network Adapter
PSID: DEL0000000027
PCI Device Name: /dev/mst/mt4125_pciconf0
Base GUID: 946dae0300e22eda
Base MAC: 946daee22eda
Versions: Current Available
FW 22.36.1010 N/A
PXE 3.6.0901 N/A
UEFI 14.29.0014 N/A

Status: No matching image found
"

Full error messages are :

"EAL: Detected CPU lcores: 24
EAL: Detected NUMA nodes: 1
EAL: Detected static linkage of DPDK
EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
EAL: Selected IOVA mode ‘PA’
EAL: VFIO support initialized
EAL: Probe PCI driver: mlx5_pci (15b3:101d) device: 0000:16:00.0 (socket 0)
EAL: Probe PCI driver: mlx5_pci (15b3:101d) device: 0000:16:00.1 (socket 0)
TELEMETRY: No legacy callbacks, legacy socket not created

Device driver name in use: mlx5_pci…
Initializing port 0 with 8 RX queues and 8 TX queues…
Ethdev port_id=0 requested Tx offloads 0x200000 doesn’t match Tx offloads capabilities 0xd96af in rte_eth_dev_configure()
EAL: Error - exiting with code: 1
Cause: Cannot configure device: err=-22, port=0
"

Thank you,

Kevin

P.S. : related topic

Hello Kevin,

Thank you for posting your inquiry on the NVIDIA Networking Community.

Based on the provided information, it appears you are using an OEM adapter. I recommend reaching out to Dell for support. Additionally, altering the OEM PSID is not advisable as it may lead to complications when seeking their assistance. Please note that the firmware downloads available on our website are specifically for adapters with Nvidia OPNs.

Thank you and regards,
NVIDIA Networking Technical Support

Thank you for the reply.
Would you mind telling me the recommended version of mlnx-ofed for Red Hat Linux 8.8 ?
Thank you,

Kevin

The latest LTS version of MLNX_OFED for RHEL 8.8 is 23.10-0.5.5.0, released in October 2023:

Thank you and regards,
NVIDIA Networking Technical Support

Thanks Juliaj,

One more question, is the firmware version for ConnectX6-DX 22.36.1010 ?
Thanks.

Hello Kevin,

To ensure compatibility with your OEM adapter, please contact your vendor for the appropriate firmware tailored to your system. They can offer the precise support needed. It’s important to note that the firmware available on our website is intended for adapters with Nvidia OPNs.

Thank you and regards,
NVIDIA Networking Technical Support

Hi,

I have an additional question for this topic.

What is the best way to check out on Linux(RedHat) machine whether ConnectX-6 DX is supporting “Accurate Scheduling” or not ? Especially, I’m curious about the supporting status.
Also, is there some prerequisites(for example, I only have ConnectX-6 DX device with Intel Xeon w7-3455 at the moment) for the feature ?

I’m asking since I can’t find any information about the feature on your site.
Kind regards,

Kevin

P.S. : I can’t get the feature(ACCURATE_SCHEDULING) working even if I configured it using mlxconfig.

[root@MELB-TEST-D18 generator]# mlxconfig -d 16:00.0 query | grep ACCURATE_
ACCURATE_TX_SCHEDULER True(1)
[root@MELB-TEST-D18 generator]# mlxconfig -d 16:00.0 query | grep REAL_TIME
REAL_TIME_CLOCK_ENABLE True(1)

lshw
*-network:0
description: Ethernet interface
product: MT2892 Family [ConnectX-6 Dx]
vendor: Mellanox Technologies
physical id: 0
bus info: pci@0000:16:00.0
logical name: ens1f0np0
version: 00
serial: ??:??:??:??:??:??
capacity: 10Gbit/s
width: 64 bits
clock: 33MHz
capabilities: pciexpress vpd msix pm bus_master cap_list rom ethernet physical 1000bt-fd 10000bt-fd autonegotiation
configuration: autonegotiation=on broadcast=yes driver=mlx5_core driverversion=23.10-0.5.5 duplex=full firmware=22.39.1002 (M
T_0000000437) ip=192.168.0.100 latency=0 link=yes multicast=yes
resources: iomemory:201f0-201ef irq:16 memory:201ffc000000-201ffdffffff memory:81000000-810fffff memory:201ffe800000-201ffeff
ffff

Hi Kevin,

I’m afraid you’ll need to contact OEM Dell for assistance with this issue, as they have control over the firmware feature in question.

Julia Jin

Thank you and regards,
NVIDIA Networking Technical Support

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.