Changes and New Features

Debugfs Directory Path Change

The debugfs directory of each interface can now be found under: /sys/kernel/debug/mlx5//, and not directly under the root of the debugfs filesystem (/sys/debug/kernel).

Deprecation of OFED Public Power PC Installation

Starting from this release, MLNX_OFED releases for Power PC are no longer available for download from the public Download Center web page. Instead, you can find it on the following page: https://network.nvidia.com/support/firmware/ibm-systemp/.

Pre-notification: Deprecation of Older Operating Systems

Starting from next release, MLNX_OFED releases will no longer support operating systems with kernels below v4.18. This includes the following systems:

• Rhel7.x

• Debian9.13

• Sles12.x

• Xenserver7.1

All HCAs

Supported in the following adapter cards unless specifically stated otherwise:

ConnectX-4/ConnectX-4 Lx/ConnectX-5/ConnectX-6/ConnectX-6 Dx/ConnectX-6 Lx/ConnectX-7/BlueField-2

ConnectX-6 Dx and above

Supported in the following adapter cards unless specifically stated otherwise:

ConnectX-6 Dx/ConnectX-6 Lx/ConnectX-7/BlueField-2

ConnectX-6 and above

Supported in the following adapter cards unless specifically stated otherwise:

ConnectX-6/ConnectX-6 Dx/ConnectX-6 Lx/ConnectX-7/BlueField-2

ConnectX-5 and above

Supported in the following adapter cards unless specifically stated otherwise:

ConnectX-5/ConnectX-6/ConnectX-6 Dx/ConnectX-6 Lx/ConnectX-7/BlueField-2

ConnectX-4 and above

Supported in the following adapter cards unless specifically stated otherwise:

ConnectX-4/ConnectX-4 Lx/ConnectX-5/ConnectX-6/ConnectX-6 Dx/ConnectX-6 Lx/ConnectX-7/BlueField-2

Rate Limiter Extension for LAG Mode

[ConnectX-5 and above] During rate configuration, this feature considers the aggregated port speed when LAG is configured, so the functions (PF,VF,SF) can utilize the maximum aggregated link speed for transmission.

Live Migration Support with IPsec Full Offload

[ConnectX-7 and above] Added an option to perform live migration on a machine with IPsec enabled. This allows having a VF on VM that has a valid IPsec state and migrate it without having to remove the IPsec.

Note: This feature is currently not supported when working in MPV mode (Multi port VHCA/Dual port). Such support will be added in future releases.

Exposing the Max SF Configuration via Devlink Resource

[All HCAs] This feature allows user applications to view the maximum number of SFs (Scalable Functions) configured on the system via general Linux API, Devlink resource.

NIC Temperature Exposure

[ConnectX-5 and above] Added support for NIC temperature exposure by implementing the hwmon kernel API.

Counters Indicating Packet Drops due to Catastrophic Steering Errors

[All HCAs] Added two counters through the Devlink health reporter - generated_pkt_steering_fail and handled_pkt_steering_fail. These counters indicate packet drops due to catastrophic steering errors.

Using the RDMA Tool to Allow Controlled QKEYs Use by Non-privileged Users

[All HCAs] To allow using controlled QKEYs by non-privileged users, the following command can now be used in the iproute2 RDMA tool, instead of the "enforce_qkey_check" module parameter, which will be deprecated:

----- $rdma system set privileged-qkey on

IPsec Configuration Support When All IPsec Operations are Offloaded to the Hardware

[ConnectX-7 and above] Added support for configuring IPsec even when all IPsec operations are completely offloaded to the hardware. This feature not only provides a significant performance improvement, but also enables the use of IPsec over RoCE packets, which are outside the network stack and cannot be used without full hardware offload. As a result, users can now leverage the benefits of IPsec protocol with RoCE V2, even when using SR-IOV VFs.

Configuration from VM/VF for Full Offload Transport Mode

[ConnectX-6 Lx and above] This feature introduces two new boolean attributes of a port function: ipsec_crypto and ipsec_packet (also known as IPsec Full Offload). The goal is to provide a level of granularity for controlling VF IPsec offload capabilities similar to what is currently offered in the software mode. This allows users to decide if they want both types of IPsec offload enabled for a VF, just one of them, or none (which is the default).

General

Bug fixes