NVIDIA MCX753436MS-HEA SmartNIC Firmware 28.40.1000

Firmware which is added at the time of manufacturing, is used to run user programs on the device and can be thought of as the software that allows hardware to run. Embedded firmware is used to control the functions of various hardware devices and systems, much like a computer’s operating system (OS) controls the function of software applications. Firmware may be written into read-only memory (ROM), erasable programmable read-only memory (EPROM) or flash memory.

The ConnectX-7 smart host channel adapter (HCA) provides up to four ports of connectivity and 400Gb/s of throughput, hardware-accelerated networking, storage, security, and manageability services at data center scale for cloud, telecommunications, AI, and enterprise workloads. ConnectX-7 empowers agile and high-performance networking solutions with features such as Accelerated Switching and Packet Processing (ASAP2), advanced RoCE, GPUDirect Storage, and in-line hardware acceleration for Transport Layer Security (TLS), IP Security (IPsec), and MAC Security (MACsec) encryption and decryption. ConnectX-7 enables organizations to meet their current and future networking needs in both high-bandwidth and high-density environments.

The ConnectX-7 smart host channel adapter (HCA), featuring the NVIDIA Quantum-2 InfiniBand architecture, provides the highest networking performance available to take on the world’s most challenging workloads. ConnectX-7 provides ultra-low latency, 400Gb/s throughput, and innovative NVIDIA In-Network Computing acceleration engines to provide additional acceleration to deliver the scalability and feature-rich technology needed for supercomputers, artificial intelligence, and hyperscale cloud data centers.

Changes and New Features:

- Enabled Single Netdev mapping to two PCIe devices (Socket Direct). Now multiple devices (PFs) of the same port can be combined under a single netdev instance. Traffic is passed through different devices belonging to different NUMA sockets, thus saving cross-NUMA traffic and allowing apps running on the same netdev from different NUMAs to still feel a sense of proximity to the device and achieve improved performance. The netdev is destroyed once any of the PFs is removed. A proper configuration would utilize the correct close NUMA when working on a certain app/CPU. Currently, this capability is limited to PFs only, and up to two devices (sockets). To enable the feature, one must configure the same Socket Direct group (non zero) for both PFs through mlxconfig SD_GROUP.
- Added a new access register (PBWS) to set the port maximum bandwidth to a value between 95% to 100%.
- Added a new access register (PBWS) to set the port maximum bandwidth to a value between 95% to 100%.
- Added support for live migration with MPV and IPSEC. This capability enables creating cross-vhca objects, however, they can only be created between affiliated GVMIs.
- NVIDIA is adding an alternative Bill of Materials (BOM) for the specified affected items (MCX713104AS-ADAT & MCX713104AC-ADAT) to enhance production yields. The new alternate BOM requires updating to a minimum firmware version of 28.39.2048.

Bug Fixes:

- Fixed an issue that prevented Congestion Control from behaving properly when GRH is used in traffic of an IB cluster.
- SPDM requests received while CPLD burn flow is in progress may be answered with incorrect responses.
- When ConnectX-7 adapter card is configured to use the Auto-Negotiation mode, 400G_8x linkup cannot be raised.
- When performing a stress toggling on a ConnectX-7 adapter card that is connected to the MMA1Z00-NS400 cable and the speed is set to 100G_1x with interleaved FEC, a long linkup time of up to 5 min may occur.
- When raising a link using 200G optical cables while connecting a ConnectX-7 to a ConnectX-7, raising a link with width less than the maximum provided by the cable with speed 25G lane is not supported.
- When raising a link using 400G optical cables while connecting a ConnectX-7 to a ConnectX-7, raising a link with width less than the maximum provided by the cable with speed 50G or 25G lane is not supported.
- WoL packets may not working properly if sent to Unicast destination MAC.
- When performing PCIe link secondary-bus-reset, disable/enable or mlxfwreset on AMD based Genoa systems, the device takes longer then expected to link up, due to a PCIe receiver termination misconfiguration.
- Disabling the Relaxed Ordered (RO) capability (relaxed_ordering_read_pci_enabled=0) using the vhca_resource_manager is currently not functional.
- In rare cases, linkup time of NDR and NDR200 with MMA4Z00-NS400 may take longer than 60 seconds.
- Added back the Digital Feedforward Equalizer (DFFE) hardware component to improve the signal integrity link.
- Fixed an issue with Selective-Repeat configuration which occasionally caused retransmission to wait for timeout instead of out-of-sequence NACK.
- Enabled the lowest minimum rate for SW DCQCN to enable congestion control to hold a larger amount of QPs without pauses or drops.
- Fixed an issue that caused the secondary ASIC run module init to fail due to missing condition.
- Fixed an issue that kept the adapter cards' quad ports UP when using breakout cables / QSFP-split-4. Now when a 4 alignment loss is noticed, the link in 25G/lane Ethernet is dropped.
- Modified PCIe switch downstream port EQLZ.PH1 timing to 3ms.
- Fixed a rare race condition in NODNIC teardown that caused commands to hang on regular PF.

About Network Adapter Firmware:

When connected, the operating system usually installs a generic firmware that helps the computer to recognize the newly attached device.

However, proper software must be applied if you want to make use of all features that the network adapter has available. This task also allows computers to properly recognize all device characteristics such as manufacturer, chipset, technology, and others.

Updating the adapter's firmware and utilities version might improve overall performance and stability, increase transfer speeds, fix different compatibility problems and several network-related errors, as well as bring various other changes.

To install this release, simply get the package, extract it if necessary, run the setup, and follow the instructions displayed on-screen. When done, don't forget to perform a system restart and reconnect the network adapter to make sure that all changes take effect properly.

Without further ado, if you intend to apply this version, click the download button and install the package. Moreover, check with our website as often as possible so that you don't miss a single new release.