NVIDIA Mellanox MCX623405AC-CDAN SmartNIC Firmware 22.32.1010

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.

Important Notes:
- Security Hardening Enhancements: This release contains important reliability improvements and security hardening enhancements. Mellanox recommends upgrading your devices firmware to this release to improve the devices’ firmware security and reliability.
- SR-IOV - Virtual Functions (VF) per Port - The maximum Virtual Functions (VF) per port is 127.
- It is recommended to enable the “above 4G decoding” BIOS setting for features that require large amount of PCIe resources. Such features are: SR-IOV with numerous VFs, PCIe Emulated Switch, and Large BAR Requests.

Changes and New Features:

- Single PF per NUMA: Added support for BMC with a single PF per NUMA in Socket-Direct adapter cards. Note: SIngle PF per NUMA should not be enabled on Multi-host.
- OpenSNAPI Communication Channel: The communication channel is used to enable communication between processes on different vHCAs regardless of their network connectivity state.
- RegC Register: Exposed an additional steering register in the hardware (reg_c_6).
- QP Resources: Added a new NvConfig parameter LOG_MAX_QUEUE to set the maximum number of work queue resources (QP, RQ, SQ...) that can be created per function. The default value is 2^17.
- Congestion Control Key: Added a Congestion Control Key to all Congestion Control MADs to authenticate that they are originated from a trusted source.
- SMP Firewall: Added an SMP firewall to block the option of sending SMPs (MADS sent on QP0 from the Subnet Manager) from unauthorized hosts to prevent fake SMPs from being recognized as the SM.
- Vendor Specific MADs Class 0x9: Vendor Specific MADs Class 0x9 is no longer supported by the firmware. If case the firmware detects such MAD, the firmware will return a "NOT SUPPORTED" error to the user.
- Match Definer Object: Added support for a new steering match definer format (format 33).
- Graceful Teardown: The teardown of hotplugged emulated device (a.k.a unplug flow) is in the reverse order of the plug flow. However, certain legacy host software stack does not support surprise removal of the PCIe PF devices. To support such host software stack, emulation manager software will perform a graceful teardown.
- AES-XTS Encryption / Decryption: Enabled disk encryption services using the aes_xts protocol to allow inline data encryption and decryption towards a remote or a local disk/NVDIM.
- TLS/XTS/Signature Padding: Blocked the VF's ability to use both padding and signature in order to prevent the NIC from hanging.
- Asserts' Severity Level: Added 3 new assert filters (Health buffer, NVlog, FW trace). The assert will be exposed now if its severity level is equal to or above the new filter. The filters are configurable by the ini file. The "Health buffer" filter is also configurable by new access register.
- SyncE: Added support for clock frequency synchronization based on Synchronous Ethernet protocol. Note: This capability is not supported with link speeds of 50G and higher, and cannot run in parallel with diagnostic counters.
- Socket-Direct Adapter Cards: Added support for: Flow Direct, LACP, and GRE Offload
- Rate Limit per VM instead of VM-TC: Enabled Rate Limit per VM instead of VM-TC. This capability is implemented by adding support to a new Scheduling element type: rate limit elements that will connect to the rate_limit and will share its rate limit.
- Cross GVMI Memory Key: Cross GVMI memory key is used to allow cross GVMI memory access using indirect memory registration which crosses vHCA context.
- Steering LAG Mode (Hash LAG): [Beta] The new LAG mode (PORT_SELECT_FT LAG (hash LAG)) distributes the packets to ports according to the hash on the packet headers, instead of distributing the packets according to the QP (queue affinity – legacy LAG) to avoid cases where the slow/fast path packets are transmitted from different ports. Identifying the right port is done by using destination type UPLINK with destination_eswitch_owner_vhca_id_valid set and destination_eswitch_owner_vhca_id indicating the PF associated with the port.
- QSHR Access Register: Added support for QSHR access register to enable Set and Query rate limit per-host per-port.
- New Software Steering ICM Resource for VXLAN Encapsulation: The firmware now exposes a new Software Steering ICM resource for VXLAN encap expand in order for the SW Steering to manage this resource directly.
- Asymmetrical VFs per PF: Added support for asymmetrical VFs per PF. To enable it: PF_NUM_OF_VF_VALID must be true, and PF_NUM_OF_VF to a none zero value.
- mlxlink Support to read/write Access Registers by LID: Added 2 new MAD access registers to enable mlxlink to read/write access registers by LID (to the whole subnet).
- Virtio-net Full Emulation: Enabled the option to dynamically modify the MSIX and the number of virtio VF device queues. Note: This modification must be done before loading the driver on the device. This new capability includes the following limitations: total queue/msix number can not exceed 2k, queue/msix per virtio vf device cannot exceed 64, the scale of virtio device is limit to below 127 from mlxconfig

Bug Fixes:

- Fixed an issue that resulted in firmware getting stuck and causing unexpected behavior when connecting an optical transceiver that support RXLOS, and the remote side port was down.
- Fixed an issue that caused the link status to be reported incorrectly and consequently caused the link to go down due to the wrong definition of the RX_LOS polarity in the INI. 
- Fixed an issue that prevented the system from creating more than 128K QPs.
- Removed the "PPS out" dependency on the "PPS in" status.
- Changed the DPORT_OWNER configuration to ignore the zero value in order to configure an external host as the "owner" uses the "host index + 1" value.
- Fixed an issue where a valid connection tracking flow triggered an unsupported connection tracking syndrome.
- Fixed an issue that caused the ATS packets to be mis-handled if hosts set RCB to 128B.
- Synced reset is not supported when NVNe Emulation is enabled. Running mlxfwreset in synced mode (sync=1) when NVME EMU is enabled will result in the tool reporting the synced reset as supported, but the reset itself will fail.
- Limited the external loopback speed to the used module's capabilities.
- Fixed an issue that resulted in cache miss being increased and performance being reduced due to the software querying the counters very frequently.
- Modified the Rx flow to go directly to QP without going thru the RX steering flow to reflect correctly the statics for the Tx and Rx.
- Fixed an assert that occurred when mlxconfig LAG_RESOURCE_ALLOCATION was set due to setting multi-port-vhca which is not allowed.
- Added rx_ts_over_crc to the PCMR access reg to control the default behavior of the FCS if that FCS is overwritten by the PTP on runtime.
- Fixed issue where configuring two identical FLEX Parser parse-graph-nodes on two different ports, can cause one of them to not work.
- Fixed an issue that resulted in wrong RNR timeout when trying to set it during the rts2rts_qp transition.
- Fixed an issue that cause the flash frequency on boot to be lower than expected (under 50Mhz) by enabling the firmware to increase it on boot2 to normal frequency. Note: On boards that use Winbond flash, the firmware is blocked if using a firmware that does not include this fix. 
- Added support for DSFP AOC (CMIS) v4 when error code is not reported by the module.
- Fixed a BER issue on the Serdes by updating the mapping of logical to physical port configuration for Link-Maintenance flow.
- Fixed issue that caused the virtio (vDPA and virtio full emulation) checksum offload not to work with packet head rewrite in steering mode. 
- Fixed and issue that resulted in QP allocation with the QPN of the previously destroyed QP due to the fact that the CreateQP does not guarantee a monotonic grow of QPN.
- Fixed the PFs iterator. Merged the regular PF iterator with the NODNiC PF iterator.
- Enabled the usage of inline actions in STEs instead of action list to avoid the hardware limitation of the maximum size of actions list (2^16).
- Enabled matching geneve_tlv_option_0_exist in the flow table entry as it is required when matching geneve_tlv_option_0_data parameter.
- Disabled the internal CQ doorbell recovery to avoid triggering the CQ doorbell recovery trigger and interrupt virtio traffic.
- Fixed an issue that caused some commands to get stuck or fail when configuring the HCA_CAP.cmdif_checksum to 0x3 and using firmware version lower than 22.31.1004. 
- Removed firmware dependency on credits reset during link reset flow.
- Fixed a rare case of a doorbell drop that caused the Rx side to get stuck that when running traffic on top of a virtio device. 
- Fixed unexpected and excessive interrupts caused by internal misconfigured EQs that took PCI bandwidth and introduced PCIe latency and as a result caused virtio Tx pps degradation.
- Fixed unexpected and excessive interrupts received by the Host when running virtio emulation application traffic due to internal misconfigured EQ in NIC.

Changed the policy of VDPA queue number capability:

- When the devices count <= 8, the VDPA queue number in cap is 256
- When the device count >=32, the VDPA queue number in cap is 64
- When the devices count is in 9~31, the VDPA queue number in cap is 128. Here the devices counts all port functions configured in mlxconfig, including pf, vf and sf.

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.

  It is highly recommended to always use the most recent driver version available.

Try to set a system restore point before installing a device driver. This will help if you installed an incorrect or mismatched driver. Problems can arise when your hardware device is too old or not supported any longer.