Note: If version 8.4 was previously installed, then it is not necessary to upgrade to version 8.4 (C).

          Note: If version 8.4 was previously installed, then it is not necessary to upgrade to version 8.4 (C).

Updated to version 9.4.1.20

Updated to version 9.4.1.20

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.

• HP FlexFabric 10Gb 2-port 556FLR-T Adapter
• HP FlexFabric 10Gb 2-port 556FLR-SFP+ Adapter

• HP FlexFabric 10Gb 2-port 556FLR-T Adapter
• HP FlexFabric 10Gb 2-port 556FLR-SFP+ Adapter

• For ConnectX-5 Adapters and above

Added support for the following features:

• Option to dump configuration space via the devlink tool in order to improve debug capabilities.
• The conventional TX descriptor (WQE or Work Queue Element) describes a single packet for transmission. Added driver support for the HW feature of multi-packet TX WQEs in XDP transmit flows. With this, the HW becomes capable of working with a new and improved WQE layout that describes several packets. In effect, this feature saves PCI bandwidth and transactions, and improves transmit packet rate.
• GENEVE encapsulation/decapsulation rules offload.
• Driver support for the hardware feature of multi-packet Tx to work with a new and improved WQE layout that describes several packets instead of a single packet for XDP transmission flows. This saves PCI bandwidth and transactions, and improves transmit packet rate.
• Updating CT (Connection Tracking) rules using the software steering mechanism.
• Updating remote mirroring rules using the software steering mechanism.
   
• For ConnectX-4 Adapters and above

Added support for the following features:

• Exposed rx_prio[p]_discards discard counters per priority that count the number of received packets dropped due to lack of buffers on the physical port.
• Reporting TSO and CSUM offload capabilities for MPLS tagged traffic and, allowed the kernel stack to use these offloads.
• mlx5e Max Combined Channels Increased the driver’s maximal combined channels value from 64 to 128 (however, note that OOB value will not cross 64). 128 is the upper bound. Lower maximal value can be seen on the host, depending on the number of cores and MSIX's configured by the firmware.
• Added the following RoCE accelerator counters:
  • roce_adp_retrans - counts the number of adaptive retransmissions for RoCE traffic.
  • roce_adp_retrans_to - counts the number of times RoCE traffic reached timeout due to adaptive retransmission.
  • roce_slow_restart - counts the number of times RoCE slow restart was used.
  • roce_slow_restart_cnps - counts the number of times RoCE slow restart generated CNP packets roce_slow_restart_trans - counts the number of times RoCE slow restart changed state to slow restart.
• User to register memory regions with a relaxed ordering access flag through experimental verbs. This can enhance performance, depending on architecture and scenario.
   
• For All HCA's

Added support for the following features:

• Output ibdev2netdev tool output was changed such that the bonding device now points at the bond instead of the slave interface.
• Monitoring and recovering from errors that occur on the RX queue, such as cookie errors and timeout.
• Improved GSO (Generic Segmentation Offload) workload performance by decreasing doorbells usage to the minimum required.
• TX CQE (Completion Queue Element) compression. Saves on outgoing PCIe bandwidth by compressing CQEs together. Disabled by default. Configurable via private flags of ethtool.
• Firmware Versions Query via Devlink :Added the option to query for running and stored firmware versions using the devlink tool.
• Firmware Flash Update via Devlink :Added the option to update the firmware image in the flash using the devlink tool. Usage: devlink dev flash file .mfa2 For further information on how to perform this update, see "Updating Firmware Using ethtool/devlink and .mfa2 File" section in MFT User Manual.
• WQE (Work Queue Element) dump, triggered by an error on Rx/Tx reporters. In addition, some dumps (not triggered by an error) can be retrieved by the user via devlink health reporters.
• GENEVE tunneled hardware offloads of TSO, CSUM and RSS.
• TCP segmentation and checksum offload support for MPLS-tagged traffic.

• For ConnectX-5 Adapters and above

Added support for the following features:

• Option to dump configuration space via the devlink tool in order to improve debug capabilities.
• The conventional TX descriptor (WQE or Work Queue Element) describes a single packet for transmission. Added driver support for the HW feature of multi-packet TX WQEs in XDP transmit flows. With this, the HW becomes capable of working with a new and improved WQE layout that describes several packets. In effect, this feature saves PCI bandwidth and transactions, and improves transmit packet rate.
• GENEVE encapsulation/decapsulation rules offload.
• Driver support for the hardware feature of multi-packet Tx to work with a new and improved WQE layout that describes several packets instead of a single packet for XDP transmission flows. This saves PCI bandwidth and transactions, and improves transmit packet rate.
• Updating CT (Connection Tracking) rules using the software steering mechanism.
• Updating remote mirroring rules using the software steering mechanism.
   
• For ConnectX-4 Adapters and above

Added support for the following features:

• Exposed rx_prio[p]_discards discard counters per priority that count the number of received packets dropped due to lack of buffers on the physical port.
• Reporting TSO and CSUM offload capabilities for MPLS tagged traffic and, allowed the kernel stack to use these offloads.
• mlx5e Max Combined Channels Increased the driver’s maximal combined channels value from 64 to 128 (however, note that OOB value will not cross 64). 128 is the upper bound. Lower maximal value can be seen on the host, depending on the number of cores and MSIX's configured by the firmware.
• Added the following RoCE accelerator counters:
  • roce_adp_retrans - counts the number of adaptive retransmissions for RoCE traffic.
  • roce_adp_retrans_to - counts the number of times RoCE traffic reached timeout due to adaptive retransmission.
  • roce_slow_restart - counts the number of times RoCE slow restart was used.
  • roce_slow_restart_cnps - counts the number of times RoCE slow restart generated CNP packets roce_slow_restart_trans - counts the number of times RoCE slow restart changed state to slow restart.
• User to register memory regions with a relaxed ordering access flag through experimental verbs. This can enhance performance, depending on architecture and scenario.
   
• For All HCA's

Added support for the following features:

• Output ibdev2netdev tool output was changed such that the bonding device now points at the bond instead of the slave interface.
• Monitoring and recovering from errors that occur on the RX queue, such as cookie errors and timeout.
• Improved GSO (Generic Segmentation Offload) workload performance by decreasing doorbells usage to the minimum required.
• TX CQE (Completion Queue Element) compression. Saves on outgoing PCIe bandwidth by compressing CQEs together. Disabled by default. Configurable via private flags of ethtool.
• Firmware Versions Query via Devlink :Added the option to query for running and stored firmware versions using the devlink tool.
• Firmware Flash Update via Devlink :Added the option to update the firmware image in the flash using the devlink tool. Usage: devlink dev flash file .mfa2 For further information on how to perform this update, see "Updating Firmware Using ethtool/devlink and .mfa2 File" section in MFT User Manual.
• WQE (Work Queue Element) dump, triggered by an error on Rx/Tx reporters. In addition, some dumps (not triggered by an error) can be retrieved by the user via devlink health reporters.
• GENEVE tunneled hardware offloads of TSO, CSUM and RSS.
• TCP segmentation and checksum offload support for MPLS-tagged traffic.

• For ConnectX-5 Adapters and above

Added support for the following features:

• Option to dump configuration space via the devlink tool in order to improve debug capabilities.
• The conventional TX descriptor (WQE or Work Queue Element) describes a single packet for transmission. Added driver support for the HW feature of multi-packet TX WQEs in XDP transmit flows. With this, the HW becomes capable of working with a new and improved WQE layout that describes several packets. In effect, this feature saves PCI bandwidth and transactions, and improves transmit packet rate.
• GENEVE encapsulation/decapsulation rules offload.
• Driver support for the hardware feature of multi-packet Tx to work with a new and improved WQE layout that describes several packets instead of a single packet for XDP transmission flows. This saves PCI bandwidth and transactions, and improves transmit packet rate.
• Updating CT (Connection Tracking) rules using the software steering mechanism.
• Updating remote mirroring rules using the software steering mechanism.
   
• For ConnectX-4 Adapters and above

Added support for the following features:

• Exposed rx_prio[p]_discards discard counters per priority that count the number of received packets dropped due to lack of buffers on the physical port.
• Reporting TSO and CSUM offload capabilities for MPLS tagged traffic and, allowed the kernel stack to use these offloads.
• mlx5e Max Combined Channels Increased the driver’s maximal combined channels value from 64 to 128 (however, note that OOB value will not cross 64). 128 is the upper bound. Lower maximal value can be seen on the host, depending on the number of cores and MSIX's configured by the firmware.
• Added the following RoCE accelerator counters:
  • roce_adp_retrans - counts the number of adaptive retransmissions for RoCE traffic.
  • roce_adp_retrans_to - counts the number of times RoCE traffic reached timeout due to adaptive retransmission.
  • roce_slow_restart - counts the number of times RoCE slow restart was used.
  • roce_slow_restart_cnps - counts the number of times RoCE slow restart generated CNP packets roce_slow_restart_trans - counts the number of times RoCE slow restart changed state to slow restart.
• User to register memory regions with a relaxed ordering access flag through experimental verbs. This can enhance performance, depending on architecture and scenario.
   
• For All HCA's

Added support for the following features:

• Output ibdev2netdev tool output was changed such that the bonding device now points at the bond instead of the slave interface.
• Monitoring and recovering from errors that occur on the RX queue, such as cookie errors and timeout.
• Improved GSO (Generic Segmentation Offload) workload performance by decreasing doorbells usage to the minimum required.
• TX CQE (Completion Queue Element) compression. Saves on outgoing PCIe bandwidth by compressing CQEs together. Disabled by default. Configurable via private flags of ethtool.
• Firmware Versions Query via Devlink :Added the option to query for running and stored firmware versions using the devlink tool.
• Firmware Flash Update via Devlink :Added the option to update the firmware image in the flash using the devlink tool. Usage: devlink dev flash file .mfa2 For further information on how to perform this update, see "Updating Firmware Using ethtool/devlink and .mfa2 File" section in MFT User Manual.
• WQE (Work Queue Element) dump, triggered by an error on Rx/Tx reporters. In addition, some dumps (not triggered by an error) can be retrieved by the user via devlink health reporters.
• GENEVE tunneled hardware offloads of TSO, CSUM and RSS.
• TCP segmentation and checksum offload support for MPLS-tagged traffic.

• For ConnectX-5 Adapters and above

Added support for the following features:

• Option to dump configuration space via the devlink tool in order to improve debug capabilities.
• The conventional TX descriptor (WQE or Work Queue Element) describes a single packet for transmission. Added driver support for the HW feature of multi-packet TX WQEs in XDP transmit flows. With this, the HW becomes capable of working with a new and improved WQE layout that describes several packets. In effect, this feature saves PCI bandwidth and transactions, and improves transmit packet rate.
• GENEVE encapsulation/decapsulation rules offload.
• Driver support for the hardware feature of multi-packet Tx to work with a new and improved WQE layout that describes several packets instead of a single packet for XDP transmission flows. This saves PCI bandwidth and transactions, and improves transmit packet rate.
• Updating CT (Connection Tracking) rules using the software steering mechanism.
• Updating remote mirroring rules using the software steering mechanism.
   
• For ConnectX-4 Adapters and above

Added support for the following features:

• Exposed rx_prio[p]_discards discard counters per priority that count the number of received packets dropped due to lack of buffers on the physical port.
• Reporting TSO and CSUM offload capabilities for MPLS tagged traffic and, allowed the kernel stack to use these offloads.
• mlx5e Max Combined Channels Increased the driver’s maximal combined channels value from 64 to 128 (however, note that OOB value will not cross 64). 128 is the upper bound. Lower maximal value can be seen on the host, depending on the number of cores and MSIX's configured by the firmware.
• Added the following RoCE accelerator counters:
  • roce_adp_retrans - counts the number of adaptive retransmissions for RoCE traffic.
  • roce_adp_retrans_to - counts the number of times RoCE traffic reached timeout due to adaptive retransmission.
  • roce_slow_restart - counts the number of times RoCE slow restart was used.
  • roce_slow_restart_cnps - counts the number of times RoCE slow restart generated CNP packets roce_slow_restart_trans - counts the number of times RoCE slow restart changed state to slow restart.
• User to register memory regions with a relaxed ordering access flag through experimental verbs. This can enhance performance, depending on architecture and scenario.
   
• For All HCA's

Added support for the following features:

• Output ibdev2netdev tool output was changed such that the bonding device now points at the bond instead of the slave interface.
• Monitoring and recovering from errors that occur on the RX queue, such as cookie errors and timeout.
• Improved GSO (Generic Segmentation Offload) workload performance by decreasing doorbells usage to the minimum required.
• TX CQE (Completion Queue Element) compression. Saves on outgoing PCIe bandwidth by compressing CQEs together. Disabled by default. Configurable via private flags of ethtool.
• Firmware Versions Query via Devlink :Added the option to query for running and stored firmware versions using the devlink tool.
• Firmware Flash Update via Devlink :Added the option to update the firmware image in the flash using the devlink tool. Usage: devlink dev flash file .mfa2 For further information on how to perform this update, see "Updating Firmware Using ethtool/devlink and .mfa2 File" section in MFT User Manual.
• WQE (Work Queue Element) dump, triggered by an error on Rx/Tx reporters. In addition, some dumps (not triggered by an error) can be retrieved by the user via devlink health reporters.
• GENEVE tunneled hardware offloads of TSO, CSUM and RSS.
• TCP segmentation and checksum offload support for MPLS-tagged traffic.

• Added support for RHEL 8.5

Updated to driver version 12.8.528.24

• Added support for RHEL 8.5

Updated to driver version 12.8.528.24

• Added support for RHEL 8.4

Updated to driver version 12.8.528.7

• Added support for RHEL 8.4

Updated to driver version 12.8.528.7

For ConnectX-4 Adapters:

• Updated mlx5 driver to use auxiliary bus in order to integrate different driver components into driver core and optimize module load/unload sequences.
• WJH (What Just Happened) in NICs: WJH allows for visibility of dropped packets (i.e., receiving notice of drop counters increase, seeing content of the dropped packets, debugging, and more). WJH is a service in devlink context and it is already implemented in the switch.
• Note: processing dropped packets (even for visibility purposes) may cause a degradation in performance and leaves the driver vulnerable for malicious attacks. The feature is disabled by default.
• Supported traps:
• VLAN mismatch: existing generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong VLAN tag
• DMAC mismatch: new generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong destination MAC
• Added support for enabling Relaxed Ordering for Kernel ULPs. Using relaxed ordering can improve performance in some setups. Since kernel ULPs are expected to support RO, it is enabled for them by default so they can benefit from it.

For ConnectX-4 Adapters and above: 

• In MLNX_OFED 5.4 GA, ConnectX-4/5/6 Ethernet network interfaces are now provided with permanent names. Prior to this release, the default interface names were provided by the kernel and udevd (ethX) remained as-is. From this release onwards, interface names are generated via new udevd rules. The generated names are now predictable, and the default names are automatically renamed to the predictable names by the udevd daemon, according to udev rules files installed by OFED.

For ConnectX-5 Adapters and above:

• Support to allow OVS (OpenVSwitch) kernel to support up to 128 matches (groups) per table and 16M entries per group.
• Offload ct_state flags rpl, inv, and rel.
• For rpl, support was added for both set and not set matching offload (i.e., +rpl and -rpl).
• For inv and rel, support was added only for the not set option (i.e., -rel and -inv).
• Scalable functions (also called subfunctions):This feature enables the user to create, configure, and deploy a scalable functions (e.g., RDMA and networking applications) and to assign them to a container when a container is started via mlxdevm tool.A scalable function can also be deployed in an untrusted guest/host system from the NIC/DPU. This enables full configuration of the function and its representors from the NIC/DPU before giving the function for a container to run in a host system.
• Scalable function QoS and QoS group via mlxdevm's rate commands were added. Run "man mlxdevm port" for details.
• "Signature API" which, on supported devices, allows application-level data-integrity checks via a signature handover mechanism. Various signature types, including CRC32 and T10-DIF, can be automatically calculated and checked, stripped, or appended during the transfer at full wire speed.
• DR support for matching on RAW tunnel headers using the misc5 parameters, This feature allows matching on each bit of the header,
• inducing reserved fields.

For ConnectX-6 Dx Adapters:

• Mapping a QP (Queue Pair) to AH over DEVX API, which enables DC/UD QPs to use multiple CC algorithms in the same data center.
• Better insertion rate in software steering was supported. This includes multi-QP which skips areas in the code that may be for debug only.
• Improved rate of updating steering rules, insertion, and deletion. The feature includes definers, multi-qp approach, and better memory usage.
• Added support for ability to allow or prevent insertion of duplicate rules, so the user can choose one of the following behaviors:
• Prevent duplicate rules, so that already-existing rule and fail can be detected.
• Allow duplicate rules, to enable updating the rule's action (this will only take effect once the previous rule is deleted). By default, duplicate rules are allowed.
• Made it so that all rule's insertion occur in a defined time using defined (export) size of Htble and decreased use of dynamic allocation.

For all HCA's:

• As of version 5.4, the driver is set so that udev rules will change the names of network interfaces created from NVIDIA adapters.The udev rules are shipped to "/lib/udev/rules.d" and may be overridden by placing a file with the same name in "/etc/udev/rules.d".
• Example: /etc/udev/rules.d/82-net-setup-link.rules
• OvS-DPDK deprecated the command "ovs-appctl dpctl/dump-e2e-stats". Instead, the command has been integrated into the existing command "ovs-appctl dpctl/offload-stats-show -m" (when e2e-cache is enabled).
• OvS-DPDK ct-ct-nat offloads is now disabled by default.
• Before version 5.4, /etc/udev/rules.d/90-ib.rules was potentially automatically edited by installation scripts in case the options —umad-dev-rw or —umad-dev-na were used. From version 5.4 and above, those changes are made in /etc/udev/rules.d/91-ib-permissions.rules which (if exist) only include the settings for those command-line options.
• ibv_query_qp_data_in_order() API: This API enables an application to check if the given QP data is guaranteed to be in order, enabling poll for data instead of poll for completion.
• Added ethtool extended link state to mlx5e. ethtool can be used to get more information to help troubleshoot the state.
• Moved all Python scripts and some other common scripts out of the mlnxofa_ kernel packages. This removed the python dependency from that package when rebuilding it and avoided unnecessary errors when rebuilding them for custom kernels.

For ConnectX-4 Adapters:

• Updated mlx5 driver to use auxiliary bus in order to integrate different driver components into driver core and optimize module load/unload sequences.
• WJH (What Just Happened) in NICs: WJH allows for visibility of dropped packets (i.e., receiving notice of drop counters increase, seeing content of the dropped packets, debugging, and more). WJH is a service in devlink context and it is already implemented in the switch.
• Note: processing dropped packets (even for visibility purposes) may cause a degradation in performance and leaves the driver vulnerable for malicious attacks. The feature is disabled by default.
• Supported traps:
• VLAN mismatch: existing generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong VLAN tag
• DMAC mismatch: new generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong destination MAC
• Added support for enabling Relaxed Ordering for Kernel ULPs. Using relaxed ordering can improve performance in some setups. Since kernel ULPs are expected to support RO, it is enabled for them by default so they can benefit from it.

For ConnectX-4 Adapters and above: 

• In MLNX_OFED 5.4 GA, ConnectX-4/5/6 Ethernet network interfaces are now provided with permanent names. Prior to this release, the default interface names were provided by the kernel and udevd (ethX) remained as-is. From this release onwards, interface names are generated via new udevd rules. The generated names are now predictable, and the default names are automatically renamed to the predictable names by the udevd daemon, according to udev rules files installed by OFED.

For ConnectX-5 Adapters and above:

• Support to allow OVS (OpenVSwitch) kernel to support up to 128 matches (groups) per table and 16M entries per group.
• Offload ct_state flags rpl, inv, and rel.
• For rpl, support was added for both set and not set matching offload (i.e., +rpl and -rpl).
• For inv and rel, support was added only for the not set option (i.e., -rel and -inv).
• Scalable functions (also called subfunctions):This feature enables the user to create, configure, and deploy a scalable functions (e.g., RDMA and networking applications) and to assign them to a container when a container is started via mlxdevm tool.A scalable function can also be deployed in an untrusted guest/host system from the NIC/DPU. This enables full configuration of the function and its representors from the NIC/DPU before giving the function for a container to run in a host system.
• Scalable function QoS and QoS group via mlxdevm's rate commands were added. Run "man mlxdevm port" for details.
• "Signature API" which, on supported devices, allows application-level data-integrity checks via a signature handover mechanism. Various signature types, including CRC32 and T10-DIF, can be automatically calculated and checked, stripped, or appended during the transfer at full wire speed.
• DR support for matching on RAW tunnel headers using the misc5 parameters, This feature allows matching on each bit of the header,
• inducing reserved fields.

For ConnectX-6 Dx Adapters:

• Mapping a QP (Queue Pair) to AH over DEVX API, which enables DC/UD QPs to use multiple CC algorithms in the same data center.
• Better insertion rate in software steering was supported. This includes multi-QP which skips areas in the code that may be for debug only.
• Improved rate of updating steering rules, insertion, and deletion. The feature includes definers, multi-qp approach, and better memory usage.
• Added support for ability to allow or prevent insertion of duplicate rules, so the user can choose one of the following behaviors:
• Prevent duplicate rules, so that already-existing rule and fail can be detected.
• Allow duplicate rules, to enable updating the rule's action (this will only take effect once the previous rule is deleted). By default, duplicate rules are allowed.
• Made it so that all rule's insertion occur in a defined time using defined (export) size of Htble and decreased use of dynamic allocation.

For all HCA's:

• As of version 5.4, the driver is set so that udev rules will change the names of network interfaces created from NVIDIA adapters.The udev rules are shipped to "/lib/udev/rules.d" and may be overridden by placing a file with the same name in "/etc/udev/rules.d".
• Example: /etc/udev/rules.d/82-net-setup-link.rules
• OvS-DPDK deprecated the command "ovs-appctl dpctl/dump-e2e-stats". Instead, the command has been integrated into the existing command "ovs-appctl dpctl/offload-stats-show -m" (when e2e-cache is enabled).
• OvS-DPDK ct-ct-nat offloads is now disabled by default.
• Before version 5.4, /etc/udev/rules.d/90-ib.rules was potentially automatically edited by installation scripts in case the options —umad-dev-rw or —umad-dev-na were used. From version 5.4 and above, those changes are made in /etc/udev/rules.d/91-ib-permissions.rules which (if exist) only include the settings for those command-line options.
• ibv_query_qp_data_in_order() API: This API enables an application to check if the given QP data is guaranteed to be in order, enabling poll for data instead of poll for completion.
• Added ethtool extended link state to mlx5e. ethtool can be used to get more information to help troubleshoot the state.
• Moved all Python scripts and some other common scripts out of the mlnxofa_ kernel packages. This removed the python dependency from that package when rebuilding it and avoided unnecessary errors when rebuilding them for custom kernels.

For ConnectX-4 Adapters:

• Updated mlx5 driver to use auxiliary bus in order to integrate different driver components into driver core and optimize module load/unload sequences.
• WJH (What Just Happened) in NICs: WJH allows for visibility of dropped packets (i.e., receiving notice of drop counters increase, seeing content of the dropped packets, debugging, and more). WJH is a service in devlink context and it is already implemented in the switch.
• Note: processing dropped packets (even for visibility purposes) may cause a degradation in performance and leaves the driver vulnerable for malicious attacks. The feature is disabled by default.
• Supported traps:
• VLAN mismatch: existing generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong VLAN tag
• DMAC mismatch: new generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong destination MAC
• Added support for enabling Relaxed Ordering for Kernel ULPs. Using relaxed ordering can improve performance in some setups. Since kernel ULPs are expected to support RO, it is enabled for them by default so they can benefit from it.

For ConnectX-4 Adapters and above: 

• In MLNX_OFED 5.4 GA, ConnectX-4/5/6 Ethernet network interfaces are now provided with permanent names. Prior to this release, the default interface names were provided by the kernel and udevd (ethX) remained as-is. From this release onwards, interface names are generated via new udevd rules. The generated names are now predictable, and the default names are automatically renamed to the predictable names by the udevd daemon, according to udev rules files installed by OFED.

For ConnectX-5 Adapters and above:

• Support to allow OVS (OpenVSwitch) kernel to support up to 128 matches (groups) per table and 16M entries per group.
• Offload ct_state flags rpl, inv, and rel.
• For rpl, support was added for both set and not set matching offload (i.e., +rpl and -rpl).
• For inv and rel, support was added only for the not set option (i.e., -rel and -inv).
• Scalable functions (also called subfunctions):This feature enables the user to create, configure, and deploy a scalable functions (e.g., RDMA and networking applications) and to assign them to a container when a container is started via mlxdevm tool.A scalable function can also be deployed in an untrusted guest/host system from the NIC/DPU. This enables full configuration of the function and its representors from the NIC/DPU before giving the function for a container to run in a host system.
• Scalable function QoS and QoS group via mlxdevm's rate commands were added. Run "man mlxdevm port" for details.
• "Signature API" which, on supported devices, allows application-level data-integrity checks via a signature handover mechanism. Various signature types, including CRC32 and T10-DIF, can be automatically calculated and checked, stripped, or appended during the transfer at full wire speed.
• DR support for matching on RAW tunnel headers using the misc5 parameters, This feature allows matching on each bit of the header,
• inducing reserved fields.

For ConnectX-6 Dx Adapters:

• Mapping a QP (Queue Pair) to AH over DEVX API, which enables DC/UD QPs to use multiple CC algorithms in the same data center.
• Better insertion rate in software steering was supported. This includes multi-QP which skips areas in the code that may be for debug only.
• Improved rate of updating steering rules, insertion, and deletion. The feature includes definers, multi-qp approach, and better memory usage.
• Added support for ability to allow or prevent insertion of duplicate rules, so the user can choose one of the following behaviors:
• Prevent duplicate rules, so that already-existing rule and fail can be detected.
• Allow duplicate rules, to enable updating the rule's action (this will only take effect once the previous rule is deleted). By default, duplicate rules are allowed.
• Made it so that all rule's insertion occur in a defined time using defined (export) size of Htble and decreased use of dynamic allocation.

For all HCA's:

• As of version 5.4, the driver is set so that udev rules will change the names of network interfaces created from NVIDIA adapters.The udev rules are shipped to "/lib/udev/rules.d" and may be overridden by placing a file with the same name in "/etc/udev/rules.d".
• Example: /etc/udev/rules.d/82-net-setup-link.rules
• OvS-DPDK deprecated the command "ovs-appctl dpctl/dump-e2e-stats". Instead, the command has been integrated into the existing command "ovs-appctl dpctl/offload-stats-show -m" (when e2e-cache is enabled).
• OvS-DPDK ct-ct-nat offloads is now disabled by default.
• Before version 5.4, /etc/udev/rules.d/90-ib.rules was potentially automatically edited by installation scripts in case the options —umad-dev-rw or —umad-dev-na were used. From version 5.4 and above, those changes are made in /etc/udev/rules.d/91-ib-permissions.rules which (if exist) only include the settings for those command-line options.
• ibv_query_qp_data_in_order() API: This API enables an application to check if the given QP data is guaranteed to be in order, enabling poll for data instead of poll for completion.
• Added ethtool extended link state to mlx5e. ethtool can be used to get more information to help troubleshoot the state.
• Moved all Python scripts and some other common scripts out of the mlnxofa_ kernel packages. This removed the python dependency from that package when rebuilding it and avoided unnecessary errors when rebuilding them for custom kernels.

For ConnectX-4 Adapters:

• Updated mlx5 driver to use auxiliary bus in order to integrate different driver components into driver core and optimize module load/unload sequences.
• WJH (What Just Happened) in NICs: WJH allows for visibility of dropped packets (i.e., receiving notice of drop counters increase, seeing content of the dropped packets, debugging, and more). WJH is a service in devlink context and it is already implemented in the switch.
• Note: processing dropped packets (even for visibility purposes) may cause a degradation in performance and leaves the driver vulnerable for malicious attacks. The feature is disabled by default.
• Supported traps:
• VLAN mismatch: existing generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong VLAN tag
• DMAC mismatch: new generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong destination MAC
• Added support for enabling Relaxed Ordering for Kernel ULPs. Using relaxed ordering can improve performance in some setups. Since kernel ULPs are expected to support RO, it is enabled for them by default so they can benefit from it.

For ConnectX-4 Adapters and above: 

• In MLNX_OFED 5.4 GA, ConnectX-4/5/6 Ethernet network interfaces are now provided with permanent names. Prior to this release, the default interface names were provided by the kernel and udevd (ethX) remained as-is. From this release onwards, interface names are generated via new udevd rules. The generated names are now predictable, and the default names are automatically renamed to the predictable names by the udevd daemon, according to udev rules files installed by OFED.

For ConnectX-5 Adapters and above:

• Support to allow OVS (OpenVSwitch) kernel to support up to 128 matches (groups) per table and 16M entries per group.
• Offload ct_state flags rpl, inv, and rel.
• For rpl, support was added for both set and not set matching offload (i.e., +rpl and -rpl).
• For inv and rel, support was added only for the not set option (i.e., -rel and -inv).
• Scalable functions (also called subfunctions):This feature enables the user to create, configure, and deploy a scalable functions (e.g., RDMA and networking applications) and to assign them to a container when a container is started via mlxdevm tool.A scalable function can also be deployed in an untrusted guest/host system from the NIC/DPU. This enables full configuration of the function and its representors from the NIC/DPU before giving the function for a container to run in a host system.
• Scalable function QoS and QoS group via mlxdevm's rate commands were added. Run "man mlxdevm port" for details.
• "Signature API" which, on supported devices, allows application-level data-integrity checks via a signature handover mechanism. Various signature types, including CRC32 and T10-DIF, can be automatically calculated and checked, stripped, or appended during the transfer at full wire speed.
• DR support for matching on RAW tunnel headers using the misc5 parameters, This feature allows matching on each bit of the header,
• inducing reserved fields.

For ConnectX-6 Dx Adapters:

• Mapping a QP (Queue Pair) to AH over DEVX API, which enables DC/UD QPs to use multiple CC algorithms in the same data center.
• Better insertion rate in software steering was supported. This includes multi-QP which skips areas in the code that may be for debug only.
• Improved rate of updating steering rules, insertion, and deletion. The feature includes definers, multi-qp approach, and better memory usage.
• Added support for ability to allow or prevent insertion of duplicate rules, so the user can choose one of the following behaviors:
• Prevent duplicate rules, so that already-existing rule and fail can be detected.
• Allow duplicate rules, to enable updating the rule's action (this will only take effect once the previous rule is deleted). By default, duplicate rules are allowed.
• Made it so that all rule's insertion occur in a defined time using defined (export) size of Htble and decreased use of dynamic allocation.

For all HCA's:

• As of version 5.4, the driver is set so that udev rules will change the names of network interfaces created from NVIDIA adapters.The udev rules are shipped to "/lib/udev/rules.d" and may be overridden by placing a file with the same name in "/etc/udev/rules.d".
• Example: /etc/udev/rules.d/82-net-setup-link.rules
• OvS-DPDK deprecated the command "ovs-appctl dpctl/dump-e2e-stats". Instead, the command has been integrated into the existing command "ovs-appctl dpctl/offload-stats-show -m" (when e2e-cache is enabled).
• OvS-DPDK ct-ct-nat offloads is now disabled by default.
• Before version 5.4, /etc/udev/rules.d/90-ib.rules was potentially automatically edited by installation scripts in case the options —umad-dev-rw or —umad-dev-na were used. From version 5.4 and above, those changes are made in /etc/udev/rules.d/91-ib-permissions.rules which (if exist) only include the settings for those command-line options.
• ibv_query_qp_data_in_order() API: This API enables an application to check if the given QP data is guaranteed to be in order, enabling poll for data instead of poll for completion.
• Added ethtool extended link state to mlx5e. ethtool can be used to get more information to help troubleshoot the state.
• Moved all Python scripts and some other common scripts out of the mlnxofa_ kernel packages. This removed the python dependency from that package when rebuilding it and avoided unnecessary errors when rebuilding them for custom kernels.

• HP FlexFabric 10Gb 2-port 556FLR-T Adapter
• HP FlexFabric 10Gb 2-port 556FLR-SFP+ Adapter

• HP FlexFabric 10Gb 2-port 556FLR-T Adapter
• HP FlexFabric 10Gb 2-port 556FLR-SFP+ Adapter

• HP FlexFabric 10Gb 2-port 556FLR-T Adapter
• HP FlexFabric 10Gb 2-port 556FLR-SFP+ Adapter

• HP FlexFabric 10Gb 2-port 556FLR-T Adapter
• HP FlexFabric 10Gb 2-port 556FLR-SFP+ Adapter

• Resolved an issue that caused the network adapter traffic to stop.
• Fixed an internal multicast loopback issue that broke LACP(Link Aggregation Control Protocol) bonding protocol.

• Resolved an issue that caused the network adapter traffic to stop.
• Fixed an internal multicast loopback issue that broke LACP(Link Aggregation Control Protocol) bonding protocol.

• Resolved an issue that caused the network adapter traffic to stop.
• Fixed an internal multicast loopback issue that broke LACP(Link Aggregation Control Protocol) bonding protocol.

• Resolved an issue that caused the network adapter traffic to stop.
• Fixed an internal multicast loopback issue that broke LACP(Link Aggregation Control Protocol) bonding protocol.

• Updated to support Service Pack for ProLiant version 2017.07.0.
  Note: If driver version 2.68.64.1 was previously installed, then it is not necessary to upgrade to version 2.68.64.1(B).

Enhancements/New Features implemented in version 2.68.64.1:

• Added support for Windows 8.1 and Windows Server 2012R2 to the build scripts.
• Add build support for new Windows Event Logging.
• Add support for automatic selection of the default driver build parameters file during the build

• Updated to support Service Pack for ProLiant version 2017.07.0.
  Note: If driver version 2.68.64.1 was previously installed, then it is not necessary to upgrade to version 2.68.64.1(B).

Enhancements/New Features implemented in version 2.68.64.1:

• Added support for Windows 8.1 and Windows Server 2012R2 to the build scripts.
• Add build support for new Windows Event Logging.
• Add support for automatic selection of the default driver build parameters file during the build

• Microsoft Windows Server 2016 - Server Core and Server with a Desktop.

Note: Revised component package. No change to driver functionality resulting from the change. If system has previously updated to driver version 2.68.64.2, then it is not necessary to update to 2.68.64.2(A).

• Microsoft Windows Server 2016 - Server Core and Server with a Desktop.

Note: Revised component package. No change to driver functionality resulting from the change. If system has previously updated to driver version 2.68.64.2, then it is not necessary to update to 2.68.64.2(A).

• HPE ProLiant DL360 Gen10 Plus Server
• HPE ProLiant DL380 Gen10 Plus Server 

• HPE ProLiant DL360 Gen10 Plus Server
• HPE ProLiant DL380 Gen10 Plus Server 

The raw driver files can be obtained by extracting the Smart Component and then extracting the Emulex installer.  Use this command:

    elxdrvr-fc-version.exe /q2 extract=2

The extracted files are located:

    C:\Users\Administrator\Documents\Emulex\Drivers\FC-version

Each kit folder has subsequent architecture folders with subsequent OS folders.  For example,

    C:\Users\Administrator\Documents\Emulex\Drivers\FC-version\x64\win2019

The raw driver files can be obtained by extracting the Smart Component and then extracting the Emulex installer.  Use this command:

    elxdrvr-fc-version.exe /q2 extract=2

The extracted files are located:

    C:\Users\Administrator\Documents\Emulex\Drivers\FC-version

Each kit folder has subsequent architecture folders with subsequent OS folders.  For example,

    C:\Users\Administrator\Documents\Emulex\Drivers\FC-version\x64\win2019

For ConnectX-4 Adapters:

• Updated mlx5 driver to use auxiliary bus in order to integrate different driver components into driver core and optimize module load/unload sequences.
• WJH (What Just Happened) in NICs: WJH allows for visibility of dropped packets (i.e., receiving notice of drop counters increase, seeing content of the dropped packets, debugging, and more). WJH is a service in devlink context and it is already implemented in the switch.
• Note: processing dropped packets (even for visibility purposes) may cause a degradation in performance and leaves the driver vulnerable for malicious attacks. The feature is disabled by default.
• Supported traps:
• VLAN mismatch: existing generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong VLAN tag
• DMAC mismatch: new generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong destination MAC
• Added support for enabling Relaxed Ordering for Kernel ULPs. Using relaxed ordering can improve performance in some setups. Since kernel ULPs are expected to support RO, it is enabled for them by default so they can benefit from it.

For ConnectX-4 Adapters and above: 

• In MLNX_OFED 5.4 GA, ConnectX-4/5/6 Ethernet network interfaces are now provided with permanent names. Prior to this release, the default interface names were provided by the kernel and udevd (ethX) remained as-is. From this release onwards, interface names are generated via new udevd rules. The generated names are now predictable, and the default names are automatically renamed to the predictable names by the udevd daemon, according to udev rules files installed by OFED.

For ConnectX-5 Adapters and above:

• Support to allow OVS (OpenVSwitch) kernel to support up to 128 matches (groups) per table and 16M entries per group.
• Offload ct_state flags rpl, inv, and rel.
• For rpl, support was added for both set and not set matching offload (i.e., +rpl and -rpl).
• For inv and rel, support was added only for the not set option (i.e., -rel and -inv).
• Scalable functions (also called subfunctions):This feature enables the user to create, configure, and deploy a scalable functions (e.g., RDMA and networking applications) and to assign them to a container when a container is started via mlxdevm tool.A scalable function can also be deployed in an untrusted guest/host system from the NIC/DPU. This enables full configuration of the function and its representors from the NIC/DPU before giving the function for a container to run in a host system.
• Scalable function QoS and QoS group via mlxdevm's rate commands were added. Run "man mlxdevm port" for details.
• "Signature API" which, on supported devices, allows application-level data-integrity checks via a signature handover mechanism. Various signature types, including CRC32 and T10-DIF, can be automatically calculated and checked, stripped, or appended during the transfer at full wire speed.
• DR support for matching on RAW tunnel headers using the misc5 parameters, This feature allows matching on each bit of the header,
• inducing reserved fields.

For ConnectX-6 Dx Adapters:

• Mapping a QP (Queue Pair) to AH over DEVX API, which enables DC/UD QPs to use multiple CC algorithms in the same data center.
• Better insertion rate in software steering was supported. This includes multi-QP which skips areas in the code that may be for debug only.
• Improved rate of updating steering rules, insertion, and deletion. The feature includes definers, multi-qp approach, and better memory usage.
• Added support for ability to allow or prevent insertion of duplicate rules, so the user can choose one of the following behaviors:
• Prevent duplicate rules, so that already-existing rule and fail can be detected.
• Allow duplicate rules, to enable updating the rule's action (this will only take effect once the previous rule is deleted). By default, duplicate rules are allowed.
• Made it so that all rule's insertion occur in a defined time using defined (export) size of Htble and decreased use of dynamic allocation.

For all HCA's:

• As of version 5.4, the driver is set so that udev rules will change the names of network interfaces created from NVIDIA adapters.The udev rules are shipped to "/lib/udev/rules.d" and may be overridden by placing a file with the same name in "/etc/udev/rules.d".
• Example: /etc/udev/rules.d/82-net-setup-link.rules
• OvS-DPDK deprecated the command "ovs-appctl dpctl/dump-e2e-stats". Instead, the command has been integrated into the existing command "ovs-appctl dpctl/offload-stats-show -m" (when e2e-cache is enabled).
• OvS-DPDK ct-ct-nat offloads is now disabled by default.
• Before version 5.4, /etc/udev/rules.d/90-ib.rules was potentially automatically edited by installation scripts in case the options —umad-dev-rw or —umad-dev-na were used. From version 5.4 and above, those changes are made in /etc/udev/rules.d/91-ib-permissions.rules which (if exist) only include the settings for those command-line options.
• ibv_query_qp_data_in_order() API: This API enables an application to check if the given QP data is guaranteed to be in order, enabling poll for data instead of poll for completion.
• Added ethtool extended link state to mlx5e. ethtool can be used to get more information to help troubleshoot the state.
• Moved all Python scripts and some other common scripts out of the mlnxofa_ kernel packages. This removed the python dependency from that package when rebuilding it and avoided unnecessary errors when rebuilding them for custom kernels.

For ConnectX-4 Adapters:

• Updated mlx5 driver to use auxiliary bus in order to integrate different driver components into driver core and optimize module load/unload sequences.
• WJH (What Just Happened) in NICs: WJH allows for visibility of dropped packets (i.e., receiving notice of drop counters increase, seeing content of the dropped packets, debugging, and more). WJH is a service in devlink context and it is already implemented in the switch.
• Note: processing dropped packets (even for visibility purposes) may cause a degradation in performance and leaves the driver vulnerable for malicious attacks. The feature is disabled by default.
• Supported traps:
• VLAN mismatch: existing generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong VLAN tag
• DMAC mismatch: new generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong destination MAC
• Added support for enabling Relaxed Ordering for Kernel ULPs. Using relaxed ordering can improve performance in some setups. Since kernel ULPs are expected to support RO, it is enabled for them by default so they can benefit from it.

For ConnectX-4 Adapters and above: 

• In MLNX_OFED 5.4 GA, ConnectX-4/5/6 Ethernet network interfaces are now provided with permanent names. Prior to this release, the default interface names were provided by the kernel and udevd (ethX) remained as-is. From this release onwards, interface names are generated via new udevd rules. The generated names are now predictable, and the default names are automatically renamed to the predictable names by the udevd daemon, according to udev rules files installed by OFED.

For ConnectX-5 Adapters and above:

• Support to allow OVS (OpenVSwitch) kernel to support up to 128 matches (groups) per table and 16M entries per group.
• Offload ct_state flags rpl, inv, and rel.
• For rpl, support was added for both set and not set matching offload (i.e., +rpl and -rpl).
• For inv and rel, support was added only for the not set option (i.e., -rel and -inv).
• Scalable functions (also called subfunctions):This feature enables the user to create, configure, and deploy a scalable functions (e.g., RDMA and networking applications) and to assign them to a container when a container is started via mlxdevm tool.A scalable function can also be deployed in an untrusted guest/host system from the NIC/DPU. This enables full configuration of the function and its representors from the NIC/DPU before giving the function for a container to run in a host system.
• Scalable function QoS and QoS group via mlxdevm's rate commands were added. Run "man mlxdevm port" for details.
• "Signature API" which, on supported devices, allows application-level data-integrity checks via a signature handover mechanism. Various signature types, including CRC32 and T10-DIF, can be automatically calculated and checked, stripped, or appended during the transfer at full wire speed.
• DR support for matching on RAW tunnel headers using the misc5 parameters, This feature allows matching on each bit of the header,
• inducing reserved fields.

For ConnectX-6 Dx Adapters:

• Mapping a QP (Queue Pair) to AH over DEVX API, which enables DC/UD QPs to use multiple CC algorithms in the same data center.
• Better insertion rate in software steering was supported. This includes multi-QP which skips areas in the code that may be for debug only.
• Improved rate of updating steering rules, insertion, and deletion. The feature includes definers, multi-qp approach, and better memory usage.
• Added support for ability to allow or prevent insertion of duplicate rules, so the user can choose one of the following behaviors:
• Prevent duplicate rules, so that already-existing rule and fail can be detected.
• Allow duplicate rules, to enable updating the rule's action (this will only take effect once the previous rule is deleted). By default, duplicate rules are allowed.
• Made it so that all rule's insertion occur in a defined time using defined (export) size of Htble and decreased use of dynamic allocation.

For all HCA's:

• As of version 5.4, the driver is set so that udev rules will change the names of network interfaces created from NVIDIA adapters.The udev rules are shipped to "/lib/udev/rules.d" and may be overridden by placing a file with the same name in "/etc/udev/rules.d".
• Example: /etc/udev/rules.d/82-net-setup-link.rules
• OvS-DPDK deprecated the command "ovs-appctl dpctl/dump-e2e-stats". Instead, the command has been integrated into the existing command "ovs-appctl dpctl/offload-stats-show -m" (when e2e-cache is enabled).
• OvS-DPDK ct-ct-nat offloads is now disabled by default.
• Before version 5.4, /etc/udev/rules.d/90-ib.rules was potentially automatically edited by installation scripts in case the options —umad-dev-rw or —umad-dev-na were used. From version 5.4 and above, those changes are made in /etc/udev/rules.d/91-ib-permissions.rules which (if exist) only include the settings for those command-line options.
• ibv_query_qp_data_in_order() API: This API enables an application to check if the given QP data is guaranteed to be in order, enabling poll for data instead of poll for completion.
• Added ethtool extended link state to mlx5e. ethtool can be used to get more information to help troubleshoot the state.
• Moved all Python scripts and some other common scripts out of the mlnxofa_ kernel packages. This removed the python dependency from that package when rebuilding it and avoided unnecessary errors when rebuilding them for custom kernels.

For ConnectX-4 Adapters:

• Updated mlx5 driver to use auxiliary bus in order to integrate different driver components into driver core and optimize module load/unload sequences.
• WJH (What Just Happened) in NICs: WJH allows for visibility of dropped packets (i.e., receiving notice of drop counters increase, seeing content of the dropped packets, debugging, and more). WJH is a service in devlink context and it is already implemented in the switch.
• Note: processing dropped packets (even for visibility purposes) may cause a degradation in performance and leaves the driver vulnerable for malicious attacks. The feature is disabled by default.
• Supported traps:
• VLAN mismatch: existing generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong VLAN tag
• DMAC mismatch: new generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong destination MAC
• Added support for enabling Relaxed Ordering for Kernel ULPs. Using relaxed ordering can improve performance in some setups. Since kernel ULPs are expected to support RO, it is enabled for them by default so they can benefit from it.

For ConnectX-4 Adapters and above: 

• In MLNX_OFED 5.4 GA, ConnectX-4/5/6 Ethernet network interfaces are now provided with permanent names. Prior to this release, the default interface names were provided by the kernel and udevd (ethX) remained as-is. From this release onwards, interface names are generated via new udevd rules. The generated names are now predictable, and the default names are automatically renamed to the predictable names by the udevd daemon, according to udev rules files installed by OFED.

For ConnectX-5 Adapters and above:

• Support to allow OVS (OpenVSwitch) kernel to support up to 128 matches (groups) per table and 16M entries per group.
• Offload ct_state flags rpl, inv, and rel.
• For rpl, support was added for both set and not set matching offload (i.e., +rpl and -rpl).
• For inv and rel, support was added only for the not set option (i.e., -rel and -inv).
• Scalable functions (also called subfunctions):This feature enables the user to create, configure, and deploy a scalable functions (e.g., RDMA and networking applications) and to assign them to a container when a container is started via mlxdevm tool.A scalable function can also be deployed in an untrusted guest/host system from the NIC/DPU. This enables full configuration of the function and its representors from the NIC/DPU before giving the function for a container to run in a host system.
• Scalable function QoS and QoS group via mlxdevm's rate commands were added. Run "man mlxdevm port" for details.
• "Signature API" which, on supported devices, allows application-level data-integrity checks via a signature handover mechanism. Various signature types, including CRC32 and T10-DIF, can be automatically calculated and checked, stripped, or appended during the transfer at full wire speed.
• DR support for matching on RAW tunnel headers using the misc5 parameters, This feature allows matching on each bit of the header,
• inducing reserved fields.

For ConnectX-6 Dx Adapters:

• Mapping a QP (Queue Pair) to AH over DEVX API, which enables DC/UD QPs to use multiple CC algorithms in the same data center.
• Better insertion rate in software steering was supported. This includes multi-QP which skips areas in the code that may be for debug only.
• Improved rate of updating steering rules, insertion, and deletion. The feature includes definers, multi-qp approach, and better memory usage.
• Added support for ability to allow or prevent insertion of duplicate rules, so the user can choose one of the following behaviors:
• Prevent duplicate rules, so that already-existing rule and fail can be detected.
• Allow duplicate rules, to enable updating the rule's action (this will only take effect once the previous rule is deleted). By default, duplicate rules are allowed.
• Made it so that all rule's insertion occur in a defined time using defined (export) size of Htble and decreased use of dynamic allocation.

For all HCA's:

• As of version 5.4, the driver is set so that udev rules will change the names of network interfaces created from NVIDIA adapters.The udev rules are shipped to "/lib/udev/rules.d" and may be overridden by placing a file with the same name in "/etc/udev/rules.d".
• Example: /etc/udev/rules.d/82-net-setup-link.rules
• OvS-DPDK deprecated the command "ovs-appctl dpctl/dump-e2e-stats". Instead, the command has been integrated into the existing command "ovs-appctl dpctl/offload-stats-show -m" (when e2e-cache is enabled).
• OvS-DPDK ct-ct-nat offloads is now disabled by default.
• Before version 5.4, /etc/udev/rules.d/90-ib.rules was potentially automatically edited by installation scripts in case the options —umad-dev-rw or —umad-dev-na were used. From version 5.4 and above, those changes are made in /etc/udev/rules.d/91-ib-permissions.rules which (if exist) only include the settings for those command-line options.
• ibv_query_qp_data_in_order() API: This API enables an application to check if the given QP data is guaranteed to be in order, enabling poll for data instead of poll for completion.
• Added ethtool extended link state to mlx5e. ethtool can be used to get more information to help troubleshoot the state.
• Moved all Python scripts and some other common scripts out of the mlnxofa_ kernel packages. This removed the python dependency from that package when rebuilding it and avoided unnecessary errors when rebuilding them for custom kernels.

For ConnectX-4 Adapters:

• Updated mlx5 driver to use auxiliary bus in order to integrate different driver components into driver core and optimize module load/unload sequences.
• WJH (What Just Happened) in NICs: WJH allows for visibility of dropped packets (i.e., receiving notice of drop counters increase, seeing content of the dropped packets, debugging, and more). WJH is a service in devlink context and it is already implemented in the switch.
• Note: processing dropped packets (even for visibility purposes) may cause a degradation in performance and leaves the driver vulnerable for malicious attacks. The feature is disabled by default.
• Supported traps:
• VLAN mismatch: existing generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong VLAN tag
• DMAC mismatch: new generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong destination MAC
• Added support for enabling Relaxed Ordering for Kernel ULPs. Using relaxed ordering can improve performance in some setups. Since kernel ULPs are expected to support RO, it is enabled for them by default so they can benefit from it.

For ConnectX-4 Adapters and above: 

• In MLNX_OFED 5.4 GA, ConnectX-4/5/6 Ethernet network interfaces are now provided with permanent names. Prior to this release, the default interface names were provided by the kernel and udevd (ethX) remained as-is. From this release onwards, interface names are generated via new udevd rules. The generated names are now predictable, and the default names are automatically renamed to the predictable names by the udevd daemon, according to udev rules files installed by OFED.

For ConnectX-5 Adapters and above:

• Support to allow OVS (OpenVSwitch) kernel to support up to 128 matches (groups) per table and 16M entries per group.
• Offload ct_state flags rpl, inv, and rel.
• For rpl, support was added for both set and not set matching offload (i.e., +rpl and -rpl).
• For inv and rel, support was added only for the not set option (i.e., -rel and -inv).
• Scalable functions (also called subfunctions):This feature enables the user to create, configure, and deploy a scalable functions (e.g., RDMA and networking applications) and to assign them to a container when a container is started via mlxdevm tool.A scalable function can also be deployed in an untrusted guest/host system from the NIC/DPU. This enables full configuration of the function and its representors from the NIC/DPU before giving the function for a container to run in a host system.
• Scalable function QoS and QoS group via mlxdevm's rate commands were added. Run "man mlxdevm port" for details.
• "Signature API" which, on supported devices, allows application-level data-integrity checks via a signature handover mechanism. Various signature types, including CRC32 and T10-DIF, can be automatically calculated and checked, stripped, or appended during the transfer at full wire speed.
• DR support for matching on RAW tunnel headers using the misc5 parameters, This feature allows matching on each bit of the header,
• inducing reserved fields.

For ConnectX-6 Dx Adapters:

• Mapping a QP (Queue Pair) to AH over DEVX API, which enables DC/UD QPs to use multiple CC algorithms in the same data center.
• Better insertion rate in software steering was supported. This includes multi-QP which skips areas in the code that may be for debug only.
• Improved rate of updating steering rules, insertion, and deletion. The feature includes definers, multi-qp approach, and better memory usage.
• Added support for ability to allow or prevent insertion of duplicate rules, so the user can choose one of the following behaviors:
• Prevent duplicate rules, so that already-existing rule and fail can be detected.
• Allow duplicate rules, to enable updating the rule's action (this will only take effect once the previous rule is deleted). By default, duplicate rules are allowed.
• Made it so that all rule's insertion occur in a defined time using defined (export) size of Htble and decreased use of dynamic allocation.

For all HCA's:

• As of version 5.4, the driver is set so that udev rules will change the names of network interfaces created from NVIDIA adapters.The udev rules are shipped to "/lib/udev/rules.d" and may be overridden by placing a file with the same name in "/etc/udev/rules.d".
• Example: /etc/udev/rules.d/82-net-setup-link.rules
• OvS-DPDK deprecated the command "ovs-appctl dpctl/dump-e2e-stats". Instead, the command has been integrated into the existing command "ovs-appctl dpctl/offload-stats-show -m" (when e2e-cache is enabled).
• OvS-DPDK ct-ct-nat offloads is now disabled by default.
• Before version 5.4, /etc/udev/rules.d/90-ib.rules was potentially automatically edited by installation scripts in case the options —umad-dev-rw or —umad-dev-na were used. From version 5.4 and above, those changes are made in /etc/udev/rules.d/91-ib-permissions.rules which (if exist) only include the settings for those command-line options.
• ibv_query_qp_data_in_order() API: This API enables an application to check if the given QP data is guaranteed to be in order, enabling poll for data instead of poll for completion.
• Added ethtool extended link state to mlx5e. ethtool can be used to get more information to help troubleshoot the state.
• Moved all Python scripts and some other common scripts out of the mlnxofa_ kernel packages. This removed the python dependency from that package when rebuilding it and avoided unnecessary errors when rebuilding them for custom kernels.

The raw driver files can be obtained by extracting the Smart Component and then extracting the Emulex installer.  Use this command:

    elxdrvr-fc-version.exe /q2 extract=2

The extracted files are located:

    C:\Users\Administrator\Documents\Emulex\Drivers\FC-version

Each kit folder has subsequent architecture folders with subsequent OS folders.  For example,

    C:\Users\Administrator\Documents\Emulex\Drivers\FC-version\x64\win2019

The raw driver files can be obtained by extracting the Smart Component and then extracting the Emulex installer.  Use this command:

    elxdrvr-fc-version.exe /q2 extract=2

The extracted files are located:

    C:\Users\Administrator\Documents\Emulex\Drivers\FC-version

Each kit folder has subsequent architecture folders with subsequent OS folders.  For example,

    C:\Users\Administrator\Documents\Emulex\Drivers\FC-version\x64\win2019

Firmware for the following devices are updated to 2.42.5056:
    764283-B21
    764284-B21

Firmware for the following device is updated to 2.42.5700:
    764285-B21

New features in firmware version 2.42.5000:

• Added support for the following features.
  • new TLV: CX3_GLOBAL_CONF to enable/disable timestamp on incoming packets through mlxconfig configuration.
  • User MAC configuration.
  • Automatically collecting mstdump before driver reset.
  • A mechanism to detect DEAD_IRISC (plastic) from TPT (iron) and raise an assert.
  • A new field is added to "set port" command which notifies the firmware what is the user_mtu size.
• Improved the debug ability for command timeout cases.

New features and changes in firmware version 2.42.5700.

• Modified the mlx_cmd_get_mlx_link_status command return value to return "Link Type = Ethernet" in Ethernet adapter cards.

Firmware for the following devices are updated to 2.42.5056:
    764283-B21
    764284-B21

Firmware for the following device is updated to 2.42.5700:
    764285-B21

New features in firmware version 2.42.5000:

• Added support for the following features.
  • new TLV: CX3_GLOBAL_CONF to enable/disable timestamp on incoming packets through mlxconfig configuration.
  • User MAC configuration.
  • Automatically collecting mstdump before driver reset.
  • A mechanism to detect DEAD_IRISC (plastic) from TPT (iron) and raise an assert.
  • A new field is added to "set port" command which notifies the firmware what is the user_mtu size.
• Improved the debug ability for command timeout cases.

New features and changes in firmware version 2.42.5700.

• Modified the mlx_cmd_get_mlx_link_status command return value to return "Link Type = Ethernet" in Ethernet adapter cards.

Firmware for the following devices are updated to 2.42.5056:
    764283-B21
    764284-B21

Firmware for the following device is updated to 2.42.5700:
    764285-B21

New features in firmware version 2.42.5000:

• Added support for the following features.
  • new TLV: CX3_GLOBAL_CONF to enable/disable timestamp on incoming packets through mlxconfig configuration.
  • User MAC configuration.
  • Automatically collecting mstdump before driver reset.
  • A mechanism to detect DEAD_IRISC (plastic) from TPT (iron) and raise an assert.
  • A new field is added to "set port" command which notifies the firmware what is the user_mtu size.
• Improved the debug ability for command timeout cases.

New features and changes in firmware version 2.42.5700.

• Modified the mlx_cmd_get_mlx_link_status command return value to return "Link Type = Ethernet" in Ethernet adapter cards.

Firmware for the following devices are updated to 2.42.5056:
    764283-B21
    764284-B21

Firmware for the following device is updated to 2.42.5700:
    764285-B21

New features in firmware version 2.42.5000:

• Added support for the following features.
  • new TLV: CX3_GLOBAL_CONF to enable/disable timestamp on incoming packets through mlxconfig configuration.
  • User MAC configuration.
  • Automatically collecting mstdump before driver reset.
  • A mechanism to detect DEAD_IRISC (plastic) from TPT (iron) and raise an assert.
  • A new field is added to "set port" command which notifies the firmware what is the user_mtu size.
• Improved the debug ability for command timeout cases.

New features and changes in firmware version 2.42.5700.

• Modified the mlx_cmd_get_mlx_link_status command return value to return "Link Type = Ethernet" in Ethernet adapter cards.

• HP FlexFabric 10Gb 2-port 556FLR-T Adapter
• HP FlexFabric 10Gb 2-port 556FLR-SFP+ Adapter

• HP FlexFabric 10Gb 2-port 556FLR-T Adapter
• HP FlexFabric 10Gb 2-port 556FLR-SFP+ Adapter

• Updated to support Service Pack for ProLiant version 2017.07.0.
  Note: If component version 2.68.64.2(A) was previously installed, then it is not necessary to upgrade to version 2.68.64.2(B).

Enhancements/New Features implemented in version 2.68.64.2(A):

Added support for:

• Microsoft Windows Server 2016 - Server Core and Server with a Desktop.
• Revised component package. No change to driver functionality resulting from the change. If system has previously updated to driver version 2.68.64.2, then it is not necessary to update to 2.68.64.2(A).

• Updated to support Service Pack for ProLiant version 2017.07.0.
  Note: If component version 2.68.64.2(A) was previously installed, then it is not necessary to upgrade to version 2.68.64.2(B).

Enhancements/New Features implemented in version 2.68.64.2(A):

Added support for:

• Microsoft Windows Server 2016 - Server Core and Server with a Desktop.
• Revised component package. No change to driver functionality resulting from the change. If system has previously updated to driver version 2.68.64.2, then it is not necessary to update to 2.68.64.2(A).

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.

 HPE Superdome Flex 280  
 HPE Superdome Flex Server  
 HPE Superdome Flex 2 

 HPE Superdome Flex 280  
 HPE Superdome Flex Server  
 HPE Superdome Flex 2 

For ConnectX-4 Adapters:

• Updated mlx5 driver to use auxiliary bus in order to integrate different driver components into driver core and optimize module load/unload sequences.
• WJH (What Just Happened) in NICs: WJH allows for visibility of dropped packets (i.e., receiving notice of drop counters increase, seeing content of the dropped packets, debugging, and more). WJH is a service in devlink context and it is already implemented in the switch.
• Note: processing dropped packets (even for visibility purposes) may cause a degradation in performance and leaves the driver vulnerable for malicious attacks. The feature is disabled by default.
• Supported traps:
• VLAN mismatch: existing generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong VLAN tag
• DMAC mismatch: new generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong destination MAC
• Added support for enabling Relaxed Ordering for Kernel ULPs. Using relaxed ordering can improve performance in some setups. Since kernel ULPs are expected to support RO, it is enabled for them by default so they can benefit from it.

For ConnectX-4 Adapters and above: 

• In MLNX_OFED 5.4 GA, ConnectX-4/5/6 Ethernet network interfaces are now provided with permanent names. Prior to this release, the default interface names were provided by the kernel and udevd (ethX) remained as-is. From this release onwards, interface names are generated via new udevd rules. The generated names are now predictable, and the default names are automatically renamed to the predictable names by the udevd daemon, according to udev rules files installed by OFED.

For ConnectX-5 Adapters and above:

• Support to allow OVS (OpenVSwitch) kernel to support up to 128 matches (groups) per table and 16M entries per group.
• Offload ct_state flags rpl, inv, and rel.
• For rpl, support was added for both set and not set matching offload (i.e., +rpl and -rpl).
• For inv and rel, support was added only for the not set option (i.e., -rel and -inv).
• Scalable functions (also called subfunctions):This feature enables the user to create, configure, and deploy a scalable functions (e.g., RDMA and networking applications) and to assign them to a container when a container is started via mlxdevm tool.A scalable function can also be deployed in an untrusted guest/host system from the NIC/DPU. This enables full configuration of the function and its representors from the NIC/DPU before giving the function for a container to run in a host system.
• Scalable function QoS and QoS group via mlxdevm's rate commands were added. Run "man mlxdevm port" for details.
• "Signature API" which, on supported devices, allows application-level data-integrity checks via a signature handover mechanism. Various signature types, including CRC32 and T10-DIF, can be automatically calculated and checked, stripped, or appended during the transfer at full wire speed.
• DR support for matching on RAW tunnel headers using the misc5 parameters, This feature allows matching on each bit of the header,
• inducing reserved fields.

For ConnectX-6 Dx Adapters:

• Mapping a QP (Queue Pair) to AH over DEVX API, which enables DC/UD QPs to use multiple CC algorithms in the same data center.
• Better insertion rate in software steering was supported. This includes multi-QP which skips areas in the code that may be for debug only.
• Improved rate of updating steering rules, insertion, and deletion. The feature includes definers, multi-qp approach, and better memory usage.
• Added support for ability to allow or prevent insertion of duplicate rules, so the user can choose one of the following behaviors:
• Prevent duplicate rules, so that already-existing rule and fail can be detected.
• Allow duplicate rules, to enable updating the rule's action (this will only take effect once the previous rule is deleted). By default, duplicate rules are allowed.
• Made it so that all rule's insertion occur in a defined time using defined (export) size of Htble and decreased use of dynamic allocation.

For all HCA's:

• As of version 5.4, the driver is set so that udev rules will change the names of network interfaces created from NVIDIA adapters.The udev rules are shipped to "/lib/udev/rules.d" and may be overridden by placing a file with the same name in "/etc/udev/rules.d".
• Example: /etc/udev/rules.d/82-net-setup-link.rules
• OvS-DPDK deprecated the command "ovs-appctl dpctl/dump-e2e-stats". Instead, the command has been integrated into the existing command "ovs-appctl dpctl/offload-stats-show -m" (when e2e-cache is enabled).
• OvS-DPDK ct-ct-nat offloads is now disabled by default.
• Before version 5.4, /etc/udev/rules.d/90-ib.rules was potentially automatically edited by installation scripts in case the options —umad-dev-rw or —umad-dev-na were used. From version 5.4 and above, those changes are made in /etc/udev/rules.d/91-ib-permissions.rules which (if exist) only include the settings for those command-line options.
• ibv_query_qp_data_in_order() API: This API enables an application to check if the given QP data is guaranteed to be in order, enabling poll for data instead of poll for completion.
• Added ethtool extended link state to mlx5e. ethtool can be used to get more information to help troubleshoot the state.
• Moved all Python scripts and some other common scripts out of the mlnxofa_ kernel packages. This removed the python dependency from that package when rebuilding it and avoided unnecessary errors when rebuilding them for custom kernels.

For ConnectX-4 Adapters:

• Updated mlx5 driver to use auxiliary bus in order to integrate different driver components into driver core and optimize module load/unload sequences.
• WJH (What Just Happened) in NICs: WJH allows for visibility of dropped packets (i.e., receiving notice of drop counters increase, seeing content of the dropped packets, debugging, and more). WJH is a service in devlink context and it is already implemented in the switch.
• Note: processing dropped packets (even for visibility purposes) may cause a degradation in performance and leaves the driver vulnerable for malicious attacks. The feature is disabled by default.
• Supported traps:
• VLAN mismatch: existing generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong VLAN tag
• DMAC mismatch: new generic trap DEVLINK_TRAP_GENERIC_ID_DMAC_MISMATCH Traps received packets with wrong destination MAC
• Added support for enabling Relaxed Ordering for Kernel ULPs. Using relaxed ordering can improve performance in some setups. Since kernel ULPs are expected to support RO, it is enabled for them by default so they can benefit from it.

For ConnectX-4 Adapters and above: 

• In MLNX_OFED 5.4 GA, ConnectX-4/5/6 Ethernet network interfaces are now provided with permanent names. Prior to this release, the default interface names were provided by the kernel and udevd (ethX) remained as-is. From this release onwards, interface names are generated via new udevd rules. The generated names are now predictable, and the default names are automatically renamed to the predictable names by the udevd daemon, according to udev rules files installed by OFED.

For ConnectX-5 Adapters and above:

• Support to allow OVS (OpenVSwitch) kernel to support up to 128 matches (groups) per table and 16M entries per group.
• Offload ct_state flags rpl, inv, and rel.
• For rpl, support was added for both set and not set matching offload (i.e., +rpl and -rpl).
• For inv and rel, support was added only for the not set option (i.e., -rel and -inv).
• Scalable functions (also called subfunctions):This feature enables the user to create, configure, and deploy a scalable functions (e.g., RDMA and networking applications) and to assign them to a container when a container is started via mlxdevm tool.A scalable function can also be deployed in an untrusted guest/host system from the NIC/DPU. This enables full configuration of the function and its representors from the NIC/DPU before giving the function for a container to run in a host system.
• Scalable function QoS and QoS group via mlxdevm's rate commands were added. Run "man mlxdevm port" for details.
• "Signature API" which, on supported devices, allows application-level data-integrity checks via a signature handover mechanism. Various signature types, including CRC32 and T10-DIF, can be automatically calculated and checked, stripped, or appended during the transfer at full wire speed.
• DR support for matching on RAW tunnel headers using the misc5 parameters, This feature allows matching on each bit of the header,
• inducing reserved fields.

For ConnectX-6 Dx Adapters:

• Mapping a QP (Queue Pair) to AH over DEVX API, which enables DC/UD QPs to use multiple CC algorithms in the same data center.
• Better insertion rate in software steering was supported. This includes multi-QP which skips areas in the code that may be for debug only.
• Improved rate of updating steering rules, insertion, and deletion. The feature includes definers, multi-qp approach, and better memory usage.
• Added support for ability to allow or prevent insertion of duplicate rules, so the user can choose one of the following behaviors:
• Prevent duplicate rules, so that already-existing rule and fail can be detected.
• Allow duplicate rules, to enable updating the rule's action (this will only take effect once the previous rule is deleted). By default, duplicate rules are allowed.
• Made it so that all rule's insertion occur in a defined time using defined (export) size of Htble and decreased use of dynamic allocation.

For all HCA's:

• As of version 5.4, the driver is set so that udev rules will change the names of network interfaces created from NVIDIA adapters.The udev rules are shipped to "/lib/udev/rules.d" and may be overridden by placing a file with the same name in "/etc/udev/rules.d".
• Example: /etc/udev/rules.d/82-net-setup-link.rules
• OvS-DPDK deprecated the command "ovs-appctl dpctl/dump-e2e-stats". Instead, the command has been integrated into the existing command "ovs-appctl dpctl/offload-stats-show -m" (when e2e-cache is enabled).
• OvS-DPDK ct-ct-nat offloads is now disabled by default.
• Before version 5.4, /etc/udev/rules.d/90-ib.rules was potentially automatically edited by installation scripts in case the options —umad-dev-rw or —umad-dev-na were used. From version 5.4 and above, those changes are made in /etc/udev/rules.d/91-ib-permissions.rules which (if exist) only include the settings for those command-line options.
• ibv_query_qp_data_in_order() API: This API enables an application to check if the given QP data is guaranteed to be in order, enabling poll for data instead of poll for completion.
• Added ethtool extended link state to mlx5e. ethtool can be used to get more information to help troubleshoot the state.
• Moved all Python scripts and some other common scripts out of the mlnxofa_ kernel packages. This removed the python dependency from that package when rebuilding it and avoided unnecessary errors when rebuilding them for custom kernels.

Updated to version 9.4.5.20

Updated to version 9.4.5.20

The raw driver files can be obtained by extracting the Smart Component and then extracting the Emulex installer.  Use this command:

    elxdrvr-fc-version.exe /q2 extract=2

The extracted files are located:

    C:\Users\Administrator\Documents\Emulex\Drivers\FC-version

Each kit folder has subsequent architecture folders with subsequent OS folders.  For example,

    C:\Users\Administrator\Documents\Emulex\Drivers\FC-version\x64\win2019

The raw driver files can be obtained by extracting the Smart Component and then extracting the Emulex installer.  Use this command:

    elxdrvr-fc-version.exe /q2 extract=2

The extracted files are located:

    C:\Users\Administrator\Documents\Emulex\Drivers\FC-version

Each kit folder has subsequent architecture folders with subsequent OS folders.  For example,

    C:\Users\Administrator\Documents\Emulex\Drivers\FC-version\x64\win2019

This product now supports the HPE ProLiant BL660c Gen9 Server. 

This product now supports the HPE ProLiant BL660c Gen9 Server. 

• HP FlexFabric 10Gb 2-port 556FLR-T Adapter
• HP FlexFabric 10Gb 2-port 556FLR-SFP+ Adapter

• HP FlexFabric 10Gb 2-port 556FLR-T Adapter
• HP FlexFabric 10Gb 2-port 556FLR-SFP+ Adapter

This Firmware package contains following firmware versions:

This Firmware package contains following firmware versions:

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.

Changes and New features in MFT version 4.17:

• Enabled Anti-rollback protection to prevent old vulnerable firmware versions from being flashed to the device.
• Flint now supports the "--activate_delay_sec" flag which performs the activation on the newly burned firmware after the specified delay.
• Remote mst device now supports cable devices. The remote cables will be shown on the mst status and can be accessed via the mlxcables tool.
• Added support for the following 
•     Parallel firmware burning. Although DMA burning is supported in Virtual Machines as well, burning in such scenarios might be slower than on Physical Machines
•     Dual Small Form Factor Pluggable(DSFP) modules in mlxlink.