Changes and New Features
On This Page
The following are the new features and changes that were added in this version.
New Adapter Card Support
Added support for ConnectX-7 adapter cards.ConnectX-7 has the same feature set as the ConnectX-6 adapter card.
Note: Software Steering is not supported over ConnectX-7. As such, OVS-DPDK is also not supported.
See Bug fixes.
See Bug fixes.
See Bug fixes.
Kernel Space GPUDirect from VM
[ConnectX-5 and above] Added support for kernel space GPUDirect from the VM. To use GDS with high performance in a VM, set the ATS capability in ib_alloc_mr.
[ConnectX-6 Dx] Added support for per flow metering using OVS or TC, PPS, and BPS.
Slow Path Metering
[ConnectX-6 Dx] Added support for slow path metering on representor using OVS or TC, PPS, and BPS.
[ConnectX-6 Dx and BlueField-2] Added support to have physical port selection based on the hash function defined by the bond so that different packets of the same flow will be egress from the same physical port.
In order to enable this feature, set this mode for both bonded devices through the below sysfs before the device is in switchdev mode:
In order to have the legacy behaviour (queue affinity based selection), echo the following:
This feature requires to set LAG_RESOURCE_ALLOCATION to 1 with mlxconfig.
Single IRQ for PCI Function
[ConnectX-4 and above] Added support for single IRQ for PCI function. To use a high number of VFs, a large amount of IRQs is required which the device cannot always support. This feature enables VFs to function with a minimum of a single IRQ instead of two.
ethtool EEPROM Support for DSFP
[ConnectX-6 Dx] Added support for reading DSFP module information. The change includes adding new options to ethtool netlink EEPROM module read API to read a specific page and bank.
Dynamic VF MSI-X Allocation
[ConnectX-5 and above] Added support for dynamic assignment of MSI-X vector count.
DV API for DMA GGA memcpy
[BlueField-2 and above] DMA memcpy is one of several Memory-to-Memory Offloads (MMO) available from BlueField-2 onwards. It utilizes the GGA modules on the DPU to perform DMA memcpy, thus improving performance. The memcpy can be done locally, on the same host, or between the host and the Arm.
Set DR Matcher Layout
[ConnectX-6 Dx] Added support for a new RDMA CORE DR API to set the DR matcher layout by calling mlx5dv_dr_matcher_set_layout.
Flex Parsers misc4
[ConnectX-5 and ConnectX-6 Dx] Added ability to expose flex parsers 4-7 provided by misc4 to extend matching ability of flex parsers. Now all flex parsers can be matched at the same time.
Software encap Action
[ConnectX-5 and above] Added support for software encap action. There is requirement for more than 1M encap actions, but currently the encap action creation uses devx, which is very slow for 1M encap actions. As such, there is a need to support a way for software to create encap actions.
For additional information on the new features, please refer to MLNX_OFED User Manual.
Customer Affection Change
ar_mgr Subnet Manager Plugin
ar_mgr subnet manager plugin is no longer supported.
For adaptive routing and SHIELD subnet manager configuration, please see the MLNX_OFED user manual.
CUDA, UCX, HCOLL
For UCX-CUDA and hcoll-cuda, CUDA was upgraded from version 10.2 to 11.2.
MLNX_OFED Verbs API Migration
As of MLNX_OFED v5.0 release (Q1 of the year 2020), MLNX_OFED Verbs API have migrated from the legacy version of user space verbs libraries (libibervs, libmlx5, etc.) to the Upstream version rdma-core.
For the list of MLNX_OFED verbs APIs that have been migrated, refer to Migration to RDMA-Core document.