What can I help you with?
NVIDIA DPDK Documentation MLNX_DPDK_22.11_2310.5.1 LTS

Scalable Functions

The scalable functions are supported starting from this release. Before using it in DPDK, the configuration should be done firstly.

To set up the Scalable Functions follow the instructions in this link: https://github.com/Mellanox/scalablefunctions/wiki/MLNX_OFED-step-by-step-guide

The section “SubFunction support” in the file “doc/guides/nics/mlx5.rst” can be used as the reference to use scalable functions in DPDK.

  • To probe a SF (9 as the index in the example):

    Copy
    Copied!
                

    -a auxiliary:mlx5_core.sf.9

  • To probe SF representors:

    Copy
    Copied!
                

    -a <PCI_BDF>,representor=sf[0-N]

  • To probe SF representors in bonding mode:

    Copy
    Copied!
                

    -a <Primary_PCI_BDF>,representor=pf[0,1]sf[0-N]

SF and VF representors cannot be probed with one “-a” parameter, but they can be attached dynamically and separately.

Copy
Copied!
            

testpmd> port attach <PCI_BDF>,representor=sf<sfnum>,dv_flow_en=1

© Copyright 2024, NVIDIA. Last updated on Jan 9, 2025.