Skip to content


Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?


Onload®️ is a high performance user-level network stack, which accelerates TCP and UDP network I/O for applications using the BSD sockets on Linux.


  • Low Application-to-application latency.
  • Binary compatible with existing applications.
  • Open Source (GPLv2.0 and BSD-2-Clause).

OpenOnload comprises a user-level shared library that intercepts network- related system calls and implements the protocol stack, and supporting kernel modules. It is compatible with the full system call API, including those aspects that are usually problematic for user-level networking, such as fork(), exec(), passing sockets through Unix domain sockets, and advancing the protocol when the application is not scheduled.

Installation and Quick Start Guide

OpenOnload is distributed as source code. Instructions for building, packaging and installing may be found in

For each interface on which Onload is to use AF_XDP, execute the following:

echo ens2f0 > /sys/module/sfc_resource/afxdp/register

Nota bene: for linux<5.11 you may need to run ulimit -l unlimited before the line above.

The application to be Onloaded should be launched by prefixing the command line with onload.


Please see

Onload with AF_XDP

Compatible network adapters, drivers and operating systems

OpenOnload can accelerate applications on operating systems with AF_XDP support. AF_XDP support needs Linux kernel version 5.3 or later. To support zero-copy, Onload needs AF_XDP network adapter drivers to implement the necessary AF_XDP primitives. Typically the latest drivers from the network adapter vendors will support these primitives.

The AF_XDP support is currently under development and is not yet at final release quality.

The following operating system distributions are known to provide an adequate level of AF_XDP support for Onload:

  • Ubuntu LTS 20.04, 22.04
  • Debian 10 with Linux kernel 5.10
  • Debian 11
  • Redhat Enterprise Linux 8.3 and newer
  • Redhat Enterprise Linux 9.x
  • Linux kernel in the range 5.4 - 5.19

If a netdriver does not support AF_XDP in native mode, Onload will try to use generic XDP mode when registering an interface. To make it work, one has to set up userland helper before registering the interface:

$ make -C ./src/tools/bpf_link_helper/
$ echo $(realpath ./src/tools/bpf_link_helper/bpf-link-helper) | sudo tee /sys/module/sfc_resource/parameters/bpf_link_helper

Building without Xilinx NICs, for AF_XDP only

OpenOnload can be built without SFC driver:

  • make: use HAVE_SFC=0 variable;
  • onload_build & onload_install: use --no-sfc parameter;
  • onload_tool reload: use --onload-only parameter.

Also, it can be built without EFCT and AUX support:

  • make: use HAVE_EFCT=0 variable;
  • onload_build: use --no-efct parameter.

Native Onload with Xilinx/AMD NICs

Onload also works with the native ef_vi hardware interface, supported by Xilinx network adapters. In this mode of operation, AF_XDP kernel and driver support is not required. This allows Onload to be used on older operating systems and take advantage of additional features. A version of the 'sfc' net driver for Xilinx network adapters is included.

Compatible Xilinx network adapters

The following adapters at least are able to support OpenOnload without AF_XDP:

  • X2541
  • X2522, X2522-25G
  • X3522
  • SFN8042
  • SFN8522, SFN8542

Compatible Linux kernels and distributions for Xilinx network adapters

This source tree is known to work with Xilinx network adapters on following Linux distributions:

  • Ubuntu LTS 18.04, LTS 20.04, LTS 22.04
  • Debian 10, 11
  • Redhat Enterprise Linux 7.9, 8.1 - 8.7, 9.0 - 9.1
  • SuSE Linux Enterprise Server 15 SP1 - SP4
  • Linux kernel in the range 4.15 - 5.19


The publicly-hosted repository is a community-supported project.

Supported releases of OpenOnload are available from

This file: (c) Copyright 2020-2022 Xilinx, Inc.