[dpdk-dev] [PATCH v3] doc: update NFP with PF support information
alejandro.lucero at netronome.com
Wed Sep 6 17:14:16 CEST 2017
NFP PMD has now support for both, PF and VFs. This patch updates
the guide and give some information about implications.
Signed-off-by: Alejandro Lucero <alejandro.lucero at netronome.com>
doc/guides/nics/nfp.rst | 78 +++++++++++++++++++++++++++++++++++++------------
1 file changed, 59 insertions(+), 19 deletions(-)
diff --git a/doc/guides/nics/nfp.rst b/doc/guides/nics/nfp.rst
index c732fb1..99a3b76 100644
@@ -1,5 +1,5 @@
.. BSD LICENSE
- Copyright(c) 2015 Netronome Systems, Inc. All rights reserved.
+ Copyright(c) 2015-2017 Netronome Systems, Inc. All rights reserved.
All rights reserved.
Redistribution and use in source and binary forms, with or without
@@ -38,31 +38,32 @@ up to 400 Gbps.
This document explains how to use DPDK with the Netronome Poll Mode
Driver (PMD) supporting Netronome's Network Flow Processor 6xxx
+(NFP-6xxx) and Netronome's Flow Processor 4xxx (NFP-4xxx).
-Currently the driver supports virtual functions (VFs) only.
+NFP is a SRIOV capable device and the PMD driver supports the physical
+function (PF) and virtual functions (VFs).
-Before using the Netronome's DPDK PMD some NFP-6xxx configuration,
+Before using the Netronome's DPDK PMD some NFP configuration,
which is not related to DPDK, is required. The system requires
-installation of **Netronome's BSP (Board Support Package)** which includes
-Linux drivers, programs and libraries.
+installation of **Netronome's BSP (Board Support Package)** along
+with some specific NFP firmware application. Netronome's NSP ABI
+version should be 0.20 or higher.
-If you have a NFP-6xxx device you should already have the code and
-documentation for doing this configuration. Contact
+If you have a NFP device you should already have the code and
+documentation for doing all this configuration. Contact
**support at netronome.com** to obtain the latest available firmware.
-The NFP Linux kernel drivers (including the required PF driver for the
-NFP) are available on Github at
+The NFP Linux netdev kernel driver for VFs is part of vanilla kernel
+since kernel version 4.5, and support for the PF since kernel version
+4.11. Support for older kernels can be obtained on Github at
**https://github.com/Netronome/nfp-drv-kmods** along with build
-DPDK runs in userspace and PMDs uses the Linux kernel UIO interface to
-allow access to physical devices from userspace. The NFP PMD requires
-the **igb_uio** UIO driver, available with DPDK, to perform correct
+NFP PMD needs to be used along with UIO ``igb_uio`` or VFIO (``vfio-pci``)
+Linux kernel driver.
Building the software
@@ -71,7 +72,7 @@ Netronome's PMD code is provided in the **drivers/net/nfp** directory.
Although NFP PMD has Netronome´s BSP dependencies, it is possible to
compile it along with other DPDK PMDs even if no BSP was installed before.
Of course, a DPDK app will require such a BSP installed for using the
+NFP PMD, along with a specific NFP firmware application.
Default PMD configuration is at **common_linuxapp configuration** file:
@@ -87,14 +88,53 @@ Driver compilation and testing
Refer to the document :ref:`compiling and testing a PMD for a NIC <pmd_build_and_test>`
+Using the PF
+NFP PMD has support for using the NFP PF as another DPDK port, but it does not
+have any functionality for controlling VFs. In fact, it is not possible to use
+the PMD with the VFs if the PF is being used by DPDK, that is, with the NFP PF
+bound to ``igb_uio`` or ``vfio-pci`` kernel drivers. Future DPDK version will
+have a PMD able to work with the PF and VFs at the same time and with the PF
+implementing VF management along with other PF-only functionalities/offloads.
+The PMD PF has extra work to do which will delay the DPDK app initialization
+like checking if a firmware is already available in the device, uploading the
+firmware if necessary, and configure the Link state properly when starting or
+stopping a PF port. Note that firmware upload is not always necessary which is
+the main delay for NFP PF PMD initialization.
+Depending on the Netronome product installed in the system, firmware files
+should be available under ``/lib/firmware/netronome``. DPDK PMD supporting the
+PF requires a specific link, ``/lib/firmware/netronome/nic_dpdk_default.nffw``,
+which should be created automatically with Netronome's Agilio products
+PF multiport support
+Some NFP cards support several physical ports with just one single PCI device.
+DPDK core is designed with the 1:1 relationship between PCI devices and DPDK
+ports, so NFP PMD PF support requires handling the multiport case specifically.
+During NFP PF initialization, the PMD will extract the information about the
+number of PF ports from the firmware and will create as many DPDK ports as
+Because the unusual relationship between a single PCI device and several DPDK
+ports, there are some limitations when using more than one PF DPDK ports: there
+is no support for RX interrupts and it is not possible either to use those PF
+ports with the device hotplug functionality.
-#. **Enable SR-IOV on the NFP-6xxx device:** The current NFP PMD works with
- Virtual Functions (VFs) on a NFP device. Make sure that one of the Physical
- Function (PF) drivers from the above Github repository is installed and
+#. **Enable SR-IOV on the NFP device:** The current NFP PMD supports the PF and
+ the VFs on a NFP device. However, it is not possible to work with both at the
+ same time because the VFs require the PF being bound to the NFP PF Linux
+ netdev driver. Make sure you are working with a kernel with NFP PF support or
+ get the drivers from the above Github repository and follow the instructions
+ for building and installing it.
Virtual Functions need to be enabled before they can be used with the PMD.
Before enabling the VFs it is useful to obtain information about the
More information about the dev