intel / ethernet-linux-iavf Goto Github PK

iavf Linux* Driver for Intel(R) Ethernet Adaptive Virtual Function
******************************************************************

August 13, 2024


Contents
^^^^^^^^

* iavf Linux* Driver for Intel(R) Ethernet Adaptive Virtual Function

  * Overview

  * Related Documentation

  * Identifying Your Adapter

  * Building and Installation

  * Command Line Parameters

  * Additional Features and Configurations

  * Performance Optimization

  * Known Issues/Troubleshooting

  * Support

  * License

  * Trademarks


Overview
========

The iavf virtual function (VF) driver supports virtual functions
generated by the physical function (PF) driver, with one or more VFs
enabled through sysfs.

The associated PF drivers for this VF driver are:

* ice

* i40e

The MTU size set on a VF should match the MTU size set on the PF. A
mismatch in MTU sizes may cause unexpected results.

SR-IOV requires the correct platform and OS support.

The guest OS loading this driver must support MSI-X interrupts.

For questions related to hardware requirements, refer to the
documentation supplied with your Intel Ethernet adapter. All hardware
requirements listed apply to use with Linux*.

Driver information can be obtained using ethtool, lspci, and ip.


Adaptive Virtual Function
-------------------------

Adaptive Virtual Function (AVF) allows the virtual function driver, or
VF, to adapt to changing feature sets of the physical function driver
(PF) with which it is associated. This allows system administrators to
update a PF without having to update all the VFs associated with it.
All AVFs have a single common device ID and branding string.

AVFs have a minimum set of features known as "base mode," but may
provide additional features depending on what features are available
in the PF with which the AVF is associated. The following are base
mode features:

* 4 Queue Pairs (QP) and associated Configuration Status Registers
  (CSRs) for Tx/Rx

* iavf descriptors and ring format

* Descriptor write-back completion

* 1 control queue, with iavf descriptors, CSRs and ring format

* 5 MSI-X interrupt vectors and corresponding iavf CSRs

* 1 Interrupt Throttle Rate (ITR) index

* 1 Virtual Station Interface (VSI) per VF

* 1 Traffic Class (TC), TC0

* Receive Side Scaling (RSS) with 64-entry indirection table and key,
  configured through the PF

* 1 unicast MAC address reserved per VF

* 8 MAC address filters for each VF on an Intel(R) Ethernet 800 Series
  devices

* 16 MAC address filters for each VF on an Intel(R) Ethernet 700
  Series device

* Stateless offloads - non-tunneled checksums

* AVF device ID

* HW mailbox is used for VF to PF communications (including on
  Windows)


Related Documentation
=====================

See the "Intel(R) Ethernet Adapters and Devices User Guide" for
additional information on . It is available on the Intel website at
https://cdrdv2.intel.com/v1/dl/getContent/705831/


Identifying Your Adapter
========================

This driver is compatible with virtual functions bound to devices
based on the following:

* Intel(R) Ethernet Controller E810-C

* Intel(R) Ethernet Controller E810-XXV

* Intel(R) Ethernet Connection E822-C

* Intel(R) Ethernet Connection E822-L

* Intel(R) Ethernet Connection E823-C

* Intel(R) Ethernet Connection E823-L

* Intel(R) Ethernet Controller I710

* Intel(R) Ethernet Controller X710

* Intel(R) Ethernet Controller XL710

* Intel(R) Ethernet Network Connection X722

* Intel(R) Ethernet Controller XXV710

* Intel(R) Ethernet Controller V710

For information on how to identify your adapter, and for the latest
Intel network drivers, refer to the Intel Support website at
https://www.intel.com/support.


Building and Installation
=========================


To Manually Build the Driver
----------------------------

1. Move the virtual function driver tar file to the directory of your
   choice. For example, use "/home/username/iavf" or
   "/usr/local/src/iavf".

2. Untar/unzip the archive, where "<x.x.x>" is the version number for
   the driver tar file:

      tar zxf iavf-<x.x.x>.tar.gz

3. Change to the driver src directory, where "<x.x.x>" is the version
   number for the driver tar:

      cd iavf-<x.x.x>/src/

4. Compile the driver module:

      make install

   The binary will be installed as:

      /lib/modules/<KERNEL VER>/updates/drivers/net/ethernet/intel/iavf/iavf.ko

   The install location listed above is the default location. This may
   differ for various Linux distributions.

5. Load the module using the modprobe command. To check the version of
   the driver and then load it:

      modinfo iavf
      modprobe iavf

   Alternately, make sure that any older iavf drivers are removed from
   the kernel before loading the new module:

      rmmod iavf; modprobe iavf

6. Assign an IP address to the interface by entering the following,
   where "ethX" is the interface name that was shown in dmesg after
   modprobe:

      ip address add <IP_address>/<netmask bits> dev <ethX>

7. Verify that the interface works. Enter the following, where
   "IP_address" is the IP address for another machine on the same
   subnet as the interface that is being tested:

      ping <IP_address>


To Build a Binary RPM Package of This Driver
--------------------------------------------

Note:

  RPM functionality has only been tested in Red Hat distributions.

1. Run the following command, where "<x.x.x>" is the version number
   for the driver tar file:

      rpmbuild -tb iavf-<x.x.x>.tar.gz

   Note:

     For the build to work properly, the currently running kernel MUST
     match the version and configuration of the installed kernel
     sources. If you have just recompiled the kernel, reboot the
     system before building.

2. After building the RPM, the last few lines of the tool output
   contain the location of the RPM file that was built. Install the
   RPM with one of the following commands, where "<RPM>" is the
   location of the RPM file:

      rpm -Uvh <RPM>

   or:

      dnf/yum localinstall <RPM>

3. If your distribution or kernel does not contain inbox support for
   auxiliary bus, you must also install the auxiliary RPM:

      rpm -Uvh <iavf RPM> <auxiliary RPM>

   or:

      dnf/yum localinstall <iavf RPM> <auxiliary RPM>

Note:

  On some distributions, the auxiliary RPM may fail to install due to
  missing kernel-devel headers. To workaround this issue, specify "--
  excludepath" during installation. For example:

     rpm -Uvh auxiliary-1.0.0-1.x86_64.rpm --excludepath=/lib/modules/3.10.0-957.el7.x86_64/source/include/linux/auxiliary_bus.h

Note:

  * To compile the driver on some kernel/arch combinations, you may
    need to install a package with the development version of libelf
    (e.g., libelf-dev, libelf-devel, elfutils-libelf-devel).

  * When compiling an out-of-tree driver, details will vary by
    distribution. However, you will usually need a kernel-devel RPM or
    some RPM that provides the kernel headers at a minimum. The RPM
    kernel-devel will usually fill in the link at "/lib/modules/'uname
    -r'/build".


Command Line Parameters
=======================

The iavf driver does not support any command line parameters.


Additional Features and Configurations
======================================


Viewing Link Messages
---------------------

Link messages will not be displayed to the console if the distribution
is restricting system messages. In order to see network driver link
messages on your console, set dmesg to eight by entering the
following:

   dmesg -n 8

Note:

  This setting is not saved across reboots.


ethtool
-------

The driver utilizes the ethtool interface for driver configuration and
diagnostics, as well as displaying statistical information. The latest
ethtool version is required for this functionality. Download it at
https://kernel.org/pub/software/network/ethtool/.


Setting VLAN Tag Stripping
--------------------------

If you have applications that require Virtual Functions (VFs) to
receive packets with VLAN tags, you can disable VLAN tag stripping for
the VF. The Physical Function (PF) processes requests issued from the
VF to enable or disable VLAN tag stripping.

Note:

  If the PF has assigned a VLAN to a VF, then requests from that VF to
  set VLAN tag stripping will be ignored.

To enable/disable VLAN tag stripping for a VF, issue the following
command from inside the VM in which you are running the VF:

   ethtool -K <ethX> rxvlan on/off

or, alternatively:

   ethtool --offload <ethX> rxvlan on/off


IEEE 802.1ad (QinQ) Support
---------------------------

The IEEE 802.1ad standard, informally known as QinQ, allows for
multiple VLAN IDs within a single Ethernet frame. VLAN IDs are
sometimes referred to as "tags," and multiple VLAN IDs are thus
referred to as a "tag stack." Tag stacks allow L2 tunneling and the
ability to separate traffic within a particular VLAN ID, among other
uses.

The following are examples of how to configure 802.1ad (QinQ):

   ip link add link eth0 eth0.24 type vlan proto 802.1ad id 24
   ip link add link eth0.24 eth0.24.371 type vlan proto 802.1Q id 371

Where "24" and "371" are example VLAN IDs.

Note:

  * 802.1ad (QinQ) is supported in 3.19 and later kernels.

  * VLAN protocols use the following EtherTypes:

    * 802.1Q = EtherType 0x8100

    * 802.1ad = EtherType 0x88A8

  * For QinQ traffic to work at MTU 1500, the L2 peer (switch port or
    another NIC) should be able to receive Ethernet frames of 1526
    bytes. Some third- party NICs support a maximum Ethernet frame
    size of 1522 bytes at MTU 1500, which will cause QinQ traffic to
    fail. To work around this issue, restrict the MTU on the Intel
    Ethernet device to 1496.


Double VLANs
------------

Devices based on the Intel(R) Ethernet 800 Series can process up to
two VLANs in a packet when all the following are installed:

* ice driver version 1.4.0 or later

* NVM version 2.4 or later

* ice DDP package version 1.3.21 or later

If you don't use the versions above, the only supported VLAN
configuration is single 802.1Q VLAN traffic.

When two VLAN tags are present in a packet, the outer VLAN tag can be
either 802.1Q or 802.1ad. The inner VLAN tag must always be 802.1Q.

Note:

  One limitation is that, for each VF, the PF can only allow VLAN
  hardware offloads (insertion and stripping) of one type, either
  802.1Q or 802.1ad.

To enable outer or single 802.1Q VLAN insertion and stripping and
disable 802.1ad VLAN insertion and stripping:

   ethtool -K <ethX> rxvlan on txvlan on rx-vlan-stag-hw-parse off
   tx-vlan-stag-hw-insert off

To enable outer or single 802.1ad VLAN insertion and stripping and
disable 802.1Q VLAN insertion and stripping:

   ethtool -K <ethX> rxvlan off txvlan off rx-vlan-stag-hw-parse on
   tx-vlan-stag-hw-insert on

To enable outer or single VLAN filtering if the VF supports modifying
VLAN filtering:

   ethtool -K <ethX> rx-vlan-filter on rx-vlan-stag-filter on

To disable outer or single VLAN filtering if the VF supports modifying
VLAN filtering:

   ethtool -K <ethX> rx-vlan-filter off rx-vlan-stag-filter off


Combining QinQ with SR-IOV VFs
------------------------------

We recommend you always configure a port VLAN for the VF from the PF.
If a port VLAN is not configured, the VF driver may only offload VLANs
via software. The PF allows all VLAN traffic to reach the VF and the
VF manages all VLAN traffic.

When the device is configured for double VLANs and the PF has
configured a port VLAN:

* The VF can only offload guest VLANs for 802.1Q traffic.

* The VF can only configure VLAN filtering rules for guest VLANs using
  802.1Q traffic.

However, when the device is configured for double VLANs and the PF has
NOT configured a port VLAN:

* You must use iavf driver version 4.1.0 or later to offload and
  filter VLANs.

* The PF turns on VLAN pruning and antispoof in the VF's VSI by
  default. The VF will not transmit or receive any tagged traffic
  until the VF requests a VLAN filter.

* The VF can offload (insert and strip) the outer VLAN tag of 802.1Q
  or 802.1ad traffic.

* The VF can create filter rules for the outer VLAN tag of both 802.1Q
  and 802.1ad traffic.

If the PF does not support double VLANs, the VF can hardware offload
single 802.1Q VLANs without a port VLAN.

When the PF is enabled for double VLANs, for iavf drivers before
version 4.1.x:

* VLAN hardware offloads and filtering are supported only when the PF
  has configured a port VLAN.

* VLAN filtering, insertion, and stripping will be software offloaded
  when no port VLAN is configured.

To see VLAN filtering and offload capabilities, use the following
command:

   ethtool -k <ethX> | grep vlan


Application Device Queues (ADQ)
-------------------------------

Application Device Queues (ADQ) allow you to dedicate one or more
queues to a specific application. This can reduce latency for the
specified application, and allow Tx traffic to be rate limited per
application.

The ADQ information contained here is specific to the iavf driver. For
more details, refer to the E810 ADQ Configuration Guide at:
https://cdrdv2.intel.com/v1/dl/getContent/609008.

Requirements:

* Kernel version: Varies by feature and the underlying PF device.
  Refer to the E810 ADQ Configuration Guide for more information on
  required kernel versions for different ADQ features on Intel(R)
  Ethernet 800 Series devices.

* Depending on the underlying PF device, ADQ cannot be enabled when
  the following features are enabled:

  * Data Center Bridging (DCB)

  * Multiple Functions per Port (MFP)

  * Sideband Filters

* If another driver (for example, DPDK) has set cloud filters, you
  cannot enable ADQ.

When ADQ is enabled:

* You cannot change RSS parameters, the number of queues, or the MAC
  address in the PF or VF. Delete the ADQ configuration before
  changing these settings.

* The driver supports subnet masks for IP addresses in the PF and VF.
  When you add a subnet mask filter, the driver forwards packets to
  the ADQ VSI instead of the main VSI.

* The VF supports a minimum rate limiting of 50Mbps, in increments of
  50Mbps up to the maximum link speed.

See Creating Traffic Class Filters in this README for more information
on configuring filters, including examples. See the E810 ADQ
Configuration Guide for detailed instructions.


Creating Traffic Classes
------------------------

Note:

  These instructions are not specific to ADQ configuration. Refer to
  the tc and tc-flower man pages for more information on creating
  traffic classes (TCs).

To create traffic classes on the interface:

1. Use the tc command to create traffic classes. You can create a
   maximum of 16 TCs from the VM on Intel(R) Ethernet 800 Series
   devices and 4 TCs from the VM on Intel(R) Ethernet 700 Series
   devices:

      tc qdisc add dev <ethX> root mqprio num_tc <tcs> map <priorities>
      queues <count1@offset1 ...> hw 1 mode channel shaper bw_rlimit
      min_rate <min_rate1 ...> max_rate <max_rate1 ...>

   Where:

   num_tc <tcs>:
      The number of TCs to use.

   map <priorities>:
      The map of priorities to TCs. You can map up to 16 priorities to
      TCs.

   queues <count1@offset1 ...>:
      For each TC, "<num queues>\@<offset>". The max total number of
      queues for all TCs is the number of cores.

   hw 1 mode channel:
      "channel" with "hw" set to 1 is a new hardware offload mode in
      mqprio that makes full use of the mqprio options, the TCs, the
      queue configurations, and the QoS parameters.

   shaper bw_rlimit:
      For each TC, sets the minimum and maximum bandwidth rates. The
      totals must be equal to or less than the port speed. This
      parameter is optional and is required only to set up the Tx
      rates.

   min_rate <min_rate1>:
      Sets the minimum bandwidth rate limit for each TC.

   max_rate <max_rate1 ...>:
      Sets the maximum bandwidth rate limit for each TC. You can set a
      min and max rate together.

   Note:

     * If you set "max_rate" to less than 50Mbps, then "max_rate" is
       rounded up to 50Mbps and a warning is logged in dmesg.

     * See the mqprio man page and the examples below for more
       information.

2. Verify the bandwidth limit using network monitoring tools such as
   "ifstat" or "sar -n DEV [interval] [number of samples]".

   Note:

     Setting up channels via ethtool ("ethtool -L") is not supported
     when the TCs are configured using mqprio.

3. Enable hardware TC offload on the interface:

      ethtool -K <ethX> hw-tc-offload on

4. Add clsact qdisc to enable adding ingress/egress filters for Rx/Tx:

      tc qdisc add dev <ethX> clsact

5. Verify successful TC creation after qdisc is created:

      tc qdisc show dev <ethX> ingress


Traffic Class Examples
~~~~~~~~~~~~~~~~~~~~~~

See the tc and tc-flower man pages for more information on traffic
control and TC flower filters.

To set up two TCs (tc0 and tc1), with 16 queues each, priorities 0-3
for tc0 and 4-7 for tc1, and max Tx rate set to 1Gbit for tc0 and
3Gbit for tc1:

   tc qdisc add dev ens4f0 root mqprio num_tc 2 map 0 0 0 0 1 1 1 1 queues
   16@0 16@16 hw 1 mode channel shaper bw_rlimit max_rate 1Gbit 3Gbit

Where:

map 0 0 0 0 1 1 1 1:
   Sets priorities 0-3 to use tc0 and 4-7 to use tc1.

queues 16@0 16@16:
   Assigns 16 queues to tc0 at offset 0 and 16 queues to tc1 at offset
   16.


Creating Traffic Class Filters
------------------------------

Note:

  These instructions are not specific to ADQ configuration.

After creating traffic classes, use the tc command to create filters
for traffic. Refer to the tc and tc-flower man pages for more
information.

To view all TC filters:

   tc filter show dev <ethX> ingress
   tc filter show dev <ethX> egress


TC Filter Examples
~~~~~~~~~~~~~~~~~~

To configure TCP TC filters, where:

protocol:
   Encapsulation protocol (valid options are IP and 802.1Q).

prio:
   Priority.

flower:
   Flow-based traffic control filter.

dst_ip:
   IP address of the device.

ip_proto:
   IP protocol to use (TCP or UDP).

dst_port:
   Destination port.

src_port:
   Source port.

skip_sw:
   Flag to add the rule only in hardware.

hw_tc:
   Route incoming traffic flow to this hardware TC. The TC count
   starts at 0. For example, "hw_tc 1" indicates that the filter is on
   the second TC.

vlan_id:
   VLAN ID.

* TCP: Destination IP + L4 Destination Port

  To route incoming TCP traffic with a matching destination IP address
  and destination port to the given TC:

     tc filter add dev <ethX> protocol ip ingress prio 1 flower dst_ip
     <ip_address> ip_proto tcp dst_port <port_number> skip_sw hw_tc 1

* TCP: Source IP + L4 Source Port

  To route outgoing TCP traffic with a matching source IP address and
  source port to the given TC associated with the given priority:

     tc filter add dev <ethX> protocol ip egress prio 1 flower src_ip
     <ip_address> ip_proto tcp src_port <port_number> action skbedit priority 1

* TCP: Destination IP + L4 Destination Port + VLAN Protocol

  To route incoming TCP traffic with a matching destination IP address
  and destination port to the given TC using the VLAN protocol
  (802.1Q):

     tc filter add dev <ethX> protocol 802.1Q ingress prio 1 flower
     dst_ip <ip address> eth_type ipv4 ip_proto tcp dst_port <port_number>
     vlan_id <vlan_id> skip_sw hw_tc 1

Note:

  You can add multiple filters to the device, using the same recipe
  (and requires no additional recipe resources), either on the same
  interface or on different interfaces. Each filter uses the same
  fields for matching, but can have different match values.

     tc filter add dev <ethX> protocol ip ingress prio 1 flower ip_proto
     tcp dst_port <port_number> skip_sw hw_tc 1

     tc filter add dev <ethX> protocol ip egress prio 1 flower ip_proto tcp
     src_port <port_number> action skbedit priority 1

  For example:

     tc filter add dev ens4f0 protocol ip ingress prio 1 flower ip_proto
     tcp dst_port 5555 skip_sw hw_tc 1

     tc filter add dev ens4f0 protocol ip egress prio 1 flower ip_proto
     tcp src_port 5555 action skbedit priority 1


RDMA in the VF
--------------

Devices based on the Intel(R) Ethernet 800 Series support RDMA in a
Linux VF, on supported Windows or Linux hosts.

The iavf driver supports the following RDMA protocols in the VF:

* iWARP (Internet Wide Area RDMA Protocol)

* RoCEv2 (RDMA over Converged Ethernet)

Note:

  RDMA in the VF is not supported on Intel(R) Ethernet X722 Series
  devices.

Refer to the README inside the irdma driver tarball for details on
configuring RDMA in the VF.

Note:

  To support VF RDMA, load the irdma driver on the host before
  creating VFs. Otherwise VF RDMA support may not be negotiated
  between the VF and PF driver.

The iavf driver allocates MSI-X resources for the VF RDMA instance
(irdma). The LAN iavf driver gets first priority and any leftover
MSI-X interrupts are used for VF RDMA.


Auxiliary Bus
-------------

Inter-Driver Communication (IDC) is the mechanism in which LAN drivers
(such as iavf) communicate with peer drivers (such as irdma). Starting
in kernel 5.11, Intel LAN and RDMA drivers use an auxiliary bus
mechanism for IDC.

RDMA functionality requires use of the auxiliary bus.

If your kernel supports the auxiliary bus, the LAN and RDMA drivers
will use the inbox auxiliary bus for IDC. For kernels lower than 5.11,
the base driver will automatically install an out-of-tree auxiliary
bus module.


Performance Optimization
========================

Driver defaults are meant to fit a wide variety of workloads, but if
further optimization is required, we recommend experimenting with the
settings in this section.


Rx Descriptor Ring Size
-----------------------

To reduce the number of Rx packet discards, increase the number of Rx
descriptors for each Rx ring using ethtool.

* Check if the interface is dropping Rx packets due to buffers being
  full ("rx_dropped.nic" can mean that there is no PCIe bandwidth):

     ethtool -S <ethX> | grep "rx_dropped"

* If the previous command shows drops on queues, it may help to
  increase the number of descriptors using "ethtool -G":

     ethtool -G <ethX> rx <N>

  Where "<N>" is the desired number of ring entries/descriptors

  This can provide temporary buffering for issues that create latency
  while the CPUs process descriptors.

Note:

  When you are handling a large number of connections in a VF, we
  recommend setting the number of Rx descriptors to 1024 or above. For
  example:

     ethtool -G <ethX> rx 2048


Known Issues/Troubleshooting
============================


Software Issues
---------------

If your Intel Ethernet Network Connection is not working after
installing the driver, verify that you have installed the correct
driver.


Linux bonding failures with VFs
-------------------------------

If you bind Virtual Functions (VFs) to an Intel(R) Ethernet 700 Series
device, the VF targets may fail when they become the active target. If
the MAC address of the VF is set by the PF (Physical Function) of the
device, when you add a target, or change the active-backup target,
Linux bonding tries to sync the backup target's MAC address to the
same MAC address as the active target. Linux bonding will fail at this
point. This issue will not occur if the VF's MAC address is not set by
the PF.

When using bonding mode 5 (i.e., balance-tlb or adaptive transmit load
balancing), if you add multiple VFs to the bond, they are assigned
duplicate MAC addresses. When the VFs are joined with the bond
interface, the Linux bonding driver sets the MAC address for the VFs
to the same value. The MAC address is based on the first active VF
added to that bond.

This results in balance-tlb mode not functioning as expected. PF
interfaces behave as expected. The presence of duplicate MAC addresses
may cause further issues, depending on your switch configuration.


Traffic is not being passed between VM and client
-------------------------------------------------

You may not be able to pass traffic between a client system and a
Virtual Machine (VM) running on a separate host if the Virtual
Function (VF, or Virtual NIC) is not in trusted mode and spoof
checking is enabled on the VF.

Note:

  This situation can occur in any combination of client, host, and
  guest operating system. See the readme for the PF driver for
  information on spoof checking and how to set the VF to trusted mode.


Using four traffic classes fails
--------------------------------

Do not try to reserve more than three traffic classes in the iavf
driver. Doing so will fail to set any traffic classes and will cause
the driver to write errors to stdout. Use a maximum of three queues to
avoid this issue.


Unexpected errors in dmesg when adding TCP filters on the VF
------------------------------------------------------------

When ADQ is configured and the VF is not in trusted mode, you may see
unexpected error messages in dmesg on the host when you try to add TCP
filters on the VF. This is due to the asynchronous design of the iavf
driver. The VF does not know whether it is trusted and appears to set
the filter, while the PF blocks the request and reports an error. See
the dmesg log in the host OS for details about the error.


Multiple log error messages on iavf driver removal
--------------------------------------------------

If you have several VFs and you remove the iavf driver, several
instances of the following log errors are written to the log:

   Unable to send opcode 2 to PF, err I40E_ERR_QUEUE_EMPTY, aq_err ok
   Unable to send the message to VF 2 aq_err 12
   ARQ Overflow Error detected


MAC address of Virtual Function changes unexpectedly
----------------------------------------------------

If a Virtual Function's MAC address is not assigned in the host, then
the VF driver will use a random MAC address. This random MAC address
may change each time the VF driver is reloaded. You can assign a
static MAC address in the host machine. This static MAC address will
survive a VF driver reload.


Driver buffer overflow fix
--------------------------

The fix to resolve CVE-2016-8105, referenced in Intel SA-00069
<https://www.intel.com/content/www/us/en/security-center/advisory
/intel-sa-00069.html>, is included in this and future versions of the
driver.


Compiling the driver
--------------------

When trying to compile the driver by running make install, the
following error may occur:

   Linux kernel source not configured - missing version.h

To solve this issue, create the "version.h" file by going to the Linux
source tree and entering:

   make include/linux/version.h


Multiple interfaces on same Ethernet broadcast network
------------------------------------------------------

Due to the default ARP behavior on Linux, it is not possible to have
one system on two IP networks in the same Ethernet broadcast domain
(non-partitioned switch) behave as expected. All Ethernet interfaces
will respond to IP traffic for any IP address assigned to the system.
This results in unbalanced receive traffic.

If you have multiple interfaces in a server, turn on ARP filtering by
entering the following:

   echo 1 > /proc/sys/net/ipv4/conf/all/arp_filter

This only works if your kernel's version is higher than 2.4.5.

Note:

  This setting is not saved across reboots. The configuration change
  can be made permanent by adding the following line to the file
  "/etc/sysctl.conf":

     net.ipv4.conf.all.arp_filter = 1

Alternatively, you can install the interfaces in separate broadcast
domains (either in different switches or in a switch partitioned to
VLANs).


Rx page allocation errors
-------------------------

Errors that read, "Page allocation failure. order:0," may occur under
stress with kernels 2.6.25 and newer. This is caused by the way the
Linux kernel reports this stressed condition.


Host may reboot after removing PF when VF is active in guest
------------------------------------------------------------

Kernel versions earlier than 3.2 do not unload the PF driver with
active VFs. Doing this will cause your VFs to stop working until you
reload the PF driver. It can also cause a spontaneous reboot of your
system.

Prior to unloading the PF driver, you must first ensure that all VFs
are no longer active. Do this by shutting down all VMs and unloading
the VF driver.


Older VF drivers on Intel Ethernet 800 Series adapters
------------------------------------------------------

Some Windows* VF drivers from Release 22.9 or older may encounter
errors when loaded on a PF, based on the Intel Ethernet 800 Series on
Linux KVM. You may see errors and the VF may not load. This issue does
not occur, starting with the following Windows VF drivers:

* v40e64, v40e65: Version 1.5.65.0 and newer

To resolve this issue, download and install the latest iavf driver.


SR-IOV virtual functions have identical MAC addresses
-----------------------------------------------------

When you create multiple SR-IOV virtual functions, the VFs may have
identical MAC addresses. Only one VF will pass traffic, and all
traffic on other VFs with identical MAC addresses will fail. This is
related to the "MACAddressPolicy=persistent" setting in
"/usr/lib/systemd/network/99-default.link".

To resolve this issue, edit the
"/usr/lib/systemd/network/99-default.link" file and change the
"MACAddressPolicy" line to "MACAddressPolicy=none". For more
information, see the systemd.link man page.


Support
=======

For general information, go to the Intel support website at
https://www.intel.com/support/

or the Intel Ethernet Linux project hosted by GitHub at
https://github.com/intel/ethernet-linux-iavf

If an issue is identified with the released source code on a supported
kernel with a supported adapter, contact Intel Customer Support at
https://www.intel.com/content/www/us/en/support/products/36773
/ethernet-products.html


License
=======

This program is free software; you can redistribute it and/or modify
it under the terms and conditions of the GNU General Public License,
version 2, as published by the Free Software Foundation.

This program is distributed in the hope it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
for more details.

You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 51 Franklin St - Fifth Floor, Boston, MA 02110-1301
USA.

The full GNU General Public License is included in this distribution
in the file called "COPYING".

Copyright (c) 2018 - 2024, Intel Corporation.


Trademarks
==========

Intel is a trademark or registered trademark of Intel Corporation or
its subsidiaries in the United States and/or other countries.

Other names and brands may be claimed as the property of others.