i40e Linux* Base Driver for the Intel(R) XL710 Ethernet Controller Family
===============================================================================
===============================================================================
September 29, 2014
===============================================================================
Contents
--------
- Overview
- Identifying Your Adapter
- Building and Installation
- Command Line Parameters
- Intel(R) i40e Ethernet Flow Director
- Additional Features & Configurations
- Known Issues
===============================================================================
Overview
--------
This document describes the i40e Linux* Base Driver for the XL710 Ethernet
Controller Family of Adapters. The Linux base driver supports the 2.6.32 and
newer kernels and includes support for Linux supported x86_64 systems.
The following features are available in supported kernels:
- Native VLANs
- Channel Bonding (teaming)
- SNMP
- Generic Receive Offload
Adapter teaming is implemented using the native Linux Channel bonding
module. This is included in supported Linux kernels. Channel Bonding
documentation can be found in the Linux kernel source:
/Documentation/networking/bonding.txt
Driver information can be obtained using the ethtool, lspci, or iproute2
ip command. Instructions on updating ethtool can be found in the
Additional Features & Configurations section later on this page.
Identifying Your Adapter
------------------------
The driver in this release is compatible with XL710 and X710-based Intel
Ethernet Network Connections.
For more information on how to identify your adapter, go to the Adapter &
Driver ID Guide at:
http://support.intel.com/support/go/network/adapter/proidguide.htm
For the latest Intel network drivers for Linux, refer to the following
website. Select the link for your adapter.
http://support.intel.com/support/go/network/adapter/home.htm
SFP+ Devices with Pluggable Optics
----------------------------------
SR Modules
----------
Intel DUAL RATE 1G/10G SFP+ SR (bailed) FTLX8571D3BCV-IT
Intel DUAL RATE 1G/10G SFP+ SR (bailed) AFBR-703SDZ-IN2
LR Modules
----------
Intel DUAL RATE 1G/10G SFP+ LR (bailed) FTLX1471D3BCV-IT
Intel DUAL RATE 1G/10G SFP+ LR (bailed) AFCT-701SDZ-IN2
QSFP+ Modules
-------------
Intel TRIPLE RATE 1G/10G/40G QSFP+ SR (bailed) E40GQSFPSR
Intel TRIPLE RATE 1G/10G/40G QSFP+ SR (bailed) E40GQSFPLR
QSFP+ 1G speed is not supported on XL710 based devices.
X710/XL710 Based SFP+ adapters support all passive and active limiting direct
attach cables that comply with SFF-8431 v4.1 and SFF-8472 v10.4 specifications.
===============================================================================
Building and Installation
-------------------------
To build a binary RPM* package of this driver, run 'rpmbuild -tb i40e*.tar.gz'.
NOTES:
- 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.
- RPM functionality has only been tested in Red Hat distributions.
1. Move the base driver tar file to the directory of your choice. For
example, use '/home/username/i40e' or '/usr/local/src/i40e'.
2. Untar/unzip the archive, where <x.x.x> is the version number for the
driver tar file:
tar zxf i40e-<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 i40e-<x.x.x>/src/
4. Compile the driver module:
# make install
The binary will be installed as:
/lib/modules/`uname -r`/kernel/drivers/net/i40e/i40e.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:
modprobe i40e <parameter>=<value>
Make sure that any older i40e drivers are removed from the kernel before
loading the new module:
rmmod i40e; modprobe i40e
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>
NOTE:
For certain distributions like (but not limited to) RedHat Enterprise
Linux 7 and Ubuntu, once the driver is installed the initrd/initramfs
file may need to be updated to prevent the OS loading old versions
of the i40e driver. The dracut utility may be used on RedHat
distributions:
# dracut --force
For Ubuntu:
# update-initramfs -u
==============================================================================
Command Line Parameters
-----------------------
In general, ethtool and other OS specific commands are used to configure user
changeable parameters after the driver is loaded. The i40e driver only supports
the max_vfs kernel parameter on older kernels that do not have the standard
sysfs interface. The only other module parameter supported is the debug
parameter that can control the default logging verbosity of the driver.
If the driver is built as a module, the following optional parameters are used
by entering them on the command line with the modprobe command using this
syntax:
modprobe i40e [<option>=<VAL1>]
For example: modprobe i40e max_vfs=7
The default value for each parameter is generally the recommended setting
unless otherwise noted.
max_vfs
-------
Valid Range: 1-32 (X710 based devices)
1-64 (XL710 based devices)
Default Value: 0
NOTE: This parameter is only used on kernel 3.7.x and below. On kernel 3.8.x
and above, use sysfs to enable VFs. For example:
#echo $num_vf_enabled > /sys/class/net/$dev/device/sriov_numvfs //enable VFs
#echo 0 > /sys/class/net/$dev/device/sriov_numvfs //disable VFs
This parameter adds support for SR-IOV. It causes the driver to spawn up
to max_vfs worth of virtual functions. Some hardware configurations
support fewer SR-IOV instances, as the whole XL710 controller (all
functions) is limited to 128 SR-IOV interfaces in total.
NOTE: When SR-IOV mode is enabled, hardware VLAN filtering and VLAN tag
stripping/insertion will remain enabled. The old VLAN filter should be
removed before the new VLAN filter is added. For example:
ip link set eth0 vf 0 vlan 100 // set vlan 100 for VF 0
ip link set eth0 vf 0 vlan 0 // delete vlan 100
ip link set eth0 vf 0 vlan 200 // set a new vlan 200 for VF 0
Intel(R) i40e Ethernet Flow Director
------------------------------------
The Flow Director performs the following tasks:
- Directs receive packets according to their flows to different queues.
- Enables tight control on routing a flow in the platform.
- Matches flows and CPU cores for flow affinity.
- Supports multiple parameters for flexible flow classification and load
balancing.
NOTES:
- The Flow Director is enabled only if the kernel supports multiple
transmit queues.
- An included script (set_irq_affinity) automates setting the IRQ to
CPU affinity.
- The i40e Linux driver does not support configuration of the mask field.
It only accepts rules that completely qualify a certain flow type.
ethtool commands:
- To enable or disable the Flow Director
# ethtool -K ethX ntuple <on|off>
When disabling ntuple filters all the user programed filters are flushed
from the driver cache and hardware. Filters must be re-added if they are
needed when ntuple is re-enabled.
- To add a filter that directs packet to queue 2, use -U or -N switch
# ethtool -N ethX flow-type tcp4 src-ip 192.168.10.1 dst-ip \
192.168.10.2 src-port 2000 dst-port 2001 action 2 [loc 1]
- To see the list of filters currently present
# ethtool <-u|-n> ethX
Application Targeted Routing (ATR) Perfect Filters
--------------------------------------------------
ATR is enabled by default when the kernel is in multiple transmit queue mode.
An ATR flow director filter rule is added when a TCP-IP flow starts and is
deleted when the flow ends. When a TCP-IP Flow Director rule is added from
ethtool (Sideband filter), ATR is turned off by the driver. To re-enable ATR,
the sideband can be disabled with the ethtool -K option. If sideband is
re-enabled after ATR is re-enabled, ATR remains enabled until a TCP-IP flow
is added. When all TCP-IP sideband rules are deleted, ATR is automatically
re-enabled.
Packets that match the ATR rules are counted in fdir_atr_match stats in
ethtool, which also can be used to verify whether ATR rules still exist.
Sideband Perfect Filters
------------------------
Sideband Perfect Filters is an interface for loading the filter table that
funnels all flow into queue_0 unless an alternative queue is specified
using "action." If action is used, any flow that matches the filter criteria
will be directed to the appropriate queue. Rules may be deleted from the
table. This is done via
ethtool -U ethX delete N
where N is the rule number to be deleted, as specified in the loc value in
the filter add command.
If the queue is defined as -1, the filter drops matching packets. To account
for Sideband filter matches, the fdir_sb_match stats in ethtool can be used.
In addition, rx-N.rx_packets shows the number of packets processed by the
Nth queue.
NOTES:
- Receive Packet Steering (RPS) and Receive Flow Steering (RFS) are not
compatible with Flow Director. If Flow Director is enabled, these will
be disabled.
- The VLAN field for Flow Director is not explicitly supported in the i40e
driver.
When filter rules are added from Sideband or ATR and the Flow Director filter
table is full, the ATR rule is turned off by the driver. Subsequently, the
Sideband filter rule is then turned off. When space becomes available in the
filter table through filter rule deletion (i.e., an ATR rule or Sideband rule
is deleted), the Sideband and ATR rule additions are turned back on.
Occasionally, when the filter table is full, you will notice HW errors when
you try to add new rules. The i40e driver will call for a filter flush and
sideband filter list replay. This will help flush any stale ATR rules and
create space.
==============================================================================
Additional Features & Configurations
------------------------------------
Configuring the Driver on Different Distributions
-------------------------------------------------
Configuring a network driver to load properly when the system is started is
distribution dependent. Typically, the configuration process involves adding
an alias line to /etc/modules.conf or /etc/modprobe.conf as well as editing
other system startup scripts and/or configuration files. Many popular Linux
distributions ship with tools to make these changes for you. To learn the
proper way to configure a network device for your system, refer to your
distribution documentation. If during this process you are asked for the
driver or module name, the name for the Linux Base Driver for the
Intel(R) 40 Gigabit PCI Express Family of Adapters is i40e.
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.
Jumbo Frames
------------
Jumbo Frames support is enabled by changing the MTU to a value larger than
the default of 1500 bytes. The maximum value for the MTU is 9710. Use the
ifconfig command to increase the MTU size. For example, enter the following
where <x> is the interface number:
# ifconfig ethx mtu 9000 up
This setting is not saved across reboots. The setting change can be made
permanent by adding MTU = 9000 to the file
/etc/sysconfig/network-scripts/ifcfg-eth<x> for RHEL or to the file
/etc/sysconfig/network/<config_file> for SLES.
The maximum MTU setting for Jumbo Frames is 9710. This value coincides with
the maximum Jumbo Frames size of 9728. This driver will attempt to use
multiple page sized buffers to receive each jumbo packet. This should help
avoid buffer starvation issues when allocating receive packets.
ethtool
-------
The driver uses the ethtool interface for driver configuration and
diagnostics, as well as displaying statistical information. The latest
ethtool version is required for this functionality.
The latest release of ethtool can be found at:
http://sourceforge.net/projects/gkernel
Supported ethtool Commands and Options
--------------------------------------
-n --show-nfc
Retrieves the receive network flow classification configurations.
rx-flow-hash tcp4|udp4|ah4|esp4|sctp4|tcp6|udp6|ah6|esp6|sctp6
Retrieves the hash options for the specified network traffic type.
-N --config-nfc
Configures the receive network flow classification.
rx-flow-hash tcp4|udp4|ah4|esp4|sctp4|tcp6|udp6|ah6|esp6|sctp6 m|v|t|s|d|f|n|r...
Configures the hash options for the specified network traffic type.
udp4 UDP over IPv4
udp6 UDP over IPv6
f Hash on bytes 0 and 1 of the Layer 4 header of the rx packet.
n Hash on bytes 2 and 3 of the Layer 4 header of the rx packet.
NAPI
----
NAPI (Rx polling mode) is supported in the i40e driver.
See ftp://robur.slu.se/pub/Linux/net-development/NAPI/usenix-paper.tgz for
more information on NAPI.
Flow Control
------------
Flow control is disabled by default. To enable it, use ethtool:
# ethtool -A eth? autoneg off rx on tx on
NOTE: You must have a flow control capable link partner.
MAC and VLAN Anti-spoofing Feature
----------------------------------
When a malicious driver attempts to send a spoofed packet, it is dropped by
the hardware and not transmitted.
NOTE: This feature can be disabled for a specific Virtual Function (VF).
ip link set <pf dev> vf <vf id> spoofchk {off|on}
Support for UDP RSS
-------------------
This feature adds an on/off switch for hashing over certain flow types. The
default setting is enabled.
IEEE 1588 PTP
-------------
Precision Time Protocol (PTP) is used to synchronize clocks in a computer
network and is supported in the i40e driver.
The I40E_PTP is a compile time flag. The user may enable PTP at compile time
by adding CFLAGS_EXTRA="-DI40E_PTP" to the make file when it is being
compiled:
make CFLAGS_EXTRA="-DI40E_PTP" install
VXLAN Overlay HW Offloading
---------------------------
VXLAN Overlay HW Offloading is enabled by default. The i40e Linux driver
features VXLAN Overlay HW Offloading support. To view and configure
VXLAN on a VXLAN-overlay offload enabled device, use the following
commands:
# ethtool -k ethX
(This command displays the offloads and their current state.)
# ethtool -K ethX tx-udp_tnl-segmentation [off|on]
(This enables/disables VXLAN support in the driver.)
For more information on configuring your network for VXLAN overlay support,
refer to the Intel Technical Brief, "Creating Overlay Networks Using Intel
Ethernet Converged Network Adapters" (Intel Networking Division, August 2013):
http://www.intel.com/content/dam/www/public/us/en/documents/technology-briefs/
overlay-networks-using-converged-network-adapters-brief.pdf
===============================================================================
********************************************************************************
Known Issues
------------
Virtual machine does not get link
---------------------------------
If the virtual machine has more than one virtual port assigned to it, and those
virtual ports are bound to different physical ports, you may not get link on all
of the virtual ports. The following command may work around the issue:
ethtool r <PF>
Where <PF> is the PF interface in the host, for example: p5p1. You may need to
run the command more than once to get link on all virtual ports.
Enabling TSO may cause data integrity issues
--------------------------------------------
Enabling TSO on kernel 3.14 or newer may cause data integrity issues.
Kernel 3.10 and older do not exhibit this behavior.
Changing the number of Rx or Tx queues with ethtool -L may cause a kernel panic
-------------------------------------------------------------------------------
Changing the number of Rx or Tx queues with ethtool -L while traffic is flowing
and the interface is up may cause a kernel panic. Bring the interface down first
to avoid the issue. For example:
ip link set ethx down
ethtool -L ethx combined 4
Adding a Flow Director Sideband rule fails incorrectly
------------------------------------------------------
If you try to add a Flow Director rule when no more sideband rule space is
available, i40e logs an error that the rule could not be added, but ethtool
returns success. You can remove rules to free up space. In addition, remove
the rule that failed. This will evict it from the driver's cache.
Flow Director Sideband Logic adds duplicate filter
--------------------------------------------------
The Flow Director Sideband Logic adds a duplicate filter in the software filter
list if the location is not specified or the specified location differs from
the previous location but has the same filter criteria. In this case, the
second of the two filters that appear is the valid one in hardware and it
decides the filter action.
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: 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 /etc/sysctl.conf file:
net.ipv4.conf.all.arp_filter = 1
The change can also be made permanent by installing the interfaces in
separate broadcast domains (either in different switches or in a switch
partitioned to VLANs).
UDP Stress Test Dropped Packet Issue
------------------------------------
Under small packets UDP stress test with the i40e driver, the Linux system may
drop UDP packets due to the fullness of socket buffers. The driver Flow
Control variables may be changed to the minimum value for controlling packet
reception, or the kernel's default buffer sizes can be increased for UDP by
changing the values in /proc/sys/net/core/rmem_default and rmem_max.
Unplugging network cable while ethtool -p is running
----------------------------------------------------
In kernel versions 2.6.32 and later, unplugging the network cable while
ethtool -p is running will cause the system to become unresponsive to keyboard
commands, except for control-alt-delete. Restarting the system appears to be
the only remedy.
Rx page allocation errors
-------------------------
"order:0" errors may occur under stress with kernels 2.6.25 and newer. This is
caused by the way the Linux kernel reports this stressed condition.
Disable GRO when routing/bridging
---------------------------------
Due to a known kernel issue, GRO must be turned off when routing/bridging.
GRO can be disabled using ethtool:
# ethtool -K ethXgro off
where ethX is the Ethernet interface to be modified.
Lower than expected performance
-------------------------------
Some PCIe x8 slots are actually configured as x4 slots. These slots have
insufficient bandwidth for full line rate with dual port and quad port
devices. In addition, if you put a PCIe Generation 3-capable adapter
into a PCIe Generation 2 slot, you cannot get full bandwidth. The driver
detects this situation and writes the following message in the system log:
"PCI-Express bandwidth available for this card is not sufficient for optimal
performance. For optimal performance a x8 PCI-Express slot is required."
If this error occurs, moving your adapter to a true PCIe Generation 3 x8 slot
will resolve the issue.
ethtool may incorrectly display SFP+ fiber module as direct attached cable
--------------------------------------------------------------------------
Due to kernel limitations, port type can only be correctly displayed on kernel
2.6.33 or newer.
Running ethtool -t ethX command causes break between PF and test client
-----------------------------------------------------------------------
When there are active VFs, "ethtool -t" performs a full diagnostic. In the
process, it resets itself and all attached VFs. The VF drivers encouter a
disruption, but are able to recover.
Enabling SR-IOV in a 64-bit Microsoft* Windows Server* 2012/R2 guest OS
under Linux KVM
------------------------------------------------------------------------
KVM Hypervisor/VMM supports direct assignment of a PCIe device to a VM. This
includes traditional PCIe devices, as well as SR-IOV-capable devices using
Intel XL710-based controllers.
Unable to obtain DHCP lease on boot with RedHat
-----------------------------------------------
For configurations where the auto-negotiation process takes more than 5
seconds, the boot script may fail with the following message:
"ethX: failed. No link present. Check cable?"
If this error appears even though the presence of a link can be confirmed
using ethtool ethX, try setting LINKDELAY to 5 seconds (LINKDELAY=5) in
/etc/sysconfig/network-scripts/ifcfg-ethX.
NOTE: Link time can take up to 30 seconds. The LINKDELAY value should be
adjusted accordingly.
Alternatively, NetworkManager can be used to configure the interfaces, which
avoids the set timeout. For configuration instructions of NetworkManager
refer to the documentation provided by your distribution.
Loading i40e driver in 3.2.x and newer kernels displays kernel tained message
-----------------------------------------------------------------------------
Due to recent kernel changes, loading an out of tree driver causes the kernel
to be tainted.
Software bridging does not work with SR-IOV Virtual Functions
-------------------------------------------------------------
SR-IOV Virtual Functions are unable to send or receive traffic between VMs
using emulated connections on a Linux Software bridge and connections that use
SR-IOV VFs.
Packet Filter cannot forward packets to emulated device when it is in bridge
mode and SR-IOV is enabled or for NPAR enabled adapter
----------------------------------------------------------------------------
In SR-IOV enabled or NPAR enabled adapters, Physical Function (PF) does not
work in bridge mode. When a bridge is created on the PF device, an emulation
device in the Virtual Machine (VM) connects to this bridge cannot receive any
unicast packets.
To avoid this from occurring, for each emulated device (software Virtual
Station Interface [VSI]) added to the bridge the MAC address of the emulated
device (software Virtual Ethernet Bridge [VEB] VSI) needs to be added manually
to the forwarding database (FDB) filter table using the iproute2 package
bridge tool. This can be done by executing the following command:
# bridge fdb add <MAC ADDR> dev <PF device interface> self permanent
The FDB entry when the emulated device is no longer in use or the guest to
which it is assigned is moved to a different host must be deleted using this
command:
# bridge fdb del <MAC ADDR> dev <PF device interface>
===============================================================================
Support
-------
For general information, go to the Intel support website at:
www.intel.com/support/
or the Intel Wired Networking project hosted by Sourceforge at:
http://sourceforge.net/projects/e1000
If an issue is identified with the released source code on the supported
kernel with a supported adapter, email the specific information related to the
issue to e1000-devel@lists.sf.net.
===============================================================================
Trademarks
----------
Intel, Itanium, and Pentium are trademarks or registered trademarks of Intel
Corporation or its subsidiaries in the United States and other countries.
* Other names and brands may be claimed as the property of others.
Download Latest Version
ice-1.14.11.tar.gz (1.4 MB)
