Skip to content

L
linux
Commit b87e7f24 authored by Jeff Kirsher's avatar Jeff Kirsher
Browse files
Options

Documentation: e1000e: Prepare documentation for RST conversion 

Before making the conversion to the RST (reStructured Text) format, there
are changes needed to the documentation so that there are no build errors.

Also fixed old/broken URLs to the correct or updated URL.
Signed-off-by: default avatarJeff Kirsher <jeffrey.t.kirsher@intel.com>
Tested-by: default avatarAndrew Bowers <andrewx.bowers@intel.com>
Tested-by: default avatarAaron Brown <aaron.f.brown@intel.com>
parent 8d59045f
Hide whitespace changes
Inline Side-by-side
Showing with 248 additions and 178 deletions
Documentation/networking/e1000e.txt
View file @ b87e7f24
.. SPDX-License-Identifier: GPL-2.0+
Linux* Driver for Intel(R) Ethernet Network Connection Linux* Driver for Intel(R) Ethernet Network Connection
====================================================== ======================================================
Intel Gigabit Linux driver. Intel Gigabit Linux driver.
Copyright(c) 1999 - 2013 Intel Corporation. Copyright(c) 2008-2018 Intel Corporation.
Contents Contents
======== ========
...@@ -12,48 +14,43 @@ Contents ...@@ -12,48 +14,43 @@ Contents
- Additional Configurations - Additional Configurations
- Support - Support
Identifying Your Adapter Identifying Your Adapter
======================== ========================
For information on how to identify your adapter, and for the latest Intel
network drivers, refer to the Intel Support website:
https://www.intel.com/support
The e1000e driver supports all PCI Express Intel(R) Gigabit Network
Connections, except those that are 82575, 82576 and 82580-based*.
* NOTE: The Intel(R) PRO/1000 P Dual Port Server Adapter is supported by
the e1000 driver, not the e1000e driver due to the 82546 part being used
behind a PCI Express bridge.
For more information on how to identify your adapter, go to the Adapter & Command Line Parameters
Driver ID Guide at: =======================
If the driver is built as a module, the following optional parameters are used
http://support.intel.com/support/go/network/adapter/idguide.htm by entering them on the command line with the modprobe command using this
syntax::
For the latest Intel network drivers for Linux, refer to the following modprobe e1000e [<option>=<VAL1>,<VAL2>,...]
website. In the search field, enter your adapter name or type, or use the
networking link on the left to search for your adapter:
http://support.intel.com/support/go/network/adapter/home.htm There needs to be a <VAL#> for each network port in the system supported by
this driver. The values will be applied to each instance, in function order.
For example::
Command Line Parameters modprobe e1000e InterruptThrottleRate=16000,16000
=======================
In this case, there are two network ports supported by e1000e in the system.
The default value for each parameter is generally the recommended setting, The default value for each parameter is generally the recommended setting,
unless otherwise noted. unless otherwise noted.
NOTES: For more information about the InterruptThrottleRate, NOTE: A descriptor describes a data buffer and attributes related to the data
RxIntDelay, TxIntDelay, RxAbsIntDelay, and TxAbsIntDelay buffer. This information is accessed by the hardware.
parameters, see the application note at:
http://www.intel.com/design/network/applnots/ap450.htm
InterruptThrottleRate InterruptThrottleRate
--------------------- ---------------------
Valid Range: 0,1,3,4,100-100000 (0=off, 1=dynamic, 3=dynamic conservative, :Valid Range: 0,1,3,4,100-100000
4=simplified balancing) :Default Value: 3
Default Value: 3
The driver can limit the amount of interrupts per second that the adapter Interrupt Throttle Rate controls the number of interrupts each interrupt
will generate for incoming packets. It does this by writing a value to the vector can generate per second. Increasing ITR lowers latency at the cost of
adapter that is based on the maximum amount of interrupts that the adapter increased CPU utilization, though it may help throughput in some circumstances.
will generate per second.
Setting InterruptThrottleRate to a value greater or equal to 100 Setting InterruptThrottleRate to a value greater or equal to 100
will program the adapter to send out a maximum of that many interrupts will program the adapter to send out a maximum of that many interrupts
...@@ -81,232 +78,305 @@ for small amounts of traffic and/or a significant percentage of small ...@@ -81,232 +78,305 @@ for small amounts of traffic and/or a significant percentage of small
packets; and "Lowest latency", for almost completely small packets or packets; and "Lowest latency", for almost completely small packets or
minimal traffic. minimal traffic.
In dynamic conservative mode, the InterruptThrottleRate value is set to 4000 - 0: Off
for traffic that falls in class "Bulk traffic". If traffic falls in the "Low Turns off any interrupt moderation and may improve small packet latency.
latency" or "Lowest latency" class, the InterruptThrottleRate is increased However, this is generally not suitable for bulk throughput traffic due
stepwise to 20000. This default mode is suitable for most applications. to the increased CPU utilization of the higher interrupt rate.
- 1: Dynamic mode
For situations where low latency is vital such as cluster or This mode attempts to moderate interrupts per vector while maintaining
grid computing, the algorithm can reduce latency even more when very low latency. This can sometimes cause extra CPU utilization. If
InterruptThrottleRate is set to mode 1. In this mode, which operates planning on deploying e1000e in a latency sensitive environment, this
the same as mode 3, the InterruptThrottleRate will be increased stepwise to parameter should be considered.
70000 for traffic in class "Lowest latency". - 3: Dynamic Conservative mode (default)
In dynamic conservative mode, the InterruptThrottleRate value is set to
In simplified mode the interrupt rate is based on the ratio of TX and 4000 for traffic that falls in class "Bulk traffic". If traffic falls in
RX traffic. If the bytes per second rate is approximately equal, the the "Low latency" or "Lowest latency" class, the InterruptThrottleRate is
interrupt rate will drop as low as 2000 interrupts per second. If the increased stepwise to 20000. This default mode is suitable for most
traffic is mostly transmit or mostly receive, the interrupt rate could applications.
be as high as 8000. - 4: Simplified Balancing mode
In simplified mode the interrupt rate is based on the ratio of TX and
Setting InterruptThrottleRate to 0 turns off any interrupt moderation RX traffic. If the bytes per second rate is approximately equal, the
and may improve small packet latency, but is generally not suitable interrupt rate will drop as low as 2000 interrupts per second. If the
for bulk throughput traffic. traffic is mostly transmit or mostly receive, the interrupt rate could
be as high as 8000.
NOTE: InterruptThrottleRate takes precedence over the TxAbsIntDelay and - 100-100000:
RxAbsIntDelay parameters. In other words, minimizing the receive Setting InterruptThrottleRate to a value greater or equal to 100
and/or transmit absolute delays does not force the controller to will program the adapter to send at most that many interrupts per second,
generate more interrupts than what the Interrupt Throttle Rate even if more packets have come in. This reduces interrupt load on the
allows. system and can lower CPU utilization under heavy load, but will increase
latency as packets are not processed as quickly.
NOTE: When e1000e is loaded with default settings and multiple adapters
are in use simultaneously, the CPU utilization may increase non- NOTE: InterruptThrottleRate takes precedence over the TxAbsIntDelay and
linearly. In order to limit the CPU utilization without impacting RxAbsIntDelay parameters. In other words, minimizing the receive and/or
the overall throughput, we recommend that you load the driver as transmit absolute delays does not force the controller to generate more
follows: interrupts than what the Interrupt Throttle Rate allows.
modprobe e1000e InterruptThrottleRate=3000,3000,3000
This sets the InterruptThrottleRate to 3000 interrupts/sec for
the first, second, and third instances of the driver. The range
of 2000 to 3000 interrupts per second works on a majority of
systems and is a good starting point, but the optimal value will
be platform-specific. If CPU utilization is not a concern, use
RX_POLLING (NAPI) and default driver settings.
RxIntDelay RxIntDelay
---------- ----------
Valid Range: 0-65535 (0=off) :Valid Range: 0-65535 (0=off)
Default Value: 0 :Default Value: 0
This value delays the generation of receive interrupts in units of 1.024 This value delays the generation of receive interrupts in units of 1.024
microseconds. Receive interrupt reduction can improve CPU efficiency if microseconds. Receive interrupt reduction can improve CPU efficiency if
properly tuned for specific network traffic. Increasing this value adds properly tuned for specific network traffic. Increasing this value adds extra
extra latency to frame reception and can end up decreasing the throughput latency to frame reception and can end up decreasing the throughput of TCP
of TCP traffic. If the system is reporting dropped receives, this value traffic. If the system is reporting dropped receives, this value may be set
may be set too high, causing the driver to run out of available receive too high, causing the driver to run out of available receive descriptors.
descriptors.
CAUTION: When setting RxIntDelay to a value other than 0, adapters may hang
CAUTION: When setting RxIntDelay to a value other than 0, adapters may (stop transmitting) under certain network conditions. If this occurs a NETDEV
hang (stop transmitting) under certain network conditions. If WATCHDOG message is logged in the system event log. In addition, the
this occurs a NETDEV WATCHDOG message is logged in the system controller is automatically reset, restoring the network connection. To
event log. In addition, the controller is automatically reset, eliminate the potential for the hang ensure that RxIntDelay is set to 0.
restoring the network connection. To eliminate the potential
for the hang ensure that RxIntDelay is set to 0.
RxAbsIntDelay RxAbsIntDelay
------------- -------------
Valid Range: 0-65535 (0=off) :Valid Range: 0-65535 (0=off)
Default Value: 8 :Default Value: 8
This value, in units of 1.024 microseconds, limits the delay in which a This value, in units of 1.024 microseconds, limits the delay in which a
receive interrupt is generated. Useful only if RxIntDelay is non-zero, receive interrupt is generated. This value ensures that an interrupt is
this value ensures that an interrupt is generated after the initial generated after the initial packet is received within the set amount of time,
packet is received within the set amount of time. Proper tuning, which is useful only if RxIntDelay is non-zero. Proper tuning, along with
along with RxIntDelay, may improve traffic throughput in specific network RxIntDelay, may improve traffic throughput in specific network conditions.
conditions.
TxIntDelay TxIntDelay
---------- ----------
Valid Range: 0-65535 (0=off) :Valid Range: 0-65535 (0=off)
Default Value: 8 :Default Value: 8
This value delays the generation of transmit interrupts in units of This value delays the generation of transmit interrupts in units of 1.024
1.024 microseconds. Transmit interrupt reduction can improve CPU microseconds. Transmit interrupt reduction can improve CPU efficiency if
efficiency if properly tuned for specific network traffic. If the properly tuned for specific network traffic. If the system is reporting
system is reporting dropped transmits, this value may be set too high dropped transmits, this value may be set too high causing the driver to run
causing the driver to run out of available transmit descriptors. out of available transmit descriptors.
TxAbsIntDelay TxAbsIntDelay
------------- -------------
Valid Range: 0-65535 (0=off) :Valid Range: 0-65535 (0=off)
Default Value: 32 :Default Value: 32
This value, in units of 1.024 microseconds, limits the delay in which a This value, in units of 1.024 microseconds, limits the delay in which a
transmit interrupt is generated. Useful only if TxIntDelay is non-zero, transmit interrupt is generated. It is useful only if TxIntDelay is non-zero.
this value ensures that an interrupt is generated after the initial It ensures that an interrupt is generated after the initial Packet is sent on
packet is sent on the wire within the set amount of time. Proper tuning, the wire within the set amount of time. Proper tuning, along with TxIntDelay,
along with TxIntDelay, may improve traffic throughput in specific may improve traffic throughput in specific network conditions.
network conditions.
Copybreak copybreak
--------- ---------
Valid Range: 0-xxxxxxx (0=off) :Valid Range: 0-xxxxxxx (0=off)
Default Value: 256 :Default Value: 256
Driver copies all packets below or equaling this size to a fresh RX The driver copies all packets below or equaling this size to a fresh receive
buffer before handing it up the stack. buffer before handing it up the stack.
This parameter differs from other parameters because it is a single (not 1,1,1
etc.) parameter applied to all driver instances and it is also available
during runtime at /sys/module/e1000e/parameters/copybreak.
To use copybreak, type::
This parameter is different than other parameters, in that it is a modprobe e1000e.ko copybreak=128
single (not 1,1,1 etc.) parameter applied to all driver instances and
it is also available during runtime at
/sys/module/e1000e/parameters/copybreak
SmartPowerDownEnable SmartPowerDownEnable
-------------------- --------------------
Valid Range: 0-1 :Valid Range: 0,1
Default Value: 0 (disabled) :Default Value: 0 (disabled)
Allows PHY to turn off in lower power states. The user can set this parameter Allows the PHY to turn off in lower power states. The user can turn off this
in supported chipsets. parameter in supported chipsets.
KumeranLockLoss KumeranLockLoss
--------------- ---------------
Valid Range: 0-1 :Valid Range: 0,1
Default Value: 1 (enabled) :Default Value: 1 (enabled)
This workaround skips resetting the PHY at shutdown for the initial This workaround skips resetting the PHY at shutdown for the initial silicon
silicon releases of ICH8 systems. releases of ICH8 systems.
IntMode IntMode
------- -------
Valid Range: 0-2 (0=legacy, 1=MSI, 2=MSI-X) :Valid Range: 0-2
Default Value: 2 :Default Value: 0
Allows changing the interrupt mode at module load time, without requiring a +-------+----------------+
recompile. If the driver load fails to enable a specific interrupt mode, the | Value | Interrupt Mode |
driver will try other interrupt modes, from least to most compatible. The +=======+================+
interrupt order is MSI-X, MSI, Legacy. If specifying MSI (IntMode=1) | 0 | Legacy |
interrupts, only MSI and Legacy will be attempted. +-------+----------------+
| 1 | MSI |
+-------+----------------+
| 2 | MSI-X |
+-------+----------------+
IntMode allows load time control over the type of interrupt registered for by
the driver. MSI-X is required for multiple queue support, and some kernels and
combinations of kernel .config options will force a lower level of interrupt
support.
This command will show different values for each type of interrupt::
cat /proc/interrupts
CrcStripping CrcStripping
------------ ------------
Valid Range: 0-1 :Valid Range: 0,1
Default Value: 1 (enabled) :Default Value: 1 (enabled)
Strip the CRC from received packets before sending up the network stack. If Strip the CRC from received packets before sending up the network stack. If
you have a machine with a BMC enabled but cannot receive IPMI traffic after you have a machine with a BMC enabled but cannot receive IPMI traffic after
loading or enabling the driver, try disabling this feature. loading or enabling the driver, try disabling this feature.
WriteProtectNVM WriteProtectNVM
--------------- ---------------
Valid Range: 0,1 :Valid Range: 0,1
Default Value: 1 :Default Value: 1 (enabled)
If set to 1, configure the hardware to ignore all write/erase cycles to the If set to 1, configure the hardware to ignore all write/erase cycles to the
GbE region in the ICHx NVM (in order to prevent accidental corruption of the GbE region in the ICHx NVM (in order to prevent accidental corruption of the
NVM). This feature can be disabled by setting the parameter to 0 during initial NVM). This feature can be disabled by setting the parameter to 0 during initial
driver load. driver load.
NOTE: The machine must be power cycled (full off/on) when enabling NVM writes NOTE: The machine must be power cycled (full off/on) when enabling NVM writes
via setting the parameter to zero. Once the NVM has been locked (via the via setting the parameter to zero. Once the NVM has been locked (via the
parameter at 1 when the driver loads) it cannot be unlocked except via power parameter at 1 when the driver loads) it cannot be unlocked except via power
cycle. cycle.
Additional Configurations Debug
========================= -----
:Valid Range: 0-16 (0=none,...,16=all)
:Default Value: 0
Jumbo Frames This parameter adjusts the level of debug messages displayed in the system logs.
------------
Jumbo Frames support is enabled by changing the MTU to a value larger than
the default of 1500. Use the ifconfig command to increase the MTU size. Additional Features and Configurations
For example: ======================================
Jumbo Frames
------------
Jumbo Frames support is enabled by changing the Maximum Transmission Unit (MTU)
to a value larger than the default value of 1500.
ifconfig eth<x> mtu 9000 up Use the ifconfig command to increase the MTU size. For example, enter the
following where <x> is the interface number::
This setting is not saved across reboots. ifconfig eth<x> mtu 9000 up
Notes: Alternatively, you can use the ip command as follows::
- The maximum MTU setting for Jumbo Frames is 9216. This value coincides ip link set mtu 9000 dev eth<x>
with the maximum Jumbo Frames size of 9234 bytes. ip link set up dev eth<x>
- Using Jumbo frames at 10 or 100 Mbps is not supported and may result in This setting is not saved across reboots. The setting change can be made
poor performance or loss of link. permanent by adding 'MTU=9000' to the file:
- Some adapters limit Jumbo Frames sized packets to a maximum of - For RHEL: /etc/sysconfig/network-scripts/ifcfg-eth<x>
4096 bytes and some adapters do not support Jumbo Frames. - For SLES: /etc/sysconfig/network/<config_file>
- Jumbo Frames cannot be configured on an 82579-based Network device, if NOTE: The maximum MTU setting for Jumbo Frames is 8996. This value coincides
MACSec is enabled on the system. with the maximum Jumbo Frames size of 9018 bytes.
ethtool NOTE: Using Jumbo frames at 10 or 100 Mbps is not supported and may result in
------- poor performance or loss of link.
The driver utilizes the ethtool interface for driver configuration and
diagnostics, as well as displaying statistical information. We
strongly recommend downloading the latest version of ethtool at:
https://kernel.org/pub/software/network/ethtool/ NOTE: The following adapters limit Jumbo Frames sized packets to a maximum of
4088 bytes:
NOTE: When validating enable/disable tests on some parts (82578, for example) - Intel(R) 82578DM Gigabit Network Connection
you need to add a few seconds between tests when working with ethtool. - Intel(R) 82577LM Gigabit Network Connection
Speed and Duplex The following adapters do not support Jumbo Frames:
----------------
Speed and Duplex are configured through the ethtool* utility. For
instructions, refer to the ethtool man page.
Enabling Wake on LAN* (WoL) - Intel(R) PRO/1000 Gigabit Server Adapter
--------------------------- - Intel(R) PRO/1000 PM Network Connection
WoL is configured through the ethtool* utility. For instructions on - Intel(R) 82562G 10/100 Network Connection
enabling WoL with ethtool, refer to the ethtool man page. - Intel(R) 82562G-2 10/100 Network Connection
- Intel(R) 82562GT 10/100 Network Connection
- Intel(R) 82562GT-2 10/100 Network Connection
- Intel(R) 82562V 10/100 Network Connection
- Intel(R) 82562V-2 10/100 Network Connection
- Intel(R) 82566DC Gigabit Network Connection
- Intel(R) 82566DC-2 Gigabit Network Connection
- Intel(R) 82566DM Gigabit Network Connection
- Intel(R) 82566MC Gigabit Network Connection
- Intel(R) 82566MM Gigabit Network Connection
- Intel(R) 82567V-3 Gigabit Network Connection
- Intel(R) 82577LC Gigabit Network Connection
- Intel(R) 82578DC Gigabit Network Connection
WoL will be enabled on the system during the next shut down or reboot. NOTE: Jumbo Frames cannot be configured on an 82579-based Network device if
For this driver version, in order to enable WoL, the e1000e driver must be MACSec is enabled on the system.
loaded when shutting down or rebooting the system.
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://www.kernel.org/pub/software/network/ethtool/
NOTE: When validating enable/disable tests on some parts (for example, 82578),
it is necessary to add a few seconds between tests when working with ethtool.
Speed and Duplex Configuration
------------------------------
In addressing speed and duplex configuration issues, you need to distinguish
between copper-based adapters and fiber-based adapters.
In the default mode, an Intel(R) Ethernet Network Adapter using copper
connections will attempt to auto-negotiate with its link partner to determine
the best setting. If the adapter cannot establish link with the link partner
using auto-negotiation, you may need to manually configure the adapter and link
partner to identical settings to establish link and pass packets. This should
only be needed when attempting to link with an older switch that does not
support auto-negotiation or one that has been forced to a specific speed or
duplex mode. Your link partner must match the setting you choose. 1 Gbps speeds
and higher cannot be forced. Use the autonegotiation advertising setting to
manually set devices for 1 Gbps and higher.
Speed, duplex, and autonegotiation advertising are configured through the
ethtool* utility.
Caution: Only experienced network administrators should force speed and duplex
or change autonegotiation advertising manually. The settings at the switch must
always match the adapter settings. Adapter performance may suffer or your
adapter may not operate if you configure the adapter differently from your
switch.
An Intel(R) Ethernet Network Adapter using fiber-based connections, however,
will not attempt to auto-negotiate with its link partner since those adapters
operate only in full duplex and only at their native speed.
Enabling Wake on LAN* (WoL)
---------------------------
WoL is configured through the ethtool* utility.
WoL will be enabled on the system during the next shut down or reboot. For
this driver version, in order to enable WoL, the e1000e driver must be loaded
prior to shutting down or suspending the system.
NOTE: Wake on LAN is only supported on port A for the following devices:
- Intel(R) PRO/1000 PT Dual Port Network Connection
- Intel(R) PRO/1000 PT Dual Port Server Connection
- Intel(R) PRO/1000 PT Dual Port Server Adapter
- Intel(R) PRO/1000 PF Dual Port Server Adapter
- Intel(R) PRO/1000 PT Quad Port Server Adapter
- Intel(R) Gigabit PT Quad Port Server ExpressModule
In most cases Wake On LAN is only supported on port A for multiple port
adapters. To verify if a port supports Wake on Lan run ethtool eth<X>.
Support Support
======= =======
For general information, go to the Intel support website at: For general information, go to the Intel support website at:
www.intel.com/support/ https://www.intel.com/support/
or the Intel Wired Networking project hosted by Sourceforge at: or the Intel Wired Networking project hosted by Sourceforge at:
http://sourceforge.net/projects/e1000 https://sourceforge.net/projects/e1000
If an issue is identified with the released source code on the supported If an issue is identified with the released source code on a supported kernel
kernel with a supported adapter, email the specific information related with a supported adapter, email the specific information related to the issue
to the issue to e1000-devel@lists.sf.net to e1000-devel@lists.sf.net.
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7