한국어

네트워킹

온누리070 플레이스토어 다운로드
    acrobits softphone
     온누리 070 카카오 프러스 친구추가온누리 070 카카오 프러스 친구추가친추
     카카오톡 채팅 상담 카카오톡 채팅 상담카톡
    
     라인상담
     라인으로 공유

     페북공유

   ◎위챗 : speedseoul


  
     PAYPAL
     
     PRICE
     

pixel.gif

    before pay call 0088 from app



http://wiki.hetzner.de/index.php/Update_Debian_igb-Treiber/en


https://downloadcenter.intel.com/product/64399/Intel-Ethernet-Controller-I210-Series



Linux* Base Driver for Intel(R) Ethernet Network Connection

===========================================================


October 24, 2012


Contents

========


- In This Release

- Identifying Your Adapter

- Upgrading

- Building and Installation

- Command Line Parameters

- Additional Configurations

- Known Issues/Troubleshooting

- Support


In This Release

===============


This file describes the e1000e Linux* Base Driver for Intel Ethernet Network

Connection.  This driver supports kernel versions 2.4.x, 2.6.x, and 3.x.  This 

driver includes support for Itanium(R)2-based systems.


NOTE: The 82546GB part requires the e1000 driver, not the e1000e driver.

 

This driver is only supported as a loadable module at this time.  Intel is

not supplying patches against the kernel source to allow for static linking

of the driver.  For questions related to hardware requirements, refer to the

documentation supplied with your Intel Gigabit adapter.  All hardware

requirements listed apply to use with Linux.


The following features are now available in supported kernels:

 - Native VLANs

 - Channel Bonding (teaming)

 - SNMP


Channel Bonding documentation can be found in the Linux kernel source:

/Documentation/networking/bonding.txt


The driver information previously displayed in the /proc filesystem is not

supported in this release.  Alternatively, you can use ethtool (version 1.6

or later), lspci, and ifconfig to obtain the same information.


Instructions on updating ethtool can be found in the section "Additional

Configurations" later in this document.


NOTE: The Intel(R) 82562v 10/100 Network Connection only provides 10/100

support.


Identifying Your Adapter

========================


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/idguide.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


Upgrading

=========


If you currently have the e1000 driver installed and need to install e1000e,

perform the following:


 - If your version of e1000 is 7.6.15.5 or less, upgrade to e1000 version 8.x,

   using the instructions in the e1000 README. 

 - Install the e1000e driver using the instructions in the Building and 

   Installation section below.

 - Modify /etc/modprobe.conf to point your PCIe devices to use the new e1000e 

   driver using alias ethX e1000e, or use your distribution's specific method 

   for configuring network adapters like RedHat's setup/system-config-network

   or SuSE's yast2.


Building and Installation

=========================


To build a binary RPM* package of this driver, run 'rpmbuild -tb

<filename.tar.gz>'.  Replace <filename.tar.gz> with the specific filename

of the driver.


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 now.


      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/e1000e or /usr/local/src/e1000e.


2. Untar/unzip archive:


     tar zxf e1000e-x.x.x.tar.gz


3. Change to the driver src directory:


     cd e1000e-x.x.x/src/


4. Compile the driver module:


     # make install


   The binary will be installed as:


     /lib/modules/<KERNEL VERSION>/kernel/drivers/net/e1000e/e1000e.[k]o


   The install locations listed above are the default locations.  They

   might not be correct for certain Linux distributions. 


5. Load the module using either the insmod or modprobe command:


     modprobe e1000e


     insmod e1000e


   Note that for 2.6 kernels the insmod command can be used if the full

   path to the driver module is specified.  For example:


     insmod /lib/modules/<KERNEL VERSION>/kernel/drivers/net/e1000e/e1000e.ko


   With 2.6 based kernels also make sure that older e1000e drivers are 

   removed from the kernel, before loading the new module:


     rmmod e1000e; modprobe e1000e


6. Assign an IP address to the interface by entering the following, where

   x is the interface number:


     ifconfig ethx <IP_address>


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>


TROUBLESHOOTING: Some systems have trouble supporting MSI and/or MSI-X 

interrupts.  If you believe your system needs to disable this style of 

interrupt, the driver can be built and installed with the command:


     # make CFLAGS_EXTRA=-DDISABLE_PCI_MSI install


Normally the driver will generate an interrupt every two seconds, so if

you can see that you're no longer getting interrupts in

cat /proc/interrupts for the ethX e1000e device, then this workaround

may be necessary.


Command Line Parameters

=======================


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 e1000e [<option>=<VAL1>,<VAL2>,...]


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:


   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, unless otherwise noted.


NOTES:  For more information about the InterruptThrottleRate,

        RxIntDelay, TxIntDelay, RxAbsIntDelay, and TxAbsIntDelay

        parameters, see the application note at:

        http://www.intel.com/design/network/applnots/ap450.htm


        A descriptor describes a data buffer and attributes related to

        the data buffer.  This information is accessed by the hardware.


InterruptThrottleRate

---------------------

Valid Range:   0,1,3,4, 100-100000 (0=off, 1=dynamic, 3=dynamic conservative,

                                    4=simplified balancing)

Default Value: 3


The driver can limit the amount of interrupts per second that the adapter

will generate for incoming packets. It does this by writing a value to the 

adapter that is based on the maximum amount of interrupts that the adapter 

will generate per second.


Setting InterruptThrottleRate to a value greater or equal to 100

will program the adapter to send out a maximum of that many interrupts

per second, even if more packets have come in. This reduces interrupt

load on the system and can lower CPU utilization under heavy load,

but will increase latency as packets are not processed as quickly.


The default behaviour of the driver previously assumed a static 

InterruptThrottleRate value of 8000, providing a good fallback value for 

all traffic types, but lacking in small packet performance and latency. 


The driver has two adaptive modes (setting 1 or 3) in which

it dynamically adjusts the InterruptThrottleRate value based on the traffic 

that it receives. After determining the type of incoming traffic in the last

timeframe, it will adjust the InterruptThrottleRate to an appropriate value 

for that traffic.


The algorithm classifies the incoming traffic every interval into

classes.  Once the class is determined, the InterruptThrottleRate value is 

adjusted to suit that traffic type the best. There are three classes defined: 

"Bulk traffic", for large amounts of packets of normal size; "Low latency",

for small amounts of traffic and/or a significant percentage of small

packets; and "Lowest latency", for almost completely small packets or 

minimal traffic.


In dynamic conservative mode, the InterruptThrottleRate value is set to 4000 

for traffic that falls in class "Bulk traffic". If traffic falls in the "Low 

latency" or "Lowest latency" class, the InterruptThrottleRate is increased 

stepwise to 20000. This default mode is suitable for most applications.


For situations where low latency is vital such as cluster or

grid computing, the algorithm can reduce latency even more when

InterruptThrottleRate is set to mode 1. In this mode, which operates

the same as mode 3, the InterruptThrottleRate will be increased stepwise to 

70000 for traffic in class "Lowest latency".


In simplified mode the interrupt rate is based on the ratio of tx and

rx traffic.  If the bytes per second rate is approximately equal, the

interrupt rate will drop as low as 2000 interrupts per second.  If the

traffic is mostly transmit or mostly receive, the interrupt rate could

be as high as 8000.


Setting InterruptThrottleRate to 0 turns off any interrupt moderation

and may improve small packet latency, but is generally not suitable

for bulk throughput traffic.


NOTE:  InterruptThrottleRate takes precedence over the TxAbsIntDelay and

       RxAbsIntDelay parameters.  In other words, minimizing the receive

       and/or transmit absolute delays does not force the controller to

       generate more interrupts than what the Interrupt Throttle Rate

       allows.


NOTE:  When e1000e is loaded with default settings and multiple adapters

       are in use simultaneously, the CPU utilization may increase non-

       linearly.  In order to limit the CPU utilization without impacting

       the overall throughput, we recommend that you load the driver as

       follows:


           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

----------

Valid Range:   0-65535 (0=off)

Default Value: 0


This value delays the generation of receive interrupts in units of 1.024

microseconds.  Receive interrupt reduction can improve CPU efficiency if

properly tuned for specific network traffic.  Increasing this value adds

extra latency to frame reception and can end up decreasing the throughput

of TCP traffic.  If the system is reporting dropped receives, this value

may be set too high, causing the driver to run out of available receive

descriptors.


CAUTION:  When setting RxIntDelay to a value other than 0, adapters may

          hang (stop transmitting) under certain network conditions.  If

          this occurs a NETDEV WATCHDOG message is logged in the system

          event log.  In addition, the controller is automatically reset,

          restoring the network connection.  To eliminate the potential

          for the hang ensure that RxIntDelay is set to 0.


RxAbsIntDelay

-------------

Valid Range:   0-65535 (0=off)

Default Value: 8


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,

this value ensures that an interrupt is generated after the initial

packet is received within the set amount of time.  Proper tuning,

along with RxIntDelay, may improve traffic throughput in specific network

conditions.


TxIntDelay

----------

Valid Range:   0-65535 (0=off)

Default Value: 8


This value delays the generation of transmit interrupts in units of

1.024 microseconds.  Transmit interrupt reduction can improve CPU

efficiency if properly tuned for specific network traffic.  If the

system is reporting dropped transmits, this value may be set too high

causing the driver to run out of available transmit descriptors.


TxAbsIntDelay

-------------

Valid Range:   0-65535 (0=off)

Default Value: 32


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,

this value ensures that an interrupt is generated after the initial

packet is sent on the wire within the set amount of time.  Proper tuning,

along with TxIntDelay, may improve traffic throughput in specific

network conditions.


copybreak

---------

Valid Range:   0-xxxxxxx (0=off)

Default Value: 256

Usage: insmod e1000e.ko copybreak=128


Driver copies all packets below or equaling this size to a fresh rx

buffer before handing it up the stack.


This parameter is different than other parameters, in that 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


SmartPowerDownEnable

--------------------

Valid Range: 0-1

Default Value:  0 (disabled)


Allows Phy to turn off in lower power states. The user can turn off

this parameter in supported chipsets.


KumeranLockLoss

---------------

Valid Range: 0-1

Default Value: 1 (enabled)


This workaround skips resetting the Phy at shutdown for the initial

silicon releases of ICH8 systems.


IntMode

-------

Valid Range: 0-2 (0=legacy, 1=MSI, 2=MSI-X)

Default Value: 2 


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

driver will try other interrupt modes, from least to most compatible.  The 

interrupt order is MSI-X, MSI, Legacy.  If specifying MSI (IntMode=1) 

interrupts, only MSI and Legacy will be attempted.


CrcStripping

------------

Valid Range: 0-1

Default Value: 1 (enabled)


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

loading or enabling the driver, try disabling this feature.


EEE

---

Valid Range:    0-1

Default Value:    1 (enabled for parts supporting EEE)

 

This option allows for the ability of IEEE802.3az (a.k.a. Energy Efficient 

Ethernet or EEE) to be advertised to the link partner on parts supporting EEE.

EEE saves energy by putting the device into a low-power state when the link is

idle, but only when the link partner also supports EEE and after the feature 

has been enabled during link negotiation.  It is not necessary to disable the 

advertisement of EEE when connected with a link partner that does not support EEE.


Node

----

Valid Range:  0-n

Default Value:  -1 (off)

0 - n: where n is the number of the NUMA node that should be used to allocate 

       memory for this adapter port.

-1: uses the driver default of allocating memory on whichever processor is 

    running insmod/modprobe.

The Node parameter will allow you to pick which NUMA node you want to have  

the adapter allocate memory from.  All driver structures, in-memory queues, 

and receive buffers will be allocated on the node specified.  This parameter

is only useful when interrupt affinity is specified, otherwise some portion 

of the time the interrupt could run on a different core than the memory is 

allocated on, causing slower memory access and impacting throughput, CPU, or

both.? 


Additional Configurations

=========================


   IEEE 1588 Precision Time Protocol (PTP) Hardware Clock (PHC)

  ------------------------------------------------------------

  Support for the IEEE 1588 Precision Time Protocol (PTP) Hardware Clock (PHC)

  is disabled by default in this out-of-tree driver even if it is enabled for 

  the in-kernel driver. 


  The feature is available only on a subset of devices supported by the 

  driver, and can only be enabled on 3.0 and newer kernels that also have

  the PTP_1588_CLOCK support compiled in statically or as a module. To enable

  the feature when compiling the driver, add 'CFLAGS_EXTRA=-DE1000E_PTP' to 

  the command line.


  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 Gigabit Family of Adapters is e1000e.


  As an example, if you install the e1000e driver for two Gigabit adapters 

  (eth0 and eth1) and want to set the interrupt mode to MSI-X and MSI 

  respectively, add the following to modules.conf or /etc/modprobe.conf:


       alias eth0 e1000e

       alias eth1 e1000e

       options e1000e IntMode=2,1


  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.  Use the ifconfig command to increase the MTU size.

  For example:


       ifconfig eth<x> mtu 9000 up


  This setting is not saved across reboots.  It can be made permanent if

  you add:


       MTU=9000


   to the file /etc/sysconfig/network-scripts/ifcfg-eth<x>.  This example

   applies to the Red Hat distributions; other distributions may store this

   setting in a different location.


  Notes:


  - To enable Jumbo Frames, increase the MTU size on the interface beyond

    1500.


  - The maximum MTU setting for Jumbo Frames is 9216.  This value coincides

    with the maximum Jumbo Frames size of 9234 bytes. 


  - Using Jumbo frames at 10 or 100 Mbps is not supported and may result in 

    poor performance or loss of link.

    

  - The following adapters limit Jumbo Frames sized packets to a maximum of

    4088 bytes:

     Intel(R) 82578DM Gigabit Network Connection

     Intel(R) 82577LM Gigabit Network Connection 


  - The following adapters do not support Jumbo Frames:

     Intel(R) PRO/1000 Gigabit Server Adapter

     Intel(R) PRO/1000 PM Network Connection

     Intel(R) 82562V 10/100 Network Connection

     Intel(R) 82566DM Gigabit Network Connection

     Intel(R) 82566DC Gigabit Network Connection

     Intel(R) 82566MM Gigabit Network Connection

     Intel(R) 82566MC Gigabit Network Connection

     Intel(R) 82562GT 10/100 Network Connection

     Intel(R) 82562G 10/100 Network Connection

     Intel(r) 82566DC-2 Gigabit Network Connection

     Intel(R) 82562V-2 10/100 Network Connection

     Intel(R) 82562G-2 10/100 Network Connection

     Intel(R) 82562GT-2 10/100 Network Connection

     Intel(R) 82578DC Gigabit Network Connection

     Intel(R) 82567V-3 Gigabit Network Connection


  - Jumbo Frames cannot be configured on an 82579-based Network device, if 

    MACSec is enabled on the system.


  ethtool

  -------

  The driver utilizes the ethtool interface for driver configuration and

  diagnostics, as well as displaying statistical information.  ethtool

  version 3 or later is required for this functionality, although we 

  strongly recommend downloading the latest version at:


  http://ftp.kernel.org/pub/software/network/ethtool/.


  NOTE: When validating enable/disable tests on some parts (82578, for example)

  you need to add a few seconds between tests when working with ethtool.


  Speed and Duplex

  ----------------

  Speed and Duplex are configured through the ethtool* utility.  ethtool is 

  included with all versions of Red Hat after Red Hat 7.2.  For other Linux 

  distributions, download and install ethtool from the following website:

  http://sourceforge.net/projects/gkernel.


  Enabling Wake on LAN* (WoL)

  ---------------------------

  WoL is configured through the ethtool* utility.  ethtool is included with

  all versions of Red Hat after Red Hat 7.2.  For other Linux distributions,

  download and install ethtool from the following website:

  http://sourceforge.net/projects/gkernel.


  For instructions on enabling WoL with ethtool, refer to the website listed

  above.


  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 when shutting down or rebooting the system.


  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 


  NAPI

  ----

  NAPI (Rx polling mode) is supported in the e1000e driver.  NAPI is enabled

  by default.


  To disable NAPI, compile the driver module, passing in a configuration option:


       # make CFLAGS_EXTRA=-DE1000E_NO_NAPI install


  See ftp://robur.slu.se/pub/Linux/net-development/NAPI/usenix-paper.tgz for 

  more information on NAPI.


Known Issues/Troubleshooting

============================


Hardware Issues:

----------------

For known hardware and troubleshooting issues, either refer to the "Release 

Notes" in your User Guide, or for more detailed information, go to the

following website:


    http://www.intel.com


In the search box enter your devices controller ID followed by "spec update" 

(i.e., 82599 spec update). The spec update file has complete information on

known hardware issues.


Software Issues:

----------------


  NOTE: After installing the driver, if your Intel Ethernet Network Connection

  is not working, verify that you have installed the correct driver.


  Intel(R) Active Management Technology 2.0, 2.1, 2.5 not supported in 

  conjunction with Linux driver

  ---------------------------------------------------------------------


  Detected Tx Unit Hang in Quad Port Adapters

  -------------------------------------------

  In some cases ports 3 and 4 don't pass traffic and report 'Detected Tx Unit

  Hang' followed by 'NETDEV WATCHDOG: ethX: transmit timed out' errors. Ports 

  1 and 2 don't show any errors and will pass traffic.


  This issue MAY be resolved by updating to the latest kernel and BIOS. The 

  user is encouraged to run an OS that fully supports MSI interrupts. You can 

  check your system's BIOS by downloading the Linux Firmware Developer Kit 

  that can be obtained at http://www.linuxfirmwarekit.org/


  Adapters with 4 ports behind a PCIe bridge 

  ------------------------------------------

  Adapters that have 4 ports behind a PCIe bridge may be incompatible with 

  some systems. The user should run the Linux firmware kit from 2686434

  http://www.linuxfirmwarekit.org to test their BIOS, if they have interrupt or 

  "missing interface" problems, especially with older kernels.


  82573(V/L/E) TX Unit Hang Messages

  ----------------------------------

  Several adapters with the 82573 chipset display "TX unit hang" messages 

  during normal operation with the e1000e driver. The issue appears both with 

  TSO enabled and disabled, and is caused by a power management function that 

  is enabled in the EEPROM. Early releases of the chipsets to vendors had the 

  EEPROM bit that enabled the feature. After the issue was discovered newer 

  adapters were released with the feature disabled in the EEPROM. 


  If you encounter the problem in an adapter, and the chipset is an 82573-based

  one, you can verify that your adapter needs the fix by using ethtool: 


  # ethtool -e eth0

  Offset          Values

  ------          ------

  0x0000          00 12 34 56 fe dc 30 0d 46 f7 f4 00 ff ff ff ff

  0x0010          ff ff ff ff 6b 02 8c 10 d9 15 8c 10 86 80 de 83

                                                           ^^

  The value at offset 0x001e (de) has bit 0 unset. This enables the problematic 

  power saving feature. In this case, the EEPROM needs to read "df" at offset 

  0x001e. 


  A one-time EEPROM fix is available as a shell script. This script will verify 

  that the adapter is applicable to the fix and if the fix is needed or not. If 

  the fix is required, it applies the change to the EEPROM and updates the 

  checksum. The user must reboot the system after applying the fix if changes 

  were made to the EEPROM. 


  Example output of the script: 


  # bash fixeep-82573-dspd.sh eth0

  eth0: is a "82573E Gigabit Ethernet Controller"

  This fixup is applicable to your hardware

  executing command: ethtool -E eth0 magic 0x109a8086 offset 0x1e value 0xdf

  Change made. You *MUST* reboot your machine before changes take effect!


  The script can be downloaded at 

  http://e1000.sourceforge.net/files/fixeep-82573-dspd.sh


  Dropped Receive Packets on Half-duplex 10/100 Networks

  ------------------------------------------------------

  If you have an Intel PCI Express adapter running at 10mbps or 100mbps, half-

  duplex, you may observe occasional dropped receive packets.  There are no

  workarounds for this problem in this network configuration.  The network must

  be updated to operate in full-duplex, and/or 1000mbps only.


  Driver Compilation

  ------------------

  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.


  Performance Degradation with Jumbo Frames

  -----------------------------------------

  Degradation in throughput performance may be observed in some Jumbo frames

  environments.  If this is observed, increasing the application's socket

  buffer size and/or increasing the /proc/sys/net/ipv4/tcp_*mem entry values

  may help.  See the specific application manual and

  /usr/src/linux*/Documentation/networking/ip-sysctl.txt for more details.


  Jumbo Frames on Foundry BigIron 8000 switch

  -------------------------------------------

  There is a known issue using Jumbo frames when connected to a Foundry

  BigIron 8000 switch.  This is a 3rd party limitation.  If you experience

  loss of packets, lower the MTU size.


  Allocating Rx Buffers when Using Jumbo Frames 

  ---------------------------------------------

  Allocating Rx buffers when using Jumbo Frames on 2.6.x kernels may fail if 

  the available memory is heavily fragmented. This issue may be seen with PCI-X 

  adapters or with packet split disabled. This can be reduced or eliminated 

  by changing the amount of available memory for receive buffer allocation, by

  increasing /proc/sys/vm/min_free_kbytes. 


  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, either 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 line:

    net.ipv4.conf.all.arp_filter = 1

  to the file /etc/sysctl.conf


      or,


  install the interfaces in separate broadcast domains (either in

  different switches or in a switch partitioned to VLANs).


  Disable rx flow control with ethtool

  ------------------------------------

  In order to disable receive flow control using ethtool, you must turn

  off auto-negotiation on the same command line.


  For example:


   ethtool -A eth? autoneg off rx off


  Unplugging network cable while ethtool -p is running

  ----------------------------------------------------

  In kernel versions 2.5.50 and later (including 2.6 kernel), 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.


  MSI-X Issues with Kernels between 2.6.19 - 2.6.21 (inclusive)

  -------------------------------------------------------------

  Kernel panics and instability may be observed on any MSI-X hardware if you 

  use irqbalance with kernels between 2.6.19 and 2.6.21. If such problems are 

  encountered, you may disable the irqbalance daemon or upgrade to a newer 

  kernel.


  Rx Page Allocation Errors

  -------------------------

  Page allocation failure. order:0 errors may occur under stress with kernels 

  2.6.25 and above. This is caused by the way the Linux kernel reports this 

  stressed condition.

  

  Network throughput degradation observed with Onboard video versus add-in 

  Video Card on  82579LM Gigabit Network Connection when used with some older

  kernels.

  ------------------------------------------------------------------------

  This issue can be worked around by specifying "pci=nommconf" in the kernel

  boot parameter or by using another kernel boot parameter 

  "memmap=128M$0x100000000" which marks 128 MB region at 4GB as reserved and

  therefore OS won't use these RAM pages.


  This issue is fixed in kernel version 2.6.21, where the kernel tries to 

  dynamically find out the mmconfig size by looking at the number of buses 

  that the mmconfig segment maps to.


  This issue won't be seen on 32bit version of EL5, as in that case, the kernel 

  sees that RAM is located around the 256MB window and avoids using the 

  mmconfig space.


  Activity LED blinks unexpectedly

  --------------------------------

  If a system based on the 82577, 82578, or 82579 controller is connected

  to a hub, the Activity LED will blink for all network traffic present on

  the hub. Connecting the system to a switch or router will filter out most

  traffic not addressed to the local port.


  Link may take longer than expected

  -----------------------------------------------------------------------

  With some Phy and switch combinations, link can take longer than expected.  

  This can be an issue on Linux distributions that timeout when checking for 

  link prior to acquiring a DHCP address; however there is usually a way to 

  work around this (e.g. set LINKDELAY in the interface configuration on RHEL).


  Tx flow control is disabled by default on 82577 and 82578-based adapters

  ------------------------------------------------------------------------


  Possible performance degradation on certain 82566 and 82577 devices

  ------------------------------------------------------------------- 

  Internal stress testing with jumbo frames shows the reliability on some

  82566 and 82567 devices is improved in certain corner cases by disabling the

  Early Receive feature. Doing so can impact Tx performance. To reduce the

  impact, the packet buffer sizes and relevant flow control settings are

  modified accordingly.


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


License

=======


Intel Gigabit Linux driver.

Copyright(c) 1999 - 2012 Intel Corporation.


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".


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.


조회 수 :
276131
등록일 :
2015.06.27
14:40:36 (*.160.89.104)
엮인글 :
http://www.webs.co.kr/index.php?document_srl=365974&act=trackback&key=fa8
게시글 주소 :
http://www.webs.co.kr/index.php?document_srl=365974
List of Articles
번호 제목 글쓴이 날짜 조회 수sort
» intel i210 driver install and compile debian admin 2015-06-27 276131
122 리눅스에 RADIUS 서버 구현하기 admin 2012-09-01 244685
121 tar로 전체시스템 백업 LINUX Backup admin 2013-04-04 220975
120 리눅스 시스템 시작시에 자동 실행할 파일 등록하기 admin 2014-03-22 205082
119 PUTTY DOWNLOAD - FREE SSH & TELNET CLIENT admin 2017-09-03 186737
118 [linux]리눅스 버전및 CPU등 환경정보 확인하기 admin 2013-04-08 185052
117 FreeRADIUS 설치(freeradius-sever-2.1.8) admin 2013-04-19 163171
116 Getting started with SSH public key cryptography admin 2013-09-05 160662
115 Site builder shootout: Drupal vs. Joomla vs. WordPress admin 2012-07-15 157880
114 How to Manually Unblock / Unban IP Address in fail2ban 차단풀기 admin 2017-09-12 157687
113 CentOS 5.4에서 RADIUS 서버 설치하기 admin 2013-04-19 153682
112 리눅스 서버 이전시 고려할 사항들 admin 2011-12-16 139717
111 linux 에티터 명령어 정리 포그라운드 전환 : ctrl + z 다시 vi 모드로 돌아오려면 fg admin 2014-03-05 139306
110 Ubuntu 14.04 원격 접속(xrdp) 설정 admin 2014-12-21 138802
109 리눅스 부팅시 자동 실행 명령스크립트 실행하기 만들기 이해 linux booting admin 2017-08-30 138671
108 openvpn Easy Windows Guide admin 2017-09-20 131650
107 Debian / Ubuntu Linux Install Kernel Headers Package admin 2015-06-27 131571
106 How To Protect SSH with fail2ban on Debian 7 데비안 ssh 방어 무작위 로그인 admin 2017-09-12 131150
105 리눅스 시작시에 부팅시 프로세스 프로그램 자동 실행 설정 admin 2017-08-30 128731
104 the world’s most widely deployed RADIUS server document admin 2017-08-29 128313
103 리눅스 데비안 네트워크 설정 유선랜 무선랜 linux debian network setting 설정 admin 2017-09-07 127923
102 파이썬 설치 및 사용하기 admin 2011-12-16 116049
101 Linux Date Command Examples to Display and Set System Date Time admin 2013-07-11 108676
100 Radius를 이용한 ssh 인증 서버 구축하기 admin 2013-04-19 105745
99 네트워크 디바이스명 변경 리눅스 eth0 eth1 admin 2017-08-29 102173
98 Build a RADIUS server on Linux admin 2013-03-23 100063
97 부팅시 자동실행 명령어 스크립트, rc.local admin 2017-08-30 97234
96 윈도우XP 알면 정말 편한 숨겨진 기능 60가지 admin 2012-05-11 94237
95 서버이전시 고려할 사항들 admin 2011-12-13 90042
94 다운된 리눅스서버 응급복구 admin 2013-03-28 88871