{"id":19062296,"url":"https://github.com/truthbk/igb-mmap-mc","last_synced_at":"2026-06-19T04:31:09.403Z","repository":{"id":2634215,"uuid":"3622570","full_name":"truthbk/igb-mmap-mc","owner":"truthbk","description":"IGB zero copy to userland + multiple traffic copies.  ","archived":false,"fork":false,"pushed_at":"2012-03-05T01:02:40.000Z","size":360,"stargazers_count":3,"open_issues_count":0,"forks_count":0,"subscribers_count":3,"default_branch":"master","last_synced_at":"2025-02-22T02:25:31.014Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"C","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/truthbk.png","metadata":{"files":{"readme":"README","changelog":null,"contributing":null,"funding":null,"license":"COPYING","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2012-03-05T00:35:25.000Z","updated_at":"2023-05-29T06:29:32.000Z","dependencies_parsed_at":"2022-08-31T19:02:33.315Z","dependency_job_id":null,"html_url":"https://github.com/truthbk/igb-mmap-mc","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/truthbk/igb-mmap-mc","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/truthbk%2Figb-mmap-mc","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/truthbk%2Figb-mmap-mc/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/truthbk%2Figb-mmap-mc/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/truthbk%2Figb-mmap-mc/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/truthbk","download_url":"https://codeload.github.com/truthbk/igb-mmap-mc/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/truthbk%2Figb-mmap-mc/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34517748,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-19T02:00:06.005Z","response_time":61,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-09T00:25:22.443Z","updated_at":"2026-06-19T04:31:09.395Z","avatar_url":"https://github.com/truthbk.png","language":"C","funding_links":[],"categories":[],"sub_categories":[],"readme":"Linux* Base Driver for Intel(R) Ethernet Network Connection\n===========================================================\n\nApril 14, 2011\n\nContents\n========\n\n- In This Release\n- Identifying Your Adapter\n- Building and Installation\n- Command Line Parameters\n- Additional Configurations\n- Known Issues/Troubleshooting\n- Support\n\nIn This Release\n===============\n\nThis file describes the igb Linux* Base Driver for Intel Ethernet Network \nConnection.  This driver supports kernel versions 2.4.x and 2.6.x. \n\nThis driver is only supported as a loadable module at this time.  Intel is\nnot supplying patches against the kernel source to allow for static linking\nof the driver.  For questions related to hardware requirements, refer to the\ndocumentation supplied with your Intel Gigabit adapter.  All hardware\nrequirements listed apply to use with Linux.\n\nThe following features are now available in supported kernels:\n - Native VLANs\n - Channel Bonding (teaming)\n - SNMP\n\nChannel Bonding documentation can be found in the Linux kernel source:\n/Documentation/networking/bonding.txt\n\nThe igb driver supports IEEE 1588 time stamping for kernels 2.6.30 and above.  \n\nThe driver information previously displayed in the /proc filesystem is not\nsupported in this release.  Alternatively, you can use ethtool (version 1.6\nor later), lspci, and ifconfig to obtain the same information.\n\nInstructions on updating ethtool can be found in the section \"Additional\nConfigurations\" later in this document.\n\nIdentifying Your Adapter\n========================\n\nFor more information on how to identify your adapter, go to the Adapter \u0026\nDriver ID Guide at:\n\n    http://support.intel.com/support/go/network/adapter/idguide.htm\n\nFor the latest Intel network drivers for Linux, refer to the following\nwebsite.  Select the link for your adapter.\n\n    http://support.intel.com/support/go/network/adapter/home.htm\n\nBuilding and Installation\n=========================\n\nTo build a binary RPM* package of this driver, run 'rpmbuild -tb\n\u003cfilename.tar.gz\u003e'.  Replace \u003cfilename.tar.gz\u003e with the specific filename\nof the driver.\n\nNOTE: For the build to work properly, the currently running kernel MUST\n      match the version and configuration of the installed kernel sources.\n      If you have just recompiled the kernel reboot the system now.\n\n      RPM functionality has only been tested in Red Hat distributions.\n\n1. Move the base driver tar file to the directory of your choice.  For\n   example, use /home/username/igb or /usr/local/src/igb.\n\n2. Untar/unzip archive:\n\n     tar zxf igb-x.x.x.tar.gz\n\n3. Change to the driver src directory:\n\n     cd igb-x.x.x/src/\n\n4. Compile the driver module:\n\n     make install\n\n   The binary will be installed as:\n\n     /lib/modules/\u003cKERNEL VERSION\u003e/kernel/drivers/net/igb/igb.[k]o\n\n   The install locations listed above are the default locations.  They\n   might not be correct for certain Linux distributions.  \n\n5. Load the module using either the insmod or modprobe command:\n\n     modprobe igb\n\n     insmod igb\n\n   Note that for 2.6 kernels the insmod command can be used if the full\n   path to the driver module is specified.  For example:\n\n     insmod /lib/modules/\u003cKERNEL VERSION\u003e/kernel/drivers/net/igb/igb.ko\n\n   With 2.6 based kernels also make sure that older igb drivers are\n   removed from the kernel, before loading the new module:\n\n     rmmod igb; modprobe igb\n\n6. Assign an IP address to the interface by entering the following, where\n   x is the interface number:\n\n     ifconfig ethx \u003cIP_address\u003e\n\n7. Verify that the interface works.  Enter the following, where \u003cIP_address\u003e\n   is the IP address for another machine on the same subnet as the\n   interface that is being tested:\n\n     ping  \u003cIP_address\u003e\n\nTROUBLESHOOTING: Some systems have trouble supporting MSI and/or MSI-X \ninterrupts. If you believe your system needs to disable this style of \ninterrupt, the driver can be built and installed with the command:\n\n     make CFLAGS_EXTRA=-DDISABLE_PCI_MSI install\n\nNormally the driver will generate an interrupt every two seconds, so if \nyou can see that you're no longer getting interrupts in cat /proc/interrupts\nfor the ethX igb device, then this workaround may be necessary.\n\nTo build igb driver with DCA:\n------------------------------\n \nIf your kernel supports DCA, the driver will build by default with DCA enabled.\n\nCommand Line Parameters\n=======================\n\nIf the driver is built as a module, the  following optional parameters\nare used by entering them on the command line with the modprobe command\nusing this syntax:\n\n     modprobe igb [\u003coption\u003e=\u003cVAL1\u003e,\u003cVAL2\u003e,...]\n\nFor example:\n\n     modprobe igb InterruptThrottleRate=16000,16000\n\n\nThe default value for each parameter is generally the recommended setting,\nunless otherwise noted.\n\nNOTES:  For more information about the InterruptThrottleRate,\n        parameter, see the application note at:\n        http://www.intel.com/design/network/applnots/ap450.htm\n\n        A descriptor describes a data buffer and attributes related to\n        the data buffer.  This information is accessed by the hardware.\n\nInterruptThrottleRate\n---------------------\nValid Range:   0,1,3,100-100000 (0=off, 1=dynamic, 3=dynamic conservative)\nDefault Value: 3\n\nThe driver can limit the amount of interrupts per second that the adapter\nwill generate for incoming packets. It does this by writing a value to the \nadapter that is based on the maximum amount of interrupts that the adapter \nwill generate per second.\n\nSetting InterruptThrottleRate to a value greater or equal to 100\nwill program the adapter to send out a maximum of that many interrupts\nper second, even if more packets have come in. This reduces interrupt\nload on the system and can lower CPU utilization under heavy load,\nbut will increase latency as packets are not processed as quickly.\n\nThe default behaviour of the driver previously assumed a static \nInterruptThrottleRate value of 8000, providing a good fallback value for \nall traffic types,but lacking in small packet performance and latency. \nThe hardware can handle many more small packets per second however, and \nfor this reason an adaptive interrupt moderation algorithm was implemented.\n\nThe driver has two adaptive modes (setting 1 or 3) in which it dynamically\nadjusts the InterruptThrottleRate value based on the traffic that it receives.\nAfter determining the type of incoming traffic in the last timeframe, it will\nadjust the InterruptThrottleRate to an appropriate value for that traffic.\n\nThe algorithm classifies the incoming traffic every interval into\nclasses.  Once the class is determined, the InterruptThrottleRate value is \nadjusted to suit that traffic type the best. There are three classes defined: \n\"Bulk traffic\", for large amounts of packets of normal size; \"Low latency\",\nfor small amounts of traffic and/or a significant percentage of small\npackets; and \"Lowest latency\", for almost completely small packets or \nminimal traffic.\n\nIn dynamic conservative mode, the InterruptThrottleRate value is set to 4000 \nfor traffic that falls in class \"Bulk traffic\". If traffic falls in the \"Low \nlatency\" or \"Lowest latency\" class, the InterruptThrottleRate is increased \nstepwise to 20000. This default mode is suitable for most applications.\n\nFor situations where low latency is vital such as cluster or\ngrid computing, the algorithm can reduce latency even more when\nInterruptThrottleRate is set to mode 1. In this mode, which operates\nthe same as mode 3, the InterruptThrottleRate will be increased stepwise to \n70000 for traffic in class \"Lowest latency\".\n\nSetting InterruptThrottleRate to 0 turns off any interrupt moderation\nand may improve small packet latency, but is generally not suitable\nfor bulk throughput traffic.\n\nNOTE:  Dynamic interrupt throttling is only applicable to adapters\n       operating in MSI or Legacy interrupt mode, using a single\n       receive queue.\n\nNOTE:  When igb is loaded with default settings and multiple adapters\n       are in use simultaneously, the CPU utilization may increase non-\n       linearly.  In order to limit the CPU utilization without impacting\n       the overall throughput, we recommend that you load the driver as\n       follows:\n\n           modprobe igb InterruptThrottleRate=3000,3000,3000\n\n       This sets the InterruptThrottleRate to 3000 interrupts/sec for\n       the first, second, and third instances of the driver.  The range\n       of 2000 to 3000 interrupts per second works on a majority of\n       systems and is a good starting point, but the optimal value will\n       be platform-specific.  If CPU utilization is not a concern, use\n       default driver settings.\n\nLLI (Low Latency Interrupts)\n---------------------------- \nLLI allows for immediate generation of an interrupt upon processing receive \npackets that match certain criteria as set by the parameters described below. \nLLI parameters are not enabled when Legacy interrupts are used. You must be \nusing MSI or MSI-X (see cat /proc/interrupts) to successfully use LLI.\n\nLLIPort\n-------\nValid Range:   0-65535\nDefault Value: 0 (disabled)\n\nLLI is configured with the LLIPort command-line parameter, which specifies \nwhich TCP port should generate Low Latency Interrupts.\n\nFor example, using LLIPort=80 would cause the board to generate an \nimmediate interrupt upon receipt of any packet sent to TCP port 80 on the \nlocal machine.\n\nWARNING: Enabling LLI can result in an excessive number of interrupts/second \nthat may cause problems with the system and in some cases may cause a kernel \npanic. \n\nLLIPush\n-------\nValid Range:   0-1\nDefault Value: 0 (disabled)\n\nCan be set to be enabled or disabled (default). It is most \neffective in an environment with many small transactions.\nNOTE: Enabling LLIPush may allow a denial of service attack.\n\nLLISize\n-------\nValid Range:   0-1500\nDefault Value: 0 (disabled)\n\nCauses an immediate interrupt if the board receives a packet smaller \nthan the specified size.\n\nIntMode\n-------\nValid Range:    0-2\nDefault Value:  2\n0 - Legacy Interrupts, single queue\n1 - MSI interrupts, single queue\n2 - MSI-X interrupts, single queue (default)\n\nAllows changing interrupt mode and MQ status at load time, without requiring\na recompile. If the driver fails to enable a specific interrupt mode, it will\nfail back to lesser capabilities.\n\nRSS\n---\nValid Range:   0-8\nDefault Value: 1\n\n0 - Assign up to whichever is less, number of CPUS or number of queues\nX - Assign X queues where X is less than the maximum number of queues\n\nNote: for 82575-based adapters the maximum number of queues is 4; for \n82576-based and newer adapters it is 8.\n\nThis parameter is also affected by the VMDq parameter in that it will limit the\nqueues more.\n\n      VMDQ\nModel 0  1  2  3+\n82575 4  4  3  1\n82576 8  2  2  2\n82580 8  1  1  1\n\nVMDQ\n----\nValid Range:  0 - 4 on 82575-based adapters; and 0 - 8 for 82576/82580-based \nadapters.\nDefault Value: 0\n\nSupports enabling VMDq pools as this is needed to support SR-IOV.\n\n0 - disabled\n1 - sets the netdev as pool 0\n2+ - add additional queues but they currently are not used.\n\nThis parameter is forced to 1 or more if the max_vfs module parameter is used. \nIn addition the number of queues available for RSS is limited if this is set to\n1 or greater.\n \nmax_vfs\n-------\nValid Range:   0-7\nDefault Value: 0\n\n  If the value is greater than 0 it will also force the VMDq parameter to be 1 \n  or more.\n\n  This parameter adds support for SR-IOV.  It causes the driver to spawn up to \n  max_vfs worth of virtual function.  \n\nQueuePairs\n----------\nValid Range:  0-1\nDefault Value:  1 (TX and RX will be paired onto one interrupt vector)\n\nIf set to 0, when MSI-X is enabled, the TX and RX will attempt to occupy\nseparate vectors.    \n\nThis option can be overridden to 1 if there are not sufficient interrupts\navailable.  This can occur if any combination of RSS, VMDQ, and max_vfs \nresults in more than 4 queues being used.\n\nNode\n----\nValid Range:   0-n\nDefault Value: -1 (off)\n\n  0 - n: where n is the number of the NUMA node that should be used to \n         allocate memory for this adapter port. \n  -1: uses the driver default of allocating memory on whichever processor is\n      running insmod/modprobe. \n\n  The Node parameter will allow you to pick which NUMA node you want to have \n  the adapter allocate memory from.  All driver structures, in-memory queues, \n  and receive buffers will be allocated on the node specified.  This parameter\n  is only useful when interrupt affinity is specified, otherwise some portion \n  of the time the interrupt could run on a different core than the memory is \n  allocated on, causing slower memory access and impacting throughput, CPU, or\n  both.\n\nEEE\n---\nValid Range:  0-1\nDefault Value: 1 (enabled)\n\n  A link between two EEE-compliant devices will result in periodic bursts of \n  data followed by periods where the link is in an idle state. This Low\n  Power Idle (LPI) state is supported in both 1Gbps and 100Mbps link speeds. \n  NOTE: EEE support requires autonegotiation.\n\nDMAC\n----\nValid Range: 0, 250, 500, 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, \n9000, 10000.\nDefault Value: 0 (disabled)\n  Enables or disables DMA Coalescing feature. Values are in usec�s and increase\n  the internal DMA Coalescing feature�s internal timer.  DMA (Direct Memory \n  Access) allows the network device to move packet data directly to the \n  system's memory, reducing CPU utilization. However, the frequency and random\n  intervals at which packets arrive do not allow the system to enter a lower \n  power state. DMA Coalescing allows the adapter to collect packets before it \n  initiates a DMA event. This may increase network latency but also increases\n  the chances that the system will enter a lower power state. \n\n  Turning on DMA Coalescing may save energy with kernel 2.6.32 and later. This\n  will impart the greatest chance for your system to consume less power. DMA \n  Coalescing is effective in helping potentially saving the platform power only\n  when it is enabled across all active ports.\n\n  InterruptThrottleRate (ITR) should be set to dynamic. When ITR=0, DMA \n  Coalescing is automatically disabled. \n \n  A whitepaper containing information on how to best configure your platform \n  is available on the Intel website.\n\nMDD (Malicious Driver Detection)\n--------------------------------\nValid Range: 0, 1; 0 = Disable, 1 = Enable\nDefault Value: 1\n\nThis parameter is only relevant for I350 devices operating in SR-IOV mode. When\nthis parameter is set, the driver detects malicious VF driver and disables its\nTX/RX queues until a VF driver reset occurs.\n \n\nAdditional Configurations\n=========================\n\n  Configuring the Driver on Different Distributions\n  -------------------------------------------------\n  Configuring a network driver to load properly when the system is started\n  is distribution dependent.  Typically, the configuration process involves\n  adding an alias line to /etc/modules.conf or /etc/modprobe.conf as well\n  as editing other system startup scripts and/or configuration files.  Many\n  popular Linux distributions ship with tools to make these changes for you.\n  To learn the proper way to configure a network device for your system,\n  refer to your distribution documentation.  If during this process you are\n  asked for the driver or module name, the name for the Linux Base Driver\n  for the Gigabit Family of Adapters is igb.\n\n  As an example, if you install the igb driver for two Gigabit adapters \n  (eth0 and eth1) and want to set the interrupt mode to MSI-X and MSI \n  respectively, add the following to modules.conf or /etc/modprobe.conf:\n\n       alias eth0 igb\n       alias eth1 igb\n       options igb IntMode=2,1\n\n  Viewing Link Messages\n  ---------------------\n  Link messages will not be displayed to the console if the distribution is\n  restricting system messages.  In order to see network driver link messages\n  on your console, set dmesg to eight by entering the following:\n\n       dmesg -n 8\n\n  NOTE: This setting is not saved across reboots.\n\n  Jumbo Frames\n  ------------\n  Jumbo Frames support is enabled by changing the MTU to a value larger than\n  the default of 1500.  Use the ifconfig command to increase the MTU size.\n  For example:\n\n       ifconfig eth\u003cx\u003e mtu 9000 up\n\n  This setting is not saved across reboots.  It can be made permanent if\n  you add:\n\n       MTU=9000\n\n   to the file /etc/sysconfig/network-scripts/ifcfg-eth\u003cx\u003e.  This example\n   applies to the Red Hat distributions; other distributions may store this\n   setting in a different location.\n\n  Notes:\n\n  - To enable Jumbo Frames, increase the MTU size on the interface beyond\n    1500.\n\n  - The maximum MTU setting for Jumbo Frames is 9216.  This value coincides\n    with the maximum Jumbo Frames size of 9234 bytes.\n\n  - Using Jumbo frames at 10 or 100 Mbps is not supported and may result in\n    poor performance or loss of link.\n\n  ethtool\n  -------\n  The driver utilizes the ethtool interface for driver configuration and\n  diagnostics, as well as displaying statistical information.  ethtool\n  version 3.0 or later is required for this functionality, although we \n  strongly recommend downloading the latest version at:\n\n  http://ftp.kernel.org/pub/software/network/ethtool/.\n\n  Enabling Wake on LAN* (WoL)\n  ---------------------------\n  WoL is configured through the ethtool* utility.  ethtool is included with\n  all versions of Red Hat after Red Hat 7.2.  For other Linux distributions,\n  download and install ethtool from the following website:\n  http://sourceforge.net/projects/gkernel.\n\n  For instructions on enabling WoL with ethtool, refer to the website listed\n  above.\n\n  WoL will be enabled on the system during the next shut down or reboot.\n  For this driver version, in order to enable WoL, the igb driver must be\n  loaded when shutting down or rebooting the system.\n\n  Wake On LAN is only supported on port A of multi-port adapters.\n\n  Wake On LAN is not supported for the Intel(R) Gigabit VT Quad Port Server \n  Adapter.\n\n  Multiqueue\n  ----------\n  In this mode, a separate MSI-X vector is allocated for each queue and one\n  for \"other\" interrupts such as link status change and errors.  All \n  interrupts are throttled via interrupt moderation.  Interrupt moderation \n  must be used to avoid interrupt storms while the driver is processing one \n  interrupt.  The moderation value should be at least as large as the expected\n  time for the driver to process an interrupt. Multiqueue is off by default.\n\n  REQUIREMENTS: MSI-X support is required for Multiqueue. If MSI-X is not \n  found, the system will fallback to MSI or to Legacy interrupts.\n  This driver supports multiqueue in kernel versions 2.6.24 and greater.\n  This driver supports receive multiqueue on all kernels that support MSI-X.\n  NOTE: Do not use MSI-X with the 2.6.19 or 2.6.20 kernels.\n\n  LRO\n  ---\n  Large Receive Offload (LRO) is a technique for increasing inbound throughput\n  of high-bandwidth network connections by reducing CPU overhead. It works by\n  aggregating multiple incoming packets from a single stream into a larger \n  buffer before they are passed higher up the networking stack, thus reducing\n  the number of packets that have to be processed. LRO combines multiple \n  Ethernet frames into a single receive in the stack, thereby potentially \n  decreasing CPU utilization for receives.\n\n  NOTE: LRO requires 2.4.22 or later kernel version.\n\n  IGB_LRO is a compile time flag. The user can enable it at compile \n  time to add support for LRO from the driver. The flag is used by adding \n  CFLAGS_EXTRA=\"-DIGB_LRO\" to the make file when it's being compiled. \n\n     make CFLAGS_EXTRA=\"-DIGB_LRO\" install\n\n  You can verify that the driver is using LRO by looking at these counters in \n  ethtool:\n\n  lro_aggregated - count of total packets that were combined\n  lro_flushed - counts the number of packets flushed out of LRO\n  lro_recycled - reflects the number of buffers returned to the ring from \n  recycling\n\n  NOTE: IPv6 and UDP are not supported by LRO.\n\n  MAC and VLAN anti-spoofing feature\n  ----------------------------------\n  When a malicious driver attempts to send a spoofed packet, it is dropped by \n  the hardware and not transmitted.  An interrupt is sent to the PF driver \n  notifying it of the spoof attempt.\n\n  When a spoofed packet is detected the PF driver will send the following \n  message to the system log (displayed by  the \"dmesg\" command):\n\n  Spoof event(s) detected on VF(n)\n\n  Where n=the VF that attempted to do the spoofing.\n\n  Setting MAC Address, VLAN and Rate Limit Using IProute2 Tool\n  ------------------------------------------------------------\n  You can set a MAC address of a Virtual Function (VF), a default VLAN and the\n  rate limit using the IProute2 tool. Download the latest version of the \n  iproute2 tool from Sourceforge if your version does not have all the\n  features you require.\n\n  \nKnown Issues/Troubleshooting\n============================\n\nFor known hardware and troubleshooting issues, refer to the following website.\n\n    http://support.intel.com/support/go/network/adapter/home.htm\n\nEither select the link for your adapter or perform a search for the adapter \nnumber. The adapter's page lists many issues. For a complete list of hardware\nissues download your adapter's user guide and read the Release Notes.  \n\n  NOTE: After installing the driver, if your Intel Ethernet Network Connection\n  is not working, verify in the \"In This Release\" section of the readme that \n  you have installed the correct driver.\n\n  Using the igb driver on 2.4 or older 2.6 based kernels\n  ------------------------------------------------------\n  Due to limited support for PCI-Express in 2.4 kernels and older 2.6 kernels,\n  the igb driver may run into interrupt related problems on some systems, \n  such as no link or hang when bringing up the device.\n\n  We recommend the newer 2.6 based kernels, as these kernels correctly\n  configure the PCI-Express configuration space of the adapter and all\n  intervening bridges. If you are required to use a 2.4 kernel, use a 2.4 kernel \n  newer than 2.4.30. For 2.6 kernels we recommend using the 2.6.21 kernel or \n  newer.\n\n  Alternatively, on 2.6 kernels you may disable MSI support in the kernel by \n  booting with the \"pci=nomsi\" option or permanently disable MSI support in \n  your kernel by configuring your kernel with CONFIG_PCI_MSI unset.\n\n  Intel(R) Active Management Technology 2.0, 2.1, 2.5 not supported in \n  conjunction with Linux driver\n  ---------------------------------------------------------------------\n\n  Driver Compilation\n  ------------------\n  When trying to compile the driver by running make install, the following\n  error may occur:\n\n    \"Linux kernel source not configured - missing version.h\"\n\n  To solve this issue, create the version.h file by going to the Linux source\n  tree and entering:\n\n    make include/linux/version.h.\n\n  Performance Degradation with Jumbo Frames\n  -----------------------------------------\n  Degradation in throughput performance may be observed in some Jumbo frames \n  environments. If this is observed, increasing the application's socket buffer\n  size and/or increasing the /proc/sys/net/ipv4/tcp_*mem entry values may help.\n  See the specific application manual and \n  /usr/src/linux*/Documentation/networking/ip-sysctl.txt for more details.\n\n  Jumbo Frames on Foundry BigIron 8000 switch\n  -------------------------------------------\n  There is a known issue using Jumbo frames when connected to a Foundry\n  BigIron 8000 switch.  This is a 3rd party limitation.  If you experience\n  loss of packets, lower the MTU size.\n\n  Multiple Interfaces on Same Ethernet Broadcast Network\n  ------------------------------------------------------\n  Due to the default ARP behavior on Linux, it is not possible to have\n  one system on two IP networks in the same Ethernet broadcast domain\n  (non-partitioned switch) behave as expected.  All Ethernet interfaces\n  will respond to IP traffic for any IP address assigned to the system.\n  This results in unbalanced receive traffic.\n\n  If you have multiple interfaces in a server, either turn on ARP\n  filtering by entering:\n\n    echo 1 \u003e /proc/sys/net/ipv4/conf/all/arp_filter\n  (this only works if your kernel's version is higher than 2.4.5),\n\n  NOTE: This setting is not saved across reboots.  The configuration\n  change can be made permanent by adding the line:\n    net.ipv4.conf.all.arp_filter = 1\n  to the file /etc/sysctl.conf\n\n      or,\n\n  install the interfaces in separate broadcast domains (either in\n  different switches or in a switch partitioned to VLANs).\n\n  Disable rx flow control with ethtool\n  ------------------------------------\n  In order to disable receive flow control using ethtool, you must turn\n  off auto-negotiation on the same command line.\n\n  For example:\n\n   ethtool -A eth? autoneg off rx off\n\n  Unplugging network cable while ethtool -p is running\n  ----------------------------------------------------\n  In kernel versions 2.5.50 and later (including 2.6 kernel), unplugging \n  the network cable while ethtool -p is running will cause the system to \n  become unresponsive to keyboard commands, except for control-alt-delete.  \n  Restarting the system appears to be the only remedy.\n\n  Trouble passing traffic with on ports 1 and 2 using RHEL3\n  ---------------------------------------------------------\n  There is a known hardware compatibility issue on some systems with RHEL3 \n  kernels. Traffic on ports 1 and 2 may be slower than expected and ping times\n  higher than expected. \n\n  This issue MAY be resolved by updating to the latest kernel and BIOS. You can \n  check your system's BIOS by downloading the Linux Firmware Developer Kit that \n  can be obtained at http://www.linuxfirmwarekit.org/\n\n  Do Not Use LRO When Routing Packets\n  -----------------------------------\n  Due to a known general compatibility issue with LRO and routing, do not use\n  LRO when routing packets.\n\n  Build error with Asianux 3.0 - redefinition of typedef 'irq_handler_t'\n  ---------------------------------------------------------------------\n  Some systems may experience build issues due to redefinition of irq_handler_t.  \n  To resolve this issue build the driver (step 4 above) using the command:\n\n    make CFLAGS_EXTRA=-DAX_RELEASE_CODE=1 install\n\n  MSI-X Issues with Kernels between 2.6.19 - 2.6.21 (inclusive)\n  -------------------------------------------------------------\n  Kernel panics and instability may be observed on any MSI-X hardware if you \n  use irqbalance with kernels between 2.6.19 and 2.6.21. If such problems are \n  encountered, you may disable the irqbalance daemon or upgrade to a newer \n  kernel.\n\n  Rx Page Allocation Errors\n  -------------------------\n  Page allocation failure. order:0 errors may occur under stress with kernels \n  2.6.25 and above. This is caused by the way the Linux kernel reports this \n  stressed condition.\n\n  Under Redhat 5.4-GA - System May Crash when Closing Guest OS Window after \n  Loading/Unloading Physical Function (PF) Driver\n  -------------------------------------------------------------------------\n  Do not remove the igb driver from Dom0 while Virtual Functions (VFs) are \n  assigned to guests. VFs must first use the xm \"pci-detach\" command to \n  hot-plug the VF device out of the VM it is assigned to or else shut down the\n  VM.\n\n  SLES10 SP3 random system panic when reloading driver\n  ---------------------------------------------------\n  This is a known SLES-10 SP3 issue. After requesting interrupts for MSI-X \n  vectors, system may panic. \n\n  Currently the only known workaround is to build the drivers with \n  CFLAGS_EXTRA=-DDISABLE_PCI_MSI if the driver need to be loaded/unloaded.\n  Otherwise the driver can be loaded once and will be safe, but unloading it \n  will lead to the issue.\n\n  Enabling SR-IOV in a 32-bit Microsoft* Windows* Server 2008 Guest OS using\n  Intel (R) 82576-based GbE or Intel (R) 82599-based 10GbE controller under KVM\n  -----------------------------------------------------------------------------\n  KVM Hypervisor/VMM supports direct assignment of a PCIe device to a VM.  This \n  includes traditional PCIe devices, as well as SR-IOV-capable devices using\n  Intel 82576-based and 82599-based controllers.\n\n  While direct assignment of a PCIe device or an SR-IOV Virtual Function (VF)\n  to a Linux-based VM running 2.6.32 or later kernel works fine, there is a\n  known issue with Microsoft Windows Server 2008 VM that results in a \"yellow \n  bang\" error. This problem is within the KVM VMM itself, not the Intel driver,\n  or the SR-IOV logic of the VMM, but rather that KVM emulates an older CPU \n  model for the guests, and this older CPU model does not support MSI-X\n  interrupts, which is a requirement for Intel SR-IOV.  \n\n  If you wish to use the Intel 82576 or 82599-based controllers in SR-IOV mode\n  with KVM and a Microsoft Windows Server 2008 guest try the following \n  workaround. The workaround is to tell KVM to emulate a different model of CPU\n  when using qemu to create the KVM guest: \n\n       \"-cpu qemu64,model=13\"\n\n\nSupport\n=======\n\nFor general information, go to the Intel support website at:\n\n    www.intel.com/support/\n\nor the Intel Wired Networking project hosted by Sourceforge at:\n\n    http://sourceforge.net/projects/e1000\n\nIf an issue is identified with the released source code on the supported\nkernel with a supported adapter, email the specific information related\nto the issue to e1000-devel@lists.sf.net\n\nLicense\n=======\n\nIntel Gigabit Linux driver.\nCopyright(c) 1999 - 2011 Intel Corporation.\n\nThis program is free software; you can redistribute it and/or modify it\nunder the terms and conditions of the GNU General Public License,\nversion 2, as published by the Free Software Foundation.\n\nThis program is distributed in the hope it will be useful, but WITHOUT\nANY WARRANTY; without even the implied warranty of MERCHANTABILITY or\nFITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for\nmore details.\n\nYou should have received a copy of the GNU General Public License along with\nthis program; if not, write to the Free Software Foundation, Inc.,\n51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.\n\nThe full GNU General Public License is included in this distribution in\nthe file called \"COPYING\".\n\nTrademarks\n==========\n\nIntel, Itanium, and Pentium are trademarks or registered trademarks of\nIntel Corporation or its subsidiaries in the United States and other\ncountries.\n\n* Other names and brands may be claimed as the property of others.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftruthbk%2Figb-mmap-mc","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftruthbk%2Figb-mmap-mc","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftruthbk%2Figb-mmap-mc/lists"}