1Linux Base Driver for 10 Gigabit Intel(R) Ethernet Network Connection 2===================================================================== 3 4March 14, 2011 5 6 7Contents 8======== 9 10- In This Release 11- Identifying Your Adapter 12- Building and Installation 13- Command Line Parameters 14- Improving Performance 15- Additional Configurations 16- Known Issues/Troubleshooting 17- Support 18 19 20 21In This Release 22=============== 23 24This file describes the ixgb Linux Base Driver for the 10 Gigabit Intel(R) 25Network Connection. This driver includes support for Itanium(R)2-based 26systems. 27 28For questions related to hardware requirements, refer to the documentation 29supplied with your 10 Gigabit adapter. All hardware requirements listed apply 30to use with Linux. 31 32The following features are available in this kernel: 33 - Native VLANs 34 - Channel Bonding (teaming) 35 - SNMP 36 37Channel Bonding documentation can be found in the Linux kernel source: 38/Documentation/networking/bonding.txt 39 40The driver information previously displayed in the /proc filesystem is not 41supported in this release. Alternatively, you can use ethtool (version 1.6 42or later), lspci, and iproute2 to obtain the same information. 43 44Instructions on updating ethtool can be found in the section "Additional 45Configurations" later in this document. 46 47 48Identifying Your Adapter 49======================== 50 51The following Intel network adapters are compatible with the drivers in this 52release: 53 54Controller Adapter Name Physical Layer 55---------- ------------ -------------- 5682597EX Intel(R) PRO/10GbE LR/SR/CX4 10G Base-LR (1310 nm optical fiber) 57 Server Adapters 10G Base-SR (850 nm optical fiber) 58 10G Base-CX4(twin-axial copper cabling) 59 60For more information on how to identify your adapter, go to the Adapter & 61Driver ID Guide at: 62 63 http://support.intel.com/support/network/sb/CS-012904.htm 64 65 66Building and Installation 67========================= 68 69select m for "Intel(R) PRO/10GbE support" located at: 70 Location: 71 -> Device Drivers 72 -> Network device support (NETDEVICES [=y]) 73 -> Ethernet (10000 Mbit) (NETDEV_10000 [=y]) 741. make modules && make modules_install 75 762. Load the module: 77 78 modprobe ixgb <parameter>=<value> 79 80 The insmod command can be used if the full 81 path to the driver module is specified. For example: 82 83 insmod /lib/modules/<KERNEL VERSION>/kernel/drivers/net/ixgb/ixgb.ko 84 85 With 2.6 based kernels also make sure that older ixgb drivers are 86 removed from the kernel, before loading the new module: 87 88 rmmod ixgb; modprobe ixgb 89 903. Assign an IP address to the interface by entering the following, where 91 x is the interface number: 92 93 ip addr add ethx <IP_address> 94 954. Verify that the interface works. Enter the following, where <IP_address> 96 is the IP address for another machine on the same subnet as the interface 97 that is being tested: 98 99 ping <IP_address> 100 101 102Command Line Parameters 103======================= 104 105If the driver is built as a module, the following optional parameters are 106used by entering them on the command line with the modprobe command using 107this syntax: 108 109 modprobe ixgb [<option>=<VAL1>,<VAL2>,...] 110 111For example, with two 10GbE PCI adapters, entering: 112 113 modprobe ixgb TxDescriptors=80,128 114 115loads the ixgb driver with 80 TX resources for the first adapter and 128 TX 116resources for the second adapter. 117 118The default value for each parameter is generally the recommended setting, 119unless otherwise noted. 120 121FlowControl 122Valid Range: 0-3 (0=none, 1=Rx only, 2=Tx only, 3=Rx&Tx) 123Default: Read from the EEPROM 124 If EEPROM is not detected, default is 1 125 This parameter controls the automatic generation(Tx) and response(Rx) to 126 Ethernet PAUSE frames. There are hardware bugs associated with enabling 127 Tx flow control so beware. 128 129RxDescriptors 130Valid Range: 64-512 131Default Value: 512 132 This value is the number of receive descriptors allocated by the driver. 133 Increasing this value allows the driver to buffer more incoming packets. 134 Each descriptor is 16 bytes. A receive buffer is also allocated for 135 each descriptor and can be either 2048, 4056, 8192, or 16384 bytes, 136 depending on the MTU setting. When the MTU size is 1500 or less, the 137 receive buffer size is 2048 bytes. When the MTU is greater than 1500 the 138 receive buffer size will be either 4056, 8192, or 16384 bytes. The 139 maximum MTU size is 16114. 140 141RxIntDelay 142Valid Range: 0-65535 (0=off) 143Default Value: 72 144 This value delays the generation of receive interrupts in units of 145 0.8192 microseconds. Receive interrupt reduction can improve CPU 146 efficiency if properly tuned for specific network traffic. Increasing 147 this value adds extra latency to frame reception and can end up 148 decreasing the throughput of TCP traffic. If the system is reporting 149 dropped receives, this value may be set too high, causing the driver to 150 run out of available receive descriptors. 151 152TxDescriptors 153Valid Range: 64-4096 154Default Value: 256 155 This value is the number of transmit descriptors allocated by the driver. 156 Increasing this value allows the driver to queue more transmits. Each 157 descriptor is 16 bytes. 158 159XsumRX 160Valid Range: 0-1 161Default Value: 1 162 A value of '1' indicates that the driver should enable IP checksum 163 offload for received packets (both UDP and TCP) to the adapter hardware. 164 165 166Improving Performance 167===================== 168 169With the 10 Gigabit server adapters, the default Linux configuration will 170very likely limit the total available throughput artificially. There is a set 171of configuration changes that, when applied together, will increase the ability 172of Linux to transmit and receive data. The following enhancements were 173originally acquired from settings published at http://www.spec.org/web99/ for 174various submitted results using Linux. 175 176NOTE: These changes are only suggestions, and serve as a starting point for 177 tuning your network performance. 178 179The changes are made in three major ways, listed in order of greatest effect: 180- Use ip link to modify the mtu (maximum transmission unit) and the txqueuelen 181 parameter. 182- Use sysctl to modify /proc parameters (essentially kernel tuning) 183- Use setpci to modify the MMRBC field in PCI-X configuration space to increase 184 transmit burst lengths on the bus. 185 186NOTE: setpci modifies the adapter's configuration registers to allow it to read 187up to 4k bytes at a time (for transmits). However, for some systems the 188behavior after modifying this register may be undefined (possibly errors of 189some kind). A power-cycle, hard reset or explicitly setting the e6 register 190back to 22 (setpci -d 8086:1a48 e6.b=22) may be required to get back to a 191stable configuration. 192 193- COPY these lines and paste them into ixgb_perf.sh: 194#!/bin/bash 195echo "configuring network performance , edit this file to change the interface 196or device ID of 10GbE card" 197# set mmrbc to 4k reads, modify only Intel 10GbE device IDs 198# replace 1a48 with appropriate 10GbE device's ID installed on the system, 199# if needed. 200setpci -d 8086:1a48 e6.b=2e 201# set the MTU (max transmission unit) - it requires your switch and clients 202# to change as well. 203# set the txqueuelen 204# your ixgb adapter should be loaded as eth1 for this to work, change if needed 205ip li set dev eth1 mtu 9000 txqueuelen 1000 up 206# call the sysctl utility to modify /proc/sys entries 207sysctl -p ./sysctl_ixgb.conf 208- END ixgb_perf.sh 209 210- COPY these lines and paste them into sysctl_ixgb.conf: 211# some of the defaults may be different for your kernel 212# call this file with sysctl -p <this file> 213# these are just suggested values that worked well to increase throughput in 214# several network benchmark tests, your mileage may vary 215 216### IPV4 specific settings 217# turn TCP timestamp support off, default 1, reduces CPU use 218net.ipv4.tcp_timestamps = 0 219# turn SACK support off, default on 220# on systems with a VERY fast bus -> memory interface this is the big gainer 221net.ipv4.tcp_sack = 0 222# set min/default/max TCP read buffer, default 4096 87380 174760 223net.ipv4.tcp_rmem = 10000000 10000000 10000000 224# set min/pressure/max TCP write buffer, default 4096 16384 131072 225net.ipv4.tcp_wmem = 10000000 10000000 10000000 226# set min/pressure/max TCP buffer space, default 31744 32256 32768 227net.ipv4.tcp_mem = 10000000 10000000 10000000 228 229### CORE settings (mostly for socket and UDP effect) 230# set maximum receive socket buffer size, default 131071 231net.core.rmem_max = 524287 232# set maximum send socket buffer size, default 131071 233net.core.wmem_max = 524287 234# set default receive socket buffer size, default 65535 235net.core.rmem_default = 524287 236# set default send socket buffer size, default 65535 237net.core.wmem_default = 524287 238# set maximum amount of option memory buffers, default 10240 239net.core.optmem_max = 524287 240# set number of unprocessed input packets before kernel starts dropping them; default 300 241net.core.netdev_max_backlog = 300000 242- END sysctl_ixgb.conf 243 244Edit the ixgb_perf.sh script if necessary to change eth1 to whatever interface 245your ixgb driver is using and/or replace '1a48' with appropriate 10GbE device's 246ID installed on the system. 247 248NOTE: Unless these scripts are added to the boot process, these changes will 249 only last only until the next system reboot. 250 251 252Resolving Slow UDP Traffic 253-------------------------- 254If your server does not seem to be able to receive UDP traffic as fast as it 255can receive TCP traffic, it could be because Linux, by default, does not set 256the network stack buffers as large as they need to be to support high UDP 257transfer rates. One way to alleviate this problem is to allow more memory to 258be used by the IP stack to store incoming data. 259 260For instance, use the commands: 261 sysctl -w net.core.rmem_max=262143 262and 263 sysctl -w net.core.rmem_default=262143 264to increase the read buffer memory max and default to 262143 (256k - 1) from 265defaults of max=131071 (128k - 1) and default=65535 (64k - 1). These variables 266will increase the amount of memory used by the network stack for receives, and 267can be increased significantly more if necessary for your application. 268 269 270Additional Configurations 271========================= 272 273 Configuring the Driver on Different Distributions 274 ------------------------------------------------- 275 Configuring a network driver to load properly when the system is started is 276 distribution dependent. Typically, the configuration process involves adding 277 an alias line to /etc/modprobe.conf as well as editing other system startup 278 scripts and/or configuration files. Many popular Linux distributions ship 279 with tools to make these changes for you. To learn the proper way to 280 configure a network device for your system, refer to your distribution 281 documentation. If during this process you are asked for the driver or module 282 name, the name for the Linux Base Driver for the Intel 10GbE Family of 283 Adapters is ixgb. 284 285 Viewing Link Messages 286 --------------------- 287 Link messages will not be displayed to the console if the distribution is 288 restricting system messages. In order to see network driver link messages on 289 your console, set dmesg to eight by entering the following: 290 291 dmesg -n 8 292 293 NOTE: This setting is not saved across reboots. 294 295 296 Jumbo Frames 297 ------------ 298 The driver supports Jumbo Frames for all adapters. Jumbo Frames support is 299 enabled by changing the MTU to a value larger than the default of 1500. 300 The maximum value for the MTU is 16114. Use the ip command to 301 increase the MTU size. For example: 302 303 ip li set dev ethx mtu 9000 304 305 The maximum MTU setting for Jumbo Frames is 16114. This value coincides 306 with the maximum Jumbo Frames size of 16128. 307 308 309 ethtool 310 ------- 311 The driver utilizes the ethtool interface for driver configuration and 312 diagnostics, as well as displaying statistical information. The ethtool 313 version 1.6 or later is required for this functionality. 314 315 The latest release of ethtool can be found from 316 http://ftp.kernel.org/pub/software/network/ethtool/ 317 318 NOTE: The ethtool version 1.6 only supports a limited set of ethtool options. 319 Support for a more complete ethtool feature set can be enabled by 320 upgrading to the latest version. 321 322 323 NAPI 324 ---- 325 326 NAPI (Rx polling mode) is supported in the ixgb driver. NAPI is enabled 327 or disabled based on the configuration of the kernel. see CONFIG_IXGB_NAPI 328 329 See www.cyberus.ca/~hadi/usenix-paper.tgz for more information on NAPI. 330 331 332Known Issues/Troubleshooting 333============================ 334 335 NOTE: After installing the driver, if your Intel Network Connection is not 336 working, verify in the "In This Release" section of the readme that you have 337 installed the correct driver. 338 339 Intel(R) PRO/10GbE CX4 Server Adapter Cable Interoperability Issue with 340 Fujitsu XENPAK Module in SmartBits Chassis 341 --------------------------------------------------------------------- 342 Excessive CRC errors may be observed if the Intel(R) PRO/10GbE CX4 343 Server adapter is connected to a Fujitsu XENPAK CX4 module in a SmartBits 344 chassis using 15 m/24AWG cable assemblies manufactured by Fujitsu or Leoni. 345 The CRC errors may be received either by the Intel(R) PRO/10GbE CX4 346 Server adapter or the SmartBits. If this situation occurs using a different 347 cable assembly may resolve the issue. 348 349 CX4 Server Adapter Cable Interoperability Issues with HP Procurve 3400cl 350 Switch Port 351 ------------------------------------------------------------------------ 352 Excessive CRC errors may be observed if the Intel(R) PRO/10GbE CX4 Server 353 adapter is connected to an HP Procurve 3400cl switch port using short cables 354 (1 m or shorter). If this situation occurs, using a longer cable may resolve 355 the issue. 356 357 Excessive CRC errors may be observed using Fujitsu 24AWG cable assemblies that 358 Are 10 m or longer or where using a Leoni 15 m/24AWG cable assembly. The CRC 359 errors may be received either by the CX4 Server adapter or at the switch. If 360 this situation occurs, using a different cable assembly may resolve the issue. 361 362 363 Jumbo Frames System Requirement 364 ------------------------------- 365 Memory allocation failures have been observed on Linux systems with 64 MB 366 of RAM or less that are running Jumbo Frames. If you are using Jumbo 367 Frames, your system may require more than the advertised minimum 368 requirement of 64 MB of system memory. 369 370 371 Performance Degradation with Jumbo Frames 372 ----------------------------------------- 373 Degradation in throughput performance may be observed in some Jumbo frames 374 environments. If this is observed, increasing the application's socket buffer 375 size and/or increasing the /proc/sys/net/ipv4/tcp_*mem entry values may help. 376 See the specific application manual and /usr/src/linux*/Documentation/ 377 networking/ip-sysctl.txt for more details. 378 379 380 Allocating Rx Buffers when Using Jumbo Frames 381 --------------------------------------------- 382 Allocating Rx buffers when using Jumbo Frames on 2.6.x kernels may fail if 383 the available memory is heavily fragmented. This issue may be seen with PCI-X 384 adapters or with packet split disabled. This can be reduced or eliminated 385 by changing the amount of available memory for receive buffer allocation, by 386 increasing /proc/sys/vm/min_free_kbytes. 387 388 389 Multiple Interfaces on Same Ethernet Broadcast Network 390 ------------------------------------------------------ 391 Due to the default ARP behavior on Linux, it is not possible to have 392 one system on two IP networks in the same Ethernet broadcast domain 393 (non-partitioned switch) behave as expected. All Ethernet interfaces 394 will respond to IP traffic for any IP address assigned to the system. 395 This results in unbalanced receive traffic. 396 397 If you have multiple interfaces in a server, do either of the following: 398 399 - Turn on ARP filtering by entering: 400 echo 1 > /proc/sys/net/ipv4/conf/all/arp_filter 401 402 - Install the interfaces in separate broadcast domains - either in 403 different switches or in a switch partitioned to VLANs. 404 405 406 UDP Stress Test Dropped Packet Issue 407 -------------------------------------- 408 Under small packets UDP stress test with 10GbE driver, the Linux system 409 may drop UDP packets due to the fullness of socket buffers. You may want 410 to change the driver's Flow Control variables to the minimum value for 411 controlling packet reception. 412 413 414 Tx Hangs Possible Under Stress 415 ------------------------------ 416 Under stress conditions, if TX hangs occur, turning off TSO 417 "ethtool -K eth0 tso off" may resolve the problem. 418 419 420Support 421======= 422 423For general information, go to the Intel support website at: 424 425 http://support.intel.com 426 427or the Intel Wired Networking project hosted by Sourceforge at: 428 429 http://sourceforge.net/projects/e1000 430 431If an issue is identified with the released source code on the supported 432kernel with a supported adapter, email the specific information related 433to the issue to e1000-devel@lists.sf.net 434