1=========================================================== 2Linux* Base Driver for Intel(R) Ethernet Network Connection 3=========================================================== 4 5Intel Gigabit Linux driver. 6Copyright(c) 1999 - 2013 Intel Corporation. 7 8Contents 9======== 10 11- Identifying Your Adapter 12- Command Line Parameters 13- Speed and Duplex Configuration 14- Additional Configurations 15- Support 16 17Identifying Your Adapter 18======================== 19 20For more information on how to identify your adapter, go to the Adapter & 21Driver ID Guide at: 22 23 http://support.intel.com/support/go/network/adapter/idguide.htm 24 25For the latest Intel network drivers for Linux, refer to the following 26website. In the search field, enter your adapter name or type, or use the 27networking link on the left to search for your adapter: 28 29 http://support.intel.com/support/go/network/adapter/home.htm 30 31Command Line Parameters 32======================= 33 34The default value for each parameter is generally the recommended setting, 35unless otherwise noted. 36 37NOTES: 38 For more information about the AutoNeg, Duplex, and Speed 39 parameters, see the "Speed and Duplex Configuration" section in 40 this document. 41 42 For more information about the InterruptThrottleRate, 43 RxIntDelay, TxIntDelay, RxAbsIntDelay, and TxAbsIntDelay 44 parameters, see the application note at: 45 http://www.intel.com/design/network/applnots/ap450.htm 46 47AutoNeg 48------- 49 50(Supported only on adapters with copper connections) 51 52:Valid Range: 0x01-0x0F, 0x20-0x2F 53:Default Value: 0x2F 54 55This parameter is a bit-mask that specifies the speed and duplex settings 56advertised by the adapter. When this parameter is used, the Speed and 57Duplex parameters must not be specified. 58 59NOTE: 60 Refer to the Speed and Duplex section of this readme for more 61 information on the AutoNeg parameter. 62 63Duplex 64------ 65 66(Supported only on adapters with copper connections) 67 68:Valid Range: 0-2 (0=auto-negotiate, 1=half, 2=full) 69:Default Value: 0 70 71This defines the direction in which data is allowed to flow. Can be 72either one or two-directional. If both Duplex and the link partner are 73set to auto-negotiate, the board auto-detects the correct duplex. If the 74link partner is forced (either full or half), Duplex defaults to half- 75duplex. 76 77FlowControl 78----------- 79 80:Valid Range: 0-3 (0=none, 1=Rx only, 2=Tx only, 3=Rx&Tx) 81:Default Value: Reads flow control settings from the EEPROM 82 83This parameter controls the automatic generation(Tx) and response(Rx) 84to Ethernet PAUSE frames. 85 86InterruptThrottleRate 87--------------------- 88 89(not supported on Intel(R) 82542, 82543 or 82544-based adapters) 90 91:Valid Range: 92 0,1,3,4,100-100000 (0=off, 1=dynamic, 3=dynamic conservative, 93 4=simplified balancing) 94:Default Value: 3 95 96The driver can limit the amount of interrupts per second that the adapter 97will generate for incoming packets. It does this by writing a value to the 98adapter that is based on the maximum amount of interrupts that the adapter 99will generate per second. 100 101Setting InterruptThrottleRate to a value greater or equal to 100 102will program the adapter to send out a maximum of that many interrupts 103per second, even if more packets have come in. This reduces interrupt 104load on the system and can lower CPU utilization under heavy load, 105but will increase latency as packets are not processed as quickly. 106 107The default behaviour of the driver previously assumed a static 108InterruptThrottleRate value of 8000, providing a good fallback value for 109all traffic types,but lacking in small packet performance and latency. 110The hardware can handle many more small packets per second however, and 111for this reason an adaptive interrupt moderation algorithm was implemented. 112 113Since 7.3.x, the driver has two adaptive modes (setting 1 or 3) in which 114it dynamically adjusts the InterruptThrottleRate value based on the traffic 115that it receives. After determining the type of incoming traffic in the last 116timeframe, it will adjust the InterruptThrottleRate to an appropriate value 117for that traffic. 118 119The algorithm classifies the incoming traffic every interval into 120classes. Once the class is determined, the InterruptThrottleRate value is 121adjusted to suit that traffic type the best. There are three classes defined: 122"Bulk traffic", for large amounts of packets of normal size; "Low latency", 123for small amounts of traffic and/or a significant percentage of small 124packets; and "Lowest latency", for almost completely small packets or 125minimal traffic. 126 127In dynamic conservative mode, the InterruptThrottleRate value is set to 4000 128for traffic that falls in class "Bulk traffic". If traffic falls in the "Low 129latency" or "Lowest latency" class, the InterruptThrottleRate is increased 130stepwise to 20000. This default mode is suitable for most applications. 131 132For situations where low latency is vital such as cluster or 133grid computing, the algorithm can reduce latency even more when 134InterruptThrottleRate is set to mode 1. In this mode, which operates 135the same as mode 3, the InterruptThrottleRate will be increased stepwise to 13670000 for traffic in class "Lowest latency". 137 138In simplified mode the interrupt rate is based on the ratio of TX and 139RX traffic. If the bytes per second rate is approximately equal, the 140interrupt rate will drop as low as 2000 interrupts per second. If the 141traffic is mostly transmit or mostly receive, the interrupt rate could 142be as high as 8000. 143 144Setting InterruptThrottleRate to 0 turns off any interrupt moderation 145and may improve small packet latency, but is generally not suitable 146for bulk throughput traffic. 147 148NOTE: 149 InterruptThrottleRate takes precedence over the TxAbsIntDelay and 150 RxAbsIntDelay parameters. In other words, minimizing the receive 151 and/or transmit absolute delays does not force the controller to 152 generate more interrupts than what the Interrupt Throttle Rate 153 allows. 154 155CAUTION: 156 If you are using the Intel(R) PRO/1000 CT Network Connection 157 (controller 82547), setting InterruptThrottleRate to a value 158 greater than 75,000, may hang (stop transmitting) adapters 159 under certain network conditions. If this occurs a NETDEV 160 WATCHDOG message is logged in the system event log. In 161 addition, the controller is automatically reset, restoring 162 the network connection. To eliminate the potential for the 163 hang, ensure that InterruptThrottleRate is set no greater 164 than 75,000 and is not set to 0. 165 166NOTE: 167 When e1000 is loaded with default settings and multiple adapters 168 are in use simultaneously, the CPU utilization may increase non- 169 linearly. In order to limit the CPU utilization without impacting 170 the overall throughput, we recommend that you load the driver as 171 follows:: 172 173 modprobe e1000 InterruptThrottleRate=3000,3000,3000 174 175 This sets the InterruptThrottleRate to 3000 interrupts/sec for 176 the first, second, and third instances of the driver. The range 177 of 2000 to 3000 interrupts per second works on a majority of 178 systems and is a good starting point, but the optimal value will 179 be platform-specific. If CPU utilization is not a concern, use 180 RX_POLLING (NAPI) and default driver settings. 181 182RxDescriptors 183------------- 184 185:Valid Range: 186 - 48-256 for 82542 and 82543-based adapters 187 - 48-4096 for all other supported adapters 188:Default Value: 256 189 190This value specifies the number of receive buffer descriptors allocated 191by the driver. Increasing this value allows the driver to buffer more 192incoming packets, at the expense of increased system memory utilization. 193 194Each descriptor is 16 bytes. A receive buffer is also allocated for each 195descriptor and can be either 2048, 4096, 8192, or 16384 bytes, depending 196on the MTU setting. The maximum MTU size is 16110. 197 198NOTE: 199 MTU designates the frame size. It only needs to be set for Jumbo 200 Frames. Depending on the available system resources, the request 201 for a higher number of receive descriptors may be denied. In this 202 case, use a lower number. 203 204RxIntDelay 205---------- 206 207:Valid Range: 0-65535 (0=off) 208:Default Value: 0 209 210This value delays the generation of receive interrupts in units of 1.024 211microseconds. Receive interrupt reduction can improve CPU efficiency if 212properly tuned for specific network traffic. Increasing this value adds 213extra latency to frame reception and can end up decreasing the throughput 214of TCP traffic. If the system is reporting dropped receives, this value 215may be set too high, causing the driver to run out of available receive 216descriptors. 217 218CAUTION: 219 When setting RxIntDelay to a value other than 0, adapters may 220 hang (stop transmitting) under certain network conditions. If 221 this occurs a NETDEV WATCHDOG message is logged in the system 222 event log. In addition, the controller is automatically reset, 223 restoring the network connection. To eliminate the potential 224 for the hang ensure that RxIntDelay is set to 0. 225 226RxAbsIntDelay 227------------- 228 229(This parameter is supported only on 82540, 82545 and later adapters.) 230 231:Valid Range: 0-65535 (0=off) 232:Default Value: 128 233 234This value, in units of 1.024 microseconds, limits the delay in which a 235receive interrupt is generated. Useful only if RxIntDelay is non-zero, 236this value ensures that an interrupt is generated after the initial 237packet is received within the set amount of time. Proper tuning, 238along with RxIntDelay, may improve traffic throughput in specific network 239conditions. 240 241Speed 242----- 243 244(This parameter is supported only on adapters with copper connections.) 245 246:Valid Settings: 0, 10, 100, 1000 247:Default Value: 0 (auto-negotiate at all supported speeds) 248 249Speed forces the line speed to the specified value in megabits per second 250(Mbps). If this parameter is not specified or is set to 0 and the link 251partner is set to auto-negotiate, the board will auto-detect the correct 252speed. Duplex should also be set when Speed is set to either 10 or 100. 253 254TxDescriptors 255------------- 256 257:Valid Range: 258 - 48-256 for 82542 and 82543-based adapters 259 - 48-4096 for all other supported adapters 260:Default Value: 256 261 262This value is the number of transmit descriptors allocated by the driver. 263Increasing this value allows the driver to queue more transmits. Each 264descriptor is 16 bytes. 265 266NOTE: 267 Depending on the available system resources, the request for a 268 higher number of transmit descriptors may be denied. In this case, 269 use a lower number. 270 271TxIntDelay 272---------- 273 274:Valid Range: 0-65535 (0=off) 275:Default Value: 8 276 277This value delays the generation of transmit interrupts in units of 2781.024 microseconds. Transmit interrupt reduction can improve CPU 279efficiency if properly tuned for specific network traffic. If the 280system is reporting dropped transmits, this value may be set too high 281causing the driver to run out of available transmit descriptors. 282 283TxAbsIntDelay 284------------- 285 286(This parameter is supported only on 82540, 82545 and later adapters.) 287 288:Valid Range: 0-65535 (0=off) 289:Default Value: 32 290 291This value, in units of 1.024 microseconds, limits the delay in which a 292transmit interrupt is generated. Useful only if TxIntDelay is non-zero, 293this value ensures that an interrupt is generated after the initial 294packet is sent on the wire within the set amount of time. Proper tuning, 295along with TxIntDelay, may improve traffic throughput in specific 296network conditions. 297 298XsumRX 299------ 300 301(This parameter is NOT supported on the 82542-based adapter.) 302 303:Valid Range: 0-1 304:Default Value: 1 305 306A value of '1' indicates that the driver should enable IP checksum 307offload for received packets (both UDP and TCP) to the adapter hardware. 308 309Copybreak 310--------- 311 312:Valid Range: 0-xxxxxxx (0=off) 313:Default Value: 256 314:Usage: modprobe e1000.ko copybreak=128 315 316Driver copies all packets below or equaling this size to a fresh RX 317buffer before handing it up the stack. 318 319This parameter is different than other parameters, in that it is a 320single (not 1,1,1 etc.) parameter applied to all driver instances and 321it is also available during runtime at 322/sys/module/e1000/parameters/copybreak 323 324SmartPowerDownEnable 325-------------------- 326 327:Valid Range: 0-1 328:Default Value: 0 (disabled) 329 330Allows PHY to turn off in lower power states. The user can turn off 331this parameter in supported chipsets. 332 333Speed and Duplex Configuration 334============================== 335 336Three keywords are used to control the speed and duplex configuration. 337These keywords are Speed, Duplex, and AutoNeg. 338 339If the board uses a fiber interface, these keywords are ignored, and the 340fiber interface board only links at 1000 Mbps full-duplex. 341 342For copper-based boards, the keywords interact as follows: 343 344- The default operation is auto-negotiate. The board advertises all 345 supported speed and duplex combinations, and it links at the highest 346 common speed and duplex mode IF the link partner is set to auto-negotiate. 347 348- If Speed = 1000, limited auto-negotiation is enabled and only 1000 Mbps 349 is advertised (The 1000BaseT spec requires auto-negotiation.) 350 351- If Speed = 10 or 100, then both Speed and Duplex should be set. Auto- 352 negotiation is disabled, and the AutoNeg parameter is ignored. Partner 353 SHOULD also be forced. 354 355The AutoNeg parameter is used when more control is required over the 356auto-negotiation process. It should be used when you wish to control which 357speed and duplex combinations are advertised during the auto-negotiation 358process. 359 360The parameter may be specified as either a decimal or hexadecimal value as 361determined by the bitmap below. 362 363============== ====== ====== ======= ======= ====== ====== ======= ====== 364Bit position 7 6 5 4 3 2 1 0 365Decimal Value 128 64 32 16 8 4 2 1 366Hex value 80 40 20 10 8 4 2 1 367Speed (Mbps) N/A N/A 1000 N/A 100 100 10 10 368Duplex Full Full Half Full Half 369============== ====== ====== ======= ======= ====== ====== ======= ====== 370 371Some examples of using AutoNeg:: 372 373 modprobe e1000 AutoNeg=0x01 (Restricts autonegotiation to 10 Half) 374 modprobe e1000 AutoNeg=1 (Same as above) 375 modprobe e1000 AutoNeg=0x02 (Restricts autonegotiation to 10 Full) 376 modprobe e1000 AutoNeg=0x03 (Restricts autonegotiation to 10 Half or 10 Full) 377 modprobe e1000 AutoNeg=0x04 (Restricts autonegotiation to 100 Half) 378 modprobe e1000 AutoNeg=0x05 (Restricts autonegotiation to 10 Half or 100 379 Half) 380 modprobe e1000 AutoNeg=0x020 (Restricts autonegotiation to 1000 Full) 381 modprobe e1000 AutoNeg=32 (Same as above) 382 383Note that when this parameter is used, Speed and Duplex must not be specified. 384 385If the link partner is forced to a specific speed and duplex, then this 386parameter should not be used. Instead, use the Speed and Duplex parameters 387previously mentioned to force the adapter to the same speed and duplex. 388 389Additional Configurations 390========================= 391 392Jumbo Frames 393------------ 394 395 Jumbo Frames support is enabled by changing the MTU to a value larger than 396 the default of 1500. Use the ifconfig command to increase the MTU size. 397 For example:: 398 399 ifconfig eth<x> mtu 9000 up 400 401 This setting is not saved across reboots. It can be made permanent if 402 you add:: 403 404 MTU=9000 405 406 to the file /etc/sysconfig/network-scripts/ifcfg-eth<x>. This example 407 applies to the Red Hat distributions; other distributions may store this 408 setting in a different location. 409 410Notes: 411 Degradation in throughput performance may be observed in some Jumbo frames 412 environments. If this is observed, increasing the application's socket buffer 413 size and/or increasing the /proc/sys/net/ipv4/tcp_*mem entry values may help. 414 See the specific application manual and /usr/src/linux*/Documentation/ 415 networking/ip-sysctl.txt for more details. 416 417 - The maximum MTU setting for Jumbo Frames is 16110. This value coincides 418 with the maximum Jumbo Frames size of 16128. 419 420 - Using Jumbo frames at 10 or 100 Mbps is not supported and may result in 421 poor performance or loss of link. 422 423 - Adapters based on the Intel(R) 82542 and 82573V/E controller do not 424 support Jumbo Frames. These correspond to the following product names:: 425 426 Intel(R) PRO/1000 Gigabit Server Adapter 427 Intel(R) PRO/1000 PM Network Connection 428 429ethtool 430------- 431 432 The driver utilizes the ethtool interface for driver configuration and 433 diagnostics, as well as displaying statistical information. The ethtool 434 version 1.6 or later is required for this functionality. 435 436 The latest release of ethtool can be found from 437 https://www.kernel.org/pub/software/network/ethtool/ 438 439Enabling Wake on LAN* (WoL) 440--------------------------- 441 442 WoL is configured through the ethtool* utility. 443 444 WoL will be enabled on the system during the next shut down or reboot. 445 For this driver version, in order to enable WoL, the e1000 driver must be 446 loaded when shutting down or rebooting the system. 447 448Support 449======= 450 451For general information, go to the Intel support website at: 452 453 http://support.intel.com 454 455or the Intel Wired Networking project hosted by Sourceforge at: 456 457 http://sourceforge.net/projects/e1000 458 459If an issue is identified with the released source code on the supported 460kernel with a supported adapter, email the specific information related 461to the issue to e1000-devel@lists.sf.net 462
