aboutsummaryrefslogtreecommitdiffstats
path: root/Documentation
diff options
context:
space:
mode:
authorAuke Kok <auke\-jan.h.kok@intel.com>2006-11-01 11:48:18 -0500
committerJeff Garzik <jeff@garzik.org>2006-12-02 00:12:01 -0500
commitde3edab4276c3c789f64dc3d78eea027709fef0e (patch)
tree093a7c20d2473d42470bea3b6617a3084891d90e /Documentation
parent25006ac61e514628b9d0f78fce0bed155f4f109c (diff)
e1000: update README for e1000
Signed-off-by: Auke Kok <auke-jan.h.kok@intel.com>
Diffstat (limited to 'Documentation')
-rw-r--r--Documentation/networking/e1000.txt451
1 files changed, 266 insertions, 185 deletions
diff --git a/Documentation/networking/e1000.txt b/Documentation/networking/e1000.txt
index 5c0a5cc03998..61b171cf5313 100644
--- a/Documentation/networking/e1000.txt
+++ b/Documentation/networking/e1000.txt
@@ -1,7 +1,7 @@
1Linux* Base Driver for the Intel(R) PRO/1000 Family of Adapters 1Linux* Base Driver for the Intel(R) PRO/1000 Family of Adapters
2=============================================================== 2===============================================================
3 3
4November 15, 2005 4September 26, 2006
5 5
6 6
7Contents 7Contents
@@ -9,6 +9,7 @@ Contents
9 9
10- In This Release 10- In This Release
11- Identifying Your Adapter 11- Identifying Your Adapter
12- Building and Installation
12- Command Line Parameters 13- Command Line Parameters
13- Speed and Duplex Configuration 14- Speed and Duplex Configuration
14- Additional Configurations 15- Additional Configurations
@@ -41,6 +42,9 @@ or later), lspci, and ifconfig to obtain the same information.
41Instructions on updating ethtool can be found in the section "Additional 42Instructions on updating ethtool can be found in the section "Additional
42Configurations" later in this document. 43Configurations" later in this document.
43 44
45NOTE: The Intel(R) 82562v 10/100 Network Connection only provides 10/100
46support.
47
44 48
45Identifying Your Adapter 49Identifying Your Adapter
46======================== 50========================
@@ -51,28 +55,27 @@ Driver ID Guide at:
51 http://support.intel.com/support/network/adapter/pro100/21397.htm 55 http://support.intel.com/support/network/adapter/pro100/21397.htm
52 56
53For the latest Intel network drivers for Linux, refer to the following 57For the latest Intel network drivers for Linux, refer to the following
54website. In the search field, enter your adapter name or type, or use the 58website. In the search field, enter your adapter name or type, or use the
55networking link on the left to search for your adapter: 59networking link on the left to search for your adapter:
56 60
57 http://downloadfinder.intel.com/scripts-df/support_intel.asp 61 http://downloadfinder.intel.com/scripts-df/support_intel.asp
58 62
59 63
60Command Line Parameters ======================= 64Command Line Parameters
65=======================
61 66
62If the driver is built as a module, the following optional parameters 67If the driver is built as a module, the following optional parameters
63are used by entering them on the command line with the modprobe or insmod 68are used by entering them on the command line with the modprobe command
64command using this syntax: 69using this syntax:
65 70
66 modprobe e1000 [<option>=<VAL1>,<VAL2>,...] 71 modprobe e1000 [<option>=<VAL1>,<VAL2>,...]
67 72
68 insmod e1000 [<option>=<VAL1>,<VAL2>,...]
69
70For example, with two PRO/1000 PCI adapters, entering: 73For example, with two PRO/1000 PCI adapters, entering:
71 74
72 insmod e1000 TxDescriptors=80,128 75 modprobe e1000 TxDescriptors=80,128
73 76
74loads the e1000 driver with 80 TX descriptors for the first adapter and 128 77loads the e1000 driver with 80 TX descriptors for the first adapter and
75TX descriptors for the second adapter. 78128 TX descriptors for the second adapter.
76 79
77The default value for each parameter is generally the recommended setting, 80The default value for each parameter is generally the recommended setting,
78unless otherwise noted. 81unless otherwise noted.
@@ -87,7 +90,7 @@ NOTES: For more information about the AutoNeg, Duplex, and Speed
87 http://www.intel.com/design/network/applnots/ap450.htm 90 http://www.intel.com/design/network/applnots/ap450.htm
88 91
89 A descriptor describes a data buffer and attributes related to 92 A descriptor describes a data buffer and attributes related to
90 the data buffer. This information is accessed by the hardware. 93 the data buffer. This information is accessed by the hardware.
91 94
92 95
93AutoNeg 96AutoNeg
@@ -96,9 +99,9 @@ AutoNeg
96Valid Range: 0x01-0x0F, 0x20-0x2F 99Valid Range: 0x01-0x0F, 0x20-0x2F
97Default Value: 0x2F 100Default Value: 0x2F
98 101
99This parameter is a bit mask that specifies which speed and duplex 102This parameter is a bit-mask that specifies the speed and duplex settings
100settings the board advertises. When this parameter is used, the Speed 103advertised by the adapter. When this parameter is used, the Speed and
101and Duplex parameters must not be specified. 104Duplex parameters must not be specified.
102 105
103NOTE: Refer to the Speed and Duplex section of this readme for more 106NOTE: Refer to the Speed and Duplex section of this readme for more
104 information on the AutoNeg parameter. 107 information on the AutoNeg parameter.
@@ -110,14 +113,15 @@ Duplex
110Valid Range: 0-2 (0=auto-negotiate, 1=half, 2=full) 113Valid Range: 0-2 (0=auto-negotiate, 1=half, 2=full)
111Default Value: 0 114Default Value: 0
112 115
113Defines the direction in which data is allowed to flow. Can be either 116This defines the direction in which data is allowed to flow. Can be
114one or two-directional. If both Duplex and the link partner are set to 117either one or two-directional. If both Duplex and the link partner are
115auto-negotiate, the board auto-detects the correct duplex. If the link 118set to auto-negotiate, the board auto-detects the correct duplex. If the
116partner is forced (either full or half), Duplex defaults to half-duplex. 119link partner is forced (either full or half), Duplex defaults to half-
120duplex.
117 121
118 122
119FlowControl 123FlowControl
120---------- 124-----------
121Valid Range: 0-3 (0=none, 1=Rx only, 2=Tx only, 3=Rx&Tx) 125Valid Range: 0-3 (0=none, 1=Rx only, 2=Tx only, 3=Rx&Tx)
122Default Value: Reads flow control settings from the EEPROM 126Default Value: Reads flow control settings from the EEPROM
123 127
@@ -127,57 +131,107 @@ to Ethernet PAUSE frames.
127 131
128InterruptThrottleRate 132InterruptThrottleRate
129--------------------- 133---------------------
130(not supported on Intel 82542, 82543 or 82544-based adapters) 134(not supported on Intel(R) 82542, 82543 or 82544-based adapters)
131Valid Range: 100-100000 (0=off, 1=dynamic) 135Valid Range: 0,1,3,100-100000 (0=off, 1=dynamic, 3=dynamic conservative)
132Default Value: 8000 136Default Value: 3
133 137
134This value represents the maximum number of interrupts per second the 138The driver can limit the amount of interrupts per second that the adapter
135controller generates. InterruptThrottleRate is another setting used in 139will generate for incoming packets. It does this by writing a value to the
136interrupt moderation. Dynamic mode uses a heuristic algorithm to adjust 140adapter that is based on the maximum amount of interrupts that the adapter
137InterruptThrottleRate based on the current traffic load. 141will generate per second.
142
143Setting InterruptThrottleRate to a value greater or equal to 100
144will program the adapter to send out a maximum of that many interrupts
145per second, even if more packets have come in. This reduces interrupt
146load on the system and can lower CPU utilization under heavy load,
147but will increase latency as packets are not processed as quickly.
148
149The default behaviour of the driver previously assumed a static
150InterruptThrottleRate value of 8000, providing a good fallback value for
151all traffic types,but lacking in small packet performance and latency.
152The hardware can handle many more small packets per second however, and
153for this reason an adaptive interrupt moderation algorithm was implemented.
154
155Since 7.3.x, the driver has two adaptive modes (setting 1 or 3) in which
156it dynamically adjusts the InterruptThrottleRate value based on the traffic
157that it receives. After determining the type of incoming traffic in the last
158timeframe, it will adjust the InterruptThrottleRate to an appropriate value
159for that traffic.
160
161The algorithm classifies the incoming traffic every interval into
162classes. Once the class is determined, the InterruptThrottleRate value is
163adjusted to suit that traffic type the best. There are three classes defined:
164"Bulk traffic", for large amounts of packets of normal size; "Low latency",
165for small amounts of traffic and/or a significant percentage of small
166packets; and "Lowest latency", for almost completely small packets or
167minimal traffic.
168
169In dynamic conservative mode, the InterruptThrottleRate value is set to 4000
170for traffic that falls in class "Bulk traffic". If traffic falls in the "Low
171latency" or "Lowest latency" class, the InterruptThrottleRate is increased
172stepwise to 20000. This default mode is suitable for most applications.
173
174For situations where low latency is vital such as cluster or
175grid computing, the algorithm can reduce latency even more when
176InterruptThrottleRate is set to mode 1. In this mode, which operates
177the same as mode 3, the InterruptThrottleRate will be increased stepwise to
17870000 for traffic in class "Lowest latency".
179
180Setting InterruptThrottleRate to 0 turns off any interrupt moderation
181and may improve small packet latency, but is generally not suitable
182for bulk throughput traffic.
138 183
139NOTE: InterruptThrottleRate takes precedence over the TxAbsIntDelay and 184NOTE: InterruptThrottleRate takes precedence over the TxAbsIntDelay and
140 RxAbsIntDelay parameters. In other words, minimizing the receive 185 RxAbsIntDelay parameters. In other words, minimizing the receive
141 and/or transmit absolute delays does not force the controller to 186 and/or transmit absolute delays does not force the controller to
142 generate more interrupts than what the Interrupt Throttle Rate 187 generate more interrupts than what the Interrupt Throttle Rate
143 allows. 188 allows.
144 189
145CAUTION: If you are using the Intel PRO/1000 CT Network Connection 190CAUTION: If you are using the Intel(R) PRO/1000 CT Network Connection
146 (controller 82547), setting InterruptThrottleRate to a value 191 (controller 82547), setting InterruptThrottleRate to a value
147 greater than 75,000, may hang (stop transmitting) adapters 192 greater than 75,000, may hang (stop transmitting) adapters
148 under certain network conditions. If this occurs a NETDEV 193 under certain network conditions. If this occurs a NETDEV
149 WATCHDOG message is logged in the system event log. In 194 WATCHDOG message is logged in the system event log. In
150 addition, the controller is automatically reset, restoring 195 addition, the controller is automatically reset, restoring
151 the network connection. To eliminate the potential for the 196 the network connection. To eliminate the potential for the
152 hang, ensure that InterruptThrottleRate is set no greater 197 hang, ensure that InterruptThrottleRate is set no greater
153 than 75,000 and is not set to 0. 198 than 75,000 and is not set to 0.
154 199
155NOTE: When e1000 is loaded with default settings and multiple adapters 200NOTE: When e1000 is loaded with default settings and multiple adapters
156 are in use simultaneously, the CPU utilization may increase non- 201 are in use simultaneously, the CPU utilization may increase non-
157 linearly. In order to limit the CPU utilization without impacting 202 linearly. In order to limit the CPU utilization without impacting
158 the overall throughput, we recommend that you load the driver as 203 the overall throughput, we recommend that you load the driver as
159 follows: 204 follows:
160 205
161 insmod e1000.o InterruptThrottleRate=3000,3000,3000 206 modprobe e1000 InterruptThrottleRate=3000,3000,3000
162 207
163 This sets the InterruptThrottleRate to 3000 interrupts/sec for 208 This sets the InterruptThrottleRate to 3000 interrupts/sec for
164 the first, second, and third instances of the driver. The range 209 the first, second, and third instances of the driver. The range
165 of 2000 to 3000 interrupts per second works on a majority of 210 of 2000 to 3000 interrupts per second works on a majority of
166 systems and is a good starting point, but the optimal value will 211 systems and is a good starting point, but the optimal value will
167 be platform-specific. If CPU utilization is not a concern, use 212 be platform-specific. If CPU utilization is not a concern, use
168 RX_POLLING (NAPI) and default driver settings. 213 RX_POLLING (NAPI) and default driver settings.
169 214
170 215
216
171RxDescriptors 217RxDescriptors
172------------- 218-------------
173Valid Range: 80-256 for 82542 and 82543-based adapters 219Valid Range: 80-256 for 82542 and 82543-based adapters
174 80-4096 for all other supported adapters 220 80-4096 for all other supported adapters
175Default Value: 256 221Default Value: 256
176 222
177This value specifies the number of receive descriptors allocated by the 223This value specifies the number of receive buffer descriptors allocated
178driver. Increasing this value allows the driver to buffer more incoming 224by the driver. Increasing this value allows the driver to buffer more
179packets. Each descriptor is 16 bytes. A receive buffer is also 225incoming packets, at the expense of increased system memory utilization.
180allocated for each descriptor and is 2048. 226
227Each descriptor is 16 bytes. A receive buffer is also allocated for each
228descriptor and can be either 2048, 4096, 8192, or 16384 bytes, depending
229on the MTU setting. The maximum MTU size is 16110.
230
231NOTE: MTU designates the frame size. It only needs to be set for Jumbo
232 Frames. Depending on the available system resources, the request
233 for a higher number of receive descriptors may be denied. In this
234 case, use a lower number.
181 235
182 236
183RxIntDelay 237RxIntDelay
@@ -187,17 +241,17 @@ Default Value: 0
187 241
188This value delays the generation of receive interrupts in units of 1.024 242This value delays the generation of receive interrupts in units of 1.024
189microseconds. Receive interrupt reduction can improve CPU efficiency if 243microseconds. Receive interrupt reduction can improve CPU efficiency if
190properly tuned for specific network traffic. Increasing this value adds 244properly tuned for specific network traffic. Increasing this value adds
191extra latency to frame reception and can end up decreasing the throughput 245extra latency to frame reception and can end up decreasing the throughput
192of TCP traffic. If the system is reporting dropped receives, this value 246of TCP traffic. If the system is reporting dropped receives, this value
193may be set too high, causing the driver to run out of available receive 247may be set too high, causing the driver to run out of available receive
194descriptors. 248descriptors.
195 249
196CAUTION: When setting RxIntDelay to a value other than 0, adapters may 250CAUTION: When setting RxIntDelay to a value other than 0, adapters may
197 hang (stop transmitting) under certain network conditions. If 251 hang (stop transmitting) under certain network conditions. If
198 this occurs a NETDEV WATCHDOG message is logged in the system 252 this occurs a NETDEV WATCHDOG message is logged in the system
199 event log. In addition, the controller is automatically reset, 253 event log. In addition, the controller is automatically reset,
200 restoring the network connection. To eliminate the potential 254 restoring the network connection. To eliminate the potential
201 for the hang ensure that RxIntDelay is set to 0. 255 for the hang ensure that RxIntDelay is set to 0.
202 256
203 257
@@ -208,7 +262,7 @@ Valid Range: 0-65535 (0=off)
208Default Value: 128 262Default Value: 128
209 263
210This value, in units of 1.024 microseconds, limits the delay in which a 264This value, in units of 1.024 microseconds, limits the delay in which a
211receive interrupt is generated. Useful only if RxIntDelay is non-zero, 265receive interrupt is generated. Useful only if RxIntDelay is non-zero,
212this value ensures that an interrupt is generated after the initial 266this value ensures that an interrupt is generated after the initial
213packet is received within the set amount of time. Proper tuning, 267packet is received within the set amount of time. Proper tuning,
214along with RxIntDelay, may improve traffic throughput in specific network 268along with RxIntDelay, may improve traffic throughput in specific network
@@ -222,9 +276,9 @@ Valid Settings: 0, 10, 100, 1000
222Default Value: 0 (auto-negotiate at all supported speeds) 276Default Value: 0 (auto-negotiate at all supported speeds)
223 277
224Speed forces the line speed to the specified value in megabits per second 278Speed forces the line speed to the specified value in megabits per second
225(Mbps). If this parameter is not specified or is set to 0 and the link 279(Mbps). If this parameter is not specified or is set to 0 and the link
226partner is set to auto-negotiate, the board will auto-detect the correct 280partner is set to auto-negotiate, the board will auto-detect the correct
227speed. Duplex should also be set when Speed is set to either 10 or 100. 281speed. Duplex should also be set when Speed is set to either 10 or 100.
228 282
229 283
230TxDescriptors 284TxDescriptors
@@ -234,7 +288,7 @@ Valid Range: 80-256 for 82542 and 82543-based adapters
234Default Value: 256 288Default Value: 256
235 289
236This value is the number of transmit descriptors allocated by the driver. 290This value is the number of transmit descriptors allocated by the driver.
237Increasing this value allows the driver to queue more transmits. Each 291Increasing this value allows the driver to queue more transmits. Each
238descriptor is 16 bytes. 292descriptor is 16 bytes.
239 293
240NOTE: Depending on the available system resources, the request for a 294NOTE: Depending on the available system resources, the request for a
@@ -248,8 +302,8 @@ Valid Range: 0-65535 (0=off)
248Default Value: 64 302Default Value: 64
249 303
250This value delays the generation of transmit interrupts in units of 304This value delays the generation of transmit interrupts in units of
2511.024 microseconds. Transmit interrupt reduction can improve CPU 3051.024 microseconds. Transmit interrupt reduction can improve CPU
252efficiency if properly tuned for specific network traffic. If the 306efficiency if properly tuned for specific network traffic. If the
253system is reporting dropped transmits, this value may be set too high 307system is reporting dropped transmits, this value may be set too high
254causing the driver to run out of available transmit descriptors. 308causing the driver to run out of available transmit descriptors.
255 309
@@ -261,7 +315,7 @@ Valid Range: 0-65535 (0=off)
261Default Value: 64 315Default Value: 64
262 316
263This value, in units of 1.024 microseconds, limits the delay in which a 317This value, in units of 1.024 microseconds, limits the delay in which a
264transmit interrupt is generated. Useful only if TxIntDelay is non-zero, 318transmit interrupt is generated. Useful only if TxIntDelay is non-zero,
265this value ensures that an interrupt is generated after the initial 319this value ensures that an interrupt is generated after the initial
266packet is sent on the wire within the set amount of time. Proper tuning, 320packet is sent on the wire within the set amount of time. Proper tuning,
267along with TxIntDelay, may improve traffic throughput in specific 321along with TxIntDelay, may improve traffic throughput in specific
@@ -288,15 +342,15 @@ fiber interface board only links at 1000 Mbps full-duplex.
288 342
289For copper-based boards, the keywords interact as follows: 343For copper-based boards, the keywords interact as follows:
290 344
291 The default operation is auto-negotiate. The board advertises all 345 The default operation is auto-negotiate. The board advertises all
292 supported speed and duplex combinations, and it links at the highest 346 supported speed and duplex combinations, and it links at the highest
293 common speed and duplex mode IF the link partner is set to auto-negotiate. 347 common speed and duplex mode IF the link partner is set to auto-negotiate.
294 348
295 If Speed = 1000, limited auto-negotiation is enabled and only 1000 Mbps 349 If Speed = 1000, limited auto-negotiation is enabled and only 1000 Mbps
296 is advertised (The 1000BaseT spec requires auto-negotiation.) 350 is advertised (The 1000BaseT spec requires auto-negotiation.)
297 351
298 If Speed = 10 or 100, then both Speed and Duplex should be set. Auto- 352 If Speed = 10 or 100, then both Speed and Duplex should be set. Auto-
299 negotiation is disabled, and the AutoNeg parameter is ignored. Partner 353 negotiation is disabled, and the AutoNeg parameter is ignored. Partner
300 SHOULD also be forced. 354 SHOULD also be forced.
301 355
302The AutoNeg parameter is used when more control is required over the 356The AutoNeg parameter is used when more control is required over the
@@ -304,7 +358,7 @@ auto-negotiation process. It should be used when you wish to control which
304speed and duplex combinations are advertised during the auto-negotiation 358speed and duplex combinations are advertised during the auto-negotiation
305process. 359process.
306 360
307The parameter may be specified as either a decimal or hexidecimal value as 361The parameter may be specified as either a decimal or hexadecimal value as
308determined by the bitmap below. 362determined by the bitmap below.
309 363
310Bit position 7 6 5 4 3 2 1 0 364Bit position 7 6 5 4 3 2 1 0
@@ -337,20 +391,19 @@ Additional Configurations
337 391
338 Configuring the Driver on Different Distributions 392 Configuring the Driver on Different Distributions
339 ------------------------------------------------- 393 -------------------------------------------------
340
341 Configuring a network driver to load properly when the system is started 394 Configuring a network driver to load properly when the system is started
342 is distribution dependent. Typically, the configuration process involves 395 is distribution dependent. Typically, the configuration process involves
343 adding an alias line to /etc/modules.conf or /etc/modprobe.conf as well 396 adding an alias line to /etc/modules.conf or /etc/modprobe.conf as well
344 as editing other system startup scripts and/or configuration files. Many 397 as editing other system startup scripts and/or configuration files. Many
345 popular Linux distributions ship with tools to make these changes for you. 398 popular Linux distributions ship with tools to make these changes for you.
346 To learn the proper way to configure a network device for your system, 399 To learn the proper way to configure a network device for your system,
347 refer to your distribution documentation. If during this process you are 400 refer to your distribution documentation. If during this process you are
348 asked for the driver or module name, the name for the Linux Base Driver 401 asked for the driver or module name, the name for the Linux Base Driver
349 for the Intel PRO/1000 Family of Adapters is e1000. 402 for the Intel(R) PRO/1000 Family of Adapters is e1000.
350 403
351 As an example, if you install the e1000 driver for two PRO/1000 adapters 404 As an example, if you install the e1000 driver for two PRO/1000 adapters
352 (eth0 and eth1) and set the speed and duplex to 10full and 100half, add 405 (eth0 and eth1) and set the speed and duplex to 10full and 100half, add
353 the following to modules.conf or modprobe.conf: 406 the following to modules.conf or or modprobe.conf:
354 407
355 alias eth0 e1000 408 alias eth0 e1000
356 alias eth1 e1000 409 alias eth1 e1000
@@ -358,9 +411,8 @@ Additional Configurations
358 411
359 Viewing Link Messages 412 Viewing Link Messages
360 --------------------- 413 ---------------------
361
362 Link messages will not be displayed to the console if the distribution is 414 Link messages will not be displayed to the console if the distribution is
363 restricting system messages. In order to see network driver link messages 415 restricting system messages. In order to see network driver link messages
364 on your console, set dmesg to eight by entering the following: 416 on your console, set dmesg to eight by entering the following:
365 417
366 dmesg -n 8 418 dmesg -n 8
@@ -369,11 +421,9 @@ Additional Configurations
369 421
370 Jumbo Frames 422 Jumbo Frames
371 ------------ 423 ------------
372 424 Jumbo Frames support is enabled by changing the MTU to a value larger than
373 The driver supports Jumbo Frames for all adapters except 82542 and 425 the default of 1500. Use the ifconfig command to increase the MTU size.
374 82573-based adapters. Jumbo Frames support is enabled by changing the 426 For example:
375 MTU to a value larger than the default of 1500. Use the ifconfig command
376 to increase the MTU size. For example:
377 427
378 ifconfig eth<x> mtu 9000 up 428 ifconfig eth<x> mtu 9000 up
379 429
@@ -390,26 +440,49 @@ Additional Configurations
390 440
391 - To enable Jumbo Frames, increase the MTU size on the interface beyond 441 - To enable Jumbo Frames, increase the MTU size on the interface beyond
392 1500. 442 1500.
393 - The maximum MTU setting for Jumbo Frames is 16110. This value coincides 443
444 - The maximum MTU setting for Jumbo Frames is 16110. This value coincides
394 with the maximum Jumbo Frames size of 16128. 445 with the maximum Jumbo Frames size of 16128.
446
395 - Using Jumbo Frames at 10 or 100 Mbps may result in poor performance or 447 - Using Jumbo Frames at 10 or 100 Mbps may result in poor performance or
396 loss of link. 448 loss of link.
449
397 - Some Intel gigabit adapters that support Jumbo Frames have a frame size 450 - Some Intel gigabit adapters that support Jumbo Frames have a frame size
398 limit of 9238 bytes, with a corresponding MTU size limit of 9216 bytes. 451 limit of 9238 bytes, with a corresponding MTU size limit of 9216 bytes.
399 The adapters with this limitation are based on the Intel 82571EB and 452 The adapters with this limitation are based on the Intel(R) 82571EB,
400 82572EI controllers, which correspond to these product names: 453 82572EI, 82573L and 80003ES2LAN controller. These correspond to the
401 Intel� PRO/1000 PT Dual Port Server Adapter 454 following product names:
402 Intel� PRO/1000 PF Dual Port Server Adapter 455 Intel(R) PRO/1000 PT Server Adapter
403 Intel� PRO/1000 PT Server Adapter 456 Intel(R) PRO/1000 PT Desktop Adapter
404 Intel� PRO/1000 PT Desktop Adapter 457 Intel(R) PRO/1000 PT Network Connection
405 Intel� PRO/1000 PF Server Adapter 458 Intel(R) PRO/1000 PT Dual Port Server Adapter
406 459 Intel(R) PRO/1000 PT Dual Port Network Connection
407 - The Intel PRO/1000 PM Network Connection does not support jumbo frames. 460 Intel(R) PRO/1000 PF Server Adapter
461 Intel(R) PRO/1000 PF Network Connection
462 Intel(R) PRO/1000 PF Dual Port Server Adapter
463 Intel(R) PRO/1000 PB Server Connection
464 Intel(R) PRO/1000 PL Network Connection
465 Intel(R) PRO/1000 EB Network Connection with I/O Acceleration
466 Intel(R) PRO/1000 EB Backplane Connection with I/O Acceleration
467 Intel(R) PRO/1000 PT Quad Port Server Adapter
468
469 - Adapters based on the Intel(R) 82542 and 82573V/E controller do not
470 support Jumbo Frames. These correspond to the following product names:
471 Intel(R) PRO/1000 Gigabit Server Adapter
472 Intel(R) PRO/1000 PM Network Connection
473
474 - The following adapters do not support Jumbo Frames:
475 Intel(R) 82562V 10/100 Network Connection
476 Intel(R) 82566DM Gigabit Network Connection
477 Intel(R) 82566DC Gigabit Network Connection
478 Intel(R) 82566MM Gigabit Network Connection
479 Intel(R) 82566MC Gigabit Network Connection
480 Intel(R) 82562GT 10/100 Network Connection
481 Intel(R) 82562G 10/100 Network Connection
408 482
409 483
410 Ethtool 484 Ethtool
411 ------- 485 -------
412
413 The driver utilizes the ethtool interface for driver configuration and 486 The driver utilizes the ethtool interface for driver configuration and
414 diagnostics, as well as displaying statistical information. Ethtool 487 diagnostics, as well as displaying statistical information. Ethtool
415 version 1.6 or later is required for this functionality. 488 version 1.6 or later is required for this functionality.
@@ -417,15 +490,14 @@ Additional Configurations
417 The latest release of ethtool can be found from 490 The latest release of ethtool can be found from
418 http://sourceforge.net/projects/gkernel. 491 http://sourceforge.net/projects/gkernel.
419 492
420 NOTE: Ethtool 1.6 only supports a limited set of ethtool options. Support 493 NOTE: Ethtool 1.6 only supports a limited set of ethtool options. Support
421 for a more complete ethtool feature set can be enabled by upgrading 494 for a more complete ethtool feature set can be enabled by upgrading
422 ethtool to ethtool-1.8.1. 495 ethtool to ethtool-1.8.1.
423 496
424 Enabling Wake on LAN* (WoL) 497 Enabling Wake on LAN* (WoL)
425 --------------------------- 498 ---------------------------
426 499 WoL is configured through the Ethtool* utility. Ethtool is included with
427 WoL is configured through the Ethtool* utility. Ethtool is included with 500 all versions of Red Hat after Red Hat 7.2. For other Linux distributions,
428 all versions of Red Hat after Red Hat 7.2. For other Linux distributions,
429 download and install Ethtool from the following website: 501 download and install Ethtool from the following website:
430 http://sourceforge.net/projects/gkernel. 502 http://sourceforge.net/projects/gkernel.
431 503
@@ -436,11 +508,17 @@ Additional Configurations
436 For this driver version, in order to enable WoL, the e1000 driver must be 508 For this driver version, in order to enable WoL, the e1000 driver must be
437 loaded when shutting down or rebooting the system. 509 loaded when shutting down or rebooting the system.
438 510
511 Wake On LAN is only supported on port A for the following devices:
512 Intel(R) PRO/1000 PT Dual Port Network Connection
513 Intel(R) PRO/1000 PT Dual Port Server Connection
514 Intel(R) PRO/1000 PT Dual Port Server Adapter
515 Intel(R) PRO/1000 PF Dual Port Server Adapter
516 Intel(R) PRO/1000 PT Quad Port Server Adapter
517
439 NAPI 518 NAPI
440 ---- 519 ----
441 520 NAPI (Rx polling mode) is supported in the e1000 driver. NAPI is enabled
442 NAPI (Rx polling mode) is supported in the e1000 driver. NAPI is enabled 521 or disabled based on the configuration of the kernel. To override
443 or disabled based on the configuration of the kernel. To override
444 the default, use the following compile-time flags. 522 the default, use the following compile-time flags.
445 523
446 To enable NAPI, compile the driver module, passing in a configuration option: 524 To enable NAPI, compile the driver module, passing in a configuration option:
@@ -457,88 +535,105 @@ Additional Configurations
457Known Issues 535Known Issues
458============ 536============
459 537
460 Jumbo Frames System Requirement 538Dropped Receive Packets on Half-duplex 10/100 Networks
461 ------------------------------- 539------------------------------------------------------
462 540If you have an Intel PCI Express adapter running at 10mbps or 100mbps, half-
463 Memory allocation failures have been observed on Linux systems with 64 MB 541duplex, you may observe occasional dropped receive packets. There are no
464 of RAM or less that are running Jumbo Frames. If you are using Jumbo 542workarounds for this problem in this network configuration. The network must
465 Frames, your system may require more than the advertised minimum 543be updated to operate in full-duplex, and/or 1000mbps only.
466 requirement of 64 MB of system memory. 544
467 545Jumbo Frames System Requirement
468 Performance Degradation with Jumbo Frames 546-------------------------------
469 ----------------------------------------- 547Memory allocation failures have been observed on Linux systems with 64 MB
470 548of RAM or less that are running Jumbo Frames. If you are using Jumbo
471 Degradation in throughput performance may be observed in some Jumbo frames 549Frames, your system may require more than the advertised minimum
472 environments. If this is observed, increasing the application's socket 550requirement of 64 MB of system memory.
473 buffer size and/or increasing the /proc/sys/net/ipv4/tcp_*mem entry values 551
474 may help. See the specific application manual and 552Performance Degradation with Jumbo Frames
475 /usr/src/linux*/Documentation/ 553-----------------------------------------
476 networking/ip-sysctl.txt for more details. 554Degradation in throughput performance may be observed in some Jumbo frames
477 555environments. If this is observed, increasing the application's socket
478 Jumbo frames on Foundry BigIron 8000 switch 556buffer size and/or increasing the /proc/sys/net/ipv4/tcp_*mem entry values
479 ------------------------------------------- 557may help. See the specific application manual and
480 There is a known issue using Jumbo frames when connected to a Foundry 558/usr/src/linux*/Documentation/
481 BigIron 8000 switch. This is a 3rd party limitation. If you experience 559networking/ip-sysctl.txt for more details.
482 loss of packets, lower the MTU size. 560
483 561Jumbo Frames on Foundry BigIron 8000 switch
484 Multiple Interfaces on Same Ethernet Broadcast Network 562-------------------------------------------
485 ------------------------------------------------------ 563There is a known issue using Jumbo frames when connected to a Foundry
486 564BigIron 8000 switch. This is a 3rd party limitation. If you experience
487 Due to the default ARP behavior on Linux, it is not possible to have 565loss of packets, lower the MTU size.
488 one system on two IP networks in the same Ethernet broadcast domain 566
489 (non-partitioned switch) behave as expected. All Ethernet interfaces 567Allocating Rx Buffers when Using Jumbo Frames
490 will respond to IP traffic for any IP address assigned to the system. 568---------------------------------------------
491 This results in unbalanced receive traffic. 569Allocating Rx buffers when using Jumbo Frames on 2.6.x kernels may fail if
492 570the available memory is heavily fragmented. This issue may be seen with PCI-X
493 If you have multiple interfaces in a server, either turn on ARP 571adapters or with packet split disabled. This can be reduced or eliminated
494 filtering by entering: 572by changing the amount of available memory for receive buffer allocation, by
495 573increasing /proc/sys/vm/min_free_kbytes.
496 echo 1 > /proc/sys/net/ipv4/conf/all/arp_filter 574
497 (this only works if your kernel's version is higher than 2.4.5), 575Multiple Interfaces on Same Ethernet Broadcast Network
498 576------------------------------------------------------
499 NOTE: This setting is not saved across reboots. The configuration 577Due to the default ARP behavior on Linux, it is not possible to have
500 change can be made permanent by adding the line: 578one system on two IP networks in the same Ethernet broadcast domain
501 net.ipv4.conf.all.arp_filter = 1 579(non-partitioned switch) behave as expected. All Ethernet interfaces
502 to the file /etc/sysctl.conf 580will respond to IP traffic for any IP address assigned to the system.
503 581This results in unbalanced receive traffic.
504 or, 582
505 583If you have multiple interfaces in a server, either turn on ARP
506 install the interfaces in separate broadcast domains (either in 584filtering by entering:
507 different switches or in a switch partitioned to VLANs). 585
508 586 echo 1 > /proc/sys/net/ipv4/conf/all/arp_filter
509 82541/82547 can't link or are slow to link with some link partners 587(this only works if your kernel's version is higher than 2.4.5),
510 ----------------------------------------------------------------- 588
511 589NOTE: This setting is not saved across reboots. The configuration
512 There is a known compatibility issue with 82541/82547 and some 590change can be made permanent by adding the line:
513 low-end switches where the link will not be established, or will 591 net.ipv4.conf.all.arp_filter = 1
514 be slow to establish. In particular, these switches are known to 592to the file /etc/sysctl.conf
515 be incompatible with 82541/82547: 593
516 594 or,
517 Planex FXG-08TE 595
518 I-O Data ETG-SH8 596install the interfaces in separate broadcast domains (either in
519 597different switches or in a switch partitioned to VLANs).
520 To workaround this issue, the driver can be compiled with an override 598
521 of the PHY's master/slave setting. Forcing master or forcing slave 59982541/82547 can't link or are slow to link with some link partners
522 mode will improve time-to-link. 600-----------------------------------------------------------------
523 601There is a known compatibility issue with 82541/82547 and some
524 # make EXTRA_CFLAGS=-DE1000_MASTER_SLAVE=<n> 602low-end switches where the link will not be established, or will
525 603be slow to establish. In particular, these switches are known to
526 Where <n> is: 604be incompatible with 82541/82547:
527 605
528 0 = Hardware default 606 Planex FXG-08TE
529 1 = Master mode 607 I-O Data ETG-SH8
530 2 = Slave mode 608
531 3 = Auto master/slave 609To workaround this issue, the driver can be compiled with an override
532 610of the PHY's master/slave setting. Forcing master or forcing slave
533 Disable rx flow control with ethtool 611mode will improve time-to-link.
534 ------------------------------------ 612
535 613 # make CFLAGS_EXTRA=-DE1000_MASTER_SLAVE=<n>
536 In order to disable receive flow control using ethtool, you must turn 614
537 off auto-negotiation on the same command line. 615Where <n> is:
538 616
539 For example: 617 0 = Hardware default
540 618 1 = Master mode
541 ethtool -A eth? autoneg off rx off 619 2 = Slave mode
620 3 = Auto master/slave
621
622Disable rx flow control with ethtool
623------------------------------------
624In order to disable receive flow control using ethtool, you must turn
625off auto-negotiation on the same command line.
626
627For example:
628
629 ethtool -A eth? autoneg off rx off
630
631Unplugging network cable while ethtool -p is running
632----------------------------------------------------
633In kernel versions 2.5.50 and later (including 2.6 kernel), unplugging
634the network cable while ethtool -p is running will cause the system to
635become unresponsive to keyboard commands, except for control-alt-delete.
636Restarting the system appears to be the only remedy.
542 637
543 638
544Support 639Support
@@ -548,24 +643,10 @@ For general information, go to the Intel support website at:
548 643
549 http://support.intel.com 644 http://support.intel.com
550 645
551 or the Intel Wired Networking project hosted by Sourceforge at: 646or the Intel Wired Networking project hosted by Sourceforge at:
552 647
553 http://sourceforge.net/projects/e1000 648 http://sourceforge.net/projects/e1000
554 649
555If an issue is identified with the released source code on the supported 650If an issue is identified with the released source code on the supported
556kernel with a supported adapter, email the specific information related 651kernel with a supported adapter, email the specific information related
557to the issue to e1000-devel@lists.sourceforge.net 652to the issue to e1000-devel@lists.sf.net
558
559
560License
561=======
562
563This software program is released under the terms of a license agreement
564between you ('Licensee') and Intel. Do not use or load this software or any
565associated materials (collectively, the 'Software') until you have carefully
566read the full terms and conditions of the file COPYING located in this software
567package. By loading or using the Software, you agree to the terms of this
568Agreement. If you do not agree with the terms of this Agreement, do not
569install or use the Software.
570
571* Other names and brands may be claimed as the property of others.