tjh.dev
Linux kernel mirror (for testing)
git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel
os
linux
1.. SPDX-License-Identifier: GPL-2.0+
2
3Linux* Base Driver for Intel(R) Ethernet Network Connection
4===========================================================
5
6Intel Gigabit Linux driver.
7Copyright(c) 1999 - 2013 Intel Corporation.
8
9Contents
10========
11
12- Identifying Your Adapter
13- Command Line Parameters
14- Speed and Duplex Configuration
15- Additional Configurations
16- Support
17
18Identifying Your Adapter
19========================
20
21For more information on how to identify your adapter, go to the Adapter &
22Driver ID Guide at:
23
24 http://support.intel.com/support/go/network/adapter/idguide.htm
25
26For the latest Intel network drivers for Linux, refer to the following
27website. In the search field, enter your adapter name or type, or use the
28networking link on the left to search for your adapter:
29
30 http://support.intel.com/support/go/network/adapter/home.htm
31
32Command Line Parameters
33=======================
34
35The default value for each parameter is generally the recommended setting,
36unless otherwise noted.
37
38NOTES:
39 For more information about the AutoNeg, Duplex, and Speed
40 parameters, see the "Speed and Duplex Configuration" section in
41 this document.
42
43 For more information about the InterruptThrottleRate,
44 RxIntDelay, TxIntDelay, RxAbsIntDelay, and TxAbsIntDelay
45 parameters, see the application note at:
46 http://www.intel.com/design/network/applnots/ap450.htm
47
48AutoNeg
49-------
50
51(Supported only on adapters with copper connections)
52
53:Valid Range: 0x01-0x0F, 0x20-0x2F
54:Default Value: 0x2F
55
56This parameter is a bit-mask that specifies the speed and duplex settings
57advertised by the adapter. When this parameter is used, the Speed and
58Duplex parameters must not be specified.
59
60NOTE:
61 Refer to the Speed and Duplex section of this readme for more
62 information on the AutoNeg parameter.
63
64Duplex
65------
66
67(Supported only on adapters with copper connections)
68
69:Valid Range: 0-2 (0=auto-negotiate, 1=half, 2=full)
70:Default Value: 0
71
72This defines the direction in which data is allowed to flow. Can be
73either one or two-directional. If both Duplex and the link partner are
74set to auto-negotiate, the board auto-detects the correct duplex. If the
75link partner is forced (either full or half), Duplex defaults to half-
76duplex.
77
78FlowControl
79-----------
80
81:Valid Range: 0-3 (0=none, 1=Rx only, 2=Tx only, 3=Rx&Tx)
82:Default Value: Reads flow control settings from the EEPROM
83
84This parameter controls the automatic generation(Tx) and response(Rx)
85to Ethernet PAUSE frames.
86
87InterruptThrottleRate
88---------------------
89
90(not supported on Intel(R) 82542, 82543 or 82544-based adapters)
91
92:Valid Range:
93 0,1,3,4,100-100000 (0=off, 1=dynamic, 3=dynamic conservative,
94 4=simplified balancing)
95:Default Value: 3
96
97The driver can limit the amount of interrupts per second that the adapter
98will generate for incoming packets. It does this by writing a value to the
99adapter that is based on the maximum amount of interrupts that the adapter
100will generate per second.
101
102Setting InterruptThrottleRate to a value greater or equal to 100
103will program the adapter to send out a maximum of that many interrupts
104per second, even if more packets have come in. This reduces interrupt
105load on the system and can lower CPU utilization under heavy load,
106but will increase latency as packets are not processed as quickly.
107
108The default behaviour of the driver previously assumed a static
109InterruptThrottleRate value of 8000, providing a good fallback value for
110all traffic types,but lacking in small packet performance and latency.
111The hardware can handle many more small packets per second however, and
112for this reason an adaptive interrupt moderation algorithm was implemented.
113
114Since 7.3.x, the driver has two adaptive modes (setting 1 or 3) in which
115it dynamically adjusts the InterruptThrottleRate value based on the traffic
116that it receives. After determining the type of incoming traffic in the last
117timeframe, it will adjust the InterruptThrottleRate to an appropriate value
118for that traffic.
119
120The algorithm classifies the incoming traffic every interval into
121classes. Once the class is determined, the InterruptThrottleRate value is
122adjusted to suit that traffic type the best. There are three classes defined:
123"Bulk traffic", for large amounts of packets of normal size; "Low latency",
124for small amounts of traffic and/or a significant percentage of small
125packets; and "Lowest latency", for almost completely small packets or
126minimal traffic.
127
128In dynamic conservative mode, the InterruptThrottleRate value is set to 4000
129for traffic that falls in class "Bulk traffic". If traffic falls in the "Low
130latency" or "Lowest latency" class, the InterruptThrottleRate is increased
131stepwise to 20000. This default mode is suitable for most applications.
132
133For situations where low latency is vital such as cluster or
134grid computing, the algorithm can reduce latency even more when
135InterruptThrottleRate is set to mode 1. In this mode, which operates
136the same as mode 3, the InterruptThrottleRate will be increased stepwise to
13770000 for traffic in class "Lowest latency".
138
139In simplified mode the interrupt rate is based on the ratio of TX and
140RX traffic. If the bytes per second rate is approximately equal, the
141interrupt rate will drop as low as 2000 interrupts per second. If the
142traffic is mostly transmit or mostly receive, the interrupt rate could
143be as high as 8000.
144
145Setting InterruptThrottleRate to 0 turns off any interrupt moderation
146and may improve small packet latency, but is generally not suitable
147for bulk throughput traffic.
148
149NOTE:
150 InterruptThrottleRate takes precedence over the TxAbsIntDelay and
151 RxAbsIntDelay parameters. In other words, minimizing the receive
152 and/or transmit absolute delays does not force the controller to
153 generate more interrupts than what the Interrupt Throttle Rate
154 allows.
155
156CAUTION:
157 If you are using the Intel(R) PRO/1000 CT Network Connection
158 (controller 82547), setting InterruptThrottleRate to a value
159 greater than 75,000, may hang (stop transmitting) adapters
160 under certain network conditions. If this occurs a NETDEV
161 WATCHDOG message is logged in the system event log. In
162 addition, the controller is automatically reset, restoring
163 the network connection. To eliminate the potential for the
164 hang, ensure that InterruptThrottleRate is set no greater
165 than 75,000 and is not set to 0.
166
167NOTE:
168 When e1000 is loaded with default settings and multiple adapters
169 are in use simultaneously, the CPU utilization may increase non-
170 linearly. In order to limit the CPU utilization without impacting
171 the overall throughput, we recommend that you load the driver as
172 follows::
173
174 modprobe e1000 InterruptThrottleRate=3000,3000,3000
175
176 This sets the InterruptThrottleRate to 3000 interrupts/sec for
177 the first, second, and third instances of the driver. The range
178 of 2000 to 3000 interrupts per second works on a majority of
179 systems and is a good starting point, but the optimal value will
180 be platform-specific. If CPU utilization is not a concern, use
181 RX_POLLING (NAPI) and default driver settings.
182
183RxDescriptors
184-------------
185
186:Valid Range:
187 - 48-256 for 82542 and 82543-based adapters
188 - 48-4096 for all other supported adapters
189:Default Value: 256
190
191This value specifies the number of receive buffer descriptors allocated
192by the driver. Increasing this value allows the driver to buffer more
193incoming packets, at the expense of increased system memory utilization.
194
195Each descriptor is 16 bytes. A receive buffer is also allocated for each
196descriptor and can be either 2048, 4096, 8192, or 16384 bytes, depending
197on the MTU setting. The maximum MTU size is 16110.
198
199NOTE:
200 MTU designates the frame size. It only needs to be set for Jumbo
201 Frames. Depending on the available system resources, the request
202 for a higher number of receive descriptors may be denied. In this
203 case, use a lower number.
204
205RxIntDelay
206----------
207
208:Valid Range: 0-65535 (0=off)
209:Default Value: 0
210
211This value delays the generation of receive interrupts in units of 1.024
212microseconds. Receive interrupt reduction can improve CPU efficiency if
213properly tuned for specific network traffic. Increasing this value adds
214extra latency to frame reception and can end up decreasing the throughput
215of TCP traffic. If the system is reporting dropped receives, this value
216may be set too high, causing the driver to run out of available receive
217descriptors.
218
219CAUTION:
220 When setting RxIntDelay to a value other than 0, adapters may
221 hang (stop transmitting) under certain network conditions. If
222 this occurs a NETDEV WATCHDOG message is logged in the system
223 event log. In addition, the controller is automatically reset,
224 restoring the network connection. To eliminate the potential
225 for the hang ensure that RxIntDelay is set to 0.
226
227RxAbsIntDelay
228-------------
229
230(This parameter is supported only on 82540, 82545 and later adapters.)
231
232:Valid Range: 0-65535 (0=off)
233:Default Value: 128
234
235This value, in units of 1.024 microseconds, limits the delay in which a
236receive interrupt is generated. Useful only if RxIntDelay is non-zero,
237this value ensures that an interrupt is generated after the initial
238packet is received within the set amount of time. Proper tuning,
239along with RxIntDelay, may improve traffic throughput in specific network
240conditions.
241
242Speed
243-----
244
245(This parameter is supported only on adapters with copper connections.)
246
247:Valid Settings: 0, 10, 100, 1000
248:Default Value: 0 (auto-negotiate at all supported speeds)
249
250Speed forces the line speed to the specified value in megabits per second
251(Mbps). If this parameter is not specified or is set to 0 and the link
252partner is set to auto-negotiate, the board will auto-detect the correct
253speed. Duplex should also be set when Speed is set to either 10 or 100.
254
255TxDescriptors
256-------------
257
258:Valid Range:
259 - 48-256 for 82542 and 82543-based adapters
260 - 48-4096 for all other supported adapters
261:Default Value: 256
262
263This value is the number of transmit descriptors allocated by the driver.
264Increasing this value allows the driver to queue more transmits. Each
265descriptor is 16 bytes.
266
267NOTE:
268 Depending on the available system resources, the request for a
269 higher number of transmit descriptors may be denied. In this case,
270 use a lower number.
271
272TxIntDelay
273----------
274
275:Valid Range: 0-65535 (0=off)
276:Default Value: 8
277
278This value delays the generation of transmit interrupts in units of
2791.024 microseconds. Transmit interrupt reduction can improve CPU
280efficiency if properly tuned for specific network traffic. If the
281system is reporting dropped transmits, this value may be set too high
282causing the driver to run out of available transmit descriptors.
283
284TxAbsIntDelay
285-------------
286
287(This parameter is supported only on 82540, 82545 and later adapters.)
288
289:Valid Range: 0-65535 (0=off)
290:Default Value: 32
291
292This value, in units of 1.024 microseconds, limits the delay in which a
293transmit interrupt is generated. Useful only if TxIntDelay is non-zero,
294this value ensures that an interrupt is generated after the initial
295packet is sent on the wire within the set amount of time. Proper tuning,
296along with TxIntDelay, may improve traffic throughput in specific
297network conditions.
298
299XsumRX
300------
301
302(This parameter is NOT supported on the 82542-based adapter.)
303
304:Valid Range: 0-1
305:Default Value: 1
306
307A value of '1' indicates that the driver should enable IP checksum
308offload for received packets (both UDP and TCP) to the adapter hardware.
309
310Copybreak
311---------
312
313:Valid Range: 0-xxxxxxx (0=off)
314:Default Value: 256
315:Usage: modprobe e1000.ko copybreak=128
316
317Driver copies all packets below or equaling this size to a fresh RX
318buffer before handing it up the stack.
319
320This parameter is different than other parameters, in that it is a
321single (not 1,1,1 etc.) parameter applied to all driver instances and
322it is also available during runtime at
323/sys/module/e1000/parameters/copybreak
324
325SmartPowerDownEnable
326--------------------
327
328:Valid Range: 0-1
329:Default Value: 0 (disabled)
330
331Allows PHY to turn off in lower power states. The user can turn off
332this parameter in supported chipsets.
333
334Speed and Duplex Configuration
335==============================
336
337Three keywords are used to control the speed and duplex configuration.
338These keywords are Speed, Duplex, and AutoNeg.
339
340If the board uses a fiber interface, these keywords are ignored, and the
341fiber interface board only links at 1000 Mbps full-duplex.
342
343For copper-based boards, the keywords interact as follows:
344
345- The default operation is auto-negotiate. The board advertises all
346 supported speed and duplex combinations, and it links at the highest
347 common speed and duplex mode IF the link partner is set to auto-negotiate.
348
349- If Speed = 1000, limited auto-negotiation is enabled and only 1000 Mbps
350 is advertised (The 1000BaseT spec requires auto-negotiation.)
351
352- If Speed = 10 or 100, then both Speed and Duplex should be set. Auto-
353 negotiation is disabled, and the AutoNeg parameter is ignored. Partner
354 SHOULD also be forced.
355
356The AutoNeg parameter is used when more control is required over the
357auto-negotiation process. It should be used when you wish to control which
358speed and duplex combinations are advertised during the auto-negotiation
359process.
360
361The parameter may be specified as either a decimal or hexadecimal value as
362determined by the bitmap below.
363
364============== ====== ====== ======= ======= ====== ====== ======= ======
365Bit position 7 6 5 4 3 2 1 0
366Decimal Value 128 64 32 16 8 4 2 1
367Hex value 80 40 20 10 8 4 2 1
368Speed (Mbps) N/A N/A 1000 N/A 100 100 10 10
369Duplex Full Full Half Full Half
370============== ====== ====== ======= ======= ====== ====== ======= ======
371
372Some examples of using AutoNeg::
373
374 modprobe e1000 AutoNeg=0x01 (Restricts autonegotiation to 10 Half)
375 modprobe e1000 AutoNeg=1 (Same as above)
376 modprobe e1000 AutoNeg=0x02 (Restricts autonegotiation to 10 Full)
377 modprobe e1000 AutoNeg=0x03 (Restricts autonegotiation to 10 Half or 10 Full)
378 modprobe e1000 AutoNeg=0x04 (Restricts autonegotiation to 100 Half)
379 modprobe e1000 AutoNeg=0x05 (Restricts autonegotiation to 10 Half or 100
380 Half)
381 modprobe e1000 AutoNeg=0x020 (Restricts autonegotiation to 1000 Full)
382 modprobe e1000 AutoNeg=32 (Same as above)
383
384Note that when this parameter is used, Speed and Duplex must not be specified.
385
386If the link partner is forced to a specific speed and duplex, then this
387parameter should not be used. Instead, use the Speed and Duplex parameters
388previously mentioned to force the adapter to the same speed and duplex.
389
390Additional Configurations
391=========================
392
393Jumbo Frames
394------------
395
396 Jumbo Frames support is enabled by changing the MTU to a value larger than
397 the default of 1500. Use the ifconfig command to increase the MTU size.
398 For example::
399
400 ifconfig eth<x> mtu 9000 up
401
402 This setting is not saved across reboots. It can be made permanent if
403 you add::
404
405 MTU=9000
406
407 to the file /etc/sysconfig/network-scripts/ifcfg-eth<x>. This example
408 applies to the Red Hat distributions; other distributions may store this
409 setting in a different location.
410
411Notes:
412 Degradation in throughput performance may be observed in some Jumbo frames
413 environments. If this is observed, increasing the application's socket buffer
414 size and/or increasing the /proc/sys/net/ipv4/tcp_*mem entry values may help.
415 See the specific application manual and /usr/src/linux*/Documentation/
416 networking/ip-sysctl.txt for more details.
417
418 - The maximum MTU setting for Jumbo Frames is 16110. This value coincides
419 with the maximum Jumbo Frames size of 16128.
420
421 - Using Jumbo frames at 10 or 100 Mbps is not supported and may result in
422 poor performance or loss of link.
423
424 - Adapters based on the Intel(R) 82542 and 82573V/E controller do not
425 support Jumbo Frames. These correspond to the following product names::
426
427 Intel(R) PRO/1000 Gigabit Server Adapter
428 Intel(R) PRO/1000 PM Network Connection
429
430ethtool
431-------
432
433 The driver utilizes the ethtool interface for driver configuration and
434 diagnostics, as well as displaying statistical information. The ethtool
435 version 1.6 or later is required for this functionality.
436
437 The latest release of ethtool can be found from
438 https://www.kernel.org/pub/software/network/ethtool/
439
440Enabling Wake on LAN* (WoL)
441---------------------------
442
443 WoL is configured through the ethtool* utility.
444
445 WoL will be enabled on the system during the next shut down or reboot.
446 For this driver version, in order to enable WoL, the e1000 driver must be
447 loaded when shutting down or rebooting the system.
448
449Support
450=======
451
452For general information, go to the Intel support website at:
453
454 http://support.intel.com
455
456or the Intel Wired Networking project hosted by Sourceforge at:
457
458 http://sourceforge.net/projects/e1000
459
460If an issue is identified with the released source code on the supported
461kernel with a supported adapter, email the specific information related
462to the issue to e1000-devel@lists.sf.net