1 .\" $NetBSD: le.4,v 1.22 2004/10/04 19:12:52 rumble Exp $
4 .\" Copyright (c) 1992, 1993
5 .\" The Regents of the University of California. All rights reserved.
7 .\" This software was developed by the Computer Systems Engineering group
8 .\" at Lawrence Berkeley Laboratory under DARPA contract BG 91-66 and
9 .\" contributed to Berkeley.
11 .\" Redistribution and use in source and binary forms, with or without
12 .\" modification, are permitted provided that the following conditions
14 .\" 1. Redistributions of source code must retain the above copyright
15 .\" notice, this list of conditions and the following disclaimer.
16 .\" 2. Redistributions in binary form must reproduce the above copyright
17 .\" notice, this list of conditions and the following disclaimer in the
18 .\" documentation and/or other materials provided with the distribution.
19 .\" 3. Neither the name of the University nor the names of its contributors
20 .\" may be used to endorse or promote products derived from this software
21 .\" without specific prior written permission.
23 .\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
24 .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
25 .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
26 .\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
27 .\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
28 .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
29 .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
30 .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
31 .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
32 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
35 .\" from: Header: le.4,v 1.2 92/10/13 05:31:33 leres Exp
36 .\" from: @(#)le.4 8.1 (Berkeley) 6/9/93
44 .Nd "AMD Am7900 LANCE and Am79C9xx PCnet Ethernet interface driver"
46 To compile this driver into the kernel,
47 place the following line in your
48 kernel configuration file:
49 .Bd -ragged -offset indent
53 Alternatively, to load the driver as a
54 module at boot time, place the following line in
56 .Bd -literal -offset indent
61 .\" .Pa /boot/device.hints :
62 .\" .Cd hint.le.0.at="isa"
63 .\" .Cd hint.le.0.port="0x280"
64 .\" .Cd hint.le.0.irq="10"
65 .\" .Cd hint.le.0.drq="0"
69 driver provides support for Ethernet adapters based on the
71 Am7990 and Am79C90 (CMOS, pin-compatible)
72 Local Area Network Controller for Ethernet
78 driver also supports PCnet adapters based on the
80 family of chips, which are single-chip implementations of a
82 chip and a DMA engine.
83 This includes a superset of the
85 bus Ethernet chip sets supported by the
90 driver treats all of these
92 bus Ethernet chip sets as an
93 .Tn AMD Am79C970 PCnet-PCI
94 and does not support the additional features like the MII bus and burst mode of
95 .Tn AMD Am79C971 PCnet-FAST
96 and greater chip sets.
99 driver should be preferred for the latter.
103 driver supports reception and transmission of extended frames for
105 Selective reception of multicast Ethernet frames is provided by a 64-bit mask;
106 multicast destination addresses are hashed to a bit entry using the Ethernet
112 .\" bus Ethernet cards supported by the
116 .\" .Bl -bullet -compact
120 .\" .Tn Novell NE2100
122 .\" .Tn Digital DEPCA
127 .\" bus Ethernet cards supported by the
131 .\" .Bl -bullet -compact
138 .\" bus Ethernet cards supported by the
142 .\" .Bl -bullet -compact
144 .\" .Tn SKNET Personal MC2
151 bus Ethernet chip sets supported by the
157 .Tn AMD Am53C974/Am79C970/Am79C974 PCnet-PCI
159 .Tn AMD Am79C970A PCnet-PCI II
161 .Tn AMD Am79C971 PCnet-FAST
163 .Tn AMD Am79C972 PCnet-FAST+
165 .Tn AMD Am79C973/Am79C975 PCnet-FAST III
167 .Tn AMD Am79C976 PCnet-PRO
169 .Tn AMD Am79C978 PCnet-Home
174 driver supports the following media types with these chip sets:
176 .Bl -tag -width xxxxxxxxxxxxxxxxxxxx
178 Enable autoselection of the media type.
182 Select AUI/BNC media.
185 The following media option is supported with these media types:
187 .Bl -tag -width xxxxxxxxxxxxxxxxxxxx
189 Select full duplex operation.
196 driver does not support selecting 100Mbps (Fast Ethernet) media types.
200 driver supports the on-board
205 The following media types are available with these:
207 .Bl -tag -width xxxxxxxxxxxxxxxxxxxx
209 Enable autoselection of the media type.
216 When using autoselection, a default media type is selected for use by
217 examining all ports for carrier.
218 The first media type with which a carrier is detected will be selected.
219 Additionally, if carrier is dropped on a port, the driver will switch
220 between the possible ports until one with carrier is found.
224 .\" driver also supports
228 .\" .Bl -bullet -compact
232 .\" and Buffered Ethernet
233 .\" (Sun P/N 501-1860)
238 .\" and Buffered Ethernet
239 .\" (Sun P/N 501-2015)
242 For further information on configuring media types and options, see
247 More packets came in from the Ethernet than there was space in the
251 .It "le%d: receive buffer error"
254 ran out of buffer space, packet dropped.
255 .It "le%d: lost carrier"
256 The Ethernet carrier disappeared during an attempt to transmit.
259 will finish transmitting the current packet,
260 but will not automatically retry transmission if there is a collision.
261 .It "le%d: excessive collisions, tdr %d"
262 The Ethernet was extremely busy or jammed,
263 outbound packets were dropped after 16 attempts to retransmit.
266 is the abbreviation of
267 .Qq Time Domain Reflectometry .
268 The optionally reported TDR value is an internal counter of the interval
269 between the start of a transmission and the occurrence of a collision.
270 This value can be used to determine the distance from the Ethernet tap to
271 the point on the Ethernet cable that is shorted or open (unterminated).
272 .It "le%d: dropping chained buffer"
273 A packet did not fit into a single receive buffer and was dropped.
276 driver allocates buffers large enough to receive maximum sized Ethernet
277 packets, this means some other station on the LAN transmitted a packet
278 larger than allowed by the Ethernet standard.
279 .It "le%d: transmit buffer error"
282 ran out of buffer space before finishing the transmission of a packet.
283 If this error occurs, the driver software has a bug.
284 .It "le%d: underflow"
287 ran out of buffer space before finishing the transmission of a packet.
288 If this error occurs, the driver software has a bug.
289 .It "le%d: controller failed to initialize"
290 Driver failed to start the
292 This is potentially a hardware failure.
293 .It "le%d: memory error"
294 RAM failed to respond within the timeout when the
296 wanted to read or write it.
297 This is potentially a hardware failure.
298 .It "le%d: receiver disabled"
301 was turned off due to an error.
302 .It "le%d: transmitter disabled"
303 The transmitter of the
305 was turned off due to an error.
317 driver was ported from
319 and first appeared in
323 version in turn was derived from the
325 driver which first appeared in
331 .An "Marius Strobl" Aq marius@FreeBSD.org .
333 .\" The Am7990 Revision C chips have a bug which causes garbage to be inserted
334 .\" in front of the received packet occasionally.
335 .\" The work-around is to ignore packets with an invalid destination address
336 .\" (garbage will usually not match), by double-checking the destination
337 .\" address of every packet in the driver.
338 .\" This work-around can be enabled with the
339 .\" .Dv LANCE_REVC_BUG
343 .\" .Dv LANCE_REVC_BUG
346 .\" driver executes one or two calls to an inline Ethernet address comparison
347 .\" function for every received packet.
350 .\" it is exactly eight instructions of 16 bits each.
351 .\" There is one comparison for each unicast packet, and two comparisons for
352 .\" each broadcast packet.
354 .\" In summary, the cost of the LANCE_REVC_BUG option is:
355 .\" .Bl -enum -compact
357 .\" loss of multicast support, and
361 .\" instructions per received packet, sometimes sixteen, depending on both the
362 .\" processor, and the type of packet.
365 .\" All sun3 systems are presumed to have this bad revision of the Am7990,
366 .\" until proven otherwise.
367 .\" Alas, the only way to prove what revision of the chip is in a particular
368 .\" system is inspection of the date code on the chip package,
369 .\" to compare against a list of what chip revisions were fabricated between
372 .\" Alas, the Am7990 chip is so old that
376 .\" the production information about it; pending a search elsewhere, we do not
377 .\" know how to identify the revision C chip from the date codes.