xref: /dragonfly/share/man/man4/rl.4 (revision cfd1aba3)
1.\" Copyright (c) 1997, 1998
2.\"	Bill Paul <wpaul@ctr.columbia.edu>. All rights reserved.
3.\"
4.\" Redistribution and use in source and binary forms, with or without
5.\" modification, are permitted provided that the following conditions
6.\" are met:
7.\" 1. Redistributions of source code must retain the above copyright
8.\"    notice, this list of conditions and the following disclaimer.
9.\" 2. Redistributions in binary form must reproduce the above copyright
10.\"    notice, this list of conditions and the following disclaimer in the
11.\"    documentation and/or other materials provided with the distribution.
12.\" 3. All advertising materials mentioning features or use of this software
13.\"    must display the following acknowledgement:
14.\"	This product includes software developed by Bill Paul.
15.\" 4. Neither the name of the author nor the names of any co-contributors
16.\"    may be used to endorse or promote products derived from this software
17.\"   without specific prior written permission.
18.\"
19.\" THIS SOFTWARE IS PROVIDED BY Bill Paul AND CONTRIBUTORS ``AS IS'' AND
20.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
21.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
22.\" ARE DISCLAIMED.  IN NO EVENT SHALL Bill Paul OR THE VOICES IN HIS HEAD
23.\" BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
24.\" CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
25.\" SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
26.\" INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
27.\" CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
28.\" ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
29.\" THE POSSIBILITY OF SUCH DAMAGE.
30.\"
31.\" $FreeBSD: src/share/man/man4/rl.4,v 1.15.2.8 2003/02/17 21:20:39 trhodes Exp $
32.\"
33.Dd November 4, 1998
34.Dt RL 4
35.Os
36.Sh NAME
37.Nm rl
38.Nd RealTek 8129/8139 fast ethernet device driver
39.Sh SYNOPSIS
40.Cd "device miibus"
41.Cd "device rl"
42.Sh DESCRIPTION
43The
44.Nm
45driver provides support for PCI ethernet adapters and embedded
46controllers based on the RealTek 8129 and 8139 fast ethernet controller
47chips.
48This includes the Allied Telesyn AT2550, Farallon NetLINE 10/100 PCI,
49Genius GF100TXR,
50NDC Communications NE100TX-E, OvisLink LEF-8129TX, OvisLink LEF-8139TX,
51Netronix Inc. EA-1210 NetEther 10/100, KTX-9130TX 10/100 Fast Ethernet,
52Encore ENL832-TX 10/100 M PCI, Longshine LCS-8038TX-R, the
53SMC EZ Card 10/100 PCI 1211-TX, and various other cheap adapters.
54It also supports the Accton EN1207D which has a
55chip labeled MPX5030 (or MPX5038) which appears to be a RealTek workalike.
56.Pp
57The RealTek controllers use bus master DMA but do not use a
58descriptor-based data transfer mechanism.
59The receiver uses a
60single fixed size ring buffer from which packets must be copied
61into mbufs.
62For transmission, there are only four outbound packet
63address registers which require all outgoing packets to be stored
64as contiguous buffers.
65Furthermore, outbound packet buffers must
66be longword aligned or else transmission will fail.
67.Pp
68The 8129 differs from the 8139 in that the 8139 has an internal
69PHY which is controlled through special direct access registers
70whereas the 8129 uses an external PHY via an MII bus.
71The 8139
72supports both 10 and 100Mbps speeds in either full or half duplex.
73The 8129 can support the same speeds and modes given an appropriate
74PHY chip.
75.Pp
76The
77.Nm
78driver supports the following media types:
79.Bl -tag -width xxxxxxxxxxxxxxxxxxxx
80.It autoselect
81Enable autoselection of the media type and options.
82This is only
83supported if the PHY chip attached to the RealTek controller
84supports NWAY autonegotiation.
85The user can manually override
86the autoselected mode by adding media options to the
87.Pa /etc/rc.conf
88file.
89.It 10baseT/UTP
90Set 10Mbps operation.
91The
92.Ar mediaopt
93option can also be used to select either
94.Ar full-duplex
95or
96.Ar half-duplex
97modes.
98.It 100baseTX
99Set 100Mbps (fast ethernet) operation.
100The
101.Ar mediaopt
102option can also be used to select either
103.Ar full-duplex
104or
105.Ar half-duplex
106modes.
107.El
108.Pp
109The
110.Nm
111driver supports the following media options:
112.Bl -tag -width xxxxxxxxxxxxxxxxxxxx
113.It full-duplex
114Force full duplex operation
115.It half-duplex
116Force half duplex operation.
117.El
118.Pp
119Note that the 100baseTX media type is only available if supported
120by the adapter.
121.Pp
122For more information on configuring this device, see
123.Xr ifconfig 8 .
124The
125.Nm
126driver supports
127.Xr polling 4 .
128.Sh DIAGNOSTICS
129.Bl -diag
130.It "rl%d: couldn't map memory"
131A fatal initialization error has occurred.
132.It "rl%d: couldn't map interrupt"
133A fatal initialization error has occurred.
134.It "rl%d: watchdog timeout"
135The device has stopped responding to the network, or there is a problem with
136the network connection (cable).
137.It "rl%d: no memory for rx list"
138The driver failed to allocate an mbuf for the receiver ring.
139.It "rl%d: no memory for tx list"
140The driver failed to allocate an mbuf for the transmitter ring when
141allocating a pad buffer or collapsing an mbuf chain into a cluster.
142.It "rl%d: chip is in D3 power state -- setting to D0"
143This message applies only to adapters which support power
144management.
145Some operating systems place the controller in low power
146mode when shutting down, and some PCI BIOSes fail to bring the chip
147out of this state before configuring it.
148The controller loses all of
149its PCI configuration in the D3 state, so if the BIOS does not set
150it back to full power mode in time, it won't be able to configure it
151correctly.
152The driver tries to detect this condition and bring
153the adapter back to the D0 (full power) state, but this may not be
154enough to return the driver to a fully operational condition.
155If
156you see this message at boot time and the driver fails to attach
157the device as a network interface, you will have to perform second
158warm boot to have the device properly configured.
159.Pp
160Note that this condition only occurs when warm booting from another
161operating system.
162If you power down your system prior to booting
163.Dx ,
164the card should be configured correctly.
165.El
166.Sh SEE ALSO
167.Xr arp 4 ,
168.Xr ifmedia 4 ,
169.Xr miibus 4 ,
170.Xr netintro 4 ,
171.Xr ng_ether 4 ,
172.Xr polling 4 ,
173.Xr ifconfig 8
174.Rs
175.%B The RealTek 8129 and 8139 datasheets
176.%O ftp.realtek.com.tw:/lancard/data sheet
177.Re
178.Sh HISTORY
179The
180.Nm
181device driver first appeared in
182.Fx 3.0 .
183.Sh AUTHORS
184The
185.Nm
186driver was written by
187.An Bill Paul Aq Mt wpaul@ctr.columbia.edu .
188.Sh BUGS
189Since outbound packets must be longword aligned, the transmit
190routine has to copy an unaligned packet into an mbuf cluster buffer
191before transmission.
192The driver abuses the fact that the cluster buffer
193pool is allocated at system startup time in a contiguous region starting
194at a page boundary.
195Since cluster buffers are 2048 bytes, they are
196longword aligned by definition.
197The driver probably should not be
198depending on this characteristic.
199.Pp
200The RealTek data sheets are of especially poor quality,
201and there is a lot of information missing
202particularly concerning the receiver operation.
203One particularly
204important fact that the data sheets fail to mention relates to the
205way in which the chip fills in the receive buffer.
206When an interrupt
207is posted to signal that a frame has been received, it is possible that
208another frame might be in the process of being copied into the receive
209buffer while the driver is busy handling the first one.
210If the driver
211manages to finish processing the first frame before the chip is done
212DMAing the rest of the next frame, the driver may attempt to process
213the next frame in the buffer before the chip has had a chance to finish
214DMAing all of it.
215.Pp
216The driver can check for an incomplete frame by inspecting the frame
217length in the header preceding the actual packet data: an incomplete
218frame will have the magic length of 0xFFF0.
219When the driver encounters
220this value, it knows that it has finished processing all currently
221available packets.
222Neither this magic value nor its significance are
223documented anywhere in the RealTek data sheets.
224