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