aboutsummaryrefslogtreecommitdiffstats
path: root/Documentation/DocBook/tulip-user.tmpl
diff options
context:
space:
mode:
Diffstat (limited to 'Documentation/DocBook/tulip-user.tmpl')
-rw-r--r--Documentation/DocBook/tulip-user.tmpl327
1 files changed, 0 insertions, 327 deletions
diff --git a/Documentation/DocBook/tulip-user.tmpl b/Documentation/DocBook/tulip-user.tmpl
deleted file mode 100644
index 6520d7a1b132..000000000000
--- a/Documentation/DocBook/tulip-user.tmpl
+++ /dev/null
@@ -1,327 +0,0 @@
1<?xml version="1.0" encoding="UTF-8"?>
2<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN"
3 "http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" []>
4
5<book id="TulipUserGuide">
6 <bookinfo>
7 <title>Tulip Driver User's Guide</title>
8
9 <authorgroup>
10 <author>
11 <firstname>Jeff</firstname>
12 <surname>Garzik</surname>
13 <affiliation>
14 <address>
15 <email>jgarzik@pobox.com</email>
16 </address>
17 </affiliation>
18 </author>
19 </authorgroup>
20
21 <copyright>
22 <year>2001</year>
23 <holder>Jeff Garzik</holder>
24 </copyright>
25
26 <legalnotice>
27 <para>
28 This documentation is free software; you can redistribute
29 it and/or modify it under the terms of the GNU General Public
30 License as published by the Free Software Foundation; either
31 version 2 of the License, or (at your option) any later
32 version.
33 </para>
34
35 <para>
36 This program is distributed in the hope that it will be
37 useful, but WITHOUT ANY WARRANTY; without even the implied
38 warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
39 See the GNU General Public License for more details.
40 </para>
41
42 <para>
43 You should have received a copy of the GNU General Public
44 License along with this program; if not, write to the Free
45 Software Foundation, Inc., 59 Temple Place, Suite 330, Boston,
46 MA 02111-1307 USA
47 </para>
48
49 <para>
50 For more details see the file COPYING in the source
51 distribution of Linux.
52 </para>
53 </legalnotice>
54 </bookinfo>
55
56 <toc></toc>
57
58 <chapter id="intro">
59 <title>Introduction</title>
60<para>
61The Tulip Ethernet Card Driver
62is maintained by Jeff Garzik (<email>jgarzik@pobox.com</email>).
63</para>
64
65<para>
66The Tulip driver was developed by Donald Becker and changed by
67Jeff Garzik, Takashi Manabe and a cast of thousands.
68</para>
69
70<para>
71For 2.4.x and later kernels, the Linux Tulip driver is available at
72<ulink url="http://sourceforge.net/projects/tulip/">http://sourceforge.net/projects/tulip/</ulink>
73</para>
74
75<para>
76 This driver is for the Digital "Tulip" Ethernet adapter interface.
77 It should work with most DEC 21*4*-based chips/ethercards, as well as
78 with work-alike chips from Lite-On (PNIC) and Macronix (MXIC) and ASIX.
79</para>
80
81<para>
82 The original author may be reached as becker@scyld.com, or C/O
83 Scyld Computing Corporation,
84 410 Severn Ave., Suite 210,
85 Annapolis MD 21403
86</para>
87
88<para>
89 Additional information on Donald Becker's tulip.c
90 is available at <ulink url="http://www.scyld.com/network/tulip.html">http://www.scyld.com/network/tulip.html</ulink>
91</para>
92
93 </chapter>
94
95 <chapter id="drvr-compat">
96 <title>Driver Compatibility</title>
97
98<para>
99This device driver is designed for the DECchip "Tulip", Digital's
100single-chip ethernet controllers for PCI (now owned by Intel).
101Supported members of the family
102are the 21040, 21041, 21140, 21140A, 21142, and 21143. Similar work-alike
103chips from Lite-On, Macronics, ASIX, Compex and other listed below are also
104supported.
105</para>
106
107<para>
108These chips are used on at least 140 unique PCI board designs. The great
109number of chips and board designs supported is the reason for the
110driver size and complexity. Almost of the increasing complexity is in the
111board configuration and media selection code. There is very little
112increasing in the operational critical path length.
113</para>
114 </chapter>
115
116 <chapter id="board-settings">
117 <title>Board-specific Settings</title>
118
119<para>
120PCI bus devices are configured by the system at boot time, so no jumpers
121need to be set on the board. The system BIOS preferably should assign the
122PCI INTA signal to an otherwise unused system IRQ line.
123</para>
124
125<para>
126Some boards have EEPROMs tables with default media entry. The factory default
127is usually "autoselect". This should only be overridden when using
128transceiver connections without link beat e.g. 10base2 or AUI, or (rarely!)
129for forcing full-duplex when used with old link partners that do not do
130autonegotiation.
131</para>
132 </chapter>
133
134 <chapter id="driver-operation">
135 <title>Driver Operation</title>
136
137<sect1><title>Ring buffers</title>
138
139<para>
140The Tulip can use either ring buffers or lists of Tx and Rx descriptors.
141This driver uses statically allocated rings of Rx and Tx descriptors, set at
142compile time by RX/TX_RING_SIZE. This version of the driver allocates skbuffs
143for the Rx ring buffers at open() time and passes the skb->data field to the
144Tulip as receive data buffers. When an incoming frame is less than
145RX_COPYBREAK bytes long, a fresh skbuff is allocated and the frame is
146copied to the new skbuff. When the incoming frame is larger, the skbuff is
147passed directly up the protocol stack and replaced by a newly allocated
148skbuff.
149</para>
150
151<para>
152The RX_COPYBREAK value is chosen to trade-off the memory wasted by
153using a full-sized skbuff for small frames vs. the copying costs of larger
154frames. For small frames the copying cost is negligible (esp. considering
155that we are pre-loading the cache with immediately useful header
156information). For large frames the copying cost is non-trivial, and the
157larger copy might flush the cache of useful data. A subtle aspect of this
158choice is that the Tulip only receives into longword aligned buffers, thus
159the IP header at offset 14 isn't longword aligned for further processing.
160Copied frames are put into the new skbuff at an offset of "+2", thus copying
161has the beneficial effect of aligning the IP header and preloading the
162cache.
163</para>
164
165</sect1>
166
167<sect1><title>Synchronization</title>
168<para>
169The driver runs as two independent, single-threaded flows of control. One
170is the send-packet routine, which enforces single-threaded use by the
171dev->tbusy flag. The other thread is the interrupt handler, which is single
172threaded by the hardware and other software.
173</para>
174
175<para>
176The send packet thread has partial control over the Tx ring and 'dev->tbusy'
177flag. It sets the tbusy flag whenever it's queuing a Tx packet. If the next
178queue slot is empty, it clears the tbusy flag when finished otherwise it sets
179the 'tp->tx_full' flag.
180</para>
181
182<para>
183The interrupt handler has exclusive control over the Rx ring and records stats
184from the Tx ring. (The Tx-done interrupt can't be selectively turned off, so
185we can't avoid the interrupt overhead by having the Tx routine reap the Tx
186stats.) After reaping the stats, it marks the queue entry as empty by setting
187the 'base' to zero. Iff the 'tp->tx_full' flag is set, it clears both the
188tx_full and tbusy flags.
189</para>
190
191</sect1>
192
193 </chapter>
194
195 <chapter id="errata">
196 <title>Errata</title>
197
198<para>
199The old DEC databooks were light on details.
200The 21040 databook claims that CSR13, CSR14, and CSR15 should each be the last
201register of the set CSR12-15 written. Hmmm, now how is that possible?
202</para>
203
204<para>
205The DEC SROM format is very badly designed not precisely defined, leading to
206part of the media selection junkheap below. Some boards do not have EEPROM
207media tables and need to be patched up. Worse, other boards use the DEC
208design kit media table when it isn't correct for their board.
209</para>
210
211<para>
212We cannot use MII interrupts because there is no defined GPIO pin to attach
213them. The MII transceiver status is polled using an kernel timer.
214</para>
215 </chapter>
216
217 <chapter id="changelog">
218 <title>Driver Change History</title>
219
220 <sect1><title>Version 0.9.14 (February 20, 2001)</title>
221 <itemizedlist>
222 <listitem><para>Fix PNIC problems (Manfred Spraul)</para></listitem>
223 <listitem><para>Add new PCI id for Accton comet</para></listitem>
224 <listitem><para>Support Davicom tulips</para></listitem>
225 <listitem><para>Fix oops in eeprom parsing</para></listitem>
226 <listitem><para>Enable workarounds for early PCI chipsets</para></listitem>
227 <listitem><para>IA64, hppa csr0 support</para></listitem>
228 <listitem><para>Support media types 5, 6</para></listitem>
229 <listitem><para>Interpret a bit more of the 21142 SROM extended media type 3</para></listitem>
230 <listitem><para>Add missing delay in eeprom reading</para></listitem>
231 </itemizedlist>
232 </sect1>
233
234 <sect1><title>Version 0.9.11 (November 3, 2000)</title>
235 <itemizedlist>
236 <listitem><para>Eliminate extra bus accesses when sharing interrupts (prumpf)</para></listitem>
237 <listitem><para>Barrier following ownership descriptor bit flip (prumpf)</para></listitem>
238 <listitem><para>Endianness fixes for >14 addresses in setup frames (prumpf)</para></listitem>
239 <listitem><para>Report link beat to kernel/userspace via netif_carrier_*. (kuznet)</para></listitem>
240 <listitem><para>Better spinlocking in set_rx_mode.</para></listitem>
241 <listitem><para>Fix I/O resource request failure error messages (DaveM catch)</para></listitem>
242 <listitem><para>Handle DMA allocation failure.</para></listitem>
243 </itemizedlist>
244 </sect1>
245
246 <sect1><title>Version 0.9.10 (September 6, 2000)</title>
247 <itemizedlist>
248 <listitem><para>Simple interrupt mitigation (via jamal)</para></listitem>
249 <listitem><para>More PCI ids</para></listitem>
250 </itemizedlist>
251 </sect1>
252
253 <sect1><title>Version 0.9.9 (August 11, 2000)</title>
254 <itemizedlist>
255 <listitem><para>More PCI ids</para></listitem>
256 </itemizedlist>
257 </sect1>
258
259 <sect1><title>Version 0.9.8 (July 13, 2000)</title>
260 <itemizedlist>
261 <listitem><para>Correct signed/unsigned comparison for dummy frame index</para></listitem>
262 <listitem><para>Remove outdated references to struct enet_statistics</para></listitem>
263 </itemizedlist>
264 </sect1>
265
266 <sect1><title>Version 0.9.7 (June 17, 2000)</title>
267 <itemizedlist>
268 <listitem><para>Timer cleanups (Andrew Morton)</para></listitem>
269 <listitem><para>Alpha compile fix (somebody?)</para></listitem>
270 </itemizedlist>
271 </sect1>
272
273 <sect1><title>Version 0.9.6 (May 31, 2000)</title>
274 <itemizedlist>
275 <listitem><para>Revert 21143-related support flag patch</para></listitem>
276 <listitem><para>Add HPPA/media-table debugging printk</para></listitem>
277 </itemizedlist>
278 </sect1>
279
280 <sect1><title>Version 0.9.5 (May 30, 2000)</title>
281 <itemizedlist>
282 <listitem><para>HPPA support (willy@puffingroup)</para></listitem>
283 <listitem><para>CSR6 bits and tulip.h cleanup (Chris Smith)</para></listitem>
284 <listitem><para>Improve debugging messages a bit</para></listitem>
285 <listitem><para>Add delay after CSR13 write in t21142_start_nway</para></listitem>
286 <listitem><para>Remove unused ETHER_STATS code</para></listitem>
287 <listitem><para>Convert 'extern inline' to 'static inline' in tulip.h (Chris Smith)</para></listitem>
288 <listitem><para>Update DS21143 support flags in tulip_chip_info[]</para></listitem>
289 <listitem><para>Use spin_lock_irq, not _irqsave/restore, in tulip_start_xmit()</para></listitem>
290 <listitem><para>Add locking to set_rx_mode()</para></listitem>
291 <listitem><para>Fix race with chip setting DescOwned bit (Hal Murray)</para></listitem>
292 <listitem><para>Request 100% of PIO and MMIO resource space assigned to card</para></listitem>
293 <listitem><para>Remove error message from pci_enable_device failure</para></listitem>
294 </itemizedlist>
295 </sect1>
296
297 <sect1><title>Version 0.9.4.3 (April 14, 2000)</title>
298 <itemizedlist>
299 <listitem><para>mod_timer fix (Hal Murray)</para></listitem>
300 <listitem><para>PNIC2 resuscitation (Chris Smith)</para></listitem>
301 </itemizedlist>
302 </sect1>
303
304 <sect1><title>Version 0.9.4.2 (March 21, 2000)</title>
305 <itemizedlist>
306 <listitem><para>Fix 21041 CSR7, CSR13/14/15 handling</para></listitem>
307 <listitem><para>Merge some PCI ids from tulip 0.91x</para></listitem>
308 <listitem><para>Merge some HAS_xxx flags and flag settings from tulip 0.91x</para></listitem>
309 <listitem><para>asm/io.h fix (submitted by many) and cleanup</para></listitem>
310 <listitem><para>s/HAS_NWAY143/HAS_NWAY/</para></listitem>
311 <listitem><para>Cleanup 21041 mode reporting</para></listitem>
312 <listitem><para>Small code cleanups</para></listitem>
313 </itemizedlist>
314 </sect1>
315
316 <sect1><title>Version 0.9.4.1 (March 18, 2000)</title>
317 <itemizedlist>
318 <listitem><para>Finish PCI DMA conversion (davem)</para></listitem>
319 <listitem><para>Do not netif_start_queue() at end of tulip_tx_timeout() (kuznet)</para></listitem>
320 <listitem><para>PCI DMA fix (kuznet)</para></listitem>
321 <listitem><para>eeprom.c code cleanup</para></listitem>
322 <listitem><para>Remove Xircom Tulip crud</para></listitem>
323 </itemizedlist>
324 </sect1>
325 </chapter>
326
327</book>
generated by cgit v1.2.2 (git 2.25.0) at 2024-12-29 18:43:24 -0500