1 2005-07-29 Andrew Lunn <andrew.lunn@ascom.ch>
3 * src/net/eth_drv.c (eth_drv_recv): Change of type to remove
6 2005-06-29 Andrew Lunn <andrew.lunn@ascom.ch>
8 * doc/driver_doc: Removed, it has been superseded by the sgml
11 2005-03-27 Andrew Lunn <andrew.lunn@ascom.ch>
13 * src/net/eth_drv.c (eth_drv_send): Fixed a compiler warning about
14 type of parameter passed to diag_printf.
16 2004-08-06 Jani Monoses <jani@iv.ro>
18 * src/stand_alone/eth_drv.c: Add parantheses to fix operator
19 precedence bugs even though they were harmless in practice.
21 2004-01-09 Jani Monoses <jani@iv.ro>
24 * src/lwip/eth_drv.c: Update to support more than one device and
25 be in sync with current lwip code.
27 2004-01-05 Gary Thomas <gary@mlbassoc.com>
29 * src/net/eth_drv.c: Names of FLASH config types changed.
31 2003-10-09 Gary Thomas <gary@mlbassoc.com>
33 * src/stand_alone/eth_drv.c: Obviate __local_enet_addr, since this
34 same information is already kept in the driver structure *__local_enet_sc
36 2003-08-19 Gary Thomas <gary@mlbassoc.com>
38 * src/stand_alone/eth_drv.c:
39 * src/net/eth_drv.c: Improve network speeds by reducing some
40 overhead and also the delay when sending a packet.
42 2003-07-23 Nick Garnett <nickg@balti.calivar.com>
46 Added an ifndef to fix warnings about NBPFILTER not being
49 2003-03-20 Mark Salter <msalter@redhat.com>
51 * cdl/eth_drivers.cdl (CYGHWR_NET_DRIVERS): Moved here from net/common.
53 2003-03-14 Nick Garnett <nickg@calivar.com>
55 * src/net/eth_drv.c: Added calls to cyg_net_show_mbufs() whenever
56 we see an out-of-mbufs problem.
58 2003-03-11 Mark Salter <msalter@redhat.com>
60 * src/stand_alone/eth_drv.c (eth_drv_stop): Uncomment, make non-static,
61 and use __local_enet_sc.
62 * include/eth_drv.h: Add extern decl for eth_drv_stop.
64 2003-02-24 Jonathan Larmour <jifl@eCosCentric.com>
66 * cdl/eth_drivers.cdl: Add doc link.
68 2003-02-07 Jonathan Larmour <jifl@eCosCentric.com>
70 * src/net/eth_drv.c: If driver can't set multi-cast, indicate it's
73 2002-08-14 Gary Thomas <gthomas@ecoscentric.com>
75 * cdl/eth_drivers.cdl: CYGNUM_IO_ETH_DRIVERS_SG_LIST_SIZE
76 needs to be global (used by all drivers, not just NET).
78 2002-08-13 Gary Thomas <gthomas@ecoscentric.com>
80 * src/net/eth_drv.c (eth_drv_send): Print a better message
81 if the scatter-gather list overflows.
83 * include/eth_drv.h (MAX_ETH_DRV_SG):
84 * cdl/eth_drivers.cdl: Add control over size of scatter-gather
85 data lists used to pass requests to physical layer. Previous
86 value was sometimes too small.
88 2002-07-31 Gary Thomas <gary@chez-thomas.org>
90 * src/net/eth_drv.c (eth_drv_start): Fix compile error when
91 built with old stack (no multicast) - introduced below.
93 2002-07-26 Gary Thomas <gary@chez-thomas.org>
94 2002-07-26 Ken Cox <jkc@redhat.com>
96 * src/net/eth_drv.c (eth_drv_start): Force multicast address
97 setup any time chip is reset/reconfigured.
99 2002-05-30 Jonathan Larmour <jlarmour@redhat.com>
101 * cdl/eth_drivers.cdl: Provide CYGINT_IO_ETH_INT_SUPPORT_REQUIRED
102 interface to indicate if interrupt support is required.
104 2002-05-28 Jonathan Larmour <jlarmour@redhat.com>
106 * src/lwip/lw.diff: Remove. Obsolete.
108 2002-05-13 Jesper Skov <jskov@redhat.com>
110 * cdl/eth_drivers.cdl: Moved the package's header files to
111 cyg/io/eth to clean up the include root directory.
113 2002-04-30 Jonathan Larmour <jlarmour@redhat.com>
115 * src/lwip/eth_drv.c: Update from Jani Monoses.
117 * cdl/eth_drivers.cdl: Move CYGINT_ISO_STRING_STRFUNCS requirement
118 into CYGPKG_IO_ETH_DRIVERS_NET where it belongs.
119 Move CYGINT_ISO_STRING_MEMFUNCS requirement into
120 CYGPKG_IO_ETH_DRIVERS_STAND_ALONE where it belongs.
122 2002-04-10 Gary Thomas <gthomas@redhat.com>
124 * src/net/eth_drv.c: Only pass mutlicast addresses down to drivers.
126 2002-04-05 Jani Monoses <jani@iv.ro>
128 * src/lwip/eth_drv.c: New file. Add eth driver support for lwip.
129 * src/lwip/lw.diff: New file. diff against lwip sources.
130 * src/lwip/README: New file.
131 * cdl/eth_drivers.cdl: Add lwip driver support.
133 2002-03-09 Gary Thomas <gthomas@redhat.com>
135 * src/net/eth_drv.c: Select 'multicast_all' if too many addresses
136 are required. Also, ignore any non-multicast addresses.
138 2002-02-22 Hugo Tyson <hmt@redhat.com>
140 * doc/ethdrv.sgml: New file. SGML-ized the existing driver-doc
141 file and brought the information up to date with deliver functions
144 2002-02-18 Gary Thomas <gthomas@redhat.com>
146 * cdl/eth_drivers.cdl:
147 Add new interface CYGINT_IO_ETH_MULTICAST which is used to
148 insure that drivers support multicast addresses if IPv6 is used.
150 2002-02-15 Gary Thomas <gthomas@redhat.com>
153 * include/eth_drv.h: Add framework for multicast address support.
155 2002-01-30 Hugo Tyson <hmt@redhat.com>
157 * cdl/eth_drivers.cdl (CYGPKG_IO_ETH_DRIVERS_WARN_FORCE_CONSOLE):
158 CDL options removed; control is now from RedBoot.
160 * src/net/eth_drv.c (START_CONSOLE): Changed to use RedBoot
161 fconfig items for deciding what console to use for special debug
162 messages such as "out of MBUFs" and the like.
164 * src/stand_alone/eth_drv.c (LOCK_APPLICATION_SCHEDULER): New
165 macro to do what it says, before passing packets to the app.
166 Used in two places: when passing a received packet, and when
167 passing a tx completion status back.
169 2002-01-28 Gary Thomas <gthomas@redhat.com>
171 * src/net/eth_drv.c (eth_drv_init): Changes for supporting new
172 FreeBSD based stack (minor API changes internal to that stack).
174 2001-12-12 Hugo Tyson <hmt@redhat.com>
176 * src/net/eth_drv.c (START_CONSOLE,END_CONSOLE): New macros akin
177 to those within RedBoot's internal net_io.c, for directing warning
178 messages to a serial line if required. All printf's wrapped in
181 * cdl/eth_drivers.cdl (CYGPKG_IO_ETH_DRIVERS_WARN_FORCE_CONSOLE):
182 and ..._NUMBER, new options to direct application warning output
183 somewhere useful (less harmful) for when you are debugging over
184 the net. Trying to print within a call which occurred from within
185 RedBoot itself is such a recursive situation as to be fatal.
186 Defaults are disabled, you must choose to set these up if
187 you want to debug your app safely over the network.
189 2001-12-12 Hugo Tyson <hmt@redhat.com>
191 * src/stand_alone/eth_drv.c (DIAG_DUMP_BUF_HDR,DIAG_DUMP_BUF_BDY):
192 Separate macros for debug dumps of ethernet header and body.
193 (eth_drv_write): Use them.
194 (eth_drv_copy_recv): Do not corrupt the static data pointer, so
195 that if we are erroneously called twice, at least what we pass is
196 a packet, rather than random memory; do not escape to SEGV. Guard
197 against NULL sg_list[].buf pointers, in case caller out of MBUFs.
198 (eth_drv_recv): Only dump packet if we actually have a buffer.
199 Only forward the packet if actually have a buffer.
201 2001-12-04 Richard Sandiford <rsandifo@redhat.com>
203 * src/stand_alone/eth_drv.c (eth_drv_read): Check that the return
204 buffer is big enough to hold the packet.
206 2001-11-28 Jonathan Larmour <jlarmour@redhat.com>
208 * doc/driver_doc: Mention preferences on how the ESA should be set.
210 2001-10-30 Jonathan Larmour <jlarmour@redhat.com>
212 * doc/driver_doc: Add description of poll, deliver and int_vector
215 2001-10-29 Hugo Tyson <hmt@redhat.com>
217 * src/net/eth_drv.c (eth_drv_tickle_devices): [CASE 106613] Rather
218 than calling can_send() in the device driver directly, call our
219 own eth_drv_send() function which will also dequeue a waiting
220 packet if we find we indeed can_send(). This helps to recover
221 from queue full situations within the if_ethersubr layer above.
222 This change is belt & braces with a similar call in the code which
223 drops a packet if the queue is full, in the main network code in
224 net/tcpip/.../src/sys/net/if_ethersubr.c (ether_output); this
225 change will recover the situation shortly even if the application
226 gives up trying to send because of ENOBUFS.
228 2001-10-18 Jonathan Larmour <jlarmour@redhat.com>
230 * cdl/eth_drivers.cdl: Rename CYGSEM_IO_ETH_DRIVERS_DEBUG to
231 CYGDBG_IO_ETH_DRIVERS_DEBUG and make it a common option.
232 Add separate CYGDBG_IO_ETH_DRIVERS_DEBUG_VERBOSITY config.
234 * include/eth_drv.h: Don't claim BSD if it isn't.
235 * include/eth_drv_stats.h: Ditto.
236 * include/netdev.h: Ditto.
237 * src/net/eth_drv.c: Ditto.
238 Also use above CDL options to conditionalize debugging (and the
240 * src/stand_alone/eth_drv.c: Use above renamed CDL options.
242 2001-10-11 Gary Thomas <gthomas@redhat.com>
244 * src/stand_alone/eth_drv.c (eth_drv_recv):
245 Defensive programming - insure valid packet size.
247 2001-10-05 Jonathan Larmour <jlarmour@redhat.com>
249 * src/net/eth_drv.c: Add default implementation of min().
251 2001-09-26 Jesper Skov <jskov@redhat.com>
253 * cdl/eth_drivers.cdl: Require the string functions.
255 * src/stand_alone/eth_drv.c: Fix warnings.
257 2001-09-25 Hugo Tyson <hmt@redhat.com>
259 * src/net/eth_drv.c (eth_drv_dsr): Assert that the "sc" given to
260 eth_drv_dsr() really is a sc from the valid list. This can save a
261 few hours debug time of a typo.
263 2001-09-13 Hugo Tyson <hmt@redhat.com>
265 * src/net/eth_drv.c (eth_drv_tickle_devices): New function to call
266 can_send() on all devices to allow them to unstick from a lost
267 interrupt or other wedge - it's up to the device to detect this,
268 of course, can_send() does not explicitly do this, it's just used
269 as an opportunity harmlessly to get control into the device so it
270 can examine its world.
272 2001-09-05 Gary Thomas <gthomas@redhat.com>
274 * src/stand_alone/eth_drv.c (eth_drv_init): Don't initialize
275 structures if ESA is zero. This is used by low level drivers
276 to indicate that a device is present, but can't currently be
277 used. The eCos driver handles this case as well.
279 2001-08-31 Hugo Tyson <hmt@redhat.com>
280 2001-08-31 Andrew Lunn <Andrew.Lunn@ascom.ch>
282 * src/net/eth_drv.c (eth_drv_tx_done): Guard against a NULL key
283 return - which can be caused by race conditions in the driver,
284 this is the neatest fixup. It's good defensive programming
285 anyway and ASCOM's tests indicate a benefit for CASE 106059.
287 2001-08-22 Gary Thomas <gthomas@redhat.com>
289 * src/stand_alone/eth_drv.c:
290 printf() is no longer a part of RedBoot. Thus all programs
291 must use diag_printf() and related functions instead.
293 * cdl/eth_drivers.cdl: Only enable warning/debug modes if RedBoot
294 is present as the debug environment (since there are implicit
295 dependencies on RedBoot functions).
297 2001-08-20 Jonathan Larmour <jlarmour@redhat.com>
299 * src/net/eth_drv.c (eth_drv_send): Move endif location to fix build
302 2001-08-17 Jonathan Larmour <jlarmour@redhat.com>
304 * cdl/eth_drivers.cdl (CYGSEM_IO_ETH_DRIVERS_DEBUG): Now booldata.
305 * src/stand_along/eth_drv.c (eth_drv_tx_done): Use different
306 value of CYGSEM_IO_ETH_DRIVERS_DEBUG for verbosity.
308 2001-08-14 Gary Thomas <gthomas@redhat.com>
310 * src/stand_alone/eth_drv.c (eth_drv_write):
312 (eth_drv_read): Better handling of stacking (layering) of drivers.
313 RedBoot (stand alone code) is designed to call into the eCos
314 stack and these changes make sure that this is done properly
315 nested/stacked. These changes also affect the behaviour positively
318 * src/net/eth_drv.c (eth_drv_send): Add locking of driver while
319 actual hardware routines are involved. Since the same driver
320 can be shared by both eCos and RedBoot, it is imperative that
321 additional locking (in the form of locking the scheduler) be
322 employed during this window to make sure that the hardware is
323 handled in complete, consistent steps. This helps with known
326 2001-08-09 Hugo Tyson <hmt@redhat.com>
328 * src/net/eth_drv.c (eth_drv_recv): Add a pair of
329 CYGARC_HAL_SAVE_GP()/CYGARC_HAL_RESTORE_GP() on entry and exit;
330 this function is intercalled between RedBoot and the application.
331 A minor re-org at the ending to accommodate these macros' nesting.
332 (eth_drv_tx_done): CYGARC_HAL_SAVE_GP()/CYGARC_HAL_RESTORE_GP().
334 * src/stand_alone/eth_drv.c (eth_drv_tx_done): Add a pair of
335 CYGARC_HAL_SAVE_GP()/CYGARC_HAL_RESTORE_GP() on entry and exit;
336 this function is intercalled between RedBoot and the application.
337 Also make printing message about "tx_done for other key" also
338 conditional on net_debug variable - it's commonplace.
339 (eth_drv_recv): CYGARC_HAL_SAVE_GP()/CYGARC_HAL_RESTORE_GP().
340 (eth_drv_copy_recv): CYGARC_HAL_SAVE_GP()/CYGARC_HAL_RESTORE_GP().
342 2001-07-03 Gary Thomas <gthomas@redhat.com>
344 * src/net/eth_drv.c (eth_drv_recv): Better cleanup when running
345 out of mbufs - could have lost mbufs before.
347 2001-06-21 David Woodhouse <dwmw2@redhat.com>
349 * src/stand_alone/eth_drv.c: Timeout when waiting for Ethernet
350 driver to finish sending packet.
351 * src/stand_alone/eth_drv.c: Don't delay if it's ready immediately.
353 2001-06-11 Gary Thomas <gthomas@redhat.com>
355 * cdl/eth_drivers.cdl: Make debug default for RedBoot.
357 2001-05-22 Jonathan Larmour <jlarmour@redhat.com>
359 * cdl/eth_drivers.cdl: New option: CYGSEM_IO_ETH_DRIVERS_PASS_PACKETS
360 which replaces the static define in...
361 * src/stand_alone_eth_drv.c: Rename ETH_DRV_PASS_PACKETS to
362 new CDL option above.
363 Also, ensure references to start_console/end_console are conditional
364 on CYGSEM_IO_ETH_DRIVERS_DEBUG.
366 2001-05-22 Hugo Tyson <hmt@redhat.com>
367 2001-05-22 Sanjay Bisen <Sanjay.Bisen@ascom.ch>
369 * src/net/eth_drv.c (eth_drv_recv): Patch from Sanjay at Ascom; it
370 leaked one mbuf if you run out of clusters. Fix is simply to
371 free m in addition to top.
373 * src/net/eth_drv.c (eth_drv_run_deliveries): Fixed a warning with
374 an int cast in HAL_CTRLC_CHECK().
376 2001-03-21 Gary Thomas <gthomas@redhat.com>
378 * cdl/eth_drivers.cdl: Fewer buffers needed in stand-alone mode.
380 2001-03-12 Hugo Tyson <hmt@redhat.com>
382 * cdl/eth_drivers.cdl: Permit defines for
383 CYGPKG_IO_ETH_DRIVERS_STAND_ALONE and CYGPKG_IO_ETH_DRIVERS_NET.
384 The generic i82559 driver needs to know, unfortunately; it calls
385 its own DSR directly to unblock after a lost interrupt. And you
386 mustn't call the DSR under RedBoot == ....STAND_ALONE.
388 2001-01-15 Jesper Skov <jskov@redhat.com>
390 * src/net/eth_drv.c (eth_drv_init): Do not use enaddr if NULL.
392 2001-01-07 Gary Thomas <gthomas@redhat.com>
394 * cdl/eth_drivers.cdl: Add interface 'CYGPKG_NET_DRIVER_FRAMEWORK'
395 to describe interdependencies between network stack and driver
398 2001-01-04 Gary Thomas <gthomas@redhat.com>
400 * src/net/eth_drv.c (eth_drv_tx_done): Update count of Tx packets.
402 2000-12-11 Gary Thomas <gthomas@redhat.com>
404 * src/net/eth_drv.c (eth_drv_run_deliveries): Support ^C when
405 using network based debug channel.
407 2000-12-02 Gary Thomas <gthomas@redhat.com>
409 * src/stand_alone/eth_drv.c (eth_drv_write): Debug: dump packet
410 before calling hardware routine for improved usability.
412 2000-10-30 Gary Thomas <gthomas@redhat.com>
414 * include/eth_drv.h: Fix compile error in stand-alone mode.
416 2000-10-17 Hugo Tyson <hmt@redhat.com>
417 2000-10-10 Andrew Lunn <Andrew.Lunn@ascom.ch>
418 * cdl/eth_drivers.cdl: Add configury to disable the warnings about
419 out of mbufs for receives.
421 * src/net/eth_drv.c: The changes required for the above.
423 2000-09-28 Hugo Tyson <hmt@redhat.com>
425 * src/net/eth_drv.c (eth_drv_run_deliveries): Remove race
426 condition; a chance to deliver could be delayed until "next time"
427 if the DSR snuck in just right.
429 2000-09-14 Hugo Tyson <hmt@redhat.com>
431 * cdl/eth_drivers.cdl: Add configury to control new features. All
432 is controlled globally by CYGPKG_IO_ETH_DRIVERS_SIMULATED_FAILURES
433 which is by default off, natch.
435 * src/net/eth_drv.c (simulate_fail): Implement simulated failures
436 in ethernet packet delivery - independent tx and rx dropping and
437 rx packet corruption. Also a "line break" drop-all feature.
439 2000-09-13 Gary Thomas <gthomas@redhat.com>
441 * src/stand_alone/eth_drv.c: Take out noisy dump when out of buffers.
443 2000-09-11 Hugo Tyson <hmt@cygnus.co.uk>
445 * src/stand_alone/eth_drv.c (eth_drv_write): If we timeout waiting
446 for the write to complete, do reset the vectors to point to the
447 application, rather just a bald "return".
449 2000-09-08 Hugo Tyson <hmt@cygnus.co.uk>
451 * include/netdev.h (NETDEVTAB_ENTRY): Work around feature of new
452 version of CYG_HAL_TABLE_ENTRY() whereby no spaces are allowed.
454 2000-09-07 Jonathan Larmour <jlarmour@redhat.com>
456 * include/netdev.h (cyg_netdevtab_entry_t): Correct syntax for
459 2000-09-04 Jonathan Larmour <jlarmour@redhat.com>
461 * include/netdev.h (cyg_netdevtab_entry_t): Apply CYG_HAL_TABLE_TYPE
463 2000-09-01 Hugo Tyson <hmt@cygnus.co.uk>
465 * src/stand_alone/eth_drv.c (eth_drv_dsr): New function, never
466 called but maybe referenced in stand_alone context, which lets
467 redboot work in the new world.
469 2000-09-01 Hugo Tyson <hmt@cygnus.co.uk>
471 * OVERVIEW: This is part of the change to the network stack to
472 greatly reduce latencies both of (other) DSRs and of thread
473 scheduling. All the work that the network stack *and* individual
474 ether drivers used to do in DSRs (including alarm callbacks and
475 data copies to/from the device memory) is moved into a "fast
476 network thread" instead. It calls a device's "deliver" function
477 to do the work that was previously in the DSR. This is a separate
478 thread so that it can be set higher priority than application
479 threads in order to minimize packet loss (depending on the
480 driver), if required (the application threads presumed to be
481 higher priority in turn than the network thread). A crucial
482 consequence of this is that we are no longer locking against DSRs,
483 so a plain mutex can be used rather than the global scheduler
484 lock, thus simplifying all the splfoo/splx() style functions.
486 These changes WILL BREAK individual device drivers until they are
487 updated AND the standalone logical ether driver in this component,
488 until it is updated also.
490 * include/eth_drv.h (ETH_DRV_SC): Add "deliver" entry to struct
491 eth_hwr_funs interface record; declare available DSR and flag for
492 "needs delivery" in SC status field.
494 * src/net/eth_drv.c (eth_drv_run_deliveries): New function,
495 performs callbacks to deliver funcs for all devs that want it.
496 (eth_drv_dsr): New function, sets flag in sc and calls up to net
497 stack to schedule the fast network thread.
498 (eth_drv_send): No need to lock scheduler here.
500 2000-08-29 Gary Thomas <gthomas@redhat.com>
502 * src/stand_alone/eth_drv.c: Use null buffer, (char *)0,
503 instead of wasting memory - low level drivers must be designed
506 2000-08-28 Gary Thomas <gthomas@redhat.com>
508 * src/stand_alone/eth_drv.c: Use new configuration parameters.
510 * cdl/eth_drivers.cdl: Add new configury to control number of
511 internal buffers used by this layer. Also exposed some of the
514 2000-08-25 Hugo Tyson <hmt@cygnus.co.uk>
516 * include/eth_drv_stats.h (ether_drv_stats): A little further
517 diddling; have a bool to say whether the dot3 info is filled in.
519 2000-08-24 Hugo Tyson <hmt@cygnus.co.uk>
521 * src/net/eth_drv.c (eth_drv_ioctl): Implement ioctl() calls.
522 sockio.h numbers SIOCGIFSTATS and SIOCGIFSTATSUD map to
523 ETH_DRV_GET_IF_STATS and ETH_DRV_GET_IF_STATS_UD respectively.
525 * include/eth_drv.h (ETH_DRV_GET_IF_STATS_UD): Add new device
526 interface ioctl() numbers, to get the struct below filled in.
527 Also pull in the definition include file (below).
529 * include/eth_drv_stats.h (ether_drv_stats): New file: Define
530 common structure for ether devices to return stats info to higher
531 up, via an ioctl() call. SNMP uses this.
533 2000-08-23 Gary Thomas <gthomas@redhat.com>
535 * src/stand_alone/eth_drv.c:
536 * include/eth_drv.h: Low level drivers now need to export a
537 function which returns the interrupt vector used by that interface.
538 This allows outside code to determine this in a portable fashion.
540 2000-08-16 Gary Thomas <gthomas@redhat.com>
542 * src/stand_alone/eth_drv.c:
543 * include/eth_drv.h: Clean up warnings (better protyping).
545 * src/stand_alone/eth_drv.c (eth_drv_write): Abandon sending
546 packet if device goes not ready for a long time (in some cases,
547 this is the only indication that the link is down).
549 2000-08-07 Gary Thomas <gthomas@redhat.com>
551 * src/net/eth_drv.c (eth_drv_recv): Add safety in case this gets
552 called [from stand-alone code] while interface is not up.
554 * src/stand_alone/eth_drv.c:
555 * include/eth_drv.h: Rework to more fully support mixed stand-alone
556 and system (eCos) stacks.
558 2000-08-03 Gary Thomas <gthomas@redhat.com>
560 * cdl/eth_drivers.cdl: Reparent within I/O, not NET. This allows
561 for stand-alone use as well as support for other stack implementations.
563 2000-07-28 Hugo Tyson <hmt@cygnus.co.uk>
565 * src/eth_drv.c (eth_drv_recv): Assert that the length we're asked
566 to deal with is at least an ether header, and also be defensive in
567 any case; discard small packets. [CASE 104206]
569 2000-07-26 Gary Thomas <gthomas@redhat.com>
572 * include/eth_drv.h: Change interfaces used by hardware layer to
573 be "soft" (pointers to functions). This will allow those drivers
574 to be shared by applications and the ROM/debug environment.
576 2000-07-15 Gary Thomas <gthomas@redhat.com>
578 * include/eth_drv.h: Add [initial] extensions to let this
579 layer work either in an eCos environment or stand-alone.
581 2000-07-11 Gary Thomas <gthomas@redhat.com>
583 * include/eth_drv.h: Add minimal PCMCIA support.
585 * src/eth_drv.c (eth_drv_netdev): New function - used to find the
586 ethernet device info for PCMCIA devices.
588 2000-06-23 Hugo Tyson <hmt@cygnus.co.uk>
590 * src/eth_drv.c (eth_drv_send): Do not consume an SG entry for
591 zero length data; tolerate overflow of the SG. Before this, pings
592 of 6000 bytes crashed the system!
594 You can now set net_debug to 2 to get quieter output; 1 gives the
595 whole packet dump as before.
597 * include/eth_drv.h (MAX_ETH_DRV_SG): Make this 16 so that an MTU
598 made of all mbufs will not overflow.
600 2000-03-28 Gary Thomas <gthomas@redhat.com>
602 * src/eth_drv.c (eth_drv_recv): Tolerate running out of MBUFs
603 instead of "panic"ing.
605 2000-03-08 Gary Thomas <gthomas@redhat.com>
607 * src/eth_drv.c: Add some function [block] comments.
608 (eth_drv_send): Use eCos scheduler lock instead of interrupt lock.
610 2000-03-08 Hugo Tyson <hmt@cygnus.co.uk>
612 * doc/driver_doc: Add some clarification about what's called when,
613 proofreading results, shorter lines so I can print it nicely.
615 2000-03-06 Gary Thomas <gthomas@redhat.com>
619 * doc/driver_doc: Remove generic "priv"ate references.
621 2000-02-29 Gary Thomas <gthomas@cygnus.co.uk>
624 * include/eth_drv.h: New expanded API for hardware drivers.
626 2000-02-18 Gary Thomas <gthomas@cygnus.co.uk>
628 * src/eth_drv.c (eth_drv_send): Disable interrupts while initiating
629 the buffer send - avoid a possible race.
631 2000-02-08 John Dallaway <jld@cygnus.co.uk>
633 * cdl/eth_drivers.cdl:
634 Reparent under CYGPKG_NET and tidy display strings.
636 //===========================================================================
637 //####ECOSGPLCOPYRIGHTBEGIN####
638 // -------------------------------------------
639 // This file is part of eCos, the Embedded Configurable Operating System.
640 // Copyright (C) 1998, 1999, 2000, 2001, 2002 Red Hat, Inc.
642 // eCos is free software; you can redistribute it and/or modify it under
643 // the terms of the GNU General Public License as published by the Free
644 // Software Foundation; either version 2 or (at your option) any later version.
646 // eCos is distributed in the hope that it will be useful, but WITHOUT ANY
647 // WARRANTY; without even the implied warranty of MERCHANTABILITY or
648 // FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
651 // You should have received a copy of the GNU General Public License along
652 // with eCos; if not, write to the Free Software Foundation, Inc.,
653 // 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA.
655 // As a special exception, if other files instantiate templates or use macros
656 // or inline functions from this file, or you compile this file and link it
657 // with other works to produce a work based on this file, this file does not
658 // by itself cause the resulting work to be covered by the GNU General Public
659 // License. However the source code for this file must still be made available
660 // in accordance with section (3) of the GNU General Public License.
662 // This exception does not invalidate any other reasons why a work based on
663 // this file might be covered by the GNU General Public License.
665 // Alternative licenses for eCos may be arranged by contacting Red Hat, Inc.
666 // at http://sources.redhat.com/ecos/ecos-license/
667 // -------------------------------------------
668 //####ECOSGPLCOPYRIGHTEND####
669 //===========================================================================