d/freebsd-dev
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
218fe3f1b0
freebsd-dev/sys/dev/ed/if_ed_pccard.c

697 lines
21 KiB
C
Raw Normal View History

Start each of the license/copyright comments with /*-, minor shuffle of lines
2005-01-06 01:43:34 +00:00
/*-
Break out the bus front end code into separate files. Note that these haven't been turned on nor has the old code been removed from if_ed.c. The next commit will address that.
1999-10-15 01:28:04 +00:00
* Copyright (c) 1995, David Greenman
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice unmodified, this list of conditions, and the following
* disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* $FreeBSD$
*/
Add an option ED_NO_MIIBUS, which causes the `ed' driver to be built without support for miibus PHYs. Most ed cards don't need miibus support, so it's useful to be able to avoid the bloat of all the mii devices for small fixed-purpose kernels.
2001-09-29 22:32:03 +00:00
#include "opt_ed.h"
Break out the bus front end code into separate files. Note that these haven't been turned on nor has the old code been removed from if_ed.c. The next commit will address that.
1999-10-15 01:28:04 +00:00
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/socket.h>
Add pccard attachment. This won't work with memory mapped ed devices yet, but that should be resolved shortly. Non memory mapped ed devices should work, but I cannot test this since my only ed card is memory mapped. Submitted by: Matt Dodd <mdodd@freebsd.org>
1999-10-25 02:55:14 +00:00
#include <sys/kernel.h>
MFPAO: Add support for AX88190, equipped in MELCO LPC3-TX.
2000-07-23 15:15:43 +00:00
#include <sys/conf.h>
#include <sys/uio.h>
Break out the bus front end code into separate files. Note that these haven't been turned on nor has the old code been removed from if_ed.c. The next commit will address that.
1999-10-15 01:28:04 +00:00
#include <sys/module.h>
Add pccard attachment. This won't work with memory mapped ed devices yet, but that should be resolved shortly. Non memory mapped ed devices should work, but I cannot test this since my only ed card is memory mapped. Submitted by: Matt Dodd <mdodd@freebsd.org>
1999-10-25 02:55:14 +00:00
#include <sys/bus.h>
#include <machine/bus.h>
Allocate memory resource to access the attribute memory of a card. Reviewed by: imp
2000-09-25 06:01:42 +00:00
#include <sys/rman.h>
#include <machine/resource.h>
Add pccard attachment. This won't work with memory mapped ed devices yet, but that should be resolved shortly. Non memory mapped ed devices should work, but I cannot test this since my only ed card is memory mapped. Submitted by: Matt Dodd <mdodd@freebsd.org>
1999-10-25 02:55:14 +00:00
#include <net/ethernet.h>
#include <net/if.h>
#include <net/if_arp.h>
#include <net/if_mib.h>
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
#include <net/if_media.h>
Add pccard attachment. This won't work with memory mapped ed devices yet, but that should be resolved shortly. Non memory mapped ed devices should work, but I cannot test this since my only ed card is memory mapped. Submitted by: Matt Dodd <mdodd@freebsd.org>
1999-10-25 02:55:14 +00:00
MFPAO: Add support for AX88190, equipped in MELCO LPC3-TX.
2000-07-23 15:15:43 +00:00
#include <dev/ed/if_edreg.h>
Break out the bus front end code into separate files. Note that these haven't been turned on nor has the old code been removed from if_ed.c. The next commit will address that.
1999-10-15 01:28:04 +00:00
#include <dev/ed/if_edvar.h>
Propagate MAC address in CIS tupple to dev/ed/ driver. Many ed-based Ethernet PC-cards can't get correct MAC address without this patch. Submitted by: Takanori Watanabe <takawata@shidahara1.planet.sci.kobe-u.ac.jp> Reviewed by: Warner Losh <imp@village.org>
2000-01-21 03:08:46 +00:00
#include <dev/pccard/pccardvar.h>
Only attach to network functions.
2005-01-20 19:37:22 +00:00
#include <dev/pccard/pccard_cis.h>
Add an option ED_NO_MIIBUS, which causes the `ed' driver to be built without support for miibus PHYs. Most ed cards don't need miibus support, so it's useful to be able to avoid the bloat of all the mii devices for small fixed-purpose kernels.
2001-09-29 22:32:03 +00:00
#ifndef ED_NO_MIIBUS
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
#include <dev/mii/mii.h>
#include <dev/mii/miivar.h>
Add an option ED_NO_MIIBUS, which causes the `ed' driver to be built without support for miibus PHYs. Most ed cards don't need miibus support, so it's useful to be able to avoid the bloat of all the mii devices for small fixed-purpose kernels.
2001-09-29 22:32:03 +00:00
#endif
Update the ed driver to probe and attach under a NEWCARD kernel (I was using a cardbus based system with pccbb providing the pcic interface). Something isn't quite right.. when the driver allocates and activates its resources, the IO space that was requested reads as all zeros (versus the original 0xff's as it normally is when there is no device responding). Also, deactivate the resources before releasing them. OLDCARD doesn't seem to care but NEWCARD/CARDBUS get rather unhappy if you release a resource that hasn't been deactivated yet. Make pcic_p.c only compile with oldcard kernels.
2000-11-25 03:36:09 +00:00
Allocate memory resource to access the attribute memory of a card. Reviewed by: imp
2000-09-25 06:01:42 +00:00
#include "card_if.h"
Add an option ED_NO_MIIBUS, which causes the `ed' driver to be built without support for miibus PHYs. Most ed cards don't need miibus support, so it's useful to be able to avoid the bloat of all the mii devices for small fixed-purpose kernels.
2001-09-29 22:32:03 +00:00
#ifndef ED_NO_MIIBUS
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
/* "device miibus" required. See GENERIC if you get errors here. */
#include "miibus_if.h"
Fix disordering of pccarddevs.h noticed by bde. Also remove a few redundant includes and fix some of the include disordering. Submitted by: bde
2004-05-27 03:49:45 +00:00
#endif
#include "pccarddevs.h"
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
Fix disordering of pccarddevs.h noticed by bde. Also remove a few redundant includes and fix some of the include disordering. Submitted by: bde
2004-05-27 03:49:45 +00:00
#ifndef ED_NO_MIIBUS
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
MODULE_DEPEND(ed, miibus, 1, 1, 1);
Add an option ED_NO_MIIBUS, which causes the `ed' driver to be built without support for miibus PHYs. Most ed cards don't need miibus support, so it's useful to be able to avoid the bloat of all the mii devices for small fixed-purpose kernels.
2001-09-29 22:32:03 +00:00
#endif
- Express hard dependencies on bus (pci, isa, pccard) and network layer (ether). - Don't abuse module names to facilitate ifconfig module loading; such abuse isn't really needed. (And if we do need type information associated with a module then we should make it explicit and not use hacks.)
2003-04-15 06:37:30 +00:00
MODULE_DEPEND(ed, ether, 1, 1, 1);
Break out the bus front end code into separate files. Note that these haven't been turned on nor has the old code been removed from if_ed.c. The next commit will address that.
1999-10-15 01:28:04 +00:00
Improve the mining of MAC address from the card: o Add a fallback location for the MAC address. Most of the early ne2000 PC Cards were built from the same parts, so most of them have the same address in the CIS to grab the MAC from. Use this address as our fallback if we don't find anything better. o Add printf, in bootverbose, noting the MAC addresses that we find along the way. # Better sanity checking of the MAC address is needed. Will have to # investigate using/creating a centralized function to do this as a number # of other PC Card drivers each have their own ad-hoc tests.
2005-02-12 16:44:15 +00:00
/*
* PC Cards should be using a network specific FUNCE in the CIS to
* communicate their MAC address to the driver. However, there are a
* large number of NE-2000ish PC Cards that don't do this. Nearly all
* of them store the MAC address at a fixed offset into attribute
* memory, without any reference at all appearing in the CIS. And
* nearly all of those store it at the same location.
Add note about what ED_DEFAULT_MAC_OFFSET applies to
2005-09-08 17:16:23 +00:00
*
* This applies only to the older, NE-2000 compatbile cards. The newer
* cards based on the AX88x90 or DL100XX chipsets have a specific place
* to look for MAC information.
Improve the mining of MAC address from the card: o Add a fallback location for the MAC address. Most of the early ne2000 PC Cards were built from the same parts, so most of them have the same address in the CIS to grab the MAC from. Use this address as our fallback if we don't find anything better. o Add printf, in bootverbose, noting the MAC addresses that we find along the way. # Better sanity checking of the MAC address is needed. Will have to # investigate using/creating a centralized function to do this as a number # of other PC Card drivers each have their own ad-hoc tests.
2005-02-12 16:44:15 +00:00
*/
#define ED_DEFAULT_MAC_OFFSET 0xff0
Add a bunch of entries from NetBSD's if_ne_pcmcia.c v1.67. We still need to verify that all the entries in pccard.conf are present in this table.
2001-01-21 08:20:23 +00:00
static const struct ed_product {
struct pccard_product prod;
int flags;
Fix the support for the ax88[17]90 to print the right type, rather than the generic ne-2000 string. This should have no effect on the actual support of the parts, just reporting what the part was. Also, rename a few functins and symbols to reflect a more generic part support that grew out of the early specific support.
2005-02-17 18:15:13 +00:00
#define NE2000DVF_DL100XX 0x0001 /* chip is D-Link DL10019/22 */
#define NE2000DVF_AX88X90 0x0002 /* chip is ASIX AX88[17]90 */
Really old PCMCIA cards stored the MAC address in the attribute memory at some offset. Unlike newer cards, the MAC address wasn't part of the CIS as a specific FUNCE. These older cards were having their MAC address show up as 0:2:4:6:8:a because that's what's in the ROM locations that would be there in a real ne2000. This patch allows one to specify the offset for the MAC address for these cards. Specify one for the IBM Ethernet II card, as it is one that has this problem. One shouldn't specify this unless the MAC address really isn't in the CIS at all. Side note: The novell probe likely shouldn't read the MAC address, and that should be moved to the bus specific attach routine(s), maybe as a convenience function in if_ed_novell.c. My IBM Ethernet II (aka Info Mover) now has a believable MAC address.
2005-02-12 00:45:22 +00:00
#define NE2000DVF_ENADDR 0x0004 /* Get MAC from attr mem */
Add support for the DLink DMF650TX Ethernet+56k modem PC Card. This gets the ethernet part of the card working, while putting appropriate hooks in place for the modem code. Other ed based lan/modem combo cards should be easy to add. Please send me info on any you'd like to see support added. Note: The 650 isn't a strictly conforming multi-function card, so special support is needed. :-(
2005-07-15 01:54:59 +00:00
#define NE2000DVF_ANYFUNC 0x0008 /* Allow any function type */
#define NE2000DVF_MODEM 0x0010 /* Has a modem/serial */
Really old PCMCIA cards stored the MAC address in the attribute memory at some offset. Unlike newer cards, the MAC address wasn't part of the CIS as a specific FUNCE. These older cards were having their MAC address show up as 0:2:4:6:8:a because that's what's in the ROM locations that would be there in a real ne2000. This patch allows one to specify the offset for the MAC address for these cards. Specify one for the IBM Ethernet II card, as it is one that has this problem. One shouldn't specify this unless the MAC address really isn't in the CIS at all. Side note: The novell probe likely shouldn't read the MAC address, and that should be moved to the bus specific attach routine(s), maybe as a convenience function in if_ed_novell.c. My IBM Ethernet II (aka Info Mover) now has a believable MAC address.
2005-02-12 00:45:22 +00:00
int enoff;
Add a bunch of entries from NetBSD's if_ne_pcmcia.c v1.67. We still need to verify that all the entries in pccard.conf are present in this table.
2001-01-21 08:20:23 +00:00
} ed_pccard_products[] = {
Eliminate unused argument in PCMCIA_CARD macro. Provide a backwards compatible way to have the extra macro by defining PCCARD_API_LEVEL 5 before including pccarddevs for driver writers that want/need to have the same driver on 5 and 6 with pccard attachments. Approved by: re (dwhite)
2005-06-24 14:36:54 +00:00
{ PCMCIA_CARD(ACCTON, EN2212), 0},
{ PCMCIA_CARD(ACCTON, EN2216), 0},
{ PCMCIA_CARD(ALLIEDTELESIS, LA_PCM), 0},
{ PCMCIA_CARD(AMBICOM, AMB8002T), 0},
{ PCMCIA_CARD(BILLIONTON, LNT10TN), 0},
{ PCMCIA_CARD(BILLIONTON, CFLT10N), 0},
Add a boatload of new device ids, gleaned from the pcmcia-cs-3.2.8 distribution. Add the appropriate devices to the man pages. Obtained from: pcmcia-cs Approved by: re (scottl)
2005-07-10 02:00:35 +00:00
{ PCMCIA_CARD(BROMAX, IPORT), 0},
{ PCMCIA_CARD(BROMAX, IPORT2), 0},
{ PCMCIA_CARD(BUFFALO, LPC2_CLT), 0},
Eliminate unused argument in PCMCIA_CARD macro. Provide a backwards compatible way to have the extra macro by defining PCCARD_API_LEVEL 5 before including pccarddevs for driver writers that want/need to have the same driver on 5 and 6 with pccard attachments. Approved by: re (dwhite)
2005-06-24 14:36:54 +00:00
{ PCMCIA_CARD(BUFFALO, LPC3_CLT), 0},
{ PCMCIA_CARD(BUFFALO, LPC3_CLX), NE2000DVF_AX88X90},
Add a boatload of new device ids, gleaned from the pcmcia-cs-3.2.8 distribution. Add the appropriate devices to the man pages. Obtained from: pcmcia-cs Approved by: re (scottl)
2005-07-10 02:00:35 +00:00
{ PCMCIA_CARD(BUFFALO, LPC4_TX), NE2000DVF_AX88X90},
Eliminate unused argument in PCMCIA_CARD macro. Provide a backwards compatible way to have the extra macro by defining PCCARD_API_LEVEL 5 before including pccarddevs for driver writers that want/need to have the same driver on 5 and 6 with pccard attachments. Approved by: re (dwhite)
2005-06-24 14:36:54 +00:00
{ PCMCIA_CARD(BUFFALO, LPC_CF_CLT), 0},
{ PCMCIA_CARD(CNET, NE2000), 0},
Add a boatload of new device ids, gleaned from the pcmcia-cs-3.2.8 distribution. Add the appropriate devices to the man pages. Obtained from: pcmcia-cs Approved by: re (scottl)
2005-07-10 02:00:35 +00:00
{ PCMCIA_CARD(COMPEX, AX88190), NE2000DVF_AX88X90},
{ PCMCIA_CARD(COMPEX, LANMODEM), 0},
Eliminate unused argument in PCMCIA_CARD macro. Provide a backwards compatible way to have the extra macro by defining PCCARD_API_LEVEL 5 before including pccarddevs for driver writers that want/need to have the same driver on 5 and 6 with pccard attachments. Approved by: re (dwhite)
2005-06-24 14:36:54 +00:00
{ PCMCIA_CARD(COMPEX, LINKPORT_ENET_B), 0},
{ PCMCIA_CARD(COREGA, ETHER_II_PCC_T), 0},
{ PCMCIA_CARD(COREGA, ETHER_II_PCC_TD), 0},
{ PCMCIA_CARD(COREGA, ETHER_PCC_T), 0},
{ PCMCIA_CARD(COREGA, ETHER_PCC_TD), 0},
{ PCMCIA_CARD(COREGA, FAST_ETHER_PCC_TX), NE2000DVF_DL100XX },
{ PCMCIA_CARD(COREGA, FETHER_PCC_TXD), NE2000DVF_AX88X90 },
{ PCMCIA_CARD(COREGA, FETHER_PCC_TXF), NE2000DVF_DL100XX },
{ PCMCIA_CARD(DAYNA, COMMUNICARD_E_1), 0},
{ PCMCIA_CARD(DAYNA, COMMUNICARD_E_2), 0},
{ PCMCIA_CARD(DLINK, DE650), 0},
{ PCMCIA_CARD(DLINK, DE660), 0 },
{ PCMCIA_CARD(DLINK, DE660PLUS), 0},
{ PCMCIA_CARD(DYNALINK, L10C), 0},
{ PCMCIA_CARD(EDIMAX, EP4000A), 0},
{ PCMCIA_CARD(EPSON, EEN10B), 0},
{ PCMCIA_CARD(EXP, THINLANCOMBO), 0},
Add a boatload of new device ids, gleaned from the pcmcia-cs-3.2.8 distribution. Add the appropriate devices to the man pages. Obtained from: pcmcia-cs Approved by: re (scottl)
2005-07-10 02:00:35 +00:00
{ PCMCIA_CARD(GREY_CELL, TDK3000), 0},
Add support for the DLink DMF650TX Ethernet+56k modem PC Card. This gets the ethernet part of the card working, while putting appropriate hooks in place for the modem code. Other ed based lan/modem combo cards should be easy to add. Please send me info on any you'd like to see support added. Note: The 650 isn't a strictly conforming multi-function card, so special support is needed. :-(
2005-07-15 01:54:59 +00:00
{ PCMCIA_CARD(GREY_CELL, DMF650TX),
It turns out that many of the ambicom cards have their CIS resources listed in different orders. Since it is easy to identify the Modem resources vs the Ethernet resources by looking at the size, use that rather than hard coded rids. For such parts, go ahead and guess which rid we should use based on the size. This guess appears reliable for the two example cards that I have with different CIS info.
2005-09-06 22:44:26 +00:00
NE2000DVF_ANYFUNC | NE2000DVF_DL100XX | NE2000DVF_MODEM},
Eliminate unused argument in PCMCIA_CARD macro. Provide a backwards compatible way to have the extra macro by defining PCCARD_API_LEVEL 5 before including pccarddevs for driver writers that want/need to have the same driver on 5 and 6 with pccard attachments. Approved by: re (dwhite)
2005-06-24 14:36:54 +00:00
{ PCMCIA_CARD(IBM, HOME_AND_AWAY), 0},
{ PCMCIA_CARD(IBM, INFOMOVER), NE2000DVF_ENADDR, 0xff0},
{ PCMCIA_CARD(IODATA3, PCLAT), 0},
Add a boatload of new device ids, gleaned from the pcmcia-cs-3.2.8 distribution. Add the appropriate devices to the man pages. Obtained from: pcmcia-cs Approved by: re (scottl)
2005-07-10 02:00:35 +00:00
{ PCMCIA_CARD(KINGSTON, CIO10T), 0},
Eliminate unused argument in PCMCIA_CARD macro. Provide a backwards compatible way to have the extra macro by defining PCCARD_API_LEVEL 5 before including pccarddevs for driver writers that want/need to have the same driver on 5 and 6 with pccard attachments. Approved by: re (dwhite)
2005-06-24 14:36:54 +00:00
{ PCMCIA_CARD(KINGSTON, KNE2), 0},
{ PCMCIA_CARD(LANTECH, FASTNETTX),NE2000DVF_AX88X90 },
{ PCMCIA_CARD(LINKSYS, COMBO_ECARD),
Fix the support for the ax88[17]90 to print the right type, rather than the generic ne-2000 string. This should have no effect on the actual support of the parts, just reporting what the part was. Also, rename a few functins and symbols to reflect a more generic part support that grew out of the early specific support.
2005-02-17 18:15:13 +00:00
NE2000DVF_DL100XX | NE2000DVF_AX88X90 },
Eliminate unused argument in PCMCIA_CARD macro. Provide a backwards compatible way to have the extra macro by defining PCCARD_API_LEVEL 5 before including pccarddevs for driver writers that want/need to have the same driver on 5 and 6 with pccard attachments. Approved by: re (dwhite)
2005-06-24 14:36:54 +00:00
{ PCMCIA_CARD(LINKSYS, ECARD_1), 0},
{ PCMCIA_CARD(LINKSYS, ECARD_2), 0},
{ PCMCIA_CARD(LINKSYS, ETHERFAST), NE2000DVF_DL100XX },
{ PCMCIA_CARD(LINKSYS, TRUST_COMBO_ECARD), 0},
{ PCMCIA_CARD(MACNICA, ME1_JEIDA), 0},
{ PCMCIA_CARD(MAGICRAM, ETHER), 0},
{ PCMCIA_CARD(MELCO, LPC3_CLX), NE2000DVF_AX88X90 },
{ PCMCIA_CARD(MELCO, LPC3_TX), NE2000DVF_AX88X90 },
{ PCMCIA_CARD(NDC, ND5100_E), 0},
{ PCMCIA_CARD(NETGEAR, FA410TXC), NE2000DVF_DL100XX},
MFp4: Reduce the size of ed a little by removing some CIS based entries (others likely can be removed too): o The D-Link DFE-670TXD doesn't need its own entry based on strings. o The Xircom CompactCard appears to be a TDK design, so list it there by ID and remove the strings. Increase the size of ed a little: o Add support for the Addtron AE-660CT and Addtron AE-660. This is a very generic NE-2000 clone (so generic that its CIS tags say NE-2000 generic card!).
2005-09-10 00:17:55 +00:00
/* Same ID as DLINK DFE-670TXD. 670 has DL10022, fa411 has ax88790 */
{ PCMCIA_CARD(NETGEAR, FA411), NE2000DVF_AX88X90 | NE2000DVF_DL100XX},
Eliminate unused argument in PCMCIA_CARD macro. Provide a backwards compatible way to have the extra macro by defining PCCARD_API_LEVEL 5 before including pccarddevs for driver writers that want/need to have the same driver on 5 and 6 with pccard attachments. Approved by: re (dwhite)
2005-06-24 14:36:54 +00:00
{ PCMCIA_CARD(NEXTCOM, NEXTHAWK), 0},
Add a boatload of new device ids, gleaned from the pcmcia-cs-3.2.8 distribution. Add the appropriate devices to the man pages. Obtained from: pcmcia-cs Approved by: re (scottl)
2005-07-10 02:00:35 +00:00
{ PCMCIA_CARD(NEWMEDIA, LANSURFER), 0},
Eliminate unused argument in PCMCIA_CARD macro. Provide a backwards compatible way to have the extra macro by defining PCCARD_API_LEVEL 5 before including pccarddevs for driver writers that want/need to have the same driver on 5 and 6 with pccard attachments. Approved by: re (dwhite)
2005-06-24 14:36:54 +00:00
{ PCMCIA_CARD(OEM2, ETHERNET), 0},
MFp4: Reduce the size of ed a little by removing some CIS based entries (others likely can be removed too): o The D-Link DFE-670TXD doesn't need its own entry based on strings. o The Xircom CompactCard appears to be a TDK design, so list it there by ID and remove the strings. Increase the size of ed a little: o Add support for the Addtron AE-660CT and Addtron AE-660. This is a very generic NE-2000 clone (so generic that its CIS tags say NE-2000 generic card!).
2005-09-10 00:17:55 +00:00
{ PCMCIA_CARD(OEM2, NE2000), 0},
Eliminate unused argument in PCMCIA_CARD macro. Provide a backwards compatible way to have the extra macro by defining PCCARD_API_LEVEL 5 before including pccarddevs for driver writers that want/need to have the same driver on 5 and 6 with pccard attachments. Approved by: re (dwhite)
2005-06-24 14:36:54 +00:00
{ PCMCIA_CARD(PLANET, SMARTCOM2000), 0 },
{ PCMCIA_CARD(PREMAX, PE200), 0},
Add a boatload of new device ids, gleaned from the pcmcia-cs-3.2.8 distribution. Add the appropriate devices to the man pages. Obtained from: pcmcia-cs Approved by: re (scottl)
2005-07-10 02:00:35 +00:00
{ PCMCIA_CARD(PSION, LANGLOBAL), 0},
Eliminate unused argument in PCMCIA_CARD macro. Provide a backwards compatible way to have the extra macro by defining PCCARD_API_LEVEL 5 before including pccarddevs for driver writers that want/need to have the same driver on 5 and 6 with pccard attachments. Approved by: re (dwhite)
2005-06-24 14:36:54 +00:00
{ PCMCIA_CARD(RACORE, ETHERNET), 0},
{ PCMCIA_CARD(RACORE, FASTENET), NE2000DVF_AX88X90},
Add a boatload of new device ids, gleaned from the pcmcia-cs-3.2.8 distribution. Add the appropriate devices to the man pages. Obtained from: pcmcia-cs Approved by: re (scottl)
2005-07-10 02:00:35 +00:00
{ PCMCIA_CARD(RACORE, 8041TX), NE2000DVF_AX88X90},
{ PCMCIA_CARD(RELIA, COMBO), 0},
Eliminate unused argument in PCMCIA_CARD macro. Provide a backwards compatible way to have the extra macro by defining PCCARD_API_LEVEL 5 before including pccarddevs for driver writers that want/need to have the same driver on 5 and 6 with pccard attachments. Approved by: re (dwhite)
2005-06-24 14:36:54 +00:00
{ PCMCIA_CARD(RPTI, EP400), 0},
{ PCMCIA_CARD(RPTI, EP401), 0},
{ PCMCIA_CARD(SMC, EZCARD), 0},
{ PCMCIA_CARD(SOCKET, EA_ETHER), 0},
Add a boatload of new device ids, gleaned from the pcmcia-cs-3.2.8 distribution. Add the appropriate devices to the man pages. Obtained from: pcmcia-cs Approved by: re (scottl)
2005-07-10 02:00:35 +00:00
{ PCMCIA_CARD(SOCKET, ES_1000), 0},
Eliminate unused argument in PCMCIA_CARD macro. Provide a backwards compatible way to have the extra macro by defining PCCARD_API_LEVEL 5 before including pccarddevs for driver writers that want/need to have the same driver on 5 and 6 with pccard attachments. Approved by: re (dwhite)
2005-06-24 14:36:54 +00:00
{ PCMCIA_CARD(SOCKET, LP_ETHER), 0},
{ PCMCIA_CARD(SOCKET, LP_ETHER_CF), 0},
{ PCMCIA_CARD(SOCKET, LP_ETH_10_100_CF), NE2000DVF_DL100XX},
{ PCMCIA_CARD(SVEC, COMBOCARD), 0},
{ PCMCIA_CARD(SVEC, LANCARD), 0},
{ PCMCIA_CARD(TAMARACK, ETHERNET), 0},
MFp4: Reduce the size of ed a little by removing some CIS based entries (others likely can be removed too): o The D-Link DFE-670TXD doesn't need its own entry based on strings. o The Xircom CompactCard appears to be a TDK design, so list it there by ID and remove the strings. Increase the size of ed a little: o Add support for the Addtron AE-660CT and Addtron AE-660. This is a very generic NE-2000 clone (so generic that its CIS tags say NE-2000 generic card!).
2005-09-10 00:17:55 +00:00
{ PCMCIA_CARD(TDK, CFE_10), 0},
Eliminate unused argument in PCMCIA_CARD macro. Provide a backwards compatible way to have the extra macro by defining PCCARD_API_LEVEL 5 before including pccarddevs for driver writers that want/need to have the same driver on 5 and 6 with pccard attachments. Approved by: re (dwhite)
2005-06-24 14:36:54 +00:00
{ PCMCIA_CARD(TDK, LAK_CD031), 0},
Add a boatload of new device ids, gleaned from the pcmcia-cs-3.2.8 distribution. Add the appropriate devices to the man pages. Obtained from: pcmcia-cs Approved by: re (scottl)
2005-07-10 02:00:35 +00:00
{ PCMCIA_CARD(TDK, DFL5610WS), 0},
{ PCMCIA_CARD(TELECOMDEVICE, LM5LT), 0 },
Eliminate unused argument in PCMCIA_CARD macro. Provide a backwards compatible way to have the extra macro by defining PCCARD_API_LEVEL 5 before including pccarddevs for driver writers that want/need to have the same driver on 5 and 6 with pccard attachments. Approved by: re (dwhite)
2005-06-24 14:36:54 +00:00
{ PCMCIA_CARD(TELECOMDEVICE, TCD_HPC100), NE2000DVF_AX88X90 },
{ PCMCIA_CARD(ZONET, ZEN), 0},
Migrate to PCMCIA_CARD() macros
2001-11-15 07:52:49 +00:00
{ { NULL } }
Update the ed driver to probe and attach under a NEWCARD kernel (I was using a cardbus based system with pccbb providing the pcic interface). Something isn't quite right.. when the driver allocates and activates its resources, the IO space that was requested reads as all zeros (versus the original 0xff's as it normally is when there is no device responding). Also, deactivate the resources before releasing them. OLDCARD doesn't seem to care but NEWCARD/CARDBUS get rather unhappy if you release a resource that hasn't been deactivated yet. Make pcic_p.c only compile with oldcard kernels.
2000-11-25 03:36:09 +00:00
};
Minor reorg of pccard attach code to fix ax88x90 case broken after fixing dl100xx case. o We no longer acquire and release resources during attach many times. We now do it once at the beginning. o Move setting the resource offsets to just after acquiring the ports in attach. o Move ax88x90 code to the end of the file, just after the dl100xx specific code. o Rename ed_pccard_Linksys to ed_pccard_dl100xx to reflect the underlying chipset. o Pass the ed_product structure into ed_pccard_{dl100xx,ax88x90} and have those routines test the flags to see if this card should be probed in that way. o transition from ed_probe_Novell to ed_probe_Novell_generic since we already have the resources setup. o Move use of ed_probe_Novell_generic into ed_pccard_dl100xx to be more consistant with ax88x90 case. o simplify the code where we probe for the chipsets
2005-09-07 03:30:58 +00:00
/*
* PC Card (PCMCIA) specific code.
*/
static int ed_pccard_probe(device_t);
static int ed_pccard_attach(device_t);
static int ed_pccard_memread(device_t dev, off_t offset, u_char *byte);
static int ed_pccard_memwrite(device_t dev, off_t offset, u_char byte);
static int ed_pccard_dl100xx(device_t dev, const struct ed_product *);
#ifndef ED_NO_MIIBUS
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
static void ed_pccard_dl10xx_mii_reset(struct ed_softc *sc);
static u_int ed_pccard_dl10xx_mii_readbits(struct ed_softc *sc, int nbits);
static void ed_pccard_dl10xx_mii_writebits(struct ed_softc *sc, u_int val,
Minor reorg of pccard attach code to fix ax88x90 case broken after fixing dl100xx case. o We no longer acquire and release resources during attach many times. We now do it once at the beginning. o Move setting the resource offsets to just after acquiring the ports in attach. o Move ax88x90 code to the end of the file, just after the dl100xx specific code. o Rename ed_pccard_Linksys to ed_pccard_dl100xx to reflect the underlying chipset. o Pass the ed_product structure into ed_pccard_{dl100xx,ax88x90} and have those routines test the flags to see if this card should be probed in that way. o transition from ed_probe_Novell to ed_probe_Novell_generic since we already have the resources setup. o Move use of ed_probe_Novell_generic into ed_pccard_dl100xx to be more consistant with ax88x90 case. o simplify the code where we probe for the chipsets
2005-09-07 03:30:58 +00:00
int nbits);
#endif
static int ed_pccard_ax88x90(device_t dev, const struct ed_product *);
Update the ed driver to probe and attach under a NEWCARD kernel (I was using a cardbus based system with pccbb providing the pcic interface). Something isn't quite right.. when the driver allocates and activates its resources, the IO space that was requested reads as all zeros (versus the original 0xff's as it normally is when there is no device responding). Also, deactivate the resources before releasing them. OLDCARD doesn't seem to care but NEWCARD/CARDBUS get rather unhappy if you release a resource that hasn't been deactivated yet. Make pcic_p.c only compile with oldcard kernels.
2000-11-25 03:36:09 +00:00
static int
Finish conversion to bus space and make ed MP safe. o Lock ed o Fix extra newline in probe messages o Eliminate gone. o Make detach less-racy. o Eliminate spl* o Switch from timeout/untimeout to callout interface. o Read/write card memory using bus_space calls. o generalize readmem so that we don't need ifs in the code. o Fix memory stuff to be consistant. o Remove OLDCARD compat stuff. o Mark interrupt as MPSAFE. # sic, hpp not tested at all # ISA and PCI attachments lightly tested
2005-08-28 23:56:25 +00:00
ed_pccard_probe(device_t dev)
Update the ed driver to probe and attach under a NEWCARD kernel (I was using a cardbus based system with pccbb providing the pcic interface). Something isn't quite right.. when the driver allocates and activates its resources, the IO space that was requested reads as all zeros (versus the original 0xff's as it normally is when there is no device responding). Also, deactivate the resources before releasing them. OLDCARD doesn't seem to care but NEWCARD/CARDBUS get rather unhappy if you release a resource that hasn't been deactivated yet. Make pcic_p.c only compile with oldcard kernels.
2000-11-25 03:36:09 +00:00
{
integrate from p4 repo: o Honor NE2000DVF_{AX88190,DL10019} flags by setting the ED_FLAGS_xxxx flag. o Mark linksys combo_ecard as ax88190 o Set the type_str to AX88190 for the ax88190 cards. This fixes ax88190 based cards, for the most part, but doesn't seem to fix the mii based dl10019 cards (aka linksys cards).
2002-07-22 21:28:46 +00:00
const struct ed_product *pp;
Only attach to network functions.
2005-01-20 19:37:22 +00:00
int error;
uint32_t fcn = PCCARD_FUNCTION_UNSPEC;
/* Make sure we're a network function */
error = pccard_get_function(dev, &fcn);
if (error != 0)
return (error);
Update the ed driver to probe and attach under a NEWCARD kernel (I was using a cardbus based system with pccbb providing the pcic interface). Something isn't quite right.. when the driver allocates and activates its resources, the IO space that was requested reads as all zeros (versus the original 0xff's as it normally is when there is no device responding). Also, deactivate the resources before releasing them. OLDCARD doesn't seem to care but NEWCARD/CARDBUS get rather unhappy if you release a resource that hasn't been deactivated yet. Make pcic_p.c only compile with oldcard kernels.
2000-11-25 03:36:09 +00:00
integrate from p4 repo: o Honor NE2000DVF_{AX88190,DL10019} flags by setting the ED_FLAGS_xxxx flag. o Mark linksys combo_ecard as ax88190 o Set the type_str to AX88190 for the ax88190 cards. This fixes ax88190 based cards, for the most part, but doesn't seem to fix the mii based dl10019 cards (aka linksys cards).
2002-07-22 21:28:46 +00:00
if ((pp = (const struct ed_product *) pccard_product_lookup(dev,
Add a bunch of entries from NetBSD's if_ne_pcmcia.c v1.67. We still need to verify that all the entries in pccard.conf are present in this table.
2001-01-21 08:20:23 +00:00
(const struct pccard_product *) ed_pccard_products,
Argh, I have fixed this cut/paste error twice before. I must have committed the wrong patch. :-( sn_pccard_products[] should have been static anyway.
2000-11-25 07:25:08 +00:00
sizeof(ed_pccard_products[0]), NULL)) != NULL) {
Don't set the description if NULL.
2003-04-10 06:08:27 +00:00
if (pp->prod.pp_name != NULL)
device_set_desc(dev, pp->prod.pp_name);
Add support for the DLink DMF650TX Ethernet+56k modem PC Card. This gets the ethernet part of the card working, while putting appropriate hooks in place for the modem code. Other ed based lan/modem combo cards should be easy to add. Please send me info on any you'd like to see support added. Note: The 650 isn't a strictly conforming multi-function card, so special support is needed. :-(
2005-07-15 01:54:59 +00:00
/*
* Some devices don't ID themselves as network, but
* that's OK if the flags say so.
*/
if (!(pp->flags & NE2000DVF_ANYFUNC) &&
fcn != PCCARD_FUNCTION_NETWORK)
return (ENXIO);
integrate from p4 repo: o Honor NE2000DVF_{AX88190,DL10019} flags by setting the ED_FLAGS_xxxx flag. o Mark linksys combo_ecard as ax88190 o Set the type_str to AX88190 for the ax88190 cards. This fixes ax88190 based cards, for the most part, but doesn't seem to fix the mii based dl10019 cards (aka linksys cards).
2002-07-22 21:28:46 +00:00
return (0);
Update the ed driver to probe and attach under a NEWCARD kernel (I was using a cardbus based system with pccbb providing the pcic interface). Something isn't quite right.. when the driver allocates and activates its resources, the IO space that was requested reads as all zeros (versus the original 0xff's as it normally is when there is no device responding). Also, deactivate the resources before releasing them. OLDCARD doesn't seem to care but NEWCARD/CARDBUS get rather unhappy if you release a resource that hasn't been deactivated yet. Make pcic_p.c only compile with oldcard kernels.
2000-11-25 03:36:09 +00:00
}
It is a little better to return ENXIO from probe/match routine than EIO.
2004-07-21 20:38:12 +00:00
return (ENXIO);
Update the ed driver to probe and attach under a NEWCARD kernel (I was using a cardbus based system with pccbb providing the pcic interface). Something isn't quite right.. when the driver allocates and activates its resources, the IO space that was requested reads as all zeros (versus the original 0xff's as it normally is when there is no device responding). Also, deactivate the resources before releasing them. OLDCARD doesn't seem to care but NEWCARD/CARDBUS get rather unhappy if you release a resource that hasn't been deactivated yet. Make pcic_p.c only compile with oldcard kernels.
2000-11-25 03:36:09 +00:00
}
Break out the bus front end code into separate files. Note that these haven't been turned on nor has the old code been removed from if_ed.c. The next commit will address that.
1999-10-15 01:28:04 +00:00
static int
o It turns out that most of the ne-2000 cards that I have got real unhappy with the latest changes. They actually have valid ROM data at location 0 of memory, just like a real NE-2000 ISA card. Use this data, if the ROM passes a few basic tests, as an additional source for the MAC address. Prefer the CIS over this source, but have it take precidence over falling back to reading the attribtue memory. o Minor cleanup of a few devices that we match on based on CIS string.
2005-02-14 22:27:03 +00:00
ed_pccard_rom_mac(device_t dev, uint8_t *enaddr)
Break out the bus front end code into separate files. Note that these haven't been turned on nor has the old code been removed from if_ed.c. The next commit will address that.
1999-10-15 01:28:04 +00:00
{
o It turns out that most of the ne-2000 cards that I have got real unhappy with the latest changes. They actually have valid ROM data at location 0 of memory, just like a real NE-2000 ISA card. Use this data, if the ROM passes a few basic tests, as an additional source for the MAC address. Prefer the CIS over this source, but have it take precidence over falling back to reading the attribtue memory. o Minor cleanup of a few devices that we match on based on CIS string.
2005-02-14 22:27:03 +00:00
struct ed_softc *sc = device_get_softc(dev);
uint8_t romdata[16];
Propagate MAC address in CIS tupple to dev/ed/ driver. Many ed-based Ethernet PC-cards can't get correct MAC address without this patch. Submitted by: Takanori Watanabe <takawata@shidahara1.planet.sci.kobe-u.ac.jp> Reviewed by: Warner Losh <imp@village.org>
2000-01-21 03:08:46 +00:00
int i;
o It turns out that most of the ne-2000 cards that I have got real unhappy with the latest changes. They actually have valid ROM data at location 0 of memory, just like a real NE-2000 ISA card. Use this data, if the ROM passes a few basic tests, as an additional source for the MAC address. Prefer the CIS over this source, but have it take precidence over falling back to reading the attribtue memory. o Minor cleanup of a few devices that we match on based on CIS string.
2005-02-14 22:27:03 +00:00
/*
* Read in the rom data at location 0. We should see one of
* two patterns. Either you'll see odd locations 0xff and
* even locations data, or you'll see odd and even locations
* mirror each other. In addition, the last two even locations
* should be 0. If they aren't 0, then we'll assume that
* there's no valid ROM data on this card and try another method
* to recover the MAC address. Since there are no NE-1000
* based PC Card devices, we'll assume we're 16-bit.
*/
ed_pio_readmem(sc, 0, romdata, 16);
if (romdata[12] != 0 || romdata[14] != 0)
return 0;
for (i = 0; i < ETHER_ADDR_LEN; i++)
enaddr[i] = romdata[i * 2];
return 1;
}
Add support for the DLink DMF650TX Ethernet+56k modem PC Card. This gets the ethernet part of the card working, while putting appropriate hooks in place for the modem code. Other ed based lan/modem combo cards should be easy to add. Please send me info on any you'd like to see support added. Note: The 650 isn't a strictly conforming multi-function card, so special support is needed. :-(
2005-07-15 01:54:59 +00:00
static int
It turns out that many of the ambicom cards have their CIS resources listed in different orders. Since it is easy to identify the Modem resources vs the Ethernet resources by looking at the size, use that rather than hard coded rids. For such parts, go ahead and guess which rid we should use based on the size. This guess appears reliable for the two example cards that I have with different CIS info.
2005-09-06 22:44:26 +00:00
ed_pccard_add_modem(device_t dev)
Add support for the DLink DMF650TX Ethernet+56k modem PC Card. This gets the ethernet part of the card working, while putting appropriate hooks in place for the modem code. Other ed based lan/modem combo cards should be easy to add. Please send me info on any you'd like to see support added. Note: The 650 isn't a strictly conforming multi-function card, so special support is needed. :-(
2005-07-15 01:54:59 +00:00
{
It turns out that many of the ambicom cards have their CIS resources listed in different orders. Since it is easy to identify the Modem resources vs the Ethernet resources by looking at the size, use that rather than hard coded rids. For such parts, go ahead and guess which rid we should use based on the size. This guess appears reliable for the two example cards that I have with different CIS info.
2005-09-06 22:44:26 +00:00
struct ed_softc *sc = device_get_softc(dev);
device_printf(dev, "Need to write this code: modem rid is %d\n",
sc->modem_rid);
Add support for the DLink DMF650TX Ethernet+56k modem PC Card. This gets the ethernet part of the card working, while putting appropriate hooks in place for the modem code. Other ed based lan/modem combo cards should be easy to add. Please send me info on any you'd like to see support added. Note: The 650 isn't a strictly conforming multi-function card, so special support is needed. :-(
2005-07-15 01:54:59 +00:00
return 0;
}
o It turns out that most of the ne-2000 cards that I have got real unhappy with the latest changes. They actually have valid ROM data at location 0 of memory, just like a real NE-2000 ISA card. Use this data, if the ROM passes a few basic tests, as an additional source for the MAC address. Prefer the CIS over this source, but have it take precidence over falling back to reading the attribtue memory. o Minor cleanup of a few devices that we match on based on CIS string.
2005-02-14 22:27:03 +00:00
static int
ed_pccard_attach(device_t dev)
{
Propagate MAC address in CIS tupple to dev/ed/ driver. Many ed-based Ethernet PC-cards can't get correct MAC address without this patch. Submitted by: Takanori Watanabe <takawata@shidahara1.planet.sci.kobe-u.ac.jp> Reviewed by: Warner Losh <imp@village.org>
2000-01-21 03:08:46 +00:00
u_char sum;
Improve the mining of MAC address from the card: o Add a fallback location for the MAC address. Most of the early ne2000 PC Cards were built from the same parts, so most of them have the same address in the CIS to grab the MAC from. Use this address as our fallback if we don't find anything better. o Add printf, in bootverbose, noting the MAC addresses that we find along the way. # Better sanity checking of the MAC address is needed. Will have to # investigate using/creating a centralized function to do this as a number # of other PC Card drivers each have their own ad-hoc tests.
2005-02-12 16:44:15 +00:00
u_char enaddr[ETHER_ADDR_LEN];
Really old PCMCIA cards stored the MAC address in the attribute memory at some offset. Unlike newer cards, the MAC address wasn't part of the CIS as a specific FUNCE. These older cards were having their MAC address show up as 0:2:4:6:8:a because that's what's in the ROM locations that would be there in a real ne2000. This patch allows one to specify the offset for the MAC address for these cards. Specify one for the IBM Ethernet II card, as it is one that has this problem. One shouldn't specify this unless the MAC address really isn't in the CIS at all. Side note: The novell probe likely shouldn't read the MAC address, and that should be moved to the bus specific attach routine(s), maybe as a convenience function in if_ed_novell.c. My IBM Ethernet II (aka Info Mover) now has a believable MAC address.
2005-02-12 00:45:22 +00:00
const struct ed_product *pp;
Finish conversion to bus space and make ed MP safe. o Lock ed o Fix extra newline in probe messages o Eliminate gone. o Make detach less-racy. o Eliminate spl* o Switch from timeout/untimeout to callout interface. o Read/write card memory using bus_space calls. o generalize readmem so that we don't need ifs in the code. o Fix memory stuff to be consistant. o Remove OLDCARD compat stuff. o Mark interrupt as MPSAFE. # sic, hpp not tested at all # ISA and PCI attachments lightly tested
2005-08-28 23:56:25 +00:00
int error, i;
struct ed_softc *sc = device_get_softc(dev);
It turns out that many of the ambicom cards have their CIS resources listed in different orders. Since it is easy to identify the Modem resources vs the Ethernet resources by looking at the size, use that rather than hard coded rids. For such parts, go ahead and guess which rid we should use based on the size. This guess appears reliable for the two example cards that I have with different CIS info.
2005-09-06 22:44:26 +00:00
u_long size;
Finish conversion to bus space and make ed MP safe. o Lock ed o Fix extra newline in probe messages o Eliminate gone. o Make detach less-racy. o Eliminate spl* o Switch from timeout/untimeout to callout interface. o Read/write card memory using bus_space calls. o generalize readmem so that we don't need ifs in the code. o Fix memory stuff to be consistant. o Remove OLDCARD compat stuff. o Mark interrupt as MPSAFE. # sic, hpp not tested at all # ISA and PCI attachments lightly tested
2005-08-28 23:56:25 +00:00
Really old PCMCIA cards stored the MAC address in the attribute memory at some offset. Unlike newer cards, the MAC address wasn't part of the CIS as a specific FUNCE. These older cards were having their MAC address show up as 0:2:4:6:8:a because that's what's in the ROM locations that would be there in a real ne2000. This patch allows one to specify the offset for the MAC address for these cards. Specify one for the IBM Ethernet II card, as it is one that has this problem. One shouldn't specify this unless the MAC address really isn't in the CIS at all. Side note: The novell probe likely shouldn't read the MAC address, and that should be moved to the bus specific attach routine(s), maybe as a convenience function in if_ed_novell.c. My IBM Ethernet II (aka Info Mover) now has a believable MAC address.
2005-02-12 00:45:22 +00:00
if ((pp = (const struct ed_product *) pccard_product_lookup(dev,
(const struct pccard_product *) ed_pccard_products,
sizeof(ed_pccard_products[0]), NULL)) == NULL)
return (ENXIO);
It turns out that many of the ambicom cards have their CIS resources listed in different orders. Since it is easy to identify the Modem resources vs the Ethernet resources by looking at the size, use that rather than hard coded rids. For such parts, go ahead and guess which rid we should use based on the size. This guess appears reliable for the two example cards that I have with different CIS info.
2005-09-06 22:44:26 +00:00
sc->modem_rid = -1;
if (pp->flags & NE2000DVF_MODEM) {
sc->port_rid = -1;
for (i = 0; i < 4; i++) {
size = bus_get_resource_count(dev, SYS_RES_IOPORT, i);
if (size == ED_NOVELL_IO_PORTS)
sc->port_rid = i;
else if (size == 8)
sc->modem_rid = i;
}
if (sc->port_rid == -1) {
device_printf(dev, "Cannot locate my ports!\n");
return (ENXIO);
}
} else {
sc->port_rid = 0;
}
Minor reorg of pccard attach code to fix ax88x90 case broken after fixing dl100xx case. o We no longer acquire and release resources during attach many times. We now do it once at the beginning. o Move setting the resource offsets to just after acquiring the ports in attach. o Move ax88x90 code to the end of the file, just after the dl100xx specific code. o Rename ed_pccard_Linksys to ed_pccard_dl100xx to reflect the underlying chipset. o Pass the ed_product structure into ed_pccard_{dl100xx,ax88x90} and have those routines test the flags to see if this card should be probed in that way. o transition from ed_probe_Novell to ed_probe_Novell_generic since we already have the resources setup. o Move use of ed_probe_Novell_generic into ed_pccard_dl100xx to be more consistant with ax88x90 case. o simplify the code where we probe for the chipsets
2005-09-07 03:30:58 +00:00
/* Allocate the port resource during setup. */
error = ed_alloc_port(dev, sc->port_rid, ED_NOVELL_IO_PORTS);
if (error)
Finish conversion to bus space and make ed MP safe. o Lock ed o Fix extra newline in probe messages o Eliminate gone. o Make detach less-racy. o Eliminate spl* o Switch from timeout/untimeout to callout interface. o Read/write card memory using bus_space calls. o generalize readmem so that we don't need ifs in the code. o Fix memory stuff to be consistant. o Remove OLDCARD compat stuff. o Mark interrupt as MPSAFE. # sic, hpp not tested at all # ISA and PCI attachments lightly tested
2005-08-28 23:56:25 +00:00
return (error);
error = ed_alloc_irq(dev, 0, 0);
Minor reorg of pccard attach code to fix ax88x90 case broken after fixing dl100xx case. o We no longer acquire and release resources during attach many times. We now do it once at the beginning. o Move setting the resource offsets to just after acquiring the ports in attach. o Move ax88x90 code to the end of the file, just after the dl100xx specific code. o Rename ed_pccard_Linksys to ed_pccard_dl100xx to reflect the underlying chipset. o Pass the ed_product structure into ed_pccard_{dl100xx,ax88x90} and have those routines test the flags to see if this card should be probed in that way. o transition from ed_probe_Novell to ed_probe_Novell_generic since we already have the resources setup. o Move use of ed_probe_Novell_generic into ed_pccard_dl100xx to be more consistant with ax88x90 case. o simplify the code where we probe for the chipsets
2005-09-07 03:30:58 +00:00
if (error)
goto bad;
/*
* Determine which chipset we are. All the PC Card chipsets have the
* ASIC and NIC offsets in the same place.
*/
sc->asic_offset = ED_NOVELL_ASIC_OFFSET;
sc->nic_offset = ED_NOVELL_NIC_OFFSET;
error = ENXIO;
if (error != 0)
error = ed_pccard_dl100xx(dev, pp);
if (error != 0 && pp->flags & NE2000DVF_AX88X90)
error = ed_pccard_ax88x90(dev, pp);
if (error != 0)
error = ed_probe_Novell_generic(dev, device_get_flags(dev));
if (error)
goto bad;
Finish conversion to bus space and make ed MP safe. o Lock ed o Fix extra newline in probe messages o Eliminate gone. o Make detach less-racy. o Eliminate spl* o Switch from timeout/untimeout to callout interface. o Read/write card memory using bus_space calls. o generalize readmem so that we don't need ifs in the code. o Fix memory stuff to be consistant. o Remove OLDCARD compat stuff. o Mark interrupt as MPSAFE. # sic, hpp not tested at all # ISA and PCI attachments lightly tested
2005-08-28 23:56:25 +00:00
error = bus_setup_intr(dev, sc->irq_res, INTR_TYPE_NET | INTR_MPSAFE,
edintr, sc, &sc->irq_handle);
Add pccard attachment. This won't work with memory mapped ed devices yet, but that should be resolved shortly. Non memory mapped ed devices should work, but I cannot test this since my only ed card is memory mapped. Submitted by: Matt Dodd <mdodd@freebsd.org>
1999-10-25 02:55:14 +00:00
if (error) {
Rework how we deal with the DL10019 and DL10022 cards (as well as the AX88190 ones, but that one only minorly): o don't set flags in the match routine. They appear to be cleared when probe/attach is called. Before this change, they were always treated as a simple ne2000, which would fail to get the right NIC address. o Lookup device again in the probe routine and probe based on the cards that you see. o Detect and report the DL10022 seprately from the DL10019 cards. While I'm here: o remove a bad printf o change another bad printf to device_printf. o minor style(9) formatting tweaks. # note: a lot of OEM entries are in the ed_pccard_products such that we can # likely remove, or collapse, many of them. This makes all of my DL100xx cards at least probe the ethernet address correctly, which it wasn't doing before. I can't seem to locate my AX88xxx based cards, so those haven't been tested, but they were busted before the change so they can't be any worse now...
2005-01-26 05:21:12 +00:00
device_printf(dev, "setup intr failed %d \n", error);
Minor reorg of pccard attach code to fix ax88x90 case broken after fixing dl100xx case. o We no longer acquire and release resources during attach many times. We now do it once at the beginning. o Move setting the resource offsets to just after acquiring the ports in attach. o Move ax88x90 code to the end of the file, just after the dl100xx specific code. o Rename ed_pccard_Linksys to ed_pccard_dl100xx to reflect the underlying chipset. o Pass the ed_product structure into ed_pccard_{dl100xx,ax88x90} and have those routines test the flags to see if this card should be probed in that way. o transition from ed_probe_Novell to ed_probe_Novell_generic since we already have the resources setup. o Move use of ed_probe_Novell_generic into ed_pccard_dl100xx to be more consistant with ax88x90 case. o simplify the code where we probe for the chipsets
2005-09-07 03:30:58 +00:00
goto bad;
Add pccard attachment. This won't work with memory mapped ed devices yet, but that should be resolved shortly. Non memory mapped ed devices should work, but I cannot test this since my only ed card is memory mapped. Submitted by: Matt Dodd <mdodd@freebsd.org>
1999-10-25 02:55:14 +00:00
}
Add comment about MAC mining
2005-02-12 08:27:46 +00:00
/*
o It turns out that most of the ne-2000 cards that I have got real unhappy with the latest changes. They actually have valid ROM data at location 0 of memory, just like a real NE-2000 ISA card. Use this data, if the ROM passes a few basic tests, as an additional source for the MAC address. Prefer the CIS over this source, but have it take precidence over falling back to reading the attribtue memory. o Minor cleanup of a few devices that we match on based on CIS string.
2005-02-14 22:27:03 +00:00
* For the older cards, we have to get the MAC address from
* the card in some way. Let's try the standard PCMCIA way
* first. If that fails, then check to see if we have valid
* data from the standard NE-2000 data roms. If that fails,
* check to see if the card has a hint about where to look in
* its CIS. If that fails, maybe we should look at some
* default value. In all fails, we should fail the attach,
* but don't right now.
Add comment about MAC mining
2005-02-12 08:27:46 +00:00
*/
Eliminate the ED_VENDOR_{PCCARD,LINKSYS} types. These cards really are NOVELL NE2000 with just a tiny quirk that's non vendor specific. Instead, use the chip_type of DL100XX instead. This is more inline with how the AX88190 support was added, and seems a little cleaner.
2005-02-11 06:02:27 +00:00
if (sc->chip_type == ED_CHIP_TYPE_DP8390) {
Improve the mining of MAC address from the card: o Add a fallback location for the MAC address. Most of the early ne2000 PC Cards were built from the same parts, so most of them have the same address in the CIS to grab the MAC from. Use this address as our fallback if we don't find anything better. o Add printf, in bootverbose, noting the MAC addresses that we find along the way. # Better sanity checking of the MAC address is needed. Will have to # investigate using/creating a centralized function to do this as a number # of other PC Card drivers each have their own ad-hoc tests.
2005-02-12 16:44:15 +00:00
pccard_get_ether(dev, enaddr);
if (bootverbose)
device_printf(dev, "CIS MAC %6D\n", enaddr, ":");
Only try to detect Linksys PCMCIA cards when we are in a pccard environment. This fixes the breakage to ISA ethernet cards. Reviewed by: peter
2000-06-29 07:31:37 +00:00
for (i = 0, sum = 0; i < ETHER_ADDR_LEN; i++)
Improve the mining of MAC address from the card: o Add a fallback location for the MAC address. Most of the early ne2000 PC Cards were built from the same parts, so most of them have the same address in the CIS to grab the MAC from. Use this address as our fallback if we don't find anything better. o Add printf, in bootverbose, noting the MAC addresses that we find along the way. # Better sanity checking of the MAC address is needed. Will have to # investigate using/creating a centralized function to do this as a number # of other PC Card drivers each have their own ad-hoc tests.
2005-02-12 16:44:15 +00:00
sum |= enaddr[i];
o It turns out that most of the ne-2000 cards that I have got real unhappy with the latest changes. They actually have valid ROM data at location 0 of memory, just like a real NE-2000 ISA card. Use this data, if the ROM passes a few basic tests, as an additional source for the MAC address. Prefer the CIS over this source, but have it take precidence over falling back to reading the attribtue memory. o Minor cleanup of a few devices that we match on based on CIS string.
2005-02-14 22:27:03 +00:00
if (sum == 0 && ed_pccard_rom_mac(dev, enaddr)) {
if (bootverbose)
device_printf(dev, "ROM mac %6D\n", enaddr,
":");
sum++;
}
Really old PCMCIA cards stored the MAC address in the attribute memory at some offset. Unlike newer cards, the MAC address wasn't part of the CIS as a specific FUNCE. These older cards were having their MAC address show up as 0:2:4:6:8:a because that's what's in the ROM locations that would be there in a real ne2000. This patch allows one to specify the offset for the MAC address for these cards. Specify one for the IBM Ethernet II card, as it is one that has this problem. One shouldn't specify this unless the MAC address really isn't in the CIS at all. Side note: The novell probe likely shouldn't read the MAC address, and that should be moved to the bus specific attach routine(s), maybe as a convenience function in if_ed_novell.c. My IBM Ethernet II (aka Info Mover) now has a believable MAC address.
2005-02-12 00:45:22 +00:00
if (sum == 0 && pp->flags & NE2000DVF_ENADDR) {
for (i = 0; i < ETHER_ADDR_LEN; i++) {
ed_pccard_memread(dev, pp->enoff + i * 2,
Improve the mining of MAC address from the card: o Add a fallback location for the MAC address. Most of the early ne2000 PC Cards were built from the same parts, so most of them have the same address in the CIS to grab the MAC from. Use this address as our fallback if we don't find anything better. o Add printf, in bootverbose, noting the MAC addresses that we find along the way. # Better sanity checking of the MAC address is needed. Will have to # investigate using/creating a centralized function to do this as a number # of other PC Card drivers each have their own ad-hoc tests.
2005-02-12 16:44:15 +00:00
enaddr + i);
sum |= enaddr[i];
Really old PCMCIA cards stored the MAC address in the attribute memory at some offset. Unlike newer cards, the MAC address wasn't part of the CIS as a specific FUNCE. These older cards were having their MAC address show up as 0:2:4:6:8:a because that's what's in the ROM locations that would be there in a real ne2000. This patch allows one to specify the offset for the MAC address for these cards. Specify one for the IBM Ethernet II card, as it is one that has this problem. One shouldn't specify this unless the MAC address really isn't in the CIS at all. Side note: The novell probe likely shouldn't read the MAC address, and that should be moved to the bus specific attach routine(s), maybe as a convenience function in if_ed_novell.c. My IBM Ethernet II (aka Info Mover) now has a believable MAC address.
2005-02-12 00:45:22 +00:00
}
Improve the mining of MAC address from the card: o Add a fallback location for the MAC address. Most of the early ne2000 PC Cards were built from the same parts, so most of them have the same address in the CIS to grab the MAC from. Use this address as our fallback if we don't find anything better. o Add printf, in bootverbose, noting the MAC addresses that we find along the way. # Better sanity checking of the MAC address is needed. Will have to # investigate using/creating a centralized function to do this as a number # of other PC Card drivers each have their own ad-hoc tests.
2005-02-12 16:44:15 +00:00
if (bootverbose)
device_printf(dev, "Hint MAC %6D\n", enaddr,
":");
}
if (sum == 0) {
for (i = 0; i < ETHER_ADDR_LEN; i++) {
ed_pccard_memread(dev, ED_DEFAULT_MAC_OFFSET +
i * 2, enaddr + i);
sum |= enaddr[i];
}
if (bootverbose)
device_printf(dev, "Fallback MAC %6D\n",
enaddr, ":");
}
if (sum == 0) {
device_printf(dev, "Cannot extract MAC address.\n");
ed_release_resources(dev);
return (ENXIO);
Really old PCMCIA cards stored the MAC address in the attribute memory at some offset. Unlike newer cards, the MAC address wasn't part of the CIS as a specific FUNCE. These older cards were having their MAC address show up as 0:2:4:6:8:a because that's what's in the ROM locations that would be there in a real ne2000. This patch allows one to specify the offset for the MAC address for these cards. Specify one for the IBM Ethernet II card, as it is one that has this problem. One shouldn't specify this unless the MAC address really isn't in the CIS at all. Side note: The novell probe likely shouldn't read the MAC address, and that should be moved to the bus specific attach routine(s), maybe as a convenience function in if_ed_novell.c. My IBM Ethernet II (aka Info Mover) now has a believable MAC address.
2005-02-12 00:45:22 +00:00
}
Stop embedding struct ifnet at the top of driver softcs. Instead the struct ifnet or the layer 2 common structure it was embedded in have been replaced with a struct ifnet pointer to be filled by a call to the new function, if_alloc(). The layer 2 common structure is also allocated via if_alloc() based on the interface type. It is hung off the new struct ifnet member, if_l2com. This change removes the size of these structures from the kernel ABI and will allow us to better manage them as interfaces come and go. Other changes of note: - Struct arpcom is no longer referenced in normal interface code. Instead the Ethernet address is accessed via the IFP2ENADDR() macro. To enforce this ac_enaddr has been renamed to _ac_enaddr. - The second argument to ether_ifattach is now always the mac address from driver private storage rather than sometimes being ac_enaddr. Reviewed by: sobomax, sam
2005-06-10 16:49:24 +00:00
bcopy(enaddr, sc->enaddr, ETHER_ADDR_LEN);
Only try to detect Linksys PCMCIA cards when we are in a pccard environment. This fixes the breakage to ISA ethernet cards. Reviewed by: peter
2000-06-29 07:31:37 +00:00
}
Propagate MAC address in CIS tupple to dev/ed/ driver. Many ed-based Ethernet PC-cards can't get correct MAC address without this patch. Submitted by: Takanori Watanabe <takawata@shidahara1.planet.sci.kobe-u.ac.jp> Reviewed by: Warner Losh <imp@village.org>
2000-01-21 03:08:46 +00:00
Replace the if_name and if_unit members of struct ifnet with new members if_xname, if_dname, and if_dunit. if_xname is the name of the interface and if_dname/unit are the driver name and instance. This change paves the way for interface renaming and enhanced pseudo device creation and configuration symantics. Approved By: re (in principle) Reviewed By: njl, imp Tested On: i386, amd64, sparc64 Obtained From: NetBSD (if_xname)
2003-10-31 18:32:15 +00:00
error = ed_attach(dev);
Minor reorg of pccard attach code to fix ax88x90 case broken after fixing dl100xx case. o We no longer acquire and release resources during attach many times. We now do it once at the beginning. o Move setting the resource offsets to just after acquiring the ports in attach. o Move ax88x90 code to the end of the file, just after the dl100xx specific code. o Rename ed_pccard_Linksys to ed_pccard_dl100xx to reflect the underlying chipset. o Pass the ed_product structure into ed_pccard_{dl100xx,ax88x90} and have those routines test the flags to see if this card should be probed in that way. o transition from ed_probe_Novell to ed_probe_Novell_generic since we already have the resources setup. o Move use of ed_probe_Novell_generic into ed_pccard_dl100xx to be more consistant with ax88x90 case. o simplify the code where we probe for the chipsets
2005-09-07 03:30:58 +00:00
if (error)
goto bad;
Add an option ED_NO_MIIBUS, which causes the `ed' driver to be built without support for miibus PHYs. Most ed cards don't need miibus support, so it's useful to be able to avoid the bloat of all the mii devices for small fixed-purpose kernels.
2001-09-29 22:32:03 +00:00
#ifndef ED_NO_MIIBUS
Finish conversion to bus space and make ed MP safe. o Lock ed o Fix extra newline in probe messages o Eliminate gone. o Make detach less-racy. o Eliminate spl* o Switch from timeout/untimeout to callout interface. o Read/write card memory using bus_space calls. o generalize readmem so that we don't need ifs in the code. o Fix memory stuff to be consistant. o Remove OLDCARD compat stuff. o Mark interrupt as MPSAFE. # sic, hpp not tested at all # ISA and PCI attachments lightly tested
2005-08-28 23:56:25 +00:00
if (sc->chip_type == ED_CHIP_TYPE_DL10019 ||
sc->chip_type == ED_CHIP_TYPE_DL10022) {
Attach the miibus for Linksys/Dlink cards from the attach routine, not from the probe routine. This was an oversight when I originally ported the miibus support to -current, though it was mostly harmless. We now set the vendor code to the new value ED_VENDOR_LINKSYS in ed_pccard_Linksys() at probe time. Then ed_pccard_attach() checks the vendor code, and sets up the miibus if appropriate. Reviewed by: imp
2001-07-22 18:58:44 +00:00
/* Probe for an MII bus, but ignore errors. */
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
ed_pccard_dl10xx_mii_reset(sc);
sc->mii_readbits = ed_pccard_dl10xx_mii_readbits;
sc->mii_writebits = ed_pccard_dl10xx_mii_writebits;
Attach the miibus for Linksys/Dlink cards from the attach routine, not from the probe routine. This was an oversight when I originally ported the miibus support to -current, though it was mostly harmless. We now set the vendor code to the new value ED_VENDOR_LINKSYS in ed_pccard_Linksys() at probe time. Then ed_pccard_attach() checks the vendor code, and sets up the miibus if appropriate. Reviewed by: imp
2001-07-22 18:58:44 +00:00
mii_phy_probe(dev, &sc->miibus, ed_ifmedia_upd,
ed_ifmedia_sts);
}
Add an option ED_NO_MIIBUS, which causes the `ed' driver to be built without support for miibus PHYs. Most ed cards don't need miibus support, so it's useful to be able to avoid the bloat of all the mii devices for small fixed-purpose kernels.
2001-09-29 22:32:03 +00:00
#endif
It turns out that many of the ambicom cards have their CIS resources listed in different orders. Since it is easy to identify the Modem resources vs the Ethernet resources by looking at the size, use that rather than hard coded rids. For such parts, go ahead and guess which rid we should use based on the size. This guess appears reliable for the two example cards that I have with different CIS info.
2005-09-06 22:44:26 +00:00
if (sc->modem_rid != -1)
ed_pccard_add_modem(dev);
Finish conversion to bus space and make ed MP safe. o Lock ed o Fix extra newline in probe messages o Eliminate gone. o Make detach less-racy. o Eliminate spl* o Switch from timeout/untimeout to callout interface. o Read/write card memory using bus_space calls. o generalize readmem so that we don't need ifs in the code. o Fix memory stuff to be consistant. o Remove OLDCARD compat stuff. o Mark interrupt as MPSAFE. # sic, hpp not tested at all # ISA and PCI attachments lightly tested
2005-08-28 23:56:25 +00:00
return (0);
Minor reorg of pccard attach code to fix ax88x90 case broken after fixing dl100xx case. o We no longer acquire and release resources during attach many times. We now do it once at the beginning. o Move setting the resource offsets to just after acquiring the ports in attach. o Move ax88x90 code to the end of the file, just after the dl100xx specific code. o Rename ed_pccard_Linksys to ed_pccard_dl100xx to reflect the underlying chipset. o Pass the ed_product structure into ed_pccard_{dl100xx,ax88x90} and have those routines test the flags to see if this card should be probed in that way. o transition from ed_probe_Novell to ed_probe_Novell_generic since we already have the resources setup. o Move use of ed_probe_Novell_generic into ed_pccard_dl100xx to be more consistant with ax88x90 case. o simplify the code where we probe for the chipsets
2005-09-07 03:30:58 +00:00
bad:
ed_release_resources(dev);
return (error);
- Fixed the conversion to bus_space interface. - Added PC-98 Cbus devices support. The original patch is submitted by chi@bd.mbn.or.jp (Chiharu Shibata) - Removed old ed driver.
2000-08-17 12:15:45 +00:00
}
MFPAO: Add support for AX88190, equipped in MELCO LPC3-TX.
2000-07-23 15:15:43 +00:00
static int
ed_pccard_memwrite(device_t dev, off_t offset, u_char byte)
{
Allocate memory resource to access the attribute memory of a card. Reviewed by: imp
2000-09-25 06:01:42 +00:00
int cis_rid;
struct resource *cis;
MFPAO: Add support for AX88190, equipped in MELCO LPC3-TX.
2000-07-23 15:15:43 +00:00
Allocate memory resource to access the attribute memory of a card. Reviewed by: imp
2000-09-25 06:01:42 +00:00
cis_rid = 0;
Minor newbus/style(9) cleanups. o Move the ax88190 code to its own function. o Move all device_method_t, driver_t and DRIVER_MODULE definitions to the end of files. o Wrap a few lines > 80 characters. o Use the same devclass for all ed drivers. This allows machines with multiple types of cards to have their cards numbered correctly. Before, you could wind up with two ed0's. o Protect if_edvar.h from multiple includes because I was there.
2000-12-26 06:38:04 +00:00
cis = bus_alloc_resource(dev, SYS_RES_MEMORY, &cis_rid, 0, ~0,
4 << 10, RF_ACTIVE | RF_SHAREABLE);
Allocate memory resource to access the attribute memory of a card. Reviewed by: imp
2000-09-25 06:01:42 +00:00
if (cis == NULL)
return (ENXIO);
Minor newbus/style(9) cleanups. o Move the ax88190 code to its own function. o Move all device_method_t, driver_t and DRIVER_MODULE definitions to the end of files. o Wrap a few lines > 80 characters. o Use the same devclass for all ed drivers. This allows machines with multiple types of cards to have their cards numbered correctly. Before, you could wind up with two ed0's. o Protect if_edvar.h from multiple includes because I was there.
2000-12-26 06:38:04 +00:00
CARD_SET_RES_FLAGS(device_get_parent(dev), dev, SYS_RES_MEMORY,
cis_rid, PCCARD_A_MEM_ATTR);
Allocate memory resource to access the attribute memory of a card. Reviewed by: imp
2000-09-25 06:01:42 +00:00
Minor newbus/style(9) cleanups. o Move the ax88190 code to its own function. o Move all device_method_t, driver_t and DRIVER_MODULE definitions to the end of files. o Wrap a few lines > 80 characters. o Use the same devclass for all ed drivers. This allows machines with multiple types of cards to have their cards numbered correctly. Before, you could wind up with two ed0's. o Protect if_edvar.h from multiple includes because I was there.
2000-12-26 06:38:04 +00:00
bus_space_write_1(rman_get_bustag(cis), rman_get_bushandle(cis),
offset, byte);
Allocate memory resource to access the attribute memory of a card. Reviewed by: imp
2000-09-25 06:01:42 +00:00
Update the ed driver to probe and attach under a NEWCARD kernel (I was using a cardbus based system with pccbb providing the pcic interface). Something isn't quite right.. when the driver allocates and activates its resources, the IO space that was requested reads as all zeros (versus the original 0xff's as it normally is when there is no device responding). Also, deactivate the resources before releasing them. OLDCARD doesn't seem to care but NEWCARD/CARDBUS get rather unhappy if you release a resource that hasn't been deactivated yet. Make pcic_p.c only compile with oldcard kernels.
2000-11-25 03:36:09 +00:00
bus_deactivate_resource(dev, SYS_RES_MEMORY, cis_rid, cis);
Allocate memory resource to access the attribute memory of a card. Reviewed by: imp
2000-09-25 06:01:42 +00:00
bus_release_resource(dev, SYS_RES_MEMORY, cis_rid, cis);
return (0);
MFPAO: Add support for AX88190, equipped in MELCO LPC3-TX.
2000-07-23 15:15:43 +00:00
}
Linksys Fast Ethernet PCCARD cards supported by the ed driver now require the addition of flag 0x80000 to their config line in pccard.conf(5). This flag is not optional. These Linksys cards will not be recognized without it. Reviewed by: imp, iwasaki
2000-12-18 15:28:53 +00:00
Really old PCMCIA cards stored the MAC address in the attribute memory at some offset. Unlike newer cards, the MAC address wasn't part of the CIS as a specific FUNCE. These older cards were having their MAC address show up as 0:2:4:6:8:a because that's what's in the ROM locations that would be there in a real ne2000. This patch allows one to specify the offset for the MAC address for these cards. Specify one for the IBM Ethernet II card, as it is one that has this problem. One shouldn't specify this unless the MAC address really isn't in the CIS at all. Side note: The novell probe likely shouldn't read the MAC address, and that should be moved to the bus specific attach routine(s), maybe as a convenience function in if_ed_novell.c. My IBM Ethernet II (aka Info Mover) now has a believable MAC address.
2005-02-12 00:45:22 +00:00
static int
ed_pccard_memread(device_t dev, off_t offset, u_char *byte)
{
int cis_rid;
struct resource *cis;
cis_rid = 0;
cis = bus_alloc_resource(dev, SYS_RES_MEMORY, &cis_rid, 0, ~0,
4 << 10, RF_ACTIVE | RF_SHAREABLE);
if (cis == NULL)
return (ENXIO);
CARD_SET_RES_FLAGS(device_get_parent(dev), dev, SYS_RES_MEMORY,
cis_rid, PCCARD_A_MEM_ATTR);
*byte = bus_space_read_1(rman_get_bustag(cis), rman_get_bushandle(cis),
offset);
bus_deactivate_resource(dev, SYS_RES_MEMORY, cis_rid, cis);
bus_release_resource(dev, SYS_RES_MEMORY, cis_rid, cis);
return (0);
}
Linksys Fast Ethernet PCCARD cards supported by the ed driver now require the addition of flag 0x80000 to their config line in pccard.conf(5). This flag is not optional. These Linksys cards will not be recognized without it. Reviewed by: imp, iwasaki
2000-12-18 15:28:53 +00:00
/*
* Probe the Ethernet MAC addrees for PCMCIA Linksys EtherFast 10/100
* and compatible cards (DL10019C Ethernet controller).
*
* Note: The PAO patches try to use more memory for the card, but that
* seems to fail for my card. A future optimization would add this back
* conditionally.
*/
static int
Minor reorg of pccard attach code to fix ax88x90 case broken after fixing dl100xx case. o We no longer acquire and release resources during attach many times. We now do it once at the beginning. o Move setting the resource offsets to just after acquiring the ports in attach. o Move ax88x90 code to the end of the file, just after the dl100xx specific code. o Rename ed_pccard_Linksys to ed_pccard_dl100xx to reflect the underlying chipset. o Pass the ed_product structure into ed_pccard_{dl100xx,ax88x90} and have those routines test the flags to see if this card should be probed in that way. o transition from ed_probe_Novell to ed_probe_Novell_generic since we already have the resources setup. o Move use of ed_probe_Novell_generic into ed_pccard_dl100xx to be more consistant with ax88x90 case. o simplify the code where we probe for the chipsets
2005-09-07 03:30:58 +00:00
ed_pccard_dl100xx(device_t dev, const struct ed_product *pp)
Linksys Fast Ethernet PCCARD cards supported by the ed driver now require the addition of flag 0x80000 to their config line in pccard.conf(5). This flag is not optional. These Linksys cards will not be recognized without it. Reviewed by: imp, iwasaki
2000-12-18 15:28:53 +00:00
{
struct ed_softc *sc = device_get_softc(dev);
u_char sum;
Rework how we deal with the DL10019 and DL10022 cards (as well as the AX88190 ones, but that one only minorly): o don't set flags in the match routine. They appear to be cleared when probe/attach is called. Before this change, they were always treated as a simple ne2000, which would fail to get the right NIC address. o Lookup device again in the probe routine and probe based on the cards that you see. o Detect and report the DL10022 seprately from the DL10019 cards. While I'm here: o remove a bad printf o change another bad printf to device_printf. o minor style(9) formatting tweaks. # note: a lot of OEM entries are in the ed_pccard_products such that we can # likely remove, or collapse, many of them. This makes all of my DL100xx cards at least probe the ethernet address correctly, which it wasn't doing before. I can't seem to locate my AX88xxx based cards, so those haven't been tested, but they were busted before the change so they can't be any worse now...
2005-01-26 05:21:12 +00:00
uint8_t id;
Minor reorg of pccard attach code to fix ax88x90 case broken after fixing dl100xx case. o We no longer acquire and release resources during attach many times. We now do it once at the beginning. o Move setting the resource offsets to just after acquiring the ports in attach. o Move ax88x90 code to the end of the file, just after the dl100xx specific code. o Rename ed_pccard_Linksys to ed_pccard_dl100xx to reflect the underlying chipset. o Pass the ed_product structure into ed_pccard_{dl100xx,ax88x90} and have those routines test the flags to see if this card should be probed in that way. o transition from ed_probe_Novell to ed_probe_Novell_generic since we already have the resources setup. o Move use of ed_probe_Novell_generic into ed_pccard_dl100xx to be more consistant with ax88x90 case. o simplify the code where we probe for the chipsets
2005-09-07 03:30:58 +00:00
int i, error;
if (!pp->flags & NE2000DVF_DL100XX)
return (ENXIO);
error = ed_probe_Novell_generic(dev, device_get_flags(dev));
if (error != 0)
return (error);
Linksys Fast Ethernet PCCARD cards supported by the ed driver now require the addition of flag 0x80000 to their config line in pccard.conf(5). This flag is not optional. These Linksys cards will not be recognized without it. Reviewed by: imp, iwasaki
2000-12-18 15:28:53 +00:00
/*
* Linksys registers(offset from ASIC base)
*
* 0x04-0x09 : Physical Address Register 0-5 (PAR0-PAR5)
* 0x0A : Card ID Register (CIR)
* 0x0B : Check Sum Register (SR)
*/
for (sum = 0, i = 0x04; i < 0x0c; i++)
sum += ed_asic_inb(sc, i);
if (sum != 0xff)
Rework how we deal with the DL10019 and DL10022 cards (as well as the AX88190 ones, but that one only minorly): o don't set flags in the match routine. They appear to be cleared when probe/attach is called. Before this change, they were always treated as a simple ne2000, which would fail to get the right NIC address. o Lookup device again in the probe routine and probe based on the cards that you see. o Detect and report the DL10022 seprately from the DL10019 cards. While I'm here: o remove a bad printf o change another bad printf to device_printf. o minor style(9) formatting tweaks. # note: a lot of OEM entries are in the ed_pccard_products such that we can # likely remove, or collapse, many of them. This makes all of my DL100xx cards at least probe the ethernet address correctly, which it wasn't doing before. I can't seem to locate my AX88xxx based cards, so those haven't been tested, but they were busted before the change so they can't be any worse now...
2005-01-26 05:21:12 +00:00
return (ENXIO); /* invalid DL10019C */
for (i = 0; i < ETHER_ADDR_LEN; i++)
Stop embedding struct ifnet at the top of driver softcs. Instead the struct ifnet or the layer 2 common structure it was embedded in have been replaced with a struct ifnet pointer to be filled by a call to the new function, if_alloc(). The layer 2 common structure is also allocated via if_alloc() based on the interface type. It is hung off the new struct ifnet member, if_l2com. This change removes the size of these structures from the kernel ABI and will allow us to better manage them as interfaces come and go. Other changes of note: - Struct arpcom is no longer referenced in normal interface code. Instead the Ethernet address is accessed via the IFP2ENADDR() macro. To enforce this ac_enaddr has been renamed to _ac_enaddr. - The second argument to ether_ifattach is now always the mac address from driver private storage rather than sometimes being ac_enaddr. Reviewed by: sobomax, sam
2005-06-10 16:49:24 +00:00
sc->enaddr[i] = ed_asic_inb(sc, 0x04 + i);
Linksys Fast Ethernet PCCARD cards supported by the ed driver now require the addition of flag 0x80000 to their config line in pccard.conf(5). This flag is not optional. These Linksys cards will not be recognized without it. Reviewed by: imp, iwasaki
2000-12-18 15:28:53 +00:00
ed_nic_outb(sc, ED_P0_DCR, ED_DCR_WTS | ED_DCR_FT1 | ED_DCR_LS);
Rework how we deal with the DL10019 and DL10022 cards (as well as the AX88190 ones, but that one only minorly): o don't set flags in the match routine. They appear to be cleared when probe/attach is called. Before this change, they were always treated as a simple ne2000, which would fail to get the right NIC address. o Lookup device again in the probe routine and probe based on the cards that you see. o Detect and report the DL10022 seprately from the DL10019 cards. While I'm here: o remove a bad printf o change another bad printf to device_printf. o minor style(9) formatting tweaks. # note: a lot of OEM entries are in the ed_pccard_products such that we can # likely remove, or collapse, many of them. This makes all of my DL100xx cards at least probe the ethernet address correctly, which it wasn't doing before. I can't seem to locate my AX88xxx based cards, so those haven't been tested, but they were busted before the change so they can't be any worse now...
2005-01-26 05:21:12 +00:00
id = ed_asic_inb(sc, 0xf);
Linksys Fast Ethernet PCCARD cards supported by the ed driver now require the addition of flag 0x80000 to their config line in pccard.conf(5). This flag is not optional. These Linksys cards will not be recognized without it. Reviewed by: imp, iwasaki
2000-12-18 15:28:53 +00:00
sc->isa16bit = 1;
Eliminate the ED_VENDOR_{PCCARD,LINKSYS} types. These cards really are NOVELL NE2000 with just a tiny quirk that's non vendor specific. Instead, use the chip_type of DL100XX instead. This is more inline with how the AX88190 support was added, and seems a little cleaner.
2005-02-11 06:02:27 +00:00
sc->vendor = ED_VENDOR_NOVELL;
Linksys Fast Ethernet PCCARD cards supported by the ed driver now require the addition of flag 0x80000 to their config line in pccard.conf(5). This flag is not optional. These Linksys cards will not be recognized without it. Reviewed by: imp, iwasaki
2000-12-18 15:28:53 +00:00
sc->type = ED_TYPE_NE2000;
Rework DL10019/DL10022 support. This tries to reset things in a more proper way, or at least the same way that NetBSD and Linux do things (I've been unable to obtain datasheets for these parts to know for sure). This has some marginal improvement in the DL10022 and DL10019 cards that I have. Also, report which type, exactly. # There's one or two ed cards that I have which still don't work, but I think # that's due to MII losage on the card that's not presently compensated # for in the MII drivers.
2005-02-14 06:54:06 +00:00
sc->chip_type = (id & 0x90) == 0x90 ?
ED_CHIP_TYPE_DL10022 : ED_CHIP_TYPE_DL10019;
Rework how we deal with the DL10019 and DL10022 cards (as well as the AX88190 ones, but that one only minorly): o don't set flags in the match routine. They appear to be cleared when probe/attach is called. Before this change, they were always treated as a simple ne2000, which would fail to get the right NIC address. o Lookup device again in the probe routine and probe based on the cards that you see. o Detect and report the DL10022 seprately from the DL10019 cards. While I'm here: o remove a bad printf o change another bad printf to device_printf. o minor style(9) formatting tweaks. # note: a lot of OEM entries are in the ed_pccard_products such that we can # likely remove, or collapse, many of them. This makes all of my DL100xx cards at least probe the ethernet address correctly, which it wasn't doing before. I can't seem to locate my AX88xxx based cards, so those haven't been tested, but they were busted before the change so they can't be any worse now...
2005-01-26 05:21:12 +00:00
sc->type_str = ((id & 0x90) == 0x90) ? "DL10022" : "DL10019";
return (0);
Linksys Fast Ethernet PCCARD cards supported by the ed driver now require the addition of flag 0x80000 to their config line in pccard.conf(5). This flag is not optional. These Linksys cards will not be recognized without it. Reviewed by: imp, iwasaki
2000-12-18 15:28:53 +00:00
}
Minor newbus/style(9) cleanups. o Move the ax88190 code to its own function. o Move all device_method_t, driver_t and DRIVER_MODULE definitions to the end of files. o Wrap a few lines > 80 characters. o Use the same devclass for all ed drivers. This allows machines with multiple types of cards to have their cards numbered correctly. Before, you could wind up with two ed0's. o Protect if_edvar.h from multiple includes because I was there.
2000-12-26 06:38:04 +00:00
Add an option ED_NO_MIIBUS, which causes the `ed' driver to be built without support for miibus PHYs. Most ed cards don't need miibus support, so it's useful to be able to avoid the bloat of all the mii devices for small fixed-purpose kernels.
2001-09-29 22:32:03 +00:00
#ifndef ED_NO_MIIBUS
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
/* MII bit-twiddling routines for cards using Dlink chipset */
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
#define DL10XX_MIISET(sc, x) ed_asic_outb(sc, ED_DL10XX_MIIBUS, \
ed_asic_inb(sc, ED_DL10XX_MIIBUS) | (x))
#define DL10XX_MIICLR(sc, x) ed_asic_outb(sc, ED_DL10XX_MIIBUS, \
ed_asic_inb(sc, ED_DL10XX_MIIBUS) & ~(x))
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
static void
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
ed_pccard_dl10xx_mii_reset(struct ed_softc *sc)
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
{
Rework DL10019/DL10022 support. This tries to reset things in a more proper way, or at least the same way that NetBSD and Linux do things (I've been unable to obtain datasheets for these parts to know for sure). This has some marginal improvement in the DL10022 and DL10019 cards that I have. Also, report which type, exactly. # There's one or two ed cards that I have which still don't work, but I think # that's due to MII losage on the card that's not presently compensated # for in the MII drivers.
2005-02-14 06:54:06 +00:00
if (sc->chip_type != ED_CHIP_TYPE_DL10022)
return;
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
ed_asic_outb(sc, ED_DL10XX_MIIBUS, ED_DL10XX_MII_RESET2);
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
DELAY(10);
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
ed_asic_outb(sc, ED_DL10XX_MIIBUS,
ED_DL10XX_MII_RESET2 | ED_DL10XX_MII_RESET1);
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
DELAY(10);
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
ed_asic_outb(sc, ED_DL10XX_MIIBUS, ED_DL10XX_MII_RESET2);
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
DELAY(10);
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
ed_asic_outb(sc, ED_DL10XX_MIIBUS,
ED_DL10XX_MII_RESET2 | ED_DL10XX_MII_RESET1);
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
DELAY(10);
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
ed_asic_outb(sc, ED_DL10XX_MIIBUS, 0);
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
}
static void
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
ed_pccard_dl10xx_mii_writebits(struct ed_softc *sc, u_int val, int nbits)
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
{
int i;
Rework DL10019/DL10022 support. This tries to reset things in a more proper way, or at least the same way that NetBSD and Linux do things (I've been unable to obtain datasheets for these parts to know for sure). This has some marginal improvement in the DL10022 and DL10019 cards that I have. Also, report which type, exactly. # There's one or two ed cards that I have which still don't work, but I think # that's due to MII losage on the card that's not presently compensated # for in the MII drivers.
2005-02-14 06:54:06 +00:00
if (sc->chip_type == ED_CHIP_TYPE_DL10022)
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
DL10XX_MIISET(sc, ED_DL10XX_MII_DIROUT_22);
Rework DL10019/DL10022 support. This tries to reset things in a more proper way, or at least the same way that NetBSD and Linux do things (I've been unable to obtain datasheets for these parts to know for sure). This has some marginal improvement in the DL10022 and DL10019 cards that I have. Also, report which type, exactly. # There's one or two ed cards that I have which still don't work, but I think # that's due to MII losage on the card that's not presently compensated # for in the MII drivers.
2005-02-14 06:54:06 +00:00
else
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
DL10XX_MIISET(sc, ED_DL10XX_MII_DIROUT_19);
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
for (i = nbits - 1; i >= 0; i--) {
if ((val >> i) & 1)
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
DL10XX_MIISET(sc, ED_DL10XX_MII_DATAOUT);
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
else
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
DL10XX_MIICLR(sc, ED_DL10XX_MII_DATAOUT);
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
DELAY(10);
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
DL10XX_MIISET(sc, ED_DL10XX_MII_CLK);
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
DELAY(10);
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
DL10XX_MIICLR(sc, ED_DL10XX_MII_CLK);
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
DELAY(10);
}
}
static u_int
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
ed_pccard_dl10xx_mii_readbits(struct ed_softc *sc, int nbits)
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
{
int i;
u_int val = 0;
Rework DL10019/DL10022 support. This tries to reset things in a more proper way, or at least the same way that NetBSD and Linux do things (I've been unable to obtain datasheets for these parts to know for sure). This has some marginal improvement in the DL10022 and DL10019 cards that I have. Also, report which type, exactly. # There's one or two ed cards that I have which still don't work, but I think # that's due to MII losage on the card that's not presently compensated # for in the MII drivers.
2005-02-14 06:54:06 +00:00
if (sc->chip_type == ED_CHIP_TYPE_DL10022)
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
DL10XX_MIICLR(sc, ED_DL10XX_MII_DIROUT_22);
Rework DL10019/DL10022 support. This tries to reset things in a more proper way, or at least the same way that NetBSD and Linux do things (I've been unable to obtain datasheets for these parts to know for sure). This has some marginal improvement in the DL10022 and DL10019 cards that I have. Also, report which type, exactly. # There's one or two ed cards that I have which still don't work, but I think # that's due to MII losage on the card that's not presently compensated # for in the MII drivers.
2005-02-14 06:54:06 +00:00
else
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
DL10XX_MIICLR(sc, ED_DL10XX_MII_DIROUT_19);
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
for (i = nbits - 1; i >= 0; i--) {
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
DL10XX_MIISET(sc, ED_DL10XX_MII_CLK);
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
DELAY(10);
val <<= 1;
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
if (ed_asic_inb(sc, ED_DL10XX_MIIBUS) & ED_DL10XX_MII_DATATIN)
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
val++;
Consistancy is the hobgoblin of small minds: o DLINK -> DL100XX
2005-09-07 03:34:24 +00:00
DL10XX_MIICLR(sc, ED_DL10XX_MII_CLK);
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
DELAY(10);
}
return val;
}
Add an option ED_NO_MIIBUS, which causes the `ed' driver to be built without support for miibus PHYs. Most ed cards don't need miibus support, so it's useful to be able to avoid the bloat of all the mii devices for small fixed-purpose kernels.
2001-09-29 22:32:03 +00:00
#endif
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
Minor reorg of pccard attach code to fix ax88x90 case broken after fixing dl100xx case. o We no longer acquire and release resources during attach many times. We now do it once at the beginning. o Move setting the resource offsets to just after acquiring the ports in attach. o Move ax88x90 code to the end of the file, just after the dl100xx specific code. o Rename ed_pccard_Linksys to ed_pccard_dl100xx to reflect the underlying chipset. o Pass the ed_product structure into ed_pccard_{dl100xx,ax88x90} and have those routines test the flags to see if this card should be probed in that way. o transition from ed_probe_Novell to ed_probe_Novell_generic since we already have the resources setup. o Move use of ed_probe_Novell_generic into ed_pccard_dl100xx to be more consistant with ax88x90 case. o simplify the code where we probe for the chipsets
2005-09-07 03:30:58 +00:00
static void
ed_pccard_ax88x90_geteprom(struct ed_softc *sc)
{
int prom[16],i;
u_char tmp;
struct {
unsigned char offset, value;
} pg_seq[] = {
/* Select Page0 */
{ED_P0_CR, ED_CR_RD2 | ED_CR_STP | ED_CR_PAGE_0},
{ED_P0_DCR, 0x01},
{ED_P0_RBCR0, 0x00}, /* Clear the count regs. */
{ED_P0_RBCR1, 0x00},
{ED_P0_IMR, 0x00}, /* Mask completion irq. */
{ED_P0_ISR, 0xff},
{ED_P0_RCR, ED_RCR_MON | ED_RCR_INTT}, /* Set To Monitor */
{ED_P0_TCR, ED_TCR_LB0}, /* loopback mode. */
{ED_P0_RBCR0, 32},
{ED_P0_RBCR1, 0x00},
{ED_P0_RSAR0, 0x00},
{ED_P0_RSAR1, 0x04},
{ED_P0_CR, ED_CR_RD0 | ED_CR_STA | ED_CR_PAGE_0},
};
/* Reset Card */
tmp = ed_asic_inb(sc, ED_NOVELL_RESET);
ed_asic_outb(sc, ED_NOVELL_RESET, tmp);
DELAY(5000);
ed_asic_outb(sc, ED_P0_CR, ED_CR_RD2 | ED_CR_STP | ED_CR_PAGE_0);
DELAY(5000);
/* Card Settings */
for (i = 0; i < sizeof(pg_seq) / sizeof(pg_seq[0]); i++)
ed_nic_outb(sc, pg_seq[i].offset, pg_seq[i].value);
/* Get Data */
for (i = 0; i < 16; i++)
prom[i] = ed_asic_inw(sc, 0);
/*
* Work around a bug I've seen on Linksys EC2T cards. On
* these cards, the node address is contained in the low order
* bytes of the prom, with the upper byte being 0. On other
* cards, the bytes are packed two per word. I'm unsure why
* this is the case, and why no other open source OS has a
* similar workaround. The Linksys EC2T card is still extremely
* popular on E-Bay, fetching way more than any other 10Mbps
* only card. I might be able to test to see if prom[7] and
* prom[15] == 0x5757, since that appears to be a reliable
* test. On the EC2T cards, I get 0x0057 in prom[14,15] instead.
*/
for (i = 0; i < ETHER_ADDR_LEN; i++)
if (prom[i] & 0xff00)
break;
if (i == ETHER_ADDR_LEN) {
for (i = 0; i < ETHER_ADDR_LEN; i++)
sc->enaddr[i] = prom[i] & 0xff;
} else {
for (i = 0; i < ETHER_ADDR_LEN; i += 2) {
sc->enaddr[i] = prom[i / 2] & 0xff;
sc->enaddr[i + 1] = (prom[i / 2] >> 8) & 0xff;
}
}
}
/*
* Special setup for AX88[17]90
*/
static int
ed_pccard_ax88x90(device_t dev, const struct ed_product *pp)
{
int error;
int iobase;
char *ts;
struct ed_softc *sc = device_get_softc(dev);
if (!pp->flags & NE2000DVF_AX88X90)
return (ENXIO);
/* XXX
* Set Attribute Memory IOBASE Register. Is this a deficiency in
* the PC Card layer, or an ax88190 specific issue? The card
* definitely doesn't work without it.
*/
iobase = rman_get_start(sc->port_res);
ed_pccard_memwrite(dev, ED_AX88190_IOBASE0, iobase & 0xff);
ed_pccard_memwrite(dev, ED_AX88190_IOBASE1, (iobase >> 8) & 0xff);
ts = "AX88190";
if (ed_asic_inb(sc, ED_ASIX_TEST) != 0) {
ed_pccard_memwrite(dev, ED_AX88790_CSR, ED_AX88790_CSR_PWRDWN);
ts = "AX88790";
}
sc->chip_type = ED_CHIP_TYPE_AX88190;
ed_pccard_ax88x90_geteprom(sc);
error = ed_probe_Novell_generic(dev, device_get_flags(dev));
if (error == 0) {
sc->vendor = ED_VENDOR_NOVELL;
sc->type = ED_TYPE_NE2000;
sc->chip_type = ED_CHIP_TYPE_AX88190;
sc->type_str = ts;
}
return (error);
}
Minor newbus/style(9) cleanups. o Move the ax88190 code to its own function. o Move all device_method_t, driver_t and DRIVER_MODULE definitions to the end of files. o Wrap a few lines > 80 characters. o Use the same devclass for all ed drivers. This allows machines with multiple types of cards to have their cards numbered correctly. Before, you could wind up with two ed0's. o Protect if_edvar.h from multiple includes because I was there.
2000-12-26 06:38:04 +00:00
static device_method_t ed_pccard_methods[] = {
/* Device interface */
Finish conversion to bus space and make ed MP safe. o Lock ed o Fix extra newline in probe messages o Eliminate gone. o Make detach less-racy. o Eliminate spl* o Switch from timeout/untimeout to callout interface. o Read/write card memory using bus_space calls. o generalize readmem so that we don't need ifs in the code. o Fix memory stuff to be consistant. o Remove OLDCARD compat stuff. o Mark interrupt as MPSAFE. # sic, hpp not tested at all # ISA and PCI attachments lightly tested
2005-08-28 23:56:25 +00:00
DEVMETHOD(device_probe, ed_pccard_probe),
DEVMETHOD(device_attach, ed_pccard_attach),
Make it possible to unload ed. Move the ed_pccard_detach routine to if_ed and rename it to ed_detach(). Tell other busses to use this routine for detach. Since I don't actually have any non-pccard ed hardware I can test with, I've only tested with my pccards. More improvements in this area likely are possible. Prodded by: rwatson
2005-02-08 05:45:35 +00:00
DEVMETHOD(device_detach, ed_detach),
Minor newbus/style(9) cleanups. o Move the ax88190 code to its own function. o Move all device_method_t, driver_t and DRIVER_MODULE definitions to the end of files. o Wrap a few lines > 80 characters. o Use the same devclass for all ed drivers. This allows machines with multiple types of cards to have their cards numbered correctly. Before, you could wind up with two ed0's. o Protect if_edvar.h from multiple includes because I was there.
2000-12-26 06:38:04 +00:00
Add an option ED_NO_MIIBUS, which causes the `ed' driver to be built without support for miibus PHYs. Most ed cards don't need miibus support, so it's useful to be able to avoid the bloat of all the mii devices for small fixed-purpose kernels.
2001-09-29 22:32:03 +00:00
#ifndef ED_NO_MIIBUS
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
/* Bus interface */
DEVMETHOD(bus_child_detached, ed_child_detached),
/* MII interface */
DEVMETHOD(miibus_readreg, ed_miibus_readreg),
DEVMETHOD(miibus_writereg, ed_miibus_writereg),
Add an option ED_NO_MIIBUS, which causes the `ed' driver to be built without support for miibus PHYs. Most ed cards don't need miibus support, so it's useful to be able to avoid the bloat of all the mii devices for small fixed-purpose kernels.
2001-09-29 22:32:03 +00:00
#endif
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
Minor newbus/style(9) cleanups. o Move the ax88190 code to its own function. o Move all device_method_t, driver_t and DRIVER_MODULE definitions to the end of files. o Wrap a few lines > 80 characters. o Use the same devclass for all ed drivers. This allows machines with multiple types of cards to have their cards numbered correctly. Before, you could wind up with two ed0's. o Protect if_edvar.h from multiple includes because I was there.
2000-12-26 06:38:04 +00:00
{ 0, 0 }
};
static driver_t ed_pccard_driver = {
"ed",
ed_pccard_methods,
sizeof(struct ed_softc)
};
- Express hard dependencies on bus (pci, isa, pccard) and network layer (ether). - Don't abuse module names to facilitate ifconfig module loading; such abuse isn't really needed. (And if we do need type information associated with a module then we should make it explicit and not use hacks.)
2003-04-15 06:37:30 +00:00
DRIVER_MODULE(ed, pccard, ed_pccard_driver, ed_devclass, 0, 0);
Add an option ED_NO_MIIBUS, which causes the `ed' driver to be built without support for miibus PHYs. Most ed cards don't need miibus support, so it's useful to be able to avoid the bloat of all the mii devices for small fixed-purpose kernels.
2001-09-29 22:32:03 +00:00
#ifndef ED_NO_MIIBUS
Add support for Dlink DL10022 to the ed driver. This is a mii part bolted to a ne-2000 chip. This is necessary for the NetGear FA-410TX and other cards. This also requires you add mii to your kernel if you have an ed driver configured. This code will result in a couple of timeout messages for ed on the impacted cards. Additional work will be needed, but this does work right now, and many people need these cards. Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
2001-03-03 08:31:37 +00:00
DRIVER_MODULE(miibus, ed, miibus_driver, miibus_devclass, 0, 0);
Add an option ED_NO_MIIBUS, which causes the `ed' driver to be built without support for miibus PHYs. Most ed cards don't need miibus support, so it's useful to be able to avoid the bloat of all the mii devices for small fixed-purpose kernels.
2001-09-29 22:32:03 +00:00
#endif
Reference in New Issue Copy Permalink