Use this page to discuss PCMCIA network adapter issues. Before posting, be sure you've checked any relevant sections of the PCMCIA-HOWTO, such as: Particularly useful bits of information to post include your laptop's make and model; the type(s) of card(s) you are using; and, of course, any relevant system log messages.

Messages Inline: 0 All Outline: 1 2 All

Question eth0: interrupt(s) dropped! messages

Date: 1999, Nov 01
From: Rick Lerner rlerner

PROBLEM:  My network PCCard appears to work fine, but /var/log/messages
contains hundreds of lines containing the following error message:

	 Nov  1 15:46:06 localhost kernel: eth0: interrupt(s) dropped! 

Any ideas what is wrong?

I recently updated my laptop from RedHat 5.1 to 6.1.  To complete the
upgrade I needed to add an ifcfg-eth0 file with the following contents:

/etc/sysconfig/network-scripts/ifcfg-eth0:
DEVICE=eth0
BOOTPROTO=dhcp
ONBOOT=no

I also added links to /etc/rc.d/init.d/pcmcia in /etc/rc.d/rc[345].d

Otherwise, the pcmcia configuration files are the ones installed by RedHat.

Thanks for any information,
  Rick Lerner
  (rick@aai.com ) Amerinex Applied Imaging, Inc.


--------------------------------------------------
DETAILS:
--------------------------------------------------

System Type:	Ergo Bigscreen4 laptop
Processor:	Pentium II, 266MHz
RAM:		96MB
Linux:		RedHat 6.1
uname -rv:	2.2.12-20 #1 Mon Sep 27 10:40:35 EDT 1999
PCMCIA Version:	cardctl version 3.0.14
Probe output:	PCI bridge probe: O2Micro OZ6836 found, 2 sockets.

cardctl status
	Socket 0:
	  5V 16-bit card present
	  Function 0: ready
	Socket 1:
	  no card

cardctl ident 
	Socket 0:
	  product info: "3Com Corporation", "3C589", "TP/BNC LAN Card Ver. 2a", "000002"
	  manfid: 0x0101, 0x0589
	  function: 6 (network)
	Socket 1:
	  no product info available


----------------------------------------------------------------------
/var/log/messages excerpt generated by
	/etc/rc.d/init.d/pcmcia start (after a stop whose output is not shown)
----------------------------------------------------------------------

Nov  1 17:50:10 localhost kernel: Linux PCMCIA Card Services 3.0.14 
Nov  1 17:50:10 localhost kernel:   kernel build: 2.2.12-20 #1 Mon Sep 27 10:40:35 EDT 1999 
Nov  1 17:50:10 localhost kernel:   options:  [pci] [cardbus] [apm] 
Nov  1 17:50:10 localhost kernel: Intel PCIC probe:  
Nov  1 17:50:10 localhost kernel:   O2Micro OZ6836 PCI-to-CardBus at bus 0 slot 3, mem 0x28000000, 2 sockets 
Nov  1 17:50:10 localhost kernel:     host opts [0]: [a 00] [b 00] [c 00] [d 83] [mhpg 18] [fifo 08] [e 0a] [no pci irq] [lat 168/176] [bus 32/34] 
Nov  1 17:50:10 localhost kernel:     host opts [1]: [a 00] [b 00] [c 50] [d 83] [mhpg 18] [fifo 08] [e 0a] [no pci irq] [lat 168/176] [bus 35/37] 
Nov  1 17:50:10 localhost kernel:     ISA irqs (scanned) = 10 polling interval = 1000 ms 
Nov  1 17:50:10 localhost cardmgr[1507]: starting, version is 3.0.14
Nov  1 17:50:10 localhost cardmgr[1507]: watching 2 sockets
Nov  1 17:50:10 localhost kernel: cs: IO port probe 0x1000-0x17ff: excluding 0x1000-0x104f 
Nov  1 17:50:10 localhost kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x100-0x107 0x200-0x207 0x220-0x22f 0x300-0x307 0x378-0x37f 0x388-0x38f 0x4d0-0x4d7 
Nov  1 17:50:10 localhost kernel: cs: IO port probe 0x0a00-0x0aff: clean. 
Nov  1 17:50:11 localhost cardmgr[1507]: initializing socket 0
Nov  1 17:50:11 localhost kernel: cs: memory probe 0xa0000000-0xa0ffffff: excluding 0xa0000000-0xa00fffff 
Nov  1 17:50:11 localhost cardmgr[1507]: socket 0: 3Com 3c589 Ethernet
Nov  1 17:50:11 localhost cardmgr[1507]: executing: 'insmod /lib/modules/2.2.12-20/pcmcia/3c589_cs.o'
Nov  1 17:50:11 localhost kernel: eth0: 3Com 3c589, io 0x310, irq 10, hw_addr 00:20:AF:DF:F6:1B 
Nov  1 17:50:11 localhost kernel:   8K FIFO split 5:3 Rx:Tx, auto xcvr 
Nov  1 17:50:11 localhost cardmgr[1507]: executing: './network start eth0'
Nov  1 17:50:12 localhost pumpd[1537]: starting at Mon Nov  1 17:50:12 1999  
Nov  1 17:50:13 localhost kernel: eth0: interrupt(s) dropped! 
Nov  1 17:50:17 localhost pumpd[1537]: configured interface eth0
Nov  1 17:50:17 localhost cardmgr[1507]: + Determining IP information for eth0... done.
Nov  1 17:50:18 localhost kernel: eth0: flipped to 10baseT 
Nov  1 17:50:24 localhost kernel: eth0: interrupt(s) dropped! 
Nov  1 17:50:56 localhost last message repeated 4 times
Nov  1 17:52:03 localhost last message repeated 9 times
Nov  1 17:53:04 localhost last message repeated 10 times

1. None Re: Question: eth0: interrupt(s) dropped! messages by David Hinds, 1999, Nov 01
(_ Ok do_scan did the trick by Rick Lerner, 1999, Nov 02

Question Billionton LNA-100B

Date: 1999, Nov 04
From: Markus Troebensberger troebs

Is a driver available for the Billionton LNA-100B ?

Have anyone experience with this card ?

If not:

I will try to write a driver. Have anyone contact to Billionton ?

1. None It is not a supported card, but... by David Hinds, 1999, Nov 04
1. Sad I have testet ... by Markus Troebensberger, 1999, Nov 05
2. Question It nearly works by Markus Troebensberger, 1999, Nov 08
1. None So close yet so far? by David Hinds, 1999, Nov 20
2. News first LNA-100B success by Miroslaw Kwasniak, 2000, Jun 30

Question Kingmax? PCMCIA 10/100 adapter

Date: 1999, Nov 06
From: Nathan Hartwell MageX

I just got a laptop. With this laptop, I ordered a 10/100 card. It's a generic card, but from the checking I've done, it seems to be a Kingmax Technologies product. Anyway, it's based on the ASIX AX88190 chip (I think). This assumption is based on information that was in a Linux driver archive that shipped with the card. Unfortunately, this driver was written for older versions of pcmcia-cs (3.0.x it seems) and Linux kernel (2.0.x). According to documention from ASIX, the AX88190 is register compatible with the 8390. Ok, so I managed to get the system to see the card, and actually work with it. However, I get tons of kernel msgs regarding interrupt routine reentrance. Since these msgs don't just go to /var/log/messages, but also get dumped on the console, I find the messages more than annoying. Anyone capable of helping here?

1. Sad These cards are not really NE2000 compatible by David Hinds, 1999, Nov 23
2. Idea AX88190 (Kingmax, PCMIA 10/100 Ethernet Card) on kernel 2.2.x by Thomas Pasch, 2000, Jan 19

None Compex/ReadyLINK 10Mbps Ethernet card: kernel `Oops' on insert

Date: 1999, Nov 08
From: David dbutler

Hi,

When I inserted a Compex/ReadyLINK 10Mbps Ethernet card
 in to a   Sony Vaio  notebook 
(Kernel 2.2.12, PCMCIA package 3.0.14)  (same problem with a 
Winbook XL - Linux Kernel 2.2.13, PCMCIA package 3.0.14), 
the kernel gave the following error mesage 

'Unable to handle kernel NULL pointer dereference ...'

Also, the kernel reports an `Oops' caused by the card manager.

I'd be happy to hear any advice.

thanks in advance,

david

1. None Re: Compex/ReadyLINK 10Mbps Ethernet card: kernel `Oops' on insert by David Hinds, 1999, Nov 08

Sad upgrading PCMCIA 3.1.1 to 3.1.2 or 3.1.3 breaks certain network adapters

Date: 1999, Nov 10
From: Patrick Hildenbrand patrick

Hi, it seams that with kernel version 2.2.13 going from 3.1.1 to 3.1.2 the following adapters will no longer work correctly:
  • IBM Ethernet Adapter II (pcnet_cs)
    various error messags on high network utilization, UDP transport not reliable (ex. host will sometimes not work, doing a tcpdump, I could see that the data gets send out, the answer returned but the system will not show the response. TCP connections are slow but work (specifiying the ip-adress).
  • Level One EPC-0100TB (pcnet_cs)
    no conectivity at all, the card gets detected and configured, eth0 is available but nothing (ping, tcp, udp) works.

going from 3.1.1 to 3.1.3 the same applies as above and in addition the IBM Etherjet cardbus adapter (tulip_cb) will no longer work. Same symptoms as for the Level One adapter. I did not try other adapters so far besides a 4lan EP100 ethernet adapter (also pcnet_cs) which just works as it has been.

The configuration had not been changed between the different test. Reinstalling the old pcmcia version did in all cases cure the problem. Hardware are IBM Thinkpad 760 and 770. TokenRing is not affected.

Anybody any hints ?

I will try to dig a bit more as I have time.

1. Ok Fixed by Patrick Hildenbrand, 1999, Dec 06

Question Keep receiving "register_netdev( ) failed: try again

Date: 1999, Nov 13
From: Al Charpentier Charpy

I have a WinBook XP5, Pentium 75, 16mb, 720mb. Slackware Linux, kernel v2.2.6 I am attempting to use a 3COM 3C589D PCMCIA Card.

Here's the output:

Nov 13 08:55:24 sys0 kernel: Linux PCMCIA Card Services 3.0.9 
Nov 13 08:55:24 sys0 kernel:   kernel build: 2.2.6 #20 Sat Apr 17 23:17:12 CDT 1999 
Nov 13 08:55:24 sys0 kernel:   options:  [pci] [cardbus] 
Nov 13 08:55:24 sys0 kernel: Intel PCIC probe:  
Nov 13 08:55:24 sys0 kernel:   Vadem VG-469 ISA-to-PCMCIA at port 0x3e0 ofs 0x00, 2 sockets 
Nov 13 08:55:24 sys0 kernel:     host opts [0]: none 
Nov 13 08:55:24 sys0 kernel:     host opts [1]: none 
Nov 13 08:55:24 sys0 kernel:     ISA irqs (default) = 3,4,5,7,9,10,11,12,15 polling interval = 1000 ms 
Nov 13 08:55:24 sys0 kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x110-0x117 0x268-0x26f 0x400-0x40f 0x420-0x427 0x440-0x447 0x460-0x467 0x480-0x497 0x4a0-0x4a7 0x4c0-0x4d7 0x4f0-0x4f7 
Nov 13 08:55:24 sys0 kernel: cs: memory probe 0x0d0000-0x0dffff: clean. 
Nov 13 08:55:24 sys0 kernel: 3c589_cs: register_netdev() failed 
Nov 13 09:01:26 sys0 login[111]: ROOT LOGIN on `tty1'
Nov 13 09:01:35 sys0 cardmgr[53]: shutting down socket 0
Nov 13 09:01:35 sys0 cardmgr[53]: executing: 'rmmod 3c589_cs'
Nov 13 09:01:39 sys0 cardmgr[53]: initializing socket 0
Nov 13 09:01:39 sys0 cardmgr[53]: socket 0: 3Com 3c589D Ethernet
Nov 13 09:01:39 sys0 cardmgr[53]: executing: 'insmod /lib/modules/2.2.6/pcmcia/3c589_cs.o'
Nov 13 09:01:39 sys0 kernel: 3c589_cs: register_netdev() failed 
Nov 13 09:01:40 sys0 cardmgr[53]: get dev info on socket 0 failed: Try again
Nov 13 09:01:47 sys0 cardmgr[53]: shutting down socket 0
Nov 13 09:01:47 sys0 cardmgr[53]: executing: 'rmmod 3c589_cs'

As you can see, PCMCIA services work fine (I have a 56k modem card that works perfectly). The 3COM is detected, however when the following is ran: insmod /lib/modules/2.2.6/pcmcia/c3589_cs.o I get: 3c589_cs: register_netdev() failed get dev info on socket 0 failed: Try again

I have attempted to exclude different io port and memory settings. The only other thing I have received is:

cs: IO port probe 0x0300-0x03ff: clean. 
cs: memory probe 0x0d0000-0x0dffff: clean. 
3c589_cs: RequestIRQ: Resource in use

I have even placed: PCIC_OPTS="do_scan=0" to disable IRQ scanning.

The card works well in other laptops running Win95/98.

Any suggestions? Thanx!

1. Question Nobody has any input? by Al Charpentier, 1999, Nov 20
1. None I feel your pain.... by Nathan Hartwell, 1999, Nov 21
2. Note This is an unusual message by David Hinds, 1999, Nov 23
1. None register_netdev( ) failed: try again by Al Charpentier, 1999, Nov 25

Question D-Link DFE-660 Problems

Date: 1999, Nov 15
From: <mike@enoch.org>

 I made the mistake of buying a D-Link DFE-660 rather than the
 650, which I should have purchased.  The 660 is listed as
 experimental, and in my case, it doesn't work.  The card is
 identified, but it does not work.  I can ping the IP of eth0,
 but not of my default router.  Swapping to a 3Com card I have
 laying around works just fine (I'm using it now).  So, here's
 the specifics:
IBM Thinkpad 1552
D-Link DFE-660
Kernel 2.2.13 (International version)
Linux PCMCIA Card Services 3.1.4
  kernel build: 2.2.13 #1 Sun Nov 14 23:59:00 EST 1999
  options:  [pci] [cardbus] [apm]
Intel PCIC probe: 
  O2Micro OZ6832/OZ6833 PCI-to-CardBus at bus 0 slot 19, mem 0x68000000, 2 sockets
    host opts [0]: [a 44] [b a1] [c 20] [d 06] [mhpg 18] [fifo 00] [e 28] [no pci irq] [lat 32/176] [bus 32/34]
    host opts [1]: [a 44] [b ad] [c 20] [d 06] [mhpg 18] [fifo 00] [e 28] [no pci irq] [lat 32/176] [bus 35/37]
    ISA irqs (default) = 3,4,7,11,12 polling interval = 1000 ms
cs: IO port probe 0x1000-0x17ff: clean.
cs: IO port probe 0x0100-0x04ff: excluding 0x3b8-0x3df 0x408-0x40f 0x480-0x48f 0x4d0-0x4d7
cs: IO port probe 0x0a00-0x0aff: clean.
cs: memory probe 0xa0000000-0xa0ffffff: clean.

And when I insert the DFE-660:

Nov 15 00:46:15 mib kernel: cs: cb_alloc(bus 35): vendor 0x1011, device 0x0019 
Nov 15 00:46:15 mib cardmgr[351]: initializing socket 1
Nov 15 00:46:15 mib cardmgr[351]: socket 1: Linksys EtherFast 10/100
Nov 15 00:46:15 mib cardmgr[351]: executing: 'insmod /lib/modules/2.2.13/pcmcia/cb_enabler.o'
Nov 15 00:46:15 mib cardmgr[351]: executing: 'insmod /lib/modules/2.2.13/pcmcia/tulip_cb.o'
Nov 15 00:46:15 mib kernel: cs: cb_config(bus 35) 
Nov 15 00:46:15 mib kernel:   fn 0 bar 1: io 0xa00-0xa7f 
Nov 15 00:46:15 mib kernel:   fn 0 bar 2: mem 0xa0081000-0xa00813ff 
Nov 15 00:46:15 mib kernel:   fn 0 rom: mem 0xa0041000-0xa0080fff 
Nov 15 00:46:15 mib kernel: tulip_attach(bus 35, function 0) 
Nov 15 00:46:15 mib kernel: tulip.c:v0.91 4/14/99 becker@cesdis.gsfc.nasa.gov (modified by danilo@cs.uni-magdeburg.de for XIRCOM CBE, fixed by Doug Ledford) 
Nov 15 00:46:15 mib kernel: eth0: Digital DS21143 Tulip rev 65 at 0xa00, 00:80:C8:BA:DF:BD, IRQ 3. 
Nov 15 00:46:15 mib kernel: eth0:  EEPROM default media type Autosense. 
Nov 15 00:46:15 mib kernel: eth0:  Index #0 - Media MII (#11) described by a 21142 MII PHY (3) block. 
Nov 15 00:46:15 mib kernel: eth0:  MII transceiver #0 config 3000 status 7809 advertising 01e1. 
Nov 15 00:46:15 mib cardmgr[351]: executing: './network start eth0'
Nov 15 00:46:21 mib kernel: eth0: Setting full-duplex based on MII#0 link partner capability of 4061. 
Nov 15 00:46:21 mib kernel: eth0: Tx hung, 9 vs. 0. 

It's the last line that seems to be the problem. I've seen 10 vs 0 and 7 vs 0, sometimes in sequence.

/sbin/cardctl ident
Socket 0:
  product info: "D-Link", "DFE-660", "CardBus Ethernet", "10/100"
  manfid: 0x0149, 0x0231
  function: 6 (network)

The card also didn't work with pcmcia-cs-3.0.14, which was why I tried 3.1.4.

So, um, is there anything I can do, short of buying a new card (I can't take this one back)? Any more debugging messages I can add?

Thanks, Mike

1. None Re: Question: D-Link DFE-660 Problems by David Hinds, 1999, Nov 15
(_ None It works! Great! by mike@enoch.org, 1999, Nov 16
(_ None They're documented in the man page by David Hinds, 1999, Nov 16
(_ None Re: They're documented in the man page by Mike Johnson, 1999, Nov 16

Question Destination Host Unreachable

Date: 1999, Nov 15
From: David Walden Walden

I have installed RedHat 6.1 on my Mitac 6020 with no problems except my Wisecom PC400 PCMCIA network card doesn't seem to be talking to the network. I am able to ping the card itself and when netstat -r is run the default gateway is shown, but I am still unable to ping anything on the network. I keep getting a response "Destination Host UnReachable". Does anyone have any suggestions?

1. None Network unreachable == routing problem by David Hinds, 1999, Nov 15
(_ None RE: Network unreachable == routing problem by David Walden, 1999, Nov 17
(_ None Another view on this problem by John Jervis, 2000, Jan 05

Question Intel EtherExpress PRO/100 Cardbus 32

Date: 1999, Nov 17
From: Mike Rudgyard MikeR

Hi,

I am trying to get the above card working on my Dell Latitude machine with the latest tulip driver and pcmcia-cs-3.1.4.

According to the supported cards, this should work. Indeed the
network comes up, and the light on the card comes on:

Nov 17 14:28:51 rock kernel: cs: cb_alloc(bus 35): vendor 0x1011, device 0x0019
Nov 17 14:28:51 rock cardmgr[360]: initializing socket 1
Nov 17 14:28:51 rock cardmgr[360]: socket 1: Intel EtherExpress PRO/100 CardBus Mobile Adapter32
Nov 17 14:28:51 rock cardmgr[360]: executing: 'insmod /lib/modules/2.2.12-20/pcmcia/cb_enabler.o'
Nov 17 14:28:51 rock cardmgr[360]: executing: 'insmod /lib/modules/2.2.12-20/pcmcia/tulip_cb.o'
Nov 17 14:28:51 rock kernel: cs: cb_config(bus 35)
Nov 17 14:28:51 rock kernel:   fn 0 bar 1: io 0x100-0x17f
Nov 17 14:28:51 rock kernel:   fn 0 bar 2: mem 0x60081000-0x6008107f
Nov 17 14:28:51 rock kernel:   fn 0 rom: mem 0x60041000-0x60080fff
Nov 17 14:28:51 rock kernel: tulip_attach(bus 35, function 0)
Nov 17 14:28:51 rock kernel: tulip.c:v0.91 4/14/99 becker@cesdis.gsfc.nasa.gov (modified by danilo@cs.uni-magdeburg.de for XIRCOM CBE, fixed by Doug Ledford)
Nov 17 14:28:51 rock kernel: eth0: Digital DS21143 Tulip rev 48 at 0x100, EEPROM not present, 00:4C:69:6E:75:79, IRQ 3.
Nov 17 14:28:51 rock cardmgr[360]: executing: './network start eth0'
Nov 17 14:28:52 rock modprobe: can't locate module block-major-22




As it points out, there is no EEPROM present in the card, which I presume is why I can't actually send or receive any
packets. If I use the tulip-diag tool, doing tulip-diag -ee
then I get:

ulip-diag.c:v1.19 10/2/99 Donald Becker (becker@cesdis.gsfc.nasa.gov)
Index #1: Found a Digital DS21143 Tulip adapter at 0x60081000.
 Port selection is 100mbps-SYM/PCS 100baseTx scrambler, full-duplex.
 Transmit started, Receive started, full-duplex.
  The Rx process state is 'Transferring Rx frame into memory'.
  The Tx process state is 'Closing Tx descriptor'.
  PCI bus error!: Unknown 7.
  The transmit unit is set to store-and-forward.
 Interrupt sources are pending!  CSR5 is ffffffff.
   Tx done indication.
   Tx complete indication.
   Tx out of buffers indication.
   Transmit Jabber indication.
   Link passed indication.
   Tx FIFO Underflow indication.
   Rx Done indication.
   Receiver out of buffers indication.
   Receiver stopped indication.
   Receiver jabber indication.
   Link changed indication.
   Timer expired indication.
   Link failed indication.
   PCI bus error indication.
   Early Rx indication.
  The NWay status register is ffffffff.
EEPROM size is 8.
WARNING: The EEPROM is missing or erased!
 This interface is missing the EEPROM.
  This is likely the non-primary interface on a multiport board.
EEPROM contents:
  ffff ffff ffff ffff ffff ffff ffff ffff
  ffff ffff ffff ffff ffff ffff ffff ffff
  ffff ffff ffff ffff ffff ffff ffff ffff
  ffff ffff ffff ffff ffff ffff ffff ffff
  ffff ffff ffff ffff ffff ffff ffff ffff
  ffff ffff ffff ffff ffff ffff ffff ffff
  ffff ffff ffff ffff ffff ffff ffff ffff
  ffff ffff ffff ffff ffff ffff ffff ffff
 ID block CRC 0xfa (vs. 0xff).
  Full contents CRC 0x6a15 (read as 0xffff).
  Internal autonegotiation state is 'Invalid state'.


Unfortunately, I can't find any details of how to program the 
EEPROM (although I can see that the above program can be used to do this...).

Has anyone any ideas ??

Thanks,

Mike
1. None The EEPROM isn't the problem by David Hinds, 1999, Nov 17
2. Question More info... by Mike Rudgyard, 1999, Nov 18
1. None Something to try... by David Hinds, 1999, Dec 30
(_ Ok Success !! by Mike Rudgyard, 2000, Jan 10
(_ Question I'd like your success by Kurt Atkinson, 2000, Feb 17
3. Sad Same thing happens with Intel PRO/100 LAN+Modem56 CardBus II card by Wayne Scott, 1999, Dec 07
1. None It's the chipset by John Kilpatrick, 2000, Apr 22

Question eth0 won't initialise on bootup.

Date: 1999, Nov 18
From: James Bramich jb

Hi,

I'm a real Linux beginner (oh god they say). I am running Red Hat 6.0 on an IBM Thinkpad 600e.

Redhat installed OK. I upgraded the kernel and card services as per the model specific doc's on the Linux on Laptops page and it seems to go OK.

However, "eth0" fails to initialise when the machine boots. I can go into the GNOME control panel and initialise it manually, but it won't do it autimatically.

Any ideas, I still haven't got the hang of the order in which everything loads up yet so I'm not sure if there are any config files or whatever I should be looking in.

Thanks for any ideas.

James.

1. None need more information by Nathan Hartwell, 1999, Nov 21
2. None Check the HOWTO... by David Hinds, 1999, Nov 22

Question Kingmax 10/100 Mbps cardbus ethernet card, help?

Date: 1999, Nov 26
From: Michael Kroes yodax

Hello,

I have a question about the kingmax 10/100Mbps cardbus ethernet adapter. I added the version info to the /etc/pcmcia/config and used the tulip_cs module.

Cardmanager now detects the card and it is configurable by ifconfig. But i can only ping computers they can also ping me, but i can't use client protocols like FTP, telnet, portscanners etc.

I had an old 10 mbps 16 bit card before and it worked fine, and stil does btw.

So perhaps someone can help me with this problem?

Sincerely yours

Michael Kroes

Question dynalink lc10 network can't initialise

Date: 1999, Nov 29
From: digit

Hi,

My dynalink LC10 pcmcia card works with windows. If checked the irq's and memory vallues and copied these ranges to linux

:/etc/pcmcia/config.opts

messages say everything works modules, irq mem but ./network start eth0 failes because eth0 doesn't exist.

I allso get twe high low beep on boot and after insert.

Using redhat 6.1 the card is in the supported database and is recognised as NE2000 compatible (messages sya this)

Is this card indeed supported? The irq 3 is really free in my case and the mem seems right to. When remomving the high memory range from config.options I get the right beeps but the card is then a mem0 device and nothing eth0 like.

Thsi is ll I know is there someone thaT KNOWS WHAT TO DO OR IS THIS CARD SIMPLY NOT SUPPORTED?

1. None The actual messages would help by David Hinds, 1999, Nov 29
(_ None messages for compaqua armada 1750 by digit@dds.nl, 1999, Nov 30
(_ None No real problems here by David Hinds, 1999, Nov 30
(_ None Thanks!! by digit@dds.nl, 1999, Dec 01

Question DFE-650 only 10 Mbit?

Date: 1999, Nov 29
From: tma

Hello, i have a rather strange problem with a D-LINK DFE-650 100TX card on an ACER Travelmate 332T.. the card itself works pretty good, but the maximum speed is ~ 1.18 mb/sec, which is exactly 10Mbit. the network/connection is about 6-8 mb/sec with win98 (spit!) using the same machine/card, so the problem seems to be the driver.. the faq lists this card as 10/100 & working, so i'm optimistic

dump of dmesg: (oh yes, and it's a somehow modified slackware 7.0)

Linux version 2.2.13 (root@flap) (gcc version 2.95.1 19990816 (release)) #2 Mon Nov 29 21:47:13 CET 1999
Detected 365819234 Hz processor.
Console: colour dummy device 80x25
Calibrating delay loop... 365.36 BogoMIPS
Memory: 62908k/65472k available (1060k kernel code, 412k reserved, 1024k data, 68k init)
VFS: Diskquotas version dquot_6.4.0 initialized
CPU: Intel Mobile Pentium II stepping 0a
Checking 386/387 coupling... OK, FPU using exception 16 error reporting.
Checking 'hlt' instruction... OK.
POSIX conformance testing by UNIFIX
mtrr: v1.35a (19990819) Richard Gooch (rgooch@atnf.csiro.au)
PCI: PCI BIOS revision 2.10 entry at 0xf0200
PCI: Using configuration type 1
PCI: Probing PCI hardware
Linux NET4.0 for Linux 2.2
Based upon Swansea University Computer Society NET3.039
NET4: Unix domain sockets 1.0 for Linux NET4.0.
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP
Starting kswapd v 1.5 
vesafb: framebuffer at 0x80800000, mapped to 0xc4000000, size 2560k
vesafb: mode is 800x600x8, linelength=800, pages=4
vesafb: protected mode interface info at c000:687c
vesafb: scrolling: redraw
Console: switching to colour frame buffer device 100x37
fb0: VESA VGA frame buffer device
Detected PS/2 Mouse Port.
Serial driver version 4.27 with no serial options enabled
ttyS00 at 0x03f8 (irq = 4) is a 16550A
ttyS02 at 0x03e8 (irq = 4) is a 16550A
pty: 256 Unix98 ptys configured
apm: BIOS version 1.2 Flags 0x0f (Driver version 1.9)
Real Time Clock Driver v1.09
Uniform Multi-Platform E-IDE driver Revision: 6.20
ALI15X3: IDE controller on PCI bus 00 dev 78
ALI15X3: not 100% native mode: will probe irqs later
    ide0: BM-DMA at 0x7090-0x7097, BIOS settings: hda:DMA, hdb:pio
ALI15X3: simplex device:  DMA disabled
ide1: ALI15X3 Bus-Master DMA disabled (BIOS)
hda: IBM-DBCA-206480, ATA DISK drive
ide2: ports already in use, skipping probe
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ALI15X3: Ultra DMA enabled
hda: IBM-DBCA-206480, 6194MB w/420kB Cache, CHS=789/255/63, UDMA(33)
Floppy drive(s): fd0 is 1.44M
FDC 0 is a National Semiconductor PC87306
Partition check:
 hda: hda1 hda2 < hda5 hda6 > hda3
VFS: Mounted root (ext2 filesystem) readonly.
Freeing unused kernel memory: 68k freed
Adding Swap: 104384k swap-space (priority -1)
parport0: PC-style at 0x378 [SPP,PS2,EPP]
lp0: using parport0 (polling).
CSLIP: code copyright 1989 Regents of the University of California
PPP: version 2.3.7 (demand dialling)
PPP line discipline registered.
PPP BSD Compression module registered
Linux PCMCIA Card Services 3.1.5
  kernel build: 2.2.13 #1 Mon Nov 29 21:17:51 CET 1999
  options:  [pci] [cardbus] [apm]
Intel PCIC probe: 
  O2Micro OZ6812 PCI-to-CardBus at bus 0 slot 19, mem 0x68000000, 1 socket
    host opts [0]: [a 24] [b 0d] [c 00] [d 83] [mhpg 18] [fifo 00] [e 2a] [no pci irq] [lat 32/176] [bus 32/34]
    ISA irqs (default) = 3,4,7,11,12 polling interval = 1000 ms
cs: IO port probe 0x1000-0x17ff: clean.
cs: IO port probe 0x0100-0x04ff: excluding 0x408-0x40f 0x480-0x48f 0x4d0-0x4d7
cs: IO port probe 0x0a00-0x0aff: clean.
cs: memory probe 0xa0000000-0xa0ffffff: clean.
eth0: NE2000 Compatible: io 0x300, irq 3, hw_addr 00:E0:98:73:A0:75
eth0: found link beat

thanks in advance, tma

1. None I don't really know by David Hinds, 1999, Nov 30
(_ Question right. by tma@ppmedia.at, 1999, Nov 30
(_ None Re: Question: right. by David Hinds, 1999, Nov 30

Question LinkSys 10baseT "Network Everywhere" NP10T?

Date: 1999, Dec 01
From: John Tunison jtunison

Hi,

I was wondering if anybody's been able to get the LinkSys 10baseT "Network Everywhere" card to work. I know it's not on the supported cards list, but it said on the box "NE2000 compatible" and it only cost $35 at CompUSA, so I thought I'd try it. (Also, I knew that LinkSys had a Linux driver page on their website. Unfortunately, it has no mention fo the Network Everywhere PC card.)

Everything appears to load correctly, but the card won't actually work. I'm _pretty_ sure I've got my ethernet settings confingured correctly.

The card seems to be detected properly:

localhost kernel: eth0: NE2000 Compatible: port 0x300, irq 5, hw_addr 00:E0:98:70:4C:8C

Any suggestions?

Thanks, -John

1. Feedback More info... by John Tunison, 1999, Dec 06
(_ None There isn't much you can do to fix install problems by David Hinds, 1999, Dec 14
(_ Ok For others with a Vaio N505... by John Tunison, 1999, Dec 14

Question RH6.1 505TR hangs returning from hibernate

Date: 1999, Dec 03
From: Henry Hollenberg <speed@rcwm.com>

I have installed RH6.1 on a Sony 505TR and it seems to work quite well except for one thing. With an active network pcmcia card in place going into a suspend or hibernate mode (Fn-F2 or Fn-F12) is the kiss of death. As soon as I try to "wake-up" the system it comes alive for a brief instant and then hangs cold. The only course of action is a power down and wait on the fsck scolding on reboot. My other pcmcia modem card does not have any trouble with suspend and resume.

Also noticed that if the adapter cable or ethernet cable are unplugged from the ethernet card while it is live this will cause the system to hang also.

I can do a 	/sbin/cardctl suspend
		hibernate
		wake up
		/sbin/cardctl resume
without crashing.

Is anyone familiar with this behavior? Is this to be expected? Thanks hghiii

ps the card in question is a Linksys etherfast 10/100:PCMPC200

1. None This driver can't handle suspend/resume events by David Hinds, 1999, Dec 30

Question Support for Intel PRO/100 CardbusII (MBLA3400)

Date: 1999, Dec 03
From: Edwin van Ouwerkerk Moria <edwin@hotobjects.nl>

Is support for this (relatively new) Intel adapter planned? None of the existing (xircom) drivers seem to work - I heard the Intel & Xircom adapters are supposedly compatible (if not identical). According to the specs, it uses an (Intel) i82559 chip.

Any info/help would be appreciated!

1. None Not yet by David Hinds, 1999, Dec 03

Question DHCP with PCMCIA

Date: 1999, Dec 06
From: Christopher Keller ckeller

I'm having a problem getting a DHCP client to automatically come up on boot up. It comes up fine after boot-up, which is why I think it's merely a configuration problem, rather than an incompatibility.

First the specifics:

Dell Inspiron 3200
Caldera 2.3 w/ kernel 2.2.10
PCMCIA 3.0.14
3COM 3CCFEM556B (ethernet + modem)

When I hook up to a network which is running a DHCP server the following happens (some is assumption, some is fact, please correct where appropriate):

Out of /etc/rc.d/rc2.d/S01network, it tries to start the DHCPcd client using the files in /etc/sysconfig/network-scripts. Specifically ifup-dhcp. THis fails, I'm assuming because the PCMCIA drivers haven't loaded yet. I haven't been able to find any error messages in the logs, but I've confirmed that the dhcpcd daemon is trying to run. I'm guessing it doesn't log because it doesn't get far enough along due to the drivers not being loaded (although I have confirmed that eth0 is up). Anyway...along comes S07pcmcia. Once again, dhcp setup is tried (this time using cardmgr). I get the following messages in /var/log/messages:
 
cardmgr[572]: + Trying to obtain network configuration via DHCP.
kernel: eth0: found link beat
kernel: eth0: autonegotiation complete: 100baseT-HD selected
kernel: eth0: remote fault detected
dhcpcd[587]: timed out waiting for a valid DHCP server response
cardmgr[572]: + No response from DHCP server
cardmgr[572]: + eth0: initialization failed

and so on....

At any rate, once the login screen comes up, I can log in as root and do a manual "network start eth0" and dhcp comes up perfectly. No problems whatsoever. Or I can accomplish the same thing by the scripts in /etc/sysconfig/network-scripts/ifup-dhcp.

So, it makes sense that the first attempt out of S01network fails, as there are no PCMCIA drivers yet loaded. But the second attempt (above) out of S07pcmcia should work. I can't believe I'm the only one trying to run DHCP on a laptop, so if someone could point me the right way, I'd much appreciate it.

The easy solution might be to write a S99dhcp file and kludge the thing together. However, if there is a problem with my knowledge or the PCMCIA packages, I'd rather solve that.

TIA....

-Chris

1. None Not sure if this is Caldera specific by David Hinds, 1999, Dec 06
(_ None Hate to say it, but even Red Hat is having trouble with DHCP and PCMCIA by Nathan Hartwell, 2000, Jan 17
(_ None Don't be so sure! by David Hinds, 2000, Jan 17
(_ None quite right....it wasn't a PCMCIA issue, but a DHCP issue by Nathan Hartwell, 2000, Jan 19

Question Possible erroneous manfid in config file?

Date: 1999, Dec 13
From: Brian Mays brian

The Corega PCC-T Ehternet PCMCIA card (NE2000 clone) identifies its manfid as 0xc00f, 0x0000.

 $ cardctl ident 0
   product info: "corega K.K.", "corega Ether PCC-T"
   manfid: 0xc00f, 0x0000
   function: 6 (network)

But in pcmcia-cs 3.1.5, the config file config has another entry which also has manfid as 0xc00f, 0x0000.

 card "Allied Telesyn AT-2800 10/100 Fast Ethernet"
   manfid 0xc00f, 0x0000
   bind "tulip_cb"

Therefore, when a PCC-T Ethernet card is used, pcmcia-cs tried to install tulip_cb modules (instead of pcnet_cs), which results in errors.

Is the manfid for the AT-2800 correct or is something else wrong?

1. None Well, one of them is wrong... by David Hinds, 2000, Jan 14

None Problems with 3c575 and CEM33 on an NEC Versa LX

Date: 1999, Dec 13
From: Brian Mays brian

This message describes problems with two pcmcia cards. One is a 3com cardbus 10/100 (3c575), the other is a Xircom pccard CEM33 (10bt ethernet+modem 33.6). These cards are being used with an NEC Versa LX laptop.

The 3com card works ok except when ftping a file (of say, 184mb) the system freezes and there are messages on the console. They are

  eth0: Host error, FIFO diagnostic register 0000.
        This message repeats 21 times
  eth0: Too much work in interrupt, status 8003.  Temporarily disabling
        functions (7ffc).

At that point, the system is completely unusable. If the card is ejected, the kernel oops's and the system is still unusable. Note that the error messages for the oops are unavailable, since it was very long and difficult to capture.

The Xircom card freezes the system. No messages are printed. When ejecting the card, the display shifts. Column 1 starts in the middle of the screen. The system is completely unusable after inserting the card.

Before upgrading to pcmcia 3.1.5, version 3.1.1 was installed. The 3com card has experienced problems with both versions. The Xircom adapter worked under the 3.1.1 version, but not 3.1.5.

1. None A favorable resolution on both counts by David Hinds, 1999, Dec 17

Warning genius MF3000 different versions?

Date: 1999, Dec 29
From: J. Schmid jds

There is an entry in the supported cards : "Genius MF3000" (fast ethernet) and a coresponding entry in pcmcia/config:

card "Genius MF3000 Fast Ethernet"
  version "CardBus", "10/100Mbps LanCard"
  bind "tulip_cb" 

I have a card labelled "Genius MF3000" which identifies itself as:

"CardBus", "10/100Mbps LANCard" ("LAN" is all capital now!!!)

And does not work properly after changing /pcmcia/config so that tulip_cb is loaded.

I can ping another machine but long packages are lossy and I don't get ftp through the line.

System: sony vaio737 kernel: 2.2.12 pcmcia: 3.1.8

1. More kludge for partial functionality of genius MF3000 by J. Schmid, 2000, Jan 02

Question NetGear FA510 ethernet card not recognized

Date: 2000, Jan 05
From: Michael Bletzinger mbletzin

A coworker and I have the same thinkpad 600E and netgear card. He is running 2.2.13 kernel and pcmcia distribution 3.1.0. I am running kernel 2.2.12 with pcmcia 3.1.8. His configuration works mine doesn't.

When he does a "cardctl ident" the following is listed:

  product info: "NETGEAR", "FA510", "Fast Ethernet CardBus Card", "1.00"
  manfid: 0x9513, 0x0081
  function: 6 (network)

On my conputer the command list "no product info found". The card was identified once while I was hot swapping cards.

        Thanks in advance
        Michael
1. None The usual response: check your system log by David Hinds, 2000, Jan 06
(_ Feedback Here are the logs by Michael Bletzinger, 2000, Jan 06
(_ Ok Changing the address worked by Michael Bletzinger, 2000, Jan 10

Question O2Micro(OZ6836/6860)+EtherFast=very slow network...

Date: 2000, Jan 07
From: Max Gotlib <max@utc.telenet.ru>

I've got a problem...
Laptop: BMx MP-995
	Mobil Pentium II 400Mhz
	O2Micro (OZ6836/OZ6860) PCI-CardBus bridge
	NIC:
		product info: "FastEthernet", "Adapter", "1.0"
		manufid: 0x0149 oxc1ab
		(FreeBSD-3.3r+PAO recognizes it as Linksys 	
		EtherFast PCMPCI00 - DL10019C based NE2000 	
		clone)
		This NIC properly operates under Win98 and
		FreeBSD-3.3r+PAO
OS: Linux (RedHat's clone) 2.2.13 kernel
PCMCIA package: 3.1.9 (tried 3.0.14, 3.1.8, 3.1.9 28 Dec,
			06 Jan - the same results)

The problem:
	NIC is recognized at bootup (it could be removed and 
inserted back - recognition is successfull again), but it 
delevers very slow performance - ping delay (across 10 and 100 
MBit/s LAN is more, than 180ms). There is no 
"routing-or-other-kinda-administration" problem - i've used 
just the same configuration in the same situation under 
FreeBSD successfully). There are no "interrupt(s) dropped!" 
messages - everithing, except performance (about 9 KB/s in FTP 
session) problems.

I've tryed virtually all combinations of i82365 and pcnet_cs options - no luck...

What can I do to fix it ?

P.S.: Here the "appendixies" are:

1. dmesg:
-----------------------------------------
Linux version 2.2.13 (root@localhost.localdomain) (gcc version egcs-2.91.66 1999
0314/Linux (egcs-1.1.2 release)) #6 pTN qNW 7 04:15:45 YEKT 2000
Detected 400921266 Hz processor.
Console: colour dummy device 80x25
Calibrating delay loop... 399.77 BogoMIPS
Memory: 63984k/66496k available (1004k kernel code, 412k reserved, 1036k data, 6
0k init)
CPU: Intel Mobile Pentium II stepping 0d
Checking 386/387 coupling... OK, FPU using exception 16 error reporting.
Checking 'hlt' instruction... OK.
POSIX conformance testing by UNIFIX
PCI: PCI BIOS revision 2.10 entry at 0xfb200
PCI: Using configuration type 1
PCI: Probing PCI hardware
Linux NET4.0 for Linux 2.2
Based upon Swansea University Computer Society NET3.039
NET4: Unix domain sockets 1.0 for Linux NET4.0.
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP, IGMP
Initializing RT netlink socket
Starting kswapd v 1.5 
parport0: PC-style at 0x378 [SPP,PS2]
vesafb: framebuffer at 0xe0000000, mapped to 0xc4800000, size 4096k
vesafb: mode is 1024x768x16, linelength=2048, pages=1
vesafb: protected mode interface info at c87c:0000
vesafb: scrolling: redraw
vesafb: directcolor: size=0:5:6:5, shift=0:11:5:0
Console: switching to colour frame buffer device 128x48
fb0: VESA VGA frame buffer device
Detected PS/2 Mouse Port.
Serial driver version 4.27 with no serial options enabled
ttyS00 at 0x03f8 (irq = 4) is a 16550A
ttyS01 at 0x02f8 (irq = 3) is a 16550A
pty: 256 Unix98 ptys configured
lp0: using parport0 (polling).
apm: BIOS version 1.2 Flags 0x07 (Driver version 1.9)
PCI_IDE: unknown IDE controller on PCI bus 00 device 39, VID=8086, DID=7199
PCI_IDE: not 100% native mode: will probe irqs later
    ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hda:DMA, hdb:DMA
    ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdc:pio, hdd:pio
hda: IBM-DARA-206000, ATA DISK drive
hdb: CD-224E, ATAPI CDROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
hda: IBM-DARA-206000, 5729MB w/418kB Cache, CHS=730/255/63, (U)DMA
hdb: ATAPI 24X CD-ROM drive, 128kB Cache
Uniform CDROM driver Revision: 2.56
Floppy drive(s): fd0 is 1.44M
FDC 0 is a post-1991 82077
Partition check:
 hda: hda1 hda2 hda3
VFS: Mounted root (ext2 filesystem) readonly.
Freeing unused kernel memory: 60k freed
Adding Swap: 112448k swap-space (priority -1)
MSDOS FS: Using codepage 866
MSDOS FS: IO charset koi8-r
VFS: freeing iocharset=koi8-r
Linux PCMCIA Card Services 3.1.9
  kernel build: 2.2.13 #6 pTN qNW 7 04:15:45 YEKT 2000
  options:  [pci] [apm]
Intel PCIC probe: 
  O2Micro OZ6836/OZ6860 PCI-to-CardBus at bus 0 slot 3, mem 0xe1000000, 2 sockets
    host opts [0]: [a 44] [b 8d] [c 00] [d 02] [mhpg 18] [fifo 00] [e 1a] [no pci irq] [lat 32/176] [bus 32/34]
    host opts [1]: [a 44] [b 81] [c 50] [d 02] [mhpg 18] [fifo 00] [e 1a] [no pci irq] [lat 32/176] [bus 35/37]
    ISA irqs (default) = 10 polling interval = 1000 ms
cs: IO port probe 0x0300-0x031f: clean.
cs: memory probe 0xa0000000-0xa0ffffff: excluding 0xa0000000-0xa00fffff
eth0: NE2000 Compatible: io 0x300, irq 10, hw_addr 00:E0:98:73:64:85
eth0: found link beat

2. /etc/sysconfig/pcmcia:

PCMCIA=yes

PCIC=i82365

PCIC_OPTS="irq_list=10 pci_int=1 pci_csc=1 do_scan=0"

CORE_OPTS=

3. /etc/pcmcia/config.opts

# i'm sure, the NIC uses 0x300 io and 10 irq.

include port 0x300-0x31f

include memory 0xd0000-0xdffff

include memory 0xa0000000-0xa0ffffff

exclude irq 4

exclude irq 3

exclude irq 7

exclude irq 9

exclude irq 11

exclude irq 12

exclude irq 13

exclude irq 14

exclude irq 15

exclude irq 0

exclude irq 1

exclude irq 2

exclude irq 5

exclude irq 6

module "pcnet_cs" opts "irq_list=10"

1. Note Ugly workaround:-) by Dag Kröper, 2000, Jul 10

Question Problems with Accton EN2220

Date: 2000, Jan 09
From: Volker Wendt <wendt@nid.de>

Hi,

I tried to use an Accton EN2220 CardBus Fast Ethernet Card. I am using pcmcia-cs.3.1.8 on Linux 2.2.13. Everything seems to work well: The card is detected correctly and all modules are loaded. But it does'nt work. The log says

Jan  9 15:06:27 jc kernel: cs: cb_alloc(bus 32): vendor 0x1011, device 0x0019
Jan  9 15:06:27 jc cardmgr[1231]: initializing socket 0
Jan  9 15:06:27 jc kernel: ROM image dump:
Jan  9 15:06:27 jc kernel:   image 0: 0x000000-0x0001ff, signature PCIR
Jan  9 15:06:27 jc cardmgr[1231]: socket 0: Accton EN2220 Fast Ethernet
Jan  9 15:06:27 jc cardmgr[1231]: executing: 'insmod /lib/modules/2.2.13/pcmcia/cb_enabler.o'
Jan  9 15:06:27 jc cardmgr[1231]: executing: 'insmod /lib/modules/2.2.13/pcmcia/tulip_cb.o'
Jan  9 15:06:27 jc kernel: cs: cb_config(bus 32)
Jan  9 15:06:27 jc kernel:   fn 0 bar 1: io 0x400-0x47f
Jan  9 15:06:27 jc kernel:   fn 0 bar 2: mem 0x600c0000-0x600c03ff
Jan  9 15:06:27 jc kernel:   fn 0 rom: mem 0x60080000-0x600bffff
Jan  9 15:06:27 jc kernel: cs: cb_enable(bus 32)
Jan  9 15:06:27 jc kernel:   bridge io map 0 (flags 0x21): 0x400-0x47f
Jan  9 15:06:27 jc kernel:   bridge mem map 0 (flags 0x1): 0x60080000-0x600c0fff
Jan  9 15:06:27 jc kernel: tulip_attach(bus 32, function 0)
Jan  9 15:06:27 jc kernel: tulip.c:v0.91 4/14/99 becker@cesdis.gsfc.nasa.gov (modified by danilo@cs.uni-magdeburg.de for XIRCOM CBE, fixed by Doug Ledford)
Jan  9 15:06:27 jc kernel: eth0: Digital DS21143 Tulip rev 65 at 0x400, FC:00:FC:00:00:00, IRQ 10.
Jan  9 15:06:27 jc cardmgr[1231]: executing: './network start eth0'
Jan  9 15:06:41 jc kernel: eth0: Tx hung, 12 vs. 9.
Jan  9 15:06:41 jc kernel: eth0: 21140 transmit timed out, status f0260000, SIA 000000c6 ffff0000 fffbffff 8ff10008, resetting...
Jan  9 15:06:47 jc kernel: eth0: Tx hung, 12 vs. 9.
Jan  9 15:06:47 jc kernel: eth0: 21140 transmit timed out, status f0260000, SIA 000000c6 ffff0000 fffbffff 8ff10008, resetting...
Jan  9 15:06:53 jc kernel: eth0: Tx hung, 12 vs. 9.
Jan  9 15:06:53 jc kernel: eth0: 21140 transmit timed out, status f0260000, SIA 000000c6 ffff0000 fffbffff 8ff10008, resetting...
Jan  9 15:07:02 jc kernel: eth0: Tx hung, 15 vs. 9.
Jan  9 15:07:02 jc kernel: eth0: 21140 transmit timed out, status f0260000, SIA 000000c6 ffff0000 fffbff7f 8ff10008, resetting...
Jan  9 15:07:08 jc kernel: eth0: Tx hung, 15 vs. 9.
Jan  9 15:07:08 jc kernel: eth0: 21140 transmit timed out, status f0260000, SIA 000000c6 ffff0000 fffbff7f 8ff10008, resetting...
Jan  9 15:07:14 jc kernel: eth0: Tx hung, 15 vs. 9.
Jan  9 15:07:14 jc kernel: eth0: 21140 transmit timed out, status f0260000, SIA 000000c6 ffff0000 fffbff7f 8ff10008, resetting...
Jan  9 15:07:23 jc kernel: eth0: Tx hung, 18 vs. 9.
Jan  9 15:07:23 jc kernel: eth0: 21140 transmit timed out, status f0260000, SIA 000000c6 ffff0000 fffbffff 8ff10008, resetting...

Thanks for any help,
  Volker

1. None Need more messages by David Hinds, 2000, Jan 09
2. None The usual mantra: more system log messages, please by David Hinds, 2000, Jan 10
3. None might be tulip problem? by J. Schmid, 2000, Jan 11

Ok New hw_info line for PCMCIA Technology OEM NE2000 Compatible

Date: 2000, Jan 10
From: sampo

I don't know where to report these for adding to autodetection list so I try here. I figured out this stuff using the instructions for configuring a new NE2000 clone.

To pcnet_cs.c following hw_info struct:

 { /* PCMCIA Technology OEM 10/100 */ 0x01c8, 0xa0, 0x0c, 0 },

To config

 card "PCMCIA Technology OEM 10/100 NE2000 Compatible Ethernet"
   manfid 0x8a01, 0xc1ab
   bind "pcnet_cs"

Basically this card is a noname OEM card. Mine does not have any indication of manufacurer (I found the string "PCMCIA Technology, Inc." by looking into cards memory space) or country of origin. I bought mine at Portugal on Sept 1999.

On the reverse side following markings were found:

 10/100 Mbps Ethernet PC Card
 P/N: KEN0100-A
 VER: XA
 S/N: 920I07430004
 16 5V

--Sampo

Question Schemes problem

Date: 2000, Jan 10
From: Ralf Eichmann ralf.eichmann

Hello,

I installed SuSE Linux 6.3 on an Asus F7400 with a D-Link 650. Since the notebook should be usable in various networks I've set up a number of configurations in /etc/pcmcia/network.opts and the corresponding lilo.conf entries.

The lilo append lines are recognized (/sbin/init.d/pcmcia prints "using scheme: ..." correctly), but eth0 is not configured properly. None of the network.opts parameters is set. However, "cardctl scheme ..." works fine. This leads to the following: I do a "cardctl scheme home", then reboot, choose the lilo config with "append="scheme=mg33"" (where mg33 is another scheme), get the message "using scheme: mg33", but "ifconfig eth0" shows the settings from the home scheme.

Is that what is meant with "The SuSE startup script is somewhat limited and does not allow PCMCIA startup variables to be overridden from the lilo boot prompt." from the SuSE section in the PCMCIA-Howto? Can I work around this (by issuing a "cardctl scheme $SCHEME" somewhere in a boot script)?

Thanks and best regards, Ralf Eichmann

Idea Problem and Workaround with dhcpcd

Date: 2000, Jan 12
From: uffti

I've just upgrade to pcmcia-cs-3.1.8 in conjuction with dhcpcd-1.3.18-pl3. I've encountered two flaws :

1. dhcpcd seems to be called wrong
   in /etc/pcmcia/network dhcpcd is called with -V
   with the assumption that it's only displays the
   version information. But it print the version info
   AND makes a dhcp request (I've checked source code
   too).
2. No possibility to set hostname
   There is no way to pass thru parameters to dhcpd

I've modified /etc/pcmcia/network the way it works for me. Maybe this should be considered in future releases :

$ diff network /etc/pcmcia/ 56,73c56,74

<           if /sbin/dhcpcd -V 2>&1 | grep -q DHCP ; then
<               /sbin/dhcpcd $DEVICE >/dev/null 2>&1 || exit 1
<           else
<               # Jump through hoops for lame 0.70-era dhcpcd
<               L=/var/run/dhcp-lock-$DEVICE
<               /bin/echo "#!/bin/sh\nrm $L" > $L ; chmod +x $L
<               /sbin/dhcpcd -c $L $DEVICE >/dev/null 2>&1
<               for t in 0 1 2 3 4 5 6 7 8 9 ; do
<                   sleep 2 ; if [ ! -e $L ] ; then break ; fi
<               done
<               rm -f $L
<               if [ -e /etc/dhcpc/resolv.conf ] ; then
<                   echo "# $DEVICE begin" > $RESOLV.N
<                   cat /etc/dhcpc/resolv.conf >> $RESOLV.N
<                   echo "# $DEVICE end" >> $RESOLV.N
<                   cat $RESOLV >> $RESOLV.N ; mv $RESOLV.N $RESOLV
<               fi
<           fi
---
>           /sbin/dhcpcd $DEVICE -h "$DHCP_HOSTNAME" || exit 1
> #         if /sbin/dhcpcd -V 2>&1 | grep -q DHCP ; then
> #             /sbin/dhcpcd $DEVICE >/dev/null 2>&1 || exit 1
> #         else
> #             # Jump through hoops for lame 0.70-era dhcpcd
> #             L=/var/run/dhcp-lock-$DEVICE
> #             /bin/echo "#!/bin/sh\nrm $L" > $L ; chmod +x $L
> #             /sbin/dhcpcd -c $L $DEVICE >/dev/null 2>&1
> #             for t in 0 1 2 3 4 5 6 7 8 9 ; do
> #                 sleep 2 ; if [ ! -e $L ] ; then break ; fi
> #             done
> #             rm -f $L
> #             if [ -e /etc/dhcpc/resolv.conf ] ; then
> #                 echo "# $DEVICE begin" > $RESOLV.N
> #                 cat /etc/dhcpc/resolv.conf >> $RESOLV.N
> #                 echo "# $DEVICE end" >> $RESOLV.N
> #                 cat $RESOLV >> $RESOLV.N ; mv $RESOLV.N $RESOLV
> #             fi
> #         fi
154,155c155,156
<               /sbin/dhcpcd -V 2>&1 | grep -q DHCP || \
<                   rm -f /var/run/dhcpcd-$DEVICE.pid
---
> #             /sbin/dhcpcd -V 2>&1 | grep -q DHCP || \
> #                 rm -f /var/run/dhcpcd-$DEVICE.pid

1. None Odd changes by David Hinds, 2000, Jan 12
(_ None yes it works by uffti, 2000, Jan 13

Sad Vaio 505V + OvisLink LFS PCM32 doesn't work

Date: 2000, Jan 18
From: Klaus Kudielka klaus.kudielka

Hello,

I have a Sony Vaio 505V and recently bought an OvisLink
LFS PCM 32 cardbus card, since it is listed in
SUPPORTED.CARDS.

Needless to say, my first try was rather unsuccessfull :-(
linux-2.2.14, pcmcia-cs-3.1.8.
Neither transmitting nor receiving works. The RX counters
ifconfig reports just stay zero, likewise the TX counters.
Only TX errors is incremented now and then. The card is
attached to a 10Mbit/s hub.

Linux PCMCIA Card Services 3.1.8
  kernel build: 2.2.14 #3 Tue Jan 18 15:46:42 UTC 2000
  options:  [pci] [cardbus] [apm]
Intel PCIC probe: 
  Ricoh RL5C475 PCI-to-CardBus at bus 0 slot 10, mem 0x68000000, 1 socket
    host opts [0]: [isa irq] [io 3/6/1] [mem 3/6/1] [no pci irq] [lat 168/176] [bus 32/34]
    ISA irqs (default) = 3,4,7,10,11,12,15 polling interval = 1000 ms
cs: IO port probe 0x1000-0x17ff: clean.
cs: IO port probe 0x0100-0x04ff: excluding 0x170-0x177 0x370-0x37f 0x388-0x38f 0x4d0-0x4d7
cs: IO port probe 0x0a00-0x0aff: clean.
cs: cb_alloc(bus 32): vendor 0x1011, device 0x0019
cs: cb_config(bus 32)
  fn 0 bar 1: io 0x400-0x47f
  fn 0 bar 2: mem 0x600c0000-0x600c03ff
  fn 0 rom: mem 0x60080000-0x600bffff
cs: cb_enable(bus 32)
  bridge io map 0 (flags 0x21): 0x400-0x47f
  bridge mem map 0 (flags 0x1): 0x60080000-0x600c0fff
tulip_attach(bus 32, function 0)
tulip.c:v0.91 4/14/99 becker@cesdis.gsfc.nasa.gov (modified by danilo@cs.uni-magdeburg.de for XIRCOM CBE, fixed by Doug Ledford)
eth0: Digital DS21143 Tulip rev 65 at 0x400, 00:E0:98:12:F9:9D, IRQ 10.
eth0:  EEPROM default media type Autosense.
eth0:  Index #0 - Media MII (#11) described by a 21142 MII PHY (3) block.
eth0:  MII transceiver #0 config 3000 status 7809 advertising 01e1.
eth0: Tx hung, 11 vs. 0.
eth0: Tx hung, 14 vs. 0.

(now executing cardctl eject)

tulip_detach(eth0)
cs: cb_disable(bus 32)
cs: cb_release(bus 32)
cs: cb_free(bus 32)

Any idea?

-Klaus
1. None Maybe try the latest beta? by David Hinds, 2000, Jan 21
(_ Sad 18-Jan-00 doesn't help. by Klaus Kudielka, 2000, Jan 21
(_ Ok 3.1.10 works by Klaus Kudielka, 2000, Feb 07

Question Problems with Kingmax KEN0100-DF Ethernet card

Date: 2000, Jan 20
From: Albio alby

My Kingmax KEN0100-DF Ethernet card is recognized by cardmngr using the module tulip_cb, but I can't access the network (with ping for example).

I controlled several times the network configuration: IP address, DNS, Gateway etc.

Giving ifconfig the eth0 is up; the ip number and netmask are correctly setted.

Can someone help me? Thanks

1. None Again... system log messages?? by David Hinds, 2000, Jan 20
(_ Note Sys Log messages by Albio, 2000, Jan 21
(_ None Driver upgrade? by David Hinds, 2000, Jan 22
(_ None It doesn't work by Albio, 2000, Jan 22
(_ None Re: It doesn't work by David Hinds, 2000, Jan 24
(_ None It's doesn't work again. by Albio, 2000, Jan 28
(_ Ok It works !!!! by Albio, 2000, Feb 05

Question Socket EA System Hang

Date: 2000, Jan 21
From: Robert Sanderson azaroth

Hi all, I've been trying unsuccessfully to get my eth0 card working with my new laptop.

System:  Dell Latitude CPi R400GT
PC Card:  Socket EA Ethernet
Kernel:  2.2.14
PCMCIA:  3.1.8
Changes:  Added exit 0; in pcmcia/network to prevent hangs (see below)
Messages:

------------------------
  
initializing socket 1
eth0: NE2000 Compatible: io 0x300 irq 3 auto xcvr hw_addr 00:C0:1B:00:5E:27
socket 1: Socket EA Ethernet
executing insmod /lib/modules/2.2.14/net/8390.o
executing insmod /lib/modules/2.2.14/pcmcia/pcnet_cs.o
executing ./network start eth0

----------------------------

At which point, the system hangs until the card is ejected.
When it is, the following messages are given:

----------------------------
Hw. address read/write mismap 0
Hw. address read/write mismap 1
Hw. address read/write mismap 2
Hw. address read/write mismap 3
Hw. address read/write mismap 4
Hw. address read/write mismap 5
eth0: interupt from stopped card

----------------------------

The card worked fine in my old laptop (Compaq Aero C33,
running linux 2.0.0), and the my modem card (psion dacom gold
card) works fine in the Dell as well as the Compaq.

In the documentation, it says that it's necessary to use
ifport to manually set the interface for EA cards - doing this
or not seems to make no difference, the system still hangs
with the same errors.

Occaisonally it won't hang when the interface is brought up,
but will on receiving any traffic. In this scenario the
messages are:

--------------------------------------

eth0: mismatched read page pointers 0 vs  e.
eth0: mismatched read page pointers 0 vs  ff.
eth0: mismatched read page pointers 0 vs  ff.
eth0: mismatched read page pointers 0 vs  ff.
...
eth0: unexpected TX-done interrupt, lasttx=0.
eth0: too much work at interrupt, status 0xff
eth0: interrupt from stopped card

---------------------------------------

Other miscellaneous messages from when the card is inserted 
but nothing has been done to it:

'probe' gives:

PCI bridge probe:  TI 1255 found, 2 sockets. 

'cardctl ident' gives:

product info: Socket Communications Inc, Socket EA PCMCIA LAN Adapter
Revision E, Ethernet ID 00c01b005e27
manfid: 0x0104, 0x000
function: 6 (network)

'cardctl status' gives:

5V 16-bit PC Card
funciton 0: [ready]   

------

This looks like a driver problem, does anyone have any 
information or ideas about it?

Thanks,

Rob Sanderson
1. None Resource conflict? by David Hinds, 2000, Jan 21
(_ More Resource Conflict, Socket EA by Robert Sanderson, 2000, Jan 24
(_ More Socket EA hang, by Robert Sanderson, 2000, Jan 25

Question Problems starting dhclient

Date: 1999, Dec 05
From: Stefan Hahne loki

I have a Toshiba Satellite 2550CDT, Kernel 2.2.13 and pcmcia-cs-3.1.5.
Compiling and installation works fine and ifconfig
shows:
	"eth0 Link encap:Ethernet	HWAddr 00:50:22:20:3F:A9
	 unspec addr:[NONE SET] Bcast:[NONE SET] MASK:[NONE SET]
	 ..."
but the dhcp-client exit with "dhclient: eth0: not found" at
boot time and whenever I insert the card!!

Any suggestions?
Thanks in advance!
1. None Check your system log, etc by David Hinds, 2000, Jan 05

Question Trust Netlink Combi don't respond to ping

Date: 2000, Jan 06
From: A. Bosio torello

I'm using an Acer Travelmate 512T, with debian slink,
Linux 2.2.13, pcmcia-cs 3.1.8.

The card is a Trust Netlink Combi, and it is recognized
as NE2000 Compatible.

This is the output of "ifconfig eth0":

eth0      Link encap:Ethernet  HWaddr 00:80:C8:8B:5C:9E  
          inet addr:192.168.1.3  Bcast:192.168.1.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:100 
          Interrupt:11 Base address:0x300

and this is the output of "netstat -r":
Kernel IP routing table
Destination     Gateway         Genmask         Flags   MSS Window  irtt Iface
localnet        *               255.255.255.0   U         0 0          0 eth0

If I do a ping from the Acer to another host, or from another
host to the Acer, noone respond.

This is the output of "tcpdump -i eth0" on the Acer, doing
a ping from the Acer to another host in the net (mu is
the Acer Travelmate, yang is the other computer).

11:07:35.711389 mu.change.fake > yang.change.fake: icmp: echo request
11:07:36.703568 mu.change.fake > yang.change.fake: icmp: echo request
11:07:40.703457 arp who-has yang.change.fake tell mu.change.fake
11:07:41.703456 arp who-has yang.change.fake tell mu.change.fake
11:07:42.703449 arp who-has yang.change.fake tell mu.change.fake
11:07:43.703448 arp who-has yang.change.fake tell mu.change.fake

And so on.
This is the output for a tcpdump on the Acer for a ping from an host in the net to the Acer:
11:06:54.933617 arp who-has mu.change.fake tell yang.change.fake
11:06:54.933719 arp reply mu.change.fake is-at 0:80:c8:8b:5c:9e
11:06:55.933600 arp who-has mu.change.fake tell yang.change.fake
11:06:55.933667 arp reply mu.change.fake is-at 0:80:c8:8b:5c:9e
11:06:56.933594 arp who-has mu.change.fake tell yang.change.fake
11:06:56.933640 arp reply mu.change.fake is-at 0:80:c8:8b:5c:9e

And this is the output for the same ping on the other host:
12:06:37.406405 arp who-has mu tell yang
12:06:38.400825 arp who-has mu tell yang
12:06:39.400781 arp who-has mu tell yang


Another thing: the "collisions" led of the Trust Netlink Combi blinks a lot...

So the problem is: am I an idiot on configuring this network card or does the card have some problems?

Thanks.
1. None The usual: system log messages?? by David Hinds, 2000, Jan 06

Question Trouble with Linksys Ethernet Card

Date: 2000, Jan 26
From: Mike jabbera

I am trying to get this card working, and every time i insert it i get the following error.

eth0: pcnet_reset_8390() did not complete. pcnet_cs: unable to read hardware net address

Anyone give me a hand with this?

1. None More system log messages? by David Hinds, 2000, Jan 27

Question Dual PCMCIA NIC system (Linksys plus NDC ) time outs and lost interrupts

Date: 2000, Jan 29
From: Dale moberg dmoberg

Report on dual PCMCIA MIC interface lost interrupt on Hewlett-Packard Omnibook 5500 CT

Second card (Linksys) yields time outs waiting for device to not be busy? First card working fine.

Message from log indicating what seems to be the core problem is

Jan 29 11:12:33 redhatomni kernel: eth1: Tx timed out, lost interrupt? TSR=0x3, ISR=0x2, t=1000.

I enclose outputs from uname, lsmod, proc/interrupts, cardctl, ifconfig, lsmod, and startup log for pcmcia modules.

[root@redhatomni net]# uname -a
Linux redhatomni 2.2.12-20 #1 Mon Sep 27 10:25:54 EDT 1999 i586 unknown
[root@redhatomni net]# cardctl -V
cardctl version 3.0.14
[root@redhatomni net]# lsmod
Module                  Size  Used by
parport_pc              5620   0  (unused)
parport                 7124   0  [parport_pc]
nls_iso8859-1           2020   1  (autoclean)
pcnet_cs                7572   2 
8390                    5920   0  [pcnet_cs]
ds                      5740   2  [pcnet_cs]
i82365                 22640   2 
pcmcia_core            39912   0  [pcnet_cs ds i82365]
[root@redhatomni net]# cat /proc/interrupts
           CPU0       
  0:    7572752          XT-PIC  timer
  1:      31645          XT-PIC  keyboard
  2:          0          XT-PIC  cascade
  3:       8964          XT-PIC  pcnet_cs
  5:          0          XT-PIC  pcnet_cs
  8:          1          XT-PIC  rtc
 12:     255634          XT-PIC  PS/2 Mouse
 13:          1          XT-PIC  fpu
 14:     561531          XT-PIC  ide0
 15:     601841          XT-PIC  ide1
NMI:          0
[root@redhatomni net]# cardctl config
Socket 0:
  Vcc = 5.0, Vpp1 = 0.0, Vpp2 = 0.0
  Interface type is memory and I/O
  IRQ 3 is exclusive, level mode, enabled
  Function 0:
    Config register base = 0x03f8
      Option = 0x60, status = 0000
    I/O window 1: 0x0300 to 0x031f, auto sized
Socket 1:
  Vcc = 5.0, Vpp1 = 0.0, Vpp2 = 0.0
  Interface type is memory and I/O
  IRQ 5 is exclusive, level mode, enabled
  Function 0:
    Config register base = 0x0fd0
      Option = 0x43, status = 0000, copy = 0000
    I/O window 1: 0x0340 to 0x035f, auto sized
[root@redhatomni net]# cardctl status
Socket 0:
  5V 16-bit card present
  Function 0: ready
Socket 1:
  5V 16-bit card present
  Function 0: ready

[root@redhatomni pcmcia]# ifconfig
eth0      Link encap:Ethernet  HWaddr 00:80:C6:F6:8E:1F  
          inet addr:192.168.88.149  Bcast:192.168.88.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:8306 errors:0 dropped:0 overruns:0 frame:21
          TX packets:927 errors:0 dropped:0 overruns:0 carrier:0
          collisions:1 txqueuelen:100 
          Interrupt:3 Base address:0x300 

eth1      Link encap:Ethernet  HWaddr 00:E0:98:06:AC:30  
          inet addr:204.210.245.55  Bcast:204.210.245.255  Mask:255.255.255.0
          UP BROADCAST RUNNING  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:2 overruns:0 frame:0
          TX packets:0 errors:45 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:100 
          Interrupt:5 Base address:0x340 

lo        Link encap:Local Loopback  
          inet addr:127.0.0.1  Mask:255.0.0.0
          UP LOOPBACK RUNNING  MTU:3924  Metric:1
          RX packets:247 errors:0 dropped:0 overruns:0 frame:0
          TX packets:247 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0 

[root@redhatomni pcmcia]# route Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 255.255.255.255 * 255.255.255.255 UH 0 0 0 eth0 204.210.245.55 * 255.255.255.255 UH 0 0 0 eth1 192.168.88.149 * 255.255.255.255 UH 0 0 0 eth0 204.210.245.0 * 255.255.255.0 U 0 0 0 eth1 localnet * 255.255.255.0 U 0 0 0 eth0 127.0.0.0 * 255.0.0.0 U 0 0 0 lo

Startup from /var/log/debug

Jan 28 20:46:59 redhatomni pcmcia: Starting PCMCIA services:
Jan 28 20:46:59 redhatomni pcmcia:  modules
Jan 28 20:46:59 redhatomni kernel: Linux PCMCIA Card Services 3.0.14 
Jan 28 20:46:59 redhatomni kernel:   kernel build: 2.2.12-20 #1 Mon Sep 27 10:25:54 EDT 1999 
Jan 28 20:46:59 redhatomni kernel:   options:  [pci] [cardbus] [apm] 
Jan 28 20:46:59 redhatomni kernel: Intel PCIC probe:  
Jan 28 20:46:59 redhatomni kernel:   Cirrus PD6729 PCI-to-PCMCIA at bus 0 slot 3, port 0x3e0, 2 sockets 
Jan 28 20:46:59 redhatomni kernel:     host opts [0]: [ring] [1/5/0] [1/20/0] 
Jan 28 20:47:00 redhatomni kernel:     host opts [1]: [ring] [1/5/0] [1/20/0] 
Jan 28 20:47:00 redhatomni kernel:     ISA irqs (default) = 3,4,5,7,9,11,12 polling interval = 1000 ms 
Jan 28 20:46:59 redhatomni pcmcia:  cardmgr.
Jan 28 20:46:59 redhatomni cardmgr[367]: starting, version is 3.0.14
Jan 28 20:47:00 redhatomni rc: Starting pcmcia succeeded
Jan 28 20:47:01 redhatomni cardmgr[367]: watching 2 sockets
Jan 28 20:47:01 redhatomni kernel: cs: IO port probe 0x1000-0x17ff: clean. 
Jan 28 20:47:01 redhatomni kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x220-0x22f 0x330-0x337 0x378-0x37f 0x388-0x38f 0x3b0-0x3ff 0x480-0x48f 
Jan 28 20:47:01 redhatomni kernel: cs: IO port probe 0x0a00-0x0aff: clean. 
Jan 28 20:47:01 redhatomni cardmgr[367]: initializing socket 0
Jan 28 20:47:01 redhatomni kernel: cs: memory probe 0x0d0000-0x0dffff: clean. 
Jan 28 20:47:01 redhatomni cardmgr[367]: socket 0: NDC Instant-Link Ethernet
Jan 28 20:47:01 redhatomni inet: inetd startup succeeded
Jan 28 20:47:02 redhatomni cardmgr[367]: executing: 'insmod /lib/modules/r[367]: executing: 'insmod /lib/modules/2.2.12-20/pcmcia/pcnet_cs.o'
Jan 28 20:47:02 redhatomni kernel: eth0: NE2000 Compatible: io 0x300, irq 3, hw_addr 00:80:C6:F6:8E:1F 
Jan 28 20:47:02 redhatomni cardmgr[367]: executing: './network start eth0'
Jan 28 20:47:03 redhatomni lpd: lpd startup succeeded
Jan 28 20:47:03 redhatomni keytable: Loading keymap: 
Jan 28 20:47:03 redhatomni keytable: Loading system font: 
Jan 28 20:47:03 redhatomni rc: Starting keytable succeeded
Jan 28 20:47:04 redhatomni cardmgr[367]: initializing socket 1
Jan 28 20:47:04 redhatomni cardmgr[367]: socket 1: NE2000 Compatible Ethernet
Jan 28 20:47:04 redhatomni kernel: eth1: NE2000 Compatible: io 0x340, irq 5, hw_addr 00:E0:98:06:AC:30 
Jan 28 20:47:04 redhatomni cardmgr[367]: executing: './network start eth1'

[omitted stuff. Now trying to ping hosts on immediate segment that eth1 sits on]

[these messages produced because device busy flag set within the 8390 module . The TSR and ISR values are in the header file, 8390.h Check 8390.c to find the format code for this printk message.

A 0x3 value seems to be a sum of ok values. So the device is flagged busy but the interrupt and transmit status registers do not appear to give anything too helpful. No other errors found. Notice that the ifconfig message has all TX errors for eth1. Is there an option I need to set?

eth0 is working fine on its segment. Can ping localhost and the address ifconfiged to the device.204.210.245.55

#

[root@redhatomni net]# Jan 29 11:02:58 redhatomni kernel: eth1: Tx timed out, lost interrupt? TSR=0x3, ISR=0x3, t=953.

Jan 29 11:03:08 redhatomni kernel: eth1: Tx timed out, lost interrupt? TSR=0x3, ISR=0x3, t=1000. 
Jan 29 11:03:48 redhatomni last message repeated 4 times
Jan 29 11:03:58 redhatomni kernel: eth1: Tx timed out, lost interrupt? TSR=0x3, ISR=0x3t=t=1000. 
Jan 29 11:12:13 redhatomni kernel: eth1: Tx timed out, lost interrupt? TSR=0x3, ISR=0x3, t=49500. 
Jan 29 11:12:23 redhatomni kernel: eth1: Tx timed out, lost interrupt? TSR=0x3, ISR=0x2, t=1000. 
Jan 29 11:12:33 redhatomni kernel: eth1: Tx timed out, lost interrupt? TSR=0x3, ISR=0x2, t=1000. 
Jan 29 11:13:28 redhatomni kernel: eth1: Tx timed out, lost interrupt? TSR=0x3, ISR=0x3, t=5500. 
Jan 29 11:13:38 redhatomni kernel: eth1: Tx timed out, lost interrupt? TSR=0x3, ISR=0x3, t=1000. 
Jan 29 11:13:58 redhatomni last message repeated 2 times

1. None Sounds like a classic interrupt conflict by David Hinds, 2000, Jan 30
(_ Ok OK, I guess next time I will just ignore proc/interruptsI by Dale moberg, 2000, Jan 30

Question Cardbus Tulip card works unreliably (system crash)

Date: 2000, Feb 03
From: <rolf.fiedler@ferrari.de>

I hvae a cardbus tulip based 10/100 eth card, and it is detected by the pcmcia core and the proper drivers are loaded.

The system is a palmax pd1000 running 2.2.14 with pcmcia 3.1.9.

once the card is inserted, it is detected and ifconfig and route is done. I can ping the network (reliably). I can do telnet and everything that does not transfer a lot of data.

However, if I try to transfer data (like a ftp download) the whole system freezes and I have to turn it off.

I don't know how to debug the problem, since the system crashes.

An ne2000 compatible ethernet card works like a charm, however it is a friend's card :-(.

On the cardbus card it says "Gericom Cardbus 10/100 Fast Ethernet"

I have tried to play around with the resource info in the pcmcia config., however without success (same behaviour). I have tried a 2.3.3? kernel, same result.

Any hints?

1. None System log messages? by David Hinds, 2000, Feb 03
1. Feedback sys log messages by rolf.fiedler@ferrari.de, 2000, Feb 04
2. More detailed bug report by rolf.fiedler@ferrari.de, 2000, Feb 07
3. Warning Problem resolved by rolf.fiedler@ferrari.de, 2000, Feb 08

Feedback Problems compiling w/ kernels 2.3.43 and up

Date: 2000, Feb 14
From: Mark Rasmussen sunbane

Has anyone else ran into the roadblock w/ the newer
kernels?  They seemed to have changed the net_device
structure in the kernel (kerneldir/include/linux/netdevice.h)
and it no longer includes tbusy, interrupt, etc. so 
pcmcia won't compile up.  I have ran into this both with
using the standalone pcmcia in the kernel and the pcmcia
as a separate package - tried 3.1.9, 3.1.10...

Any workarounds for this?

Thanks,
Mark

Note Supported?

Date: 2000, Feb 15
From: Alex Pearsall rebelpacket

Hello, I hate to be a bother, but ive read the supported card list, and looked for two pcmcia ethernet cards that i hope are supported. Im not sure, i think one might be supported, but i figure you guys would know better than me. the two cards are: Linksys - Ethernet Card EC2T 10bT UMAX - 250TX 10/100 bT

Any information would be GREATLY appricated. Thanks..

                   -Alex

Question Argosy 10/100 EN-235

Date: 2000, Feb 17
From: Robert Rossmann rbr

I have an Argosy En-235 fast Etherent network card which is not supported be the current pcmcia driver. The en-220 is supported. The cardmgr wirtes the following code into the message log

cardmgr[63]: unsupported card in socket 1

kernel:cs: cb_alloc vendor 0x10ex device 0x8138

cardmgr[63]: no produkt info available

So i have no produkt info and can't set the card by editing the /etc/pcmcia/config file

Any hints pls

Robert

1. More Argosy 10/100 more info by Robert Rossmann, 2000, Feb 17
(_ None Re: More: Argosy 10/100 more info by David Hinds, 2000, Feb 17
(_ More Used both tulip_cb and epic_cb by Robert Rossmann, 2000, Feb 18
(_ More More info by Robert Rossmann, 2000, Feb 21
(_ None You are currently out of luck by David Hinds, 2000, Feb 21

Question I get "Network Unreachable" trying to ping others...

Date: 2000, Feb 17
From: Captain Panic panic

I read the PCMCIA-howto and it mentions that if you get a "Network Unreachable" trying to ping others on the LAN it means something with your routing. But I can't figure out what's wrong. Also I can ping myself fine. I have my self as ip 90.0.0.1 set through linuxconf . I am running RH 6.1 on a compaq computer with a 10baseT pcmcia ethernet card (8390.o). Can't seem to figure it out. Below is a copy of route. And ifconfig looks good with itherface eth0 and lo.

Dest Gateway Genmask Flags Iface

90.0.0.1 * 255.255.255.255 UH eth0

90.0.0.0 * 255.0.0.0 U eth0

127.0.0.0 * 255.0.0.0 U lo

Thanks in advance. Bri

Question card works fine, but 'tcpdump' cant see any of the traffic

Date: 2000, Feb 17
From: Ron Flory <ron.flory@adtran.com>

hi-

 I've been using Dave's pcmcia stuff for a while now, (thanks Dave!).

 I have a PII RedHat 6.1 based laptop with a 10/100 network adapater that works just fine, however I'm a developer working with networks and I'd like to use the system for network sniffing.  For some reason, tcpdump is not able to see any of the traffic on eth0, even traffic from this machine.

 tcpdump seems to work file on all my other machines (7-8 of various types), but for some reason its deaf on this one.

 Any hints?  (something in /proc perhaps ?).

 I actually use ethereal as my packet sniffer, however since tcpdump is used by ethereal, and most everybody already has tcpdump, I'll reduce the problem to a single user program.

 Thanks-

ron

1. More Untitled by Ron Flory, 2000, Feb 17

Note Problem with ADDTRON Fast Ethernet card

Date: 2000, Feb 17
From: Andreas-Johann Ulvestad aj

I am currently having problems with my new pcmcia card. It is made by ADDTRON, and I belive it uses an intel chip (pcmcia loads the i82365 module, but it doesn't seem to work - I get 'Delaying eth0 initialization.' when trying to configure the interface.

--LOG--
Linux PCMCIA Card Services 3.1.8
 kernel build: 2.2.15-2.5.0 #1 Sat Feb 5 00:13:43 EST 2000
 options: [pci] [cardbus] [apmd] [pnp]
PnP: PNP BIOS installation structure at 0xc00f7fc0
PnP: PNP BIOS version 1.0, entry at f0000:a954, dseg at 400
PCI routing table version 1.0 at 0xfdf70
  unknown PCI interrupt router 1045:c700
Intel PCIC probe:
  TI 1131 PCI-to-CardBus at bus 0 slot 10, mem 0x68000000, 2 sockets
    host opts: [0]: [ring] [pci + serial irq] [no pci irq] [lat 168/176] [bus 32/34]
     host opts [1]: [ring] [pci + serial irq] [no pci irq] [lat 168/176] [bus 35/37]
     ISA irqs (scanned) = 3,4,7,9 polling interval = 1000 ms
cardmgr[646]: starting, version is 3.1.8
cardmgr[646]: wathcing 2 sockets
kernel: cs: IO port probe 0x1000-0x17ff: clean.
kernel: cs: IO port probe 0x0100-0x04ff: exluding 0x378-0x37f
kernel: cs: IO port probe 0x0a00-0x0aff: clean.
cardmgr[646]: initializing socket 0
kernel: cs: memory probe 0xa000000-0xa0fffff: clean.
cardmgr[646]: unsupported card in socket 0
cardmgr[646]: product info: "PCMCIA", "FastEthernet", "V", "1.0"
cardmgr[646]: manfid: 0x021b, 0x0202 function: 6 (network)

--END LOG--

Any ideas?
1. None It isn't an Intel network device by David Hinds, 2000, Feb 17
(_ More Done that, doesn't work by Andreas-Johann Ulvestad, 2000, Feb 17
(_ More af-680 ? by Poul Erik Laursen, 2000, Feb 18
1. None That card is not a true NE2000 by David Hinds, 2000, Feb 18
2. Question Someone got AF-680 (Addtron PCMCIA NIC) work? by Armando Montiel, 2000, Jul 10
1. None The regular pcnet_cs driver won't work by David Hinds, 2000, Jul 10
... 1 Message(s)

Question Beffalo LPC3-TX 10/100 NIC

Date: 2000, Feb 18
From: Ian Bone ianbone

Since I've had no responce from the company, thought I would try in here...can anyone help me, please.....Thanks.



Further to my comments below, I have spent more time with the Kernel and PCMCIA package drivers, and have the following message displayed when I insert the card:

cardmgr[128]: initializing socket 0
cardmgr[128]: socket 0: Buffalo LPC3-TX 10/10 PCMCIA-NIC
cardmgr[128]: executing : 'insmod /lib/modules/2.3.45/net/8390.o'
cardmgr[128]: executing : 'insmod /lib/modules/2.3.45/pcmcia/pcnet_cs.o'
pcnet_cs: ReleaseWindow: Bad handle
pcnet_cs: unable to read hardware net address
cardmgr[128]: get dev info on socket 0 failed: no such device

I also get something similar inserted in my /var/log/messages log file.

As you can see I have added the card definition tot he /etc/pcmcia/config
file as:

card "Baffalo LPC3-TX 10/100 PCMCIA-NIC"
   manfid 0x8a01, 0xc1ab
   bind "pcnet_cs"

Following instructions to try binding the card to the memory_cs module to
get the net address gave a similar "ReleaseWindow: Bad handle" error.

Using the DOS drivers on the CD that came with the card, I got an I/O Base
address of 0300H, and hardware address of 00:A0:0C:40:1E:0E, so I used this
in the pcnet_cs.c file in the hw_info definition and added
{ /* Buffalo LPC3-TX */ 0x0300, 0x00, 0xa0, 0x0c, 0 },

which compiled fine, although I doubt if I have the memory offset correct?

Can you offer assistance, please?

Many thanks in advance.


----- Original Message -----
From: Ian Bone
To: techsupport@techworks.co.uk
Sent: Wednesday, February 16, 2000 11:37 PM
Subject: LPC3-TX Under Linux


I am having a great deal of difficulty in getting my new LPC3-TX PCMCIA
network card to work under Linux RedHat 6.1. I have installed the latest
kernel 2.3.45, the latest PCMCIA package, version 3.1.11, and have
downloaded your Linux drivers from your web site, but I can not get them to
compile? Can you try compiling them yourself and see if they compile for
you. It looks like they have been ftp'ed using binary instead of ASCII at
some point? perhaps they could be placed in gziped tar file instead of a DOS
pkzip format?

Any comments would be greatly appreciated.

I am not a C programmer but it seems that my Makefile is not correct:
', needed by 'pcnet_cs.o'. Stop.'

Which looks like the Makefile has bad characters at the end of each line, or
the line:
pcnet_cs.o : pcnet_cs.c melco.h meclo_oem.h
is in the wrong format?

I have also tried to go down the route of not using your drivers and
configuring the PCMCIA drivers to handle this card, but I have not been
clever enough to get this to work either.

Thanks in advance.

Ian Bone
Database Administrator
National Australia Group
ianbone@ibone.co.uk
1. None Bad kernel choice by David Hinds, 2000, Feb 18

Sad ND5120-E Card seems to be alive, but cannot ping.

Date: 2000, Feb 22
From: Martin Andrews mdda

I believe that my card is 'alive', and pinging outside the machine causes the little lights to flash.  However, no other machine will respond to pings (nor will the Presario respond to their pings).

This is an install of RedHat 6.1 out-of-the-box.  Should I be recompiling the PCMCIA from your source?

Many thanks for any help you can give me.
Martin
:-)

Machine : Compaq Presario 1622, SohoWare ND5120-E 10bT card

:::: dmesg :::::
Linux version 2.2.12-20 (root@porky.devel.redhat.com) (gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)) #1 Mon Sep 27 10:25:54 EDT 1999
...
Linux PCMCIA Card Services 3.0.14
  kernel build: 2.2.12-20 #1 Mon Sep 27 10:25:54 EDT 1999
  options:  [pci] [cardbus] [apm]
Intel PCIC probe: 
  TI 1131 PCI-to-CardBus at bus 0 slot 10, mem 0x68000000, 2 sockets
    host opts [0]: [ring] [pci + serial irq] [no pci irq] [lat 168/176] [bus 32/34]
    host opts [1]: [ring] [pci + serial irq] [no pci irq] [lat 168/176] [bus 35/37]
    ISA irqs (scanned) = 3,4,7,10 status change on irq 10
cs: IO port probe 0x1000-0x14ff: clean.
cs: memory probe 0x60000000-0x60ffffff: clean.
eth0: NE2000 Compatible: io 0x1000, irq 3, hw_addr 00:C0:F0:1C:FF:FF
eth0: bogus packet size: 8481, status=0x21 nxpg=0x40.

:::: ifconfig status ::::
Socket 0:
  no card
Socket 1:
  5V 16-bit card present
  Function 0: ready

:::: ifconfig ident ::::
Socket 0:
  no product info available
Socket 1:
  product info: "NDC", "Ethernet", "A", "004743118001"
  function: 6 (network)

:::: ifconfig eth0 ::::
eth0      Link encap:Ethernet  HWaddr 00:C0:F0:1C:FF:FF  
          inet addr:192.168.1.12  Bcast:192.168.1.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:26 dropped:0 overruns:0 frame:1
          TX packets:9 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:100 
          Interrupt:3 Base address:0x1000 

:::: route ::::
Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
192.168.1.12    *               255.255.255.255 UH    0      0        0 eth0
192.168.1.0     *               255.255.255.0   U     0      0        0 eth0
127.0.0.0       *               255.0.0.0       U     0      0        0 lo

Question Dynalink L100C32 not recognized

Date: 2000, Feb 22
From: Jasper Spit jasper

Hi,

I have a Dynalink L100C32 pcmcia network adapter which isn't recognized by cardmgr. This is the output I get :

cardctl config:

Socket 0:
  Vcc 3.3V  Vpp1 3.3V  Vpp2 3.3v
  interface type is "cardbus"
  function 0:
Socket 1:
  not configured

cardctl ident:

Socket 0:
  product info: "PCMCIA", "CardBus Ethernet 10/100 (R5)"
  function 6 (network)
Socket 1:
  no product info available

when inserting the card, dmesg gives me this :

cs: cb_alloc(bus 32): vendor 0x10ec, device 0x8138

that's all :(

I already tried to modify the pcmcia config file by adding a new entry using manfid 0x10ec, 0x8138, but with no luck. (yes, I made cardmgr re-read the configfile).

Can anybody tell me how I can get this card to work ?

Kernel = 2.2.13

Thanks !!

jasper

1. None Wrong ID numbers by David Hinds, 2000, Feb 22
(_ Sad No luck by Jasper Spit, 2000, Feb 22

Question route add unnecessary for dhclient?

Date: 2000, Feb 24
From: Adam Goode <goodea@rpi.edu>

On line 60 of etc/network, there is the line

/sbin/route add default dev $DEVICE

which presumably adds the current device as the default gateway, when dhcp is enabled.

When I insert my 3com ethernet card, configured as a static interface in network.opts, it works properly as default route. When I plug in my wavelan card, it takes over as default route, which means it knocks out default route for the ethernet, and makes it go over the wavelan. Right now, I am not near a access point, so dhclient finds nothing on the wavelan and goes to sleep. But the card is configured as default route, which I don't want. When I comment out the route add line in etc/network, it works correctly. Is the line necessary? Doesn't dhclient automatically add the route as necessary? Seems like it, check out /etc/dhclient-script.

I am running Debian woody on x86.

Adam Goode

1. None I'm not sure by David Hinds, 2000, Feb 24
(_ Feedback dhclient for Linux by Adam Goode, 2000, Feb 24
(_ None Package name confusion, and a fix by David Hinds, 2000, Feb 24
(_ More Hmmm, almost, but not quite by Adam Goode, 2000, Feb 24
(_ None Argh by David Hinds, 2000, Feb 24

Question Wisecom PC-400 10/100 - Doesn`t work

Date: 2000, Feb 24
From: Francesc Bassas frankb

Hi to everybody,

I have a Dell Inspiron 3700 with a Wisecom PC-400 with 10/100 Ethernet.

My Linux version is 2.2.13 pcmcia version is 3.1.11

The output of "cardctl ident" is:

Socket 1:
  product info: "PCMCIA", "10/100 Ethernet Card", "", ""
  manfid: 0x8a01, 0xc1ab
  function: 6 (network)

In the startup I get this message:

 kernel: pcnet_cs: unable to read hardware net address

Reading the documentation I saw that one can hardwire the hardware address with an option, a that's what I did.

It seemed to work; in startup it got the address and it assigned an IRQ and a I/O port to the card.

I did a "ping" to a machine and it worked; but when I tried to do a "ftp" or "telnet" it blocked. Afterwards, I noticed that no machine can do a "ping" to this one; this machine doesn`t seem to answer.

I tried a NE2000 10 Mbps card, and it worked perfect.

Could anyone help me? Thanks in advance.

1. None The wrong driver is being loaded by David Hinds, 2000, Feb 24
(_ Sad It fails anyway by Francesc Bassas, 2000, Feb 25
(_ None Sorry, ignore that suggestion by David Hinds, 2000, Apr 04

News NetGear FA410TXC Setup

Date: 2000, Feb 24
From: Jochen Friedrich jochen1

Hi,

i am using a recently bought FA410TXC with pcmcia 3.1.10. The card is found, but apparently the transmitter isn't enable by default:

Feb 19 16:05:50 localhost cardmgr[380]: initializing socket 1
Feb 19 16:05:50 localhost cardmgr[380]: socket 1: NetGear FA410TXC Fast Ethernet
Feb 19 16:05:50 localhost cardmgr[380]: executing: 'insmod /lib/modules/2.2.14/net/8390.o'
Feb 19 16:05:51 localhost cardmgr[380]: executing: 'insmod /lib/modules/2.2.14/pcmcia/pcnet_cs.o'
Feb 19 16:05:51 localhost kernel: eth0: NE2000 Compatible: io 0x300, irq 7, hw_addr 00:E0:98:77:A6:21
Feb 19 16:05:51 localhost cardmgr[380]: executing: './network start eth0'
Feb 19 16:09:43 localhost automount[775]: running expiration on path /home
Feb 19 16:09:43 localhost automount[775]: expired /home/xedds10
Feb 19 16:32:39 localhost kernel: eth0: found link beat
Feb 19 16:32:54 localhost kernel: eth0: Tx timed out, cable problem? TSR=0x42, ISR=0x0, t=614.
Feb 19 16:33:04 localhost kernel: eth0: Tx timed out, cable problem? TSR=0x42, ISR=0x0, t=1000.
Feb 19 16:33:10 localhost kernel: eth0: lost link beat
Feb 19 16:33:14 localhost kernel: eth0: Tx timed out, cable problem? TSR=0x42, ISR=0x0, t=1000.
Feb 19 16:33:54 localhost last message repeated 4 times
Feb 19 16:34:14 localhost last message repeated 2 times
Feb 19 16:34:21 localhost kernel: eth0: interrupt from stopped card

I also recognize, that no interrupts are generated by the card, at all.

I had a closer look at the packet driver they ship and came up with this user space program to enable the transmitter and select speed + full duplex:

usage:

fa_select <interface> <option>

option:
0: 10 MBPS/HALF
1: 10 MBPS/FULL
2: 100MBPS/HALF
3: 100MBPS/FULL

I hope it's useful to someone...

--
Jochen


#include <sys/types.h>
#include <sys/socket.h>
#include <sys/ioctl.h>
#include <net/if.h>
#include <stdio.h>
#include <errno.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>

inline unsigned char
inb (unsigned short port)
{
  unsigned char _v;

  __asm__ __volatile__ ("inb %w1,%0":"=a" (_v):"Nd" (port));
  return _v;
}

inline void
outb (unsigned char value, unsigned short port)
{
  __asm__ __volatile__ ("outb %b0,%w1"::"a" (value), "Nd" (port));
}

static int sockets_open(void)
{
    int sock;
    if ((sock = socket(AF_INET, SOCK_DGRAM, 0)) != -1)
        return sock;
    else if ((sock = socket(AF_IPX, SOCK_DGRAM, 0)) != -1)
        return sock;
    else if ((sock = socket(AF_AX25, SOCK_DGRAM, 0)) != -1)
        return sock;
    else
        return socket(AF_APPLETALK, SOCK_DGRAM, 0);
}
void write_bit(int port, int bit)
{
    outb((bit << 6) + 0x20, port);
    usleep(1);
    outb((bit << 6) + 0xa0, port);
    usleep(1);
    outb((bit << 6) + 0x20, port);
}
int read_bit(int port)
{
    int i;
    outb(0,    port);
    usleep(1);
    outb(0x80, port);
    usleep(1);
    i = inb(port);
    outb(0,    port);
    return (i & 0x10) >> 4;
}
void reset(int port)
{
    outb(0x08, port);
    usleep(1);
    outb(0x0C, port);
    usleep(1);
    outb(0x08, port);
    usleep(1);
    outb(0x0C, port);
    outb(0x00, port);
}
int reada(int port, int adr)
{
    int i,j;

    for (i=0; i<0x20; i++)
      write_bit(port, 1);

    write_bit(port, 0);
    write_bit(port, 1);
    write_bit(port, 1);
    write_bit(port, 0);

    write_bit(port, 0);
    write_bit(port, 0);
    write_bit(port, 0);
    write_bit(port, 0);
    write_bit(port, 0);

    write_bit(port, (adr & 0x10) >> 4);
    write_bit(port, (adr & 0x08) >> 3);
    write_bit(port, (adr & 0x04) >> 2);
    write_bit(port, (adr & 0x02) >> 1);
    write_bit(port, (adr & 0x01) >> 0);

    j = read_bit(port);
    if (j == 1)
      j = read_bit(port);
    for (i=0; i<16; i++) {
      j = (j << 1) + read_bit(port);
    }
    write_bit(port, 1);
    return j;
}
int writea(int port, int adr, int val)
{
    int i,j;

    outb(0x08, port);
    usleep(1);
    outb(0x0C, port);
    usleep(1);
    outb(0x08, port);
    usleep(1);
    outb(0x0C, port);
    outb(0x00, port);

    for (i=0; i<0x20; i++)
      write_bit(port, 1);

    write_bit(port, 0);
    write_bit(port, 1);
    write_bit(port, 0);
    write_bit(port, 1);

    write_bit(port, 0);
    write_bit(port, 0);
    write_bit(port, 0);
    write_bit(port, 0);
    write_bit(port, 0);

    write_bit(port, (adr & 0x10) >> 4);
    write_bit(port, (adr & 0x08) >> 3);
    write_bit(port, (adr & 0x04) >> 2);
    write_bit(port, (adr & 0x02) >> 1);
    write_bit(port, (adr & 0x01) >> 0);

    write_bit(port, 1);
    write_bit(port, 0);

    write_bit(port, (val & 0x8000) >> 15);
    write_bit(port, (val & 0x4000) >> 14);
    write_bit(port, (val & 0x2000) >> 13);
    write_bit(port, (val & 0x1000) >> 12);
    write_bit(port, (val & 0x0800) >> 11);
    write_bit(port, (val & 0x0400) >> 10);
    write_bit(port, (val & 0x0200) >> 9);
    write_bit(port, (val & 0x0100) >> 8);
    write_bit(port, (val & 0x0080) >> 7);
    write_bit(port, (val & 0x0040) >> 6);
    write_bit(port, (val & 0x0020) >> 5);
    write_bit(port, (val & 0x0010) >> 4);
    write_bit(port, (val & 0x0008) >> 3);
    write_bit(port, (val & 0x0004) >> 2);
    write_bit(port, (val & 0x0002) >> 1);
    write_bit(port, (val & 0x0001) >> 0);

    write_bit(port, 1);
    return 0;
}
void main(int argc, char **argv)
{
    int skfd, i, sub, led;
    struct ifreq ifr;

    skfd = sockets_open();
    if (skfd == -1) {
        perror("socket");
        exit(1);
    }
    strcpy(ifr.ifr_name, argv[1]);
    if (ioctl(skfd, SIOCGIFMAP, &ifr) < 0) {
        perror("ioctl");
        exit(1);
    }
    i = atoi(argv[2]);
    switch(i) {
      case 0:
        sub = 0x0000;
        break;
      case 1:
        sub = 0x0100;
        break;
      case 2:
        sub = 0x2000;
        break;
      default:
        sub = 0x2100;
        break;
    }
    ioperm(ifr.ifr_map.base_addr+0x1c, 1, 1);
    reset(ifr.ifr_map.base_addr+0x1c);
    writea(ifr.ifr_map.base_addr+0x1c, 0, 0x8000);
    writea(ifr.ifr_map.base_addr+0x1c, 0, sub);
    close(skfd);
    exit(0);
}

Question DFE-650 not always Linux-compatible?

Date: 2000, Feb 25
From: Anthony Jameson AJameson

Subject: DFE-650 not always Linux-compatible?
Reply-to: Anthony Jameson <jameson@cs.uni-sb.de>
BCC: jameson@sol.cs.uni-sb.de
--text follows this line--
Hi,

The D-Link DFE-650TX is well known as a Linux-compatible Fast
Ethernet card. For example, two such cards have been found to
work fine on my Toshiba Satellite 4090XCDT running S.u.S.e. Linux 6.0 

But just a few days ago I bought a third such card for myself,
and it's been nothing but trouble, even as its supposedly
identical twins continue to work fine. It's recognized by the
card manager, and the pcnet_cs driver is duly loaded. But it's
impossible to connect to the network: I always get the error
message "No route to host". In /var/log/messages, there's always a message like this: 

Feb 24 22:14:40 helena kernel: eth0: Tx timed out, cable problem? TSR=0x42, ISR=0x0, t=1000.

Followng the advice in the PCMCIA how-to, I've tried juggling
the IRQs and the ports, but the result is always the same.

And this is not a unique experience: I've corresponded with a
Linux expert in northern Germany who spent days combating
exactly the same problem with his new DFE-650TX card, before giving up and returning the card to the dealer.

So it looks as if D-Link may have made some change in their
card that renders it Linux-incompatible (though they still
claim Linux-compatibility for the card on the packaging and on their web site).

            Anthony Jameson

Here's some debugging information:

Kernel version: 2.0.36
cardctl version 3.0.6

Output of "cardctl ident":
Socket 0:
  no product info available
Socket 1:
  product info: "D-Link", "DFE-650", "Fast Ethernet", "Rev. D1"
  manfid: 0x0149, 0x0230
  function: 6 (network)

Output of "cardctl status":
Socket 0:
  no card
Socket 1:
  5V 16-bit card present
  Function 0: 

Typical sequence in /var/log/messages, starting with a system reboot:
Feb 24 22:13:51 helena syslogd 1.3-3: restart.
Feb 24 22:13:52 helena kernel: klogd 1.3-3, log source = /proc/kmsg started.
Feb 24 22:13:52 helena kernel: Inspecting /boot/System.map
Feb 24 22:13:52 helena kernel: Loaded 5086 symbols from /boot/System.map.
Feb 24 22:13:52 helena kernel: Symbols match kernel version 2.0.36.
Feb 24 22:13:52 helena kernel: Loaded 18 symbols from 6 modules.
Feb 24 22:13:52 helena kernel: Linux PCMCIA Card Services 3.0.6
Feb 24 22:13:52 helena kernel:   kernel build: 2.0.36 #1 Fri Dec 11 16:26:51 /etc/localtime 1998
Feb 24 22:13:52 helena kernel:   options:  [pci] [cardbus]
Feb 24 22:13:52 helena kernel: Intel PCIC probe: 
Feb 24 22:13:52 helena kernel:   Toshiba ToPIC97 PCI-to-CardBus at bus 0 slot 2, mem 0x68000000, 2 sockets
Feb 24 22:13:52 helena kernel:     host opts [0]: [slot 0xf0] [ccr 0x10] [cdr 0x86] [rcr 0x02] [no pci irq] [lat 168/176] [bus 20/20]
Feb 24 22:13:52 helena kernel:     host opts [1]: [slot 0xf0] [ccr 0x11] [cdr 0x86] [rcr 0x02] [no pci irq] [lat 168/176] [bus 21/21]
Feb 24 22:13:52 helena kernel:     ISA irqs (default) = 3,4,5,7,9,10,11,12 polling interval = 1000 ms
Feb 24 22:13:52 helena kernel: cs: IO port probe 0x1000-0x17ff: clean.
Feb 24 22:13:52 helena kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x378-0x37f 0x4d0-0x4d7
Feb 24 22:13:52 helena kernel: cs: IO port probe 0x0a00-0x0aff: clean.
Feb 24 22:13:52 helena kernel: cs: memory probe 0x0d0000-0x0dffff: clean.
Feb 24 22:13:52 helena kernel: eth0: NE2000 Compatible: port 0x300, irq 9, hw_addr 00:E0:98:77:94:FD
Feb 24 22:13:53 helena ypbind[124]: unable to register (YPBINDPROG, YPBINDVERS, udp). : Connection refused 
# Is the above relevant here?
Feb 24 22:13:54 helena lpd[147]: restarted
Feb 24 22:14:18 helena apmd[255]: Version 3.0final (APM BIOS 1.1, Linux driver 1.2)
Feb 24 22:14:18 helena apmd[255]: Charge: * * * (94% 2:38)
Feb 24 22:14:22 helena su: (to jameson) root on /dev/ttyp2
Feb 24 22:14:32 helena apmd[255]: Now using AC Power
Feb 24 22:14:40 helena kernel: eth0: Tx timed out, cable problem? TSR=0x42, ISR=0x0, t=1000.
Feb 24 22:15:45 helena kernel: eth0: Tx timed out, cable problem? TSR=0x42, ISR=0x0, t=6500.
Feb 24 22:17:45 helena kernel: eth0: Tx timed out, cable problem? TSR=0x42, ISR=0x0, t=12000.

/etc/pcmcia/config.opts: (Don't know if there's anything nonstandard here.)
#
# Local PCMCIA Configuration File
#
# System resources available for PCMCIA devices
#
include port 0x100-0x4ff, port 0x1000-0x17ff
include memory 0xc0000-0xfffff
include memory 0xa0000000-0xa0ffffff, memory 0x60000000-0x60ffffff
#
# Extra port range for IBM Token Ring
#
include port 0xa00-0xaff
#
# Resources we should not use, even if they appear to be available
#
# First built-in serial port
exclude irq 4
# Second built-in serial port
exclude irq 3

# First built-in parallel port
exclude irq 7
1. Idea Try This by Jochen Friedrich, 2000, Feb 25
1. Feedback Hope others can try by Anthony Jameson, 2000, Feb 26
1. Feedback tried and it works by Ingo Ciechowski, 2000, Feb 26
... 3 Message(s)
2. Question bad performance with Jochen's patch and DFE-650 by Ingo Ciechowski, 2000, Feb 26
1. None Performance should be equally bad under Windows by David Hinds, 2000, Feb 27
... 8 Message(s)
2. Agree NetGear FA410TXC Tool works for D-Link DE-650 by Ingo Ciechowski, 2000, Feb 26
1. Angry HELP! *Still* can't get D-Link DFE-650TX going! (Day 5) by Harold Boyer, 2000, Aug 18
(_ None Re: Angry: HELP! *Still* can't get D-Link DFE-650TX going! (Day 5) by David Hinds, 2000, Aug 21

Question D-Link DE-650 PCMCIA ethernet card

Date: 2000, Feb 25
From: Phil Brooke pjbrooke

Hi,

I have a problem with a

  D-Link DE-650 PCMCIA ethernet card (+RJ45 `tail')

when using this with a Gateway 2000 Solo 2100 laptop running Debian 2.0, kernel version 2.0.34 and PCMCIA Card Services 3.1.11.

The symptoms are:

- The card appears to configure okay. (Two high beeps, and ifconfig seems okay -- output shortly.)

- No traffic appears to get off the card.

- The light on the tail with the RJ45 socket never comes on.

ifconfig reports:

 eth0     Link encap:Ethernet  HWaddr 00:80:C8:85:81:39  
          inet addr:192.168.16.99  Bcast:192.168.16.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
          Collisions:0 
          Interrupt:3 Base address:0x300 

/proc/net/dev never shows any receive packets; only transmit packets.

The only other anomaly I can find is related to when I remove the card. Sometimes, there is a series of messages (from 19:46:14) like this when the card is ejected:

 Feb 24 19:36:47 diamond syslogd 1.3-3#26: restart.
 Feb 24 19:36:47 diamond kernel: Linux PCMCIA Card Services 3.1.11 
 Feb 24 19:36:47 diamond kernel:   kernel build: 2.0.34 #2 Tue Aug 3 20:22:30 GMT 1999 
 Feb 24 19:36:47 diamond kernel:   options:  [pci] [cardbus] [apm] 
 Feb 24 19:36:47 diamond kernel: Intel PCIC probe:  
 Feb 24 19:36:47 diamond kernel:   Cirrus PD6729 PCI-to-PCMCIA at slot 00:13, port 0xfcfc 
 Feb 24 19:36:47 diamond kernel:     host opts [0]: [ring] [1/3/2] [1/9/2] 
 Feb 24 19:36:47 diamond kernel:     host opts [1]: [ring] [1/3/2] [1/9/2] 
 Feb 24 19:36:47 diamond kernel:     ISA irqs (default) = 3,4,5,9,10,11,12 polling interval = 1000 ms 
 Feb 24 19:36:48 diamond kernel: cs: IO port probe 0x0100-0x03ff: excluding 0x330-0x337 0x370-0x37f 0x398-0x39f 
 Feb 24 19:36:48 diamond kernel: cs: IO port probe 0x0a00-0x0aff: clean. 
 Feb 24 19:36:48 diamond kernel: cs: memory probe 0x0d0000-0x0dffff: clean. 
 Feb 24 19:36:48 diamond kernel: loading device 'eth0'... 
 Feb 24 19:36:48 diamond kernel: eth0: NE2000 Compatible: io 0x300, irq 3, hw_addr 00:80:C8:85:81:39 
 Feb 24 19:46:14 diamond kernel: eth0: interrupt(s) dropped! 
 Feb 24 19:46:14 diamond kernel: eth0: mismatched read page pointers  0 vs 4c. 
 Feb 24 19:46:14 diamond kernel: eth0: mismatched read page pointers  0 vs ff. 
 Feb 24 19:46:14 diamond last message repeated 7 times
 Feb 24 19:46:14 diamond kernel: eth0: unexpected TX-done interrupt, lasttx=20. 
[lots of lines snipped]
 Feb 24 19:46:15 diamond kernel: eth0: Too much work at interrupt, status 0xff 
 Feb 24 19:46:15 diamond kernel: eth0: interrupt from stopped card 

I've checked that the network cable to the hub is okay. The card and tail were reported as working okay before I bought it (however, I have no means to check it with another OS). I *think* I've excluded IRQ conflicts (the laptop IR port was allegedly on IRQ 3; that's now disabled).

The laptop has happily used a modem card before.

I've now run out of things to check. Have I missed anything I can do? (I know the answer is almost certainly `yes', but I can't think of what they are.)

Many thanks in advance,

Phil.

-- Phillip J. Brooke pjb@mithlond48.freeserve.co.uk

1. None Maybe a bad card?? by David Hinds, 2000, Feb 25
1. Note Possibly... by Phil Brooke, 2000, Feb 27
(_ None Bad hardware seems likely by David Hinds, 2000, Feb 28
2. Idea can you try it under Windows? by Ingo Ciechowski, 2000, Feb 27

Idea Nit in 3.0.14 pcmcia/network script

Date: 2000, Feb 26
From: Glen Ditchfield gjditchfield

My network.opts script used to contain

    # Host's IP address, netmask, network address, broadcast address
    IPADDR="192.168.0.1"
    #NETMASK=""
    #NETWORK=""
    #BROADCAST=""

This worked, but /var/log/messages contained warning messages:

	...cardmgr[361]: executing: './network start eth0'
	...cardmgr[361]: + Usage:
	...cardmgr[361]: +   ifconfig [-a] [-i] [-v] <interface> [[<AF>] <address>]
	...cardmgr[361]: +   [add <address>[/<prefixlen>]]
	...

The problem was that NETWORK wasn't set when network executed the lines

        # Basic network setup
        BC=${BROADCAST:+broadcast $BROADCAST}
        /sbin/ifconfig $DEVICE up $IPADDR netmask $NETMASK $BC

Shouldn't the 'netmask' parameter be guarded in the same way as 'broadcast'?

***************
*** 75,86 ****
      if [ "$IPADDR" ] ; then

  	# Basic network setup
  	BC=${BROADCAST:+broadcast $BROADCAST}
! 	/sbin/ifconfig $DEVICE up $IPADDR netmask $NETMASK $BC

  	if [ "$NETWORK" ] ; then
  	    /sbin/ifuser $DEVICE $NETWORK || \
! 		/sbin/route add -net $NETWORK netmask $NETMASK dev $DEVICE
  	elif [ "$GATEWAY" ] ; then
  	    /sbin/ifuser $DEVICE $GATEWAY || \
  		/sbin/route add $GATEWAY $DEVICE
--- 75,87 ----
      if [ "$IPADDR" ] ; then

  	# Basic network setup
+ 	NM=${NETMASK:+netmask $NETMASK}
  	BC=${BROADCAST:+broadcast $BROADCAST}
! 	/sbin/ifconfig $DEVICE up $IPADDR $NM $BC

  	if [ "$NETWORK" ] ; then
  	    /sbin/ifuser $DEVICE $NETWORK || \
! 		/sbin/route add -net $NETWORK $NM dev $DEVICE
  	elif [ "$GATEWAY" ] ; then
  	    /sbin/ifuser $DEVICE $GATEWAY || \
  		/sbin/route add $GATEWAY $DEVICE

1. None You're right by David Hinds, 2000, Feb 28

Sad eth0: Tx hung, 11 vs. 0. error with Abocom FE2000 and similar with D-Link DFE-650 cards

Date: 2000, Feb 26
From: Ingo Ciechowski inki

I guess I'm about to get crazy upon this problem. If I only was somewhat deeper in driver development I'd love to figure it out by myself, but I know similar problems have been posted earlier and you sure know how to get it going.

I've got a "Gericom Hydro-Speed" no-name notebook with AMD K6-2 CPU running SuSE Linux 2.2.13 and meanwhile the latest experimental pcmcia package version 3.1.12 (tried with 3.1.10 and 3.1.11 as well).

This system shall work with an Abocom FE2000 10/100 fast ethernet cardbus network card, but it ends finally with " eth0: Tx hung, 11 vs. 0. " messages as described below.
I finally decided to purchase the D-Link DFE-650 16bit fast ethernet card which was in the list of supported devices and ended up with " eth0: Tx timed out, cable problem? TSR=0x42, ISR=0x0, t=902. " messages which look pretty similar for me :-((

BTW: Both cards work well under Windows98 and I also have a 16bit Farallon/3Com 10BaseT card which works under linux.

I'd love to get the FE2000 32bit card to work - and if it only was the D-Link DFE-650 16bit card that'd be great as well.


In /etc/pcmcia/network.opts I inserted some local network information which works fine with the 3Com card:
 ..
    IPADDR="194.162.209.41"
    NETMASK="255.255.255.192"
    NETWORK="194.162.209.0"
    BROADCAST="194.162.209.63"
    # Gateway address for static routing
    GATEWAY="194.162.209.1"
 ..

And since I read that this helped someone else with a similar problem I commented out the line
 exclude irq 3
in /etc/pcmcia/config.opts - also excluded other interrupts as well, but with no success either.


Here's what gets logged in /var/log/messages upon startup with no card inserted:

 Feb 26 16:29:31 training2 kernel: Linux PCMCIA Card Services 3.1.12
 Feb 26 16:29:31 training2 kernel: kernel build: 2.2.13 #3 Sat Feb 26 16:13:07 CET 2000
 Feb 26 16:29:31 training2 kernel: options: [pci] [cardbus] [pnp]
 Feb 26 16:29:31 training2 kernel: PCI routing table version 1.0 at 0xfdf70
 Feb 26 16:29:31 training2 kernel: unknown PCI interrupt router 1106:0686
 Feb 26 16:29:31 training2 kernel: PnP: PNP BIOS installation structure at 0xc00f7580
 Feb 26 16:29:31 training2 kernel: PnP: PNP BIOS version 1.0, entry at f0000:a0e5, dseg at 400
 Feb 26 16:29:31 training2 kernel: Intel PCIC probe:
 Feb 26 16:29:31 training2 kernel: O2Micro OZ6832/OZ6833 PCI-to-CardBus at slot 00:0a, mem 0x68000000
 Feb 26 16:29:31 training2 kernel: host opts [0]: [a 24] [b 21] [c 20] [d 02] [mhpg 18] [fifo 00] [e 2a] [no
 Feb 26 16:29:31 training2 kernel: host opts [1]: [a 24] [b 21] [c 20] [d 02] [mhpg 18] [fifo 00] [e 2a] [no
 Feb 26 16:29:31 training2 kernel: ISA irqs (default) = 3,9,10,11 polling interval = 1000 ms
 Feb 26 16:29:31 training2 kernel: cs: IO port probe 0x0c00-0x0cff: clean.
 Feb 26 16:29:31 training2 kernel: cs: IO port probe 0x0800-0x08ff: clean.
 Feb 26 16:29:31 training2 kernel: cs: IO port probe 0x0100-0x04ff: clean.
 Feb 26 16:29:31 training2 kernel: cs: IO port probe 0x0a00-0x0aff: clean.


The following information was gathered about the Abocom FE2000 card :
                                        
 /var/log/messages:
 Feb 26 16:58:45 training2 kernel: cs: cb_alloc(bus 35): vendor 0x1011, device 0x0019
 Feb 26 16:58:45 training2 cardmgr[467]: initializing socket 1
 Feb 26 16:58:45 training2 cardmgr[467]: socket 1: Linksys EtherFast 10/100
 Feb 26 16:58:45 training2 cardmgr[467]: executing: 'insmod /lib/modules/2.2.13/pcmcia/cb_enabler.o'
 Feb 26 16:58:45 training2 cardmgr[467]: executing: 'insmod /lib/modules/2.2.13/pcmcia/tulip_cb.o'
 Feb 26 16:58:45 training2 kernel: cs: cb_config(bus 35)
 Feb 26 16:58:45 training2 kernel: fn 0 bar 1: io 0x200-0x27f
 Feb 26 16:58:45 training2 kernel: fn 0 bar 2: mem 0xa00c0000-0xa00c03ff
 Feb 26 16:58:45 training2 kernel: fn 0 rom: mem 0xa0080000-0xa00bffff
 Feb 26 16:58:45 training2 kernel: cs: cb_enable(bus 35)
 Feb 26 16:58:45 training2 kernel: bridge io map 0 (flags 0x21): 0x200-0x27f
 Feb 26 16:58:45 training2 kernel: bridge mem map 0 (flags 0x1): 0xa0080000-0xa00c0fff
 Feb 26 16:58:45 training2 kernel: tulip_reap()
 Feb 26 16:58:45 training2 kernel: tulip_attach(bus 35, function 0)
 Feb 26 16:58:45 training2 kernel: tulip.c:v0.91g-ppc 7/16/99 becker@cesdis.gsfc.nasa.gov (modified by danilo@cs.uni-magdeburg.de for XIRCOM CBE, fixed by Doug Ledford)
 Feb 26 16:58:45 training2 kernel: eth0: Digital DS21143 Tulip rev 65 at 0x200, 00:E0:98:7A:6A:40, IRQ 9.
 Feb 26 16:58:45 training2 kernel: eth0: EEPROM default media type Autosense.
 Feb 26 16:58:45 training2 kernel: eth0: Index #0 - Media MII (#11) described by a 21142 MII PHY (3) block.
 Feb 26 16:58:45 training2 kernel: eth0: MII transceiver #0 config 3000 status 7809 advertising 01e1.
 Feb 26 16:58:45 training2 cardmgr[467]: executing: './network start eth0'
 Feb 26 16:58:49 training2 kernel: eth0: Tx hung, 11 vs. 0.
  

 bash-2.03# ifconfig eth0
 eth0 Link encap:Ethernet HWaddr 00:E0:98:7A:6A:40
          inet addr:194.162.209.41 Bcast:194.162.209.63 Mask:255.255.255.192
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:1 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:100
          Interrupt:9 Base address:0x200


 (while trying to ping 194.162.209.1)
 bash-2.03# tcpdump -nlieth0
 User level filter, protocol ALL, datagram packet socket
 tcpdump: listening on eth0
 17:03:47.209806 arp who-has 194.162.209.1 tell 194.162.209.41 (0:e0:98:7a:6a:40)
 17:03:48.201157 arp who-has 194.162.209.1 tell 194.162.209.41 (0:e0:98:7a:6a:40)



 bash-2.03# cardctl config
 Socket 0:
  not configured
 Socket 1:
  Vcc 3.3V Vpp1 3.3V Vpp2 3.3V
  interface type is "cardbus"
  irq 9 [exclusive] [level]
  function 0:
    io 0x0200-0x027f [32bit]

 bash-2.03# tulip-diag -a
 tulip-diag.c:v1.03 5/20/98 Donald Becker (becker@cesdis.gsfc.nasa.gov)
 Chip Index #1: Found a DC21142 Tulip card at PCI bus 35, device 0 I/O 0x200.
 Digital DS21142 Tulip chip registers at 0x200:
  f8a08000 ffffffff ffffffff 00ab6810 00ab6a10 f0000146 b20e0000 f3fe0000
  e0000000 ffffcbf8 ffffffff 00000000 000000c6 ffff0000 fff80000 8ff90000
  ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff
  ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff
 The Rx process state is 'Stopped'.
 The Tx process state is 'Stopped'.
 Transmit stopped, Receive stopped, half-duplex.
 The transmit threshold is 128.
 Port selection is MII, half-duplex.
 EEPROM transceiver/media description for the DC21142 chip.
 
 Leaf node at offset 128, default media type 0000 (10baseT).
 0 transceiver description blocks:

 bash-2.03# tulip-diag -e
 tulip-diag.c:v1.03 5/20/98 Donald Becker (becker@cesdis.gsfc.nasa.gov)
 Chip Index #1: Found a DC21142 Tulip card at PCI bus 35, device 0 I/O 0x200.
 EEPROM transceiver/media description for the DC21142 chip.
 
 Leaf node at offset 128, default media type 0000 (10baseT).
  0 transceiver description blocks:

 bash-2.03# tulip-diag -m
 tulip-diag.c:v1.03 5/20/98 Donald Becker (becker@cesdis.gsfc.nasa.gov)
 Chip Index #1: Found a DC21142 Tulip card at PCI bus 35, device 0 I/O 0x200.
 EEPROM transceiver/media description for the DC21142 chip.
 
 Leaf node at offset 128, default media type 0000 (10baseT).
 0 transceiver description blocks:
 MII PHY found at address 0, status 0x782d.
 MII PHY #0 transceiver registers:
   3000 782d 0040 6212 01e1 0021 0000 0000
   0000 0000 0000 0000 0000 0000 0000 0000
   1000 0000 0000 0000 0000 0000 0100 0000
   003c d006 0f00 ff00 002c 0000 0080 000b.
  Internal autonegotiation state is 'Autonegotiation disabled'.

-------------------------------------------------------------------------------------------

And here's what I figured out about the D-Link DFE-650 card:

 /var/log/messages:
 Feb 26 17:09:18 training2 cardmgr[467]: initializing socket 1
 Feb 26 17:09:18 training2 cardmgr[467]: socket 1: KTI ETHER-C16 Fast ethernet
 Feb 26 17:09:18 training2 cardmgr[467]: executing: 'insmod /lib/modules/2.2.13/net/8390.o
'
 Feb 26 17:09:18 training2 cardmgr[467]: executing: 'insmod /lib/modules/2.2.13/pcmcia/pcn
et_cs.o'
 Feb 26 17:09:18 training2 kernel: eth0: NE2000 Compatible: io 0x300, irq 9, hw_addr 00:E0
:98:77:92:8F
 Feb 26 17:09:18 training2 cardmgr[467]: executing: './network start eth0'
 Feb 26 17:09:20 training2 kernel: eth0: found link beat
 Feb 26 17:09:38 training2 kernel: eth0: Tx timed out, cable problem? TSR=0x42, ISR=0x0, t=895.
 Feb 26 17:09:48 training2 kernel: eth0: Tx timed out, cable problem? TSR=0x42, ISR=0x0, t=1000.


 bash-2.03# ifconfig eth0
 eth0 Link encap:Ethernet HWaddr 00:E0:98:77:92:8F
          inet addr:194.162.209.41 Bcast:194.162.209.63 Mask:255.255.255.192
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:0 errors:4 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:100
          Interrupt:9 Base address:0x300


 (while trying to ping 194.162.209.1)
 bash-2.03# tcpdump -nlieth0
 User level filter, protocol ALL, datagram packet socket
tcpdump: listening on eth0
 17:09:58.931129 arp who-has 194.162.209.1 tell 194.162.209.41 (0:e0:98:77:92:8f)
 17:09:58.981151 arp who-has 194.162.209.1 tell 194.162.209.41 (0:e0:98:77:92:8f)
 17:10:08.931127 arp who-has 194.162.209.1 tell 194.162.209.41 (0:e0:98:77:92:8f)


 bash-2.03# cardctl config
 Socket 0:
  not configured
 Socket 1:
  Vcc 5.0V Vpp1 0.0V Vpp2 0.0V
  interface type is "memory and I/O"
  irq 9 [exclusive] [level]
  function 0:
    config base 0x0400
      option 0x60 status 0x00 copy 0x00
    io 0x0300-0x031f [auto]



---


Of course I'm willing to add any additional useful information immediately if that'd help to solve my problem.


Thanks a lot for your patience,

Ingo
Ciechowski@cis-computer.com
1. News D-Link DFE-650 problem solved - FE2000 still not working by Ingo Ciechowski, 2000, Feb 26

Idea Fiberline FL-4680 working!

Date: 2000, Feb 28
From: CW Zuckschwerdt zany

I recently bought a
 FIBERLINE FL-4680
 16-Bit PCMCIA NIC

It is working great with pcnet_cs driver.

Although cpu load on my 166MHz pentium is high I reach the full 10,000 KBit/sec.

This PC-Card was amazingly cheap! I paid EUR 35,- (about the same amount in USD)

Question No success with either 10/100 Etherjet nor 3Com cards on Thinkpad

Date: 2000, Mar 01
From: lyle borg-graham lyle

I have gotten the same "SIOCADDRT: Network is unreachable" message with both the IBM 10/100 EtherJet card and the 3Com 10/100 3CXFE575CT card. The machine is a Thinkpad 600E, running Red Hat 6.1, kernel 2.2.14, and using the pcmcia-cs-3.1.11 distribution. Apparently all the network configuration files are ok. I note some bizzare things, such as the ability to ping locally even when the card is not installed.

As perhaps a side note, at some point in this process the cardmgr started sending messages directly to the screen (i.e. "Network unreachable") during bootup, and *not* to /var/log/messages as it did before.

Thanks very much for any ideas!

Lyle

Included below are some (hopefully) useful tidbits (with the 3Com card installed):

   [root@cognilap ~]# uname -rv
   2.2.14 #8 Tue Feb 29 17:13:24 CET 2000

   [root@cognilap ~]# cardctl -V
   cardctl version 3.1.11

   [root@cognilap ~]# sh -x /etc/pcmcia/network start eth0
   + [ -r ./shared ]
   + . /etc/pcmcia/shared
   ++ [ 2 -lt 2 ]
   ++ ACTION=start
   ++ DEVICE=eth0
   + get_info eth0
   + [ -d /var/state/pcmcia ]
   ++ cat /var/state/pcmcia/scheme
   + SCHEME=
   + STAB=/var/state/pcmcia/stab
   + [ ! ]
   + SCHEME=default
   + grep_stab eth0
   + local CLASS DEV
   + read SOCKET CLASS DRIVER INSTANCE DEV MAJOR MINOR
   + [ eth0 = ]
   + read SOCKET CLASS DRIVER INSTANCE DEV MAJOR MINOR
   + [ eth0 = Fast ]
   + read SOCKET CLASS DRIVER INSTANCE DEV MAJOR MINOR
   + [ eth0 = eth0 ]
   + return 0
   ++ /sbin/ifconfig eth0
   ++ sed -n -e s/.*addr \([^ ]*\) */\1/p
   + HWADDR=00:00:86:51:E6:E4
   + ADDRESS=default,1,0,00:00:86:51:E6:E4
   + . /etc/pcmcia/network.opts
   ++ [ -f /etc/sysconfig/network-scripts/ifcfg-eth0 ]
   ++ BOOTP=n
   ++ IPADDR=128.122.146.142
   ++ NETWORK=128.122.146.0
   ++ NETMASK=255.255.255.0
   ++ BROADCAST=128.122.255.255
   ++ GATEWAY=128.122.146.1
   ++ DOMAIN=cns.nyu.edu
   + RESOLV=/etc/resolv.conf
   + test
   + is_true
   + [ = y -o = Y -o = yes -o = YES ]
   + is_true n
   + [ n = y -o n = Y -o n = yes -o n = YES ]
   + is_true
   + [ = y -o = Y -o = yes -o = YES ]
   + [ 128.122.146.142 ]
   + BC=broadcast 128.122.255.255
   + /sbin/ifconfig eth0 up 128.122.146.142 netmask 255.255.255.0 broadcast 128.122.255.255
   + [ 128.122.146.0 ]
   + /sbin/ifuser eth0 128.122.146.0
   + /sbin/route add -net 128.122.146.0 netmask 255.255.255.0 dev eth0
   + test 128.122.146.1
   + /sbin/route add default gw 128.122.146.1 metric 1
   SIOCADDRT: File exists
   + [ cns.nyu.edu ]
   + echo # eth0 begin
   + test cns.nyu.edu
   + echo domain cns.nyu.edu
   + test
   + echo # eth0 end
   + sed -e /# eth0 begin/,/# eth0 end/d /etc/resolv.conf
   + mv /etc/resolv.conf.N /etc/resolv.conf
   + [ ]
   + [ ]
   + start_fn eth0
   + /sbin/ifup eth0
   SIOCADDRT: Network is unreachable
   + exit 0


Now, with the cable *unplugged* (128.122.154.50 is another machine):

   [root@cognilap ~]# ping 128.122.154.50
   PING 128.122.154.50 (128.122.154.50) from 128.122.154.50 : 56(84) bytes of data.
   64 bytes from 128.122.154.50: icmp_seq=0 ttl=255 time=0.1 ms
   64 bytes from 128.122.154.50: icmp_seq=1 ttl=255 time=0.1 ms
   64 bytes from 128.122.154.50: icmp_seq=2 ttl=255 time=0.1 ms
   64 bytes from 128.122.154.50: icmp_seq=3 ttl=255 time=0.0 ms

   --- 128.122.154.50 ping statistics ---
   4 packets transmitted, 4 packets received, 0% packet loss
   round-trip min/avg/max = 0.0/0.0/0.1 ms



In the file /etc/pcmcia/network.opts:

       if [ -f /etc/sysconfig/network-scripts/ifcfg-eth0 ] ; then
           start_fn () {
               /sbin/ifup $1
           }
           stop_fn () {
               /sbin/ifdown $1
           }
       fi



       case "$ADDRESS" in
       *,*,*,*)

           BOOTP="n"
           IPADDR="128.122.146.142"
           NETWORK="128.122.146.0"
           NETMASK="255.255.255.0"
   BROADCAST="128.122.255.255"
           GATEWAY="128.122.146.1"
           DOMAIN="cns.nyu.edu"

           ;;
       esac


In the file /etc/pcmcia/config.opts:

#
# Local PCMCIA Configuration File
#
# System resources available for PCMCIA devices
#
include port 0x100-0x4ff, port 0x1000-0x17ff
include memory 0xc0000-0xfffff
include memory 0xa0000000-0xa0ffffff, memory 0x60000000-0x60ffffff
#
# Extra port range for IBM Token Ring
#
include port 0xa00-0xaff
#
# Resources we should not use, even if they appear to be available
#
# First built-in serial port
exclude irq 4
# Second built-in serial port
#exclude irq 3
# First built-in parallel port
exclude irq 7
#
# Additional stuff from the PCMCIA-HOWTO
exclude port 0x230-0x233
exclude irq 5
exclude port 0x2f8-0x2ff

# Modifications from A. Krumeich
include port 0x600-0x6ff
include irq 11

# Examples of options for loadable modules
#
# To fix sluggish network with IBM ethernet adapter...
#module "pcnet_cs" opts "mem_speed=600"
#
# Options for Xircom Netwave driver...
#module "netwave_cs" opts "domain=0x100 scramble_key=0x0"
#
# Options for IBM Token Ring adapters
#module "ibmtr_cs" opts "mmiobase=0xd0000 srambase=0xd4000"
1. More Addendum by lyle borg-graham, 2000, Mar 01
2. None This is a configuration error by David Hinds, 2000, Mar 01
1. None Progress, but... by lyle borg-graham, 2000, Mar 01

Question CardBus and Product Info

Date: 2000, Mar 01
From: Sergio Talens-Oliag sto

I've seen in a message from David Hinds that the CardBus cards that do not provide product info are not supported under linux.

I have a Dynalink L100C32-2 and I know of other cards with the same problem (Argosy 10/100 EN-235, Conceptronics CON100TC).

The questions are:

  • Does anyone knows what changes are needed to make the pcmcia package work with this kind of cards?
  • Is there any documentation I can read to try to add support for them?

Thanks in advance,

    Sergio

1. None Re: Question: CardBus and Product Info by David Hinds, 2000, Mar 06
(_ News Dynalink L100C32 and Product Info by Sergio Talens-Oliag, 2000, Mar 07

Question Best PC Card Ethernet Interface for 10/100, DHCP under RH6.1

Date: 2000, Mar 07
From: Dave Aiello dave_aiello

I am new to Linux, although I am an experienced Solaris end user. I bought a RedHat 6.1 certified laptop from Dell the other day, with the idea of totally immersing myself in Linux. I do not want to spend the time fooling with hardware configurations unless I have to.

My configuration is a Dell Inspiron 7500 with a 3com 3CCFE575 running on the RH 6.1 installation provided by Dell with minimal configuration edits.

I want a 10/100 Ethernet PC Card that is capable of running DHCP without any hiccups. I am willing to go out and buy the most trouble free PC Card of this nature (and abandon the 3CCFE575 if necessary).

The problem I am having is that I cannot determine from reading these message forums if any PC Card NICs can live up to my wishlist. Are there any obvious choices that would satisfy me?

Thanks,

Dave Aiello
CTDATA

Question Thinkpad 600E and IBM etherjet 100/10

Date: 2000, Mar 11
From: Andrew Fong jester

I have installed the latest version of the pcmcia package. /var/log/messages reports that there is a card in the slot. I get two high tone beeps reporting that the card was intialized. I am not able to ping anything besides the local machine. I did change the memory addresses in /etc/pcmcia/config.opts. Any help would be appriciated. -Andrew Fong

1. More I dont get lights by Andrew Fong, 2000, Mar 11
2. None More information? by David Hinds, 2000, Mar 13

Question Argosy EN-235 continued.

Date: 2000, Mar 15
From: Robert Rossmann rbr

I installed the new pcmcia package 3.1.13 and now the card is found but it won't work. This are the entries in the /var/log/messages file

This is for tulip

Mar 15 14:22:07 armin cardmgr[785]: starting, version is 3.1.13
Mar 15 14:22:07 armin cardmgr[785]: watching 2 sockets
Mar 15 14:22:07 armin kernel: cs: IO port probe 0x1000-0x17ff: clean.
Mar 15 14:22:07 armin kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x200-0x207 0x220-0x22f 0x300-0x307 0x378-0x37f 0x388-0x38f 0x398-0x39f 0x400-0x40f 0x480-0x48f 0x4c0-0x4df
Mar 15 14:22:07 armin kernel: cs: IO port probe 0x0a00-0x0aff: clean.
Mar 15 14:22:07 armin cardmgr[785]: initializing socket 0
Mar 15 14:22:07 armin cardmgr[785]: socket 0: Argosy EN-235 Fast Ethernet
Mar 15 14:22:07 armin cardmgr[785]: executing: 'insmod /lib/modules/2.2.13/pcmcia/cb_enabler.o'
Mar 15 14:22:07 armin cardmgr[785]: executing: 'insmod /lib/modules/2.2.13/pcmcia/tulip_cb.o'
Mar 15 14:22:07 armin kernel: cs: cb_config(bus 2)
Mar 15 14:22:07 armin kernel:   fn 0 bar 1: io 0xa00-0xaff
Mar 15 14:22:07 armin kernel:   fn 0 bar 2: mem 0x5042000-0x50421ff
Mar 15 14:22:07 armin kernel: cs: cb_enable(bus 2)
Mar 15 14:22:07 armin kernel:   bridge io map 0 (flags 0x21): 0xa00-0xaff
Mar 15 14:22:07 armin kernel:   bridge mem map 0 (flags 0x1): 0x5042000-0x5042fff
Mar 15 14:22:07 armin kernel: tulip_reap()
Mar 15 14:22:07 armin kernel: tulip_attach(bus 2, function 0)
Mar 15 14:22:07 armin kernel: tulip.c:v0.91g-ppc 7/16/99 becker@cesdis.gsfc.nasa.gov (modified by danilo@cs.uni-magdeburg.de for XIRCOM CBE, fixed by Doug Ledford)
Mar 15 14:22:07 armin kernel: eth0: Digital DS21143 Tulip rev 16 at 0xa00, EEPROM not present, 00:4C:69:6E:75:79, IRQ 9.
Mar 15 14:22:07 armin kernel: eth0:  Missing EEPROM, this interface may not work correctly!
Mar 15 14:22:07 armin cardmgr[785]: executing: './network start eth0'
Mar 15 14:22:10 armin kernel: eth0: Tx hung, 11 vs. 0.
Mar 15 14:22:10 armin kernel: eth0: 21140 transmit timed out, status 0801fbfc, SIA 0000000f 00000000 00000018 58fab388, resetting...
Mar 15 14:22:16 armin kernel: eth0: Tx hung, 11 vs. 0.
Mar 15 14:22:16 armin kernel: eth0: 21140 transmit timed out, status 0801fbfc, SIA 0000000f 00000000 00000018 58fab388, resetting...
Mar 15 14:22:22 armin kernel: eth0: Tx hung, 11 vs. 0.
Mar 15 14:22:22 armin kernel: eth0: 21140 transmit timed out, status 0801fbfc, SIA 0000000f 00000000 00000018 58fab388, resetting...
Mar 15 14:22:28 armin kernel: eth0: Tx hung, 11 vs. 0.
Mar 15 14:22:28 armin kernel: eth0: 21140 transmit timed out, status 0801fbfc, SIA 0000000f 00000000 00000018 58fab388, resetting...
Mar 15 14:22:34 armin kernel: eth0: Tx hung, 11 vs. 0.
Mar 15 14:22:34 armin kernel: eth0: 21140 transmit timed out, status 0801fbfc, SIA 0000000f 00000000 00000018 58fab388, resetting...
Mar 15 14:22:35 armin cardmgr[785]: executing: './network check eth0'
Mar 15 14:22:36 armin cardmgr[785]: shutting down socket 0
Mar 15 14:22:36 armin cardmgr[785]: executing: './network stop eth0'
Mar 15 14:22:36 armin kernel: tulip_detach(eth0)
Mar 15 14:22:36 armin kernel: cs: cb_disable(bus 2)
Mar 15 14:22:36 armin kernel: cs: cb_release(bus 2)
Mar 15 14:22:36 armin cardmgr[785]: executing: 'rmmod tulip_cb'
Mar 15 14:22:36 armin kernel: tulip_reap()
Mar 15 14:22:36 armin cardmgr[785]: executing: 'rmmod cb_enabler'
Mar 15 14:22:36 armin cardmgr[785]: exiting

This is for epic

Mar 15 14:24:08 armin cardmgr[835]: starting, version is 3.1.13
Mar 15 14:24:08 armin cardmgr[835]: watching 2 sockets
Mar 15 14:24:08 armin kernel: cs: IO port probe 0x1000-0x17ff: clean.
Mar 15 14:24:08 armin kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x200-0x207 0x220-0x22f 0x300-0x307 0x378-0x37f 0x388-0x38f 0x398-0x39f 0x400-0x40f 0x480-0x48f 0x4c0-0x4df
Mar 15 14:24:08 armin kernel: cs: IO port probe 0x0a00-0x0aff: clean.
Mar 15 14:24:08 armin cardmgr[835]: initializing socket 0
Mar 15 14:24:08 armin cardmgr[835]: socket 0: Argosy EN-235 Fast Ethernet
Mar 15 14:24:08 armin cardmgr[835]: executing: 'insmod /lib/modules/2.2.13/pcmcia/cb_enabler.o'
Mar 15 14:24:08 armin cardmgr[835]: executing: 'insmod /lib/modules/2.2.13/pcmcia/epic_cb.o'
Mar 15 14:24:08 armin kernel: epic100.c:v1.06 2/10/98 Donald Becker http://cesdis.gsfc.nasa.gov/linux/drivers/epic100.html
Mar 15 14:24:08 armin kernel: cs: cb_config(bus 2)
Mar 15 14:24:08 armin kernel:   fn 0 bar 1: io 0xa00-0xaff
Mar 15 14:24:08 armin kernel:   fn 0 bar 2: mem 0x5042000-0x50421ff
Mar 15 14:24:08 armin kernel: cs: cb_enable(bus 2)
Mar 15 14:24:08 armin kernel:   bridge io map 0 (flags 0x21): 0xa00-0xaff
Mar 15 14:24:08 armin kernel:   bridge mem map 0 (flags 0x1): 0x5042000-0x5042fff
Mar 15 14:24:08 armin kernel: epic_attach(bus 2, function 0)
Mar 15 14:24:08 armin kernel: eth0: SMSC EPIC/C 83c175 at 0xa00, IRQ 9, 00:00:40:00:aa:4a.
Mar 15 14:24:08 armin kernel: eth0: ***WARNING***: No MII transceiver found!
Mar 15 14:24:08 armin cardmgr[835]: executing: './network start eth0'
Mar 15 14:24:16 armin cardmgr[835]: executing: './network check eth0'
Mar 15 14:24:17 armin cardmgr[835]: shutting down socket 0
Mar 15 14:24:17 armin cardmgr[835]: executing: './network stop eth0'
Mar 15 14:24:17 armin kernel: epic_detach(eth0)
Mar 15 14:24:17 armin kernel: cs: cb_disable(bus 2)
Mar 15 14:24:17 armin kernel: cs: cb_release(bus 2)
Mar 15 14:24:17 armin cardmgr[835]: executing: 'rmmod epic_cb'
Mar 15 14:24:17 armin cardmgr[835]: executing: 'rmmod cb_enabler'
Mar 15 14:24:17 armin cardmgr[835]: exiting

Any clues?

Thx

Robert

1. None That doesn't look so good to me by David Hinds, 2000, Mar 15
(_ More No wrong HW address by Robert Rossmann, 2000, Mar 16
(_ None So it isn't a tulip at all by David Hinds, 2000, Mar 16

Question Unsupported Tulip card only works with DHCP

Date: 2000, Mar 18
From: Mike McDonald <mcdonald@aeac.org>

I've got an unsupported Intel 21142/3 based 10/100 PC card that works fine using the tulip driver as long as I configure it using DHCP.

When I configure the card manually, it won't work until I reset it by issuing a "ifconfig eth0 down up". The only problem is that when I do this, I lose the default route.

My guess is that pump is somehow resetting the card when it configures it.

I know the card isn't supported, but I'd still like to use it if possible. Is there an easy way to force the card to reset without losing routes?

Thanks,
Mike

Notebook: Gateway Solo 9300 running Redhat 6.1
Card: SHOMITI 100TX_CardBus PS1101
Manfid: 0x013f, 0x4120

Syslog Messages(when not working):
Mar 17 14:35:18 localhost kernel: eth0: 21140 transmit timed out, status f0260000, SIA 40a1d2cd ffff0001 fffbffff 8ff40008, resetting...
Mar 17 14:35:18 localhost kernel: eth0: transmit timed out, switching to 100baseTx media.
...

1. None I know it isn't a real solution... by David Hinds, 2000, Apr 28

Question trouble with netgear FA410TXC

Date: 2000, Mar 23
From: Matthew Austin yogi

I am running redhat 6.1 distro on a sony vaio picturebook (PCG-C1XS) and I cannot seem to get connectivity to my network.

I have assigned myself a fixed IP address to isolate the issue (was unable to get a DHCP address anyway ;-))

The adapter comes up and is recognized but no connectivity. I can ping the machine at the ip address I have assigned it but nothing on my network, including machines on the same subnet. The dongle lights just flash on and off simultaneous about every second.

I downloaded and compiled the pcmcia-3.1.14 package in the hopes that perhaps there was some issue with the redhat package. I added the recommended changes to network.opts to allow me to continue to use linuxconf.

I have forced a change in io port and IRQ to see if that was an issue as recommended by changing config.opts . . . no change.

Since compiling the new package I have been getting 2 new messages.

/var/log/messages

eth0: found link beat
eth0: lost link beat

Any thoughts?

1. Feedback Hardware Version by Jochen Friedrich, 2000, Mar 25
1. Sad tried code but no go, MAC address within problem range by Matthew Austin, 2000, Mar 27
(_ More What version by Jochen Friedrich, 2000, Mar 31
1. Sad still no go even with your binary . . . here's the details by Matthew Austin, 2000, Mar 31
2. None ioperm() vs iopl() issue by David Hinds, 2000, Mar 31
... 1 Message(s)
2. Question 410TXC and OpenBSD by Morgan Neubauer, 2000, Mar 26

Question Using regular kernel tulip driver with pcmcia-cs

Date: 2000, Mar 31
From: RichardA

Hello,

I'm trying to get my SMC CardBus card, which uses a DEC "tulip" chip, to work on my laptop.

I have already tried pcmcia-cs 3.1.13 with the 2.3.99-pre2 kernel and the tulip 0.91g-ppc driver from pcmcia-cs, i.e. the standalone PCMCIA support, for which I disabled PCMCIA in the kernel. Unfortunately, this version of the tulip code does not work with my card.

However, there is also the different version of the tulip code which comes with the kernel and is supposed to support "regular" PCI NICs as well as CardBus cards. I have compiled a 2.3.99-pre2 kernel and pcmcia-cs in attempt to use it for my card.

I enabled "hot-swappable" (yes), "PCMCIA" (module) and "CardBus" (yes) support in the kernel configuration as well as choosing the regular tulip DS2x14x driver in "network devices". pcmcia-cs did not compile any of its drivers, which is fine since I only need the kernel tulip driver.

Now, when I try to use the kernel PCMCIA code, it decides to exclude more memory ranges than the standalone version. This results in my card not being identified. Instead, it is listed as an "anonymous memory" card.

Here are kernel logs for the two cases. First, the way it looks when PCMCIA support in the kernel is disabled and the standalone pcmcia-cs is used:

Mar 24 21:07:37 elessar pcmcia: Starting PCMCIA services:
Mar 24 21:07:37 elessar pcmcia:  modules
Mar 24 21:07:37 elessar kernel: Linux PCMCIA Card Services 3.1.13
Mar 24 21:07:37 elessar kernel:   kernel build: 2.3.99-pre2 #8 Fri Mar 24 18:54:56 CET 2000
Mar 24 21:07:37 elessar kernel:   options:  [pci] [cardbus]
Mar 24 21:07:37 elessar kernel: Intel PCIC probe: PCI: Assigned IRQ 10 to device 00:03.1 [PIIX-NEW]
Mar 24 21:07:37 elessar kernel:
Mar 24 21:07:37 elessar kernel:   O2Micro OZ6836/OZ6860 PCI-to-CardBus at slot 00:03, mem 0x10000000
Mar 24 21:07:37 elessar kernel:     host opts [0]: [ring] [pci+isa] [pci irq 10] [lat 168/176] [bus
1/2]
Mar 24 21:07:37 elessar kernel:     host opts [1]: [ring] [pci+isa] [pci irq 10] [lat 168/176] [bus
3/4]
Mar 24 21:07:37 elessar kernel:     ISA irqs (default) = 3,4,5,7,9,12 PCI status changes
Mar 24 21:07:37 elessar pcmcia:  cardmgr.
Mar 24 21:07:37 elessar cardmgr[280]: starting, version is 3.1.13
Mar 24 21:07:37 elessar rc: Starting pcmcia succeeded
Mar 24 21:07:37 elessar cardmgr[280]: watching 2 sockets
Mar 24 21:07:37 elessar kernel: cs: IO port probe 0x1000-0x17ff: excluding 0x1000-0x104f
Mar 24 21:07:37 elessar kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x100-0x107 0x200-0x207 0
x220-0x22f 0x300-0x307 0x378-0x37f 0x388-0x38f 0x398-0x39f 0x4d0-0x4d7
[snip]
Mar 24 21:08:36 elessar cardmgr[280]: initializing socket 0
Mar 24 21:08:36 elessar cardmgr[280]: socket 0: SMC EZ CardBus 10/100 PC Card
Mar 24 21:08:36 elessar kernel: cs: cb_alloc(bus 1): vendor 0x1011, device 0x0019
Mar 24 21:08:36 elessar kernel: ROM image dump:
Mar 24 21:08:36 elessar kernel:   image 0: 0x000000-0x0001ff, signature PCIR
Mar 24 21:08:36 elessar cardmgr[280]: executing: 'insmod /lib/modules/2.3.99-pre2/pcmcia/cb_enabler.
o'
Mar 24 21:08:36 elessar cardmgr[280]: executing: 'insmod /lib/modules/2.3.99-pre2/pcmcia/tulip_cb.o'
Mar 24 21:08:36 elessar kernel: cs: cb_config(bus 1)
Mar 24 21:08:36 elessar kernel:   fn 0 bar 1: io 0x400-0x47f
Mar 24 21:08:36 elessar kernel:   fn 0 bar 2: mem 0xa00c0000-0xa00c03ff
Mar 24 21:08:36 elessar kernel:   fn 0 rom: mem 0xa0080000-0xa00bffff
Mar 24 21:08:36 elessar kernel: cs: cb_enable(bus 1)
Mar 24 21:08:36 elessar kernel:   bridge io map 0 (flags 0x21): 0x400-0x47f
Mar 24 21:08:36 elessar kernel:   bridge mem map 0 (flags 0x1): 0xa0080000-0xa00c0fff
Mar 24 21:08:36 elessar kernel: tulip_reap()
Mar 24 21:08:36 elessar kernel: tulip_attach(bus 1, function 0)
Mar 24 21:08:36 elessar kernel: tulip.c:v0.91g-ppc 7/16/99 becker@cesdis.gsfc.nasa.gov (modified by
danilo@cs.uni-magdeburg.de for XIRCOM CBE, fixed by Doug Ledford)
Mar 24 21:08:36 elessar kernel: eth0: Digital DS21143 Tulip rev 65 at 0x400, 00:E0:29:55:E1:12, IRQ
10.
Mar 24 21:08:36 elessar kernel: eth0:  EEPROM default media type Autosense.
Mar 24 21:08:36 elessar kernel: eth0:  Index #0 - Media MII (#11) described by a 21142 MII PHY (3) b
lock.
Mar 24 21:08:36 elessar kernel: eth0:  MII transceiver #1 config 3000 status 7809 advertising 00a1.
Mar 24 21:08:36 elessar kernel: eth0:  Advertising 01e1 on PHY 1, previously advertising 00a1.
Mar 24 21:08:36 elessar kernel: eth0:  Advertising 01e1 (to advertise is 01e1).
Mar 24 21:08:36 elessar cardmgr[280]: executing: './network start eth0'
[ping local machine]
Mar 24 21:11:36 elessar kernel: spurious 8259A interrupt: IRQ7.

(various versions of the tulip driver have failed in weird & wonderful ways. The kernel one seems to be most actively developed ATM, which is why I want to try it. BTW, /proc/interrupts shows that IRQ 7 is indeed not used.)

For the second case, with PCMCIA support in the kernel, the following is logged:

Mar 24 22:20:56 elessar pcmcia: Starting PCMCIA services:
Mar 24 22:20:56 elessar pcmcia:  modules
Mar 24 22:20:56 elessar kernel: Linux PCMCIA Card Services 3.1.11
Mar 24 22:20:56 elessar kernel:   kernel build: 2.3.99-pre2 #1 Fri Mar 24 22:02:38 CET 2000
Mar 24 22:20:56 elessar kernel:   options:  [pci] [cardbus] [pm]
Mar 24 22:20:56 elessar kernel: Intel PCIC probe:
Mar 24 22:20:56 elessar kernel:   Intel i82365sl A step ISA-to-PCMCIA at port 0x3e0 ofs 0x00, 2 sock
ets
Mar 24 22:20:56 elessar kernel:     host opts [0]: none
Mar 24 22:20:56 elessar kernel:     host opts [1]: none
Mar 24 22:20:56 elessar kernel:     ISA irqs (scanned) = 3,4,5,7 polling interval = 1000 ms
Mar 24 22:20:56 elessar pcmcia:  cardmgr.
Mar 24 22:20:56 elessar rc: Starting pcmcia succeeded
Mar 24 22:20:56 elessar cardmgr[280]: starting, version is 3.1.11
Mar 24 22:20:57 elessar cardmgr[280]: watching 2 sockets
Mar 24 22:20:57 elessar kernel: cs: IO port probe 0x1000-0x17ff: clean.
Mar 24 22:20:57 elessar kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x100-0x107 0x200-0x207 0
x220-0x22f 0x300-0x307 0x378-0x37f 0x388-0x38f 0x398-0x39f 0x4d0-0x4d7
[snip]
Mar 24 22:21:55 elessar kernel: cs: memory probe 0x0d0000-0x0dffff: excluding 0xdc000-0xdffff
Mar 24 22:21:55 elessar kernel: cs: memory probe 0x0c0000-0x0cffff: excluding 0xc0000-0xcbfff
Mar 24 22:21:55 elessar cardmgr[280]: initializing socket 0
Mar 24 22:21:55 elessar cardmgr[280]: socket 0: Anonymous Memory
Mar 24 22:21:55 elessar cardmgr[280]: module /lib/modules/2.3.99-pre2/pcmcia/memory_cs.o not availab
le
Mar 24 22:21:56 elessar cardmgr[280]: get dev info on socket 0 failed: Resource temporarily unavaila
ble

AFAICT the memory ranges reported in the "memory probe" messages are not used by anything else!

Is the differing behaviour due to a difference between the pcmcia-cs 3.1.11 in the kernel and the 3.1.13 I used for the first case above? What can I do to fix the problem?

My /etc/pcmcia/config.opts for both cases was:

include port 0x100-0x4ff, port 0x1000-0x17ff
include memory 0xc0000-0xfffff, memory 0xa0000000-0xa0ffffff
exclude irq 4
exclude irq 7
card "SMC EZ CardBus 10/100 PC Card"
  manfid 0x01bf, 0x2220
  bind "tulip_cb"

When the above things failed, I tried to make the standalone pcmcia-cs load and use the kernel tulip driver. However, I don't really know how I am supposed to do this. I tried changing my /etc/pcmcia/config.opts file to load the normal tulip driver like this:

include port 0x100-0x4ff, port 0x1000-0x17ff
include memory 0xc0000-0xfffff, memory 0xa0000000-0xa0ffffff
exclude irq 4
exclude irq 7
device "tulip2_cb"
  class "network" module "cb_enabler", "net/tulip"                                                                               opts "tulip_debug=6"
card "SMC EZ CardBus 10/100 PC Card"
  manfid 0x01bf, 0x2220
  bind "tulip2_cb"

Not very surprisingly, this doesn't work; the kernel tulip driver just complains I/O ports (0x80@0x0) unavailable, aborting and refuses to load.

At this point I'm really stuck. I have been trying to get this card to work for four months now - any help would be very much appreciated!!

Thanks for reading this far,
Even more thanks for any help you can give,

  Richard

1. None Configuration on 2.3.* by David Hinds, 2000, Mar 31

Question KingMax 10/100 Ethernet

Date: 2000, Apr 04
From: Ryan Marquis rmarquis

Any chance of getting this into the PCMCIA package? I did read above (November '99) that it probably wouldn't be included, but I was wondering if anyone was going to reconsider...

I notice they've done an update up to kernel v. 2.2.6. However, predictably, their stuff doesn't compile with pcmcia 3.1.8.

Thanks!

1. None Not in the immediate future by David Hinds, 2000, Apr 04

Question Possible need for a FORWARDING="y/n" setting

Date: 2000, Apr 05
From: Dagmar d'Surreal Dagmar_dSurreal

Okay, I've got a set of wireless ethernet cards, and while frankly I would not recommend people doing this excepting in certain situations, I found a need last night for turning on ip_forwarding for the interface created by pcmcia-cs for them. This normally defaults to off (unless you turn on forwarding globally, in which case you should go stand in the corner now) but some people might have a genuine need to do this.

I know this probably seems like a minor thing, but I want to know if anyone else has any opinions on the subject of whether or not they think an IP_FORWARDING="y/n" option would be apropos for the /etc/pcmcia/network.opts file.

(For the record, I would not normally use forwarding blatantly across the interface. I'm working with Free S/WAN to get an encrypted connection authenticated by secret key which DOES have forwarding turned on)

1. None Another way to do it: use start_fn() by David Hinds, 2000, Apr 05
(_ Note Untitled by Dagmar d'Surreal, 2000, Apr 06

Question IBM ethernet card don't work

Date: 2000, Apr 07
From: borg1 borg1

 have a Acer travel mate 514t and a IBM credit card adapter ethernet adapter 2 pcmcia network card.
I installed the installed the pcmcia-cs 3.1.13 and i am using linux mandrake 7.0 with kernel 2.2.14.When i insert the card it finds it and beeps and the cardmgr tells me the name of the card so it must be able to see it.But the network card will not work i have configured the network and the laptop becomes slow when i insert the card.

 The error given in dmesg is

eth0: Tx timed out, cable problem? TSR=0x3, ISR=0x0, t=2000. eth0: pcnet_reset_8390() did not complete.
Hw. address read/write mismap 2
Hw. address read/write mismap 4
[root@laptop /]#

the card has two network fittings one a RJ 45 which i use and one bnc 10 base T
I hardly use the bnc one but do on the odd occasion some times.

And the led lights up when i insert the card and the access led blinks some times.

card ident

[root@laptop /]# cardctl ident
Socket 0:
  no product info available
Socket 1:
  product info: "IBM Corp.", "Ethernet", "0934214", "PC-NIC ENCC 990010759-001A" manfid: 0x00a4, 0x0002
  function: 6 (network)
[root@laptop /]#

dmesg

eth0: pcnet_reset_8390() did not complete.
eth0: NE2000 Compatible: io 0x300, irq 3, mem 0xc805c000, auto xcvr, hw_addr 00:04:AC:24:1E:24
Hw. address read/write mismap 2
Hw. address read/write mismap 4
[root@laptop /]#


the loptop runs slowley when the card is inserted and some times the hard disk will go insane and fill the screen up with the
Hw. address read/write mismap 2
Hw. address read/write mismap 4
and then stops when you take the card out but this onley happens then you have the card in slot 1
1. None Maybe a resource conflict by David Hinds, 2000, Apr 07

Question SMC EZ CardBus 10/100 PC Card

Date: 2000, Apr 09
From: Andreas-Johann Ulvestad aj

I am currently using a pcmcia ethernet adapter, SMC EZ CardBus 10/100 PC Card is reported by cardmgr when booting or inserting the card. This is all well, but inserting "alias eth0 pcnet_cs" in /etc/conf.modules, I get "Delaying eth0 initialization" when ifup'ing eth0.

Is this the correct module to be using?
1. None Wrong driver; also, don't list PCMCIA drivers in conf.modules by David Hinds, 2000, Apr 10
(_ More hmm....ok? by Andreas-Johann Ulvestad, 2000, Apr 10
(_ None Check your system log by David Hinds, 2000, Apr 10

Angry Intel EtherExpress PRO/100 Mobile, model MBLA3200

Date: 2000, Apr 09
From: Jim Perry aziwoqpd

Heya... I was trying to get Linux installed on my laptop
today, and I couldn't get the bloody PCMCIA ethernet driver
to work.  I tried shoving it onto different IRQs and different
ports, but I kept getting the same error.  Anyway, I'm using Linux-2.2.13, the latest pcmcia (pcmcia-cs-3.1.13 i think?).
Here's the logs:

Apr  9 15:57:35 slackware kernel: Linux PCMCIA Card Services 3.1.13 
Apr  9 15:57:35 slackware kernel:   kernel build: 2.2.13 #1 SMP Sun Apr 9 14:10:46 PDT 2000 
Apr  9 15:57:35 slackware kernel:   options:  [pci] [cardbus] [pnp] 
Apr  9 15:57:35 slackware kernel: PCI routing table version 1.0 at 0xf6980 
Apr  9 15:57:35 slackware kernel: PnP: PNP BIOS installation structure at 0xc00fdab0 
Apr  9 15:57:35 slackware kernel: PnP: PNP BIOS version 1.0, entry at f0000:4b88, dseg at f0000 
Apr  9 15:57:46 slackware kernel: Intel PCIC probe:  
Apr  9 15:57:46 slackware kernel:   Ricoh RL5C465 PCI-to-CardBus at slot 00:13, mem 0xfedff000 
Apr  9 15:57:46 slackware kernel:     host opts [0]: [io 4/6/2] [mem 4/6/2] [pci irq 9] [lat 64/176] [bus 1/1] 
Apr  9 15:57:46 slackware kernel:     PCI irq 9 test failed 
Apr  9 15:57:46 slackware kernel:     ISA irqs (default) = 10,11,15 polling interval = 1000 ms 
Apr  9 15:57:56 slackware cardmgr[255]: starting, version is 3.1.13
Apr  9 15:57:56 slackware kernel: cs: IO port probe 0x1000-0x17ff: clean. 
Apr  9 15:57:56 slackware cardmgr[255]: watching 1 sockets
Apr  9 15:58:05 slackware kernel: cs: cb_alloc(bus 1): vendor 0x1011, device 0x0019 
Apr  9 15:58:05 slackware kernel: ROM image dump: 
Apr  9 15:58:05 slackware kernel:   image 0: 0x000000-0x0001ff, signature PCIR 
Apr  9 15:58:05 slackware kernel: cs: cb_config(bus 1) 
Apr  9 15:58:05 slackware kernel:   fn 0 bar 1: io 0x1000-0x107f 
Apr  9 15:58:05 slackware kernel:   fn 0 bar 2: mem 0x600c0000-0x600c007f 
Apr  9 15:58:05 slackware kernel:   fn 0 rom: mem 0x60080000-0x600bffff 
Apr  9 15:58:05 slackware kernel: cs: cb_enable(bus 1) 
Apr  9 15:58:05 slackware kernel:   bridge io map 0 (flags 0x21): 0x1000-0x107f 
Apr  9 15:58:05 slackware kernel:   bridge mem map 0 (flags 0x1): 0x60080000-0x600c0fff 
Apr  9 15:58:05 slackware kernel: tulip_reap() 
Apr  9 15:58:05 slackware kernel: tulip_attach(bus 1, function 0) 
Apr  9 15:58:05 slackware kernel: tulip.c:v0.91g-ppc 7/16/99 becker@cesdis.gsfc.nasa.gov (modified by danilo@cs.uni-magdeburg.de for XIRCOM CBE, fixed by Doug Ledford) 
Apr  9 15:58:05 slackware kernel: eth0: Digital DS21143 Tulip rev 48 at 0x1000, EEPROM not present, 00:4C:69:6E:75:79, IRQ 11. 
Apr  9 15:58:05 slackware cardmgr[255]: initializing socket 0
Apr  9 15:58:05 slackware cardmgr[255]: socket 0: Intel EtherExpress PRO/100 CardBus Mobile Adapter32
Apr  9 15:58:05 slackware cardmgr[255]: executing: 'insmod /lib/modules/2.2.13/pcmcia/cb_enabler.o'
Apr  9 15:58:05 slackware cardmgr[255]: executing: 'insmod /lib/modules/2.2.13/pcmcia/tulip_cb.o'
Apr  9 15:58:05 slackware cardmgr[255]: executing: './network start eth0'
Apr  9 15:58:50 slackware kernel: eth0: Tx hung, 9 vs. 0. 
Apr  9 15:58:50 slackware kernel: eth0: 21140 transmit timed out, status f0678006, SIA 000000c6 ffff0001 fffbff7f 8ff04008, resetting... 
Apr  9 15:58:56 slackware kernel: eth0: Tx hung, 9 vs. 0. 
Apr  9 15:58:56 slackware kernel: eth0: 21140 transmit timed out, status f0678106, SIA 000000c6 ffff0000 fffbff7f 8ff04008, resetting... 
Apr  9 15:59:02 slackware kernel: eth0: Tx hung, 9 vs. 0. 
Apr  9 15:59:02 slackware kernel: eth0: 21140 transmit timed out, status f0678106, SIA 000000c6 ffff0000 fffbff7f 8ff04008, resetting... 
Apr  9 15:59:08 slackware kernel: eth0: Tx hung, 9 vs. 0. 
Apr  9 15:59:08 slackware kernel: eth0: 21140 transmit timed out, status f0678106, SIA 000000c6 ffff0000 fffbff7f 8ff04008, resetting... 
Apr  9 15:59:14 slackware kernel: eth0: Tx hung, 9 vs. 0. 
Apr  9 15:59:14 slackware kernel: eth0: 21140 transmit timed out, status f0678106, SIA 000000c6 ffff0000 fffbff7f 8ff04008, resetting... 
(i get mad and take the card out)
Apr  9 15:59:15 slackware kernel: tulip_detach(eth0) 
Apr  9 15:59:15 slackware kernel: cs: cb_disable(bus 1) 
Apr  9 15:59:15 slackware kernel: cs: cb_release(bus 1) 
Apr  9 15:59:15 slackware kernel: cs: cb_free(bus 1) 
Apr  9 15:59:15 slackware kernel: tulip_reap() 
Apr  9 15:59:15 slackware cardmgr[255]: shutting down socket 0

Also, I tried running tulip-diag while it was sitting there 
printing out those "hung" statements.  This is what I got:
tulip-diag.c:v1.19 10/2/99 Donald Becker (becker@cesdis.gsfc.nasa.gov)
Index #1: Found a Digital DS21143 Tulip adapter at 0x1000.
 Port selection is 10mpbs-serial, full-duplex.
 Transmit stopped, Receive stopped, full-duplex.
  The Rx process state is 'Stopped'.
  The Tx process state is 'Stopped'.
  The transmit threshold is 72.
 Interrupt sources are pending!  CSR5 is f0018106.
   Tx complete indication.
   Tx out of buffers indication.
   Receiver stopped indication.
  The NWay status register is 000000c6.
EEPROM size is 6.
 A simplifed EEPROM data table was found.
 The EEPROM does not contain transceiver control information.
EEPROM contents:
  8086 0001 0087 0000 0000 0000 0000 0000
  00c4 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 4448
 ID block CRC 0xc4 (vs. 0xc4).
  Full contents CRC 0x4448 (read as 0x4448).
You have an Intel CardBus card with an incomplete EEPROM.
The station address is 00:a0:c9:7a:d8:a5
New EEPROM contents would be:
  8086 0001 0087 0000 0000 0000 0000 0000
  00c4 0103 a000 7ac9 a5d8 1e00 0000 0800
  8604 0002 08af 00a5 0286 af04 a508 8800
  0304 08af 00a5 8061 0488 af05 a508 6100
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 0000
  0000 0000 0000 0000 0000 0000 0000 4448
 ID CRC 0xc4 (vs. 0xc4), complete CRC 4448.
  Internal autonegotiation state is 'Autonegotiation disabled'.

Any ideas?  I heard something about reflashing the EEPROM,
but I didn't do that for two reasons.. Mainly, it sounds
dangerous, and the more important, I don't know how.
Oh, and incidentially, I'm on a 100baseT network, not a
10baseT network.  Tried doing an "ifport eth0 100baseT", but
I just got an ioctl not allowed error.  "ifport eth0" by itself claims
that the adapter is on "auto" mode.

Thanks,
-jim
1. Ok EtherExpress fixed! by Jim Perry, 2000, Apr 15
(_ Agree That fixes my card too. But the EEPROM rewrite seems essential. by Kurt Atkinson, 2000, May 24

Question DHCP and manually set MAC address

Date: 2000, Apr 12
From: Andreas Hotz ahotz

To use my laptop on the company network, I have to use DHCP. No problem so far. But I have to use DHCP with a manually configured MAC address. Don't ask me why this is so, it just is that way. The DHCP server won't answer me with thy MAC address burned into the adapter.

I could not find a suitable variable in the PCMCIA config scripts. I think I have to tweak a little with ifconfig. Maybe in the start_fn function in network.opts? Or should I change config instead. Any tips which is the best way to do it?

1. None There has got to be another way by David Hinds, 2000, Apr 12
1. More I agree, however... by Andreas Hotz, 2000, Apr 13
(_ None Ok, maybe I don't know what I'm talking about by David Hinds, 2000, Apr 13
2. Ok It works! by Andreas Hotz, 2000, Apr 17

Sad Toshiba+TDK Ethernet+SuSe -> SIOCSIFADRR unknown device

Date: 2000, Apr 12
From: Alexei Gorbatov alexx

I am trying to connect TDK DFL 3410/12 Ethernet/Modem card on the Toshiba Tecra 8000 using SuSe Linux 6.3 The system recognizes ethernet card and load driver succesfully. However, at the moment of mounting ethernet it gives messages: SIOCSIFADRR unknown device.............fault eth0 unknown device....................fault etc. I exclude irqs 3,5, and 9 from ./pcmcia/config.opts Unfortunately it does not resolve the problem. Please, help!!!!

1. None System log messages, please? by David Hinds, 2000, Apr 12
1. Sad log message by Alexei Gorbatov, 2000, Apr 13
2. None Unsupported card by David Hinds, 2000, Apr 13
1. More Xircom by Alexei Gorbatov, 2000, Apr 14

Question kernel: eth0: transmit timed out, tx_status 00 status e000.

Date: 2000, Apr 11
From: Bernard BARBIER BernardJB

hello,

Could some body tell me what's wrong ? My network interface works well 
during 3 or 4 minutes and then stops, and restart...

in order : 
- error messages : 

kernel: eth0: transmit timed out, tx_status 00 status e000.
kernel: eth0: Tx Ring full, refusing to send buffer. 
cardmgr[334]: + ioctl: Operation not supported

- pcmcia-cs-3.1.13
- kernel : linux-2.2.5  (redhat)
- laptop : Dell inspiron 7500
- files :
/etc/sysconfig/pcmcia
/proc/interrupt
/var/log/messages

Any help would be grately appreciated ! 

From France, BERNARD!

---


/etc/sysconfig/pcmcia
PCMCIA=yes
PCIC=i82365
PCIC_OPTS="do_scan=0 irq_list=3,5"
CORE_OPTS=

 /proc/interrupts 
           CPU0       
  0:     328922          XT-PIC  timer
  1:       2815          XT-PIC  keyboard
  2:          0          XT-PIC  cascade
  3:     606111          XT-PIC  eth0
  8:          2          XT-PIC  rtc
 12:     111621          XT-PIC  PS/2 Mouse
 13:          1          XT-PIC  fpu
 14:     619092          XT-PIC  ide0
 15:          5          XT-PIC  ide1
NMI:          0


here is my last boot /var/log/messages

pcmcia: Starting PCMCIA services:
pcmcia:  modules
kernel: Linux PCMCIA Card Services 3.1.13 
kernel:   kernel build: 2.2.5-15 #1 Mon Apr 19 23:00:46 EDT 1999 
kernel:   options:  [pci] [cardbus] [apm] 
kernel: PCI routing table version 1.0 at 0xfdf50 
kernel:   00:04.0 -> irq 11 
kernel:   00:04.1 -> irq 11 
kernel: Intel PCIC probe:  
TI 1225 PCI-to-CardBus at slot 00:04, mem 0x68000000 
kernel:     host opts [0]: [ring] [serial pci & irq] [pci irq 11] [lat 168/32] [bus 32/34] 
kernel:     host opts [1]: [ring] [serial pci & irq] [pci irq 11] [lat 168/32] [bus 35/37] 
kernel:     PCI irq 11 seems to be wedged! 
kernel:     PCI irq 11 test failed 
kernel:     ISA irqs (default) = 3 polling interval = 1000 ms 
pcmcia:  cardmgr.
rc: Starting pcmcia succeeded
cardmgr[334]: starting, version is 3.1.13
cardmgr[334]: watching 2 sockets
kernel: cs: IO port probe 0x1000-0x17ff: excluding 0x1000-0x1087 0x1400-0x14ff 
kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x378-0x37f 0x4d0-0x4d7 
kernel: cs: IO port probe 0x0a00-0x0aff: clean. 
inet: inetd startup succeeded
kernel: cs: cb_alloc(bus 32): vendor 0x10b7, device 0x5257 
cardmgr[334]: initializing socket 0
cardmgr[334]: socket 0: 3Com 3CCFE575CT/3CXFE575CT Fast EtherLink XL
cardmgr[334]: executing: 'insmod /lib/modules/2.2.5-15/pcmcia/cb_enabler.o'
cardmgr[334]: executing: 'insmod /lib/modules/2.2.5-15/pcmcia/3c575_cb.o'
kernel: 3c59x.c:v0.99L 5/28/99 Donald Becker http://cesdis.gsfc.nasa.gov/linux/drivers/vortex.html 
kernel: cs: cb_config(bus 32) 
kernel:   fn 0 bar 1: io 0x200-0x27f 
kernel:   fn 0 bar 2: mem 0xa0021000-0xa002107f 
kernel:   fn 0 bar 3: mem 0xa0020000-0xa002007f 
kernel:   fn 0 rom: mem 0xa0000000-0xa001ffff 
kernel: vortex_attach(bus 32, function 0, device 5257) 
kernel: eth0: 3Com 3CCFE575CT Cyclone CardBus at 0x200,  00:50:da:55:27:2d, IRQ 3 
kernel: eth0: CardBus functions mapped a0020000->c8849000 
kernel:   8K byte-wide RAM 5:3 Rx:Tx split, MII interface. 
kernel:   MII transceiver found at address 0, status 7809. 
kernel:   Enabling bus-master transmits and whole-frame receives. 
cardmgr[334]: executing: './network start eth0'
lpd: lpd startup succeeded
keytable: Loading keymap: 
cardmgr[334]: + ioctl: Operation not supported

Sad eth0:interrupt(s) dropped!

Date: 2000, Mar 31
From: Ray jxray

/etc/sysconfig/pcmcia PCMCIA=yes PCIC=i82365 PCIC_OPTS="do_scan=0" #this dose not work for me. CORE_OPTS=

#cardctl ident
Socket 0:
  product info: "PCMCIA", "100BASE"
  manfid: 0xffff, 0x1090
  function: 6 (network)
Socket 1:
  no product info available

no other devices was using the irq my card was using. i tried all irqs which can be used, but get the same massage.

i think pcmcia-cs-3.1.13 can not support my poor card...

this card works perfectly in windoz using 0x320 irq 10

can any body help? thanX.

1. None So give us a hint: what brand and model of card is this? by David Hinds, 2000, Mar 31
(_ Feedback laplop manufactured in China. & my config files by Ray, 2000, Apr 01
(_ None I'm afraid you are out of luck for now by David Hinds, 2000, Apr 03
(_ Ok I'll be waiting for your new drivers...:) by Ray, 2000, Apr 04

Question Problem with IBM 10/100 Etherjet Cardbus Adapter

Date: 2000, Apr 18
From: Torsten Tittmann tti

I use a Thinkpad 385XD with an IBM 10/100 Etherjet Cardbus Adapter and
an IBM International Modem with GSM PC Card.

The installation is an SuSE 6.4 (Kernel 2.2.14 with APM Support)
with cardctl 3.1.11.

It looks all fine during bootup and card detecting, but any access to the LAN fail.
I connected directly to our switch and the switch receive frames.
The LEDs of the cableadapter blinks too.
But there is no response to ping, traceroute, ...

After adding "debug=255" to the tulip_cs module I get the following messages.

/var/log/messages:
Apr 18 13:43:12 laptop-tti kernel: Linux PCMCIA Card Services 3.1.11
Apr 18 13:43:12 laptop-tti kernel:   kernel build: 2.2.14 #1 Sat Mar 25 00:46:59 
GMT 2000
Apr 18 13:43:12 laptop-tti kernel:   options:  [pci] [cardbus] [apm]
Apr 18 13:43:12 laptop-tti kernel: PCI routing table version 1.0 at 0xfb150
Apr 18 13:43:12 laptop-tti kernel:   00:02.0 -> irq 11
Apr 18 13:43:12 laptop-tti kernel:   00:02.1 -> irq 11
Apr 18 13:43:12 laptop-tti kernel: Intel PCIC probe: 
Apr 18 13:43:12 laptop-tti kernel:   TI 1250A PCI-to-CardBus at slot 00:02, mem 
0x50000000
Apr 18 13:43:12 laptop-tti kernel:     host opts [0]: [ring] [isa irq] [pci irq 
11] [lat 168/176] [bus 1/3]
Apr 18 13:43:12 laptop-tti kernel:     host opts [1]: [ring] [isa irq] [pci irq 
11] [lat 168/176] [bus 4/6]
Apr 18 13:43:12 laptop-tti kernel:     ISA irqs (scanned) = 3,4,7,9,10,15 PCI 
status changes
Apr 18 13:43:12 laptop-tti kernel: cs: IO port probe 0x0c00-0x0cff: excluding 
0xcf8-0xcff
Apr 18 13:43:12 laptop-tti kernel: cs: IO port probe 0x0800-0x08ff: clean.
Apr 18 13:43:12 laptop-tti kernel: cs: IO port probe 0x0100-0x04ff: excluding 
0x200-0x207 0x220-0x22f 0x388-0x38f 0x3b8-0x3df 0x4d0-0x4d7
Apr 18 13:43:12 laptop-tti kernel: cs: cb_alloc(bus 4): vendor 0x115d, device 
0x0003
Apr 18 13:43:12 laptop-tti kernel: ROM image dump:
Apr 18 13:43:12 laptop-tti kernel:   image 0: 0x000000-0xffffffff, signature 
PCIR
Apr 18 13:43:12 laptop-tti kernel: cs: cb_config(bus 4)
Apr 18 13:43:12 laptop-tti kernel:   fn 0 bar 1: io 0x100-0x17f
Apr 18 13:43:12 laptop-tti kernel:   fn 0 bar 2: mem 0x6000d000-0x6000d7ff
Apr 18 13:43:12 laptop-tti kernel:   fn 0 bar 3: mem 0x6000c000-0x6000c7ff
Apr 18 13:43:12 laptop-tti kernel:   fn 0 rom: mem 0x60008000-0x6000bfff
Apr 18 13:43:12 laptop-tti kernel: cs: cb_enable(bus 4)
Apr 18 13:43:12 laptop-tti kernel:   bridge io map 0 (flags 0x21): 0x100-0x17f
Apr 18 13:43:12 laptop-tti kernel:   bridge mem map 0 (flags 0x1): 0x60008000-
0x6000dfff
Apr 18 13:43:12 laptop-tti kernel: tulip_reap()
Apr 18 13:43:12 laptop-tti kernel: tulip_attach(bus 4, function 0)
Apr 18 13:43:12 laptop-tti kernel: tulip.c:v0.91g-ppc 7/16/99 
becker@cesdis.gsfc.nasa.gov (modified by danilo@cs.uni-magdeburg.de for XIRCOM 
CBE, fixed by Doug Ledford)
Apr 18 13:43:12 laptop-tti kernel: eth0: Xircom Cardbus Adapter (DEC 21143 
compatible mode) rev 3 at 0x100, 00:04:AC:90:DD:73, IRQ 11.
Apr 18 13:43:12 laptop-tti kernel: eth0:  MII transceiver #0 config 3100 status 
7809 advertising 01e1.
Apr 18 13:43:12 laptop-tti kernel: eth0: tulip_open() irq 11.
Apr 18 13:43:12 laptop-tti kernel: eth0: interrupt  csr5=0x60678006 new 
csr5=0x60660000.
Apr 18 13:43:12 laptop-tti kernel: eth0: The transmitter stopped.  CSR5 is 
60678006, CSR6 324c2002, new CSR6 32400000.
Apr 18 13:43:12 laptop-tti kernel: eth0: interrupt  csr5=0x60670004 new 
csr5=0x60660000.
Apr 18 13:43:12 laptop-tti kernel: eth0: interrupt  csr5=0x60660000 new 
csr5=0x60660000.
Apr 18 13:43:12 laptop-tti kernel: eth0: exiting interrupt, csr5=0x60660000.
Apr 18 13:43:12 laptop-tti kernel: eth0: Done tulip_open(), CSR0 fea08000, CSR5 
60660000 CSR6 324c2002.
Apr 18 13:43:12 laptop-tti kernel: eth0: interrupt  csr5=0x60670004 new 
csr5=0x60660000.
...

/var/log/warn:
Apr 18 13:43:12 laptop-tti kernel: eth0: The transmitter stopped.  CSR5 is 
60678006, CSR6 324c2002, new CSR6 32400000.



can somebody help me?

Thanks a lot
                Torsten

Question Problem with IBM token ring driver and 3c389

Date: 2000, Apr 18
From: Rogan Dawes rdawes

Hi

I am trying to get a 3Com 3c389 Tokenlink Velocity Card working with Linux 2.2.14. I have downloaded the 3.1.8 PCMCIA software, but have detected a problem:

My NT installation detects memory used at 0xd4000 and 0xd0000, and shows this as allocated to the token ring driver, but Linux seems to be trying to use other fairly arbitrary addresses (e.g. 0xc....)

I tried hard coding the memory ranges into the driver, to verify where the problem was, and the driver installed and worked like a champ.

I also tried passing the parameters to the module using /etc/modules.conf, but this did not seem to get the values into the driver ( perhaps the cardmgr was passing its own parameters?)

The laptop is a Dell Latitude CPiA300.

What follows is an abridged NT diags dump, as well as the dump_cis output.

Does anyone have any suggestions for me?

Drivers Report
----------------------------------------------------------------------
IBM Token Ring Adapter (ISA/PCMCIA) Driver (NDIS) Running   (Automatic)
   C:\WINNT\System32\drivers\ibmtoks.sys
   Error Severity: Normal
   Service Flags: Kernel Driver, Shared Process
IRQ and Port Report
----------------------------------------------------------------------
Devices                       Vector Level  Affinity
----------------------------------------------------------------------
IbmTok                            10    10 0x00000000
----------------------------------------------------------------------
Devices                       Physical Address  Length
----------------------------------------------------------------------
IbmTok                          0x00000a20  0x0000000004

DMA and Memory Report Devices Physical Address Length ---------------------------------------------------------------------- IbmTok 0x000d4000 0x00002000 IbmTok 0x000d0000 0x00004000 Pcmcia 0x08000000 0x00002000

Here is the dump_cis:
Socket 0:
  no CIS present

Socket 1:
  dev_info
    ROM 250ns, 8kb
    fn_specific 250ns, 64kb
  vers_1 4.1, "3Com", "TokenLink Velocity PC Card", "85H3636",
    "B3ENIPSKJRT"
  config base 0x0800 mask 0x000b last_index 0x23
  cftable_entry 0x20 [default]
    [mwait]
    Vcc Vnom 5V Iavg 300mA
    timing wait 1us
    io 0x0000-0x0003 [lines=2] [8bit]
    irq mask 0xffff [level]
    memory 0x0000-0x1fff @ 0x0000
  cftable_entry 0x21
    memory 0x400000-0x6010ff @ 0x0000
  cftable_entry 0x22
    memory 0x800000-0xa010ff @ 0x0000
  cftable_entry 0x23
    memory 0x0000-0x2010ff @ 0x0000
  manfid 0x00a4, 0x001e
  funcid network_adapter [post] [rom]
  lan_technology token_ring
  lan_speed 16 mb/sec
  lan_speed 4 mb/sec
  lan_media unshielded_twisted_pair
  lan_media shielded_twisted_pair
  lan_node_id 95 67 fb 24 a0 00
  checksum 0x0000-0x00af = 0x75

Another thing that I find funny, is that I cannot get useful information out of tcpdump. I have got a version of tcpdump 3.3 modified to support Token Ring, but all it ever shows me is broadcast traffic. It never even shows me my own traffic! Weird! Suggestions?

Thanks a lot!

Rogan

1. None See /etc/pcmcia/config.opts by David Hinds, 2000, Apr 18

None Problem with Fiberline PCM-FE1500 Fast Ethernet card

Date: 2000, Apr 22
From: digithi

Hi, I've bought a Fiberline PCM-FE1500 16Bit PCMCIA Fast Ethernet Card; I have pcmcia_cs 3.1.14 and kernel 2.2.14, The module pcnet_cs loads correctly: "eth0: NE2000 Compatible: io 0x300, irq 4, hw_addr 00:E0:...."

my Problem: The performance of this card under linux is really low: I reach 300kb/s in a switched 100MBit network und only 60kb/s in a 10MBit network ! The cable is ok and ifconfig doesn't show any problems or errors. With windows, I reach in a 10Mbit and 100MBit network 700kb/s.

can anyone help me with my problem ?

1. None No error messages? Then I'm not sure. by David Hinds, 2000, Apr 24
(_ Ok possible Half/Fullduplex problem by thimo.e@digithi.han.de, 2000, Apr 25

Question Anyone getting the Intel PRO/100 LAN+Modem56 Cardbus II to work?

Date: 2000, Apr 22
From: John Kilpatrick <john@hypergeek.net>

I've got the patch that can be found at http://www.tux.org/hypermail/linux-eepro100/1999-Jun/0010.html (parts of which I had to apply by hand since I couldn't get the patch to work). This patch is applied against eepro100.c from 2.3.5.

NOTE: EEPROM_SAVE is a typo, I think...it should be EEPROM_SIZE.

I've been trying all night, and from 2.3.99-pre5 to 2.3.25 I could not get it to compile (i'll try 2.3.5 tomorrow). I can't get it to compile to save my soul...I get all sorts of errors against spinlock, or it compiles NOTHING in clients. (And typing make clients says there is nothing to do.) Which means I'm configuring something wrong.

So if anyone has somehow gotten this card to work, could you let me know? I'd like to try to get it working. :-)

Thanks, John

Sad Bizarre Linksys NP100 problem

Date: 2000, Apr 23
From: Larry Carter lcarter

I get normal cardmgr startup (i.e. good double beep) and all my logs/config files look good, yet the card will not talk to the lan (dongle lights do not even light up). EXCEPT that sometimes I can get the card to work if I remove it and re-insert it. The card will work fine, network lights glow, I can ping the other boxes on my net, but eventually (I have not been able to get a consistent time frame), the card just seems to shut off (lights go off). cardctl says all is fine, ifconfig looks good, I can ping my local IP but not anything else on the network! This card works great with Win95 as the only other difference. I have tried it in both slots and at different IRQ's (even though I know 3 is free) to no avail.

Here are the only things I can think may be the problem.

1. My i82365 chipset is one that is not listed anywhere (Omega Micro) and may be the problem (or part of it).
2. The card is going into some kind of low power state that Linux can't get it out of. However, I have had the card "turn off" in the middle of pinging other boxes on the network so I don't think this could the problem.
3. The NP100 is "dumbing down" to 10Mbs. Though I doubt this would be an issue (I do have a dual speed hub).

It is driving me nuts. I really, really need to have networking on this system and I cannot seem to figure this out. Does anyone have any pointers?

Here are the details:
Laptop: CTX EzBook 750cs, P150, 32MB
OS: Caldera OpenLinux eDesktop 2.4 (SuSE 6.3 seems to have same problem)
NIC: Linksys (Network Everywhere) NP100 PCMCIA
PCMCIA: tried both 3.1.8 and 3.1.14

Here is /var/log/messages output for cardmgr:
Apr 22 18:38:14 willow kernel: Linux PCMCIA Card Services 3.1.14
Apr 22 18:38:14 willow kernel: kernel build: 2.2.14 #1 SMP Thu Feb 17 11:49:42 MST 2000
Apr 22 18:38:14 willow kernel: options: [pci]
Apr 22 18:38:14 willow kernel: Intel PCIC probe:
Apr 22 18:38:14 willow kernel: Omega Micro 82c092G rev 00 PCI-to-PCMCIA at slot 00:03, port 0x3e0
Apr 22 18:38:14 willow kernel: host opts [0]: none
Apr 22 18:38:14 willow kernel: host opts [1]: none
Apr 22 18:38:14 willow kernel: ISA irqs (scanned) = 3,7,9,11,15 status change on irq 15
Apr 22 18:38:14 willow cardmgr[2007]: starting, version is 3.1.14
Apr 22 18:38:14 willow cardmgr[2007]: watching 2 sockets
Apr 22 18:38:14 willow kernel: cs: IO port probe 0x0c00-0x0cff: excluding 0xcf8-0xcff
Apr 22 18:38:14 willow kernel: cs: IO port probe 0x0800-0x08ff: clean.
Apr 22 18:38:14 willow kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x01e0-0x1e7 0x208-0x20f 0x378-0x37f 0x388-0x38f 0x4d0-0x4d7
Apr 22 18:38:14 willow kernel: cs: IO port probe 0x0a00-0x0aff: clean.
Apr 22 18:38:14 willow cardmgr[2007]: initializing socket 1
Apr 22 18:38:14 willow cardmgr[2007]: socket 1: NE2000 Compatible Ethernet
Apr 22 18:38:14 willow kernel: cs: memory probe 0x0d0000-0x0dffff: clean.
Apr 22 18:38:14 willow cardmgr[2007]: executing: 'insmod /lib/modules/2.2.14/net/8390.o'
Apr 22 18:38:14 willow cardmgr[2007]: executing: 'insmod /lib/modules/2.2.14/pcmcia/pcnet_cs.o'
Apr 22 18:38:14 willow kernel: eth0: NE2000 Compatible: io 0x300, irq 3, hw_addr 00:E0:98:77:13:C5
Apr 22 18:38:14 willow cardmgr[2007]: executing: './network start eth0'

Here is cardctl status:
Socket 0:
  no card
Socket 1:
  5V 16-bit PC Card
  function 0: [ready]

Here is cardctl config:
Socket 0:
  not configured
Socket 1:
  Vcc 5.0V Vpp1 0.0V Vpp2 0.0V
  interface type is "memory and I/O"
  irq 3 [exclusive] [level]
  function 0:
    config base 0x0400
      option 0x50
    io 0x0300-0x031f [auto]

Here is cardctl ident:
Socket 0:
  no product info available
Socket 1:
  product info: "Network Everywhere", "Fast Ethernet 10/100 PC Card", "2.0", " "
  manfid: 0x0149, 0xc1ab
  function: 6 (network)

Here is /var/state/pcmcia/stab:
Socket 0: empty
Socket 1: NE2000 Compatible Ehternet
1 network pcnet_cs 0 eth0

/etc/pcmcia/config.opts is standard from 3.1.14 (I didn't modify it)

here is the important stuff from /etc/pcmcia/network.opts:
IF_PORT=""
BOOTP="n"
DHCP="N"
DHCP_HOSTNAME=""
IPADDR="10.1.1.50"
NETMASK="255.255.255.0"
NETWORK="10.1.1.0"
BROADCAST="10.1.1.255"
GATEWAY="10.1.1.1"
DOMAIN="tron.net"

everything else is as installed from 3.1.14.
1. Idea Look at #46 and #47 by Jochen Friedrich, 2000, Apr 23
1. Sad Didn't seem to help. by Larry Carter, 2000, Apr 23
(_ Note Possible interrupt conflict? by Jochen Friedrich, 2000, Apr 23
(_ Feedback IRQ change didn't help by Larry Carter, 2000, Apr 23
2. Angry Bad card!!! by Larry Carter, 2000, Apr 24

None Slow connections with Genius MF3000 ?

Date: 2000, Apr 27
From: Thomas Moestl whoo

I own a Genius MF3000 Ethernet Cardbus adaptor that I am using with a Sony Vaio PCG-F270 (Mobile Pentium II-333, 128 MB RAM, pcmcia-cs 3.1.14 reports a Ricoh RL5C478 PCI-to-Cardbus bridge). It is working, but the maximum outgoing data rate I get is about 3MB/s. When sending, there are no collisions/errors. It can receive at up to 8MB/s however, but then I am getting collisions.

The card is reported as follows:

Apr 26 17:29:10 flux cardmgr[84]: initializing socket 0
Apr 26 17:29:10 flux cardmgr[84]: socket 0: Tulip-based CardBus Fast Ethernet
Apr 26 17:29:10 flux kernel: cs: cb_alloc(bus 32): vendor 0x1011, device 0x0019
Apr 26 17:29:10 flux kernel: ROM image dump:
Apr 26 17:29:10 flux kernel: image 0: 0x000000-0x0001ff, signature PCIR
Apr 26 17:29:10 flux cardmgr[84]: executing: 'insmod /lib/modules/2.2.14/pcmcia/cb_enabler.o'
Apr 26 17:29:10 flux cardmgr[84]: executing: 'insmod /lib/modules/2.2.14/pcmcia/tulip_cb.o'
Apr 26 17:29:11 flux kernel: cs: cb_config(bus 32)
Apr 26 17:29:11 flux kernel: fn 0 bar 1: io 0x200-0x27f
Apr 26 17:29:11 flux kernel: fn 0 bar 2: mem 0x600c0000-0x600c03ff
Apr 26 17:29:11 flux kernel: fn 0 rom: mem 0x60080000-0x600bffff
Apr 26 17:29:11 flux kernel: cs: cb_enable(bus 32)
Apr 26 17:29:11 flux kernel: bridge io map 0 (flags 0x21): 0x200-0x27f
Apr 26 17:29:11 flux kernel: bridge mem map 0 (flags 0x1): 0x60080000-0x600c0fff
Apr 26 17:29:11 flux kernel: tulip_reap()
Apr 26 17:29:11 flux kernel: tulip_attach(bus 32, function 0)
Apr 26 17:29:11 flux kernel: tulip.c:v0.91g-ppc 7/16/99 becker@cesdis.gsfc.nasa.gov (modified by danilo@cs.uni-magdeburg.de for XIRCOM CBE, fixed by Doug Ledford)
Apr 26 17:29:11 flux kernel: eth0: Digital DS21143 Tulip rev 65 at 0x200, 00:A0:0C:90:7C:5F, IRQ 3.
Apr 26 17:29:11 flux kernel: eth0: EEPROM default media type Autosense.
Apr 26 17:29:11 flux kernel: eth0: Index #0 - Media 10baseT (#0) described by a 21142 Serial PHY (2) block.
Apr 26 17:29:11 flux kernel: eth0: Index #1 - Media 10baseT-FD (#4) described by a 21142 Serial PHY (2) block.
Apr 26 17:29:11 flux kernel: eth0: Index #2 - Media 100baseTx (#3) described by a 21143 SYM PHY (4) block.
Apr 26 17:29:11 flux kernel: eth0: Index #3 - Media 100baseTx-FD (#5) described by a 21143 SYM PHY (4) block.
Apr 26 17:29:11 flux cardmgr[84]: executing: './network start eth0'

I have tried to explicitely set the media types, but it does not change anything.
Anybody has the same problem or some idea? Could it be a limitation of the bridge/card/notebook?
1. None I don't know if this is unusual by David Hinds, 2000, Jun 03
(_ Question Seems to be an issue with my notebook by Thomas Moestl, 2000, Jun 03

Question Multicast and promisc issues

Date: 2000, Apr 27
From: David Shaw <dshaw@akamai.com>

I'm using a Xircom PCMCIA ethernet card (xirc2ps_cs) on a Dell Latitude CPx. MULTICAST is enabled (so says ifconfig). However, I can't receive multicast packets unless the interface is switched to PROMISC or ALLMULTI. I do see that the multicast group was joined (via/proc/net/igmp), but no packets arrive.

I've tried different ethernet cards with different drivers (the wavelan for example) and it has the same behavior.

I can duplicate the problem on both 3.1.10 and 3.1.14 using kernel 2.2.14.

Since this doesn't happen using a non-PCMCIA ethernet card, it seems like there is a bug somewhere (either in my configuration or in pcmcia-cs)

1. None I'm afraid it probably isn't tested much by David Hinds, 2000, Apr 27
(_ None Re: I'm afraid it probably isn't tested much by David Shaw, 2000, Apr 27
(_ None I don't think I can explain that by David Hinds, 2000, Apr 27

Question DFE660tx-problems with non-auto-negotiation

Date: 2000, Apr 28
From: Christian Froemmel ckf

I have the following problem with the D-Link DFE-660TX-Card: When I use the auto-negotiation of this card, the link is established, the half/full-duplex-led is in the beginning off, just when I transfer data the led starts to pulse. The switch is 100mbit-FD and there the half/full-duplex-led is not pulsing. mii-diag says, the card is on auto-negotiation and the switch advertised 100baseTx-FD 100baseTx 19BaseT-FD and 10baseT. When I set manually the card to option 14 of the tulip-driver (MII-100baseTx-FD) the link is not established and nothing happens (the mii-diag says: link was firstly broken, but now reestablished). Yes, I´ve changed the PCIC_OPTS to "pci_int=1 pci_csc=1" ;)

The laptop is an ibm-thinkpad 4187 with the O2Micro OZ6812 (rev 5) Cardbus-Controller. With windows the full-duplex is automatically set, nothing pulses and everything is going fine there (but its not my preferred OS).

Or is there somewhere a tool, where I can turn off the autonegotiation by default and force it to 100BaseTx-Fd (I did not found anything to setup at dlinks webpage, just diagnostic tools) ?

If anyone needs more debug-info or has THE idea, please write.

regards christian

1. None It isn't clear to me if you have a problem by David Hinds, 2000, Apr 28

Ok Success with NetGear FA510C

Date: 2000, Apr 30
From: Chris Love ChrisLove

This card wasn't expressly on the supported cards list, but is working fine on a Dell Inspiron 3800, Redhat Linux 6.2.

As cited on the "Unsupported PCMCIA Card list" (http://home.snafu.de/wehe/pcmcia_linux.html), the following entry in /etc/pcmcia/config was required:

	card "NetGear FA510C Fast Ethernet CardBus Card"
	manfid: 0x9513, 0x0081
	bind: tulip_cb

Older releases may need to compile the latest tulip driver; Netgear includes rev 0.91 of the tulip.c on the driver disk.

Kudos to all who have worked on this stuff! My early impressions are that linux on laptops is pretty slick.

	Chris

Question 2 PCMCIA cards in one laptop. Help please...

Date: 2000, Apr 30
From: J Toy gee308

Hi, I have 2 Linksys Etherfast PCMCIA cards. Im trying to use my Laptop as an IPMASQ server. My Linux distribution is redhat 6.2. My laptop is an HP Omnibook 4000 CT with a 486 50 mhz with 16 mb of RAM and a 503 MB HD. I can't get both PCMCIA cards to work at the same time. If I pull one out, the other works and I can switch them so the other works. ifconfig reports that both cards have different IRQs.When I do ifconfig while both cards are in, it shows that they are both running. My test to see if they are both working is: while both cards are in laptop, ping local IPs with one card attached to a LAN cable. Then switch LAN cable to other card. The IPs I gave the PCMCIA cards were both 192.168.0.x/24. I'm sure that it is possible to do this, but I'm not sure how. My current IPMASQ server has 2 Linksys etherfast NICs. I didn't need to do a special config, but things are probably different with PCMCIA. Also, how can I set an ethernet card's hardware address so that it boots with the same set hardware address everytime? I know how to set the harware address for the card, but after rebooting, it goes back to its default address( this is normal, but I need to know how and where to put the script to make it boot with a different hardware address).If you can, please email me your information as I don't surf the web often, only check email.

Thanks for the help in advance. Jason gee308@mediaone.net

1. None Sounds like a fishy test procedure by David Hinds, 2000, May 01
2. Feedback Setting another MAC address is no problem by Andreas Hotz, 2000, May 05

Question setting ethernet PCMCIA address

Date: 2000, May 01
From: J Toy gee308

Hi, I need help setting my ethernet Linksys etherfast PCMCIA card's hardware address automatically when it boots. I know how to set my harwade address with ifconfig eth0 hw ether FF:FFF:FF:FFF:FF:FF, but when I reboot, it resets. How can I make this automatic? Thanks Jason gee308@mediaone.net

1. None Don't know why you need it, but here it is by David Hinds, 2000, May 01
2. Feedback Setting the hardware address from network.opts by Andreas Hotz, 2000, May 05

Question LANPRO Realtek 8139 B/C supported?

Date: 2000, May 03
From: Gabriele Paciucci <paciucci@caldera.it>

Dear PCMCIA developer,
	Is the LANPRO Realtek 8139 B/C supported? I've downloaded a rtl8139.c file from realtek (with CardBus support), I've compiled in a 2.2.14 tree, with pcmcia-cs 3.1.14 (i think!!!). The card manager find my Intel PC Card controller but when i try to make modprobe rtl8139.o the kernel give me a beautiful "Resource Busy".

Any idea?

Thanks in advance...

1. None Not currently supported by David Hinds, 2000, May 03
1. None Re: Not currently supported by Gabriele Paciucci, 2000, May 03
2. Question Unidentified PCMCIA Card (mayby Realtek) by Gregor Suhr, 2000, Jun 28
1. None I'd go ahead and try it out by David Hinds, 2000, Jul 06

Question Hw. address read/write mishap 2 with i1500 and IBM Ethernet II

Date: 2000, May 09
From: Tim Gerla timg

I've been having problems for a long time with my IBM Thinkpad i1500 machine and this IBM Ethernet II network card. It works fine with a very specific configuration of kernel 2.2.12 and tools 3.1.8.

I'm currently trying to run kernel 2.3.99 with the tools 3.1.8, with teh built in pcmcia modules.

As soon as I insert the card, it is detected properly, I get an irq, io address, etc. The hardware address shows up properly, but I get the following messages in my syslog:

coruscant kernel: eth0: pcnet_reset_8390() did not complete.
(NE2000 Compatible stuff detected here)

Then I get:

Hw. address read/write mishap 2
Hw. address read/write mishap 4

An ifconfig shows everything in order, but I can't ping any outside hosts. I can ping myself, though. From the outside, it looks like a cable problem, but it is not.

For reference, the same exact problem occurs with a similar NE2000 compatible Kingston adapter. (KNE-PCM/M)

It's also happened with 2.2.13 and a recompiled tools set. I had to revert to 3.1.8 and 2.2.12 for things to work properly though.

An IBM modem card works fine. Another odd symptom: If I insert the modem card in slot 0, then insert the network card into slot 1, the entire kernel freezes. If I try popping out the network card, it goes into an infinite loop of "eth0: interrupt from stopped card." Nothing stops this, not even the magic sysrq key.

Any ideas? I'm getting sort of frustrated stuck with 2.2.12.

Thanks!

-Tim
timg@rrv.net
1. None Need to narrow down where the problem appears by David Hinds, 2000, May 09

Question OZ6832: "eth0: interrupt(s) dropped!"

Date: 2000, May 09
From: Axel Thimm Axel

Hello,

I have an as it seems common problem for O2Micro (at least on
Acer notebooks).

If my network card already receives packets during the
initialization of pcmcia-cs (either at boot time or
when plugging in the card with the connector attached),
I get dropped interrupts and some exit status 0x80.

If I start pcmcia with the cards connector detached and
attach it after the two beeps everything is perfect.

I already tried every possible solution to no extend,
including
- excluding all possible interrupts that might conflict
  (and it should not be an interrupt conflict, as I get
   almost 1MB/s on this cheap 10Mbit card)
- module "pcnet_cs" opts "mem_speed=1000 delay_output=1 delay_time=20"
- any combination in
  PCIC_OPTS="pci_int=9 irq_list=11,9 pci_csc=0 do_scan=0"

If it were not for trying to patch up the kernel with the
swsusp patch, I'd call this bug cosmetic, but it messes up
the suspend/resume cycle (I can overcome this by remembering
to remove the card before suspend, but I forget and the
machine messes up.)

System info:
Hardware: Acer TravelMate 513T, O2Micro 6832
          Longshine NE2000 compatible TypeII card.
Software: i686-pc-linux-gnu with 2.2.14 kernel (RedHat 6.2 but
          non RH modified kernel)
          pcmcia-cs-3.1.14

Symptoms:
May  9 20:06:08 pua cardmgr[3108]: executing: './network start eth0'
May  9 20:06:10 pua kernel: eth0: interrupt(s) dropped! 
May  9 20:06:10 pua kernel: eth0: Too much work at interrupt, status 0x80 
May  9 20:06:21 pua last message repeated 562 times

Thanks, Axel Thimm.

1. None Re: Question: OZ6832: "eth0: interrupt(s) dropped!" by David Hinds, 2000, May 11

Question Has anyone successfully compiled Kingmax-10/100 modified sources under 2.2.x?

Date: 2000, May 10
From: Carlos Betancourt cbetan

If there exists drivers for this cards, although non standard, why is it so hard to compile them? I mean, at least someone should have successfully compiled then, otherwise those drivers wouldn't exist. I'm using kernel 2.2.14 and pcmcia-cs-3.1.14. There is a 8390.c source for 2.2.x kernels within a message addressed to this list on november '99; however, it doesn't compile. I know that this cards are not 100% NE2000 compatible, and it is not officially supported by the kernel nor pcmcia-cs, but there must be a way to make them work with the modified sources. Please, help me, I'm desperate. I have to keep using windows in order to use my cable modem, and I've not money to buy another net card. It's a nightmare being forced to use windows. I would gladly write up a mini-howto for this cards, but I must get it to work first. The kingmax chipset is included in lots of generic network cards, so a lot of linux users have or will have troubles with these bitches. TIA

Carlos Betancourt carlos.betancourt@chello.be

1. Sad Untitled by Chris Lo, 2000, Jun 09
(_ None I wouldn't hold your breath by David Hinds, 2000, Jun 09

Question Genius MF3000 stalls 3.1.14

Date: 2000, May 16
From: Philip Martin philip

 kernel 2.2.15
 pcmcia 3.1.14

I have a Genius MF3000 network adapter that gets identified as

 May 16 21:26:29 debian1 cardmgr[155]: initializing socket 0
 May 16 21:26:29 debian1 cardmgr[155]: socket 0: Tulip-based CardBus Fast Ethernet
 May 16 21:26:29 debian1 kernel: cs: cb_alloc(bus 32): vendor 0x1011, device 0x0019
 May 16 21:26:29 debian1 kernel: ROM image dump:
 May 16 21:26:29 debian1 kernel:   image 0: 0x000000-0x0001ff, signature PCIR
 May 16 21:26:29 debian1 cardmgr[155]: executing: 'insmod /lib/modules/2.2.15/pcmcia/cb_enabler.o'
 May 16 21:26:29 debian1 cardmgr[155]: executing: 'insmod /lib/modules/2.2.15/pcmcia/tulip_cb.o'
 May 16 21:26:29 debian1 kernel: cs: cb_config(bus 32)
 May 16 21:26:29 debian1 kernel:   fn 0 bar 1: io 0x200-0x27f
 May 16 21:26:29 debian1 kernel:   fn 0 bar 2: mem 0x600c0000-0x600c03ff
 May 16 21:26:29 debian1 kernel:   fn 0 rom: mem 0x60080000-0x600bffff  
 May 16 21:26:29 debian1 kernel: cs: cb_enable(bus 32)
 May 16 21:26:29 debian1 kernel:   bridge io map 0 (flags 0x21): 0x200-0x27f
 May 16 21:26:29 debian1 kernel:   bridge mem map 0 (flags 0x1): 0x60080000-0x600c0fff
 May 16 21:26:29 debian1 kernel: tulip_reap()
 May 16 21:26:29 debian1 kernel: tulip_attach(bus 32, function 0)
 May 16 21:26:29 debian1 kernel: tulip.c:v0.91g-ppc 7/16/99 becker@cesdis.gsfc.nasa.gov (modified by danilo@cs.uni-magdeburg.de for XIRCOM CBE, fixed by Doug Ledford)
 May 16 21:26:29 debian1 kernel: eth0: Digital DS21143 Tulip rev 65 at 0x200, 00: A0:0C:90:6D:9A, IRQ 10.
 May 16 21:26:29 debian1 kernel: eth0:  EEPROM default media type Autosense.
 May 16 21:26:29 debian1 kernel: eth0:  Index #0 - Media 10baseT (#0) described by a 21142 Serial PHY (2) block.
 May 16 21:26:29 debian1 kernel: eth0:  Index #1 - Media 10baseT-FD (#4) described by a 21142 Serial PHY (2) block.
 May 16 21:26:29 debian1 kernel: eth0:  Index #2 - Media 100baseTx (#3) described by a 21143 SYM PHY (4) block.
 May 16 21:26:29 debian1 kernel: eth0:  Index #3 - Media 100baseTx-FD (#5) described by a 21143 SYM PHY (4) block.
 May 16 21:26:29 debian1 cardmgr[155]: executing: './network start eth0'

Once the network interface is up and running I can use NFS and ssh successfully. However transferring large files with scp or network printing does not work--the transfer stalls. Setting the tulip_cb debug option I see Transmit errors:

 May 16 20:42:04 debian1 kernel:  In tulip_rx(), entry 13 005a0320.
 May 16 20:42:04 debian1 kernel: eth0: In tulip_rx(), entry 13 005a0320.
 May 16 20:42:04 debian1 kernel: eth0: interrupt  csr5=0xf0660000 new csr5=0xf0660000.
 May 16 20:42:04 debian1 kernel: eth0: exiting interrupt, csr5=0xf0660000.
 May 16 20:42:04 debian1 kernel: eth0: interrupt  csr5=0xf0670004 new csr5=0xf0660000.
 May 16 20:42:04 debian1 kernel: eth0: interrupt  csr5=0xf0660000 new csr5=0xf0660000.
 May 16 20:42:04 debian1 kernel: eth0: exiting interrupt, csr5=0xf0660000.
 May 16 20:42:04 debian1 kernel: eth0: interrupt  csr5=0xf0670040 new csr5=0xf0660000.
 May 16 20:42:04 debian1 kernel:  In tulip_rx(), entry 14 005a0320.
 May 16 20:42:04 debian1 kernel: eth0: In tulip_rx(), entry 14 005a0320.
 May 16 20:42:04 debian1 kernel: eth0: interrupt  csr5=0xf0660000 new csr5=0xf0660000.
 May 16 20:42:04 debian1 kernel: eth0: exiting interrupt, csr5=0xf0660000.
 May 16 20:42:04 debian1 kernel: eth0: interrupt  csr5=0xf0670004 new csr5=0xf0660000.
 May 16 20:42:04 debian1 kernel: eth0: Transmit error, Tx status 7fffb200.
 May 16 20:42:04 debian1 kernel: eth0: interrupt  csr5=0xf0660000 new csr5=0xf0660000.
 May 16 20:42:04 debian1 kernel: eth0: exiting interrupt, csr5=0xf0660000.
 May 16 20:42:04 debian1 kernel: eth0: interrupt  csr5=0xf0670004 new csr5=0xf0660000.
 May 16 20:42:04 debian1 kernel: eth0: Transmit error, Tx status 7fffb200.
 May 16 20:42:04 debian1 kernel: eth0: interrupt  csr5=0xf0660000 new csr5=0xf0660000.

When this happens the tulip_diag program reports:

 tulip-diag.c:v1.19 10/2/99 Donald Becker (becker@cesdis.gsfc.nasa.gov)
 Index #1: Found a Digital DS21143 Tulip adapter at 0x200.
  Port selection is 100mbps-SYM/PCS 100baseTx scrambler, full-duplex.
  Transmit started, Receive started, full-duplex.
   The Rx process state is 'Waiting for packets'.
   The Tx process state is 'Idle'.
   The transmit threshold is 128.
   The NWay status register is 45e1d2cd.
   Internal autonegotiation state is 'Negotiation complete'.

An strace of the scp process shows it stalled in a write():

 write(6, "\376\f\0\211\300PW\350\370&\375\377\203\304\20\350\20,"..., 4096) = 4096
 read(3, "\350\37\34\375\377\350\312\37\375\377\215\266\0\0\0\0\213"..., 4096) = 4096
 write(6, "\350\37\34\375\377\350\312\37\375\377\215\266\0\0\0\0\213"..., 4096) = 4096
 read(3, "\350\17\r\375\377\203\304\20\203\304\370\203\304\364\213"..., 4096) = 4096
 write(6, "\350\17\r\375\377\203\304\20\203\304\370\203\304\364\213"..., 4096) = 4096
 read(3, "\203\304\364\213E\10\203\300\4P\350Q\2\24\0\203\304\20"..., 4096) = 4096
 write(6, "\203\304\364\213E\10\203\300\4P\350Q\2\24\0\203\304\20"..., 4096) = 4096
 read(3, "\354\24S\213]\10\307\3|\236 \10\203\304\370j\2\215C\20"..., 4096) = 4096
 write(6, "\354\24S\213]\10\307\3|\236 \10\203\304\370j\2\215C\20"..., 4096 
                                                                           ^it hangs here

If I cancel the scp file transfer I can then do an NFS copy at several Mb/second.

I also have a 3Com 3CCFEM556B which works fine, so the problem would appear to lie with the tulip driver.

Any ideas?

Philip

1. None I think I need a bit more information by David Hinds, 2000, Jun 06
(_ None Re: I think I need a bit more information by Philip Martin, 2000, Jun 06

Sad When will pcnet_cs work with Asix AX88190 Chipset?

Date: 2000, May 24
From: Ray jxray

i installed 3.1.15. pcnet_cs told me this poor card was in the blacklist. :( so, when will this card be supported? or, will it be supported?

1. None We're working on it by David Hinds, 2000, May 25
1. Ok Thank u. by Ray, 2000, May 26
2. Note It works with 3.1.15 by Carlos Betancourt, 2000, Jun 09

Question O2Micro OZ6836: Extremly slow Ethernet-Card

Date: 2000, May 31
From: Dag Kröper dagkro

Hi!

Can everyone help me ????

I've a problem with my Ethernet-Card. The card will be recognized correctly, but it has a very slow Data-Transfer
(about 10 KBytes/sec)

Infos (about my laptop) (I hope they are sufficient):

CPU: Mobile-Pentium II 333 MHz
Chip-Set: Intel MX
RAM: 64 MB
HDD: 4,5 GB Fujitsu
Card-Bus-Controller: O2Micro OZ6836
Ethernet-Card: Longshine LS 8534 TB-A1 10base2 or 10baseT

cardctl config:
================
Socket 0:
  not configured
Socket 1:
  Vcc 5.0V Vpp1 0.0V Vpp2 0.0V
  interface type is "memory and I/O"
  irq 10 [exclusive] [level]
  function 0:
    config base 0x0fd0
      option 0x41 status 0x00 copy 0x00
    io 0x0300-0x031f [auto]

cardctl ident
=============
Socket 0:
  no product info available
Socket 1:
  product info: "Ethernet", "Adapter", "2.0"
  manfid: 0x0149, 0xc1ab
  function: 6 (network)

dmesg:
======
Linux PCMCIA Card Services 3.1.15
  kernel build: 2.2.15 #4 Tue May 30 15:37:59 MEST 2000
  options: [pci] [cardbus] [apm]
PCI routing table version 1.0 at 0xfdf00
  00:03.0 -> irq 11
  00:03.1 -> irq 11
Intel PCIC probe:
  O2Micro OZ6836/60 rev 62 PCI-to-CardBus at slot 00:03, mem 0xe1000000
    host opts [0]: [ring] [pci/way] [pci irq 11] [lat 32/176] [bus 32/34]
    host opts [1]: [ring] [pci/way] [pci irq 11] [lat 32/176] [bus 35/37]
    ISA irqs (default) = 3,4,7,9,10,12 PCI status changes
cs: IO port probe 0x1000-0x17ff: clean.
cs: IO port probe 0x0100-0x04ff: excluding 0x408-0x40f 0x4d0-0x4d7
cs: IO port probe 0x0a00-0x0aff: clean.
cs: memory probe 0xa0000000-0xa0ffffff: excluding 0xa0000000-0xa00fffff
eth0: NE2000 Compatible: io 0x300, irq 10, hw_addr 00:E0:98:7B:D2:A3
1. None The messages look good by David Hinds, 2000, May 31
1. None Re: The messages look good by Dag Kröper, 2000, Jun 02
2. Idea Solve Problem :-) by Dag Kröper, 2000, Jul 10

Question Question about epic_cb driver

Date: 2000, Jun 02
From: Brian Mays brian

Looking at the Configure script, it appears that the epic_cb client is built if CARDBUS support is enabled:

   725	if [ "$CONFIG_CARDBUS" = "y" ] ; then
   726	    if grep CARDBUS ${LINUX}/drivers/net/epic100.c >/dev/null ; then
   727		echo DO_EPIC_CB=y >> $CONFIG_MK
   728	    fi
   729	fi

I have found that if epic support (CONFIG_EPIC100) has not been compiled into the kernel, the following unresolved symbol results:

    # depmod -ae
    depmod: *** Unresolved symbols in /lib/modules/2.2.15/pcmcia/epic_cb.o
    depmod:         epic100_probe

Shouldn't the epic_cb be built only if epic support has been built into the kernel? Have I missed something here?

Thanks,

Brian

1. None The epic_cb driver got broken in 2.2.15 by David Hinds, 2000, Jun 02
(_ None Re: The epic_cb driver got broken in 2.2.15 by Brian Mays, 2000, Jun 02

Feedback Corega 10M EtherII PCC-T failed detection

Date: 2000, Jun 03
From: Michael Fitzpatrick leahcim

I bought a Corega 10M PCMCIA PCC-T card @ Linux Expo 2000.

Plugged it in and got an Oops from CardMgr loading tulip_cb drivers.

Turns out this card isn't in the config file (only a slight difference in version info from another Corega card though)

Biggest problem is the manuf ID it reports clashes with another card in the database.

It was trivial for me to add an entry for this card (it uses pcnet_cs like the other Corega cards already defined) and comment out the clashing card, and once that's done the card works without problems.

Info returned from cardctl ident for Corega :-

Socket 1:
  product info: "corega K.K.", "corega EtherII PCC-T"
  manfid: 0xc00f, 0x0000
  function: 6 (network)

config entry for manfid clash :-

#card "Allied Telesyn AT-2800 10/100 Fast Ethernet"
# manfid 0xc00f, 0x0000
# bind "tulip_cb"

Entry I added to config :-

card "Corega PCC-T EtherII"
  version "corega K.K.", "corega EtherII PCC-T"
  bind "pcnet_cs"

1. None Apologies by Michael Fitzpatrick, 2000, Jun 03

Question Abocom FE1500 supported ???

Date: 2000, Jun 06
From: Julien pwaro

Hi,

I plan to buy an Abocom FE1500 Ethernet card. On the abocom's web site, it is claimed that this card is fully supported on Linux. But on the list of supported cards by pcmcia-cs, I have only found the FE1000... So what's the truth ? Is it really supported or not ?

Thank you if you can help me ;-)

Mr Pwaro

1. None Probably the wrong place to ask by David Hinds, 2000, Jun 06
(_ Ok FE-1500 works !!! by Julien, 2000, Jun 26

Question OvisLink LFS-PCM-32 not working unless in promiscous mode

Date: 2000, Jun 13
From: Miguel mtavares

I've notice this strange behaviour when using the refered in the subject.

After inserting the device and cardmgr beeps twice, so I guess that it is being recognize, I canot ping to the outside of my computer. But, if I start a tcpdump (getting eth0 in promiscuous mode) then I'm able to ping, althoug the network connectiong doesn't seem very reliable.

This board is being identified as a tulip_cb, so my guess is that I'm probably missing some configuration parameters.

More information (that I hope can help): - The computer is a Toshiba 4070CDS - I'm using Suse 6.4 with the upgraded pcmcia package ('thou I wouldn't mind having to compile and install it myself).

		Best regards to all

				Miguel

1. None You may have to live with that for now by David Hinds, 2000, Jun 14

None Minor mistake in /etc/pcmcia/network script

Date: 2000, Jun 15
From: Brian Mays brian

There is a typo in the etc/network script. The line

    if test -x /usr/sbin/ipmasq && istrue "$IPMASQ" ; then

should read

    if test -x /usr/sbin/ipmasq && is_true "$IPMASQ" ; then

1. None Hmmm... by David Hinds, 2000, Jun 15

Sad invalid operand: 0000

Date: 2000, Jun 24
From: Nuitari

I get this error whenever cardmgr detects my 
3Com TokenLink III (3c689)

It happends with both the 2.3.99-pre9 and 2.4.0-test1-ac22-classproc kernels.

The token ring driver is built in the kernel.
Before I updated to 3.1.17, I think I was using version 3.1.9 (not sure).

The box is an Ascentia 910N from AST (part number 501589-301).

Note that the serial part (an 14.4 kbps data/fax modem from AST, part number 234091-001) does not cause any problem at all.


[nuitari@gandalf nuitari]$ cat bug.cardmgr 
Jun 24 02:50:33 zifnab cardmgr[684]: starting, version is 3.1.11
Jun 24 02:50:33 zifnab cardmgr[684]: watching 2 sockets
Jun 24 02:50:33 zifnab kernel: cs: IO port probe 0x0c00-0x0cff: clean.
Jun 24 02:50:33 zifnab kernel: cs: IO port probe 0x0800-0x08ff: clean.
Jun 24 02:50:33 zifnab kernel: cs: IO port probe 0x0100-0x04ff: excluding
0x100-0x107
Jun 24 02:50:33 zifnab kernel: cs: IO port probe 0x0a00-0x0aff: clean.
Jun 24 02:50:33 zifnab cardmgr[684]: initializing socket 1
Jun 24 02:50:33 zifnab cardmgr[684]: socket 1: IBM Token Ring Adapter
Jun 24 02:50:33 zifnab kernel: invalid operand: 0000
Jun 24 02:50:33 zifnab kernel: CPU:    0
Jun 24 02:50:33 zifnab kernel: EIP:    0010:[ibmtr_probe+32/128]
Jun 24 02:50:33 zifnab kernel: EFLAGS: 00010a07
Jun 24 02:50:33 zifnab kernel: eax: c02b2fe0   ebx: c14c2600
ecx: c1a5d30c   edx: c14c2600
Jun 24 02:50:33 zifnab kernel: esi: 0000002b   edi: c03a7a24
ebp: c10d2400   esp: c03a79a4
Jun 24 02:50:33 zifnab kernel: ds: 0018   es: 0018   ss: 0018
Jun 24 02:50:33 zifnab kernel: Process cardmgr (pid: 684,
stackpage=c03a7000)
Jun 24 02:50:33 zifnab kernel: Stack: c01c94cf c14c2600 c03a79dc c01df6a8
c14c2600 c14c2600 c10d2400 c14c2600
Jun 24 02:50:33 zifnab kernel:        c1a5d358 c1a5d300 c10d249c c10d24c8
c14c2600 c10d2400 000cc000 c03a0000
Jun 24 02:50:33 zifnab kernel:        08002101 c01dff0b c10f6ba0 c03a7b00
c03a7b44 c03a7b44 c10f6ba0 42490104
Jun 24 02:50:33 zifnab kernel: Call Trace: [register_trdev+15/48]
[ibmtr_config+760/976] [tok_interrupt+443/3584]
[pcmcia_get_next_tuple+125/560] [pcmcia_get_next_tuple+163/560]
[<ffffc0c6>] [pcmcia_validate_cis+278/336]
Jun 24 02:50:33 zifnab kernel:        [msstab+28574/29086]
[kfree_skbmem+38/128] [__kfree_skb+265/272] [ibmtr_event+202/416]
[pcmcia_register_client+503/528] [register_netdevice+397/416]
[tr_configure+0/96] [init_netdev+182/208]
Jun 24 02:50:33 zifnab kernel:        [CardServices+414/848]
[list.927+6827/11800] [ibmtr_attach+364/416] [ibmtr_event+0/416]
[bind_request+352/432] [bind_request+392/432] [ds_ioctl+1213/1488]
[do_anonymous_page+92/272]
Jun 24 02:50:34 zifnab kernel:        [do_no_page+48/176]
[handle_mm_fault+240/352] [do_page_fault+391/1360] [msstab+28574/29086]
[proc_get_inode+65/288] [proc_lookup+154/176] [vsprintf+822/880]
[free_page_and_swap_cache+470/480]
Jun 24 02:50:34 zifnab kernel:        [zap_page_range+358/480]
[unmap_fixup+283/304] [do_munmap+612/640] [sys_ioctl+359/400]
[system_call+52/64] [stext+43/309]

Jun 24 02:50:34 zifnab kernel: Code: ff eb 57 50 57 e8 56 00 00 00 83 c4
08 85 c0 75 44 31 c0 eb


uname -rv
2.4.0-test1-ac22-class #1 Sat Jun 24 00:47:33 EDT 2000

cardctl -V
cardctl version 3.1.11

[nuitari@gandalf nuitari]$ 

Sad Again: eth0: interrupt(s) dropped!

Date: 2000, Jun 26
From: Michael Brandt michab

hi,

i just installed linux (suse64) on my laptop using pcmcia 3.1.17. eth0 is set up correcty via dhcp but i'm getting these messages:

kernel: eth0: Reentering the interrupt handler! isr=0x0 imr=0x0.
last message repeated 3 times
kernel: eth0: interrupt(s) dropped!

additionaly i'm not able to build up an connection....

under win irq_11 is assigned to my graphiccard, irq_10 seem to be free..

how can i fix this ?? any ideas ??

thx,
micha

some settings & messages:
---------------------------------

PCIC=i82365
PCIC_OPTS="pci_int=1 pci_irq_list=11 irq_list=10"


config:
---------
Socket 0:
  Vcc 5.0V  Vpp1 0.0V  Vpp2 0.0V
  interface type is "memory and I/O"
  irq 10 [exclusive] [level]
  function 0:
    config base 0x03c0
      option 0x47 status 0x00
    io 0x0300-0x031f [auto]
Socket 1:
  not configured

ident:
---------
Socket 0:
  product info: "PCMCIA", "FAST ETHERNET CARD"
  manfid: 0xffff, 0x1090
  function: 6 (network)
Socket 1:
  no product info available

status:
----------
Socket 0:
  5V 16-bit PC Card
  function 0: [ready]
Socket 1:
  no card

messages:
-----------
kernel: Linux PCMCIA Card Services 3.1.17
kernel:   kernel build: 2.2.14 #1 Mon Mar 13 10:54:26 GMT 2000
kernel:   options:  [pci] [cardbus] [apm]
kernel: PCI routing table version 1.0 at 0xfdf80
kernel: Intel PCIC probe: 
kernel:   O2Micro OZ6832/33 rev 34 PCI-to-CardBus at slot 00:0a, mem 0x68000000
kernel:     host opts [0]: [ring] [pci/way] [pci irq 11] [lat 168/176] [bus 32/34]
kernel:     host opts [1]: [ring] [pci/way] [no pci irq] [lat 168/176] [bus 35/37]
kernel:     PCI irq 11 test failed
kernel:     ISA irqs (default) = 10 polling interval = 1000 ms
cardmgr[253]: starting, version is 3.1.17
cardmgr[253]: watching 2 sockets
kernel: cs: IO port probe 0x0c00-0x0cff: excluding 0xcf8-0xcff
kernel: cs: IO port probe 0x0800-0x08ff: clean.
kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x398-0x39f 0x4d0-0x4d7
kernel: cs: IO port probe 0x0a00-0x0aff: clean.
cardmgr[253]: initializing socket 0
kernel: cs: memory probe 0xa0000000-0xa0ffffff: excluding 0xa0000000-0xa00fffff
cardmgr[253]: socket 0: NE2000 Compatible Fast Ethernet
cardmgr[253]: executing: 'modprobe 8390'
cardmgr[253]: executing: 'modprobe pcnet_cs mem_speed=600 hw_addr=0x00,0x50,0x22,0xF2,0x1D,0x58'
kernel: pcnet_cs: sorry, the AX88190 chipset is not supported.
kernel: eth0: NE2000 Compatible: io 0x300, irq 10, hw_addr 00:50:22:F2:1D:58
cardmgr[253]: executing: './network start eth0'
kernel: eth0: interrupt from stopped card
kernel: eth0: interrupt(s) dropped!
kernel: eth0: Reentering the interrupt handler! isr=0x0 imr=0x0.
last message repeated 3 times
kernel: eth0: interrupt(s) dropped!

1. None This network chipset isn't supported for the moment by David Hinds, 2000, Jul 05

Question No such file or directory?

Date: 2000, Jun 28
From: John Cusick <zen@sprynet.com>

Am attempting to configure PCMCIA network card and receive following respons to ./network start eth0 command:

  cat: /var/lib/pcmcia/scheme: No such file or directory

Previous cardmgr commands in start up script appear to be successful.

System details: IBM TPAD 770ED, Suse 6.4 upgraded to 2.2.16 kernel, PCMCIA 3.1.17, Intel CardBus 32 Pro100
1. None Update the PCMCIA init script... by David Hinds, 2000, Jul 05
(_ Feedback Readonly Variable? by John Cusick, 2000, Jul 06
(_ None That message is not an error by David Hinds, 2000, Jul 06
(_ Feedback Other message is "failed services" by John Cusick, 2000, Jul 06
(_ Note Intel CardBus 32 Works! by John Cusick, 2000, Jul 07

Note Addition to config file for IBM EtherJet/Modem combo

Date: 2000, Jul 03
From: Peter A. Friend pafriend

I have an IBM ThinkPad 600X with the 10/100 EtherJet CardBus Ready Port Adapter with 56k Modem. I haven't attempted anything with the modem yet, but I got the NIC to work by adding this line to /etc/pcmcia/config

card "IBM 10/100 EtherJet CardBus Adapter with 56k Modem"
  manfid 0x00a4, 0x0103
  bind "tulip_cb"

I got this info by running:

cardctl ident

This has worked for me on RedHat 6.1 and Mandrake 7.0.

Question Question about Compex Linkport card.

Date: 2000, Jul 06
From: Alexandr Berezniatsky Alexandr

I have pcmcia card -Compex Linkport/ENET-A I would like to install Linux on my PC. So I would like to know if my card works in Linux. Thanks!

1. None You can check for yourself by David Hinds, 2000, Jul 06

Sad Problems with Compex Linkport

Date: 2000, Jul 09
From: Alexandr Berezniatsky Alexandr

I have pcmcia card -Compex Linkport /ENET-A. (I/O Base Address default 320H,Interrupt default 11) (Linux kernel 2.0.34, PCMCIA package 3.0.0) When I insert my card ,Linux detects it but it detects incorrect card (while booting kernel it is written LinkSys Ethercard 10/100Tbase,I/O=300H,IRQ=3) then system gives information that my card wasn't correctly set. What can be done to solve this problem? Thanks!

1. None Need more info by David Hinds, 2000, Jul 10
(_ Note additional info by Alexandr Berezniatsky, 2000, Jul 10
(_ None Re: Note: additional info by David Hinds, 2000, Jul 10
(_ Ok Thanks!!!!:) by Alexandr Berezniatsky, 2000, Jul 11

Question Psion Gold Card 10-100Mb onto Toshiba(!) 310CDS with Redhad 6.1

Date: 2000, Jul 14
From: Mike Jones nimpo

Looked at SUPPORTED.CARDS for Fast Ethernet Psion Gold Card and found it under driver: smc91c92_cs. Updated to pcmcia-cs-3.1.18 to cope with this card.

But I get "Yikes! bad chip signature" error:
dmesg > dmesg.txt

um... other info:
/sbin/cardctl ident > ident.txt
/sbin/cardctl config > config.txt
/sbin/ifup eth0 > ifup.txt

Oh by the way, have Psion Gold card 56K modem in other slot which still works!

Any suggestions?
Mike.

1. None Your card is not the same by David Hinds, 2000, Jul 14
(_ More It pretends to be a netglobal card. by Mike Jones, 2000, Jul 14
(_ None Re: More: It pretends to be a netglobal card. by David Hinds, 2000, Jul 14
(_ Note smb91c92_cs failed with "exclude port 0x100-0x1ff" by Mike Jones, 2000, Jul 14
(_ More No joy yet, small progress by Mike Jones, 2000, Jul 16
(_ None Oh, of course... by David Hinds, 2000, Jul 17
(_ Sad Not Jack of Diamonds ... I thought: 7 of Diamonds (but looks like I'm Wrong) by Mike Jones, 2000, Jul 24
(_ More Summary of what I've done so far (that's not worked) by Mike Jones, 2000, Jul 25
(_ Ok Got it to work Fix enclosed... by Mike Jones, 2000, Jul 27

Question Failure to recognise D-Link DE650

Date: 2000, Jul 17
From: Mike Dwerryhouse MikeDw

I have a Dell XPi100 Laptop, on which I have been running Slackware 3.6 (Kernel 2.0.35, pcmcia-cs 3.0.5) and successfully using a D-Link DE650 Combo card. I have been trying to upgrade to Slack 7.1 (kernel 2.2.16, pcmcia-cs 3.1.16). The card apparently gives no ident string - cardctl ident says "no product info available" on both installations. However, the older system successfully identifies and configures it using pcnet_cs. The new system fails to recognise it.

Can I persuade the new setup to recognise the card ? Perhaps by editing a config file ?. Or can I manually load the drivers. If I do a modprobe pcnet_cs, nothing happens. No eth0 device.

I currently have the system dual booting, so I can try both installations. I also have a modem card which is successfully identified and configured on both setups.

Question Hamlet FE1500 poor performance problem.

Date: 2000, Jul 18
From: Alessandro De Salvo desalvo

I have a brand new Hamlet FE1500 FastEthernet card. It is correctly configured and recognized as a NE2000 compatible ethernet card, using the version 3.1.18 of the pcmcia card services. The problem is that the performance is really bad!!! This means that I cannot achieve more than ~40-50 Kb/sec on a 10Mbits network under Linux. The problem disappear using Windows (~640 Kb/s, which is reasonable), so this is not a problem of my notebook/card. I tried to use the same interrupts and memory addresses as in Windows but it did not work, and the strangest thing is that in Linux the card does not receive any incoming traffic other than its own traffic (that should be not the case since I am connected to a shared network)! Does anyone has a suggestion/solution? Thanks in advance

1. None Need more detailed information by David Hinds, 2000, Jul 18
(_ None More informations... by Alessandro De Salvo, 2000, Jul 18
(_ None Syslog dump by Alessandro De Salvo, 2000, Jul 20
(_ None Maybe a driver upgrade will help by David Hinds, 2000, Jul 20
(_ None Upgrade done, but nothing different... by Alessandro De Salvo, 2000, Jul 21
(_ None Back to kernel 2.2.16-3. by Alessandro De Salvo, 2000, Jul 21
(_ None More syslog + other infos by Alessandro De Salvo, 2000, Jul 21
(_ None The problem appears to be only in RX by Alessandro De Salvo, 2000, Jul 24

Sad Strange 10/100 card Allied Telesyn 2800TX

Date: 2000, Jul 21
From: Mircea Ciocan cmircea

	Please help with this card. Is an Allied Telesyn 
AT-2800TX V2 or at leastthat is what the label and windoze reports it.
	But when used with the latest pcmcia code .18, is not recognized and and in syslog there is a record that say:
"Unsuported adapter:
Allied Telesin K. K. CentreCom LA-100-CardBus-T V.2
manfid: 0xc00f, 0x0000 function: 6 (network)"
	And that's it, no intitialization whatsoever, no i/o, irq or memory range, nothing, the dumping tools say, so I can't use tulip-diag program of Donald Becker to see if is tulip compatible card (AT2800 is listed  as supported via tulip_cs driver).
	Now the Telesyn/Telesis K.K. sound dubious to me, maybe is a clone, even was bought from an authorised distributor of AT :( but anyhow, for what is worth, that is the values that windoze allocates for it:

	IRQ=11
	I/O=0x1000-0x107f
	MEM=0x02100000-0x021003ff
	MEM=0x020C000-0x020Cfff
	Is some way of force init the card to try some tulip tests on it or I just bought myself a piece of windoze shit ( misleaded by the fact that is mentioned in compat. list)
	I mention that in the same notebook a 3Com modem 56K works great ( acttualy I was afraid that the modem will be incompatible :)

1. None Just edit /etc/pcmcia/config by David Hinds, 2000, Jul 21

Sad MELCO LPC3-TX "eth0: interrupt(s) dropped!"

Date: 2000, Jul 23
From: Gary Chan gchan1

The problem of "eth0: interrupt(s) dropped!" cannot be solved.
I have already tried all the possible solution.
I tried all "irq" and used "do_scan=0" parameters.

I love LINUX and don't want to use MS-Window any more.
PLease help me.
Many thanks.

Hardware:
LAN Card: Buffalo MELCO LPC3-TX PCMCIA NIC 10/100.
Laptop  : NEC VERSA NOTE

The following is syslog dump.
----------------------------------------------
Jul 23 10:00:46 localhost kernel:   kernel build: 2.2.14-1mdklinus #1 Tue Jan 4 21:39:25 CET 2000 
Jul 23 10:00:46 localhost kernel:   options:  [pci] [cardbus] [apm] 
Jul 23 10:00:46 localhost kernel: PCI routing table version 1.0 at 0xfdf70 
Jul 23 10:00:46 localhost kernel: Intel PCIC probe:  
Jul 23 10:00:46 localhost kernel:   TI 1221 rev 00 PCI-to-CardBus at slot 00:0a, mem 0x68000000 
Jul 23 10:00:46 localhost kernel:     host opts [0]: [serial pci & irq] [pci irq 10] [lat 168/176] [bus 32/34] 
Jul 23 10:00:46 localhost kernel:     host opts [1]: [serial pci & irq] [pci irq 10] [lat 168/176] [bus 35/37] 
Jul 23 10:00:46 localhost kernel:     ISA irqs (default) = 3,4,7,9,11 PCI status changes 
Jul 23 10:00:46 localhost cardmgr[3198]: starting, version is 3.1.19
Jul 23 10:00:46 localhost cardmgr[3198]: watching 2 sockets
Jul 23 10:00:46 localhost kernel: cs: IO port probe 0x0100-0x03ff: excluding 0x378-0x37f 
Jul 23 10:00:47 localhost cardmgr[3198]: initializing socket 0
Jul 23 10:00:47 localhost kernel: cs: memory probe 0xa000000-0xa0fffff: clean. 
Jul 23 10:00:47 localhost cardmgr[3198]: socket 0: Melco LPC3-TX Fast Ethernet
Jul 23 10:00:47 localhost cardmgr[3198]: executing: 'modprobe 8390'
Jul 23 10:00:47 localhost cardmgr[3198]: executing: 'modprobe pcnet_cs'
Jul 23 10:00:47 localhost kernel: eth0: NE2000 Compatible: io 0x300, irq 3, hw_addr 00:A0:0C:43:70:AA 
Jul 23 10:00:47 localhost cardmgr[3198]: executing: './network start eth0'
Jul 23 10:00:49 localhost kernel: eth0: interrupt(s) dropped! 
Jul 23 10:00:49 localhost kernel: eth0: Reentering the interrupt handler! isr=0x0 imr=0x0. 
Jul 23 10:00:50 localhost last message repeated 3 times
Jul 23 10:00:54 localhost kernel: eth0: interrupt(s) dropped! 
Jul 23 10:00:54 localhost kernel: eth0: Reentering the interrupt handler! isr=0x0 imr=0x0. 
Jul 23 10:01:00 localhost PAM_pwdb[3224]: (su) session opened for user news by (uid=0)
Jul 23 10:01:00 localhost PAM_pwdb[3224]: (su) session closed for user news
Jul 23 10:01:00 localhost kernel: eth0: interrupt(s) dropped! 
Jul 23 10:01:00 localhost kernel: eth0: Reentering the interrupt handler! isr=0x0 imr=0x0. 
Jul 23 10:01:13 localhost kernel: eth0: interrupt(s) dropped! 
Jul 23 10:01:13 localhost kernel: eth0: Reentering the interrupt handler! isr=0x0 imr=0x0. 
Jul 23 10:01:39 localhost kernel: eth0: interrupt(s) dropped! 
Jul 23 10:01:39 localhost kernel: eth0: Reentering the interrupt handler! isr=0x0 imr=0x0. 
Jul 23 10:01:47 localhost dhcpcd[3217]: timed out waiting for a valid DHCP server response 
Jul 23 10:01:47 localhost cardmgr[3198]: start cmd exited with status 1
------------------------------------------



Best regards.

1. None It is more than an interrupt issue by David Hinds, 2000, Jul 24
(_ Ok Will it be supported in the near future? by Gary Chan, 2000, Jul 25
(_ None Nope by David Hinds, 2000, Jul 25

Sad Network card doesn't work after warm boot Win98->Linux

Date: 2000, Jul 24
From: Bernd Eggink beggink

My EZ PC Card 10 doesn't come up after a reboot from Win to Linux (SuSE 6.4). It reports an error like

  .../pcmcia/i82365.o: init_module: Device or resource busy

I have to power the notebook off and on again to get it working. No problems with warm boots from Linux to Linux, however. Both Linux and Win98 use IRQ 10. Is there a way to force a "hard reset" of the card during startup?

Thanks, Bernd

1. None I think this is fixed in the 3.1.19 drivers by David Hinds, 2000, Jul 24

Question NE2000 Compatible Fast Ethernet says: get dev info on socket 1 failed: No such device

Date: 2000, Jul 30
From: Baltasar Parreira Balta

I can't identify the card, the only thing it says on the face of card is: Fast Ethernet PC Card and some picture with E=mc2, no brand, no model, etc, can anyone help me, or have the same card?

None "eth0: Tx hung, 9 vs 0." from NETGEAR FA510c

Date: 2000, Aug 02
From: David Brooks dbrooks

I recently purchased a NETGEAR 510c network adapter. The box it was in actually advertises it as "Linux Compatible", which was a delightfull sight to behold in a mainstream retail establishment!

However, I have had been unable to get it to work. The card is detected fine, although there is no entry for it in the pcmcia configuration file, this is not a problem for it shares the same manfid as both the "UMAX250 Fast Ethernet" and the "Ambicom AMB8100 Fast Ethernet", which both also use the tulip_cb driver.

The card is detected [syslog output provided below], and then delivers the inevitable error of

"eth0: Tx hung, 9 vs. 0. "

After this, an attempt to ping a machine on the network [and this is quite strange] -DOES- suceed in at least lighting the activity lights on the hub... tcpdump run on another system demonstrates that the card is sucessfully sending out ARP requests for the desired system, but it fails to be capable of doing anything else, and after about 6 arp requests, the card then fails altogether and must be ejected+inserted to even do that much again.

And advice would be greatly appreciated. Is it possible that I just have a bad card and should exchange this one? Or should I abbandon hope of cardbus 10/100 on my laptop? [I'm certainly not about to stop running linux ;) ]

For whomever may have any assistance or advice to profer, thank you in advance for your time.

--David Brooks

Aug  2 00:13:45 note kernel: Linux PCMCIA Card Services 3.1.18 
Aug  2 00:13:45 note kernel:   kernel build: 2.2.16 #2 Tue Aug 1 23:28:29 EDT 2000 
Aug  2 00:13:45 note kernel:   options:  [pci] [cardbus] [apm] 
Aug  2 00:13:45 note kernel: PCI routing table version 1.0 at 0xfd3d0 
Aug  2 00:13:45 note kernel: Intel PCIC probe:  
Aug  2 00:13:45 note kernel:   Ricoh RL5C466 rev 02 PCI-to-CardBus at slot 00:13, mem 0xf0000000 
Aug  2 00:13:45 note kernel:     host opts [0]: [io 7/15/2] [mem 3/6/2] [no pci irq] [lat 32/176] [bus 32/34] 
Aug  2 00:13:45 note kernel:     host opts [1]: [io 7/15/2] [mem 3/6/2] [no pci irq] [lat 32/176] [bus 35/37] 
Aug  2 00:13:45 note kernel:     ISA irqs (default) = 3,4,5,7,9,10,11,15 polling interval = 1000 ms 
Aug  2 00:13:46 note cardmgr[601]: starting, version is 3.1.18
Aug  2 00:13:46 note cardmgr[601]: watching 2 sockets
Aug  2 00:13:46 note kernel: cs: IO port probe 0x1000-0x17ff: clean. 
Aug  2 00:13:46 note kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x220-0x22f 0x2a0-0x2af 0x330-0x337 0x370-0x37f 0x388-0x38
f 0x398-0x39f 0x3e0-0x3e7 0x4d0-0x4d7
Aug  2 00:13:46 note kernel: cs: IO port probe 0x0a00-0x0aff: clean. 
Aug  2 00:14:01 note kernel: cs: cb_alloc(bus 32): vendor 0x1011, device 0x0019 
Aug  2 00:14:01 note cardmgr[601]: initializing socket 0
Aug  2 00:14:01 note cardmgr[601]: socket 0: UMAX Technologies UMAX250 Fast Ethernet
Aug  2 00:14:02 note cardmgr[601]: executing: 'modprobe cb_enabler'
Aug  2 00:14:02 note cardmgr[601]: executing: 'modprobe tulip_cb'
Aug  2 00:14:02 note kernel: cs: cb_config(bus 32) 
Aug  2 00:14:02 note kernel:   fn 0 bar 1: io 0x100-0x17f 
Aug  2 00:14:02 note kernel:   fn 0 bar 2: mem 0x600c0000-0x600c03ff 
Aug  2 00:14:02 note kernel:   fn 0 rom: mem 0x60080000-0x600bffff 
Aug  2 00:14:02 note kernel:   irq 3 
Aug  2 00:14:02 note kernel: tulip_attach(device 20:00.0) 
Aug  2 00:14:02 note kernel: tulip.c:v0.91g-ppc 7/16/99 becker@cesdis.gsfc.nasa.gov (modified by danilo@cs.uni-magdeburg.de for XI
RCOM CBE, fixed by Doug Ledford)
Aug  2 00:14:02 note kernel: eth0: Digital DS21143 Tulip rev 65 at 0x100, 00:10:7A:15:CB:26, IRQ 3. 
Aug  2 00:14:02 note kernel: eth0:  EEPROM default media type Autosense. 
Aug  2 00:14:02 note kernel: eth0:  Index #0 - Media MII (#11) described by a 21142 MII PHY (3) block. 
Aug  2 00:14:02 note kernel: eth0:  MII transceiver #1 config 1000 status 7849 advertising 01e1. 
Aug  2 00:14:02 note cardmgr[601]: executing: './network start eth0'
Aug  2 00:14:24 note kernel: eth0: Setting full-duplex based on MII#1 link partner capability of 45e1. 
Aug  2 00:14:24 note kernel: eth0: Tx hung, 9 vs. 0. 

1. None Interrupt setup problem by David Hinds, 2000, Aug 02
1. Ok Problem solved; card working! by David Brooks, 2000, Aug 03
2. Question Another Netgear FA510c problem by Ed Hill, 2000, Aug 08

Question DLink DFE 660 - Bad performance

Date: 2000, Aug 02
From: hvx

Hi!

Im using a DLink DFE-660 card with is an 100MBit card, but I get only about 200k/s. I tried 2 cards, a new and an old one, neither one of them was faster than that. Apart from that, the card seems to work just fine.

Here is some output from differnt commands, maybe somebody can give me a little advise how to get more speed out of this card. Thank you, M.Melms

Im using linux 2.2.16 with cardmgr 3.1.19.

bobby:~ # ifconfig eth0
eth0      Link encap:Ethernet  HWaddr 00:50:BA:74:55:50  
          inet addr:192.168.100.80  Bcast:192.168.100.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:1362 errors:0 dropped:0 overruns:0 frame:0
          TX packets:2281 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:100 
          Interrupt:10 Base address:0x200 

When I put in the card, /var/log/messages sais:

Aug  2 21:53:35 bobby kernel: cs: cb_alloc(bus 35): vendor 0x1011, device 0x0019
Aug  2 21:53:35 bobby cardmgr[78]: initializing socket 1
Aug  2 21:53:35 bobby cardmgr[78]: socket 1: Linksys EtherFast 10/100
Aug  2 21:53:36 bobby cardmgr[78]: executing: 'modprobe cb_enabler'
Aug  2 21:53:36 bobby cardmgr[78]: executing: 'modprobe tulip_cb'
Aug  2 21:53:36 bobby kernel: cs: cb_config(bus 35)
Aug  2 21:53:36 bobby kernel:   fn 0 bar 1: io 0x200-0x27f
Aug  2 21:53:36 bobby kernel:   fn 0 bar 2: mem 0x600c0000-0x600c03ff
Aug  2 21:53:36 bobby kernel:   fn 0 rom: mem 0x60080000-0x600bffff
Aug  2 21:53:36 bobby kernel:   irq 10
Aug  2 21:53:36 bobby kernel: cs: cb_enable(bus 35)
Aug  2 21:53:36 bobby kernel:   bridge io map 0 (flags 0x21): 0x200-0x27f
Aug  2 21:53:36 bobby kernel:   bridge mem map 0 (flags 0x1): 0x60080000-0x600c0
fff
Aug  2 21:53:36 bobby kernel: tulip_attach(device 23:00.0)
Aug  2 21:53:36 bobby kernel: tulip.c:v0.91g-ppc 7/16/99 becker@cesdis.gsfc.nasa
.gov (modified by danilo@cs.uni-magdeburg.de for XIRCOM CBE, fixed by Doug Ledfo rd) Aug 2 21:53:36 bobby kernel: eth0: Digital DS21143 Tulip rev 65 at 0x200, 00:50 :BA:74:55:50, IRQ 10.
Aug  2 21:53:36 bobby kernel: eth0:  EEPROM default media type Autosense.
Aug  2 21:53:36 bobby kernel: eth0:  Index #0 - Media MII (#11) described by a 2
1142 MII PHY (3) block. Aug 2 21:53:36 bobby kernel: eth0: MII transceiver #0 config 3000 status 7809 advertising 01e1. Aug 2 21:53:36 bobby cardmgr[78]: executing: './network start eth0'

1. None First guess: problem negotiating correct media type by David Hinds, 2000, Aug 02
(_ Feedback thanks, but it didnt work by hvx@cryptnet.de, 2000, Aug 02
(_ None Re: Feedback: thanks, but it didnt work by David Hinds, 2000, Aug 02
(_ Feedback Re: Feedback: thanks, but it didnt work by hvx@cryptnet.de, 2000, Aug 02

Angry pcnet_cs, card works. SLOW!

Date: 2000, Aug 04
From: <jguerin@home.com>

I managed to get my Hawking 10/100 'Fast' ethernet card to work today. The only problem is it's FAR from being fast. Transfer rates on a local LAN usually go about 20-30K/s. SLOW! I saw some people had the same problem before and there was some ugly Windows kludge to get it to work. Unfortunately, I don't have Windows so that's going to be a little harder to fix. I'm hoping maybe somewhere there's a variable or option I can set to fix this problem (sync v. async?). Does anyone know of any such thing? I think I'm using the pcnet_cs module. Everything is stock out of the 3.1.19 (or was it 18) PCMCIA source. HELP please ^_^ muchas gracias. I'd prefer email CCs too, if possible.

Jon

Ok Toshiba Satellite Pro 4200 series, Micronet SP162A ethernet card, Redhat 6.2 - Some thoughts

Date: 2000, Aug 04
From: David Martin dcmartin

This is just a note about my experiences getting things up and going on some brand new (delivered on Monday) hardware. I thought I would post it to "get it on the record"- someone else might mind it useful (or not).

Firstly about the Satellite-Pro 4220 XCDT. Mine has the ToPIC95 (rev 07) bridge chipset, and despite reports of others having problems with Toshiba chipsets, I have had not problems with it (thankfully). Just to add to the confusion I have heard rumours that there are actually several versions of this model - mine is "Made in Europe" with a ToPIC95 and Trident Cyber video adaptor, there is also a "Made in Japan" variant with a ToPIC97 and Savage S3 video adaptor, so go figure. The only thing I had to do was exclude IRQ 9, as I had problems with SanDisk compactflash cards on that IRQ (Microsoft's SanDisk driver also hiccupped on IRQ 9 under windows 98) - something is obviously lurking on the IRQ line. Other than that, plain sailing.

Onto Redhat 6.2. This was my first installation of Redhat, my first installation with pcmcia hardware and my first x86 installation (only ever worked with PPC and Alpha ports in the past). It ships with pcmcia version 3.1.8, but the package is incomplete - it does not contain ide_info, as I discovered when using SanDisk compactflash cards. Also the pcmcia source is included as part of the kernel source rpm, so if you upgarde or replace it (as I did, read on) it technically breaks the rpm. Also, I have to say that I am not too impressed by Redhat's modified pcmcia network startup arrangement, either. I scrapped it in favour of the supplied pcmcia one.

Lastly the card. The Micronet SP162A PCMCIA Fast Ethernet Adaptor (to give it its complete name) was not an adaptor we had seen before and no information was available about what chipset it used. The accompanying driver disk included a README for linux that implied that if you installed the pcmcia system it would work...and it did, kind of. With pcmcia version 3.1.8 cardmgr autodetected it as a "NE2000 compatible" (from the manufacturer ID string, I think) and picked up the pcnet_cs diver. Output from cardctl showed

Socket 1:
  product info: "Dual Speed", "10/100 PC Card", "1.0", " "
  manfid: 0x0149, 0xc1ab
  function: 6 (network)

Configuration worked ok, but the driver did not appear to work correctly - it produced a 'eth0: lost link beat' (from the driver watchdog reset ?) message immediately after carrier was detected on the Lan at 10Mbps. The interface still worked, you could shift packets using known IP addresses, but TX behaviour seemed a bit odd and arp and dns were hung-up somehow. On the assumption that the card/driver marriage was not a happy one, I upgraded to 3.1.15 - it appeared tp contain the most recent version of the pcnet_cs driver. Compilation and installation was a doddle, and the card worked straight away, so I am guessing that it is based on one of the newer D-Link chipsets that have been recently added to the pcnet_cs driver.

So, overall, it was relatively painless exercise. I have to dip my hat to David Hinds and the other developers, the pcmcia system is a neat piece of work that works way better than I expected it would. Cheers, gents (and ladies)!!!



David Martin.

Warning FYI Linksys PCMPC200 V2 card; different from older versions

Date: 2000, Aug 06
From: Nik Conwell nikc

FYI - Be aware that the Linksys EtherFast 10/100 CardBus PCMPC200 has changed and is now V2 which apparently isn't supported yet. The box has blue edges and says something like new V2 in the upper left corner.

Details:

I'm runing RH 6.2, kernel-pcmcia-cs-2.2.16-3.

kernel: Linux PCMCIA Card Services 3.1.8
kernel:   kernel build: 2.2.16-3 #1 Mon Jun 19 18:10:14 EDT 2000
kernel: Intel PCIC probe:
kernel:   Toshiba ToPIC97 PCI-to-CardBus at bus 0 slot 19, mem 0x68000000, 2 sockets
kernel:     host opts [0]: [slot 0xf0] [ccr 0x10] [cdr 0x86] [rcr 0x02] [no pci irq] [l at 168/176] [bus 20/20]
kernel:     host opts [1]: [slot 0xf0] [ccr 0x20] [cdr 0x86] [rcr 0x02] [no pci irq] [l at 168/176] [bus 21/21]
kernel:     ISA irqs (default) = 3,4,5,7,9,10,12 polling interval = 1000 ms
pcmcia: cardmgr. cardmgr[488]: starting, version is 3.1.8 rc: Starting pcmcia succeeded cardmgr[488]: watching 2 sockets
kernel: cs: IO port probe 0x1000-0x17ff: clean.
kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x220-0x22f 0x330-0x337 0x378-0x37f 0x388-0x38f 0x480-0x48f 0x4d0-0x4d7
kernel: cs: IO port probe 0x0a00-0x0aff: clean.
inet: inetd startup succeeded kernel: cs: cb_alloc(bus 20): vendor 0x13d1, device 0xab02 cardmgr[488]: initializing socket 0 cardmgr[488]: unsupported card in socket 0 cardmgr[488]: product info: "CardBus", "Fast Ethernet", "V1.0", "" cardmgr[488]: manfid: 0x13d1, 0xab02 function: 6 (network)

The manfid is different from the pcmcia config for the EtherFast 10/100. Being a tricky guy, I tried adding an entry for the new manfid, using the tulip_cb driver, and was rewarded with:

kernel: cs: cb_config(bus 20)
kernel:   fn 0 bar 1: io 0xa00-0xaff
kernel:   fn 0 bar 2: mem 0x60060000-0x600603ff
kernel:   fn 0 rom: mem 0x60040000-0x6005ffff
kernel: tulip_attach(bus 20, function 0)
kernel: tulip.c:v0.91 4/14/99 becker@cesdis.gsfc.nasa.gov (modified by danilo@cs.uni-magdeburg.de for XIRCOM CBE, fixed by Doug Ledford)
kernel: eth0: Digital DS21143 Tulip rev 17 at 0xa00, 10:80:00:80:00:80, IRQ 3.
kernel: eth0:  EEPROM default media type 10baseT.
kernel: eth0:  MII transceiver #1 config 3000 status 7849 advertising 01e1.
cardmgr[1210]: executing: './network start eth0' pumpd[1242]: starting at (uptime 0 days, 0:18:06) Sat Aug 5 17:59:54 2000
kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000000
kernel: current->tss.cr3 = 02721000, %cr3 = 02721000
kernel: *pde = 00000000
kernel: Oops: 0000
kernel: CPU:    0
kernel: EIP:    0010:[lockd:__insmod_lockd_S.bss_L444+161553/21737519]
kernel: EFLAGS: 00010246
kernel: eax: fffffff8   ebx: ffffffff   ecx: 00000000   edx: c3ffab58
kernel: esi: 00000000   edi: c593ada0   ebp: c2723e4c   esp: c2723e00
kernel: ds: 0018   es: 0018   ss: 0018
kernel: Process pump (pid: 1242, process nr: 48, stackpage=c2723000)
kernel: Stack: c593ada0 00000008 00000286 c26e1000 00000246 00000003 00000008 00000000
kernel:        00000000 00000282 c593ada0 c3ffab58 00000a00 c3ffab50 c0383800 00000a00
kernel:        c593ada0 c01ded87 00000214 c2723ea0 c6857c04 c593ada0 00000001 c593ada0
kernel: Call Trace: [__delay+19/40] [lockd:__insmod_lockd_S.bss_L444+160452/21738620] [lockd:__insmod_lockd_S.bss_L444+161399/21737673] [dev_open+32/128] [dev_change_flags+84/264] [devinet_ioctl+635/1536] [sock_ioctl+0/32]
kernel:        [sk_alloc+17/60] [inet_ioctl+402/524] [sock_ioctl+26/32] [sys_ioctl+414/452] [system_call+52/56]
kernel: Code: 8a 06 8b 55 08 88 42 3c 83 7d 0c 00 74 15 8b 45 e8 0f b6 48
cardmgr[1210]: + Determining IP information for eth0... done.

The card wasn't accessible after that...

I took the card back to CompUSA and exchanged it for a not V2 PCMPC200 (which they had tried to unload on me first, but I was too smart for them...). The older PCMPC200 works just fine in my system.

I've sent a copy to the Linksys tech support as well(support@linksys.com)

Keep up the good work. I'm eternally grateful. -nik

nik@bu.edu

None 3COM 3CXFE575BT sending with no flow control?

Date: 2000, Aug 07
From: Roberto Jung Drebes <drebes@aton.inf.ufrgs.br>

Hi there,

My 3CXFE575BT is still crazy when it has to send data in a burst, giving me the "Tx Ring full, refusing to send buffer" error. It works ok when I'm receiving HTTP, FTP traffic, and it's even ok when I'm sending or receiving ssh traffic, but I just need to start a bursty data transfer leaving the machine to get the error. I'm now using pcmcia-cs-3.1.19, kernel 2.2.16, have tried using all combinations of pci_int and pci_csc in PCIC_OPTs, with no success. The brige is: 00:11.0 CardBus bridge: Texas Instruments PCI1220 (rev 02)

Well, first, is there any information I can get from the error message? Like, why the buffer is getting full? I know nothing about TCP, but why isn't the flow control working (I belive there is something like a sliding window in TCP)? Or a debug mode (The i82365 man page says something of pc_debug parameter, but only when the module is compiled with debug info, how do I do it?)

<update>I tried increasing the size of the Tx ring to 32, and the problem still exists, only it takes more time to print the message. I believe that perhaps the card is not receiving some packets, which would contain the acks, and so it cannot advance the ring and queue them again. I will test if I can see all the incoming packets with tcpdump, but it seems difficult. I'm also using debug level 7 now.</update>

If there is no easy fix, and I think it really doesn't have since I'm having this problem for a long time, I would like to have some information on using PCMCIA with 2.3/4 kernels. I remember seeing that some PCMCIA support was included in the kernel in such versions, so does this make the pcmcia-cs useless? Or do they work together? If so, should I do the same, ie, unpack it, ./Configure && make all?

Thanks in advance,

1. None It is not a flow control issue by David Hinds, 2000, Aug 08
(_ None Re: It is not a flow control issue by Roberto Jung Drebes, 2000, Aug 08

Question 3c574 prob, no h/w address

Date: 2000, Aug 08
From: Joerg Schmitz-Linneweber jsl123

I'm trying to use a (no name/oem brand) "Allnet ALL0142+" Fast Ethernet adapter. I tried several modules and decided that the 3c574_cs worked best :-) Indeed it seems to find some config info in the card but no hardware address! I even treied to set it via ifconfig (like someone on the list did) but with no result. To give somemore info: I tried different io ranges and irq's (io > 0x300, io > 0x400, irq=5,9,10). Appended is a log with some kernel info (and debug info from 3c574_cs, this is the reason why cardmgr claims its not the right version). Beside of the pcmcia tools, i used the latest (low-level) drivers. Perhaps I'm only missing a very small thing? BTW: I had to arrange /etc/pcmcia/config and config.opts of course to let cardmgr reco. the card. Did any one know which manufac. uses these id's: 0xffff, 0x021b? BTW2: The last error messages in the log are the results from trying to ping someone on the net. I know it's hard to use a network adapter with a MAC of all zeros but why not give it a try? ;-)

Thanks in advance,
  Joerg

--- schnipp --------------------------------------------

Aug  8 12:28:26 passport kernel: Linux PCMCIA Card Services 3.1.19
Aug  8 12:28:26 passport kernel:   kernel build: 2.2.14 #1 Tue May 9 00:33:32 GMT 2000
Aug  8 12:28:26 passport kernel:   options:  [pci] [cardbus]
Aug  8 12:28:27 passport rcpcmcia: /sbin/insmod /lib/modules/2.2.14/pcmcia/i82365.o
Aug  8 12:28:27 passport kernel: Intel PCIC probe: 
Aug  8 12:28:27 passport kernel:   TI 1131 rev 01 PCI-to-CardBus at slot 00:04, mem 0x68000000
Aug  8 12:28:27 passport kernel:     host opts [0]: [isa irq] [no pci irq] [lat 32/176] [bus 1/1]
Aug  8 12:28:27 passport kernel:     host opts [1]: [isa irq] [no pci irq] [lat 32/176] [bus 2/2]
Aug  8 12:28:27 passport kernel:     ISA irqs (scanned) = 5,9,10,11 status change on irq 11
Aug  8 12:28:27 passport rcpcmcia: /sbin/insmod /lib/modules/2.2.14/pcmcia/ds.o
Aug  8 12:28:27 passport cardmgr[1547]: starting, version is 3.1.16
Aug  8 12:28:27 passport cardmgr[1547]: watching 2 sockets
Aug  8 12:28:27 passport cardmgr[1547]: Card Services release does not match
Aug  8 12:28:27 passport kernel: cs: IO port probe 0x0c00-0x0cff: excluding 0xcf8-0xcff
Aug  8 12:28:27 passport kernel: cs: IO port probe 0x0800-0x08ff: clean.
Aug  8 12:28:27 passport kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x178-0x17f 0x378-0x37f 0x408-0x40f 0x480-0x48f 0x4d0-0x4d7
Aug  8 12:28:27 passport kernel: cs: IO port probe 0x0a00-0x0aff: clean.
Aug  8 12:28:27 passport cardmgr[1547]: initializing socket 0
Aug  8 12:28:27 passport kernel: cs: memory probe 0xa0000000-0xa0ffffff: clean.
Aug  8 12:28:27 passport cardmgr[1547]: socket 0: Allnet ALL0142+
Aug  8 12:28:27 passport cardmgr[1547]: executing: 'modprobe 3c574_cs pc_debug=255'
Aug  8 12:28:28 passport kernel: 3c574_cs.c v1.08 9/24/98 Donald Becker/David Hinds, becker@cesdis.gsfc.nasa.gov.
Aug  8 12:28:28 passport kernel: 3c574_cs.c 1.000 1998/1/8 Donald Becker, becker@cesdis.gsfc.nasa.gov.
Aug  8 12:28:28 passport kernel: 
Aug  8 12:28:28 passport kernel: 3c574_attach()
Aug  8 12:28:28 passport kernel: 3c574_event(0x000004)
Aug  8 12:28:28 passport kernel: 3c574_config(0xc23d1000)
Aug  8 12:28:28 passport kernel: eth0: Fast-Ethernet at io 0x300, irq 5, hw_addr 00:00:00:00:00:00.
Aug  8 12:28:28 passport kernel:   ASIC rev 0, 16K FIFO split 5:3 Rx:Tx, MII interface.
Aug  8 12:28:28 passport kernel:   MII transceiver at index 1, status 0.
Aug  8 12:28:28 passport kernel:   MII transceiver at index 2, status 0.
Aug  8 12:28:28 passport cardmgr[1547]: executing: './network start eth0'
Aug  8 12:28:28 passport kernel: eth0: opened, status 0000.
Aug  8 12:28:28 passport kernel: eth0: updating the statistics.
Aug  8 12:28:41 passport kernel: eth0: el3_start_xmit(length = 42) called, status 0000.
Aug  8 12:28:41 passport kernel: eth0: transmit error: status 0xff
Aug  8 12:28:41 passport last message repeated 31 times
Aug  8 12:28:48 passport kernel: eth0: Transmit timed out!
Aug  8 12:28:48 passport kernel:   irq status 0000, rx status ffff, tx status ff, tx free 0000
Aug  8 12:28:48 passport kernel:   diagnostics: fifo 0f00 net 4005 ethernet 0000 media ffff
Aug  8 12:28:48 passport kernel: eth0: el3_start_xmit(length = 42) called, status 0000.
Aug  8 12:28:48 passport kernel: eth0: transmit error: status 0xff
--- schnapp ------------------------------------------------

1. None This is the wrong driver by David Hinds, 2000, Aug 08

Question Psion GoldCard Cardbus Network adapter problem with Mandrake 7.1

Date: 2000, Aug 10
From: Jan Calmered <jan.calmered@sbs.siemens.se>

I have a new installation of Mandrake 7.1 (RedHat mostly) 
on my mobil, Siemens 510AGP and trying to get a PCMCIA 
Network adapter going (Psion GoldCard 56k + Ethernet, Cardbus)

I am using the 2.2.15 kernel and the 3.1.19 PCMCIA package.

Two problems:

1. After cold boot the system loads the core PCMCIA OK but
not the card driver. checking the card with cardctl status
 show the card to be "busy", ejecting and inserting makes cardctl ident show:
Socket 0:
  product info: "Psion", "10-100Mb MF CardBus", "400062-01C", "123456789abc"
  manfid: 0x016c, 0x0021
  function: 6 (network)
Socket 1:
  no product info available

2. But still no driver loaded as fas as I can see

I have added the card as a "10-100Mb MF Cardbus" in the
config file to bind it to pcnet_cs but no change

There seems to be a failure of IRQ 9 in the log file which 
I have tried to exclude in the config.opts but no change

Enclose is the dmesg, lsmod  output

any clues?

Regards

Jan

dmesg =
Linux version 2.2.15-4mdk (chmou@kenobi.mandrakesoft.com) (gcc version 2.95.3 19991030 (prerelease)) #1 Wed May 10 15:31:30 CEST 2000
Detected 267277459 Hz processor.
Console: colour VGA+ 80x25
Calibrating delay loop... 532.48 BogoMIPS
Memory: 192660k/196608k available (1208k kernel code, 416k reserved, 2252k data, 72k init, 0k bigmem)
Dentry hash table entries: 32768 (order 6, 256k)
Buffer cache hash table entries: 262144 (order 8, 1024k)
Page cache hash table entries: 65536 (order 6, 256k)
VFS: Diskquotas version dquot_6.4.0 initialized
CPU: Intel Pentium II (Deschutes) stepping 02
Checking 386/387 coupling... OK, FPU using exception 16 error reporting.
Checking 'hlt' instruction... OK.
POSIX conformance testing by UNIFIX
mtrr: v1.35a (19990819) Richard Gooch (rgooch@atnf.csiro.au)
PCI: PCI BIOS revision 2.10 entry at 0xf5200
PCI: Using configuration type 1
PCI: Probing PCI hardware
PCI: Enabling I/O for device 00:3a
Linux NET4.0 for Linux 2.2
Based upon Swansea University Computer Society NET3.039
NET4: Unix domain sockets 1.0 for Linux NET4.0.
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP, IGMP
TCP: Hash tables configured (ehash 262144 bhash 65536)
Initializing RT netlink socket
Starting kswapd v 1.5 
Detected PS/2 Mouse Port.
Serial driver version 4.27 with MANY_PORTS MULTIPORT SHARE_IRQ enabled
ttyS00 at 0x03f8 (irq = 4) is a 16550A
pty: 256 Unix98 ptys configured
apm: BIOS version 1.2 Flags 0x03 (Driver version 1.12)
Real Time Clock Driver v1.09
RAM disk driver initialized:  16 RAM disks of 4096K size
Uniform Multi-Platform E-IDE driver Revision: 6.30
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
PIIX4: IDE controller on PCI bus 00 dev 39
PIIX4: not 100% native mode: will probe irqs later
    ide0: BM-DMA at 0xfcd0-0xfcd7, BIOS settings: hda:pio, hdb:pio
    ide1: BM-DMA at 0xfcd8-0xfcdf, BIOS settings: hdc:pio, hdd:pio
hda: TOSHIBA MK1214GAP, ATA DISK drive
hdc: TOSHIBA CD-ROM XM-1602B, ATAPI CDROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
hda: TOSHIBA MK1214GAP, 11513MB w/0kB Cache, CHS=1467/255/63
hdc: ATAPI 20X CD-ROM drive, 128kB Cache
Uniform CDROM driver Revision: 2.56
Floppy drive(s): fd0 is 1.44M
FDC 0 is a National Semiconductor PC87306
md driver 0.90.0 MAX_MD_DEVS=256, MAX_REAL=12
raid5: measuring checksumming speed
raid5: MMX detected, trying high-speed MMX checksum routines
   pII_mmx   :   589.407 MB/sec
   p5_mmx    :   615.696 MB/sec
   8regs     :   477.774 MB/sec
   32regs    :   251.841 MB/sec
using fastest function: p5_mmx (615.696 MB/sec)
scsi : 0 hosts.
scsi : detected total.
md.c: sizeof(mdp_super_t) = 4096
Partition check:
 hda: hda1 hda2 < hda5 hda6 hda7 hda8 hda9 hda10 >
autodetecting RAID arrays
autorun ...
... autorun DONE.
VFS: Mounted root (ext2 filesystem) readonly.
Freeing unused kernel memory: 72k freed
Adding Swap: 136512k swap-space (priority -1)
usb.c: registered new driver hub
usb.c: registered new driver usb-storage
USB Mass Storage support registered.
usb.c: registered new driver keyboard
cs89x0.c: Module autoprobing not allowed.
cs89x0.c: Append io=0xNNN
Linux PCMCIA Card Services 3.1.19
  kernel build: 2.2.15-4mdk #1 Wed May 10 15:31:30 CEST 2000
  options:  [pci] [cardbus] [apm]
PCI routing table version 1.0 at 0xf57d0
  00:0a.0 -> irq 9
  00:0a.1 -> irq 9
Intel PCIC probe: 
  O2Micro OZ6832/33 rev 02 PCI-to-CardBus at slot 00:0a, mem 0x68000000
    host opts [0]: [pci irq 9] [lat 168/176] [bus 32/34]
    host opts [1]: [pci irq 9] [lat 168/176] [bus 35/37]
    PCI irq 9 test failed
    ISA irqs (default) = 3,4,5,7,11,12 polling interval = 1000 ms
cs: IO port probe 0x0c00-0x0cff: excluding 0xcf8-0xcff
cs: IO port probe 0x0800-0x08ff: excluding 0x800-0x807
cs: IO port probe 0x0100-0x04ff: excluding 0x200-0x207 0x220-0x22f 0x330-0x337 0x378-0x37f 0x388-0x38f 0x398-0x39f 0x4d0-0x4d7
cs: IO port probe 0x0a00-0x0aff: clean.
IrDA (tm) Protocols for Linux-2.2 (Dag Brattli)
cs: socket 0 timed out during reset
usb-uhci.c: $Revision: 1.228 $ time 15:46:55 May 10 2000
usb-uhci.c: High bandwidth mode enabled
usb-uhci.c: Intel USB controller: setting latency timer to 0
usb-uhci.c: USB UHCI at I/O 0xfce0, IRQ 10
usb-uhci.c: Detected 2 ports
usb.c: new USB bus registered, assigned bus number 1
usb.c: USB new device connect, assigned device number 1
hub.c: USB hub found
hub.c: 2 ports detected
cs: socket 0 timed out during reset
unloading PCMCIA Card Services
Linux PCMCIA Card Services 3.1.19
  kernel build: 2.2.15-4mdk #1 Wed May 10 15:31:30 CEST 2000
  options:  [pci] [cardbus] [apm]
PCI routing table version 1.0 at 0xf57d0
Intel PCIC probe: 
  O2Micro OZ6832/33 rev 02 PCI-to-CardBus at slot 00:0a, mem 0x68000000
    host opts [0]: [pci irq 9] [lat 168/176] [bus 32/34]
    host opts [1]: [pci irq 9] [lat 168/176] [bus 35/37]
    PCI irq 9 test failed
    ISA irqs (default) = 3,4,5,7,11 polling interval = 1000 ms
cs: IO port probe 0x0c00-0x0cff: excluding 0xcf8-0xcff
cs: IO port probe 0x0800-0x08ff: excluding 0x800-0x807
cs: IO port probe 0x0100-0x04ff: excluding 0x200-0x207 0x220-0x22f 0x330-0x337 0x378-0x37f 0x388-0x38f 0x398-0x39f 0x4d0-0x4d7
cs: IO port probe 0x0a00-0x0aff: clean.
cs: cb_alloc(bus 32): vendor 0x10b8, device 0x0006
ROM image dump:
  image 0: 0x000000-0x000bff, signature PCIR
epic100.c:v1.07h 8/18/99 Donald Becker http://cesdis.gsfc.nasa.gov/linux/drivers/epic100.html
cs: cb_config(bus 32)
cs: could not allocate 512 IO ports for CardBus socket 0
cs: cb_release(bus 32)
epic_cb: RequestIO: Out of resource
cs89x0.c: Module autoprobing not allowed.
cs89x0.c: Append io=0xNNN

lsmod =

Module                  Size  Used by
serial_cb               1148   0  (unused)
epic_cb                10440   0  (unused)
cb_enabler              2536   2  [serial_cb epic_cb]
ds                      6440   2  [cb_enabler]
i82365                 22948   2 
pcmcia_core            45856   0  [cb_enabler ds i82365]
soundcore               3748   0  (autoclean) (unused)
usb-uhci               26660   0  (unused)
autofs                  9604   1  (autoclean)
irda                   84289   1 
nls_iso8859-1           2276   4  (autoclean)
nls_cp437               3784   4  (autoclean)
vfat                   11164   4  (autoclean)
fat                    32864   4  (autoclean) [vfat]
keybdev                 1832   0  (unused)
usbkbd                  2292   0  (unused)
input                   2880   0  [keybdev usbkbd]
usb-storage            10120   0  (unused)
usbcore                27236   0  [usb-uhci usbkbd usb-storage]
supermount             15112   2  (autoclean)
1. None IO port allocation problem by David Hinds, 2000, Aug 10
(_ Feedback Thanks, allocating more IO helped by Jan Calmered, 2000, Aug 14
(_ Question To quick to be happy, seems to have some IRQ problem still by Jan Calmered, 2000, Aug 14
(_ None Try some other irq_mode values by David Hinds, 2000, Aug 14
(_ None Tried irq_mode=3 and got it to stop complaining about IRQ9 but still not OK by Jan Calmered, 2000, Aug 16
(_ None Several issues by David Hinds, 2000, Aug 16
(_ Ok Ethernet working using 3.1.20 release, still some minor problems by Jan Calmered, 2000, Aug 17

Question Can use card only once

Date: 2000, Aug 13
From: Maik Hassel Maik

Any of my pcmcia network-cards are correctly detected when iserted at startup. as soon as I eject them and re-insert any card, they are identified as "unknown-memory" cards. Could anybody help me?

1. None Not enough information to tell anything by David Hinds, 2000, Aug 16
1. None Here it is.... by Maik Hassel, 2000, Aug 16
(_ None Still not enough information by David Hinds, 2000, Aug 16
(_ None Here is all I have... by Maik Hassel, 2000, Aug 22
(_ None These drivers are too old for submitting a bug report by David Hinds, 2000, Aug 22
2. None Done - no difference by Maik Hassel, 2000, Sep 05

Question IBM EtherJet combo only works after bouncing eth0

Date: 2000, Aug 16
From: Peter A. Friend pafriend

A while back I posted a message saying I had gotten the IBM EtherJet Modem/Ethernet card working by adding the ident string to the /etc/pcmcia/config file. This has been working fine, but since I move around a lot, I have been using scripts after boot-up to up the interface and add the default route.

I recently started using the scheme support, which works great, except that I have to bounce eth0 to get things working. After booting, the ifconfig output looks normal, as does the route table. I can't ping anything. There are 0 errors in /var/log/messages, and the test_network in debug-tools reports nothing abnormal, except that the default gateway cannot be pinged. Again, after a 'ifconfig eth0 down up' everything works flawlessly.

I'm stumped, any ideas on what I should be looking at?

IBM ThinkPad 600X Linux Mandrake 7.0, kernel 2.2.16, pcmcia 3.1.17.

TIA,

Peter Friend

Question DFE-650 Cannot read hardware net address

Date: 2000, Aug 22
From: Ben Marsh blm

I have read all the material I can and fiddled as much as I can and yet I still can't get my DFE-650 pc card working on my laptop. It was identified as Anonymous then I fiddled with the config.opts and get the identity recognised and the correct module loading but I get

pcnet_cs: unable to read net address

I have PC Card Services v3.1.8 Kernel 2.2.14-12 Redhat 6.2 Intel PCIC i82365sl B step ISA-to-PCMCIA at port 0x3e0 ofs 0x00, 2 sockets

Anything else helpful?

1. None Your workaround didn't work by David Hinds, 2000, Aug 22

Question intel pro 10/100/56k combo card

Date: 2000, Aug 22
From: Tim Newsham <newsham@lava.net>

Hi,

   I got a laptop recently with an intel pro 10/100 ethernet
plus 56k modem combo card in it.  This card was not detected
when I installed the redhat-6.2 distribution.  After 
installation I checked and got the following information
about the card:

    product info: "INTEL(R)", "PRO/100 LAN+MODEM56 CARDBUS
                   II", "MBLA3356", "1.00"
    manfid: 0x0089, 0x1103
    function: 6 (network)

This card was not listed in /etc/pcmcia/config, so I added
an entry mimicking the entry for the other manfid for
intel pro/100 lan + modem card.  It tried to load the
xirc* driver, which reported that my card was "an old CE"
card.  I don't know what the old CE card was, or what kind
of driver it woudl take, so I stopped there.

Can anyone suggest something to try at this point?  Is there
a driver that will work with this card?  If so, what kind
of entry do I want to add to the config file?  Has this
already been done in the distribution?  If not, how do I
submit a change once this is working?

Thanks in advance.

                                         Tim N.
1. None You can't fix this by editing the config file by David Hinds, 2000, Aug 22
(_ None Re: You can't fix this by editing the config file by Tim Newsham, 2000, Aug 23

Question Misidentifying 'D-Link DFE-650' as 'KTI Ether-C16'

Date: 2000, Aug 22
From: William Crandall WilliamCrandall

Greetings,

System:	Acer Extensa 710TE 
PCcard:	D-Link DFE-650
[Setup:	Works without a hitch under Win98]

Kernel:	2.2.14 #1 Mon Mar 13 10:54:26 GMT 2000 (Suse 6.4)
Driver:	cardctl version 3.1.19

No changes to /etc/pcmcia or other startup scripts.


Startup (/var/log/messages):

Aug 22 11:10:32 wcra kernel: Linux PCMCIA Card Services 3.1.19
Aug 22 11:10:32 wcra kernel:   kernel build: 2.2.14 #1 Mon Mar 13 10:54:26 GMT 2000
Aug 22 11:10:32 wcra kernel:   options:  [pci] [cardbus] [apm] [pnp]
Aug 22 11:10:32 wcra kernel: PCI routing table version 1.0 at 0xfdf60
Aug 22 11:10:32 wcra kernel:   00:06.0 -> irq 11
Aug 22 11:10:32 wcra kernel:   00:06.1 -> irq 11
Aug 22 11:10:32 wcra kernel: PnP: PNP BIOS installation structure at 0xc00f77b0
Aug 22 11:10:32 wcra kernel: PnP: PNP BIOS version 1.0, entry at f0000:a515, dseg at 400
Aug 22 11:10:32 wcra kernel: Intel PCIC probe: 
Aug 22 11:10:32 wcra kernel:   TI 1250A rev 02 PCI-to-CardBus at slot 00:06, mem 0x68000000
Aug 22 11:10:32 wcra kernel:     host opts [0]: [pci + serial irq] [pci irq 11] [lat 168/176] [bus 32/34]
Aug 22 11:10:32 wcra kernel:     host opts [1]: [pci + serial irq] [pci irq 11] [lat 168/176] [bus 35/37]
Aug 22 11:10:32 wcra kernel:     ISA irqs (scanned) = 9,10 PCI status changes
Aug 22 11:10:32 wcra kernel: cs: IO port probe 0x0c00-0x0cff: clean.
Aug 22 11:10:32 wcra kernel: cs: IO port probe 0x0800-0x08ff: clean.
Aug 22 11:10:32 wcra kernel: cs: IO port probe 0x0100-0x04ff: clean.
Aug 22 11:10:32 wcra kernel: cs: IO port probe 0x0a00-0x0aff: clean.
Aug 22 11:10:32 wcra kernel: cs: memory probe 0xa0000000-0xa0ffffff: clean.
Aug 22 11:10:32 wcra kernel: eth0: NE2000 (DL10022 rev 30): io 0x300, irq 9, hw_addr 00:E0:98:77:54:FD


Operating info:

bash-2.03# lsmod
Module                  Size  Used by
usb-uhci               17000   0  (unused)
pcnet_cs               10636   0  (unused)
8390                    6228   0  [pcnet_cs]
ds                      6632   2  [pcnet_cs]
i82365                 23392   2 
pcmcia_core            52672   0  [pcnet_cs ds i82365]
serial                 42612   1  (autoclean)
memstat                 1476   0  (unused)
usbcore                42344   1  [usb-uhci]

bash-2.03# cardctl config
Socket 0:
  not configured
Socket 1:
  Vcc 5.0V  Vpp1 0.0V  Vpp2 0.0V
  interface type is "memory and I/O"
  irq 9 [exclusive] [level]
  function 0:
    config base 0x0400
      option 0x60 status 0x00 copy 0x00
    io 0x0300-0x031f [auto]

bash-2.03# cardctl ident
Socket 0:
  no product info available
Socket 1:
  product info: "D-Link", "DFE-650", "Fast Ethernet", "Rev. D1"
  manfid: 0x0149, 0x0230
  function: 6 (network)


This is the *correct* identity.
But no access to the network (ping <server>).

Cardinfo gives a warning; then all fields are blank:

bash-2.03# cardinfo
Card Services release does not match!       

Resetting the card controller:

bash-2.03# cardctl suspend
bash-2.03# cardctl reset

	... generates these /var/log/messages:

Aug 22 11:37:16 wcra cardmgr[80]: executing: './network suspend eth0'
Aug 22 11:37:32 wcra cardmgr[80]: shutting down socket 1
Aug 22 11:37:32 wcra cardmgr[80]: executing: './network stop eth0'
Aug 22 11:37:32 wcra cardmgr[80]: executing: 'modprobe -r pcnet_cs'
Aug 22 11:37:32 wcra cardmgr[80]: executing: 'modprobe -r 8390'
Aug 22 11:37:32 wcra cardmgr[80]: initializing socket 1
Aug 22 11:37:32 wcra cardmgr[80]: socket 1: KTI ETHER-C16 Fast ethernet
Aug 22 11:37:33 wcra cardmgr[80]: executing: 'modprobe 8390'
Aug 22 11:37:33 wcra cardmgr[80]: executing: 'modprobe pcnet_cs'
Aug 22 11:37:33 wcra kernel: eth0: NE2000 (DL10022 rev 30): io 0x300, irq 9, hw_addr 00:E0:98:77:54:FD
Aug 22 11:37:33 wcra cardmgr[80]: executing: './network start eth0'

So it seems the 'KTI ETHER-C-16' driver is installed, 
but cardctl still thinks its the 'D-Link':

bash-2.03# date
Tue Aug 22 11:44:38 CEST 2000
bash-2.03# cardctl ident
Socket 0:
  no product info available
Socket 1:
  product info: "D-Link", "DFE-650", "Fast Ethernet", "Rev. D1"
  manfid: 0x0149, 0x0230
  function: 6 (network)


Thanks in advance for any suggestions.

Regards,

William BC Crandall
Wiesbaden . Germany
wcra@beest.midat.de  and/or
crandall@wiesbaden.vistec.net
1. None The correct driver *is* loaded by David Hinds, 2000, Aug 22

Question Compex Linkport/TX32 and EEPROM

Date: 2000, Aug 26
From: Marco Dubbeld marco

When I insert my Linkport TX32 card cardmgr reports in messages:

cardmgr[9623]: initializing socket 0
cardmgr[9623]: socket 0: Compex LinkPort/TX32A
kernel: cs: cb_alloc(bus 32): vendor 0x10ec, device 0x8138
cardmgr[9623]: executing: 'modprobe cb_enabler'
cardmgr[9623]: executing: 'modprobe tulip_cb 10baseT debu
kernel: fn 0 bar 1: io 0xa00-0xaff
kernel: fn 0 bar 2: mem 0x60001000-0x600011ff
kernel: irq 11
kernel: tulip_attach(device 20:00.0)
kernel: tulip.c:v0.91g-ppc 7/16/99 becker@cesdis.gsfc.nasord)
kernel: eth0: Digital DS21143 Tulip rev 16 at 0xa00, EEPROM not present, 00:4C:69:6E:75:79, IRQ 11.
kernel: eth0: Missing EEPROM, this interface may not work correctly!
cardmgr[9623]: executing: './network start eth0'
kernel: eth0: Tx hung, 11 vs. 0.
kernel: eth0: 21140 transmit timed out, status 0801fbfc,
SIA 0000000f 00000000 00000018 58fab388, resetting...

And then the interface doesn't work.
What can I do?

Thanks.
1. None "Tx Hung" often means an interrupt delivery problem by David Hinds, 2000, Aug 28
(_ None more info Compex TX32, "Tx hung" problem by Marco Dubbeld, 2000, Aug 29
(_ None Wait a second: this is not the right driver! by David Hinds, 2000, Sep 01

Feedback Addition to pcmcia/config file required.

Date: 2000, Aug 27
From: Rudi Heitbaum rudi

Addition to the config file to enable the EtherExpress PrO/100 Lan+Modem56 CardBus II.

card "Intel EtherExpress PRO/100 Lan+Modem56 CardBus II"
  manfid 0x0089, 0x1103
  bind "eepro100_cb" to 0, "serial_cb" to 1

The other thing I found is if minicom is set to use hardware flow control the laptop hangs.

/sbin/cardctl ident 
  product info: "INTEL(R)", "PRO/100 LAN+MODEM56 CARDBUS II", "MBLA3456", "1.00"  manfid: 0x0089, 0x1103

Regards

Rudi

Sad 2.4.0 test 7 broke pcnet_cs

Date: 2000, Aug 28
From: Heather Stern starshine

I decided to give the latest test kernel a shot since PCMCIA is supposed to be a bit handier now that it's in the real tree. Looks like they missed something in the Great ReOrg though. And the amazing thing is, they only missed the most commonly used module for me :( pcnet_cs doesn't work because it's missing these symbols:

 Aug 27 18:18:06 venus cardmgr[108]: insmod exited with status 1
 Aug 27 18:18:06 venus cardmgr[108]: executing: 'modprobe pcnet_cs'
 Aug 27 18:18:06 venus cardmgr[108]: + /lib/modules/2.4.0-test7/pcmcia/pcnet_cs.o: unresolved symbol ei_interrupt_Rb93d8fa5
 Aug 27 18:18:06 venus cardmgr[108]: + /lib/modules/2.4.0-test7/pcmcia/pcnet_cs.o: unresolved symbol ei_open_R7308bf72
 Aug 27 18:18:06 venus cardmgr[108]: + /lib/modules/2.4.0-test7/pcmcia/pcnet_cs.o: unresolved symbol NS8390_init_R73472e03
 Aug 27 18:18:06 venus cardmgr[108]: + /lib/modules/2.4.0-test7/pcmcia/pcnet_cs.o: unresolved symbol ethdev_init_R98ed9a17
 Aug 27 18:18:06 venus cardmgr[108]: + /lib/modules/2.4.0-test7/pcmcia/pcnet_cs.o: insmod /lib/modules/2.4.0-test7/pcmcia/pcnet_cs.o failed
 Aug 27 18:18:06 venus cardmgr[108]: + /lib/modules/2.4.0-test7/pcmcia/pcnet_cs.o: insmod pcnet_cs failed
 Aug 27 18:18:06 venus cardmgr[108]: modprobe exited with status 255

This was on a pentium 233 MMX. Haven't had a chance to try it compiled for other CPUs. Card services are version 3.1.19.

Thanks ... looking forward to using a new kernel one of these days.

Heather *

1. None Makefile bug by David Hinds, 2000, Sep 01

Question Dynalink LC100-32-2

Date: 2000, Aug 30
From: Mitsuo <mfuchida@hotmail.com>

Hi friends

Does anybody know how to make this pcmcia works (Dynalink
LC100-32-2), or if this card is actually supported by linux?
I did search www.dynalink.com site for support (2-tech) and
i didn´t make it by myself (i´m just a linux newbie), so 
would anyone mind helping me?

Thank´s advanced and regards

Question Silicom pcmcia drivers: make error ?

Date: 2000, Nov 24
From: Graham FennerBarbour grahamfennerbarbour

I am trying to install the Silicom pcmcia drivers on my Laptech (?) laptop that is running Redhat 6.2.

Cardmgr is already installed (from setup) and seems to be working fine, issuing the apopriate beeps when the card is inserted etc.

I always get an error when running the make files.

Unfortunately, I am a Linux newbie (but to nothing else), so I assume the problem is not really for this forum, but know nowhere else to begin.

Has anyone encountered this problem?

PCMCIA Network Adapter Issues


Add Message to: "PCMCIA Network Adapter Issues"

Members Subscribe Admin Mode Show Frames Help for HyperNews at pcmcia-cs.sourceforge.net 1.10