Re: pcmcia-cs (cardmgr) problem - "eth0: command 0x5800 did not complete!"

Just for confirmation, both the unit battery and pcmcia sleeve battery were
fully charged.
I also tested each pcmcia slot, with/without DC adapter power, with/without
LCD light
(assuming, perhaps, the odd 3600 series heat issues were in effect).

Each test yielded the same results, so I'm lead to believe that it's not a
battery/power issue in this case.

Also, I ran through the packages to make sure that nothing critical was
missing or mismatched.
All installed binaries and modules are aligned with the current kernel
version, etc.

This particular piece of the error message caught my attention:

> eth0: 3Com 3c589, io 0xf6000300, irq 42, hw_addr 00:00:00:00:00:00

irq 42? (it's 9 in /proc/interrupts) and a hwaddr of 00:00:00:00:00:00

Something is being naughty. Again, I may not be seeing something obvious.

Any other suggestions?

----- Original Message -----
From: "heissenhuber" <heissenhuber_at_iabg.de>
To: "mowse" <mowse_at_mowse.ne.mediaone.net>
Cc: <familiar_at_handhelds.org>
Sent: Thursday, August 09, 2001 2:28 AM
Subject: Re: [Familiar] pcmcia-cs (cardmgr) problem - "eth0: command 0x5800
did not complete!"

> Hi,
> I've had similar problems. The reason was simply that the battery of the
> jacket was not charged.
>
> Florian
>
>
> > Working with the following:
> >
> > 2.4.6-rmk1-np2-devfs
> > task-bootstrap-2.4.6-rmk1-np2-devfs-hh4.jffs2
> > pcmcia-modules-2.4.6-rmk1-np2-devfs_hh4_arm.ipk
> > several pcmcia ethernet/80211 cards
> >
> > Upon insertion of a pcmcia NIC into a dual-slot sleeve, the following
> > occured:
> >
> > # bitsy_pcmcia_configure_socket: socket=0 vcc=50 vpp=50 reset=0
> > bitsy_pcmcia_configure_socket: socket=0 vcc=50 vpp=50 reset=1
> > bitsy_pcmcia_configure_socket: socket=0 vcc=50 vpp=50 reset=0
> > cardmgr[76]: initializing socket 0
> > cardmgr[76]: socket 0: 3Com 589 Ethernet
> > cardmgr[76]: executing: 'modprobe 3c589_cs'
> > bitsy_pcmcia_configure_socket: socket=0 vcc=50 vpp=0 reset=0
> > bitsy_pcmcia_configure_socket: socket=0 vcc=50 vpp=0 reset=0
> > eth0: 3Com 3c589, io 0xf6000300, irq 42, hw_addr 00:00:00:00:00:00
> > 8K FIFO split 5:3 Rx:Tx, auto xcvr
> > cardmgr[76]: executing: './network start eth0'
> > NETDEV WATCHDOG: eth0: transmit timed out
> > eth0: Transmit timed out!
> > irq status 0808, rx status 0808, tx status 08 tx free 0808
> > diagnostics: fifo 0808 net 0808 ethernet 0808 media 0808
> > eth0: command 0x5800 did not complete!
> > NETDEV WATCHDOG: eth0: transmit timed out
> > eth0: Transmit timed out!
> > irq status 0808, rx status 0808, tx status 08 tx free 0808
> > diagnostics: fifo 0808 net 0808 ethernet 0808 media 0808
> > eth0: command 0x5800 did not complete!
> > NETDEV WATCHDOG: eth0: transmit timed out
> > eth0: Transmit timed out!
> > irq status 0808, rx status 0808, tx status 08 tx free 0808
> > diagnostics: fifo 0808 net 0808 ethernet 0808 media 0808
> > eth0: command 0x5800 did not complete!
> >
> > ...<repeats timeout message a few more times>...
> >
> > cardmgr[76]: + Operation failed.
> >
> > ...<The interface is brought down and the associated modules are
> > unloaded>...
> >
> > Oddly, I was able to bring up the interface, ifconfig/route the
interface,
> > and ping via loopback and the newly assigned ip (which, from my
> > understanding, is routed through loopback by linux tcp/ip
implementation).
> > However, after repeated "time out" errors, cardmgr bombs and the
interface
> > is brought down. This occured with the 3com 3c589d, the D-Link DWL650,
the
> > SMC2632W, and the Cisco Aironet 340. Has anyone else experienced this
> > situation within a similar environment? Am I forgetting something
silly?
> >
> > Thanks in advance,
> > mowse_at_mowse.ne.mediaone.net
>
> --
> _______________________________________________
> Familiar mailing list
> Familiar_at_handhelds.org
> http://handhelds.org/mailman/listinfo/familiar
> irc://irc.openprojects.net #familiar
>
Received on Wed Aug 08 2001 - 23:50:58 EDT