Problems w/ 2.6.0-test5 and orinocco device

In short, I can bring up dev eth2 which is a wireless net card (orinocco), but if I bring it down, it breaks. I can't do anything with it anymore; I have to reboot before I can use it again. This didn't happen in -test4. Once in a while in -test4 and below it did break but broke differently. Mail me if you want details on it.

orinoco.c 0.13e (David Gibson <[email protected]> and others) orinoco_cs.c 0.13e (David Gibson <[email protected]> and others) orinoco_plx.c 0.13e (Daniel Barlow <[email protected]>, David Gibson <[email protected]>)

eth2: Station identity 001f:0001:0008:000a
eth2: Looks like a Lucent/Agere firmware version 8.10
eth2: Ad-hoc demo mode supported
eth2: IEEE standard IBSS ad-hoc mode supported
eth2: WEP supported, 104-bit key
eth2: MAC address 00:02:2D:5C:18:9F
eth2: Station name "HERMES I"
eth2: ready
eth2: index 0x01: Vcc 3.3, irq 9, io 0x0100-0x013f
eth2: New link status: Connected (0001) spurious 8259A interrupt: IRQ7.
eth2: New link status: AP Out of Range (0004)
eth2: New link status: AP In Range (0005)

Sorry I've taken a while to reply - I've had other things on my plate and not much time for orinoco work. Any details you have would be useful.

In particular any error messages would be handy - since the crash might prevent them getting to syslog, try switching to a text console and running "dmesg -n 8" to request that all messages, including debug level ones go to the console immediately. Then trigger the bug and see if any messages appear on the console.

David Gibson			| For every complex problem there is a
[email protected]	| solution which is simple, neat and
				| wrong.
http://www.ozlabs.org/people/dgibson