Messages in this thread | | | Date | Sat, 18 Mar 2000 09:44:48 -0800 (PST) | From | Linus Torvalds <> | Subject | Re: Cardbus oops (after resume) |
| |
On Sat, 18 Mar 2000, Wakko Warner wrote: > > But, I received an oops. Here's what happened: > I'm on battery power (9% left). As soon as 8% hits, it goes to suspend. No > problem, power handy, plug in and unsuspend. I was in X playing an mp3 with > xmms. Everything resumed except the network. This card is a 3cxfe575ct (I > hope I got the number right). I figure, no problem, eject and reinsert. I > got a light on the hub and my IP is assigned via bootp. No IP. I ran > ifconfig eth0 and it hung (D state). My window manager (icewm) reports > network traffic on it's bar. It went in D state as well. I have an oops > that happened when I ran ifconfig. I noticed the decoded oops said > something about apci (I don't have that enabled!)
The "ACPI" here is not the ACPI configuration option the kernel talks about, but simply the internal driver knowledge about the power management interface on the 3c575 network card, which is ACPI-enabled.
What seems to be happening is that the PCMCIA layer has ejected the card and freed the PCI device structure, but the driver still used it. Boom.
The problem seems to be that the vortex driver doesn't detach very nicely. I would LOVE to have the vortex driver too just converted to the new PCI driver interfaces which are less likely to have this kind of confusion (synchronization between PCI events and PCMCIA events).
Linus
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |