On Tue, Jan 12, 2010 at 08:48:21PM +0100, Matthias Drochner wrote: > But I think there is a more general problem: The driver reads a pointer > which can become invalid immediately after the lock is released, no > matter how the locking is done. Or am I missing something?
No, that is a problem affecting virtually every driver. We need to take care to get driver subroutines out of the device_t+softc before we tear them down. There are many softc entry points, so fixing the problem throughout the kernel looks like a hard problem. Dave -- David Young OJC Technologies [email protected] Urbana, IL * (217) 278-3933
