devfs_d_revalidate_wait adds to a waitqueue but never removes from it
again so we there's one entry full of reused stack space added on
each call (I wonder how this ever worked).

The function has a few more bugs (it effectivly does a sleep_on instead
of checking for the actual even and can't deal with negative dentries
at all), but I just had breakfast and don't want to poke into devfs
internals deeper - I still hope Adam's smalldevfs will get merged
anyway..

fix yet another manual reschedule point

into home.transmeta.com:/home/torvalds/v2.5/linux

Also add abort functionality

- New style ide-default driver
- Don't attach non existant drives
- DRIVER()==NULL checks can go
- Ioctl checks that were missing are now in

This isnt perfect, there is a race left somewhere still but its closer.

Make drive->id not NULL an invariant
Clean up a few things from that
hwif specific queue length
initialisation/IRQ cleanups

Note: this changes the default blocks limit per I/O to 256. I've still seen
no credible evidence that its a problem and "other OS's" do it.

This is the first of a set of changes to make DRIVER(drive)!=NULL an
invariant

Yes NEC use *different* function numbers!!