> I found that a kernel I had kicking around from 3/21 sources, did not hang,
> while a kernel built today does.  Ross and Charles suggested I start cvs
> updating and find where it breaks.
> 
> Unfortunately, my scratch disk beat them to it.  So now I am without spare disk
> space on my machine to build anything until the replacement arrives.  Perhaps
> someone else with CVS and a fast alpha could take a few hours and try to narrow
> it down.
> 
> My apologies..
> 

No problem. I was told that Charles was looking at it, so I stopped
looking at it myself.

Who *is* on it? If nobody currently is, I'll go back to tracking the
actual problem down- unless Bernd's conclusion that it's a compiler bug is
what folks are happy with.

-matt