On Wed, 2 May 2001, Toru Nishimura wrote:

> >>   I can't think of any way to fix this other than setting the SR in
> >> cpu_fork().
> >
> > Should be fine, and is exactly what the i386 and alpha kernels do.
> 
> I guess that proc0 == uvm_scheduler() is left running at splhigh state,
> at least in pmax case.

  It shouldn't be - the kernel is running at spl0() in most of main()
[I've verified that it's at spl0() when it forks the kthread processes]
and I don't see any splhigh() before it calls uvm_scheduler().

--
Michael L. Hitch			mhitch@montana.edu
Computer Consultant
Information Technology Center
Montana State University	Bozeman, MT	USA