On Fri, Jul 20, 2001 at 12:07:11PM +0930, Brett Lymn wrote:
> According to Andreas Gustafsson:
> >
> >There is a problem with using these instructions in the NetBSD kernel:
> >they use the floating point registers,
> >
> 
> True for the older processors - I believe that later processors do not
> overload the FPU registers in this manner.

Would be worse. We don't save the MMX registers inside the kernel, either, 
if they're different from the FP registers...

	-is