On Wed, Nov 28, 2001 at 11:09:46AM -0800, Jonathan Stone wrote:
> 
> Micro-optimizing like this, with the assumption of a small subset of
> contemporary CPUs is all very well, if that's the only target. NetBSD
> has to run well on more CPUs -- even contemporary CPUs -- than dreamt
> of in your philosophy.
> 
> Perhaps we should provide two or three hashes, and have
> architecture-specific defopt's to select a "good" one for the
> kernel target CPU.

yes. overridable in some way, so that this stays benchmarkable... and, e.g.,
even within a CPU family,, different members have different performance
characteristics.

Regards,
	Ignatios