On Fri, Nov 01, 2002 at 05:38:22PM +1300, Gregory McGarry wrote:
> - use the existing code
> - use the C version
> - use Charles' untested table version
> 
> Are there strong objections to going to C versions?  In all likelihood
> the runqueue will be redesigned for different schedulers and to improve
> the performance on multiprocessor systems anyway.

I have no objections to using a C version, it's easier to maintain.
If it really turns out to be a performance hit (I don't think so),
it's always possible to come up with an asm version for a release.

- Frank