Messages in this thread | | | Date | Fri, 15 Mar 2002 03:03:22 +0100 (MET) | From | Erich Focht <> | Subject | Re: [Lse-tech] Re: Node affine NUMA scheduler |
| |
On Thu, 14 Mar 2002, Gerrit Huizenga wrote:
> Beware of optimizing for the benchmark. In real life, exec is likely > a much better place for the balancing.
:-) Thanks for the warning. I'll play around with the possibilities by (at least temporarilly) including a node_policy or balance_policy. I'm having in mind some big OpenMP codes which could benefit of initial balancing which they won't get in do_exec().
Regards, Erich
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |