Re: [rfc][patch] sched: remove smpnice
From: Martin Bligh
Date: Tue Feb 07 2006 - 18:35:31 EST
I think that the problems with excessive idling with the early versions
of my modifications to Con's patch showed that the load balancing code
is fairly sensitive to the average load per normal task not being
approximately 1. My latest patches restore this state of affairs and
kernbench testing indicates that the excessive idling has gone away (see
Martin J Bligh's message of 2006/01/29 11:52 "Re: -mm seems
significantly slower than mainline on kernbench thread").
I *think* the latest slowdown in -mm was due to some TSC side effects
from John's patches - see his other patch earlier today to fix (oops,
I forgot to reply to that ..)
So AFAICS, all issues with Peter's stuff were fixed.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/