Re: [patch] HT scheduler, sched-2.5.68-A9

Ingo Molnar (mingo@redhat.com)
Wed, 23 Apr 2003 05:12:29 -0400 (EDT)


On Tue, 22 Apr 2003, Martin J. Bligh wrote:

> Hmmm. When the machine is loaded up fully, this seems OK, but for lower
> loads, it seems to have a significant degredation. This is on my normal
> 16-way machine, no HT at all ... just checking for degredations. At a
> guess, maybe it's bouncing stuff around between runqueues under low
> loads?

hm - it should not cause any change in the non-HT case. Oh, there's one
thing that could make a difference on SMP - could you change
AGRESSIVE_IDLE_STEAL back to 0?

> Are there things embedded in here that change stuff for non-HT machines?

not intentionally, but apparently there's some sort of effect. The idle
steal thing could explain it.

Ingo

-
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/