RE: + sched-use-tasklet-to-call-balancing.patch added to -mm tree
From: Chen, Kenneth W
Date: Fri Nov 10 2006 - 21:40:46 EST
Christoph Lameter wrote on Friday, November 10, 2006 5:01 PM
> On Fri, 10 Nov 2006, Ingo Molnar wrote:
>
> > ok, that's what i suspected - what made the difference wasnt the fact
> > that it was moved out of irqs-off section, but that it was running
> > globally, instead of in parallel on every cpu. I have no conceptual
> > problem with single-threading the more invasive load-balancing bits.
> > (since it has to touch every runqueue anyway there's probably little
> > parallelism possible) But it's a scary change nevertheless, it
> > materially affects every SMP system's balancing characteristics.
>
> We saw multiple issues. The first we saw was interrupt holdoff related
> since IPIs took a long time to complete. The other was that multiple
> load balance actions in multiple CPUs seem to serialize on the locks
> trying each to move tasks off the same busy processor. So both better be
> addressed.
So designate only one CPU within a domain to do load balance between groups
for that specific sched domain should in theory fix the 2nd problem you
identified. Did you get a chance to look at the patch Suresh posted?
-
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/