Re: [RFCv3 PATCH 29/48] sched: Highest energy aware balancing sched_domain level pointer

From: Morten Rasmussen
Date: Tue Mar 24 2015 - 06:52:33 EST


On Mon, Mar 23, 2015 at 04:16:06PM +0000, Peter Zijlstra wrote:
> On Wed, Feb 04, 2015 at 06:31:06PM +0000, Morten Rasmussen wrote:
> > Add another member to the family of per-cpu sched_domain shortcut
> > pointers. This one, sd_ea, points to the highest level at which energy
> > model is provided. At this level and all levels below all sched_groups
> > have energy model data attached.
>
> Here might be a good point to discuss what it means for the topology to
> have partial energy model information.

Partial energy model information is supported as in you can have energy
model data for just lower level sched domains and leave scheduling
between higher level sched_domains to be done the non-energy-aware
(current) way. For example, have energy-aware scheduling working within
each socket on a multi-socket system and let normal performance
(spreading) load-balancing work between sockets. Or within a NUMA node
and not between NUMA nodes as I mentioned in my reply to the next patch.
--
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/