Re: [RFC PATCH v2 5/7] sched: nominate preferred wakeup cpu

[Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>]

* Peter Zijlstra <a.p.zijlstra@chello.nl> [2008-09-08 15:21:31]:

> On Mon, 2008-09-08 at 18:51 +0530, Vaidyanathan Srinivasan wrote:
> > When the system utilisation is low and more cpus are idle,
> > then the process waking up from sleep should prefer to
> > wakeup an idle cpu from semi-idle cpu package (multi core
> > package) rather than a completely idle cpu package which
> > would waste power.
> > 
> > Use the sched_mc balance logic in find_busiest_group() to
> > nominate a preferred wakeup cpu.
> > 
> > This info can be sored in appropriate sched_domain, but
> > updating this info in all copies of sched_domain is not
> > practical.  For now lets try with a global variable.
> > 
> > Signed-off-by: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
> > ---
> > 
> >  kernel/sched.c |   10 ++++++++++
> >  1 files changed, 10 insertions(+), 0 deletions(-)
> > 
> > diff --git a/kernel/sched.c b/kernel/sched.c
> > index 569fc8d..4ae79f5 100644
> > --- a/kernel/sched.c
> > +++ b/kernel/sched.c
> > @@ -3380,6 +3380,9 @@ out_balanced:
> >  
> >  	if (this == group_leader && group_leader != group_min) {
> >  		*imbalance = min_load_per_task;
> > +		if (sched_mc_power_savings >= POWERSAVINGS_BALANCE_WAKEUP)
> > +			sched_mc_preferred_wakeup_cpu =
> > +					first_cpu(group_leader->cpumask);
> >  		return group_min;
> >  	}
> >  #endif
> > @@ -6911,6 +6914,13 @@ static void sched_domain_node_span(int node, cpumask_t *span)
> >  int sched_smt_power_savings = 0, sched_mc_power_savings = 0;
> >  
> >  /*
> > + * Preferred wake up cpu nominated by sched_mc balance that will be used when
> > + * most cpus are idle in the system indicating overall very low system
> > + * utilisation. Triggered at POWERSAVINGS_BALANCE_WAKEUP (2).
> > + */
> > +unsigned int sched_mc_preferred_wakeup_cpu;
> 
> This cannot be a global variable, what happens when we have two disjoint
> load-balance domains?

Agreed this is certainly a problem.  I tried adding this to the
sched_domain, but accessing the correct 'copy' for sched_domain that
holds this variable from any cpu is not fast.  

Thank you for pointing this out.  I will find a alternative
implementation.

--Vaidy