Re: [PATCH 2/5] sched: factor out code to should_we_balance()

[Peter Zijlstra]

On Tue, 2013-04-02 at 12:00 +0200, Peter Zijlstra wrote:
> On Tue, 2013-04-02 at 18:50 +0900, Joonsoo Kim wrote:
> > 
> > It seems that there is some misunderstanding about this patch.
> > In this patch, we don't iterate all groups. Instead, we iterate on
> > cpus of local sched_group only. So there is no penalty you mentioned.
> 
> OK, I'll go stare at it again..

Ah, I see, you're doing should_we_balance() _before_
find_busiest_group() and instead you're doing another for_each_cpu() in
there.

I'd write the thing like:

static bool should_we_balance(struct lb_env *env)
{
	struct sched_group *sg = env->sd->groups;
	struct cpumask *sg_cpus, *sg_mask;
	int cpu, balance_cpu = -1;

	if (env->idle == CPU_NEWLY_IDLE)
		return true;

	sg_cpus = sched_group_cpus(sg);
	sg_mask = sched_group_mask(sg);

	for_each_cpu_and(cpu, sg_cpus, env->cpus) {
		if (!cpumask_test_cpu(cpu, sg_mask))
			continue;

		if (!idle_cpu(cpu))
			continue;

		balance_cpu = cpu;
		break;
	}

	if (balance_cpu == -1)
		balance_cpu = group_balance_cpu(sg);

	return balance_cpu == env->dst_cpu;
}

I also considered doing the group_balance_cpu() first to avoid having
to do the idle_cpu() scan, but that's a slight behavioural change
afaict.

--
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/