Re: [PATCH v2] sched: smart wake-affine

[Michael Wang <wangyun@linux.vnet.ibm.com>]

On 07/03/2013 04:50 PM, Peter Zijlstra wrote:
> On Wed, Jul 03, 2013 at 02:10:32PM +0800, Michael Wang wrote:
>> +static int wake_wide(struct task_struct *p)
>> +{
>> +	int factor = nr_cpus_node(cpu_to_node(smp_processor_id()));
> 
> That's still a cpumask_weight() in there... 

Exactly, although the scale is lower but still it's a mask...

> we should avoid that. How about something like the below:

Amazing, cost will be lowest when we got such counter, will make them a
patch set and test again ;-)

Regards,
Michael Wang

> 
>> +
>> +	/*
>> +	 * Yeah, it's the switching-frequency, could means many wakee or
>> +	 * rapidly switch, use factor here will just help to automatically
>> +	 * adjust the loose-degree, so bigger node will lead to more pull.
>> +	 */
>> +	if (p->nr_wakee_switch > factor) {
>> +		/*
>> +		 * wakee is somewhat hot, it needs certain amount of cpu
>> +		 * resource, so if waker is far more hot, prefer to leave
>> +		 * it alone.
>> +		 */
>> +		if (current->nr_wakee_switch > (factor * p->nr_wakee_switch))
>> +			return 1;
>> +	}
>> +
>> +	return 0;
>> +}
> 
> ---
> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> index 9b1f2e5..166ab9b 100644
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -5081,18 +5083,23 @@ static void destroy_sched_domains(struct sched_domain *sd, int cpu)
>   * two cpus are in the same cache domain, see cpus_share_cache().
>   */
>  DEFINE_PER_CPU(struct sched_domain *, sd_llc);
> +DEFINE_PER_CPU(int, sd_llc_size);
>  DEFINE_PER_CPU(int, sd_llc_id);
> 
>  static void update_top_cache_domain(int cpu)
>  {
>  	struct sched_domain *sd;
>  	int id = cpu;
> +	int size = 1;
> 
>  	sd = highest_flag_domain(cpu, SD_SHARE_PKG_RESOURCES);
> -	if (sd)
> +	if (sd) {
>  		id = cpumask_first(sched_domain_span(sd));
> +		size = cpumask_weight(sched_domain_span(sd));
> +	}
> 
>  	rcu_assign_pointer(per_cpu(sd_llc, cpu), sd);
> +	per_cpu(sd_llc_size, cpu) = size;
>  	per_cpu(sd_llc_id, cpu) = id;
>  }
> 
> diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
> index ef0a7b2..c992f58 100644
> --- a/kernel/sched/sched.h
> +++ b/kernel/sched/sched.h
> @@ -595,6 +595,7 @@ static inline struct sched_domain *highest_flag_domain(int cpu, int flag)
>  }
> 
>  DECLARE_PER_CPU(struct sched_domain *, sd_llc);
> +DECLARE_PER_CPU(int, sd_llc_size);
>  DECLARE_PER_CPU(int, sd_llc_id);
> 
>  struct sched_group_power {
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -3175,7 +3175,7 @@ static inline unsigned long effective_lo
> 
>  static int wake_wide(struct task_struct *p)
>  {
> -	int factor = nr_cpus_node(cpu_to_node(smp_processor_id()));
> +	int factor = this_cpu_read(sd_llc_size);
> 
>  	/*
>  	 * Yeah, it's the switching-frequency, could means many wakee or
> 
> --
> 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/
>