Date: Mon, 21 Aug 2017 14:44:58 +0100
From: Juri Lelli <juri.lelli@arm.com>
To: Byungchul Park <byungchul.park@lge.com>
Cc: peterz@infradead.org, mingo@kernel.org, joel.opensrc@gmail.com,
        linux-kernel@vger.kernel.org, juri.lelli@gmail.com,
        rostedt@goodmis.org, kernel-team@lge.com
Subject: Re: [PATCH v8 1/2] sched/deadline: Add support for SD_PREFER_SIBLING
 on find_later_rq()
Message-ID: <20170821134458.gocjoobaybb45egi@e106622-lin>
References: <1503044519-2804-1-git-send-email-byungchul.park@lge.com>
 <1503044519-2804-2-git-send-email-byungchul.park@lge.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1503044519-2804-2-git-send-email-byungchul.park@lge.com>
User-Agent: NeoMutt/20170113 (1.7.2)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2701
Lines: 86

Hi,

On 18/08/17 17:21, Byungchul Park wrote:
> It would be better to try to check other siblings first if
> SD_PREFER_SIBLING is flaged when pushing tasks - migration.
> 
> Signed-off-by: Byungchul Park <byungchul.park@lge.com>

Mmm, this looks like Peter's proposed patch, maybe add (at least) a
Suggested-by: him ?

https://marc.info/?l=linux-kernel&m=150176183807073

Also, I'm not sure what Peter meant with

"But still this isn't quite right, because when we consider this for SMT
(as was the intent here) we'll happily occupy a full sibling core over
finding an empty one."

since we are still using the later_mask, which should not include full
cores (unless it is the one with the lates deadline)?

> ---
>  kernel/sched/deadline.c | 55 ++++++++++++++++++++++++++++++++++++++++++++++---
>  1 file changed, 52 insertions(+), 3 deletions(-)
> 
> diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
> index 0223694..115250b 100644
> --- a/kernel/sched/deadline.c
> +++ b/kernel/sched/deadline.c
> @@ -1319,12 +1319,35 @@ static struct task_struct *pick_earliest_pushable_dl_task(struct rq *rq, int cpu
>  
>  static DEFINE_PER_CPU(cpumask_var_t, local_cpu_mask_dl);
>  
> +/*
> + * Find the first cpu in: mask & sd & ~prefer
> + */
> +static int find_cpu(const struct cpumask *mask,
> +		    const struct sched_domain *sd,
> +		    const struct sched_domain *prefer)
> +{
> +	const struct cpumask *sds = sched_domain_span(sd);
> +	const struct cpumask *ps  = prefer ? sched_domain_span(prefer) : NULL;
> +	int cpu;
> +
> +	for_each_cpu(cpu, mask) {
> +		if (!cpumask_test_cpu(cpu, sds))
> +			continue;
> +		if (ps && cpumask_test_cpu(cpu, ps))
> +			continue;
> +		break;
> +	}
> +
> +	return cpu;
> +}
> +
>  static int find_later_rq(struct task_struct *task)
>  {
> -	struct sched_domain *sd;
> +	struct sched_domain *sd, *prefer = NULL;
>  	struct cpumask *later_mask = this_cpu_cpumask_var_ptr(local_cpu_mask_dl);
>  	int this_cpu = smp_processor_id();
>  	int cpu = task_cpu(task);
> +	int fallback_cpu = -1;
>  
>  	/* Make sure the mask is initialized first */
>  	if (unlikely(!later_mask))
> @@ -1376,8 +1399,7 @@ static int find_later_rq(struct task_struct *task)
>  				return this_cpu;
>  			}
>  
> -			best_cpu = cpumask_first_and(later_mask,
> -							sched_domain_span(sd));
> +			best_cpu = find_cpu(later_mask, sd, prefer);
>  			/*
>  			 * Last chance: if a cpu being in both later_mask
>  			 * and current sd span is valid, that becomes our
> @@ -1385,6 +1407,26 @@ static int find_later_rq(struct task_struct *task)
>  			 * already under consideration through later_mask.
>  			 */

It seems that the comment above should be updated as well.

Thanks,

- Juri