Date: Mon, 8 Jan 2018 23:18:51 +0100
From: Peter Zijlstra <peterz@infradead.org>
To: subhra mazumdar <subhra.mazumdar@oracle.com>
Cc: linux-kernel@vger.kernel.org, mingo@redhat.com,
        steven.sistare@oracle.com, dhaval.giani@oracle.com
Subject: Re: [RFC PATCH V2] sched: Improve scalability of select_idle_sibling
 using SMT balance
Message-ID: <20180108221851.GV29822@worktop.programming.kicks-ass.net>
References: <20180108221237.31761-1-subhra.mazumdar@oracle.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20180108221237.31761-1-subhra.mazumdar@oracle.com>
User-Agent: Mutt/1.5.22.1 (2013-10-16)
Sender: linux-kernel-owner@vger.kernel.org

On Mon, Jan 08, 2018 at 02:12:37PM -0800, subhra mazumdar wrote:
> @@ -2751,6 +2763,31 @@ context_switch(struct rq *rq, struct task_struct *prev,
>  	       struct task_struct *next, struct rq_flags *rf)
>  {
>  	struct mm_struct *mm, *oldmm;
> +	int this_cpu = rq->cpu;
> +	struct sched_domain *sd;
> +	int prev_busy, next_busy;
> +
> +	if (rq->curr_util == UTIL_UNINITIALIZED)
> +		prev_busy = 0;
> +	else
> +		prev_busy = (prev != rq->idle);
> +	next_busy = (next != rq->idle);
> +
> +	/*
> +	 * From sd_llc downward update the SMT utilization.
> +	 * Skip the lowest level 0.
> +	 */
> +	sd = rcu_dereference_sched(per_cpu(sd_llc, this_cpu));
> +	if (next_busy != prev_busy) {
> +		for_each_lower_domain(sd) {
> +			if (sd->level == 0)
> +				break;
> +			sd_context_switch(sd, rq, next_busy - prev_busy);
> +		}
> +	}
> +

No, we're not going to be adding atomic ops here. We've been arguing
over adding a single memory barrier to this path, atomic are just not
going to happen.

Also this is entirely the wrong way to do this, we already have code
paths that _know_ if they're going into or coming out of idle.