Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Mon, 22 Oct 2018 18:56:31 +0200
From:   Peter Zijlstra <peterz@infradead.org>
To:     Steve Sistare <steven.sistare@oracle.com>
Cc:     mingo@redhat.com, subhra.mazumdar@oracle.com,
        dhaval.giani@oracle.com, rohit.k.jain@oracle.com,
        daniel.m.jordan@oracle.com, pavel.tatashin@microsoft.com,
        matt@codeblueprint.co.uk, umgwanakikbuti@gmail.com,
        riel@redhat.com, jbacik@fb.com, juri.lelli@redhat.com,
        linux-kernel@vger.kernel.org, Steven Rostedt <rostedt@goodmis.org>
Subject: Re: [PATCH 04/10] sched/fair: Dynamically update cfs_overload_cpus
Message-ID: <20181022165631.GE3117@worktop.programming.kicks-ass.net>
References: <1540220381-424433-1-git-send-email-steven.sistare@oracle.com>
 <1540220381-424433-5-git-send-email-steven.sistare@oracle.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1540220381-424433-5-git-send-email-steven.sistare@oracle.com>
User-Agent: Mutt/1.5.22.1 (2013-10-16)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Oct 22, 2018 at 07:59:35AM -0700, Steve Sistare wrote:
> An overloaded CPU has more than 1 runnable task.  When a CFS task wakes
> on a CPU, if h_nr_running transitions from 1 to more, then set the CPU in
> the cfs_overload_cpus bitmap.  When a CFS task sleeps, if h_nr_running
> transitions from 2 to less, then clear the CPU in cfs_overload_cpus.

> @@ -5111,8 +5147,12 @@ static inline void hrtick_update(struct rq *rq)
>  		update_cfs_group(se);
>  	}
>  
> -	if (!se)
> +	if (!se) {
>  		add_nr_running(rq, 1);
> +		if (prev_nr == 1)
> +			overload_set(rq);
> +
> +	}

This is very similar to what RT already has, except they also track if
the tasks in question is in fact migratable (p->nr_cpus_allowed > 1).

And similarly, it might make sense to use the sparse mask thing over
there.