Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Mon, 26 Oct 2020 17:20:29 +0100
From:   Vincent Guittot <vincent.guittot@linaro.org>
To:     Rik van Riel <riel@surriel.com>
Cc:     Chris Mason <clm@fb.com>, Peter Zijlstra <peterz@infradead.org>,
        Johannes Weiner <hannes@cmpxchg.org>,
        linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] fix scheduler regression from "sched/fair: Rework
 load_balance()"
Message-ID: <20201026162029.GA11367@vingu-book>
References: <DB4481A8-FD4E-4879-9CD2-275ABAFC09CF@fb.com>
 <CAKfTPtBiOFXwV9SkZ=YBw16xoS6LSrKVR4sFX6r2hZPZ9_5-+A@mail.gmail.com>
 <0014CA62-A632-495A-92B0-4B14C8CA193C@fb.com>
 <20201026142455.GA13495@vingu-book>
 <465597a2250d69346cff73dd07817794d3e80244.camel@surriel.com>
 <CAKfTPtCVzass7GM5oj3o3y0ru4HQViWZc2+D-RpFoLvg=__FrA@mail.gmail.com>
 <334f491d2887a6ed7c5347d5125412849feb8a0a.camel@surriel.com>
 <CAKfTPtAfKn0jzOpPNR4NUb0zLs02iLQq2_UCDSCEwhTB2LDAig@mail.gmail.com>
 <20201026120445.6a5dbbbe@imladris.surriel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <20201026120445.6a5dbbbe@imladris.surriel.com>
User-Agent: Mutt/1.9.4 (2018-02-28)
Precedence: bulk

Le lundi 26 oct. 2020 ? 12:04:45 (-0400), Rik van Riel a ?crit :
> On Mon, 26 Oct 2020 16:42:14 +0100
> Vincent Guittot <vincent.guittot@linaro.org> wrote:
> > On Mon, 26 Oct 2020 at 16:04, Rik van Riel <riel@surriel.com> wrote:
> 
> > > Could utilization estimates be off, either lagging or
> > > simply having a wrong estimate for a task, resulting
> > > in no task getting pulled sometimes, while doing a
> > > migrate_task imbalance always moves over something?  
> > 
> > task and cpu utilization are not always up to fully synced and may lag
> > a bit which explains that sometimes LB can fail to migrate for a small
> > diff
> 
> OK, running with this little snippet below, I see latencies
> improve back to near where they used to be:
> 
> Latency percentiles (usec) runtime 150 (s)
>         50.0th: 13
>         75.0th: 31
>         90.0th: 69
>         95.0th: 90
>         *99.0th: 761
>         99.5th: 2268
>         99.9th: 9104
>         min=1, max=16158
> 
> I suspect the right/cleaner approach might be to use
> migrate_task more in !CPU_NOT_IDLE cases?
> 
> Running a task to an idle CPU immediately, instead of refusing
> to have the load balancer move it, improves latencies for fairly
> obvious reasons.
> 
> I am not entirely clear on why the load balancer should need to
> be any more conservative about moving tasks than the wakeup
> path is in eg. select_idle_sibling.


what you are suggesting is something like:
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 4978964e75e5..3b6fbf33abc2 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -9156,7 +9156,8 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
         * emptying busiest.
         */
        if (local->group_type == group_has_spare) {
-               if (busiest->group_type > group_fully_busy) {
+               if ((busiest->group_type > group_fully_busy) &&
+                   !(env->sd->flags & SD_SHARE_PKG_RESOURCES)) {
                        /*
                         * If busiest is overloaded, try to fill spare
                         * capacity. This might end up creating spare capacity

which also fixes the problem for me and alignes LB with wakeup path regarding the migration
in the LLC

> 
> 
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 35bdc0cccfa6..60acf71a2d39 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -7415,7 +7415,7 @@ static int detach_tasks(struct lb_env *env)
>  		case migrate_util:
>  			util = task_util_est(p);
>  
> -			if (util > env->imbalance)
> +			if (util > env->imbalance && env->idle == CPU_NOT_IDLE)

we don't want to be so agressive outside LLC and such bypass will defeat the purpose of
using utilization instead of nr_running.


>  				goto next;
>  
>  			env->imbalance -= util;