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;
Content-Type: text/plain;
        charset=us-ascii
Subject: Re: [PATCH] sched/fair: Ensure _sum and _avg values stay consistent
From:   Sachin Sant <sachinp@linux.vnet.ibm.com>
In-Reply-To: <20210624111815.57937-1-odin@uged.al>
Date:   Thu, 24 Jun 2021 19:18:09 +0530
Cc:     Peter Zijlstra <peterz@infradead.org>,
        Vincent Guittot <vincent.guittot@linaro.org>,
        Dietmar Eggemann <dietmar.eggemann@arm.com>,
        Naresh Kamboju <naresh.kamboju@linaro.org>,
        Ingo Molnar <mingo@redhat.com>,
        Juri Lelli <juri.lelli@redhat.com>,
        Steven Rostedt <rostedt@goodmis.org>,
        Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
        Daniel Bristot de Oliveira <bristot@redhat.com>,
        linux-kernel@vger.kernel.org
Message-Id: <5EA7BE86-F610-424F-B443-69C87944E48A@linux.vnet.ibm.com>
References: <20210624111815.57937-1-odin@uged.al>
To:     Odin Ugedal <odin@uged.al>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Precedence: bulk


> On 24-Jun-2021, at 4:48 PM, Odin Ugedal <odin@uged.al> wrote:
>=20
> The _sum and _avg values are in general sync together with the PELT
> divider. They are however not always completely in perfect sync,
> resulting in situations where _sum gets to zero while _avg stays
> positive. Such situations are undesirable.
>=20
> This comes from the fact that PELT will increase period_contrib, also
> increasing the PELT divider, without updating _sum and _avg values to
> stay in perfect sync where (_sum =3D=3D _avg * divider). However, such PE=
LT
> change will never lower _sum, making it impossible to end up in a
> situation where _sum is zero and _avg is not.
>=20
> Therefore, we need to ensure that when subtracting load outside PELT,
> that when _sum is zero, _avg is also set to zero. This occurs when
> (_sum < _avg * divider), and the subtracted (_avg * divider) is bigger
> or equal to the current _sum, while the subtracted _avg is smaller than
> the current _avg.
>=20
> Reported-by: Sachin Sant <sachinp@linux.vnet.ibm.com>
> Reported-by: Naresh Kamboju <naresh.kamboju@linaro.org>
> Signed-off-by: Odin Ugedal <odin@uged.al>
> ---
>=20

Thank You for the fix. Works for me.

Tested-by: Sachin Sant <sachinp@linux.vnet.ibm.com>

Thanks
-Sachin

> Reports and discussion can be found here:
>=20
> https://lore.kernel.org/lkml/2ED1BDF5-BC0C-47CD-8F33-9A46C738F8CF@linux.v=
net.ibm.com/
> https://lore.kernel.org/lkml/CA+G9fYsMXELmjGUzw4SY1bghTYz_PeR2diM6dRp2J37=
bBZzMSA@mail.gmail.com/
>=20
> kernel/sched/fair.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>=20
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index bfaa6e1f6067..def48bc2e90b 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -3688,15 +3688,15 @@ update_cfs_rq_load_avg(u64 now, struct cfs_rq *cf=
s_rq)
>=20
> 		r =3D removed_load;
> 		sub_positive(&sa->load_avg, r);
> -		sub_positive(&sa->load_sum, r * divider);
> +		sa->load_sum =3D sa->load_avg * divider;
>=20
> 		r =3D removed_util;
> 		sub_positive(&sa->util_avg, r);
> -		sub_positive(&sa->util_sum, r * divider);
> +		sa->util_sum =3D sa->util_avg * divider;
>=20
> 		r =3D removed_runnable;
> 		sub_positive(&sa->runnable_avg, r);
> -		sub_positive(&sa->runnable_sum, r * divider);
> +		sa->runnable_sum =3D sa->runnable_avg * divider;
>=20
> 		/*
> 		 * removed_runnable is the unweighted version of removed_load so we
> --=20
> 2.32.0
>=20