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;
From:   Roman Gushchin <guro@fb.com>
To:     Rik van Riel <riel@surriel.com>
CC:     "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "linux-mm@kvack.org" <linux-mm@kvack.org>,
        Kernel Team <Kernel-team@fb.com>,
        Johannes Weiner <hannes@cmpxchg.org>, Chris Mason <clm@fb.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Michal Hocko <mhocko@suse.com>,
        "hange-folder>?" <toggle-mailboxes@castle.dhcp.thefacebook.com>
Subject: Re: [PATCH] mm,slab,vmscan: accumulate gradual pressure on small
 slabs
Thread-Topic: [PATCH] mm,slab,vmscan: accumulate gradual pressure on small
 slabs
Thread-Index: AQHUt0DNW/N3hHmpU0+8/TF4YJ27XaXFFWyA
Date:   Mon, 28 Jan 2019 19:45:09 +0000
Message-ID: <20190128194502.GA30061@castle.DHCP.thefacebook.com>
References: <20190128143535.7767c397@imladris.surriel.com>
In-Reply-To: <20190128143535.7767c397@imladris.surriel.com>
Accept-Language: en-US
Content-Language: en-US
x-ms-exchange-messagesentrepresentingtype: 1
x-originating-ip: [2620:10d:c090:180::1:b22f]
x-ms-publictraffictype: Email
x-microsoft-exchange-diagnostics: 1;BYAPR15MB2536;20:iBlr0/ip04jZsBqu1kZesOvkbYzfkihbuHywrAyaOPIl1xNGpPr3njQa2W1EJ5Ptmr3juhsJK2Era7OdEEDD4s5+clVuNnQYgqPqWLifAJhrDZiaE4NQJ1lobyr3aNMuzBlT0M7uMTehuBEdhd0B0aiexK3gvT1L1iFFZ4jxdjc=
x-ms-office365-filtering-correlation-id: 53b5d469-98d8-49c0-e4f3-08d685591c70
x-microsoft-antispam: BCL:0;PCL:0;RULEID:(2390118)(7020095)(4652040)(8989299)(4534185)(4627221)(201703031133081)(201702281549075)(8990200)(5600110)(711020)(4605077)(2017052603328)(7153060)(7193020);SRVR:BYAPR15MB2536;
x-ms-traffictypediagnostic: BYAPR15MB2536:
x-microsoft-antispam-prvs: <BYAPR15MB2536D4A764A076F0217CD1A3BE960@BYAPR15MB2536.namprd15.prod.outlook.com>
x-forefront-prvs: 0931CB1479
x-forefront-antispam-report: SFV:NSPM;SFS:(10019020)(346002)(366004)(39860400002)(136003)(376002)(396003)(189003)(199004)(76176011)(52116002)(97736004)(81156014)(446003)(25786009)(46003)(8676002)(53936002)(81166006)(476003)(7736002)(305945005)(54906003)(6916009)(6246003)(4326008)(33896004)(2906002)(86362001)(11346002)(316002)(99286004)(256004)(186003)(486006)(8936002)(68736007)(105586002)(71200400001)(106356001)(14454004)(33656002)(1076003)(229853002)(386003)(9686003)(6512007)(71190400001)(6486002)(6436002)(478600001)(6506007)(102836004)(6116002)(142933001);DIR:OUT;SFP:1102;SCL:1;SRVR:BYAPR15MB2536;H:BYAPR15MB2631.namprd15.prod.outlook.com;FPR:;SPF:None;LANG:en;PTR:InfoNoRecords;MX:1;A:1;
received-spf: None (protection.outlook.com: fb.com does not designate
 permitted sender hosts)
x-ms-exchange-senderadcheck: 1
x-microsoft-antispam-message-info: jw0lE7XldqByeR1JkYot52Iu4wRPJ3J5Ia3XGfa5JSJb6Dc2kAUenDNnYcQ8o6sBaPiHCpXnlNcTroIqkOeiIljsDKYoc6GUEWoU0UO+COpK87V0BQwUwUHAfm5ZTNZJyosapuExSkRD6LpADwYMhPH2peMwMMkJEWVGPyh+981RwD/Toz64qKgd5Woh30f71K+hSW2lifewdKjLxaRE22N2WPQ+JPJAPiwQEtk7FuNEytrm2ZWen8bSPuTWDxGL7xsfmCMcEq3+LWkBK2eACUtdhJJSIK2HhFgpcGrL9AjsxvIuFi6vTQjmjC3GSW72SAEj7z7KuPMwNl5RZf0ORV5wNTHEoWb3A8rHvrZgi+wyS/vXIyXB8/txJzZ4yhKpxbDFrLYB2tfigkShnoJII4NxfUhcB3vXwEE9JhPfBIg=
Content-Type: text/plain; charset="us-ascii"
Content-ID: <76A242FEEB281C4FB87A8E48DEF7F215@namprd15.prod.outlook.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: 53b5d469-98d8-49c0-e4f3-08d685591c70
X-MS-Exchange-CrossTenant-originalarrivaltime: 28 Jan 2019 19:45:08.7122
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 8ae927fe-1255-47a7-a2af-5f3a069daaa2
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BYAPR15MB2536
X-OriginatorOrg: fb.com
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:,, definitions=2019-01-28_10:,,
 signatures=0
X-Proofpoint-Spam-Reason: safe
X-FB-Internal: Safe
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Mon, Jan 28, 2019 at 02:35:35PM -0500, Rik van Riel wrote:
> There are a few issues with the way the number of slab objects to
> scan is calculated in do_shrink_slab.  First, for zero-seek slabs,
> we could leave the last object around forever. That could result
> in pinning a dying cgroup into memory, instead of reclaiming it.
> The fix for that is trivial.
>=20
> Secondly, small slabs receive much more pressure, relative to their
> size, than larger slabs, due to "rounding up" the minimum number of
> scanned objects to batch_size.
>=20
> We can keep the pressure on all slabs equal relative to their size
> by accumulating the scan pressure on small slabs over time, resulting
> in sometimes scanning an object, instead of always scanning several.
>=20
> This results in lower system CPU use, and a lower major fault rate,
> as actively used entries from smaller caches get reclaimed less
> aggressively, and need to be reloaded/recreated less often.
>=20
> Fixes: 4b85afbdacd2 ("mm: zero-seek shrinkers")
> Fixes: 172b06c32b94 ("mm: slowly shrink slabs with a relatively small num=
ber of objects")
> Cc: Johannes Weiner <hannes@cmpxchg.org>
> Cc: Chris Mason <clm@fb.com>
> Cc: Roman Gushchin <guro@fb.com>
> Cc: kernel-team@fb.com
> Tested-by: Chris Mason <clm@fb.com>

Hi, Rik!

There is a couple of formatting issues (see below), but other than that
the patch looks very good to me. Thanks!

Acked-by: Roman Gushchin <guro@fb.com>

> ---
>  include/linux/shrinker.h |  1 +
>  mm/vmscan.c              | 16 +++++++++++++---
>  2 files changed, 14 insertions(+), 3 deletions(-)
>=20
> diff --git a/include/linux/shrinker.h b/include/linux/shrinker.h
> index 9443cafd1969..7a9a1a0f935c 100644
> --- a/include/linux/shrinker.h
> +++ b/include/linux/shrinker.h
> @@ -65,6 +65,7 @@ struct shrinker {
> =20
>  	long batch;	/* reclaim batch size, 0 =3D default */
>  	int seeks;	/* seeks to recreate an obj */
> +	int small_scan;	/* accumulate pressure on slabs with few objects */
>  	unsigned flags;
> =20
>  	/* These are for internal use */
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index a714c4f800e9..0e375bd7a8b6 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -488,18 +488,28 @@ static unsigned long do_shrink_slab(struct shrink_c=
ontrol *shrinkctl,
>  		 * them aggressively under memory pressure to keep
>  		 * them from causing refetches in the IO caches.
>  		 */
> -		delta =3D freeable / 2;
> +		delta =3D (freeable + 1)/ 2;
                                      ^
                                      A space is missing here.
>  	}
> =20
>  	/*
>  	 * Make sure we apply some minimal pressure on default priority
> -	 * even on small cgroups. Stale objects are not only consuming memory
> +	 * even on small cgroups, by accumulating pressure across multiple
> +	 * slab shrinker runs. Stale objects are not only consuming memory
>  	 * by themselves, but can also hold a reference to a dying cgroup,
>  	 * preventing it from being reclaimed. A dying cgroup with all
>  	 * corresponding structures like per-cpu stats and kmem caches
>  	 * can be really big, so it may lead to a significant waste of memory.
>  	 */
> -	delta =3D max_t(unsigned long long, delta, min(freeable, batch_size));
> +	if (!delta) {
> +		shrinker->small_scan +=3D freeable;
> +
> +		delta =3D shrinker->small_scan >> priority;
> +		shrinker->small_scan -=3D delta << priority;
> +
> +		delta *=3D 4;
> +		do_div(delta, shrinker->seeks);
> +

This empty line can be removed, I believe.

> +	}
> =20
>  	total_scan +=3D delta;
>  	if (total_scan < 0) {
>=20