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:     Greg Thelen <gthelen@google.com>
CC:     Johannes Weiner <hannes@cmpxchg.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Michal Hocko <mhocko@kernel.org>,
        "Vladimir Davydov" <vdavydov.dev@gmail.com>,
        Tejun Heo <tj@kernel.org>,
        "linux-mm@kvack.org" <linux-mm@kvack.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "stable@vger.kernel.org" <stable@vger.kernel.org>
Subject: Re: [PATCH v2] writeback: use exact memcg dirty counts
Thread-Topic: [PATCH v2] writeback: use exact memcg dirty counts
Thread-Index: AQHU5ldV+o6YfPDV3E6lTVYX7xBNtKYjEiEA
Date:   Fri, 29 Mar 2019 20:38:50 +0000
Message-ID: <20190329203844.GA24069@tower.DHCP.thefacebook.com>
References: <20190329174609.164344-1-gthelen@google.com>
In-Reply-To: <20190329174609.164344-1-gthelen@google.com>
Accept-Language: en-US
Content-Language: en-US
x-ms-exchange-messagesentrepresentingtype: 1
x-originating-ip: [2620:10d:c090:200::2:7c]
x-ms-publictraffictype: Email
x-ms-office365-filtering-correlation-id: 9bdeb2b8-affb-4cd2-3452-08d6b4868d17
x-microsoft-antispam: BCL:0;PCL:0;RULEID:(2390118)(7020095)(4652040)(8989299)(5600127)(711020)(4605104)(4534185)(4627221)(201703031133081)(201702281549075)(8990200)(2017052603328)(7153060)(7193020);SRVR:BYAPR15MB2536;
x-ms-traffictypediagnostic: BYAPR15MB2536:
x-microsoft-antispam-prvs: <BYAPR15MB25360863BEB5783D0D374006BE5A0@BYAPR15MB2536.namprd15.prod.outlook.com>
x-forefront-prvs: 0991CAB7B3
x-forefront-antispam-report: SFV:NSPM;SFS:(10019020)(39860400002)(136003)(376002)(396003)(346002)(366004)(189003)(199004)(102836004)(71200400001)(46003)(6506007)(6436002)(6246003)(86362001)(386003)(8936002)(81156014)(106356001)(71190400001)(81166006)(316002)(14454004)(6486002)(6916009)(186003)(68736007)(33656002)(53936002)(486006)(14444005)(99286004)(476003)(229853002)(4326008)(11346002)(8676002)(105586002)(256004)(5660300002)(6116002)(305945005)(2906002)(446003)(97736004)(478600001)(6512007)(1076003)(52116002)(54906003)(25786009)(76176011)(9686003)(7736002)(14143004);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: V5q8AZpHDQqmxMENzJ3YprUvJIC6CVEeJ7d2p77eTKDQVxV7p9aeh+Etx6ojaLGncm6MQdEb93TKQO0xmotrWuN054tiZh8ed/dYwzNDjLZ4tGcM2O417FrukQzdTxYyietOEvlbv9ykKIeMVENnQSBvuPlJFUDYeBbwhqeaRC7cCJX699mHzki6fDNdztHsDfQ4ldMPU1Kx9J1Vl0uvoAGIgl5zuB6qi2rrzvebGzrSRP8fy6UlToxQ4rBhqcnAlqHVU6OlDl73nIBwLBuWy/KoozWDtD6yRy8+g7DqZPAqpoBTXJ/qQQUdZoXzd6BaWwp21/e1ERnT+W+0SC4ZMBBNb05EDqMMFf0daX9hgSfAx2LgqU4r7AfLki/mEjUZ5BVzovFA5nvNjpMn7fLiLvLEzAv/4gCagj9DmcY9tIk=
Content-Type: text/plain; charset="us-ascii"
Content-ID: <EBA624BDEC2CAE4C9D89E7FF1151A9B7@namprd15.prod.outlook.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: 9bdeb2b8-affb-4cd2-3452-08d6b4868d17
X-MS-Exchange-CrossTenant-originalarrivaltime: 29 Mar 2019 20:38:50.9822
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 8ae927fe-1255-47a7-a2af-5f3a069daaa2
X-MS-Exchange-CrossTenant-mailboxtype: HOSTED
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BYAPR15MB2536
X-OriginatorOrg: fb.com
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:,, definitions=2019-03-29_12:,,
 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 Fri, Mar 29, 2019 at 10:46:09AM -0700, Greg Thelen wrote:
> Since commit a983b5ebee57 ("mm: memcontrol: fix excessive complexity in
> memory.stat reporting") memcg dirty and writeback counters are managed
> as:
> 1) per-memcg per-cpu values in range of [-32..32]
> 2) per-memcg atomic counter
> When a per-cpu counter cannot fit in [-32..32] it's flushed to the
> atomic.  Stat readers only check the atomic.
> Thus readers such as balance_dirty_pages() may see a nontrivial error
> margin: 32 pages per cpu.
> Assuming 100 cpus:
>    4k x86 page_size:  13 MiB error per memcg
>   64k ppc page_size: 200 MiB error per memcg
> Considering that dirty+writeback are used together for some decisions
> the errors double.
>=20
> This inaccuracy can lead to undeserved oom kills.  One nasty case is
> when all per-cpu counters hold positive values offsetting an atomic
> negative value (i.e. per_cpu[*]=3D32, atomic=3Dn_cpu*-32).
> balance_dirty_pages() only consults the atomic and does not consider
> throttling the next n_cpu*32 dirty pages.  If the file_lru is in the
> 13..200 MiB range then there's absolutely no dirty throttling, which
> burdens vmscan with only dirty+writeback pages thus resorting to oom
> kill.
>=20
> It could be argued that tiny containers are not supported, but it's more
> subtle.  It's the amount the space available for file lru that matters.
> If a container has memory.max-200MiB of non reclaimable memory, then it
> will also suffer such oom kills on a 100 cpu machine.
>=20
> The following test reliably ooms without this patch.  This patch avoids
> oom kills.
>=20
>   $ cat test
>   mount -t cgroup2 none /dev/cgroup
>   cd /dev/cgroup
>   echo +io +memory > cgroup.subtree_control
>   mkdir test
>   cd test
>   echo 10M > memory.max
>   (echo $BASHPID > cgroup.procs && exec /memcg-writeback-stress /foo)
>   (echo $BASHPID > cgroup.procs && exec dd if=3D/dev/zero of=3D/foo bs=3D=
2M count=3D100)
>=20
>   $ cat memcg-writeback-stress.c
>   /*
>    * Dirty pages from all but one cpu.
>    * Clean pages from the non dirtying cpu.
>    * This is to stress per cpu counter imbalance.
>    * On a 100 cpu machine:
>    * - per memcg per cpu dirty count is 32 pages for each of 99 cpus
>    * - per memcg atomic is -99*32 pages
>    * - thus the complete dirty limit: sum of all counters 0
>    * - balance_dirty_pages() only sees atomic count -99*32 pages, which
>    *   it max()s to 0.
>    * - So a workload can dirty -99*32 pages before balance_dirty_pages()
>    *   cares.
>    */
>   #define _GNU_SOURCE
>   #include <err.h>
>   #include <fcntl.h>
>   #include <sched.h>
>   #include <stdlib.h>
>   #include <stdio.h>
>   #include <sys/stat.h>
>   #include <sys/sysinfo.h>
>   #include <sys/types.h>
>   #include <unistd.h>
>=20
>   static char *buf;
>   static int bufSize;
>=20
>   static void set_affinity(int cpu)
>   {
>   	cpu_set_t affinity;
>=20
>   	CPU_ZERO(&affinity);
>   	CPU_SET(cpu, &affinity);
>   	if (sched_setaffinity(0, sizeof(affinity), &affinity))
>   		err(1, "sched_setaffinity");
>   }
>=20
>   static void dirty_on(int output_fd, int cpu)
>   {
>   	int i, wrote;
>=20
>   	set_affinity(cpu);
>   	for (i =3D 0; i < 32; i++) {
>   		for (wrote =3D 0; wrote < bufSize; ) {
>   			int ret =3D write(output_fd, buf+wrote, bufSize-wrote);
>   			if (ret =3D=3D -1)
>   				err(1, "write");
>   			wrote +=3D ret;
>   		}
>   	}
>   }
>=20
>   int main(int argc, char **argv)
>   {
>   	int cpu, flush_cpu =3D 1, output_fd;
>   	const char *output;
>=20
>   	if (argc !=3D 2)
>   		errx(1, "usage: output_file");
>=20
>   	output =3D argv[1];
>   	bufSize =3D getpagesize();
>   	buf =3D malloc(getpagesize());
>   	if (buf =3D=3D NULL)
>   		errx(1, "malloc failed");
>=20
>   	output_fd =3D open(output, O_CREAT|O_RDWR);
>   	if (output_fd =3D=3D -1)
>   		err(1, "open(%s)", output);
>=20
>   	for (cpu =3D 0; cpu < get_nprocs(); cpu++) {
>   		if (cpu !=3D flush_cpu)
>   			dirty_on(output_fd, cpu);
>   	}
>=20
>   	set_affinity(flush_cpu);
>   	if (fsync(output_fd))
>   		err(1, "fsync(%s)", output);
>   	if (close(output_fd))
>   		err(1, "close(%s)", output);
>   	free(buf);
>   }
>=20
> Make balance_dirty_pages() and wb_over_bg_thresh() work harder to
> collect exact per memcg counters.  This avoids the aforementioned oom
> kills.
>=20
> This does not affect the overhead of memory.stat, which still reads the
> single atomic counter.
>=20
> Why not use percpu_counter?  memcg already handles cpus going offline,
> so no need for that overhead from percpu_counter.  And the
> percpu_counter spinlocks are more heavyweight than is required.
>=20
> It probably also makes sense to use exact dirty and writeback counters
> in memcg oom reports.  But that is saved for later.
>=20
> Cc: stable@vger.kernel.org # v4.16+
> Signed-off-by: Greg Thelen <gthelen@google.com>

Hi , Greg!

Looks good to me!
Reviewed-by: Roman Gushchin <guro@fb.com>

Thanks!