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:     Andrew Morton <akpm@linux-foundation.org>,
        "linux-mm@kvack.org" <linux-mm@kvack.org>
CC:     Michal Hocko <mhocko@kernel.org>,
        Johannes Weiner <hannes@cmpxchg.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Kernel Team <Kernel-team@fb.com>,
        Yafang Shao <laoar.shao@gmail.com>
Subject: Re: [PATCH] Partially revert "mm/memcontrol.c: keep local VM counters
 in sync with the hierarchical ones"
Thread-Topic: [PATCH] Partially revert "mm/memcontrol.c: keep local VM
 counters in sync with the hierarchical ones"
Thread-Index: AQHVVJVnJCTepTfXpE2m8ikuyJkTSKcJXEGA
Date:   Fri, 23 Aug 2019 22:33:01 +0000
Message-ID: <20190823223257.GA22200@tower.DHCP.thefacebook.com>
References: <20190817004726.2530670-1-guro@fb.com>
In-Reply-To: <20190817004726.2530670-1-guro@fb.com>
Accept-Language: en-US
Content-Language: en-US
x-ms-exchange-messagesentrepresentingtype: 1
x-originating-ip: [2620:10d:c090:200::1:a7ed]
x-ms-publictraffictype: Email
x-ms-office365-filtering-correlation-id: c905d369-4a39-45ea-8c19-08d72819db19
x-microsoft-antispam: BCL:0;PCL:0;RULEID:(2390118)(7020095)(4652040)(8989299)(5600166)(711020)(4605104)(1401327)(4534185)(4627221)(201703031133081)(201702281549075)(8990200)(2017052603328)(7193020);SRVR:DM6PR15MB2745;
x-ms-traffictypediagnostic: DM6PR15MB2745:
x-ms-exchange-transport-forked: True
x-microsoft-antispam-prvs: <DM6PR15MB274551CEFBEFE869E5DE9ACCBEA40@DM6PR15MB2745.namprd15.prod.outlook.com>
x-ms-oob-tlc-oobclassifiers: OLM:9508;
x-forefront-prvs: 0138CD935C
x-forefront-antispam-report: SFV:NSPM;SFS:(10019020)(979002)(136003)(396003)(346002)(366004)(39860400002)(376002)(199004)(189003)(486006)(1076003)(76176011)(71200400001)(229853002)(8936002)(81166006)(316002)(6486002)(5660300002)(99286004)(4326008)(11346002)(2501003)(6246003)(476003)(33656002)(102836004)(446003)(52116002)(66946007)(66556008)(256004)(66476007)(7736002)(305945005)(386003)(64756008)(2906002)(6512007)(9686003)(25786009)(6116002)(81156014)(186003)(110136005)(14454004)(54906003)(478600001)(46003)(53936002)(71190400001)(8676002)(6436002)(86362001)(6506007)(66446008)(969003)(989001)(999001)(1009001)(1019001);DIR:OUT;SFP:1102;SCL:1;SRVR:DM6PR15MB2745;H:DM6PR15MB2635.namprd15.prod.outlook.com;FPR:;SPF:None;LANG:en;PTR:InfoNoRecords;A:1;MX: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: dMVyv4pQNaGQyc88hF9G59a9d3p+SuMh0jQlyjAikJ5/2Tl8qPMq3i3GebJ3ydAOdB9wOiUNf9kZl6OthJu8VIWnu5hYyFuIP0QHRe/al7GEAm745ElQKYIcv61bNEyBzN1qVCCKoR9Bis4NukUzXm7hjNfyhv6mRDh0W1+qipCXfA5m0ZYzgA76v1EIjLabk7mPMjnBmnGbzxgsvVksSgvHnUxUDmKZXsbNn43sn3ZeRl6roOs6LCvHdgFD8Ji/D44px7B52ofDmVPfQCkC/DtG02ZtkXUx0Cj0XaRDc74S2RuvoybvOmoNLgRNwLDxsVBIpIIiBcXuHNxuNmprjyGzEX26Ns/yv14R5HbJ9GT9DkFVoPeUUMik9AFJgREOvSVHRMF/Y1axIoDeIFgWYGfH+LBvQolOouVXL2vwQPo=
Content-Type: text/plain; charset="us-ascii"
Content-ID: <8DAE87D3646DAB46AC49BFB584993A74@namprd15.prod.outlook.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: c905d369-4a39-45ea-8c19-08d72819db19
X-MS-Exchange-CrossTenant-originalarrivaltime: 23 Aug 2019 22:33:01.2931
 (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-CrossTenant-userprincipalname: fJdLFisKEouwkubPfC/rHJuoNJRM606PMkHM4PHcP6boNv8FdQ5cnuPl8jINPh09
X-MS-Exchange-Transport-CrossTenantHeadersStamped: DM6PR15MB2745
X-OriginatorOrg: fb.com
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:,, definitions=2019-08-23_10:,,
 signatures=0
X-Proofpoint-Spam-Details: rule=fb_default_notspam policy=fb_default score=0 priorityscore=1501
 malwarescore=0 suspectscore=0 phishscore=0 bulkscore=0 spamscore=0
 clxscore=1015 lowpriorityscore=0 mlxscore=0 impostorscore=0
 mlxlogscore=851 adultscore=0 classifier=spam adjust=0 reason=mlx
 scancount=1 engine=8.0.1-1906280000 definitions=main-1908230211
X-FB-Internal: deliver
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, Aug 16, 2019 at 05:47:26PM -0700, Roman Gushchin wrote:
> Commit 766a4c19d880 ("mm/memcontrol.c: keep local VM counters in sync
> with the hierarchical ones") effectively decreased the precision of
> per-memcg vmstats_local and per-memcg-per-node lruvec percpu counters.
>=20
> That's good for displaying in memory.stat, but brings a serious regressio=
n
> into the reclaim process.
>=20
> One issue I've discovered and debugged is the following:
> lruvec_lru_size() can return 0 instead of the actual number of pages
> in the lru list, preventing the kernel to reclaim last remaining
> pages. Result is yet another dying memory cgroups flooding.
> The opposite is also happening: scanning an empty lru list
> is the waste of cpu time.
>=20
> Also, inactive_list_is_low() can return incorrect values, preventing
> the active lru from being scanned and freed. It can fail both because
> the size of active and inactive lists are inaccurate, and because
> the number of workingset refaults isn't precise. In other words,
> the result is pretty random.
>=20
> I'm not sure, if using the approximate number of slab pages in
> count_shadow_number() is acceptable, but issues described above
> are enough to partially revert the patch.
>=20
> Let's keep per-memcg vmstat_local batched (they are only used for
> displaying stats to the userspace), but keep lruvec stats precise.
> This change fixes the dead memcg flooding on my setup.
>=20
> Fixes: 766a4c19d880 ("mm/memcontrol.c: keep local VM counters in sync wit=
h the hierarchical ones")
> Signed-off-by: Roman Gushchin <guro@fb.com>
> Cc: Yafang Shao <laoar.shao@gmail.com>
> Cc: Johannes Weiner <hannes@cmpxchg.org>

Any other concerns/comments here?

I'd prefer to fix the regression: we're likely leaking several pages
of memory for each created and destroyed memory cgroup. Plus
all internal structures, which are measured in hundreds of kb.

Thanks!