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:     Johannes Weiner <hannes@cmpxchg.org>
CC:     Roman Gushchin <guroan@gmail.com>,
        "linux-mm@kvack.org" <linux-mm@kvack.org>,
        Kernel Team <Kernel-team@fb.com>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Tejun Heo <tj@kernel.org>, Rik van Riel <riel@surriel.com>,
        Michal Hocko <mhocko@kernel.org>
Subject: Re: [PATCH 5/5] mm: spill memcg percpu stats and events before
 releasing
Thread-Topic: [PATCH 5/5] mm: spill memcg percpu stats and events before
 releasing
Thread-Index: AQHU1TmwoulMxqBT9U+dUzzlGgn+iaYGuAGAgAAeXAA=
Date:   Mon, 11 Mar 2019 19:27:08 +0000
Message-ID: <20190311192702.GA6622@tower.DHCP.thefacebook.com>
References: <20190307230033.31975-1-guro@fb.com>
 <20190307230033.31975-6-guro@fb.com> <20190311173825.GE10823@cmpxchg.org>
In-Reply-To: <20190311173825.GE10823@cmpxchg.org>
Accept-Language: en-US
Content-Language: en-US
x-ms-exchange-messagesentrepresentingtype: 1
x-originating-ip: [2620:10d:c090:200::1:b487]
x-ms-publictraffictype: Email
x-ms-office365-filtering-correlation-id: a6654dd5-30c3-4a13-2ba5-08d6a6578d09
x-microsoft-antispam: BCL:0;PCL:0;RULEID:(2390118)(7020095)(4652040)(8989299)(4534185)(4627221)(201703031133081)(201702281549075)(8990200)(5600127)(711020)(4605104)(2017052603328)(7153060)(7193020);SRVR:BYAPR15MB2966;
x-ms-traffictypediagnostic: BYAPR15MB2966:
x-microsoft-exchange-diagnostics: 1;BYAPR15MB2966;20:uVdq8nnCJU/DKhoO42G+rN8y75eaJX0WlHNcmrNkKifUg4J73FP1juqcyt3FgnrKCFyVSglXVzB/uh9RpaWUBKStwQYFOpJKxWKdxTcbpXBRE/DytsPXQAWSKL1Fq2c+MunbAOGaccLM3O+VCc9CE8GFNltSy52oqtgnjp6f99o=
x-microsoft-antispam-prvs: <BYAPR15MB296680A644850C258224DF69BE480@BYAPR15MB2966.namprd15.prod.outlook.com>
x-forefront-prvs: 09730BD177
x-forefront-antispam-report: SFV:NSPM;SFS:(10019020)(396003)(136003)(366004)(376002)(346002)(39860400002)(189003)(199004)(25786009)(14454004)(52116002)(6486002)(102836004)(229853002)(6246003)(4326008)(105586002)(99286004)(33656002)(68736007)(14444005)(5660300002)(256004)(106356001)(86362001)(6512007)(6436002)(53936002)(54906003)(478600001)(316002)(9686003)(6116002)(386003)(305945005)(7736002)(76176011)(186003)(6506007)(46003)(446003)(81166006)(8936002)(476003)(11346002)(81156014)(6916009)(8676002)(1076003)(97736004)(71200400001)(71190400001)(2906002)(486006);DIR:OUT;SFP:1102;SCL:1;SRVR:BYAPR15MB2966;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: U1xSphB1K7QRdUxX5/itZYhwbdVyAA7iTiUs8xtlRBbRf+v6/Rb4Ee8s4wMgre2Dzs+hmuSNPtOIYsdJJBJVjlQXf9h78mXeifquMrQZpmQ+IpgVlAMhXYFwjYYAZr+kMHjAYIJRNXxU4VBAxw/XNUHRvq7Oy2KSQpdDl08Y8aV2oLHVInBPb4amR32hNl1DLBtdUyHJPZY0RARHI/Mq7DX4ckNKXAkRlrLwJHIXSnC9wpV8iJrm8gnzvCa5Teu27+JbzoBv8c7/Ao+oP/rD2hDrWj0Q8QEkxThF4wcs5O/HOc4uo9z7y5UtsLftoDPu+LZA95cIRi5hBA2AKxxA2sr0wHg2RNezZBpVCG8IqL+bLq2EelYNVT8j/dLcl1fApIUn7p+b3zlraGOM7knzIsf755S725BDRE8SrePX3GY=
Content-Type: text/plain; charset="us-ascii"
Content-ID: <A8B3421037A6B442A7D3FEE88FE09D90@namprd15.prod.outlook.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: a6654dd5-30c3-4a13-2ba5-08d6a6578d09
X-MS-Exchange-CrossTenant-originalarrivaltime: 11 Mar 2019 19:27:08.1890
 (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: BYAPR15MB2966
X-OriginatorOrg: fb.com
X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:,, definitions=2019-03-11_14:,,
 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, Mar 11, 2019 at 01:38:25PM -0400, Johannes Weiner wrote:
> On Thu, Mar 07, 2019 at 03:00:33PM -0800, Roman Gushchin wrote:
> > Spill percpu stats and events data to corresponding before releasing
> > percpu memory.
> >=20
> > Although per-cpu stats are never exactly precise, dropping them on
> > floor regularly may lead to an accumulation of an error. So, it's
> > safer to sync them before releasing.
> >=20
> > To minimize the number of atomic updates, let's sum all stats/events
> > on all cpus locally, and then make a single update per entry.
> >=20
> > Signed-off-by: Roman Gushchin <guro@fb.com>
> > ---
> >  mm/memcontrol.c | 52 +++++++++++++++++++++++++++++++++++++++++++++++++
> >  1 file changed, 52 insertions(+)
> >=20
> > diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> > index 18e863890392..b7eb6fac735e 100644
> > --- a/mm/memcontrol.c
> > +++ b/mm/memcontrol.c
> > @@ -4612,11 +4612,63 @@ static int mem_cgroup_css_online(struct cgroup_=
subsys_state *css)
> >  	return 0;
> >  }
> > =20
> > +/*
> > + * Spill all per-cpu stats and events into atomics.
> > + * Try to minimize the number of atomic writes by gathering data from
> > + * all cpus locally, and then make one atomic update.
> > + * No locking is required, because no one has an access to
> > + * the offlined percpu data.
> > + */
> > +static void mem_cgroup_spill_offlined_percpu(struct mem_cgroup *memcg)
> > +{
> > +	struct memcg_vmstats_percpu __percpu *vmstats_percpu;
> > +	struct lruvec_stat __percpu *lruvec_stat_cpu;
> > +	struct mem_cgroup_per_node *pn;
> > +	int cpu, i;
> > +	long x;
> > +
> > +	vmstats_percpu =3D memcg->vmstats_percpu_offlined;
> > +
> > +	for (i =3D 0; i < MEMCG_NR_STAT; i++) {
> > +		int nid;
> > +
> > +		x =3D 0;
> > +		for_each_possible_cpu(cpu)
> > +			x +=3D per_cpu(vmstats_percpu->stat[i], cpu);
> > +		if (x)
> > +			atomic_long_add(x, &memcg->vmstats[i]);
> > +
> > +		if (i >=3D NR_VM_NODE_STAT_ITEMS)
> > +			continue;
> > +
> > +		for_each_node(nid) {
> > +			pn =3D mem_cgroup_nodeinfo(memcg, nid);
> > +			lruvec_stat_cpu =3D pn->lruvec_stat_cpu_offlined;
> > +
> > +			x =3D 0;
> > +			for_each_possible_cpu(cpu)
> > +				x +=3D per_cpu(lruvec_stat_cpu->count[i], cpu);
> > +			if (x)
> > +				atomic_long_add(x, &pn->lruvec_stat[i]);
> > +		}
> > +	}
> > +
> > +	for (i =3D 0; i < NR_VM_EVENT_ITEMS; i++) {
> > +		x =3D 0;
> > +		for_each_possible_cpu(cpu)
> > +			x +=3D per_cpu(vmstats_percpu->events[i], cpu);
> > +		if (x)
> > +			atomic_long_add(x, &memcg->vmevents[i]);
> > +	}
>=20
> This looks good, but couldn't this be merged with the cpu offlining?
> It seems to be exactly the same code, except for the nesting of the
> for_each_possible_cpu() iteration here.
>=20
> This could be a function that takes a CPU argument and then iterates
> the cgroups and stat items to collect and spill the counters of that
> specified CPU; offlining would call it once, and this spill code here
> would call it for_each_possible_cpu().
>=20
> We shouldn't need the atomicity of this_cpu_xchg() during hotunplug,
> the scheduler isn't even active on that CPU anymore when it's called.

Good point!
I initially tried to adapt the cpu offlining code, but it didn't work
well: the code became too complex and ugly. But the opposite can be done
easily: mem_cgroup_spill_offlined_percpu() can take a cpumask,
and the cpu offlining code will look like:
	for_each_mem_cgroup(memcg)
		mem_cgroup_spill_offlined_percpu(memcg, cpumask);
I'll master a separate patch.

Thank you!