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;
IronPort-SDR: JuuGMzE/j6eAuRBnDgU+wuTdg9DG4B2rG0XO9xWC9U1Wxb/xyUG3fEVzNOj1T5cB6Hh/q7kau7
 +H7KkrLwphOw==
IronPort-SDR: UvoCD/G3oFC1HVTHJ6vaE/NUBFNH4ZT3KDYInbi9H2iXGvUi85HWZ2jm74cl567fkIHr3Z/afD
 AQHd1hSqrmxg==
From:   "Williams, Dan J" <dan.j.williams@intel.com>
To:     "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "david@redhat.com" <david@redhat.com>
CC:     "akpm@linux-foundation.org" <akpm@linux-foundation.org>,
        "Huang, Ying" <ying.huang@intel.com>,
        "linux-mm@kvack.org" <linux-mm@kvack.org>,
        "richard.weiyang@gmail.com" <richard.weiyang@gmail.com>,
        "hannes@cmpxchg.org" <hannes@cmpxchg.org>,
        "minchan@kernel.org" <minchan@kernel.org>,
        "mhocko@suse.com" <mhocko@suse.com>,
        "mgorman@techsingularity.net" <mgorman@techsingularity.net>,
        "stable@vger.kernel.org" <stable@vger.kernel.org>
Subject: Re: [PATCH v2 1/3] mm/shuffle: don't move pages between zones and
 don't read garbage memmaps
Thread-Topic: [PATCH v2 1/3] mm/shuffle: don't move pages between zones and
 don't read garbage memmaps
Thread-Index: AQHWRqNlMdaKs5n8vkexxU3ANW17sA==
Date:   Sat, 20 Jun 2020 01:37:45 +0000
Message-ID: <52c25f52d25d989b54e974f7c5c1c7de1bae674f.camel@intel.com>
References: <20200619125923.22602-1-david@redhat.com>
         <20200619125923.22602-2-david@redhat.com>
In-Reply-To: <20200619125923.22602-2-david@redhat.com>
Accept-Language: en-US
Content-Language: en-US
user-agent: Evolution 3.32.5 (3.32.5-1.fc30) 
Content-Type: text/plain; charset="utf-7"
Content-ID: <7C683C8E3AA38147BEEAEE67C0A4ACCE@namprd11.prod.outlook.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: 227d9224-0083-4e53-6206-08d814ba883a
X-MS-Exchange-CrossTenant-originalarrivaltime: 20 Jun 2020 01:37:45.3066
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 46c98d88-e344-4ed4-8496-4ed7712e255d
X-MS-Exchange-CrossTenant-mailboxtype: HOSTED
X-MS-Exchange-CrossTenant-userprincipalname: UyKu58XgkJHlPaoFIIICNBBojSZNZYEHwyNRp3eOAqNr4jeYpZHb1ydqpsdbu51wOQA+4CDnLEJbCMmsD9BblIwtR672rgkXeHgHNsgpxmE=
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BN8PR11MB3780
X-OriginatorOrg: intel.com
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, 2020-06-19 at 14:59 +-0200, David Hildenbrand wrote:
+AD4- Especially with memory hotplug, we can have offline sections (with a
+AD4- garbage memmap) and overlapping zones. We have to make sure to only
+AD4- touch initialized memmaps (online sections managed by the buddy) and
+AD4- that
+AD4- the zone matches, to not move pages between zones.
+AD4-=20
+AD4- To test if this can actually happen, I added a simple
+AD4- 	BUG+AF8-ON(page+AF8-zone(page+AF8-i) +ACEAPQ- page+AF8-zone(page+AF8=
-j))+ADs-
+AD4- right before the swap. When hotplugging a 256M DIMM to a 4G x86-64 VM
+AD4- and
+AD4- onlining the first memory block +ACI-online+AF8-movable+ACI- and the =
second
+AD4- memory
+AD4- block +ACI-online+AF8-kernel+ACI-, it will trigger the BUG, as both z=
ones (NORMAL
+AD4- and MOVABLE) overlap.
+AD4-=20
+AD4- This might result in all kinds of weird situations (e.g., double
+AD4- allocations, list corruptions, unmovable allocations ending up in the
+AD4- movable zone).
+AD4-=20
+AD4- Fixes: e900a918b098 (+ACI-mm: shuffle initial free memory to improve
+AD4- memory-side-cache utilization+ACI-)
+AD4- Acked-by: Michal Hocko +ADw-mhocko+AEA-suse.com+AD4-
+AD4- Cc: stable+AEA-vger.kernel.org +ACM- v5.2+-
+AD4- Cc: Andrew Morton +ADw-akpm+AEA-linux-foundation.org+AD4-
+AD4- Cc: Johannes Weiner +ADw-hannes+AEA-cmpxchg.org+AD4-
+AD4- Cc: Michal Hocko +ADw-mhocko+AEA-suse.com+AD4-
+AD4- Cc: Minchan Kim +ADw-minchan+AEA-kernel.org+AD4-
+AD4- Cc: Huang Ying +ADw-ying.huang+AEA-intel.com+AD4-
+AD4- Cc: Wei Yang +ADw-richard.weiyang+AEA-gmail.com+AD4-
+AD4- Cc: Mel Gorman +ADw-mgorman+AEA-techsingularity.net+AD4-
+AD4- Signed-off-by: David Hildenbrand +ADw-david+AEA-redhat.com+AD4-

Looks good to me.

Acked-by: Dan Williams +ADw-dan.j.williams+AEA-intel.com+AD4-