Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   "Huang, Ying" <ying.huang@intel.com>
To:     Muchun Song <songmuchun@bytedance.com>
Cc:     Miaohe Lin <linmiaohe@huawei.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Mike Kravetz <mike.kravetz@oracle.com>,
        Yang Shi <shy828301@gmail.com>,
        Matthew Wilcox <willy@infradead.org>, Zi Yan <ziy@nvidia.com>,
        Minchan Kim <minchan@kernel.org>,
        Alistair Popple <apopple@nvidia.com>,
        ave.hansen@linux.intel.com, o451686892@gmail.com,
        Mina Almasry <almasrymina@google.com>,
        John Hubbard <jhubbard@nvidia.com>,
        Ralph Campbell <rcampbell@nvidia.com>,
        Peter Xu <peterx@redhat.com>,
        =?utf-8?B?SE9SSUdVQ0hJIE5BT1lBKOWggA==?=
         =?utf-8?B?5Y+jIOebtOS5nyk=?= <naoya.horiguchi@nec.com>,
        Michal Hocko <mhocko@suse.com>, riel@redhat.com,
        Linux Memory Management List <linux-mm@kvack.org>,
        LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 15/16] mm/migration: fix possible do_pages_stat_array
 racing with memory offline
References: <20220304093409.25829-1-linmiaohe@huawei.com>
        <20220304093409.25829-16-linmiaohe@huawei.com>
        <875yoq5od4.fsf@yhuang6-desk2.ccr.corp.intel.com>
        <CAMZfGtXKa73ZVLtJeE5O40rOgn7oZkxP5CB+q5t--ihUerfKew@mail.gmail.com>
Date:   Mon, 07 Mar 2022 15:42:43 +0800
In-Reply-To: <CAMZfGtXKa73ZVLtJeE5O40rOgn7oZkxP5CB+q5t--ihUerfKew@mail.gmail.com>
        (Muchun Song's message of "Mon, 7 Mar 2022 15:01:12 +0800")
Message-ID: <87r17emcng.fsf@yhuang6-desk2.ccr.corp.intel.com>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/27.1 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain; charset=ascii
Precedence: bulk

Muchun Song <songmuchun@bytedance.com> writes:

> On Mon, Mar 7, 2022 at 1:22 PM Huang, Ying <ying.huang@intel.com> wrote:
>>
>> Miaohe Lin <linmiaohe@huawei.com> writes:
>>
>> > When follow_page peeks a page, the page could be reclaimed under heavy
>> > memory pressure
>>
>> I don't think that memory pressure and reclaiming will be an issue.
>
> I think he means a page first to be reclaimed then to be offline
> could encounter this issue and reclaiming is a precondition.

I don't think reclaiming is a precondition.  It seems possible that the
virtual page is migrated, then the physical page is offlined.

Best Regards,
Huang, Ying

> Thanks.
>
>>
>> > and thus be offlined while it's still being used by the
>> > do_pages_stat_array().
>>
>> "offline" seems a possible problem.