From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-8.2 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_PATCH,MAILING_LIST_MULTI,SIGNED_OFF_BY,SPF_HELO_NONE,SPF_PASS, UNPARSEABLE_RELAY,USER_AGENT_SANE_1 autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id AA065C2BA19 for ; Fri, 10 Apr 2020 00:03:19 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id 6645520768 for ; Fri, 10 Apr 2020 00:03:19 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 6645520768 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=linux.alibaba.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id D7B038E0025; Thu, 9 Apr 2020 20:03:18 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id D03878E0003; Thu, 9 Apr 2020 20:03:18 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id BF25D8E0025; Thu, 9 Apr 2020 20:03:18 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0191.hostedemail.com [216.40.44.191]) by kanga.kvack.org (Postfix) with ESMTP id 9F71A8E0003 for ; Thu, 9 Apr 2020 20:03:18 -0400 (EDT) Received: from smtpin15.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay05.hostedemail.com (Postfix) with ESMTP id 57AE0181AEF39 for ; Fri, 10 Apr 2020 00:03:18 +0000 (UTC) X-FDA: 76689995676.15.fan88_911c23052df1d X-HE-Tag: fan88_911c23052df1d X-Filterd-Recvd-Size: 5712 Received: from out4436.biz.mail.alibaba.com (out4436.biz.mail.alibaba.com [47.88.44.36]) by imf06.hostedemail.com (Postfix) with ESMTP for ; Fri, 10 Apr 2020 00:03:16 +0000 (UTC) X-Alimail-AntiSpam:AC=PASS;BC=-1|-1;BR=01201311R381e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=e01f04397;MF=yang.shi@linux.alibaba.com;NM=1;PH=DS;RN=7;SR=0;TI=SMTPD_---0Tv4dqDG_1586476980; Received: from US-143344MP.local(mailfrom:yang.shi@linux.alibaba.com fp:SMTPD_---0Tv4dqDG_1586476980) by smtp.aliyun-inc.com(127.0.0.1); Fri, 10 Apr 2020 08:03:03 +0800 Subject: Re: [PATCHv2 5/8] khugepaged: Allow to callapse a page shared across fork From: Yang Shi To: "Kirill A. Shutemov" Cc: akpm@linux-foundation.org, Andrea Arcangeli , Zi Yan , linux-mm@kvack.org, linux-kernel@vger.kernel.org, "Kirill A. Shutemov" References: <20200403112928.19742-1-kirill.shutemov@linux.intel.com> <20200403112928.19742-6-kirill.shutemov@linux.intel.com> <20200408131044.xzlheacvslrbwrja@box> <107630f5-bbde-3f78-23e9-6f6b3113d709@linux.alibaba.com> Message-ID: <61569ead-2bbd-6d18-c04c-3251218a3fcd@linux.alibaba.com> Date: Thu, 9 Apr 2020 17:03:00 -0700 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.12; rv:52.0) Gecko/20100101 Thunderbird/52.7.0 MIME-Version: 1.0 In-Reply-To: <107630f5-bbde-3f78-23e9-6f6b3113d709@linux.alibaba.com> Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-US Content-Transfer-Encoding: quoted-printable X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: On 4/8/20 11:51 AM, Yang Shi wrote: > > > On 4/8/20 6:10 AM, Kirill A. Shutemov wrote: >> On Mon, Apr 06, 2020 at 01:50:56PM -0700, Yang Shi wrote: >>> >>> On 4/3/20 4:29 AM, Kirill A. Shutemov wrote: >>>> The page can be included into collapse as long as it doesn't have=20 >>>> extra >>>> pins (from GUP or otherwise). >>>> >>>> Signed-off-by: Kirill A. Shutemov >>>> --- >>>> =C2=A0=C2=A0 mm/khugepaged.c | 25 ++++++++++++++----------- >>>> =C2=A0=C2=A0 1 file changed, 14 insertions(+), 11 deletions(-) >>>> >>>> diff --git a/mm/khugepaged.c b/mm/khugepaged.c >>>> index 57ff287caf6b..1e7e6543ebca 100644 >>>> --- a/mm/khugepaged.c >>>> +++ b/mm/khugepaged.c >>>> @@ -581,11 +581,18 @@ static int=20 >>>> __collapse_huge_page_isolate(struct vm_area_struct *vma, >>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 } >>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 /* >>>> -=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 * cannot use mapco= unt: can't collapse if there's a gup pin. >>>> -=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 * The page must on= ly be referenced by the scanned process >>>> -=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 * and page swap ca= che. >>>> +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 * Check if the pag= e has any GUP (or other external) pins. >>>> +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 * >>>> +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 * The page table t= hat maps the page has been already=20 >>>> unlinked >>>> +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 * from the page ta= ble tree and this process cannot get >>>> +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 * additinal pin on= the page. >>>> +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 * >>>> +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 * New pins can com= e later if the page is shared across fork, >>>> +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 * but not for the = this process. It is fine. The other=20 >>>> process >>>> +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 * cannot write to = the page, only trigger CoW. >>>> =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 *= / >>>> -=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 if (page_count(page) !=3D= 1 + PageSwapCache(page)) { >>>> +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 if (total_mapcount(page)= + PageSwapCache(page) !=3D >>>> +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0 page_count(page)) { >>> This check looks fine for base page, but what if the page is=20 >>> PTE-mapped THP? >>> The following patch made this possible. >>> >>> If it is PTE-mapped THP and the page is in swap cache, the refcount=20 >>> would be >>> 512 + the number of PTE-mapped pages. >>> >>> Shall we do the below change in the following patch? >>> >>> extra_pins =3D PageSwapCache(page) ? nr_ccompound(page) - 1 : 0; >>> if (total_mapcount(page) + PageSwapCache(page) !=3D page_count(page) = - >>> extra_pins) { >>> ... >> Looks like you're right. >> >> It would be nice to have a test case to demonstrate the issue. >> >> Is there any way to trigger moving the page to swap cache? I don't=20 >> see it >> immediately. > > It sounds not easy to trigger since it totally depends on timing, I'm=20 > wondering we may have to use MADV_PAGEOUT? Something below off the top=20 > of my head may trigger this? > > > =C2=A0=C2=A0=C2=A0 CPU=C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0 A=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2= =A0 =C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0= CPU=C2=A0=C2=A0=C2=A0 B =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0=20 > =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 CPU=C2=A0=C2=A0= =C2=A0 C > In parent: > MADV_HUGEPAGE > page fault to fill with THP > fork > =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2= =A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0= =C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2= =A0=C2=A0 =C2=A0=C2=A0=C2=A0 In Child: > MADV_NOHUGEPAGE > MADV_DONTNEED (split pmd) > MADV_PAGEOUT > =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2= =A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0= =C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2= =A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 ->=20 > add_to_swap > =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2= =A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0= =C2=A0 khugepaged scan parent and try to=20 > collapse PTE-mapped > > =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2= =A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0= =C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2= =A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 ->=20 > try_to_unmap > > When doing MADV_DONTNEED we need make sure head page is unmapped since=20 > MADV_PAGEOUT would call page_mapcount(page) to skip shared mapping. This can't trigger it since MADV_PAGEOUT would isolate the page from lru=20 so khugepaged_scan_pmd() can't reach the refcount check at all. If try_to_unmap() or pageout() fails, the page may stay in swap cache.=20 But I didn't think of an easy way to make this happen. > >> >