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=-5.5 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,NICE_REPLY_A,SPF_HELO_NONE, SPF_PASS,UNPARSEABLE_RELAY,USER_AGENT_SANE_1 autolearn=no 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 BD401C433E1 for ; Tue, 21 Jul 2020 13:51:34 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id 8C072206E9 for ; Tue, 21 Jul 2020 13:51:34 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 8C072206E9 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 0B86C6B0002; Tue, 21 Jul 2020 09:51:34 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 069646B0003; Tue, 21 Jul 2020 09:51:34 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id EA74E6B0005; Tue, 21 Jul 2020 09:51:33 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0022.hostedemail.com [216.40.44.22]) by kanga.kvack.org (Postfix) with ESMTP id D27486B0002 for ; Tue, 21 Jul 2020 09:51:33 -0400 (EDT) Received: from smtpin07.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay02.hostedemail.com (Postfix) with ESMTP id 39BB53F10F96 for ; Tue, 21 Jul 2020 13:51:33 +0000 (UTC) X-FDA: 77062220466.07.loaf60_3b0fbd026f2d Received: from filter.hostedemail.com (10.5.16.251.rfc1918.com [10.5.16.251]) by smtpin07.hostedemail.com (Postfix) with ESMTP id 9897F180364CF for ; Tue, 21 Jul 2020 13:51:32 +0000 (UTC) X-HE-Tag: loaf60_3b0fbd026f2d X-Filterd-Recvd-Size: 8219 Received: from out4436.biz.mail.alibaba.com (out4436.biz.mail.alibaba.com [47.88.44.36]) by imf21.hostedemail.com (Postfix) with ESMTP for ; Tue, 21 Jul 2020 13:51:31 +0000 (UTC) X-Alimail-AntiSpam:AC=PASS;BC=-1|-1;BR=01201311R161e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=e01e01422;MF=alex.shi@linux.alibaba.com;NM=1;PH=DS;RN=18;SR=0;TI=SMTPD_---0U3PtznR_1595339485; Received: from IT-FVFX43SYHV2H.lan(mailfrom:alex.shi@linux.alibaba.com fp:SMTPD_---0U3PtznR_1595339485) by smtp.aliyun-inc.com(127.0.0.1); Tue, 21 Jul 2020 21:51:25 +0800 Subject: Re: [PATCH v16 16/22] mm/mlock: reorder isolation sequence during munlock From: Alex Shi To: Alexander Duyck , Johannes Weiner Cc: Andrew Morton , Mel Gorman , Tejun Heo , Hugh Dickins , Konstantin Khlebnikov , Daniel Jordan , Yang Shi , Matthew Wilcox , kbuild test robot , linux-mm , LKML , cgroups@vger.kernel.org, Shakeel Butt , Joonsoo Kim , Wei Yang , "Kirill A. Shutemov" References: <1594429136-20002-1-git-send-email-alex.shi@linux.alibaba.com> <1594429136-20002-17-git-send-email-alex.shi@linux.alibaba.com> <6e37ee32-c6c5-fcc5-3cad-74f7ae41fb67@linux.alibaba.com> <7a931661-e096-29ee-d97d-8bf96ba6c972@linux.alibaba.com> Message-ID: Date: Tue, 21 Jul 2020 21:51:24 +0800 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:68.0) Gecko/20100101 Thunderbird/68.7.0 MIME-Version: 1.0 In-Reply-To: <7a931661-e096-29ee-d97d-8bf96ba6c972@linux.alibaba.com> Content-Type: text/plain; charset=utf-8 X-Rspamd-Queue-Id: 9897F180364CF X-Spamd-Result: default: False [0.00 / 100.00] X-Rspamd-Server: rspam01 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: =E5=9C=A8 2020/7/21 =E4=B8=8B=E5=8D=885:26, Alex Shi =E5=86=99=E9=81=93: >=20 >=20 > =E5=9C=A8 2020/7/21 =E4=B8=8A=E5=8D=882:51, Alexander Duyck =E5=86=99=E9= =81=93: >>> Look into the __split_huge_page_tail, there is a tiny gap between tai= l page >>> get PG_mlocked, and it is added into lru list. >>> The TestClearPageLRU could blocked memcg changes of the page from sto= pping >>> isolate_lru_page. >> I get that there is a gap between the two in __split_huge_page_tail. >> My concern is more the fact that you are pulling the bit testing >> outside of the locked region when I don't think it needs to be. The >> lock is being taken unconditionally, so why pull the testing out when >> you could just do it inside the lock anyway? My worry is that you >> might be addressing __split_huge_page_tail but in the process you >> might be introducing a new race with something like >> __pagevec_lru_add_fn. >=20 > Yes, the page maybe interfered by clear_page_mlock and add pages to wro= ng lru > list. >=20 >> >> If I am not mistaken the Mlocked flag can still be cleared regardless >> of if the LRU bit is set or not. So you can still clear the LRU bit >> before you pull the page out of the list, but it can be done after >> clearing the Mlocked flag instead of before you have even taken the >> LRU lock. In that way it would function more similar to how you >> handled pagevec_lru_move_fn() as all this function is really doing is >> moving the pages out of the unevictable list into one of the other LRU >> lists anyway since the Mlocked flag was cleared. >> >=20 > Without the lru bit guard, the page may be moved between memcgs, luckly= , > lock_page would stop the mem_cgroup_move_account with BUSY state cost. > whole new change would like the following, I will testing/resend again. >=20 Hi Johannes, It looks like lock_page_memcg() could be used to replace lock_page(), whi= ch could change retry into spinlock wait. Would you like to give some commen= ts? Thank Alex > Thanks! > Alex >=20 > @@ -182,7 +179,7 @@ static void __munlock_isolation_failed(struct page = *page) > unsigned int munlock_vma_page(struct page *page) > { > int nr_pages; > - pg_data_t *pgdat =3D page_pgdat(page); > + struct lruvec *lruvec; >=20 > /* For try_to_munlock() and to serialize with page migration */ > BUG_ON(!PageLocked(page)); > @@ -190,11 +187,11 @@ unsigned int munlock_vma_page(struct page *page) > VM_BUG_ON_PAGE(PageTail(page), page); >=20 > /* > - * Serialize with any parallel __split_huge_page_refcount() whi= ch > + * Serialize split tail pages in __split_huge_page_tail() which > * might otherwise copy PageMlocked to part of the tail pages b= efore > * we clear it in the head page. It also stabilizes hpage_nr_pa= ges(). > */ > - spin_lock_irq(&pgdat->lru_lock); > + lruvec =3D lock_page_lruvec_irq(page); >=20 > if (!TestClearPageMlocked(page)) { > /* Potentially, PTE-mapped THP: do not skip the rest PT= Es */ > @@ -205,15 +202,15 @@ unsigned int munlock_vma_page(struct page *page) > nr_pages =3D hpage_nr_pages(page); > __mod_zone_page_state(page_zone(page), NR_MLOCK, -nr_pages); >=20 > - if (__munlock_isolate_lru_page(page, true)) { > - spin_unlock_irq(&pgdat->lru_lock); > + if (__munlock_isolate_lru_page(page, lruvec, true)) { > + unlock_page_lruvec_irq(lruvec); > __munlock_isolated_page(page); > goto out; > } > __munlock_isolation_failed(page); >=20 > unlock_out: > - spin_unlock_irq(&pgdat->lru_lock); > + unlock_page_lruvec_irq(lruvec); >=20 > out: > return nr_pages - 1; > @@ -293,23 +290,27 @@ static void __munlock_pagevec(struct pagevec *pve= c, struct zone *zone) > int nr =3D pagevec_count(pvec); > int delta_munlocked =3D -nr; > struct pagevec pvec_putback; > + struct lruvec *lruvec =3D NULL; > int pgrescued =3D 0; >=20 > pagevec_init(&pvec_putback); >=20 > /* Phase 1: page isolation */ > - spin_lock_irq(&zone->zone_pgdat->lru_lock); > for (i =3D 0; i < nr; i++) { > struct page *page =3D pvec->pages[i]; >=20 > + /* block memcg change in mem_cgroup_move_account */ > + lock_page(page); > + lruvec =3D relock_page_lruvec_irq(page, lruvec); > if (TestClearPageMlocked(page)) { > /* > * We already have pin from follow_page_mask() > * so we can spare the get_page() here. > */ > - if (__munlock_isolate_lru_page(page, false)) > + if (__munlock_isolate_lru_page(page, lruvec, fa= lse)) { > + unlock_page(page); > continue; > - else > + } else > __munlock_isolation_failed(page); > } else { > delta_munlocked++; > @@ -321,11 +322,14 @@ static void __munlock_pagevec(struct pagevec *pve= c, struct zone *zone) > * pin. We cannot do it under lru_lock however. If it's > * the last pin, __page_cache_release() would deadlock. > */ > + unlock_page(page); > pagevec_add(&pvec_putback, pvec->pages[i]); > pvec->pages[i] =3D NULL; > } > - __mod_zone_page_state(zone, NR_MLOCK, delta_munlocked); > - spin_unlock_irq(&zone->zone_pgdat->lru_lock); > + if (lruvec) { > + __mod_zone_page_state(zone, NR_MLOCK, delta_munlocked); > + unlock_page_lruvec_irq(lruvec); > + } >=20 > /* Now we can release pins of pages that we are not munlocking = */ > pagevec_release(&pvec_putback); >=20