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=-16.6 required=3.0 tests=BAYES_00,DKIM_INVALID, DKIM_SIGNED,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_GIT 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 10A11C433EF for ; Thu, 16 Sep 2021 13:53:28 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id AE1B460EB4 for ; Thu, 16 Sep 2021 13:53:27 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org AE1B460EB4 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=bytedance.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=kvack.org Received: by kanga.kvack.org (Postfix) id 575F16B0074; Thu, 16 Sep 2021 09:53:27 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 524066B0082; Thu, 16 Sep 2021 09:53:27 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 39D8E900002; Thu, 16 Sep 2021 09:53:27 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0212.hostedemail.com [216.40.44.212]) by kanga.kvack.org (Postfix) with ESMTP id 275936B0074 for ; Thu, 16 Sep 2021 09:53:27 -0400 (EDT) Received: from smtpin35.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay04.hostedemail.com (Postfix) with ESMTP id DCED82FE06 for ; Thu, 16 Sep 2021 13:53:26 +0000 (UTC) X-FDA: 78593578812.35.F02D501 Received: from mail-pf1-f175.google.com (mail-pf1-f175.google.com [209.85.210.175]) by imf25.hostedemail.com (Postfix) with ESMTP id 967A7B00018B for ; Thu, 16 Sep 2021 13:53:26 +0000 (UTC) Received: by mail-pf1-f175.google.com with SMTP id y17so5900521pfl.13 for ; Thu, 16 Sep 2021 06:53:26 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=bytedance-com.20210112.gappssmtp.com; s=20210112; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=bEZAbShFT9/AGDoEdWVnP7By+WvjKrjWlaNhUM27Rf4=; b=3N+GUfr9XPjof0X7F5chysIJDn4foyS7LT7VQVGGK38nKAUyIWdPNtPVOONliIiyQg QhDI5WPPSlAe+bB82BFYrG9+s8QX0XqAl5ktF+7ou0SW6jJBfRIq1RwB/Tf5Rd5A6ehp i/2+oQ5pNkx7icbNgJEtmadfKnR9/ZYiSlRtz795E1jS6fjLntIOw3yVbjqY1sv+SZTf LNcddoncBWbBmLinB64zLewRXzvQcv6TDlb4vlywjk5AuY0QteMrGQGcUmVRF4sCmO2w HPLi6lcZAEQ572sOyF0NsFn2kXqxQCdOgamCGpSvMBRfMAecqNs/BZdK5ikHJy3dePWT K2Qw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=bEZAbShFT9/AGDoEdWVnP7By+WvjKrjWlaNhUM27Rf4=; b=Spu5leeW+L/v/TY7+iMh+Be3MDfIwhsrk1bLdotMEu7rwsEn/EI3c6t8pAbd5XoNno SdyFTco34dawWq6lif08Ox0PbnZ/1PG/GmOAlAe2K0MHmioYC7GI0fmn3GHFg+yGNC0U RFAd2gsRHXdu3XqiM9z7KJrvh+a30LyafmCZaHlIINe3SFtfdUxgr1doLr/K3uyT3B6w P7IoaHqed86WgmOIvAMrQRIB35L0k2L5R7XVI5Sf9GCd79kX4NkrjDEt00ftY5GVO6CM nBC/gGGyuQKjfIjRZ1zksF2vkP0bcZK/qmHp2r4m1D+4W+uPTos5D7NmO2r6IR6XNOUd LOug== X-Gm-Message-State: AOAM531J0+Wy+jgruMtpa3RH0izDVBons/T6DLc8eChaAQWOqVSBURUq NUohkxr8Gz3Jh5XM4HqxBAejSw== X-Google-Smtp-Source: ABdhPJyLFGvZlc7EjoFMDLGWq7+nu0XeRMDLx70MlWsHKWRfFwb/wq5DG37kKh6r7m4bAqo6W+6D9Q== X-Received: by 2002:a62:a20d:0:b029:35b:73da:dc8d with SMTP id m13-20020a62a20d0000b029035b73dadc8dmr5363626pff.54.1631800405700; Thu, 16 Sep 2021 06:53:25 -0700 (PDT) Received: from localhost.localdomain ([139.177.225.226]) by smtp.gmail.com with ESMTPSA id o9sm3617443pfh.217.2021.09.16.06.53.20 (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128); Thu, 16 Sep 2021 06:53:25 -0700 (PDT) From: Muchun Song To: guro@fb.com, hannes@cmpxchg.org, mhocko@kernel.org, akpm@linux-foundation.org, shakeelb@google.com, vdavydov.dev@gmail.com Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, duanxiongchun@bytedance.com, fam.zheng@bytedance.com, bsingharora@gmail.com, shy828301@gmail.com, alexs@kernel.org, smuchun@gmail.com, zhengqi.arch@bytedance.com, Muchun Song Subject: [PATCH v2 13/13] mm: lru: use lruvec lock to serialize memcg changes Date: Thu, 16 Sep 2021 21:47:48 +0800 Message-Id: <20210916134748.67712-14-songmuchun@bytedance.com> X-Mailer: git-send-email 2.21.0 (Apple Git-122) In-Reply-To: <20210916134748.67712-1-songmuchun@bytedance.com> References: <20210916134748.67712-1-songmuchun@bytedance.com> MIME-Version: 1.0 X-Rspamd-Server: rspam04 X-Rspamd-Queue-Id: 967A7B00018B X-Stat-Signature: tyqnzmbenks3xdgfy9g9hptd58hzaxj7 Authentication-Results: imf25.hostedemail.com; dkim=pass header.d=bytedance-com.20210112.gappssmtp.com header.s=20210112 header.b=3N+GUfr9; spf=pass (imf25.hostedemail.com: domain of songmuchun@bytedance.com designates 209.85.210.175 as permitted sender) smtp.mailfrom=songmuchun@bytedance.com; dmarc=pass (policy=none) header.from=bytedance.com X-HE-Tag: 1631800406-590148 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: As described by commit fc574c23558c ("mm/swap.c: serialize memcg changes in pagevec_lru_move_fn"), TestClearPageLRU() aims to serialize mem_cgroup_move_account() during pagevec_lru_move_fn(). Now lock_page_lruvec*() has the ability to detect whether page memcg has been changed. So we can use lruvec lock to serialize mem_cgroup_move_account() during pagevec_lru_move_fn(). This change is a partial revert of the commit fc574c23558c ("mm/swap.c: serialize memcg changes in pagevec_lru_move_fn"). And pagevec_lru_move_fn() is more hot compare with mem_cgroup_move_account(), removing an atomic operation would be an optimization. Also this change would not dirty cacheline for a page which isn't on the LRU. Signed-off-by: Muchun Song --- mm/compaction.c | 1 + mm/memcontrol.c | 31 +++++++++++++++++++++++++++++++ mm/swap.c | 41 +++++++++++------------------------------ mm/vmscan.c | 9 ++++----- 4 files changed, 47 insertions(+), 35 deletions(-) diff --git a/mm/compaction.c b/mm/compaction.c index c4ba41de8591..9e74f96f9879 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -529,6 +529,7 @@ static struct lruvec *compact_lock_page_irqsave(struc= t page *page, =20 spin_lock_irqsave(&lruvec->lru_lock, *flags); out: + /* See the comments in lock_page_lruvec(). */ if (unlikely(lruvec_memcg(lruvec) !=3D page_memcg(page))) { spin_unlock_irqrestore(&lruvec->lru_lock, *flags); goto retry; diff --git a/mm/memcontrol.c b/mm/memcontrol.c index e46fc01c6164..ab65d1c975cc 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1298,12 +1298,38 @@ struct lruvec *lock_page_lruvec(struct page *page= ) lruvec =3D mem_cgroup_page_lruvec(page); spin_lock(&lruvec->lru_lock); =20 + /* + * The memcg of the page can be changed by any the following routines: + * + * 1) mem_cgroup_move_account() or + * 2) memcg_reparent_objcgs() + * + * The possible bad scenario would like: + * + * CPU0: CPU1: CPU2: + * lruvec =3D mem_cgroup_page_lruvec() + * + * if (!isolate_lru_page()) + * mem_cgroup_move_account() + * + * memcg_reparent_objcgs() + * + * spin_lock(&lruvec->lru_lock) + * ^^^^^^ + * wrong lock + * + * Either CPU1 or CPU2 can change page memcg, so we need to check + * whether page memcg is changed, if so, we should reacquire the + * new lruvec lock. + */ if (unlikely(lruvec_memcg(lruvec) !=3D page_memcg(page))) { spin_unlock(&lruvec->lru_lock); goto retry; } =20 /* + * When we reach here, it means that the page_memcg(page) is stable. + * * Preemption is disabled in the internal of spin_lock, which can serve * as RCU read-side critical sections. */ @@ -1321,6 +1347,7 @@ struct lruvec *lock_page_lruvec_irq(struct page *pa= ge) lruvec =3D mem_cgroup_page_lruvec(page); spin_lock_irq(&lruvec->lru_lock); =20 + /* See the comments in lock_page_lruvec(). */ if (unlikely(lruvec_memcg(lruvec) !=3D page_memcg(page))) { spin_unlock_irq(&lruvec->lru_lock); goto retry; @@ -1341,6 +1368,7 @@ struct lruvec *lock_page_lruvec_irqsave(struct page= *page, unsigned long *flags) lruvec =3D mem_cgroup_page_lruvec(page); spin_lock_irqsave(&lruvec->lru_lock, *flags); =20 + /* See the comments in lock_page_lruvec(). */ if (unlikely(lruvec_memcg(lruvec) !=3D page_memcg(page))) { spin_unlock_irqrestore(&lruvec->lru_lock, *flags); goto retry; @@ -5841,7 +5869,10 @@ static int mem_cgroup_move_account(struct page *pa= ge, obj_cgroup_get(to->objcg); obj_cgroup_put(from->objcg); =20 + /* See the comments in lock_page_lruvec(). */ + spin_lock(&from_vec->lru_lock); page->memcg_data =3D (unsigned long)to->objcg; + spin_unlock(&from_vec->lru_lock); =20 __unlock_page_objcg(from->objcg); =20 diff --git a/mm/swap.c b/mm/swap.c index 18d44f978b2e..fa2352f0f9d3 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -189,14 +189,8 @@ static void pagevec_lru_move_fn(struct pagevec *pvec= , for (i =3D 0; i < pagevec_count(pvec); i++) { struct page *page =3D pvec->pages[i]; =20 - /* block memcg migration during page moving between lru */ - if (!TestClearPageLRU(page)) - continue; - lruvec =3D relock_page_lruvec_irqsave(page, lruvec, &flags); (*move_fn)(page, lruvec); - - SetPageLRU(page); } if (lruvec) unlock_page_lruvec_irqrestore(lruvec, flags); @@ -206,7 +200,7 @@ static void pagevec_lru_move_fn(struct pagevec *pvec, =20 static void pagevec_move_tail_fn(struct page *page, struct lruvec *lruve= c) { - if (!PageUnevictable(page)) { + if (PageLRU(page) && !PageUnevictable(page)) { del_page_from_lru_list(page, lruvec); ClearPageActive(page); add_page_to_lru_list_tail(page, lruvec); @@ -302,7 +296,7 @@ void lru_note_cost_page(struct page *page) =20 static void __activate_page(struct page *page, struct lruvec *lruvec) { - if (!PageActive(page) && !PageUnevictable(page)) { + if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page)) { int nr_pages =3D thp_nr_pages(page); =20 del_page_from_lru_list(page, lruvec); @@ -355,12 +349,9 @@ static void activate_page(struct page *page) struct lruvec *lruvec; =20 page =3D compound_head(page); - if (TestClearPageLRU(page)) { - lruvec =3D lock_page_lruvec_irq(page); - __activate_page(page, lruvec); - unlock_page_lruvec_irq(lruvec); - SetPageLRU(page); - } + lruvec =3D lock_page_lruvec_irq(page); + __activate_page(page, lruvec); + unlock_page_lruvec_irq(lruvec); } #endif =20 @@ -515,6 +506,9 @@ static void lru_deactivate_file_fn(struct page *page,= struct lruvec *lruvec) bool active =3D PageActive(page); int nr_pages =3D thp_nr_pages(page); =20 + if (!PageLRU(page)) + return; + if (PageUnevictable(page)) return; =20 @@ -552,7 +546,7 @@ static void lru_deactivate_file_fn(struct page *page,= struct lruvec *lruvec) =20 static void lru_deactivate_fn(struct page *page, struct lruvec *lruvec) { - if (PageActive(page) && !PageUnevictable(page)) { + if (PageLRU(page) && PageActive(page) && !PageUnevictable(page)) { int nr_pages =3D thp_nr_pages(page); =20 del_page_from_lru_list(page, lruvec); @@ -568,7 +562,7 @@ static void lru_deactivate_fn(struct page *page, stru= ct lruvec *lruvec) =20 static void lru_lazyfree_fn(struct page *page, struct lruvec *lruvec) { - if (PageAnon(page) && PageSwapBacked(page) && + if (PageLRU(page) && PageAnon(page) && PageSwapBacked(page) && !PageSwapCache(page) && !PageUnevictable(page)) { int nr_pages =3D thp_nr_pages(page); =20 @@ -1033,20 +1027,7 @@ static void __pagevec_lru_add_fn(struct page *page= , struct lruvec *lruvec) */ void __pagevec_lru_add(struct pagevec *pvec) { - int i; - struct lruvec *lruvec =3D NULL; - unsigned long flags =3D 0; - - for (i =3D 0; i < pagevec_count(pvec); i++) { - struct page *page =3D pvec->pages[i]; - - lruvec =3D relock_page_lruvec_irqsave(page, lruvec, &flags); - __pagevec_lru_add_fn(page, lruvec); - } - if (lruvec) - unlock_page_lruvec_irqrestore(lruvec, flags); - release_pages(pvec->pages, pvec->nr); - pagevec_reinit(pvec); + pagevec_lru_move_fn(pvec, __pagevec_lru_add_fn); } =20 /** diff --git a/mm/vmscan.c b/mm/vmscan.c index f38ec21babf3..e9f4a2360465 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -4672,18 +4672,17 @@ void check_move_unevictable_pages(struct pagevec = *pvec) nr_pages =3D thp_nr_pages(page); pgscanned +=3D nr_pages; =20 - /* block memcg migration during page moving between lru */ - if (!TestClearPageLRU(page)) + lruvec =3D relock_page_lruvec_irq(page, lruvec); + + if (!PageLRU(page) || !PageUnevictable(page)) continue; =20 - lruvec =3D relock_page_lruvec_irq(page, lruvec); - if (page_evictable(page) && PageUnevictable(page)) { + if (page_evictable(page)) { del_page_from_lru_list(page, lruvec); ClearPageUnevictable(page); add_page_to_lru_list(page, lruvec); pgrescued +=3D nr_pages; } - SetPageLRU(page); } =20 if (lruvec) { --=20 2.11.0