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=-12.6 required=3.0 tests=BAYES_00,DKIM_INVALID, DKIM_SIGNED,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH,MAILING_LIST_MULTI, SIGNED_OFF_BY,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 1BAEBC388F9 for ; Sun, 8 Nov 2020 14:13:25 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id A2A2020760 for ; Sun, 8 Nov 2020 14:13:24 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=fail reason="signature verification failed" (2048-bit key) header.d=bytedance-com.20150623.gappssmtp.com header.i=@bytedance-com.20150623.gappssmtp.com header.b="Q2qHU7V4" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org A2A2020760 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=bytedance.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id 4B70E6B006C; Sun, 8 Nov 2020 09:13:24 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id 43F7C6B0072; Sun, 8 Nov 2020 09:13:24 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 2E1336B0073; Sun, 8 Nov 2020 09:13:24 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0121.hostedemail.com [216.40.44.121]) by kanga.kvack.org (Postfix) with ESMTP id 022B86B006C for ; Sun, 8 Nov 2020 09:13:23 -0500 (EST) Received: from smtpin17.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay01.hostedemail.com (Postfix) with ESMTP id ABFBD180AD801 for ; Sun, 8 Nov 2020 14:13:23 +0000 (UTC) X-FDA: 77461443486.17.shop69_0e0b9cb272e4 Received: from filter.hostedemail.com (10.5.16.251.rfc1918.com [10.5.16.251]) by smtpin17.hostedemail.com (Postfix) with ESMTP id 8D988180D0181 for ; Sun, 8 Nov 2020 14:13:23 +0000 (UTC) X-HE-Tag: shop69_0e0b9cb272e4 X-Filterd-Recvd-Size: 9267 Received: from mail-pj1-f68.google.com (mail-pj1-f68.google.com [209.85.216.68]) by imf25.hostedemail.com (Postfix) with ESMTP for ; Sun, 8 Nov 2020 14:13:22 +0000 (UTC) Received: by mail-pj1-f68.google.com with SMTP id gi3so1894489pjb.3 for ; Sun, 08 Nov 2020 06:13:22 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=bytedance-com.20150623.gappssmtp.com; s=20150623; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=rXSn3D2YKe0+X0WLHIvZOcLAPiLQlYOefZQ1A7s7AtE=; b=Q2qHU7V4lYexamo2M7caIRpE5GFV3Xh9Dk2KgSI8RavPh3p0jsS341WkT4TBpMGbtE Jzm2zAXu/zi9SY2X2TooQNmx4aZd0F/+YLyeSXgCAuoChxZdzZo4qYZjjS4y+C0bwJgw fJdiDqyEAHK8bu3R0kCfB/lu/JoHInS27ZAoq9WHbVrvbBISB3Oa3ZB6TLd9uI3Jajii tSmllNlMLCdBSQE/Lk+UmIf5b1YGGibwA+vx1tW7zOz6vypKM8QVQ2FfOYASx6+vgAF5 Xgr8764EmXA5nktDLN5X8f4dCWd0c+WME/e/gU8TlAwqgs5ogGJsQYqiVFcjnT3ySdUv 8uVg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=rXSn3D2YKe0+X0WLHIvZOcLAPiLQlYOefZQ1A7s7AtE=; b=ie2DoLBieoqRJOPE5xqW/3ds/1V8lMvhGoN9Npnd8Jj/9Xxae747q0Npe2cnU9vHOV Rf4TTgk6o52i4tmQyfwudhtCapHcxYNBTkb5Kp3GMUrsDS5LjAPB4ctFV8jBf81TxVqK Qt3Kixo3QMZfDtHo+DFn1J8rSD8rYiyv1J7g/Vd6KLmsD/epw2eXv3r9hbQInN4dA2w8 RcMgA04eeYDPo5KX8jFOVYVShUGT94ZBn0mtfxLeHDbgecgWepHsFunuq1MWKH5vWQxs MyxOjXM2QfwAXq84Uf/CVZke5b6tKt1OE8lBuy/ohqB73Wop+zD+u5H9qPBgzzp5i/J3 WJ+A== X-Gm-Message-State: AOAM530NM/egFevXjGpcpqVwVXAJ17QnTiJYeD0e754+I6rAN0/e7kEl MP+3WpYV/I6/luEpfR2KY3i2Ag== X-Google-Smtp-Source: ABdhPJxP2eZ+jU77mxWdGYFFsWcVpA5y1yjM2Ntif+Ct0/qeAd9IIoR/2NeAhAX+KQKP5LIv69d+aA== X-Received: by 2002:a17:902:8e8b:b029:d2:4276:1df0 with SMTP id bg11-20020a1709028e8bb02900d242761df0mr9025848plb.62.1604844801946; Sun, 08 Nov 2020 06:13:21 -0800 (PST) Received: from localhost.localdomain ([103.136.220.94]) by smtp.gmail.com with ESMTPSA id z11sm8754047pfk.52.2020.11.08.06.13.12 (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128); Sun, 08 Nov 2020 06:13:21 -0800 (PST) From: Muchun Song To: corbet@lwn.net, mike.kravetz@oracle.com, tglx@linutronix.de, mingo@redhat.com, bp@alien8.de, x86@kernel.org, hpa@zytor.com, dave.hansen@linux.intel.com, luto@kernel.org, peterz@infradead.org, viro@zeniv.linux.org.uk, akpm@linux-foundation.org, paulmck@kernel.org, mchehab+huawei@kernel.org, pawan.kumar.gupta@linux.intel.com, rdunlap@infradead.org, oneukum@suse.com, anshuman.khandual@arm.com, jroedel@suse.de, almasrymina@google.com, rientjes@google.com, willy@infradead.org, osalvador@suse.de, mhocko@suse.com Cc: duanxiongchun@bytedance.com, linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, Muchun Song Subject: [PATCH v3 10/21] mm/hugetlb: Defer freeing of hugetlb pages Date: Sun, 8 Nov 2020 22:11:02 +0800 Message-Id: <20201108141113.65450-11-songmuchun@bytedance.com> X-Mailer: git-send-email 2.21.0 (Apple Git-122) In-Reply-To: <20201108141113.65450-1-songmuchun@bytedance.com> References: <20201108141113.65450-1-songmuchun@bytedance.com> MIME-Version: 1.0 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: In the subsequent patch, we will allocate the vmemmap pages when free huge pages. But update_and_free_page() is be called from a non-task context(and hold hugetlb_lock), we can defer the actual freeing in a workqueue to prevent use GFP_ATOMIC to allocate the vmemmap pages. Signed-off-by: Muchun Song --- mm/hugetlb.c | 101 ++++++++++++++++++++++++++++++++++++++++++++++++++++-= ------ 1 file changed, 89 insertions(+), 12 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 27f0269aab70..ded7f0fbde35 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -1220,7 +1220,7 @@ static void destroy_compound_gigantic_page(struct p= age *page, __ClearPageHead(page); } =20 -static void free_gigantic_page(struct page *page, unsigned int order) +static void __free_gigantic_page(struct page *page, unsigned int order) { /* * If the page isn't allocated using the cma allocator, @@ -1287,11 +1287,14 @@ static struct page *alloc_gigantic_page(struct hs= tate *h, gfp_t gfp_mask, { return NULL; } -static inline void free_gigantic_page(struct page *page, unsigned int or= der) { } +static inline void __free_gigantic_page(struct page *page, + unsigned int order) { } static inline void destroy_compound_gigantic_page(struct page *page, unsigned int order) { } #endif =20 +static void __free_hugepage(struct hstate *h, struct page *page); + #ifdef CONFIG_HUGETLB_PAGE_FREE_VMEMMAP #include =20 @@ -1574,6 +1577,64 @@ static void free_huge_page_vmemmap(struct hstate *= h, struct page *head) =20 free_vmemmap_page_list(&free_pages); } + +/* + * As update_and_free_page() is be called from a non-task context(and ho= ld + * hugetlb_lock), we can defer the actual freeing in a workqueue to prev= ent + * use GFP_ATOMIC to allocate a lot of vmemmap pages. + * + * update_hpage_vmemmap_workfn() locklessly retrieves the linked list of + * pages to be freed and frees them one-by-one. As the page->mapping poi= nter + * is going to be cleared in update_hpage_vmemmap_workfn() anyway, it is + * reused as the llist_node structure of a lockless linked list of huge + * pages to be freed. + */ +static LLIST_HEAD(hpage_update_freelist); + +static void update_hpage_vmemmap_workfn(struct work_struct *work) +{ + struct llist_node *node; + struct page *page; + + node =3D llist_del_all(&hpage_update_freelist); + + while (node) { + page =3D container_of((struct address_space **)node, + struct page, mapping); + node =3D node->next; + page->mapping =3D NULL; + __free_hugepage(page_hstate(page), page); + + cond_resched(); + } +} +static DECLARE_WORK(hpage_update_work, update_hpage_vmemmap_workfn); + +static inline void __update_and_free_page(struct hstate *h, struct page = *page) +{ + /* No need to allocate vmemmap pages */ + if (!free_vmemmap_pages_per_hpage(h)) { + __free_hugepage(h, page); + return; + } + + /* + * Defer freeing to avoid using GFP_ATOMIC to allocate vmemmap + * pages. + * + * Only call schedule_work() if hpage_update_freelist is previously + * empty. Otherwise, schedule_work() had been called but the workfn + * hasn't retrieved the list yet. + */ + if (llist_add((struct llist_node *)&page->mapping, + &hpage_update_freelist)) + schedule_work(&hpage_update_work); +} + +static inline void free_gigantic_page(struct hstate *h, struct page *pag= e) +{ + __free_gigantic_page(page, huge_page_order(h)); +} #else static inline void hugetlb_vmemmap_init(struct hstate *h) { @@ -1591,17 +1652,39 @@ static inline void vmemmap_pgtable_free(struct hs= tate *h, struct page *page) static inline void free_huge_page_vmemmap(struct hstate *h, struct page = *head) { } + +static inline void __update_and_free_page(struct hstate *h, struct page = *page) +{ + __free_hugepage(h, page); +} + +static inline void free_gigantic_page(struct hstate *h, struct page *pag= e) +{ + /* + * Temporarily drop the hugetlb_lock, because + * we might block in __free_gigantic_page(). + */ + spin_unlock(&hugetlb_lock); + __free_gigantic_page(page, huge_page_order(h)); + spin_lock(&hugetlb_lock); +} #endif =20 static void update_and_free_page(struct hstate *h, struct page *page) { - int i; - if (hstate_is_gigantic(h) && !gigantic_page_runtime_supported()) return; =20 h->nr_huge_pages--; h->nr_huge_pages_node[page_to_nid(page)]--; + + __update_and_free_page(h, page); +} + +static void __free_hugepage(struct hstate *h, struct page *page) +{ + int i; + for (i =3D 0; i < pages_per_huge_page(h); i++) { page[i].flags &=3D ~(1 << PG_locked | 1 << PG_error | 1 << PG_referenced | 1 << PG_dirty | @@ -1613,14 +1696,8 @@ static void update_and_free_page(struct hstate *h,= struct page *page) set_compound_page_dtor(page, NULL_COMPOUND_DTOR); set_page_refcounted(page); if (hstate_is_gigantic(h)) { - /* - * Temporarily drop the hugetlb_lock, because - * we might block in free_gigantic_page(). - */ - spin_unlock(&hugetlb_lock); destroy_compound_gigantic_page(page, huge_page_order(h)); - free_gigantic_page(page, huge_page_order(h)); - spin_lock(&hugetlb_lock); + free_gigantic_page(h, page); } else { __free_pages(page, huge_page_order(h)); } @@ -2057,7 +2134,7 @@ static struct page *alloc_fresh_huge_page(struct hs= tate *h, =20 if (vmemmap_pgtable_prealloc(h, page)) { if (hstate_is_gigantic(h)) - free_gigantic_page(page, huge_page_order(h)); + free_gigantic_page(h, page); else put_page(page); return NULL; --=20 2.11.0