From nobody Mon Jun 22 23:59:14 2026 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 15AE0C433F5 for ; Tue, 15 Mar 2022 04:06:02 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1344494AbiCOEHL (ORCPT ); Tue, 15 Mar 2022 00:07:11 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43672 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229469AbiCOEHJ (ORCPT ); Tue, 15 Mar 2022 00:07:09 -0400 Received: from loongson.cn (mail.loongson.cn [114.242.206.163]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id D427F1CB18 for ; Mon, 14 Mar 2022 21:05:57 -0700 (PDT) Received: from localhost.localdomain (unknown [10.2.5.185]) by mail.loongson.cn (Coremail) with SMTP id AQAAf9AxSs2dEDBibo0JAA--.30070S2; Tue, 15 Mar 2022 12:05:49 +0800 (CST) From: Bibo Mao To: Andrew Morton Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, David Hildenbrand , Yang Shi Subject: [PATCH v2] mm/khugepaged: sched to numa node when collapse huge page Date: Tue, 15 Mar 2022 00:05:49 -0400 Message-Id: <20220315040549.4122396-1-maobibo@loongson.cn> X-Mailer: git-send-email 2.31.1 MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-CM-TRANSID: AQAAf9AxSs2dEDBibo0JAA--.30070S2 X-Coremail-Antispam: 1UD129KBjvJXoW7AF1DAF4DGF18Cw13CFyUJrb_yoW8Xw4fpF WDJ3yDCrWDXrykKw1Iqw1DZryrtrZ5tFWvqF15Aan2yr98Jr10ga4UZayUAFy7JrWkGFWf ArWYvrn09F48X3DanT9S1TB71UUUUUUqnTZGkaVYY2UrUUUUjbIjqfuFe4nvWSU5nxnvy2 9KBjDU0xBIdaVrnUUvcSsGvfC2KfnxnUUI43ZEXa7xR_UUUUUUUUU== X-CM-SenderInfo: xpdruxter6z05rqj20fqof0/ Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" collapse huge page will copy huge page from general small pages, dest node is calculated from most one of source pages, however THP daemon is not scheduled on dest node. The performance may be poor since huge page copying across nodes, also cache is not used for target node. With this patch, khugepaged daemon switches to the same numa node with huge page. It saves copying time and makes use of local cache better. With this patch, specint 2006 base performance is improved with 6% on Loongson 3C5000L platform with 32 cores and 8 numa nodes. Signed-off-by: Bibo Mao --- mm/khugepaged.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/mm/khugepaged.c b/mm/khugepaged.c index 131492fd1148..12d1e6a5eaa6 100644 --- a/mm/khugepaged.c +++ b/mm/khugepaged.c @@ -1066,6 +1066,7 @@ static void collapse_huge_page(struct mm_struct *mm, struct vm_area_struct *vma; struct mmu_notifier_range range; gfp_t gfp; + const struct cpumask *cpumask; =20 VM_BUG_ON(address & ~HPAGE_PMD_MASK); =20 @@ -1079,6 +1080,13 @@ static void collapse_huge_page(struct mm_struct *mm, * that. We will recheck the vma after taking it again in write mode. */ mmap_read_unlock(mm); + + /* sched to specified node before huage page memory copy */ + if (task_node(current) !=3D node) { + cpumask =3D cpumask_of_node(node); + if (unlikely(!cpumask_empty(cpumask))) + set_cpus_allowed_ptr(current, cpumask); + } new_page =3D khugepaged_alloc_page(hpage, gfp, node); if (!new_page) { result =3D SCAN_ALLOC_HUGE_PAGE_FAIL; --=20 2.31.1