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 Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 7991CC00140 for ; Fri, 12 Aug 2022 08:49:25 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S237718AbiHLItV (ORCPT ); Fri, 12 Aug 2022 04:49:21 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:34028 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S237684AbiHLItM (ORCPT ); Fri, 12 Aug 2022 04:49:12 -0400 Received: from mga04.intel.com (mga04.intel.com [192.55.52.120]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id E1EAEA9242 for ; Fri, 12 Aug 2022 01:49:09 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1660294149; x=1691830149; h=from:to:cc:subject:date:message-id:mime-version: content-transfer-encoding; bh=+nxW0bEQ9sv51ybHeEN8SKdGEjqYmcdcOKMXIfq1QOk=; b=bQMzTJW/miXpv3xHcoC4aXrfYS+qotPTts7dF79qPrMkccTd2ctDffSB 4nIRAKzuRv2tP/PUAjxqqFe0OjSHofeK3pbHr/2qZTzrl0ymPYEkBemLb /OeYZ0nv3E0xo5kPP1Yr45nIkaYnqQZzjnA1yjShhf98/Q6TuweJEPMEu WnYmjDydJ321tI+VY8yb2NkYvDXyomdXJUWpraS0Zl8R6HhEN6vD/evnw lGt4w8+GgEJR526DYhUNO9Ef1je7v5S7Df3XrZqsMW3q8XWaKuV4lGAYX 59ZerqeJtGAeLZdUVKb6eBKgaWXLqvVah3YKt9Z9/StNQtISb7kj5mLeP g==; X-IronPort-AV: E=McAfee;i="6400,9594,10436"; a="290309761" X-IronPort-AV: E=Sophos;i="5.93,231,1654585200"; d="scan'208";a="290309761" Received: from fmsmga001.fm.intel.com ([10.253.24.23]) by fmsmga104.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 12 Aug 2022 01:49:09 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.93,231,1654585200"; d="scan'208";a="748128925" Received: from unknown (HELO dsg-bmc-sherry-3.sh.intel.com) ([10.239.138.101]) by fmsmga001.fm.intel.com with ESMTP; 12 Aug 2022 01:49:07 -0700 From: Haiyue Wang To: linux-mm@kvack.org, linux-kernel@vger.kernel.org Cc: akpm@linux-foundation.org, david@redhat.com, linmiaohe@huawei.com, ying.huang@intel.com, songmuchun@bytedance.com, naoya.horiguchi@linux.dev, Haiyue Wang Subject: [PATCH v1] mm: migration: fix the FOLL_GET failure on following huge page Date: Fri, 12 Aug 2022 16:49:21 +0800 Message-Id: <20220812084921.409142-1-haiyue.wang@intel.com> X-Mailer: git-send-email 2.37.1 MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Not all huge page APIs support FOLL_GET option, so the __NR_move_pages will fail to get the page node information for huge page. This is an temporary solution to mitigate the racing fix. After supporting follow huge page by FOLL_GET is done, this fix can be reverted safely. Fixes: 4cd614841c06 ("mm: migration: fix possible do_pages_stat_array racing with memory offline") Signed-off-by: Haiyue Wang --- mm/migrate.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/mm/migrate.c b/mm/migrate.c index 6a1597c92261..581dfaad9257 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1848,6 +1848,7 @@ static void do_pages_stat_array(struct mm_struct *mm, unsigned long nr_pages, for (i = 0; i < nr_pages; i++) { unsigned long addr = (unsigned long)(*pages); + unsigned int foll_flags = FOLL_DUMP; struct vm_area_struct *vma; struct page *page; int err = -EFAULT; @@ -1856,8 +1857,12 @@ static void do_pages_stat_array(struct mm_struct *mm, unsigned long nr_pages, if (!vma) goto set_status; + /* Not all huge page follow APIs support 'FOLL_GET' */ + if (!is_vm_hugetlb_page(vma)) + foll_flags |= FOLL_GET; + /* FOLL_DUMP to ignore special (like zero) pages */ - page = follow_page(vma, addr, FOLL_GET | FOLL_DUMP); + page = follow_page(vma, addr, foll_flags); err = PTR_ERR(page); if (IS_ERR(page)) @@ -1865,7 +1870,8 @@ static void do_pages_stat_array(struct mm_struct *mm, unsigned long nr_pages, if (page && !is_zone_device_page(page)) { err = page_to_nid(page); - put_page(page); + if (foll_flags & FOLL_GET) + put_page(page); } else { err = -ENOENT; } -- 2.37.1