From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S964798AbdEOLZw (ORCPT ); Mon, 15 May 2017 07:25:52 -0400 Received: from mga02.intel.com ([134.134.136.20]:41273 "EHLO mga02.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934440AbdEOLZm (ORCPT ); Mon, 15 May 2017 07:25:42 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.38,344,1491289200"; d="scan'208";a="101694528" From: "Huang, Ying" To: Andrew Morton Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, Huang Ying , Johannes Weiner Subject: [PATCH -mm -v11 5/5] mm, THP, swap: Enable THP swap optimization only if has compound map Date: Mon, 15 May 2017 19:25:22 +0800 Message-Id: <20170515112522.32457-6-ying.huang@intel.com> X-Mailer: git-send-email 2.11.0 In-Reply-To: <20170515112522.32457-1-ying.huang@intel.com> References: <20170515112522.32457-1-ying.huang@intel.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Huang Ying If there is no compound map for a THP (Transparent Huge Page), it is possible that the map count of some sub-pages of the THP is 0. So it is better to split the THP before swapping out. In this way, the sub-pages not mapped will be freed, and we can avoid the unnecessary swap out operations for these sub-pages. Cc: Johannes Weiner Signed-off-by: "Huang, Ying" --- mm/vmscan.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/mm/vmscan.c b/mm/vmscan.c index a5355022dc2f..f7e949ac9756 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1125,10 +1125,19 @@ static unsigned long shrink_page_list(struct list_head *page_list, !PageSwapCache(page)) { if (!(sc->gfp_mask & __GFP_IO)) goto keep_locked; - /* cannot split THP, skip it */ - if (PageTransHuge(page) && - !can_split_huge_page(page, NULL)) - goto activate_locked; + if (PageTransHuge(page)) { + /* cannot split THP, skip it */ + if (!can_split_huge_page(page, NULL)) + goto activate_locked; + /* + * Split pages without a PMD map right + * away. Chances are some or all of the + * tail pages can be freed without IO. + */ + if (!compound_mapcount(page) && + split_huge_page_to_list(page, page_list)) + goto activate_locked; + } if (!add_to_swap(page)) { if (!PageTransHuge(page)) goto activate_locked; -- 2.11.0