Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751306AbbLHRNV (ORCPT ); Tue, 8 Dec 2015 12:13:21 -0500 Received: from mail-wm0-f53.google.com ([74.125.82.53]:38885 "EHLO mail-wm0-f53.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750884AbbLHRNT (ORCPT ); Tue, 8 Dec 2015 12:13:19 -0500 From: Michal Hocko To: Andrew Morton Cc: Johannes Weiner , Minchan Kim , "Kirill A. Shutemov" , Vladimir Davydov , , LKML , Michal Hocko Subject: [PATCH mmotm] memcg: Ignore partial THP when moving task Date: Tue, 8 Dec 2015 18:13:09 +0100 Message-Id: <1449594789-15866-1-git-send-email-mhocko@kernel.org> X-Mailer: git-send-email 2.6.2 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2863 Lines: 69 From: Michal Hocko After "mm: rework mapcount accounting to enable 4k mapping of THPs" it is possible to have a partial THP accessible via ptes. Memcg task migration code is not prepared for this situation and uncharges the tail page from the original memcg while the original THP is still charged via the head page which is not mapped to the moved task. The page counter of the origin memcg will underflow when the whole THP is uncharged later on and lead to: WARNING: CPU: 0 PID: 1340 at mm/page_counter.c:26 page_counter_cancel+0x34/0x40() reported by Minchan Kim. This patch prevents from the underflow by skipping any partial THP pages in mem_cgroup_move_charge_pte_range. PageTransCompound is checked when we do pte walk. This means that a process might leave a partial THP behind in the original memcg if there is no other process mapping it via pmd but this is considered acceptable because it shouldn't happen often and this is not considered a memory leak because the original THP is still accessible and reclaimable. Moreover the task migration has always been racy and never guaranteed to move all pages. Reported-by: Minchan Kim Acked-by: Johannes Weiner Signed-off-by: Michal Hocko --- Hi, this is a patch tested by Minchan in the original thread [1]. I have only replaced PageCompound with PageTransCompound because other similar fixes in mmotm used this one. The underlying implementation is the same. Johannes, I have kept your a-b but let me know if you are not OK with the changelog. This is mmotm only material. It can be merged into the page which has introduced the issue but maybe it is worth having on its own for documentation purposes. I will leave the decision to you Andrew. [1] http://lkml.kernel.org/r/20151201133455.GB27574@bbox mm/memcontrol.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 79a29d564bff..143c933f0b81 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -4895,6 +4895,14 @@ static int mem_cgroup_move_charge_pte_range(pmd_t *pmd, switch (get_mctgt_type(vma, addr, ptent, &target)) { case MC_TARGET_PAGE: page = target.page; + /* + * We can have a part of the split pmd here. Moving it + * can be done but it would be too convoluted so simply + * ignore such a partial THP and keep it in original + * memcg. There should be somebody mapping the head. + */ + if (PageCompound(page)) + goto put; if (isolate_lru_page(page)) goto put; if (!mem_cgroup_move_account(page, false, -- 2.6.2 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/