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=-7.0 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_PATCH,MAILING_LIST_MULTI,SIGNED_OFF_BY,SPF_HELO_NONE,SPF_PASS, UNPARSEABLE_RELAY,URIBL_BLOCKED autolearn=unavailable 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 9B80AC16A69 for ; Wed, 22 May 2019 03:26:29 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 79C43217D9 for ; Wed, 22 May 2019 03:26:29 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728422AbfEVD02 (ORCPT ); Tue, 21 May 2019 23:26:28 -0400 Received: from out30-56.freemail.mail.aliyun.com ([115.124.30.56]:54643 "EHLO out30-56.freemail.mail.aliyun.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727733AbfEVD01 (ORCPT ); Tue, 21 May 2019 23:26:27 -0400 X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R141e4;CH=green;DM=||false|;FP=0|-1|-1|-1|0|-1|-1|-1;HT=e01e04400;MF=yang.shi@linux.alibaba.com;NM=1;PH=DS;RN=11;SR=0;TI=SMTPD_---0TSM5PZQ_1558495581; Received: from US-143344MP.local(mailfrom:yang.shi@linux.alibaba.com fp:SMTPD_---0TSM5PZQ_1558495581) by smtp.aliyun-inc.com(127.0.0.1); Wed, 22 May 2019 11:26:22 +0800 Subject: Re: [v3 PATCH 2/2] mm: vmscan: correct some vmscan counters for THP swapout To: "Huang, Ying" Cc: hannes@cmpxchg.org, mhocko@suse.com, mgorman@techsingularity.net, kirill.shutemov@linux.intel.com, josef@toxicpanda.com, hughd@google.com, shakeelb@google.com, akpm@linux-foundation.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org References: <1558431642-52120-1-git-send-email-yang.shi@linux.alibaba.com> <1558431642-52120-2-git-send-email-yang.shi@linux.alibaba.com> <87ftp7cmds.fsf@yhuang-dev.intel.com> From: Yang Shi Message-ID: <200a8dc9-f9cd-e039-fefe-1971271e27c3@linux.alibaba.com> Date: Wed, 22 May 2019 11:26:17 +0800 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.12; rv:52.0) Gecko/20100101 Thunderbird/52.7.0 MIME-Version: 1.0 In-Reply-To: <87ftp7cmds.fsf@yhuang-dev.intel.com> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Content-Language: en-US Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 5/22/19 9:23 AM, Huang, Ying wrote: > Yang Shi writes: > >> Since commit bd4c82c22c36 ("mm, THP, swap: delay splitting THP after >> swapped out"), THP can be swapped out in a whole. But, nr_reclaimed >> and some other vm counters still get inc'ed by one even though a whole >> THP (512 pages) gets swapped out. >> >> This doesn't make too much sense to memory reclaim. For example, direct >> reclaim may just need reclaim SWAP_CLUSTER_MAX pages, reclaiming one THP >> could fulfill it. But, if nr_reclaimed is not increased correctly, >> direct reclaim may just waste time to reclaim more pages, >> SWAP_CLUSTER_MAX * 512 pages in worst case. >> >> And, it may cause pgsteal_{kswapd|direct} is greater than >> pgscan_{kswapd|direct}, like the below: >> >> pgsteal_kswapd 122933 >> pgsteal_direct 26600225 >> pgscan_kswapd 174153 >> pgscan_direct 14678312 >> >> nr_reclaimed and nr_scanned must be fixed in parallel otherwise it would >> break some page reclaim logic, e.g. >> >> vmpressure: this looks at the scanned/reclaimed ratio so it won't >> change semantics as long as scanned & reclaimed are fixed in parallel. >> >> compaction/reclaim: compaction wants a certain number of physical pages >> freed up before going back to compacting. >> >> kswapd priority raising: kswapd raises priority if we scan fewer pages >> than the reclaim target (which itself is obviously expressed in order-0 >> pages). As a result, kswapd can falsely raise its aggressiveness even >> when it's making great progress. >> >> Other than nr_scanned and nr_reclaimed, some other counters, e.g. >> pgactivate, nr_skipped, nr_ref_keep and nr_unmap_fail need to be fixed >> too since they are user visible via cgroup, /proc/vmstat or trace >> points, otherwise they would be underreported. >> >> When isolating pages from LRUs, nr_taken has been accounted in base >> page, but nr_scanned and nr_skipped are still accounted in THP. It >> doesn't make too much sense too since this may cause trace point >> underreport the numbers as well. >> >> So accounting those counters in base page instead of accounting THP as >> one page. >> >> This change may result in lower steal/scan ratio in some cases since >> THP may get split during page reclaim, then a part of tail pages get >> reclaimed instead of the whole 512 pages, but nr_scanned is accounted >> by 512, particularly for direct reclaim. But, this should be not a >> significant issue. >> >> Cc: "Huang, Ying" >> Cc: Johannes Weiner >> Cc: Michal Hocko >> Cc: Mel Gorman >> Cc: "Kirill A . Shutemov" >> Cc: Hugh Dickins >> Cc: Shakeel Butt >> Signed-off-by: Yang Shi >> --- >> v3: Removed Shakeel's Reviewed-by since the patch has been changed significantly >> Switched back to use compound_order per Matthew >> Fixed more counters per Johannes >> v2: Added Shakeel's Reviewed-by >> Use hpage_nr_pages instead of compound_order per Huang Ying and William Kucharski >> >> mm/vmscan.c | 40 ++++++++++++++++++++++++++++------------ >> 1 file changed, 28 insertions(+), 12 deletions(-) >> >> diff --git a/mm/vmscan.c b/mm/vmscan.c >> index b65bc50..1044834 100644 >> --- a/mm/vmscan.c >> +++ b/mm/vmscan.c >> @@ -1250,7 +1250,7 @@ static unsigned long shrink_page_list(struct list_head *page_list, >> case PAGEREF_ACTIVATE: >> goto activate_locked; >> case PAGEREF_KEEP: >> - stat->nr_ref_keep++; >> + stat->nr_ref_keep += (1 << compound_order(page)); >> goto keep_locked; >> case PAGEREF_RECLAIM: >> case PAGEREF_RECLAIM_CLEAN: >> @@ -1294,6 +1294,17 @@ static unsigned long shrink_page_list(struct list_head *page_list, >> goto activate_locked; >> } >> >> + /* >> + * Account all tail pages when THP is added >> + * into swap cache successfully. >> + * The head page has been accounted at the >> + * first place. >> + */ >> + if (PageTransHuge(page)) >> + sc->nr_scanned += >> + ((1 << compound_order(page)) - >> + 1); >> + > The "if" here could be changed to "else if" because if add_to_swap() > fails we don't need to call PageTransHuge() here. But this isn't a big > deal. This could be moved to the beginning according to Johannes. > > You have analyzed the code and found that nr_dirty, nr_unqueued_dirty, > nr_congested and nr_writeback are file cache related and not impacted by > THP swap out. How about add your findings in the patch description? Yes, sure. Will add in v4. > > Best Regards, > Huang, Ying >