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=-2.3 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 autolearn=no 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 167CBCA9EA0 for ; Tue, 22 Oct 2019 10:02:17 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id D370E214B2 for ; Tue, 22 Oct 2019 10:02:16 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org D370E214B2 Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=profihost.ag Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id 6D1556B0003; Tue, 22 Oct 2019 06:02:16 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 6827B6B0006; Tue, 22 Oct 2019 06:02:16 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 570256B0007; Tue, 22 Oct 2019 06:02:16 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0037.hostedemail.com [216.40.44.37]) by kanga.kvack.org (Postfix) with ESMTP id 39EF16B0003 for ; Tue, 22 Oct 2019 06:02:16 -0400 (EDT) Received: from smtpin23.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay05.hostedemail.com (Postfix) with SMTP id BF36A181AF5CC for ; Tue, 22 Oct 2019 10:02:15 +0000 (UTC) X-FDA: 76070980230.23.beds52_4dd99967ecc3c X-HE-Tag: beds52_4dd99967ecc3c X-Filterd-Recvd-Size: 3601 Received: from cloud1-vm154.de-nserver.de (cloud1-vm154.de-nserver.de [178.250.10.56]) by imf09.hostedemail.com (Postfix) with ESMTP for ; Tue, 22 Oct 2019 10:02:14 +0000 (UTC) Received: (qmail 5421 invoked from network); 22 Oct 2019 12:02:13 +0200 X-Fcrdns: No Received: from phoffice.de-nserver.de (HELO [10.11.11.182]) (185.39.223.5) (smtp-auth username hostmaster@profihost.com, mechanism plain) by cloud1-vm154.de-nserver.de (qpsmtpd/0.92) with (ECDHE-RSA-AES256-GCM-SHA384 encrypted) ESMTPSA; Tue, 22 Oct 2019 12:02:13 +0200 X-GeoIP-Country: DE Subject: Re: lot of MemAvailable but falling cache and raising PSI To: Vlastimil Babka , Michal Hocko Cc: "linux-mm@kvack.org" , l.roehrs@profihost.ag, cgroups@vger.kernel.org, Johannes Weiner References: <52235eda-ffe2-721c-7ad7-575048e2d29d@profihost.ag> <20190910082919.GL2063@dhcp22.suse.cz> <132e1fd0-c392-c158-8f3a-20e340e542f0@profihost.ag> <20190910090241.GM2063@dhcp22.suse.cz> <743a047e-a46f-32fa-1fe4-a9bd8f09ed87@profihost.ag> <20190910110741.GR2063@dhcp22.suse.cz> <364d4c2e-9c9a-d8b3-43a8-aa17cccae9c7@profihost.ag> <20190910125756.GB2063@dhcp22.suse.cz> <20190910132418.GC2063@dhcp22.suse.cz> <2fe81a9e-5d29-79cf-f747-c66ae35defd0@profihost.ag> <4f6f1bc9-08f4-d53a-8788-a761be769757@suse.cz> <76ad5b29-815b-3d87-cefa-ec5b222279f1@profihost.ag> From: Stefan Priebe - Profihost AG Message-ID: <1430bb64-ef9b-f6a1-fb2c-1ca351e7950e@profihost.ag> Date: Tue, 22 Oct 2019 12:02:13 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.9.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Language: de-DE Content-Transfer-Encoding: 7bit X-User-Auth: Auth by hostmaster@profihost.com through 185.39.223.5 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: Am 22.10.19 um 09:48 schrieb Vlastimil Babka: > On 10/22/19 9:41 AM, Stefan Priebe - Profihost AG wrote: >>> Hi, could you try the patch below? I suspect you're hitting a corner >>> case where compaction_suitable() returns COMPACT_SKIPPED for the >>> ZONE_DMA, triggering reclaim even if other zones have plenty of free >>> memory. And should_continue_reclaim() then returns true until twice the >>> requested page size is reclaimed (compact_gap()). That means 4MB >>> reclaimed for each THP allocation attempt, which roughly matches the >>> trace data you preovided previously. >>> >>> The amplification to 4MB should be removed in patches merged for 5.4, so >>> it would be only 32 pages reclaimed per THP allocation. The patch below >>> tries to remove this corner case completely, and it should be more >>> visible on your 5.2.x, so please apply it there. >>> >> is there any reason to not apply that one on top of 4.19? >> >> Greets, >> Stefan >> > > It should work, cherrypicks fine without conflict here. OK but does not work ;-) mm/compaction.c: In function '__compaction_suitable': mm/compaction.c:1451:19: error: implicit declaration of function 'zone_managed_pages'; did you mean 'node_spanned_pages'? [-Werror=implicit-function-declaration] alloc_flags, zone_managed_pages(zone))) ^~~~~~~~~~~~~~~~~~ node_spanned_pages Greets, Stefan