From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ed1-f69.google.com (mail-ed1-f69.google.com [209.85.208.69]) by kanga.kvack.org (Postfix) with ESMTP id 4E6908E0001 for ; Tue, 18 Dec 2018 08:58:36 -0500 (EST) Received: by mail-ed1-f69.google.com with SMTP id t2so12556781edb.22 for ; Tue, 18 Dec 2018 05:58:36 -0800 (PST) Received: from mx1.suse.de (mx2.suse.de. [195.135.220.15]) by mx.google.com with ESMTPS id h88si1747735edc.299.2018.12.18.05.58.34 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 18 Dec 2018 05:58:34 -0800 (PST) Subject: Re: [PATCH 09/14] mm, compaction: Ignore the fragmentation avoidance boost for isolation and compaction References: <20181214230310.572-1-mgorman@techsingularity.net> <20181214230310.572-10-mgorman@techsingularity.net> <20181218135156.GK29005@techsingularity.net> From: Vlastimil Babka Message-ID: Date: Tue, 18 Dec 2018 14:58:33 +0100 MIME-Version: 1.0 In-Reply-To: <20181218135156.GK29005@techsingularity.net> Content-Type: text/plain; charset=iso-8859-15 Content-Language: en-US Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org List-ID: To: Mel Gorman Cc: Linux-MM , David Rientjes , Andrea Arcangeli , Linus Torvalds , Michal Hocko , ying.huang@intel.com, kirill@shutemov.name, Andrew Morton , Linux List Kernel Mailing On 12/18/18 2:51 PM, Mel Gorman wrote: > On Tue, Dec 18, 2018 at 01:36:42PM +0100, Vlastimil Babka wrote: >> On 12/15/18 12:03 AM, Mel Gorman wrote: >>> When pageblocks get fragmented, watermarks are artifically boosted to pages >>> are reclaimed to avoid further fragmentation events. However, compaction >>> is often either fragmentation-neutral or moving movable pages away from >>> unmovable/reclaimable pages. As the actual watermarks are preserved, >>> allow compaction to ignore the boost factor. >> >> Right, I should have realized that when reviewing the boost patch. I >> think it would be useful to do the same change in >> __compaction_suitable() as well. Compaction has its own "gap". >> > > That gap is somewhat static though so I'm a bit more wary of it. However, Well, watermark boost is dynamic, but based on allocations stealing from other migratetypes, not reflecting compaction chances of success. > the check in __isolate_free_page looks too agressive. We isolate in > units of COMPACT_CLUSTER_MAX yet the watermark check there is based on > the allocation request. That means for THP that we check if 512 pages > can be allocated when only somewhere between 1 and 32 is needed for that > compaction cycle to complete. Adjusting that might be more appropriate? AFAIU the code in __isolate_free_page() reflects that if there's less than 512 free pages gap, we might form a high-order page for THP but won't be able to allocate it afterwards due to watermark.