From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753677AbbIYKcI (ORCPT ); Fri, 25 Sep 2015 06:32:08 -0400 Received: from mx2.suse.de ([195.135.220.15]:51916 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750898AbbIYKcG (ORCPT ); Fri, 25 Sep 2015 06:32:06 -0400 Subject: Re: [PATCH v2 5/9] mm/compaction: allow to scan nonmovable pageblock when depleted state To: Joonsoo Kim , Andrew Morton References: <1440382773-16070-1-git-send-email-iamjoonsoo.kim@lge.com> <1440382773-16070-6-git-send-email-iamjoonsoo.kim@lge.com> Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, Mel Gorman , Rik van Riel , David Rientjes , Minchan Kim , Joonsoo Kim From: Vlastimil Babka Message-ID: <560522A2.50609@suse.cz> Date: Fri, 25 Sep 2015 12:32:02 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.0.1 MIME-Version: 1.0 In-Reply-To: <1440382773-16070-6-git-send-email-iamjoonsoo.kim@lge.com> Content-Type: text/plain; charset=iso-8859-2 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 08/24/2015 04:19 AM, Joonsoo Kim wrote: [...] > > Because we just allow freepage scanner to scan non-movable pageblock > in very limited situation, more scanning events happen. But, allowing > in very limited situation results in a very important benefit that > memory isn't fragmented more than before. Fragmentation effect is > measured on following patch so please refer it. AFAICS it's measured only for the whole series in the cover letter, no? Just to be sure I didn't overlook something. > Signed-off-by: Joonsoo Kim > --- > include/linux/mmzone.h | 1 + > mm/compaction.c | 27 +++++++++++++++++++++++++-- > 2 files changed, 26 insertions(+), 2 deletions(-) > > diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h > index e13b732..5cae0ad 100644 > --- a/include/linux/mmzone.h > +++ b/include/linux/mmzone.h > @@ -545,6 +545,7 @@ enum zone_flags { > */ > ZONE_FAIR_DEPLETED, /* fair zone policy batch depleted */ > ZONE_COMPACTION_DEPLETED, /* compaction possiblity depleted */ > + ZONE_COMPACTION_SCANALLFREE, /* scan all kinds of pageblocks */ "SCANALLFREE" is hard to read. Otherwise yeah, I agree scanning unmovable pageblocks is necessary sometimes, and this seems to make a reasonable tradeoff.