From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752948Ab1LTJbh (ORCPT ); Tue, 20 Dec 2011 04:31:37 -0500 Received: from cantor2.suse.de ([195.135.220.15]:40438 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751827Ab1LTJba (ORCPT ); Tue, 20 Dec 2011 04:31:30 -0500 Date: Tue, 20 Dec 2011 09:31:25 +0000 From: Mel Gorman To: Minchan Kim Cc: Andrew Morton , linux-mm , LKML , KOSAKI Motohiro , Rik van Riel , Johannes Weiner , Andrea Arcangeli Subject: Re: [PATCH] consider swap space when we decide compaction goes or not Message-ID: <20111220093125.GO3487@suse.de> References: <1324363653-18220-1-git-send-email-minchan@kernel.org> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline In-Reply-To: <1324363653-18220-1-git-send-email-minchan@kernel.org> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Dec 20, 2011 at 03:47:33PM +0900, Minchan Kim wrote: > It's pointless to go reclaiming when we have no swap space > and lots of anon pages in inactive list. > > Cc: KOSAKI Motohiro > Cc: Mel Gorman > Cc: Rik van Riel > Cc: Johannes Weiner > Cc: Andrea Arcangeli > Signed-off-by: Minchan Kim > --- > mm/vmscan.c | 5 +++-- > 1 files changed, 3 insertions(+), 2 deletions(-) > > diff --git a/mm/vmscan.c b/mm/vmscan.c > index 23256e8..cd5400c 100644 > --- a/mm/vmscan.c > +++ b/mm/vmscan.c > @@ -2015,8 +2015,9 @@ static inline bool should_continue_reclaim(struct zone *zone, > * inactive lists are large enough, continue reclaiming > */ > pages_for_compaction = (2UL << sc->order); > - inactive_lru_pages = zone_nr_lru_pages(zone, sc, LRU_INACTIVE_ANON) + > - zone_nr_lru_pages(zone, sc, LRU_INACTIVE_FILE); > + inactive_lru_pages = zone_nr_lru_pages(zone, sc, LRU_INACTIVE_FILE); > + if (nr_swap_pages > 0) > + inactive_lru_pages += zone_nr_lru_pages(zone, sc, LRU_INACTIVE_ANON); > if (sc->nr_reclaimed < pages_for_compaction && > inactive_lru_pages > pages_for_compaction) > return true; The changelog does not win a prize for detail but it makes sense. Without this patch, it is possible when swap is disabled to continue trying to reclaim when there are only anonymous page in the system even though that will not make any progress. Acked-by: Mel Gorman -- Mel Gorman SUSE Labs