Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752523Ab0ATJwI (ORCPT ); Wed, 20 Jan 2010 04:52:08 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752536Ab0ATJwG (ORCPT ); Wed, 20 Jan 2010 04:52:06 -0500 Received: from gir.skynet.ie ([193.1.99.77]:35528 "EHLO gir.skynet.ie" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751986Ab0ATJwB (ORCPT ); Wed, 20 Jan 2010 04:52:01 -0500 Date: Wed, 20 Jan 2010 09:51:48 +0000 From: Mel Gorman To: David Rientjes Cc: Andrea Arcangeli , Christoph Lameter , Adam Litke , Avi Kivity , linux-kernel@vger.kernel.org, linux-mm@kvack.org Subject: Re: [PATCH 7/7] Do not compact within a preferred zone after a compaction failure Message-ID: <20100120095148.GE5154@csn.ul.ie> References: <1262795169-9095-1-git-send-email-mel@csn.ul.ie> <1262795169-9095-8-git-send-email-mel@csn.ul.ie> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.17+20080114 (2008-01-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3618 Lines: 97 On Wed, Jan 13, 2010 at 03:28:16PM -0800, David Rientjes wrote: > On Wed, 6 Jan 2010, Mel Gorman wrote: > > > The fragmentation index may indicate that a failure it due to external > > fragmentation, a compaction run complete and an allocation failure still > > fail. There are two obvious reasons as to why > > > > o Page migration cannot move all pages so fragmentation remains > > o A suitable page may exist but watermarks are not met > > > > In the event of compaction and allocation failure, this patch prevents > > compaction happening for a short interval. It's only recorded on the > > preferred zone but that should be enough coverage. This could have been > > implemented similar to the zonelist_cache but the increased size of the > > zonelist did not appear to be justified. > > > > Signed-off-by: Mel Gorman > > --- > > include/linux/mmzone.h | 7 +++++++ > > mm/page_alloc.c | 15 ++++++++++++++- > > 2 files changed, 21 insertions(+), 1 deletions(-) > > > > diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h > > index 30fe668..1d6ccbe 100644 > > --- a/include/linux/mmzone.h > > +++ b/include/linux/mmzone.h > > @@ -328,6 +328,13 @@ struct zone { > > unsigned long *pageblock_flags; > > #endif /* CONFIG_SPARSEMEM */ > > > > +#ifdef CONFIG_MIGRATION > > + /* > > + * If a compaction fails, do not try compaction again until > > + * jiffies is after the value of compact_resume > > + */ > > + unsigned long compact_resume; > > +#endif > > CONFIG_COMPACTION? > Yep > > > > ZONE_PADDING(_pad1_) > > > > diff --git a/mm/page_alloc.c b/mm/page_alloc.c > > index 7275afb..9c86606 100644 > > --- a/mm/page_alloc.c > > +++ b/mm/page_alloc.c > > @@ -1729,7 +1729,7 @@ __alloc_pages_direct_reclaim(gfp_t gfp_mask, unsigned int order, > > cond_resched(); > > > > /* Try memory compaction for high-order allocations before reclaim */ > > - if (order) { > > + if (order && time_after(jiffies, preferred_zone->compact_resume)) { > > *did_some_progress = try_to_compact_pages(zonelist, > > order, gfp_mask, nodemask); > > if (*did_some_progress != COMPACT_INCOMPLETE) { > > @@ -1748,6 +1748,19 @@ __alloc_pages_direct_reclaim(gfp_t gfp_mask, unsigned int order, > > * but not enough to satisfy watermarks. > > */ > > count_vm_event(COMPACTFAIL); > > + > > + /* > > + * On failure, avoid compaction for a short time. > > + * XXX: This is very unsatisfactory. The failure > > + * to compact has nothing to do with time > > + * and everything to do with the requested > > + * order, the number of free pages and > > + * watermarks. How to wait on that is more > > + * unclear, but the answer would apply to > > + * other areas where the VM waits based on > > + * time. > > + */ > > + preferred_zone->compact_resume = jiffies + HZ/50; > > } > > } > > > > This will need to be moved to (another) inline function dependent on > CONFIG_COMPACTION since we don't have zone->compact_resume without it; > it's probably better to seperate the function out rather than add #ifdef's > within __alloc_pages_direct_reclaim(). > Moved to an inline called defer_compaction() -- Mel Gorman Part-time Phd Student Linux Technology Center University of Limerick IBM Dublin Software Lab -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/