From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from psmtp.com (na3sys010amx123.postini.com [74.125.245.123]) by kanga.kvack.org (Postfix) with SMTP id 53D076B005A for ; Thu, 6 Sep 2012 08:55:40 -0400 (EDT) Received: by dadi14 with SMTP id i14so1182271dad.14 for ; Thu, 06 Sep 2012 05:55:39 -0700 (PDT) Date: Thu, 6 Sep 2012 20:55:26 +0800 From: Shaohua Li Subject: Re: [patch 1/2]compaction: check migrated page number Message-ID: <20120906125526.GA1025@kernel.org> References: <20120906104404.GA12718@kernel.org> <20120906121725.GQ11266@suse.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20120906121725.GQ11266@suse.de> Sender: owner-linux-mm@kvack.org List-ID: To: Mel Gorman Cc: linux-mm@kvack.org, akpm@linux-foundation.org, aarcange@redhat.com On Thu, Sep 06, 2012 at 01:17:25PM +0100, Mel Gorman wrote: > On Thu, Sep 06, 2012 at 06:44:04PM +0800, Shaohua Li wrote: > > > > isolate_migratepages_range() might isolate none pages, for example, when > > zone->lru_lock is contended and compaction is async. In this case, we should > > abort compaction, otherwise, compact_zone will run a useless loop and make > > zone->lru_lock is even contended. > > > > It might also isolate no pages because the range was 100% allocated and > there were no free pages to isolate. This is perfectly normal and I suspect > this patch effectively disables compaction. What problem did you observe > that this patch is aimed at? I'm running a random swapin/out workload. When memory is fragmented enough, I saw 100% cpu usage. perf shows zone->lru_lock is heavily contended in isolate_migratepages_range. I'm using slub(I didn't see the problem with slab), the allocation is for radix_tree_node slab, which needs 4 pages. Even If I just apply the second patch, the system is still in 100% cpu usage. The spin_is_contended check can't cure the problem completely. Trace shows compact_zone will run a useless loop and each loop contend the lru_lock. With this patch, the cpu usage becomes normal (about 20% utilization). -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org