Subject: Re: [PATCH 1/3] mm: kswapd: Stop high-order balancing when any
 suitable zone is balanced
From: Shaohua Li <shaohua.li@intel.com>
To: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Mel Gorman <mel@csn.ul.ie>, Simon Kirby <sim@hostway.ca>,
        Dave Hansen <dave@linux.vnet.ibm.com>, linux-mm <linux-mm@kvack.org>,
        linux-kernel <linux-kernel@vger.kernel.org>
In-Reply-To: <20101201122638.ABBF.A69D9226@jp.fujitsu.com>
References: <20101201115401.ABB1.A69D9226@jp.fujitsu.com>
	 <1291173628.12777.65.camel@sli10-conroe>
	 <20101201122638.ABBF.A69D9226@jp.fujitsu.com>
Content-Type: text/plain; charset="UTF-8"
Date: Wed, 01 Dec 2010 15:40:27 +0800
Message-ID: <1291189227.12777.79.camel@sli10-conroe>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2772
Lines: 46

On Wed, 2010-12-01 at 11:28 +0800, KOSAKI Motohiro wrote:
> > On Wed, 2010-12-01 at 10:59 +0800, KOSAKI Motohiro wrote:
> > > > On Wed, 2010-12-01 at 10:23 +0800, KOSAKI Motohiro wrote:
> > > > > > On Wed, 2010-12-01 at 01:15 +0800, Mel Gorman wrote:
> > > > > > > When the allocator enters its slow path, kswapd is woken up to balance the
> > > > > > > node. It continues working until all zones within the node are balanced. For
> > > > > > > order-0 allocations, this makes perfect sense but for higher orders it can
> > > > > > > have unintended side-effects. If the zone sizes are imbalanced, kswapd
> > > > > > > may reclaim heavily on a smaller zone discarding an excessive number of
> > > > > > > pages. The user-visible behaviour is that kswapd is awake and reclaiming
> > > > > > > even though plenty of pages are free from a suitable zone.
> > > > > > > 
> > > > > > > This patch alters the "balance" logic to stop kswapd if any suitable zone
> > > > > > > becomes balanced to reduce the number of pages it reclaims from other zones.
> > > > > > from my understanding, the patch will break reclaim high zone if a low
> > > > > > zone meets the high order allocation, even the high zone doesn't meet
> > > > > > the high order allocation. This, for example, will make a high order
> > > > > > allocation from a high zone fallback to low zone and quickly exhaust low
> > > > > > zone, for example DMA. This will break some drivers.
> > > > > 
> > > > > Have you seen patch [3/3]? I think it migigate your pointed issue.
> > > > yes, it improves a lot, but still possible for small systems.
> > > 
> > > Ok, I got you. so please define your "small systems" word? 
> > an embedded system with less memory memory, obviously
> 
> Typical embedded system don't have multiple zone. It's not obvious.
IIRC, ARM supports highmem. But you are right, slub doen't allocate from
highmem.
> > > we can't make
> > > perfect VM heuristics obviously, then we need to compare pros/cons.
> > if you don't care about small system, let's consider a NORMAL i386
> > system with 896m normal zone, and 896M*3 high zone. normal zone will
> > quickly exhaust by high order high zone allocation, leave a latter
> > allocation which does need normal zone fail.
> 
> Not happen. slab don't allocate from highmem and page cache allocation
> is always using order-0. When happen high order high zone allocation?
ok, thanks, I missed this. then how about a x86_64 box with 896M DMA32
and 896*3M NORMAL? some pci devices can only dma to DMA32 zone.

--
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/