Date: Fri, 22 Jan 2010 08:54:20 +0900
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
To: balbir@linux.vnet.ibm.com
Cc: "linux-mm@kvack.org" <linux-mm@kvack.org>,
       "akpm@linux-foundation.org" <akpm@linux-foundation.org>,
       rientjes@google.com, minchan.kim@gmail.com,
       "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] oom-kill: add lowmem usage aware oom kill handling
Message-Id: <20100122085420.38d8adea.kamezawa.hiroyu@jp.fujitsu.com>
In-Reply-To: <4B5872EC.5040603@linux.vnet.ibm.com>
References: <20100121145905.84a362bb.kamezawa.hiroyu@jp.fujitsu.com>
	<4B5872EC.5040603@linux.vnet.ibm.com>
Organization: FUJITSU Co. LTD.
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1699
Lines: 41

On Thu, 21 Jan 2010 20:59:48 +0530
Balbir Singh <balbir@linux.vnet.ibm.com> wrote:

> On Thursday 21 January 2010 11:29 AM, KAMEZAWA Hiroyuki wrote:
> > A patch for avoiding oom-serial-killer at lowmem shortage.
> > Patch is onto mmotm-2010/01/15 (depends on mm-count-lowmem-rss.patch)
> > Tested on x86-64/SMP + debug module(to allocated lowmem), works well.
> > 
> > ==
> > From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
> > 
> > One cause of OOM-Killer is memory shortage in lower zones.
> > (If memory is enough, lowmem_reserve_ratio works well. but..)
> > 
> > In lowmem-shortage oom-kill, oom-killer choses a vicitim process
> > on their vm size. But this kills a process which has lowmem memory
> > only if it's lucky. At last, there will be an oom-serial-killer.
> > 
> > Now, we have per-mm lowmem usage counter. We can make use of it
> > to select a good? victim.
> 
> Have you seen any use cases that need this change? Or is it mostly via
> code review and to utilize the availability of lowmem rss? Do we often
> run into lowmem shortage triggering OOM?
> 
  - I saw lowmem OOM killer very frequently on x86-32 box in my cusotmers.
  - I saw lowmem OOM killer somemtimes on ia64 box in my customers.

I know this helps oom handling in x86-32+Highmem environments.

For my _new_ customers, almost all devices are connected to 64bit PCI bus.
So, this is not for my customers ;) But OOM-Killer should handle this case.

Thanks,
-Kame

--
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/