Re: [RFC][PATCH 2/5] add SWAP_HAS_CACHE flag to swap_map

From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
To: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: "linux-mm@kvack.org" <linux-mm@kvack.org>,
	"balbir@linux.vnet.ibm.com" <balbir@linux.vnet.ibm.com>,
	"hugh.dickins@tiscali.co.uk" <hugh.dickins@tiscali.co.uk>,
	"hannes@cmpxchg.org" <hannes@cmpxchg.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [RFC][PATCH 2/5] add SWAP_HAS_CACHE flag to swap_map
Date: Thu, 28 May 2009 10:44:48 +0900	[thread overview]
Message-ID: <20090528104448.17c4b37c.kamezawa.hiroyu@jp.fujitsu.com> (raw)
In-Reply-To: <20090528104013.e410235e.nishimura@mxp.nes.nec.co.jp>

On Thu, 28 May 2009 10:40:13 +0900
Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> wrote:

> > > So I think it should be like:
> > > 
> > > 	read_swap_cache_async()
> > > 		:
> > > 		valid = swapcache_prepare(entry);
> > > 		if (!valid)
> > > 			break;
> > > 		if (valid == -EAGAIN);
> > > 			continue;
> > > 
> > > to let the context that succeeded in swapcache_prepare() do add_to_swap_cache().
> > > 
> > 
> > What you reccomend is code like this ?
> > 
> > ==
> > 	ret = swapcache_prapare(entry);
> > 	if (ret == -ENOENT)
> > 		break;    /* unused swap entry */
> > 	if (ret == -EBUSY)
> > 		continue; /* to call find_get_page() again */
> > ==
> > 
> Yes.
> By current version of your patch, read_swap_cache_async() might return NULL
> if concurrent read_swap_cache_async() exists. It is different from current behavior.
> And this means swapin_readahead() might fail(it calls read_swap_cache_async()
> twice, though) and can cause oom, right ?
> 
Good point. I fixed this in a patch in my box ;)
Thank you fot good review.

Thanks,
-Kame