From: Arnd Bergmann <arnd@arndb.de>
To: Dave Hansen <dave@linux.vnet.ibm.com>
Subject: Re: [RFC][PATCH 1/4] checkpoint-restart: general infrastructure
Date: Sat, 9 Aug 2008 08:37:07 +0200
User-Agent: KMail/1.9.9
Cc: containers@lists.linux-foundation.org, Theodore Tso <tytso@mit.edu>,
       linux-kernel@vger.kernel.org, Oren Laadan <orenl@cs.columbia.edu>
References: <20080807224033.FFB3A2C1@kernel> <200808090039.20289.arnd@arndb.de> <1218242614.19082.65.camel@nimitz>
In-Reply-To: <1218242614.19082.65.camel@nimitz>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="iso-8859-15"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
Message-Id: <200808090837.07417.arnd@arndb.de>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1076
Lines: 22

On Saturday 09 August 2008, Dave Hansen wrote:
> On Sat, 2008-08-09 at 00:39 +0200, Arnd Bergmann wrote:
> > The main problem I see with that would be atomicity: If you want multiple
> > processes to keep interacting with each other, you need to save them at
> > the same point in time, which gets harder as you split your interface into
> > more than a single file descriptor.
> 
> It could take ages to write out a checkpoint even to a single fd, so I
> suspect we'd have the exact same kinds of issues either way.

I guess either way, you have to SIGSTOP (or similar) all the tasks you want
to checkpoint atomically before you start saving the contents.
If you use a single fd, you can do that under the covers, when using a 
more complex file system, it seems more logical to require an explicit
interface for this.

	Arnd <><
--
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/