From mboxrd@z Thu Jan  1 00:00:00 1970
From: Jens Axboe <axboe@kernel.dk>
Subject: Re: [PATCH 6/7 v3] overlay: hybrid overlay filesystem prototype
Date: Fri, 24 Sep 2010 16:33:05 +0200
Message-ID: <4C9CB6A1.9010904@kernel.dk>
References: <20100920180404.939991832@szeredi.hu> <20100920180447.854260354@szeredi.hu> <20100922232110.GC26268@shell>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Cc: Miklos Szeredi <miklos@szeredi.hu>, linux-fsdevel@vger.kernel.org,
	linux-kernel@vger.kernel.org, neilb@suse.de,
	viro@zeniv.linux.org.uk
To: Valerie Aurora <vaurora@redhat.com>
Return-path: <linux-kernel-owner@vger.kernel.org>
In-Reply-To: <20100922232110.GC26268@shell>
Sender: linux-kernel-owner@vger.kernel.org
List-Id: linux-fsdevel.vger.kernel.org

On 2010-09-23 01:21, Valerie Aurora wrote:
> On Mon, Sep 20, 2010 at 08:04:10PM +0200, Miklos Szeredi wrote:
>> From: Miklos Szeredi <mszeredi@suse.cz>
>>
>> This overlay filesystem is a hybrid of entirely filesystem based
>> (unionfs, aufs) and entierly VFS based (union mounts) solutions.
> 
> [...]
> 
>> +static int ovl_copy_up_data(struct path *old, struct path *new, loff_t len)
>> +{
>> +	struct file *old_file;
>> +	struct file *new_file;
>> +	int error = 0;
>> +
>> +	if (len == 0)
>> +		return 0;
>> +
>> +	old_file = path_open(old, O_RDONLY);
>> +	if (IS_ERR(old_file))
>> +		return PTR_ERR(old_file);
>> +
>> +	new_file = path_open(new, O_WRONLY);
>> +	if (IS_ERR(new_file)) {
>> +		error = PTR_ERR(new_file);
>> +		goto out_fput;
>> +	}
>> +
>> +	/* FIXME: copy up sparse files efficiently */
>> +	while (len) {
>> +		loff_t offset = new_file->f_pos;
>> +		size_t this_len = OVL_COPY_UP_CHUNK_SIZE;
>> +		long bytes;
>> +
>> +		if (len < this_len)
>> +			this_len = len;
>> +
>> +		if (signal_pending_state(TASK_KILLABLE, current))
>> +			return -EINTR;
>> +
>> +		bytes = do_splice_direct(old_file, &offset, new_file, this_len,
>> +				 SPLICE_F_MOVE);
> 
> Interruptible copyup is good.  But it looks like splice setup is kind
> of heavyweight and we should do it as seldom as possible.
> 
> What about implementing splice flag SPLICE_F_INTERRUPTIBLE instead?

The pipe alloc and such? That is lazily done and sticks around.

-- 
Jens Axboe