From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43)
	id 1M1dHD-0001D7-Dc
	for qemu-devel@nongnu.org; Wed, 06 May 2009 05:18:55 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43)
	id 1M1dHB-0001CT-A5
	for qemu-devel@nongnu.org; Wed, 06 May 2009 05:18:54 -0400
Received: from [199.232.76.173] (port=52112 helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1M1dHA-0001CD-Ub
	for qemu-devel@nongnu.org; Wed, 06 May 2009 05:18:53 -0400
Received: from dd21438.kasserver.com ([85.13.141.110]:44297)
	by monty-python.gnu.org with esmtp (Exim 4.60)
	(envelope-from <martin.mohring@opensuse.org>) id 1M1dHA-0007Nv-ER
	for qemu-devel@nongnu.org; Wed, 06 May 2009 05:18:52 -0400
Message-ID: <4A0155F6.9000801@opensuse.org>
Date: Wed, 06 May 2009 11:18:46 +0200
From: Martin Mohring <martin.mohring@opensuse.org>
MIME-Version: 1.0
Subject: Re: [Qemu-devel] [PATCH] linux-user: implement pipe2 syscall
References: <20090505133048.GA29646@kos.to>	<20090505225809.GJ7574@shareable.org>
	<20090506080023.GA7230@kos.to>
In-Reply-To: <20090506080023.GA7230@kos.to>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
List-Id: qemu-devel.nongnu.org
List-Unsubscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Riku Voipio <riku.voipio@iki.fi>
Cc: qemu-devel@nongnu.org

Riku Voipio wrote:
> On Tue, May 05, 2009 at 11:58:09PM +0100, Jamie Lokier wrote:
>   
>> Riku Voipio wrote:
>>     
>>> implement pipe2 syscall. instead of calling pipe2 directly
>>> (which was introduced in 2.6.27), emulate the flag functionality
>>> with fcntl.
>>>       
>
>   
>> This is wrong with multiple threads if the flag contains FD_CLOEXEC.
>> If that situation is possible, please don't do this.
>>     
>
>   
>> The point of pipe2() with FD_CLOEXEC is to be atomic: make sure
>> another thread can never see the file descriptor with FD_CLOEXEC not set.
>>     
>
>   
How do we want to handle if "O_CLOEXEC" is not defined on the host OS,
because kernel too old (Debian Etch, CentOS 5)? Should it then return
ENOSYS?
>> If you can't guarantee that, it's better to return ENOSYS as every
>> application using pipe2() like this has a fallback to use pipe() and
>> FD_CLOEXEC itself, and probably has application logic to protect
>> against the race condition.
>>     
>
>   
>> If there's only one thread, or if you can arrange to block any
>> concurrent clone/fork/execve calls in other threads (in QEMU) during
>> the race window, then it's fine to emulate it with fcntl.
>>     
>
> We haven't returned from the pipe2 syscall when setting the flag with fcntl.
> Before returning from the syscall, the pipe file descriptors could point
> to anything (unitialized memory, zeros, ...)
>
>
>
>