From owner-freebsd-current@FreeBSD.ORG  Mon Apr  2 16:21:39 2012
Return-Path: <owner-freebsd-current@FreeBSD.ORG>
Delivered-To: freebsd-current@freebsd.org
Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34])
	by hub.freebsd.org (Postfix) with ESMTP id 4C4791065670
	for <freebsd-current@freebsd.org>; Mon,  2 Apr 2012 16:21:39 +0000 (UTC)
	(envelope-from alexandre.martins@netasq.com)
Received: from work.netasq.com (gateway.netasq.com [91.212.116.2])
	by mx1.freebsd.org (Postfix) with ESMTP id DED218FC0C
	for <freebsd-current@freebsd.org>; Mon,  2 Apr 2012 16:21:38 +0000 (UTC)
Received: from pc-alex.netasq.com (unknown [10.2.40.253])
	by work.netasq.com (Postfix) with ESMTPS id BADAE740019
	for <freebsd-current@freebsd.org>;
	Mon,  2 Apr 2012 18:21:27 +0200 (CEST)
From: Alexandre Martins <alexandre.martins@netasq.com>
Organization: NETASQ
To: freebsd-current@freebsd.org
Date: Mon, 2 Apr 2012 18:21:37 +0200
User-Agent: KMail/1.13.7 (FreeBSD/8.2-RELEASE; KDE/4.7.4; amd64; ; )
MIME-Version: 1.0
Content-Type: Text/Plain;
  charset="us-ascii"
Content-Transfer-Encoding: 7bit
Message-Id: <201204021821.37437.alexandre.martins@netasq.com>
Subject: Potential deadlock on mbuf
X-BeenThere: freebsd-current@freebsd.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Discussions about the use of FreeBSD-current
	<freebsd-current.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-current>, 
	<mailto:freebsd-current-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-current>
List-Post: <mailto:freebsd-current@freebsd.org>
List-Help: <mailto:freebsd-current-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-current>,
	<mailto:freebsd-current-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Mon, 02 Apr 2012 16:21:39 -0000

Dear,

I have currently having troubles with a basic socket stress.

The socket are setup to use non-blocking I/O.

During this stress-test, the kernel is running mbuf exhaustion, the goal is to 
see system limits.

If the program make a write on a socket during this mbuf exhaustion, it become 
blocked in "write" system call. The status of the process is "zonelimit" and 
whole network I/O fall in timeout.

I have found the root cause of the block  :
http://svnweb.freebsd.org/base/head/sys/kern/uipc_socket.c?view=markup#l1279

So, the question is : Why m_uiotombuf is called with a blocking parameter 
(M_WAITOK) even if is for a non-blocking socket ?

Then, if M_NOWAIT is used, maybe it will be usefull to have an 'ENOMEM' error.

Regards

-- 
Alexandre Martins