From owner-freebsd-hackers@FreeBSD.ORG Fri Nov 12 04:09:25 2004 Return-Path: Delivered-To: freebsd-hackers@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 7B98B16A4CE for ; Fri, 12 Nov 2004 04:09:25 +0000 (GMT) Received: from firetide.com (firetide.com [64.62.241.102]) by mx1.FreeBSD.org (Postfix) with SMTP id 5249C43D49 for ; Fri, 12 Nov 2004 04:09:25 +0000 (GMT) (envelope-from kmcatee@firetide.com) Received: from fbsd01 ([67.52.72.226]) by firetide.com for ; Thu, 11 Nov 2004 20:09:20 -0800 From: Kip McAtee To: freebsd-hackers@freebsd.org Date: Thu, 11 Nov 2004 18:09:20 -1000 User-Agent: KMail/1.6 MIME-Version: 1.0 Content-Disposition: inline Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Message-Id: <200411111809.20550.kmcatee@firetide.com> cc: dcornejo@firetide.com cc: hsu@freebsd.org Subject: 5.3-BETA5 m_dup() question X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 12 Nov 2004 04:09:25 -0000 Greetings- I have a symptom where an mbuf is being allocated and the (int *) ref_count is not a virtual address. I have a few new instances of calls to m_dup() that might be causing the problem downstream. Could this be because the network is allocating from a 'packet zone' and m_dup() is not? #define EXT_PACKET 3 /* came out of Packet zone */ That is, given a multicast input packet like this: mh_flags = 0x403 {MH_ext = {ext_buf = 0xc14c3000, ext_free = 0x0, ext_args = 0x0, ext_size = 0x800, ref_cnt = 0xc14b5564, ext_type = 0x3} after a call to m_dup(), I have a canonical mbuf with the MH_ext fields uninitialized. Later mb_ctor_pack() is tripping over constructing a new packet with a non-virtual address in ref_cnt. Thanks- Kip