From owner-freebsd-net@FreeBSD.ORG Wed May 19 04:41:41 2010 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 67D96106566B for ; Wed, 19 May 2010 04:41:41 +0000 (UTC) (envelope-from robin@icir.org) Received: from fruitcake.ICSI.Berkeley.EDU (fruitcake.ICSI.Berkeley.EDU [192.150.186.11]) by mx1.freebsd.org (Postfix) with ESMTP id 520D18FC1B for ; Wed, 19 May 2010 04:41:41 +0000 (UTC) Received: from empire.icsi.berkeley.edu (empire.ICSI.Berkeley.EDU [192.150.186.169]) by fruitcake.ICSI.Berkeley.EDU (8.12.11.20060614/8.12.11) with ESMTP id o4J4L0YF016270 for ; Tue, 18 May 2010 21:21:00 -0700 (PDT) Received: by empire.icsi.berkeley.edu (Postfix, from userid 502) id D181F1BAF06; Tue, 18 May 2010 21:21:00 -0700 (PDT) Date: Tue, 18 May 2010 21:21:00 -0700 From: Robin Sommer To: freebsd-net@freebsd.org Message-ID: <20100519042100.GB6390@icir.org> References: <20100514180151.GA8175@icir.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100514180151.GA8175@icir.org> User-Agent: Mutt/1.5.19 (2009-01-05) Subject: Crashes with ixgbe on FreeBSD 8 X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 19 May 2010 04:41:41 -0000 Hi all, I'm running into crashes with the ixgbe driver on a FreeBSD 8.0-RELEASE system equipped with an Intel X520-SR2 (and using a Supermicro X8DTU-F motherboard). I was wondering whether somebody might have seen this before. I updated the driver to the latest version I found on Intel's download page, which is 2.0.1. However, the moment I run tcpdump on my ix1 interface (which is connected to a switch's monitor port), I see almost immediate crashes of the kind below. Sometimes tcpdump outputs a few 10s of packets before the system crashes, but most of the time it doesn't even get that far. I saw that 8.0-STABLE comes with a more recent version of the driver. Any chance that that might fix the problem? Any other thoughts? Thanks, Robin --------- cut ------------------------------------------------------- syslog from crash: May 14 16:42:24 bart kernel: ix1: link state changed to UP May 14 16:43:34 bart kernel: ix1: promiscuous mode enabled May 14 16:43:35 bart kernel: May 14 16:43:35 bart kernel: May 14 16:43:35 bart kernel: Fatal trap 12: page fault while in kernel mode May 14 16:43:35 bart kernel: cpuid = 5; apic id = 05 May 14 16:43:35 bart kernel: fault virtual address = 0xffff80403d38d3b8 May 14 16:43:35 bart kernel: fault code = supervisor read data, page not present May 14 16:43:35 bart kernel: instruction pointer = 0x20:0xffffffff808437fe May 14 16:43:35 bart kernel: May 14 16:43:35 bart kernel: stack pointer = 0x28:0xffffff82b34d5970 May 14 16:43:35 bart kernel: frame pointer = 0x28:0xffffff82b34d5980 May 14 16:43:35 bart kernel: code segment = base 0x0, limit 0xfffff, type 0x1b May 14 16:43:35 bart kernel: = DPL 0, pres 1, long 1, def32 0, gran 1 May 14 16:43:35 bart kernel: processor eflags = interrupt enabled, Driver initalization: ix0: port 0xec00-0xec1f mem 0xf8f80000-0xf8ffffff,0xf8f7c006 ix0: Using MSIX interrupts with 33 vectors ix0: [ITHREAD] [...] ix0: Ethernet address: 00:1b:21:55:7f:d8 ix0: PCI Express Bus: Speed 5.0Gb/s Width x8 ix1: port 0xe880-0xe89f mem 0xf8e80000-0xf8efffff,0xf8e7c006 ix1: Using MSIX interrupts with 33 vectors ix1: [ITHREAD] [...] ix1: Ethernet address: 00:1b:21:55:7f:d9 ix1: PCI Express Bus: Speed 5.0Gb/s Width x8 ifconfig ix1 ix1: flags=8843 metric 0 mtu 1500 options=5bb ether 00:1b:21:55:7f:d9 media: Ethernet autoselect (10Gbase-SR ) status: active -- Robin Sommer * Phone +1 (510) 666-2886 * robin@icir.org ICSI/LBNL * Fax +1 (510) 666-2956 * www.icir.org