From owner-svn-src-head@FreeBSD.ORG Fri Oct 29 13:42:53 2010 Return-Path: Delivered-To: svn-src-head@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id A88BC106566C; Fri, 29 Oct 2010 13:42:53 +0000 (UTC) (envelope-from mike@sentex.net) Received: from smarthost2.sentex.ca (smarthost2-6.sentex.ca [IPv6:2607:f3e0:80:80::2]) by mx1.freebsd.org (Postfix) with ESMTP id 93DC58FC20; Fri, 29 Oct 2010 13:42:52 +0000 (UTC) Received: from lava.sentex.ca (pyroxene.sentex.ca [199.212.134.18]) by smarthost2.sentex.ca (8.14.4/8.14.4) with ESMTP id o9TDgiYQ039344 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Fri, 29 Oct 2010 09:42:44 -0400 (EDT) (envelope-from mike@sentex.net) Received: from mdt-xp.sentex.net (simeon.sentex.ca [192.168.43.27]) by lava.sentex.ca (8.14.4/8.14.4) with ESMTP id o9TDggVQ062469; Fri, 29 Oct 2010 09:42:42 -0400 (EDT) (envelope-from mike@sentex.net) Message-Id: <201010291342.o9TDggVQ062469@lava.sentex.ca> X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Fri, 29 Oct 2010 09:42:34 -0400 To: "Bjoern A. Zeeb" , Denis Ahrens From: Mike Tancsa In-Reply-To: <20101029013606.I66242@maildrop.int.zabbadoz.net> References: <0C099B90-AFB9-422F-AB75-AFC138A36D20@berlin.ccc.de> <20101029013606.I66242@maildrop.int.zabbadoz.net> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed X-Scanned-By: MIMEDefang 2.67 on 205.211.164.50 Cc: svn-src-head@FreeBSD.org, Randy Bush , Philipp =?iso-8859-1?Q?W=FCnsche?= Subject: Re: pending em0 commits for FreeBSD 8.1 STABLE X-BeenThere: svn-src-head@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SVN commit messages for the src tree for head/-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 29 Oct 2010 13:42:53 -0000 Actually, I dont think its an em bug as I can trigger this crash with the igb nic. Perhaps these are 2 different issues being hit ? Anyways, I tried with igb from HEAD in case there is some common code path between em and igb (2.0.1 vs 2.0.4). But the issue is still there In the netperf cluster its trivial to trigger this bug. Try this on camel and leopard1. Note bear1 is still on the console debugger as well in case you want to compare 2 different machines. camel1# route add -inet6 2607:f3e0:0:80::290 2607:f3e0:9::107 add host 2607:f3e0:0:80::290: gateway 2607:f3e0:9::107 camel1# Then start a ping on camel1. Then, with the ping is going, on leopard1 do an ifconfig em0 down. A few seconds later, the panic below camel1# ping6 offsite.sentex.ca PING6(56=40+8+8 bytes) 2607:f3e0:9::113 --> 2607:f3e0:0:80::290 16 bytes from 2607:f3e0:0:80::290, icmp_seq=0 hlim=60 time=3.988 ms 16 bytes from 2607:f3e0:0:80::290, icmp_seq=1 hlim=60 time=3.321 ms 16 bytes from 2607:f3e0:0:80::290, icmp_seq=2 hlim=60 time=3.834 ms 16 bytes from 2607:f3e0:0:80::290, icmp_seq=3 hlim=60 time=3.206 ms 16 bytes from 2607:f3e0:0:80::290, icmp_seq=4 hlim=60 time=3.076 ms 16 bytes from 2607:f3e0:0:80::290, icmp_seq=5 hlim=60 time=5.182 ms 16 bytes from 2607:f3e0:0:80::290, icmp_seq=6 hlim=60 time=3.437 ms 16 bytes from 2607:f3e0:0:80::290, icmp_seq=7 hlim=60 time=3.308 ms 16 bytes from 2607:f3e0:0:80::290, icmp_seq=8 hlim=60 time=3.555 ms 16 bytes from 2607:f3e0:0:80::290, icmp_seq=9 hlim=60 time=2.929 ms 16 bytes from 2607:f3e0:0:80::290, icmp_seq=10 hlim=60 time=3.903 ms 16 bytes from 2607:f3e0:0:80::290, icmp_seq=11 hlim=60 time=5.421 ms Fatal trap 9: general protection fault while in kernel mode cpuid = 0; apic id = 00 instruction pointer = 0x20:0xffffffff808950c6 stack pointer = 0x28:0xffffff800006e950 frame pointer = 0x28:0xffffff800006e9a0 code segment = base 0x0, limit 0xfffff, type 0x1b = DPL 0, pres 1, long 1, def32 0, gran 1 processor eflags = interrupt enabled, resume, IOPL = 0 current process = 12 (swi1: netisr 0) [thread pid 12 tid 100011 ] Stopped at bcopy+0x16: repe movsq (%rsi),%es:(%rdi) db> bt Tracing pid 12 tid 100011 td 0xffffff00024fd3e0 bcopy() at bcopy+0x16 ip6_input() at ip6_input+0x28f swi_net() at swi_net+0x151 intr_event_execute_handlers() at intr_event_execute_handlers+0x66 ithread_loop() at ithread_loop+0xb2 fork_exit() at fork_exit+0x12a fork_trampoline() at fork_trampoline+0xe --- trap 0, rip = 0, rsp = 0xffffff800006ed30, rbp = 0 --- db> and using the igb driver from HEAD Fatal trap 9: general protection fault while in kernel mode cpuid = 2; apic id = 06 instruction pointer = 0x20:0xffffffff80881506 stack pointer = 0x28:0xffffff800006e950 frame pointer = 0x28:0xffffff800006e9a0 code segment = base 0x0, limit 0xfffff, type 0x1b = DPL 0, pres 1, long 1, def32 0, gran 1 processor eflags = interrupt enabled, resume, IOPL = 0 current process = 12 (swi1: netisr 0) [thread pid 12 tid 100011 ] Stopped at bcopy+0x16: repe movsq (%rsi),%es:(%rdi) db> bt Tracing pid 12 tid 100011 td 0xffffff00024fd3e0 bcopy() at bcopy+0x16 ip6_input() at ip6_input+0x28f swi_net() at swi_net+0x151 intr_event_execute_handlers() at intr_event_execute_handlers+0x66 ithread_loop() at ithread_loop+0xb2 fork_exit() at fork_exit+0x12a fork_trampoline() at fork_trampoline+0xe --- trap 0, rip = 0, rsp = 0xffffff800006ed30, rbp = 0 --- db> -------------------------------------------------------------------- Mike Tancsa, tel +1 519 651 3400 Sentex Communications, mike@sentex.net Providing Internet since 1994 www.sentex.net Cambridge, Ontario Canada www.sentex.net/mike