[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20080811192739.GB5791@martell.zuzino.mipt.ru>
Date: Mon, 11 Aug 2008 23:27:39 +0400
From: Alexey Dobriyan <adobriyan@...il.com>
To: Vegard Nossum <vegard.nossum@...il.com>
Cc: "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
LKML <linux-kernel@...r.kernel.org>
Subject: Re: latest -git: kernel hangs when pulling the plug on 8139too
On Mon, Aug 11, 2008 at 06:49:13PM +0200, Vegard Nossum wrote:
> I am experiencing a system hang as soon as pull the cable out of eth0.
> Kernel version is unmodified latest -git
> (v2.6.27-rc2/796aadeb1b2db9b5d463946766c5bbfd7717158c) and my network
> driver is 8139too (for both eth0 and eth1).
>
> I am running netconsole on eth0 (the interface which I disconnect).
>
> A bit of serial console (ttyS0) output:
>
> Linux version 2.6.27-rc2-00325-g796aade (vegardno@....ifi.uio.no) (gcc
> version 4.1.2 20071124 (Red Hat 4.1.2-42)) #2 SMP PREEMPT Mon Aug 11
> 18:29:29 CEST 2008
> ...
> Kernel command line: ro root=/dev/VolGroup00/LogVol00 rhgb
> console=tty0 console=ttyS0,115200 ignore_loglevel debug initcall_debug
> nmi_watchdog=1 panic=30 sysrq_always_enabled
> netconsole=@....168.0.172/,@192.168.0.11/ 3
> ...
> calling rtl8139_init_module+0x0/0x20
> 8139too Fast Ethernet driver 0.9.28
> 8139too 0000:02:01.0: PCI INT A -> GSI 18 (level, low) -> IRQ 18
> eth0: RealTek RTL8139 at 0xf881ac00, 00:10:a7:09:48:52, IRQ 18
> eth0: Identified 8139 chip type 'RTL-8139C'
> 8139too 0000:02:05.0: PCI INT A -> GSI 21 (level, low) -> IRQ 21
> eth1: RealTek RTL8139 at 0xf881c000, 00:16:ec:ee:ad:b9, IRQ 21
> eth1: Identified 8139 chip type 'RTL-8100B/8139D'
> initcall rtl8139_init_module+0x0/0x20 returned 0 after 37 msecs
> ...
> eth0: link down
> [hangs here; num lock + scroll lock flashing]
> [reboots automatically, probably due to NMI watchdog?]
>
> There is no detailed crash output on either tty0 or ttyS0. Keyboard
> sysrq does not work.
>
> Is it expected that pulling the plug on an interface running
> netconsole should hang the machine?
Of course, no!
> Is there anything I can do to help narrow down the problem?
I have two realteks and one of them showed very similar symptoms:
cable unplug and immediate panic without anything. If this is a same
bug, it's pretty old.
> I'd like to avoid a bisect (esp. since I don't know the latest good
> version), but it is a possible solution.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists