Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 14 Feb 2006 20:01:54 -0800
From:      Joseph Olatt <joji@eskimo.com>
To:        Peter Jeremy <peterjeremy@optushome.com.au>
Cc:        freebsd-stable@freebsd.org
Subject:   Re: kernel panic on 6.1-PRERELEASE with Lexar Jumpdrive2
Message-ID:  <20060214200154.A3559@eskimo.com>
In-Reply-To: <20060214090748.GA900@turion.vk2pj.dyndns.org>; from peterjeremy@optushome.com.au on Tue, Feb 14, 2006 at 08:07:48PM %2B1100
References:  <20060213194406.A29000@eskimo.com> <20060214090748.GA900@turion.vk2pj.dyndns.org>

index | next in thread | previous in thread | raw e-mail

On Tue, Feb 14, 2006 at 08:07:48PM +1100, Peter Jeremy wrote:
> On Mon, 2006-Feb-13 19:44:07 -0800, Joseph Olatt wrote:
> >I've been having kernel panics when I use one particular USB flash drive
> >(LEXAR MEDIA JUMPDRIVE2, rev 2.00/1.25, addr 2) with FreeBSD
> >6.1-PRERELEASE (See [1] below). All I have to do to get the kernel to 
> >panic is to insert and remove the drive from the USB port 2 or 3 times
> 
> You left out details of the panic and backtrace.
>  
> >This problem happens only if the "ehci" device is enabled in the kernel
> 
> Note that echi(4) states:
> BUGS
>      The driver is not finished and is quite buggy.
> 
> >So I suspect that it is a problem with the "ehci" driver. Unfortunately
> >I'm not being able to get a crash dump as the system starts dumping and
> >then freezes.
> 
> Try adding the following options so you can get backtraces without dumps:
> options         KDB
> options         KDB_TRACE
> options         DDB
> 

Using Peter Jeremy's suggestion, I was able to get a backtrace. Since I
couldn't figure out an easier way to capture the backtrace, I
transcribed by hand from the console. Here is the output:

### Begin Backtrace ###

dt> bt
Tracing pid 39 tid 100030 td 0x32bcc00
probedone(c3a56800, c3606000) at probedone+0x4b2
camisr(c0926a78) at camisr+0x27d
ithread_loop(c33c3400, d44aad38) at ithread_loop+0x159
fork_exit(c0631330, c33c3400, d44aad38) at fork_exit+0x70
fork_trampoline() at fork_trampoline+0x8
--- trap 0x1, eip = 0, esp = 0xd44aadbc, ebp = 0 ---

### End Backtrace ###

I still wish I could get the system to dump and save the core. I still
haven't figured out why the system freezes in the middle of a core dump.

It looks like the the function "probedone" is defined in:
/usr/src/sys/cam/cam_xpt.c

I'm not familiar with DDB. If anyone would like more information, I'll
be glad to get the system to panic again and follow your instructions.

regards,
joseph


home | help

Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060214200154.A3559>