Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 30 Nov 2008 12:46:58 -0800
From:      Kris Kennaway <kris@FreeBSD.org>
To:        Holger Kipp <hk@alogis.com>
Cc:        stable@freebsd.org
Subject:   Re: panic: spin lock held too long on 7.1-PRERELEASE (sio)
Message-ID:  <4932FBC2.5060807@FreeBSD.org>
In-Reply-To: <20081130124938.GA44403@intserv.int1.b.intern>
References:  <20081130124938.GA44403@intserv.int1.b.intern>

next in thread | previous in thread | raw e-mail | index | archive | help
Holger Kipp wrote:
> Hi,
> 
> I currently encounter
> 
> spin lock 0xc0c83ba8 (sio) held by 0xc6337460 (tid 100005) too long
> panic: spin lock held too long
> cpuid=3
> 
> on hp server with two dual-core intel xeon (3.8GHz) and 3GB RAM.
> System is 7.1-PRELEASE last fetched 29.11.2008, 17:58 UTC.
> 
> This happens with VSCom Card that was previously in use in a
> 5.3-STABLE system (also hp server, but older hardware) without
> problems. The card is a 8-port serial card using puc0 and puc1
> with 4 ports each.
> 
> Interestingly, problem seems to occur only if sendfax is using
> devices attache to puc1. Works without problems at the moment
> with sendfax only using puc0-sios. See dmesg below.
> 
> Any ideas what might be causing this, or is this even related
> to kern/121322, 118044 or 117913?

Enable WITNESS in your kernel and try to reproduce the problem.  It 
should give additional information about the condition leading to the 
deadlock.

Kris




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4932FBC2.5060807>