Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 20 May 2005 16:05:32 -0700
From:      Peter Wemm <peter@wemm.org>
To:        freebsd-amd64@freebsd.org
Cc:        amd64@freebsd.org
Subject:   Re: help with GPF on 5.4-STABLE
Message-ID:  <200505201605.33275.peter@wemm.org>
In-Reply-To: <20050520135046.T8229@carver.gumbysoft.com>
References:  <1116566651.1588.17.camel@server.mcneil.com> <20050520135046.T8229@carver.gumbysoft.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Friday 20 May 2005 01:59 pm, Doug White wrote:
> kern.timeout.c line 530 is
>
> 530         mtx_unlock_spin(&callout_lock);
>
> I'm not sure what in there would generate a GPF. 

Allowing an invalid address to reach the AGU (address generation unit) 
will cause a GPF with no specific details.  Its a real pain in the butt 
because it should cause a SIGBUS with the bad address in siginfo.  But 
since GPF doesn't tell us anything useful, we can't do this.   

Anyway...  the usual cause is that the upper 16 bits are not all 0's or 
all 1's.  ie: its a reference outside the sign extended 48 bit virtual 
address space.  The disassemble that you asked for will show us which 
register we should be looking at for what the actual address was..
-- 
Peter Wemm - peter@wemm.org; peter@FreeBSD.org; peter@yahoo-inc.com
"All of this is for nothing if we don't go to the stars" - JMS/B5



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200505201605.33275.peter>