Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 2 Mar 2011 10:18:52 -0500
From:      John Baldwin <jhb@freebsd.org>
To:        Mike Tancsa <mike@sentex.net>
Cc:        freebsd-stable@freebsd.org, Jeremy Chadwick <freebsd@jdc.parodius.com>
Subject:   Re: CPU0: local APIC error 0x40 CPU1: local APIC error 0x40
Message-ID:  <201103021018.52403.jhb@freebsd.org>
In-Reply-To: <4D6E412F.6080208@sentex.net>
References:  <4D6DA259.4050307@sentex.net> <201103020755.54147.jhb@freebsd.org> <4D6E412F.6080208@sentex.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wednesday, March 02, 2011 8:07:59 am Mike Tancsa wrote:
> On 3/2/2011 7:55 AM, John Baldwin wrote:
> > 
> > Hmm, the interrupt pins on the each lapic look fine (they all either have a 
> > legal vector, are using NMI delivery, or are masked).
> > 
> > All of the places that send IPIs have the interrupt vectors hard-coded as 
> > constant values in the code.
> > 
> > Unfortunately there is no register that tells us which illegal vector was 
> > posted.
> > 
> > Were you doing anything related to changing the state of device interrupts 
> > (cpuset -x, kldload, kldunload, etc.) when this happened?
> 
> Hi,
> 	No, nothing at all.  I checked the logs again and nothing unusual
> leading up to it, nor was anything recorded on the serial console other
> than that error.  Do you think its just a hardware issue?

No, was trying to think if there was a scenario where an I/O APIC pin or MSI
message could specify an illegal vector.

Can you reproduce this at all?

-- 
John Baldwin



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201103021018.52403.jhb>