xfs
[Top] [All Lists]

Re: Kernel oops with current xfs CVS code and smp...

To: Jordan Breeding <ledzep37@xxxxxxxxx>
Subject: Re: Kernel oops with current xfs CVS code and smp...
From: Krzysztof Rusocki <kszysiu@xxxxxxxxxxxxxxxxx>
Date: Tue, 19 Feb 2002 15:39:24 +0100
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: <3C722814.416C9A85@xxxxxxxxx>
References: <3C722814.416C9A85@xxxxxxxxx>
Sender: owner-linux-xfs@xxxxxxxxxxx
User-agent: Mutt/1.3.25i
On Tue, Feb 19, 2002 at 04:25:24AM -0600, Jordan Breeding wrote:
> Hello,
> 
>   I have an smp system and have been using the Robert M. Love preempt
> patches for some time now with no problem.  I have also been wanting to
> try out the xfs filesystem for some time now, so since the cvs tree was
> recently updated to the 2.5.5-pre1 kernel which has ALSA and preempt
> included I decided to try making an xfs system.  Everything was going
> great, kernel build, recreating my filesystems, etc. until I needed to
> do the final reboot into my new kernel.  The resulting smp and preempt
> kernel (also has current acpi and Robert M. Love's networking patches to
> allow contributions to /dev/random) will boot just fine but soon after
> booting the kernel and the userland init level stuff starts up the
> kernel will oops everytime.  The only way I have found to make the
> kernel not oops was to boot it using the "nosmp" argument.  I assume
> that xfs is tested fairly well with smp since it has been out for a
> while but I am guessing it has no gotten a lot of testing against the
> preempt stuff.  Anyway if anyone knows what the problem is or how to fix
> it I would appreciate it greatly.  I am attaching a copy of the oops
> (processed and raw) as well as my .config file.  Thanks for the help and
> the great file system.

Hi Jordan,

You're not first reporting oops with such backtrace.
Turn off spinlock debugging, since it is not compatible with XFS.

Cheers,
Krzysztof


<Prev in Thread] Current Thread [Next in Thread>