Re: 2.6.17-rt8 crash amd64

[hui]

On Tue, Aug 01, 2006 at 07:19:56PM -0700, Bill Huey wrote:
> On Tue, Aug 01, 2006 at 09:31:42PM -0400, Steven Rostedt wrote:
> > On Tue, 2006-08-01 at 18:18 -0700, Bill Huey wrote:
> > > [   42.124525]        <ffffffff8029ae98>{atomic_dec_and_spin_lock+21}
> > > [   42.131032]        <ffffffff8025fd89>{schedule+236}
> > > [   42.136195]        <ffffffff8026078f>{rt_lock_slowlock+351}
> > > [   42.142086]        <ffffffff8026117d>{__lock_text_start+13}
> > > [   42.147966]        <ffffffff8029ae98>{atomic_dec_and_spin_lock+21}
> > > [   42.154476]        <ffffffff8020c4e9>{dput+57}
> > > [   42.159194]        <ffffffff802093f3>{__link_path_walk+1710}
> > > [   42.165166]        <ffffffff802617ad>{_raw_spin_unlock+46}
> > > [   42.170961]        <ffffffff8020db81>{link_path_walk+103}
> > > [   42.176672]        <ffffffff8020be5a>{do_path_lookup+644}
> > > [   42.182379]        <ffffffff80223829>{__user_walk_fd+63}
> > > [   42.187994]        <ffffffff8023fce4>{vfs_lstat_fd+33}
> > > [   42.193434]        <ffffffff8022b3e4>{sys_newlstat+34}
> > > [   42.198871]        <ffffffff8025ce3d>{error_exit+0}
> > > [   42.204040]        <ffffffff8025bf22>{system_call+126}
> > 
> > This back trace is definitely ugly.  Do you get this all the time? And
> > if so, could you compile in frame pointers and try again.  (I'll dig
> > through this in the mean time.) 
> 
> Not sure, I'm getting hard reboots as well from what looks like more
> atomic scheduling violations. I'll tweek my kernel config to be more
> friendly about these things. It looked like it was in the rtmutex code,
> which is why I CCed you.
> 
> Any other configuration suggestions ?

It's already compiling with frame pointers. Unfortunately, this is about
as good it gets for a stack trace unless you've got another suggestion. 

bill

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/