Re: rtmutex assert failure (was [Patch] restore the RCU callback...)

[Bill Huey (hui) <billh@gnuppy.monkey.org>]

On Wed, Aug 23, 2006 at 10:14:27AM -0700, Robert Crocombe wrote:
> On 8/22/06, hui Bill Huey <billh@gnuppy.monkey.org> wrote:
> >I turned off the tracing in the latency tracking stuff and a relatively
> >small patch is here against -rt8:
> >
> >        http://mmlinux.sourceforge.net/public/against-2.6.17-rt8-0.diff
> 
> I'm going to assume that the #error here:
> 
> +#ifdef CONFIG_LATENCY_TRACE
> +#error
> +       stop_trace();
> +#endif
> 
> is to see if I'm awake.  No, but gcc is.  I just removed it (?).
> 
> End result is as with the previous patch: nothing to serial console,
> and just the single line moaning about line 471 in blah blah blah.

I still to have streamline my patch consolidation process, so your getting
things like this as an artifact of that right now, sorry. That was the
correct change.

It might be the case that I need to remove some of that stuff and then
reupload the patch. I'm kind of guessing at the problem right now with
your stack dumps and it looks like the changes are causing more problems
than it is helping. Let me change that and upload something else to you
today.

Your case is particularly difficult since I don't have a machine as big as
yours that can replicate that problem and I can't do the normal kernel
stuff that I want to on it. However, I do have a good idea of what the
problem is and I'm trying to get reacquainted with the rt mutex code to
see if I can narrow it down.

bill