Subject: Re: a process hanged?
To: Frank van der Linden <firstname.lastname@example.org>
From: Bill Sommerfeld <email@example.com>
Date: 11/19/2001 16:15:06
> > I made new kernel, but it doesn't work well. It looks like
> > kernel diagnostic routines found deadlock or something, but
> > I'm not sure. Where should I start to look over? Thanks.
> Unfortunately, RAIDframe and LOCKDEBUG don't mix. RAIDframe contains
> quite a few instances where it calls either malloc or pool_get
> with the WAITOK flag, which is correctly discovered by the
> LOCKDEBUG code as incorrect.
It should be "obvious" but a consequence of this is that SMP and
RAIDframe don't currently mix.
In general, if feature "FOO" doesn't work with LOCKDEBUG enabled, an
SMP kernel using "FOO" is extremely likely to run into trouble; the
most likely symptom is a CPU getting "stuck", eventually leading to a