Re: "sleeping without a mutex" panic with FixitCD from today's SNAP

Dmitry Morozovsky wrote:
> On Sat, 10 Apr 2004, Scott Long wrote:
> 
> SL> Cordula's Web wrote:
> SL> >>Trying to boot freshly compiled (via current-on-stable vay, mostly described in
> SL> >>http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&oe=UTF-8&c2coff=1&selm=20040319034905.GA975%40doom.homeunix.org)
> SL> >>-current LiveFS CD on three different machines (all i686-like, from AMD Duron
> SL> >>600 on VIA KT133 to AthlonXP 2k+ on nVidia nForce2; all with ATACI CD/DVD) I
> SL> >>got 100% reproducible ddb panic after mounting CD with "Fixit" menu item:
> SL> >>
> SL> >>panic: sleeping without a mutex
> SL> >>at line 144 in file /usr/src/sys/kern/kern_synch.c
> SL> >
> SL> >
> SL> > Same here, while trying to upgrade from 5.2.1-RELEASE to yesterday's
> SL> > and today's -CURRENT. It happens here:
> SL> >
> SL> > # make installkernel KERNCONF=GENERIC
> SL> > # reboot   (in single user mode)
> SL> > # mount -a
> SL> > # mergemaster -p
> SL> > # make installworld
> SL> > <panic: sleeping without a mutex
> SL> > at line 144 in file /usr/src/sys/kern/kern_synch.c>
> SL> >
> SL> > The panic is 100% reproducible.
> SL>
> SL> Providing a backtrace from the debugger here is very important.  Any
> SL> chance that you could do that?
> 
> My chances are pretty low, as I have no -current systems runnning. I cat write
> ddb's traceback though if it's helpful (supposedly, I also can make serial
> console for that).
> 
> Sincerely,
> D.Marck                                     [DM5020, MCK-RIPE, DM3-RIPN]

DDB traceback is fine.  Without some sort of traceback, it's impossible
to know what is triggering the message in msleep that you are seeing.

Thanks,

Scott