On Sep 20, 2007, at 18:50, Martin Husemann wrote:
> On Thu, Sep 20, 2007 at 06:20:42PM -0400, Chris Ross wrote:
>> unmounting file systems...panic: switch wchan
>
> That panic means the scheduler is trying to switch to a lwp that  
> has l_wchan
> still set - i.e. has not properly been awoken after a ltsleep().

   Ahh.  Okay.  Is this the sort of thing that would be possible to  
fix?  Or, is it something that is just completely different after the  
idlelwp branch merge, such that it's not worth thinking about?

   Let me know if you can think of any way to have it work as  
intended, rather than fail.  And, will it work properly when I try it  
the next time?

             - Chris