re: How to get a crash dump with recursive panic?

[matthew green <mrg%eterna.com.au@localhost>]

matthew green writes:
> 
> i haven't looked closely, but i'm pretty sure there is a bug in
> the sparc64 crash handling where it sets doing_shutdown and that
> triggers the "Skipping crash dump on recursive panic" message,
> when infact it hasn't gotten that far yet.
> 
> someone should fix it..  it's sparc64 specific AFAICT.

ah, i wonder if this should be reverted now:

revision 1.44
date: 2000-09-28 11:54:39 -0700;  author: eeh;  state: Exp;  lines: +4 -1;
Fix the `trap type 0x10' problem.  Apparently, something has changed and
now makes use of lockmgr().  Since we broke into the debugger from an interrupt,
we don't have a curproc, so when lockmgr() tries to find our PID it barfs,
causing another break into DDB which then calls the mysterious code that
calls lockmgr() again.....

So set doing_shutdown while we're in DDB so lockmgr ignores locks.