re: mostly working SMP again

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

   >   hmm, maybe that panic from scsipi wasn't all that unrelated after  
   > all
   >   - - I got another one:
   >   sdstart(): dequeued wrong buf
   >   scsipi_put_xs()
   >   scsipi_complete()
   >   scsipi_done()
   >   ncr53c9x_done()
   >   ncr53c9x_intr()
   >
   >
   > hmmm..i don't have any disks or cdrom in my sparc box..
   
   hmm, it took an interrupt in uvm_fault_internal() - is that supposed  
   to happen?
   Also, the MP kernel made it all the way through filesystem checks /  
   log replays and only crapped out when sshd started.
   
   Any idea what's happening here?

what's the stack trace like around uvm_fault?