Having got my hands on a couple of Blade 100's, I thought I'd see if I
could see why we fail with OBP 4.17.1.  Using a kernel compiled with
PCI_CONFIG_DUMP and DEBUG=1 in sparc64/dev/psycho.c, I captured the
kernel startup output.  However, I can't see anything obviously wrong
with the output from the machine with OBP 4.17.1.  The output files are
at:

  http://www.coris.org.uk/misc/B100-4.0.0-boot.text
  http://www.coris.org.uk/misc/B100-4.17.1-boot.text

if anybody cares to look and spot the obvious problem ;-)  There are a
couple of differences between the machines - the one with 4.0.0 has a
dual SCSI card in and the quad HME cards are in different slots.

Thanks,

J

-- 
  My other computer also runs NetBSD    /        Sailing at Newbiggin
        http://www.netbsd.org/        /   http://www.newbigginsailingclub.org/