Just to see how MacBSD was doing for the Quadra 950, I decided to try
booting HWDIRECT.356 just to see what would happen.  As I expected, it
didn't finish the booting process; however, it got rather further than I
expected it to, so I thought I'd report things to the list.

Important/interesting observations:
* ADB - the HWDIRECT code seems to think that the Quadra 950 has
  IIsi-class ADB.  It does not; it has IOP-class (IIfx-class) ADB.  Despite
  this misapprehension, though, the boot process survives the ADB
  initializion stage, which is really somewhat impressive, especially
  compared to the hangs earlier versions of this code manifested there.

* SCSI - The esp driver sucessfully detects and probes both SCSI busses.
  However, with my external CD-ROM drive turned on (see second boot log),
  the boot process dies after receiving an excessive number of parity errors
  during the probing process.  (I've reported this as a PR.)

* SERIAL - the kernel boot reports finding a zsc0, despite the fact that on
  this machine I'm pretty sure the zsc0 lives behind the IOP controllers.
  Is the machine actually doing an initialization of the zsc when it prints
  those messages, or is it just assuming it can find it?  (If it's actually
  finding the zsc, I'm somewhat pleasantly surprised -- I didn't think it
  was that easy).  Also, could this be related to the crash right after
  this point?

* NUBUS - I hang just after the nubus scan begins, I assume.  I have two
  nubus cards -- a SuperMac video card (which is clearly working, at least
  to some extent, being my console display) and a Hustler 2-port accelerated
  serial card, which I suspect is more likely to be the cause of the
  problem.  What can I do to help track this down?


Here are the kernel messages without the CD-ROM drive; note these were typed
in by hand so there are inevitably some typos.  We pick up just after the
copyright message.

NetBSD 1.2G (HWDIRECT) #356: Wed Aug 13 09:50:03 PDT 1997
    root@bigjohn.noip.org:/usr/src/sys/arch/mac68k/compile/HWDIRECT
Apple Macintosh Quadra 950 (68040)
real mem = 67108864
avail mem = 6236792
using 204 buffers containing 835584 bytes of memory
mrg: 'Quadra 950 class ROMs' ROM glue, tracing off, debug off, silent traps
mrg: I/O map kludge for ROMs that use hardware addresses directly
adb: using IIsi series hardware support
adb: cleanup: (length=0x2):  0xff  0x03
adb: no devices found
adb: ADBReInit complete
mainbus0 (root)
obio0 at mainbus0
adb0 at obio0 (ADB event device)
asc0 at obio0: Apple Sound Chip
sn0 at obio0 address 08:00:07:34:64:6f
esp0 at obio0: address 0x8ae000: NCR53C96, 25 MHz, SCSI ID 7
scsibus0 at esp0: 8 targets
sd0 at scsibus0 targ0 lun0: <QUANTUM, PD425S, 606_> SCSI2 0/direct fixed
sd0: 406MB, 1520 cyl, 9 head, 60 sec, 512 bytes/sect x 833150 sectors
sd1 at scsibus0 targ2 lun0: <QUANTUM, FIREBALL1080S, 1Q09> SCSI2 0/direct fixed
sd1: 1042MB, 3835 cyl, 4 head, 139 sec, 512 bytes/sect x 2134305 sectors
st1 at scsibus0 targ4 lun0: <ARCHIVE, Python 2801-XXX, 2.50> SCSI2 1/sequential removable
st0: drive empty
esp1 at obio0: address 0x8ae402: NCR53C86, 25 MHz, SCSI ID 7
scsibus1 at esp1: 8 targets
zsc0 at obio0 chiptype 0
zstty0 at zsc0 channel 0
zstty0 at zsc0 channel 1
nubus0 at mainbus 0
<hang>


And here is the boot log with the external drive on; we are the same through
the beginning of the scan of scsibus1.

scsibus1 at esp1: 8 targets
esp1: SCSI bus parity error
esp1: SCSI bus parity error
data interrupt, but no count left.esp1: SCSI bus parity error
data interrupt, but no count left.esp1: SCSI bus parity error
data interrupt, but no count left.esp1: SCSI bus parity error
data interrupt, but no count left.esp1: SCSI bus parity error
data interrupt, but no count left.esp1: SCSI bus parity error
data interrupt, but no count left.esp1: SCSI bus parity error
data interrupt, but no count left.esp1: SCSI bus parity error
data interrupt, but no count left.esp1: SCSI bus parity error
data interrupt, but no count left.esp1: SCSI bus parity error
data interrupt, but no count left.esp1: SCSI bus parity error
data interrupt, but no count left.esp1: SCSI bus parity error
data interrupt, but no count left.esp1: SCSI bus parity error
data interrupt, but no count left.esp1: SCSI bus parity error
data interrupt, but no count left.esp1: SCSI bus parity error
data interrupt, but no count left.esp1: SCSI bus parity error
data interrupt, but no count left.data interrupt, but no count left.esp1: SCSI bus parity error
<hang>


Any advice anyone can give would be greatly appreciated, and any help I can
give tracking down problems is gladly given.

Allan Briggs: are you still working on the IIfx and 900-class Quadras?  Is
there anything I can help test yet?


-- 
Jonathan Lennox
lennox@cs.columbia.edu