Whereas I can't tell where the SPARC64 CD leaves me on my Ultra 5+:

well, i've never booted NetBSD from a cdrom so there you go :-)
   
   ---------------------------------------------------------------------------
   ok  boot cdrom
   ...
   wd0 at pciide0 channel 0 drive 0: <ST39140A>
   wd0: drive supports 16-sector pio transfers, lba addressing
   wd0: 8693 MB, 17662 cyl, 16 head, 63 sec, 512 bytes/sect x 17803440 sectors
   wd0: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 2
   wd0(pciide0:0:0): using PIO mode 4, DMA mode 2 (using DMA data transfers)
   pciide0: secondary channel configured to native-PCI mode
   pciide0: disabling secondary channel (no drives)
   pcons at mainbus0
   No counter-timer -- using %tick at 333MHz as system clock.
   Using tick -- intr in 3330000 cycles...done.
   raidattach: Asked for 4 units
   Kernelized RAIDframe activated
   root device:
   ---------------------------------------------------------------------------
   
   At this point I don't know what to do.  Typing "? yields
   
   use one of : raid0[a-h] raid1[a-h] raid2[a-h] raid3[a-h] hme0 wd0[a-h] halt
   root device:

hmm.. it it was from, i'd expect to see a "cd0" or something available.. the
problem here is probably that the cd driver doesn't call the MI disklabel
reader code, so sun labels are ignored on cdroms, so it can't see any partitions?
perhaps it (netbsd/sparc64) doens't know about booting from cdrom yet... we do not
use the md0-style install yet.
   
   Where is the "md0a" choice for the "microroot" install I would expect here?

no.  se above.
   
   I have no clue what to use/do at this point.  if I specify "wd0", I get asked
   for "dump device" and "file system" type, after which I see (if I type in
   "wd0b" and "ffs", respectively - even though I'm convinced these answers are
   wrong anyway)
   
   root on wd0a dumps on wd0b
   pciide0:0:0: lost interrupt
   	type: ata tc_bcount: 512: tc_skip:0
   pciide0:0:0: lost interrupt
   	type: ata tc_bcount: 512: tc_skip:0
   pciide0:0:0: lost interrupt
   	type: ata tc_bcount: 512: tc_skip:0
   pciide0:0:0: lost interrupt
   	type: ata tc_bcount: 512: tc_skip:0
   pciide0:0:0: lost interrupt
   	type: ata tc_bcount: 512: tc_skip:0
   pciide0:0:0: bus-master DMA error: missing interrupt, status=0x20
   wd0: transfer error, downgrading to PIO mode 4
   wd0(pciide0:0:0): using PIO mode 4
   wd0c: DMA error reading fsbn 0 ( wd0 bn 0; cn 0 tn 0 sn 0), retring
   
   Followed by another 4 sets of "lost interrupt" messages and then a
   
   wd0c: device timeout reading fsbn 0 (wd0 bn 0; cn 0 tn 0 sn 0), retrying
   
   Eventually it returns (after several more of the above cycles) with
   
   wd0: disk label read error
   cannot mount root, error = 6
   root device (default wd0a): 
   
   What now?  Why is the kernel choking on reading the internal EIDE hard disk?

this is the same problem that i and others have reported on the pciide driver.
i've not seen it for cdroms, but i've never booted from a cdrom.  it happens
when the OFW driver touches the pciide controller before NetBSD does, on a 
certain class of machine, with a certain class of disks.. (this also affects
some macpcc machines).  i have a U5 that will boot properly from the internal
sun-supplied 4G ide drive, but will not boot from a seagate 15G drive (even if
i replace the internal drive completely).
   
   Why are there no "md0[a-h]" and/or "cd0[a-h]" choices for "root device", as
   one might expect?

there should be cd0, but i'm fairly sure this won't work right anyway..
   
   I realize there are "traditional" methods of installing (load a miniroot and
   boot from it; or set it all up manually from Solaris, as per the INSTALL
   file), but I was really looking forward to being able to do the full install
   while being booted off of the CD-ROM ... (also, there's no "miniroot.fs" as
   per the INSTALL document, I don't know what to do with what's in "misc" -
   "ofwboot" et al. - and finally, there's "ramdisk.fs.gz" but you can't use that
   from a CD-R ... )

please remember that the sparc64 port is very new, and installers are some
of the most immature parts of the system.  there are much more important
tasks we've been working on... :-)


.mrg.