I've built GENERIC kernels from 2000-02-19 source just fine, but I
tried to build a very stripped-down kernel for an IPX and got

ld -Ttext F0004000 -e start -n -X -o netbsd ${SYSTEM_OBJ} vers.o
audio.o: In function `audioopen':
audio.o(.text+0x9ec): undefined reference to `audio_cd'
audio.o(.text+0x9f8): undefined reference to `audio_cd'
audio.o(.text+0xa18): undefined reference to `audio_cd'
audio.o: In function `audioclose':
audio.o(.text+0xb00): undefined reference to `audio_cd'
audio.o(.text+0xb04): undefined reference to `audio_cd'
audio.o(.text+0xb84): more undefined references to `audio_cd' follow
autoconf.o: In function `cpu_configure':
autoconf.o(.text+0xe34): undefined reference to `obio_find_rom_map'
autoconf.o: In function `findzs':
autoconf.o(.text+0x1744): undefined reference to `obio_find_rom_map'
cache.o: In function `hypersparc_cache_enable':
cache.o(.text+0x22c): undefined reference to `dvma_cachealign'
cache.o(.text+0x240): undefined reference to `dvma_cachealign'
*** Error code 1

What's up, anyone know?  There is some potential for pilot error, in
that I stripped out all the devices that didn't show up in the GENERIC
kernel's boot messages, but now realize I didn't strip out the options
lines for SUN4M and SUN4...but though that should produce a suboptimal
kernel, it surely shouldn't produce an unlinkable kernel!

Am I just missing some fixes, or is there something worse here?

I'm now building a GENERIC kernel to run nm on, and building another
test kernel without options SUN4 and SUN4M...will report results in the
morning.

					der Mouse

			       mouse@rodents.montreal.qc.ca
		     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B