> Here is what I get when I make the second time (first time generates
> lots of stuff then failes here -- the second make starts off where
> it failed last time)
> [ ... ]

OK, given everything else you've sent me, this makes no sense.

Send me the following information, in private e-mail:

	(1) dmesg from your system.

	(2) 'df' of the file system containing your kernel tree.

	(3) 'df' of your GENERIC kernel compile dir.

	(4) 'mount' output from your system.

	(5) a shar file of all of the dec*.h files in your kernel
	    compile dir (i.e. shar dec*.h there, send the result to
	    me).

	(6) your cpuconf.c file.

	(7) a preprocessed version of cpuconf.c, i.e. the output of:

cc -E -g -O2 -Werror -Wall -Wstrict-prototypes -Wmissing-prototypes \
-Wno-format -mno-fp-regs -I. -I../../../../arch -I../../.../.. -nostdinc \
-DEXEC_ECOFF -DEXEC_ELF64 -DEXEC_SCRIPT -DSWAPPAGER -DVNODEPAGER \
-DDEVPAGER -DDEC_3000_500 -DDEC_3000_300 -DDEC_2100_A50 -DDEC_KN20AA \
-DDEC_AXPPCI_33 -DDEC_EB164 -DTIMEZONE="5*60" -DDST="1" -DDEBUG \
-DDIAGNOSTIC -DKTRACE -DFIFO -DFFS -DMFS -DNFSSERVER -DNFSCLIENT -DINET \
-DCOMPAT_43 -DCOMPAT_10 -DCOMPAT_11 -DCOMPAT_12 -DLKM -DEISAVERBOSE \
-DPCIVERBOSE -DTCVERBOSE -DUCONSOLE -DMAXUSERS=8 -D_KERNEL  -Dalpha \
../../../../arch/alpha/alpha/cpuconf.c


Hopefully, _something_ there will give me a clue as to what's going
on.

It's worth noting that everything is building fine for a bunch of
other people (myself included 8-), so i strongly believe that it's
_some_ kind of misconfiguration or something on your system.



> > > > you might make sure your 'make' binary is up to date, as well. it
> > couldn't be causing these problems, but it could cause you other
> > problems later.
> The make and config are pretty much uptodate.
> 
> What are the files which normally contain these defines?

Those dec*.h files generated in your kernel compile directory define
them.  Those files are included by cpuconf.c.



cgd