[Don't reply to me directly, as I'll be going offline soon for a
 week or so, but I'll check the archives when I get the chance]

I would hope that the following build failure which I got when
crossbuilding on FreeBSD for a NetBSD Sparc (not sparc64) build
has long since been fixed -- my source was updated last the
morning of 30.Aug before starting the failed build.


[ ... ]
#   compile  GENERIC/scsipi_base.o
/dist/build/../obj/NetBSD/tools/obj.sparc/tools.FreeBSD-4.7-STABLE-i386/bin/spar
c--netbsdelf-gcc -mno-fpu -ffreestanding -pipe -O1 -Werror -Wall -Wno-main -Wno-
format-zero-length -Wpointer-arith -Wmissing-prototypes -Wstrict-prototypes -Wno
-sign-compare -fno-zero-initialized-in-bss -I. -I/dist/src/NetBSD-src/src/sys/ar
ch -I/dist/src/NetBSD-src/src/sys -nostdinc -DRASTERCONSOLE -DLKM -DMAXUSERS=32
-D_KERNEL -D_KERNEL_OPT -c /dist/src/NetBSD-src/src/sys/dev/scsipi/scsipi_base.c
/dist/src/NetBSD-src/src/sys/dev/scsipi/scsipi_base.c: In function `scsipi_init'
:
/dist/src/NetBSD-src/src/sys/dev/scsipi/scsipi_base.c:107: error: `PAGE_SIZE' un
declared (first use in this function)
/dist/src/NetBSD-src/src/sys/dev/scsipi/scsipi_base.c:107: error: (Each undeclar
ed identifier is reported only once
/dist/src/NetBSD-src/src/sys/dev/scsipi/scsipi_base.c:107: error: for each funct
ion it appears in.)
*** Failed target:  scsipi_base.o


I've looked very quickly in the latest RCS source files I've just
updated to see if there's been a change in the obvious places,
but I don't see anything.  However, I should note that it seems
the mirror site from which I update (via CVSup from cvsup.no )
or somewhere upstream from it is now updating in a way which
causes the last-modified-time of every file in the kernel source
to be changed at each update, whether or not a new delta or tag
has been applied, which sort of makes it difficult to see at a
glance which of the foo.c,v files has been recently changed.

Apologies in advance for probably wasting your time with the
above failure that's no doubt been fixed somewhere I haven't
thought to look in the source.  I'm not online enough to offer
timely reports of failures or to try a new build before sending
a report, without losing another week of timeliness...


thanks
barry bouwsma