Subject: Two build problems
To: None <port-sun3@NetBSD.ORG>
From: der Mouse <mouse@Collatz.McRCIM.McGill.EDU>
List: port-sun3
Date: 07/31/1995 10:09:10
I'm trying to do a full rebuild from source on a Sun-3.  As you can
probably guess since I'm sending this message, I've had a little
trouble.

The source tree was supped a couple of weeks ago.  I'd be using a newer
source tree, but shortly after I got this one, sup went nuts deleting
directories right and left, and only recently has it recreated them,
and it hasn't yet populated them with files again.  So I'm stuck
working with an old source tree.

First, a missing directory, or perhaps a stray SUBDIRS entry:

===> libnet
cd: can't cd to /usr/src/sys/arch/sun3/stand/libnet

Sure enough, that directory doesn't exist.  This I "fixed" by removing
libnet from sys/arch/sun3/stand/Makefile.  Then I restarted and got
something more serious:

===> xlint
===> lint1
cc -O -I. -DXXX_BROKEN_GCC   -c cgram.c
cc -O -I. -DXXX_BROKEN_GCC   -c scan.c
cc -O -I. -DXXX_BROKEN_GCC   -c mem1.c
cc -O -I. -DXXX_BROKEN_GCC   -c mem.c
cc -O -I. -DXXX_BROKEN_GCC   -c err.c
cc -O -I. -DXXX_BROKEN_GCC   -c main1.c
cc -O -I. -DXXX_BROKEN_GCC   -c decl.c
cc -O -I. -DXXX_BROKEN_GCC   -c tree.c
cc -O -I. -DXXX_BROKEN_GCC   -c func.c
cc -O -I. -DXXX_BROKEN_GCC   -c init.c
cc -O -I. -DXXX_BROKEN_GCC   -c emit.c
cc -O -I. -DXXX_BROKEN_GCC   -c emit1.c
cc   -o lint1 cgram.o scan.o mem1.o mem.o err.o main1.o decl.o tree.o func.o init.o emit.o emit1.o -ll
tree.o: Undefined symbol `___fixunsxfdi' referenced from text segment
tree.o: Undefined symbol `___floatdixf' referenced from text segment
tree.o: Undefined symbol `___floatdixf' referenced from text segment
tree.o: Undefined symbol `___fixxfdi' referenced from text segment
*** Error code 1

Stop.

This too I dealt with by yanking a directory from a SUBDIRS list, in
this case xlint from usr.bin/Makefile.

Are these known problems that would be fixed if I could get a clean sup
to build from, or should I send-pr them?  (There's another problem -
tail keeps coredumping - but that is still under investigation.)

					der Mouse

			    mouse@collatz.mcrcim.mcgill.edu