[gnu.utils.bug] gas for Sequent Symmetry?

daniel@ux1.cso.uiuc.edu (02/10/90)

/* Written  8:25 pm  Feb  8, 1990 by daniel@ux1.cso.uiuc.edu in ux1.cso.uiuc.edu:gnu.misc.discuss */
/* ---------- "gas for Sequent Symmetry?" ---------- */
I have been trying to get g++ running on a Sequent Symmetry.  When I
went to compile libg++-1.36.3, I got an assembler error compiling
builtin.cc so I installed gas (as /usr/local/lib/gcc-as).  The library
then completely compiled with no problem but when ranlib looked at it,
it complained:
	. . . . . .  . . .  . .. ... . . .
	gcc -I/usr/gnu/lib/gcc-include -I/usr/include -I/cso/staff/daniel/Gnu/libg++-1.36.3/g++-include -g -O -fstrength-reduce -fdelayed-branch  -Wall  -pipe -c gnulib3.c
gnulib3.c: In function __main:
gnulib3.c:69: warning: control reaches end of non-void function
	rm -f libg++.a
	ar r libg++.a AllocRing.o builtin.o File.o  stream.o streambuf.o  PlotFile.o  SFile.o  Obstack.o  regex.o String.o  Integer.o Rational.o Complex.o  BitSet.o BitString.o Random.o  RNG.o ACG.o MLCG.o Sample.o  Fix.o Fix16.o Fix24.o Curses.o GetOpt.o EH.o  malloc.o  gnulib3.o
ar: creating libg++.a
	ranlib libg++.a
ranlib: libg++.a(AllocRing.o): old NS32000 format .o file
*** Error code 1
Make: .  Stop.
*** Error code 1
Make: .  Stop.

How do I get gas to produce the correct .o file format?

-- Daniel Pommert.
 pommert@uiuc.edu
/* End of text from ux1.cso.uiuc.edu:gnu.misc.discuss */

daniel@ux1.cso.uiuc.edu (02/10/90)

I should note that I generated this as an Intel 80386 product.  Here is
the make a3896 output:

	gcc -g -I. -c as.c
	gcc -g -I. -Derror=as_fatal -c xrealloc.c
	gcc -g -I. -Derror=as_fatal -c xmalloc.c
	gcc -g -I. -Derror=as_fatal -c hash.c
	gcc -g -I. -c hex-value.c
	gcc -g -I. -c atof-generic.c
	gcc -g -I. -c append.c
	gcc -g -I. -DNO_VARARGS   -c messages.c
	gcc -g -I. -c expr.c
	gcc -g -I. -c app.c
	gcc -g -I. -c frags.c
	gcc -g -I. -c input-file.c
	gcc -g -I. -c input-scrub.c
	gcc -g -I. -c output-file.c
	gcc -g -I. -c subsegs.c
	gcc -g -I. -c symbols.c
	gcc -g -I. -c version.c
	gcc -g -I. -c flonum-const.c
	gcc -g -I. -c flonum-copy.c
	gcc -g -I. -c flonum-mult.c
	gcc -g -I. -c strstr.c
	gcc -g -I. -c bignum-copy.c
	gcc -g -I. -c gdb.c
	gcc -g -I. -c gdb-file.c
	gcc -g -I. -c gdb-symbols.c
	gcc -g -I. -c gdb-blocks.c
	gcc -g -I. -c obstack.c
	gcc -g -I. -c gdb-lines.c
	gcc -g -I. -DNO_VARARGS   -c i386.c
	gcc -g -I. -c atof-i386.c
	gcc -g -I. -c write.c
	gcc -g -I. -c read.c
	gcc -o a386 -g -I. as.o		xrealloc.o	xmalloc.o
	hash.o		hex-value.o    atof-generic.o	append.o
	messages.o	expr.o		app.o          frags.o	
	input-file.o	input-scrub.o	output-file.o	              
	subsegs.o	symbols.o				
	version.o      flonum-const.o	flonum-copy.o	flonum-mult.o
	strstr.o	bignum-copy.o  gdb.o		gdb-file.o
	gdb-symbols.o	gdb-blocks.o	obstack.o      gdb-lines.o
	i386.o  atof-i386.o  write.o  read.o 

I also looked through /usr/include/a.out.h and can't figure out what
makes a file NS32000 as opposed to i386.  The source for ranlib is
locked away in a private directory and I haven't bugged someone with
enough system privileges to look at it.

-- Daniel Pommert.
 pommert@uiuc.edu

jw@sics.se (Johan Widen) (02/12/90)

>>>>> In article <41600002@ux1.cso.uiuc.edu>, daniel@ux1.cso.uiuc.edu writes:

daniel> I have been trying to get g++ running on a Sequent Symmetry.  When I
daniel> went to compile libg++-1.36.3, I got an assembler error compiling
daniel> builtin.cc so I installed gas (as /usr/local/lib/gcc-as).  The library
daniel> then completely compiled with no problem but when ranlib looked at it,
daniel> it complained:
daniel> 	ranlib libg++.a
daniel> ranlib: libg++.a(AllocRing.o): old NS32000 format .o file
daniel> *** Error code 1

A version of gas (1.34) that works on the Sequent Symmetry is available for
anonymous ftp from
	sics.se	(192.16.123.90)


--
Johan Widen
SICS, PO Box 1263, S-164 28 KISTA, SWEDEN	Internet: jw@sics.se
Tel: +46 8 752 15 32	Ttx: 812 61 54 SICS S	Fax: +46 8 751 72 30