🚀 go-pugleaf

	I retrieved gcc-2.3.1.tar.Z from prep and incountered the following
compilation problems:

On my Sun386i running SunOS 4.0.1 after typing:

->configure --target=sun386i-sunos4.0
	and
->make LANGUAGES=c
	
	compiling...

rm -f tmplibgcc2.a
set -e; \
for name in _muldi3 _divdi3 _moddi3 _udivdi3 _umoddi3 _negdi2  _lshrdi3 _lshldi3 _ashldi3 _ashrdi3   _udiv_w_sdiv _udivmoddi4 _cmpdi2 _ucmpdi2 _floatdidf _floatdisf  _fixunsdfsi _fixunssfsi _fixunsdfdi _fixdfdi _fixunssfdi _fixsfdi  __gcc_bcmp _varargs _eprintf _builtin_new _caps_New _builtin_del  _bb _shtab _clear_cache _trampoline __main _exit _ctors; \
do \
  echo ${name}; \
  ./xgcc -B./ -O2  -DIN_GCC   -g  -g1 -I. -I. -I./config -c -DL${name} \
      ./libgcc2.c -o ${name}.o; \
  if [ $? -eq 0 ] ; then true; else exit 1; fi; \
  ar rc tmplibgcc2.a ${name}.o; \
  rm -f ${name}.o; \
done
_muldi3

The compilation gets to this point and then hangs.  After 1 hour I did a ps ux
and found:

arras     2213 90.7  3.6  320  408 p2 R    19:37  73:49 ./cc1 /usr/tmp/cca02211.i -quiet -dumpbase libgcc2.c -g -g1 -O2 -o /usr/tmp/cca02211.s
arras     2210  0.0  0.0   24    0 p2 IW   19:37   0:00 sh -ce set -e; \for name in _muldi3 _divdi3 _moddi3 _udivdi3 _umoddi3 _negdi2  _lshrdi3 _lshldi3 _ashldi3 _a (there's more)
arras     2211  0.0  0.5   48   56 p2 I    19:37   0:00 ./xgcc -B./ -O2 -DIN_GCC -g -g1 -I. -I. -I./config -c -DL_muldi3 ./libgcc2.c -o _muldi3.o

The CPU load was 1.0 1.0 1.0.  Gcc 2.2.2 compiled very nicely on the same
machine.

	==	==	==	==	==	==	==	==

On my SPARCstation ELC running SunOS 4.1.2:

When I type

->ls -l gcc-2.3.1/cp-parse.*
-rw-r--r--  1 arras      304228 Sep 18 08:43 gcc-2.3.1/cp-parse.c
-rw-r--r--  1 arras        1871 Sep 18 08:43 gcc-2.3.1/cp-parse.h
-r--r--r--  1 arras      121744 Oct 19 23:19 gcc-2.3.1/cp-parse.y

I get a date for cp-parse.c which is before cp-parse.y meaning, I need
bison.  I got bison-1.19 and continued the make and everything went well,
and it produced a compiler that seems to work.  I got this error when
making the stage 3 compiler, probably because cp-parse.y is for C++ and I
used LANGUAGES=c for both the stage 1 and 2 compilers.

Not a problem, but a question:

	The INSTALL instructions say to use CFLAGS="-g -O", I compiled
gcc 2.3.1 with CFLAGS="-O2" because I don't intend on doing any debugging.
I typed `make compare' and received NO errors (or diffs).  Is there any
consensus agaist doing this?  Thank you for a tremendously well done job!

                                    __    __
Michael Arras                      /  \  |  |  Bavarian Research
arras@forwiss.uni-erlangen.de     /    \ |  |  Center for
Artificial Neural Network and    /  /\  \|  |  Knowledge-Based
Fuzzy Logic Research Group      |   __   |  |  Systems__________
Tel (09131) 691-211             |__|  |__|__|
I speak for myself.             F O R W I S S  ERLANGEN * MUNICH * PASSAU